Handbook of Neurosurgery 7th Ed

18. Spine & spinal cord

18.1. Low back pain and radiculopathy

Key concepts¹:

• low back pain is common, and in ≈ 85% of cases, no specific diagnosis can be made

• initial assessment is geared to detecting “red flags” (indicating potentially serious pathology), and in the absence of these, imaging studies and further testing of patients is usually not helpful during the first 4 weeks of low back symptoms

• relief of discomfort is usually best achieved with nonprescription pain meds and/or spinal manipulation

• while activities may need to be modified, bed rest beyond 4 days may be more harmful than helpful, and patients are encouraged to return to work or their normal daily activities as soon as possible

• 89-90% of patients with low back problems will improve within 1 month even without treatment (including patients with sciatica from disc herniation)

Low back pain (LBP) is extremely prevalent, and is the second most common reason for people to seek medical attention². LBP accounts for ≈ 15% of all sick leave from work, and is the most common cause of disability for persons < 45 yrs age³. Estimates of lifetime prevalence range from 60-90%, and the annual incidence is 5%⁴. Only 1% of patients will have nerve-root symptoms, and only 1-3% have lumbar disc herniation. The prognosis for most cases of LBP is good, and improvement usually occurs with little or no medical intervention.

DEFINITIONS/CLASSIFICATIONS

radiculopathy	dysfunction of a nerve root (signs and symptoms may include: pain in the distribution of that nerve root, dermatomal sensory disturbances, weakness of muscles innervated by that nerve root, and hypoactive muscle stretch reflexes of the same muscles)
mechanical low back pain	AKA “musculoskeletal” back pain (both non-specific terms). The most common form of low back pain. May result from strain of the paraspinal muscles and/or ligaments, irritation of facet joints… Excludes anatomically identifiable causes (e.g. tumor, disc herniation…)

INTERVERTEBRAL DISC

The function of the intervertebral disc is to permit stable motion of the spine while supporting and distributing loads under movement.

Anatomy

Anulus fibrosus (anulus may alternatively be spelled annulus, but fibrosus is the only correct spelling and is distinct from fibrosis)⁵: the multilaminated ligament that encompasses the periphery of the disc space. Attaches to the end-plate cartilage and ring apophyseal bone. Blends centrally with the nucleus pulposus.

Nucleus pulposus: the central portion of the disc. A remnant of the notocord.

Capsule⁵: combined fibers of the anulus fibrosus and the posterior longitudinal ligament (this term is useful because these 2 structures may not be distinguishable on imaging studies).

NOMENCLATURE FOR DISC PATHOLOGY

Historically, the terminology has been contentious and nonstandardized. Many diagnostic labels are used inconsistently (e.g. spondylosis, sprain, strain, musculoskeletal pain, myofascial pain…). A subset of nomenclature proposed by a task force⁵ is shown in Table 18-1, which is useful primarily for consistent terminology related to radiographic reports, research….

Degenerated disc: some reports indicate that these can cause radicular pain possibly by an inflammatory mechanism⁶, but this is not universally accepted.

Vacuum disc: gas in the disc space (empty space on imaging), usually indicates disc degeneration, not infection.

Non-Standard terms

Included for completeness, but are not recommended because they may be confusing or ambiguous⁵.

Contained herniation⁵: displaced disc tissue that is entirely contained within an uninterrupted (but possibly distended) anulus or capsule (see page 428 for definition of capsule). It may be difficult to distinguish this on currently available imaging studies from an uncontained herniation which is underneath the posterior longitudinal ligament.

Ruptured disc: colloquial term usually intended to be equivalent to herniated disc.

Table 18-1 Nomenclature for lumbar disc pathology⁵

Term	Description
anular tears AKA anular fissures	separations between anular fibers, avulsions of fibers from their VB insertions, or breaks through fibers that extend radially, transversely, or concentrically
degeneration	desiccation, fibrosis, narrowing of the disc space, diffuse bulging of the anulus beyond the disc space, extensive fissuring (numerous anular tears), mucinous degeneration of the anulus, defects & sclerosis of end-plates, & osteophytes at the vertebral apophyses
degenerative disc disease	clinical syndrome of symptoms related to degenerative changes in the intervertebral disc (described above), also often considered to encompass degenerative changes outside the disc as well
bulging disc	generalized displacement of disc material (arbitrarily defined as > 50% or 180°) beyond the peripheral limits of the disc space*. Not considered a form of herniation. May be a normal finding, not usually symptomatic
herniation	localized displacement of disc material (< 50% or 180°) beyond the limits of the intervertebral disc space*
	focal: < 25% of the disc circumference
	broad-based: 25-50% of the disc circumference
	protrusion: the fragment does not have a “neck” that is narrower than the fragment in any dimension
	extrusion: the fragment has a “neck” that is narrower than the fragment in at least 1 dimension. 2 subtypes A. sequestration: the fragment has lost continuity with the disc of origin (AKA free fragment) B. migration: the fragment is displaced away from the site of extrusion, regardless of whether sequestered or not
	intravertebral herniation (AKA Schmorl’s node): disc herniates in the cranio-caudal direction through the cartilaginous end-plate into the VB (see page 455)

* intervertebral disc space: bounded by VB endplates in the craniocaudal dimension, and by the outer edges of the vertebral ring apophyses (exclusive of osteophytes) in the peripheral direction

Recommended classification

It is recommended¹ that acute back problems be classified into one of the 3 categories shown in Table 18-2 based on the history and physical exam (see Initial assessment of the patient with back pain below). Further evaluation, treatment, and even some information regarding prognosis can be based on this simple classification. A major goal is to detect “red flags” that may indicate potentially serious spinal or nonspinal pathology (see page 432).

Table 18-2 AHCPR classification of back problems

Clinical category	Description
potentially serious spinal condition	includes spinal tumor, infection, fracture, or cauda equina syndrome (see text)
sciatica	pain along the course of the sciatic nerve, usually resulting from nerve root compromise
nonspecific back symptoms	symptoms occurring primarily in the back that suggest neither nerve root compromise nor a serious underlying condition

NOMENCLATURE FOR SPINE PATHOLOGY OUTSIDE THE DISC

Vertebral body marrow changes:

Associated with degenerative or inflammatory changes. Modic’s classification⁷ of MRI characteristics is shown in Table 18-3.

Kyphosis: Focal kyphosis across a fracture is measured using the Cobb angle: the angle between two lines, one drawn parallel to the superior endplate of the VB one level above the injured VB, the other line parallel to the inferior endplate of the VB one level below (showed the best overall intraobserver and interobserver reliability⁸).

Scoliosis: Named for the side of the convexity of the curvature (i.e. dextroscoliosis = convex to right, levoscoliosis = convex to left). Cobb angle: may also be used (using the superior endplate of the uppermost vertebra involved and the inferior endplate of the lowest vertebra involved). To make the lines meet within the borders of the film, it is some-times helpful to draw perpendiculars to these lines and measure the angles between those (see Figure 18-1).

DISABILITY, PAIN AND OUTCOME DETERMINATIONS

Disability scales for low back pain have been developed to assess outcomes for research purposes. Some widely used measures include:

1. visual analogue scale: used for any type of pain. The patient is asked to mark their pain level on a line divided into segments with sequential labels 0 (no pain) to 10 (the worst pain)

2. Oswestry disability index (ODI)⁹: a categorical ordinal scale that is used for low back pain. There are 4 English versions in wide use¹⁰, version 2.011 is recommended¹⁰. It consists 10 questions related to activities of daily living. Each item is scored 0-5 (5 being the most disability) and the total is multiplied by 2% to obtain the final score (range: 0-100%). The interpretation of the final score is shown in Table 18-4. A score > 45% is essentially completely disabled. A score in the teens is very functional

3. Short Form 36 (SF36)¹³

4. Roland–Morris disability questionnaire¹²

Figure 18-1 Cobb angle AP schematic showing levoscoliosis with Cobb angle?

Table 18-4 Oswestry disability index score

Score	Interpretation
0-20%	minimal disability: can cope with most daily activities
21-40%	moderate disability: pain and difficulty with sitting, lifting & standing. The patient may be disabled from work
41-60%	severe disability: pain is the main problem, but other areas are affected
61-80%	crippled: back pain impinges on all aspects of the patient’s life
81-100%	these patients are either bed-bound or else are exaggerating their symptoms

DIFFERENTIAL DIAGNOSIS

The differential diagnosis of low back pain (see Low back pain, page 1192) overlaps with that of myelopathy. In ≈ 85% of cases of LBP no specific diagnosis can be made¹⁴.

INITIAL ASSESSMENT OF THE PATIENT WITH BACK PAIN

Initial assessment consists of a history and physical exam focused on identifying serious underlying conditions such as: fracture, tumor, infection or cauda equina syndrome. Serious conditions presenting as low back problems are relatively rare.

HISTORY

The following information has been found to be helpful in identifying patients with serious underlying conditions such as cancer and spinal infection¹. Table 18-5 shows the sensitivity and specificity.

1. age

2. history of cancer (especially malignancies that are prone to skeletal metastases: prostate, breast, kidney, thyroid, lung)

3. unexplained weight loss

4. immunosuppression: from steroids, organ transplant medication, or HIV

5. prolonged use of steroids

6. duration of symptoms

7. responsiveness to previous therapy

8. pain that is worse at rest

9. history of skin infection: especially furuncle

10. history of IV drug abuse

11. UTI or other infection

12. pain radiating below the knee

13. persistent numbness or weakness in the legs

14. history of significant trauma. In a young patient: MVA, a fall from a height, or a direct blow to the back. In an older patient: minor falls, heavy lifting or even severe coughing can cause a fracture especially in the presence of osteoporosis

15. findings consistent with cauda equina syndrome (see page 446):

A. bladder dysfunction (usually urinary retention, or over-flow incontinence) or fecal incontinence

B. saddle anesthesia: see page 446

C. unilateral or bilateral leg weakness or pain

16. psychological and socioeconomic factors may influence the patient’s report of symptoms (also see page 436), and one should inquire about:

A. work status

B. typical job tasks

C. educational level

D. pending litigation

E. worker’s compensation or disability issues

F. failed previous treatments

G. substance abuse

H. depression

PHYSICAL EXAMINATION

Less helpful than the history in identifying patients who may be harboring conditions such as cancer, but may be more helpful in detecting spinal infections.

1. spinal infection (see page 376): findings that suggest this as a possibility (but are also common in patients without infection)

A. fever: common in epidural abscess and vertebral osteomyelitis, less common in discitis

B. vertebral tenderness

C. very limited range of spinal motion

2. findings of possible neurologic compromise: the following physical findings will identify most cases of clinically significant nerve root compromise due to L4-5 or L5-S1 HLD which comprise > 90% of cases of radiculopathy due to HLD (limiting the exam to the following might not detect the much less common upper lumbar disc herniations, which may be difficult to detect on PE, see page 453)

A. dorsiflexion strength of ankle and great toe: weakness suggests L5 and some L4 dysfunction

B. achilles reflex: diminished reflex suggests S1 root dysfunction

C. light touch sensation of the foot:

1. diminished over medial malleolus and medial foot: suggests L4

2. diminished over dorsum of foot: suggests L5

3. diminished over lateral malleolus and lateral foot: suggests S1

D. straight leg raising (SLR) (also check for crossed SLR): see page 443

“RED FLAGS” IN THE HISTORY AND PHYSICAL EXAM FOR LOW BACK PROBLEMS

Based upon the above history and physical exam, the findings in Table 18-6 would suggest the possibility of a serious underlying condition as the cause of the low back problem. Also, thoracic region pain is relatively uncommon and should raise the index of suspicion.

FURTHER EVALUATION

For over 95% of patients with acute low back problems, no further testing within the first 4 weeks of symptoms is required¹.

In the absence of any of the “red flag” conditions shown in Table 18-6, no further testing is recommended (even for patients suspected of having a HLD) and the treatment is similar for most patients with an acute episode of low back problems.

Simple laboratory tests including CBC and ESR are sufficiently efficacious and inexpensive that they should be obtained when there is a suspicion of back related tumor or infection.

Table 18-6 “Red flags” for patients with low back problems

Condition	Red flags
cancer or infection	1. age > 50 or < 20 yrs 2. history of cancer 3. unexplained weight loss 4. immunosuppression (see text) 5. UTI, IV drug abuse, fever or chills 6. back pain not improved with rest
spinal fracture	1. history of significant trauma (see text) 2. prolonged use of steroids 3. age > 70 yrs
cauda equina syndrome or severe neurologic compromise	1. acute onset of urinary retention or overflow incontinence 2. fecal incontinence or loss of anal sphincter tone 3. saddle anesthesia 4. global or progressive weakness in the LEs

FURTHER EVALUATION OF PATIENTS WITH LOW BACK PROBLEMS

Except for those exhibiting “red flags” (see above), special diagnostic tests are usually not needed during the first month of symptoms since it is not possible to predict which patients will improve (as most do) and which will not.

TESTS FOR EVIDENCE OF PHYSIOLOGIC DYSFUNCTION

Electrodiagnostics for low back problems: If the diagnosis of radiculopathy seems likely on clinical grounds, electrophysiologic testing is not recommended¹.

1. needle EMG: (see page 269) can assess acute and chronic nerve root dysfunction, myelopathy and myopathy, and may be useful for patients with suspicion of other conditions (e.g. neuropathy) or when a reliable strength exam is not possible. Reduced recruitment may be seen within the first several days of onset, however, spontaneous activity (see page 270) takes 10-21 days to develop ( less helpful in the first ≈ 3 weeks). Also, not usually helpful with normal muscle strength exam. Accuracy is highly operator dependent and improves with knowledge about imaging studies and clinical information¹⁵. For findings in radiculopathy, see page 270

2. H-reflex (see page 270): measures sensory conduction through nerve roots. Correlates with achilles reflex. Use is limited to assessing S1 radiculopathy¹⁶

3. SSEPs: (see page 266) assesses afferent fibers which travel in peripheral nerve and the posterior column of the spinal cord. May be abnormal in conditions affecting the dorsal columns with impaired joint position and proprioception (e.g. cervical spondylotic spinal myelopathy)

4. nerve conduction studies (including NCVs): helps identify acute and chronic entrapment neuropathies that may mimic radiculopathy

5. ✖ not recommended for assessing acute low back problems¹

A. F-wave response (see page 270): measures motor conduction through nerve roots, used to assess proximal neuropathies

B. surface EMG: assesses acute and chronic recruitment patterns during static or dynamic tasks using surface (instead of needle) electrodes

Bone scan for low back problems: Description: injection of a - radiolabeled compound (usually technetium-99m) that is taken up by metabolically active bone. A gamma camera localizes regions of uptake. Total radiation dose is ≈ to a set of lumbar spine x-rays¹. Contraindicated during pregnancy, and breast feeding must be briefly suspended following a bone scan due to presence of radiotracer in the breast milk.

A moderately sensitive test which may be used in evaluating low back pain when spinal tumor¹⁷, infection¹⁸, or occult fracture is suspected from “red flags” (see Table 18-6) on history or examination, or results of lab tests or plain x-rays. Not very specific, but may locate occult lesions and help differentiate these conditions from degenerative changes. A positive bone scan suggesting one of these conditions usually must be confirmed by other diagnostic tests or procedures (no studies have compared bone scans to CT or MRI).

Low yield in patients with longstanding low back problems and normal plain x-rays and laboratory tests (especially ESR)¹⁷.

SPECT scans may provide additional information to a bone scan.

Thermography for low back problems: Not recommended¹. Did not accurately predict absence or presence of nerve root compression seen at surgery¹⁹, and may be positive in a significant percentage of asymptomatic patients²⁰.

RADIOGRAPHIC EVALUATION

Diagnosing lumbar spinal stenosis or herniated intervertebral disc is usually helpful only in potential surgical candidates²¹. This includes patients with appropriate clinical syndromes who have not responded satisfactorily to adequate non-surgical treatment over a sufficient period of time, and who have no medical contraindications to surgery. Radiologic confirmation of these diagnoses usually requires CT, myelography, MRI, or some combination (see below). NB: myelography²², CT²³, or MRI²⁴ may also show bulging or herniated lumbar discs (HLD) or spinal stenosis in asymptomatic patients (e.g. 24% of asymptomatic patients have herniated discs on MRI and 4% have spinal stenosis; these numbers become 36% and 21% respectively in patients 60-80 years old)²⁵. Thus, these tests must be interpreted in light of clinical findings, and the anatomic level and side should correspond to the history, examination, and/or other physiologic data. Diagnostic radiology is of limited benefit as the initial evaluation in the majority of spinal disorders²⁶.

In the absence of red flags for serious conditions, imaging studies are not recommended in the first month of symptoms¹. For patients who have had previous back surgery, MRI with contrast is probably the best test. Myelography (with or without CT) is invasive and has increased risk of complications, and is therefore indicated only in situations where MRI cannot be done or is inadequate, and surgery is anticipated.

Σ

Patients for whom radiographic imaging is recommended are those with: • suspected benign conditions with symptoms of great enough severity to consider surgery persisting beyond 4 weeks, including: back related leg symptoms and clinically specific signs of nerve root compromise a history of neurogenic claudication (see page 477) or other finding suggestive of lumbar spinal stenosis • red flags: physical examination or other test results suggesting other serious conditions affecting the spine (e.g. cauda equina syndrome, fracture, infection, tumor, or other mass lesions or defects)

Recommendations for use of MRI and discography to select patients for fusion are shown in PRACTICE GUIDELINE 18-1.

PRACTICE GUIDELINE 18-1 MRI & DISCOGRAPHY FOR PATIENT SELECTION FOR LUMBAR FUSION*

Level II²⁷:

• MRI is recommended as the initial diagnostic test

• normal appearing discs on MRI should not be considered for discography or treatment

• lumbar discography should not be used as a stand-alone test

• to consider a disc level for treatment, if discography is used, there should be a concordant pain response^† and associated abnormalities on MRI^‡

Level III²⁷: discography should be reserved for equivocal MRI findings, especially at levels adjacent to unequivocally abnormal levels

* for recommendations on use of facet injections, see PRACTICE GUIDELINE 18-2, page 437

^† concordant pain response: pain identical or very similar to the patient’s usual pain complaints (NB: discography can produce severe LBP in patients with no prior complaints^{28, 29})

^‡ abnormal disc morphology on MRI: loss of T2WI signal intensity (“black disc”), disc space collapse, Modic changes (see page 430), and high-intensity zones (these findings also frequently occur in asymptomatic patients 30)

PLAIN LUMBOSACRAL X-RAYS

Unexpected findings occurred in only 1 in 2500 adults < 50 years age³¹. Diagnosis of surgical conditions of disc herniation and spinal stenosis cannot be made from plain films. Various congenital abnormalities of uncertain significance may be identified (e.g. spina bifida occulta), and evidence of degenerative changes (including osteophytes) are as frequent in symptomatic as in asymptomatic patients. Gonadal radiation is significant. Seldom indicated during pregnancy.

Recommendation

Not recommended for routine evaluation of patients with acute low back problems during the first month of symptoms unless a “red flag” is present (see below). Reserve LS x-rays for patients with a likelihood of having spinal malignancy, infection, inflammatory spondylitis, or clinically significant fracture. In these cases, plain x-rays are often just a starting point, and further study (CT, MRI…) may be indicated even if the plain x-rays are normal. “Red flags” for these conditions include the following:

1. age > 70 years, or < 20 yrs

2. systemically ill patients

3. temp > 100°F (or > 38° C)

4. history of malignancy

5. recent infection

6. patients with neurologic deficits suggesting possible cauda equina syndrome (saddle anesthesia, urinary incontinence or retention, LE weakness, see page 446)

7. heavy alcohol or IV drug abusers

8. diabetics

9. immunosuppressed patients (including prolonged treatment with corticosteroids)

10. recent urinary tract or spinal surgery

11. recent trauma: any age with significant trauma, or > 50 yrs old with mild trauma

12. unrelenting pain at rest

13. persistent pain for more than ≈ 4 weeks

14. unexplained weight loss

When spine x-rays are indicated, AP and lateral views are usually adequate³². Obliques and coned-down L5-S1 views more than double the radiation exposure, and add information in only 4-8% of cases³³, and can be obtained in specific instances where warranted (e.g. to diagnose spondylolysis when spondylolisthesis is found on the lateral film).

MRI

Unless contraindicated, MRI has supplanted CT and myelography for diagnosing most cases of disc herniation and spinal stenosis. Specificity and sensitivity for HLD are on the same order as CT/myelography, which is better than myelography alone^{1, 34, 35}.

Advantages:

1. provides information in sagittal plane (can easily evaluate cauda equina)

2. provides information regarding tissue outside of the spinal canal (e.g. extreme lateral disc herniation (see page 453), tumors…)

3. non-invasive and does not involve ionizing radiation

Disadvantages:

1. patients in severe pain or with claustrophobia may have difficulty holding still

2. dose not visualize bone well

3. poor for studying blood early (e.g. spinal epidural hematoma)

4. expensive (note: may be more cost effective than myelography if post-myelogram overnight hospitalization is avoided, and especially if a rare complication from myelography occurs)

5. interpretation with scoliosis is more difficult, may be partially compensated by contouring plane through which axial cuts are taken. Myelogram/CT may be superior

6. a number of contraindications: see Contraindications to MRI, page 130

Findings:

In addition to demonstrating herniated lumbar disc (HLD) outside of the disc interspace compressing nerve root or thecal sac, MRI can demonstrate signal changes within the interspace suggestive of disc degeneration³⁶ (loss of signal intensity on T2WI, loss of disc space height) and is useful in diagnosing infections and tumors.

LUMBOSACRAL CT

Not considered state of the art. If technically adequate images can be obtained (e.g. good quality scanner, images not obscured by artifact from patient movement or obesity), CT can demonstrate most spine pathology. For HLD, sensitivity is 80-95%, and specificity is 68-88%^{37, 38}. However, even some large disc herniations will be missed with plain CT. CT studies for HLD tend to be less satisfactory in the elderly. More utility with fractures.

Disc material has density (Hounsfield units) ≈ twice that of the thecal sac. Associated findings with herniated disc include:

1. loss of epidural fat (normally seen as low density in the anterolateral canal)

2. loss of normal “convexity” of thecal sac (indentation by herniated disc)

Advantages:

1. images soft tissue to a degree that may be adequate

2. excellent bony detail

3. non-invasive

4. outpatient evaluation

5. evaluates for extreme lateral disc herniation to some degree

6. evaluates paraspinal soft tissue (e.g. to rule out tumor, paraspinal abscess…)

7. advantages over MRI: faster scanning (significant in patients who have difficulty laying still for long time), less expensive, less claustrophobic, fewer contraindications (see Contraindications to MRI, page 130)

Disadvantages:

1. does not evaluate sagittal plane (may be partially ameliorated by eliminating skip regions and then utilizing computerized sagittal reconstructions)

2. evaluates only those levels that are scanned:

A. higher cuts must be taken through the conus medullaris to avoid missing occasional pathology there

B. performing cuts only through the disc spaces (a common practice) may miss pathology between the disc spaces

3. sensitivity is significantly lower than MRI or myelogram/CT

MYELOGRAPHY

With water soluble contrast, sensitivity (62-100%) and specificity (83-94%)^39-42 are similar to CT for detection of HLD. When combined with post-myelographic CT scan (myelogram/CT), the sensitivity and especially specificity increase⁴³. A herniated disk in the large space between thecal sac and posterior border of vertebral bodies at L5-S1 (insensitive space) may not be seen on myelography (CT or MRI are usually better at detecting this).

Advantages:

1. provides information in sagittal plane (unlike plain CT)

2. evaluates cauda equina (unlike routine CT)

3. provides “functional” information about degree of stenosis (a high-degree block will allow flow of dye only after certain position changes)

Disadvantages:

1. occasionally requires overnight hospitalization

2. may miss pathology outside of the dura (including far laterally herniated disc), sensitivity is improved with post-myelographic CT

3. invasive

A. drugs e.g. warfarin must be stopped, and sometimes converted to heparin

B. with occasional side effects (post LP H/A, N/V, rare seizures)

4. iodine allergic patients

A. requires iodine allergy prep

B. may still be risky (especially in severely iodine allergic patients)

Findings:

HLD produces extradural filling defect at the level of the intervertebral disc. Massive disc herniation or severe lumbar stenosis may produce a total or near total block. In some cases of HLD, the finding may be very subtle and may consist of a cut-off of the filling (with contrast) of the nerve root sleeve (compared to normal nerve(s) on contralateral side or at other levels). Another subtle finding may be a “dual shadow” on lateral view.

BONE SCAN

See page 432

DISCOGRAPHY

Injection of water-soluble contrast agent directly into the nucleus pulposus of the intervertebral disc being studied. Results of the test depend on volume of dye accepted into the disc, the pressure needed to inject the dye, the configuration of the dye (including leakage from the confines of the disc space) on radiographic imaging (plain x-rays produce the so-called “discogram”, CT scan may also be utilized), and reproduction of the patient’s pain on injection. Some of the basis for performing a discogram is to identify levels that may produce “discogenic pain” or “painful disc syndrome”, a controversial point (see PRACTICE GUIDELINE 18-1, page 434).

Critique:

Invasive. Interpretation is equivocal, and complications may occur (disc space infection, disc herniation, and significant radiation exposure with CT-discography). May be abnormal in asymptomatic patients^28,29 (as any of the above tests may be) although the false positive rate may not be quite this high⁴⁴. See PRACTICE GUIDELINE 18-1, page 434 for recommendations.

PSYCHOSOCIAL FACTORS

Although some patients with chronic LPB (> 3 months duration) may have started off with a diagnosable condition, psychological and socioeconomic factors (such as depression, secondary gain…) may come to play a significant role in perpetuating or amplifying pain. Psychological factors, especially elevated hysteria or hypochondriasis scales on the Minnesota Multiphasic Personality Inventory (MMPI) were found to be a better predictor of outcome than findings on radiographic imaging in one study¹⁵. A screening scale of 5 factors has been proposed⁴⁵ (positive findings in any 3 suggests psychological distress):

1. pain on simulated axial loading: press on top of head

2. inconsistent performance: e.g. difficulty tolerating straight leg raising (SLR) while supine, but no difficulty when sitting

3. overreaction during the physical exam

4. inappropriate tenderness that is superficial or widespread } these two items may not be reliable, the others are potentially reliable⁴⁶

5. motor or sensory abnormalities not corresponding to anatomic confines } these two items may not be reliable, the others are potentially reliable⁴⁶

However, the usefulness of this information is limited, and no effective interventions have been identified to address these factors. Therefore the AHCPR panel was unable to recommend specific assessment tools or interventions¹.

TREATMENT

An initial period of nonsurgical management (see “Conservative” treatment below) is indicated except in the following circumstances where urgent surgery is indicated: symptoms of a cauda equina syndrome (urinary retention, saddle anesthesia…, see Cauda equina syndrome, page 446), progressive neurologic deficit, or profound motor weakness. A relative indication for proceeding to surgery without conservative management is severe pain that cannot be controlled with adequate pain medication (rare).

If specific diagnoses such as herniated intervertebral lumbar disc or symptomatic lumbar stenosis are made, surgical treatment for these conditions may be considered if the patient fails to improve satisfactorily. In cases where no specific diagnosis can be made, management consists of conservative treatment and following the patient to rule out the possible development of symptoms suggestive of a more serious diagnosis that may not have initially been evident.

“CONSERVATIVE” TREATMENT

This term has regrettably come to be used for non-surgical management. With slight modification, similar approaches can be used for mechanical low back pain, as well as for acute radiculopathy from disc herniation.

PRACTICE GUIDELINE 18-2 INJECTION THERAPY FOR LOW-BACK PAIN

Therapeutic recommendations

Level III⁴⁷: lumbar epidural injections or trigger point injections are not recommended for long-term relief of chronic LBP. These techniques or facet injections may be used to provide temporary relief in select patients

Diagnostic recommendations

Level III⁴⁷: lumbar facet injections

• may predict the response to radiofrequency facet ablation

• ✖ not recommended as a diagnostic tool to predict the response to lumbar fusion

Recommendations (based on AHCPR findings¹ in the absence of “red flags”^A):

A. some key literature citations are given here, primarily those from the better studies that support the Agency for Health Care Policy and Research (AHCPR) panel recommendations. However, refer to Bigos et al.¹ for full analysis and list of references

1. activity modifications: no studies were found that met the panels review criteria for adequate evidence. However, the following information was felt to be useful:

A. bed rest: for 2-3 days maximum

1. the theoretical objective is to reduce symptoms by reducing pressure on the nerve roots and/or intradiscal pressures which is lowest in the supine semi-Fowler position⁴⁸, and also to reduce movements which are experienced as painful by the patient

2. deactivation from prolonged bed rest (> 4 days) appears to be worse for patients (producing weakness, stiffness, and increased pain) than a gradual return to normal activities⁴⁹

3. recommendations: the majority of patients with low back problems will not require bed rest. Bed rest for 2-4 days may be an option for those with severe initial radicular symptoms, however, this may be no better than watchful waiting⁵⁰ and may be harmful⁵¹

B. activity modification

1. the goal is to achieve a tolerable level of discomfort while continuing sufficient physical activity to minimize disruption of daily activities

2. risk factors: although there is not agreement on their exact role, the following were identified as having an increased incidence of low back problems. Jobs requiring heavy or repetitive lifting, total body vibration (from vehicles or industrial machinery), asymmetric postures, or postures sustained for long periods (including prolonged sitting)

3. recommendations: temporarily limit heavy lifting, prolonged sitting, and bending or twisting of the back. Establish activity goals to help focus attention on expected return to full functional status

C. exercise (may be part of a physical therapy program):

1. during the 1st month of symptoms, low-stress aerobic exercise can minimize debility due to inactivity. In the first 2 weeks, utilize exercises that minimally stress the back: walking, bicycling, or swimming

2. conditioning exercises for trunk muscles (especially back extensors, and possibly abdominal muscles) are helpful if symptoms persist (during the first 2 weeks, these exercises may aggravate symptoms)

3. there is no evidence to support stretching of back muscles, or to recommend back-specific exercise machines over traditional exercise

4. recommended exercise quotas that are gradually escalated results in better outcome than having patients simply stop when pain occurs⁵²

2. analgesics:

A. for the initial short-term period, acetaminophen (APAP) or NSAIDs (see page 44) may be used. In one study 53 of acute LBP, NSAIDs did not add any benefit to APAP + standard education (see below)

B. stronger analgesics (mostly opioids, see page 46) may be required for severe pain, usually severe radicular pain. For non-specific back pain, there was no earlier return to full activity than with NSAIDs or APAP1. Opioids should not be used > 2-3 weeks, at which time NSAIDs should be instituted

3. muscle relaxants (see page 50)

A. muscle spasms have not been proven to cause pain, and the most commonly used muscle relaxants have no peripheral effect on muscle spasm

B. probably more effective than placebo, but have not been shown to be more effective than NSAIDs

C. potential for side effects: drowsiness (in up to 30%). Most manufacturers recommend use for < 2-3 weeks. Agents such as chlorzoxazone (Parafon Forte® and others) may be associated with risk of serious and potentially fatal hepatotoxicity⁵⁴

4. education: (may be provided as part of a physical therapy program)

A. explanation of the condition to the patient⁵⁵ in understandable terms, and positive reassurance that the condition will almost certainly subside⁵⁶ have been shown to be more effective than many other forms of treatment

B. proper posture, sleeping positions, lifting techniques… should be conveyed to the patient. Formal “back school” seems to be marginally effective⁵⁷

5. spinal manipulation therapy (SMT): defined as manual therapy in which loads are applied to the spine using long or short lever methods with the selected joint being taken to its end range of voluntary motion, followed by application of an impulse loading (may be part of a physical therapy program)

A. may be helpful for patients with acute low back problems without radiculopathy when used in the first month of symptoms (efficacy after 1 month is unproven) for a period not to exceed 1 month. One study⁵³ found no added benefit to APA + standard education

B. insufficient evidence to recommend SMT in the presence of radiculopathy

C. SMT should not be used in the face of severe or progressive neurologic deficit until serious conditions have been ruled out

D. ✖ reports of arterial dissection (especially vertebral artery - see page 985) and CVA, myelopathy & subdural hematoma with cervical SMT and cauda equina syndrome with lumbar SMT^58-60 and the uncertainty of benefits have led to the questioning of the use of SMT⁵⁸ (especially cervical)

6. epidural injections:

A. epidural steroid injections (ESI): there is no evidence that this is effective in treating acute radiculopathy⁶¹. Prospective studies yield varied results⁶². Some improvement at 3 & 6 weeks may occur (but no functional benefit, and no change in the need for surgery), with no benefit at 3 months⁶³. The response in chronic back pain is poor in comparison to acute pain. ESI may be an option for short-term relief of radicular pain when control on oral medications is inadequate or for patients who are not surgical candidates

B. there is no evidence to support the use of epidural injections of steroids, local anesthetics and/or opioids for LBP without radiculopathy

C. efficacy with conditions such as lumbar spinal stenosis are conflicting 62

✖ Not recommended by the AHCPR panel¹ for treatment of acute low back problems in the absence of “red flags” (see Table 18-6, page 432):

1. medications

A. oral steroids: no difference was found at one week and 1 year after randomization to receive 1 week therapy with oral dexamethasone or placebo⁶⁴

B. colchicine: conflicting evidence shows either some⁶⁵ or no⁶⁶ therapeutic benefit. Side effects of N/V and diarrhea¹

C. antidepressant medications: most studies of these medications were for chronic back pain. Some methodologically flawed studies failed to show benefits when compared to placebo for chronic (not acute) LBP⁶⁷

2. physical treatments

A. TENS (transcutaneous electrical nerve stimulation): not statistically significantly better than placebo, and added no benefit to exercise alone⁶⁸

B. traction (including pelvic traction): not demonstrated to be effective⁶⁹

C. physical agents and modalities: including heat (including diathermy), ice, ultrasound. Benefit is insufficiently proven, however, self-administered home programs for application of heat or cold may be considered. Ultrasound and diathermy should not be used in pregnancy

D. lumbar corsets and support belts: not proven beneficial for acute back problems. Prophylactic use has been advocated, but this is controversial⁷⁰

E. biofeedback: has not been studied for acute back problems. Primarily advocated for chronic LBP, where effectiveness is controversial⁷¹

3. injection therapy

A. trigger point and ligamentous injections: the theory that trigger points cause or perpetuate LBP is controversial and disputed by many experts. Injections of local anesthetic are of equivocal efficacy

B. (zygapophyseal) facet joint injections: theoretical basis is that there exists a “facet syndrome” producing LBP which is aggravated by spine extension, with no nerve root tension signs (see page 443). No studies have adequately investigated injections for pain < 3 months duration. For chronic LBP, neither the agent nor the location (intrafacet or pericapsular) made a significant difference in outcomes^{72, 73}

C. epidural injections in the absence of radiculopathy: see above

D. acupuncture: no studies were found that evaluated the use in acute back problems. All randomized clinical trials found were for patients with chronic LBP, and even the best studies were felt to be mediocre and contradictory. Meta-analysis found acupuncture was more effective in relieving chronic LBP than sham or no treatment⁷⁴, but there was no comparison to other therapies

SURGICAL TREATMENT

Indications for surgery for herniated lumbar disc:

See page 445.

Indications for fusion for chronic LBP without stenosis or spondylolisthesis:

Very controversial. Guidelines are shown in PRACTICE GUIDELINE 18-3.

PRACTICE GUIDELINE 18-3 LUMBAR FUSION FOR LBP WITHOUT STENOSIS OR SPONDYLOLISTHESIS

Level I⁷⁵: lumbar fusion is recommended for carefully selected patients* with disabling LBP due to one- or two-level degenerative disease without stenosis or spondylolisthesis

Level III^{75, 77}: an intensive course of PT and cognitive therapy is recommended as a option for patients with LBP in whom conventional medical management has failed

* in the primary quoted study 76 patients had chronic LBP for ≥ 2 years and had radiologic evidence of disc degeneration at L4-L5, L5-S1, or both, and had failed best medical management

PRACTICE GUIDELINE 18-4 CHOICE OF FUSION TECHNIQUE

Level II⁷⁸: for ALIF or ALIF + instrumentation, the addition of a posterolateral fusion is not recommended*

Level III⁷⁸:

• either a posterolateral fusion or an interbody fusion (PLIF, TLIF or ALIF) are options for patients with LBP due to DDD at 1 or 2 levels

• an interbody graft is an option to improve fusion rates and functional outcome^†

✖ the use of multiple approaches (anterior + posterior) is not recommended as a routine option for LBP without deformity

* the demonstrated benefit does not outweigh the additional time and blood loss involved

^† caution: the improvement in fusion rate and outcome is marginal, and interbody fusion is associated with an increased complication rate, especially with combined approaches (e.g. 360° fusion)

TYPE OF SURGICAL TREATMENT

The type of surgical procedure chosen is tailored to the specific condition identified. Examples are shown in Table 18-7. Discussion of some options is also provided below.

Lumbar spinal fusion

Although there is no consensus on the indications⁷⁹, lumbar spinal fusion (LSF) is accepted treatment for fracture/dislocation or instability resulting from tumor or infection.

For degenerative spine disease, practice parameters have been developed and are included herein. Pain associated with Modic type 1 changes (see page 430) may respond to stabilization procedures, the other types do not exhibit this association.

Table 18-7 Surgical options for low back problems

Condition	Surgical treatment options
“routine” HLD	• standard discectomy and microdiscectomy are of similar efficacy • intradiscal procedures: nucleotome, laser disc decompression. Not recommended (see page 447) • chymopapain: acceptable, but less efficacious than above. Significant risk of anaphylaxis (see page 447). Use has been largely abandoned
foraminal or far lateral HLD	• partial or total facetectomy (see page 454) • extracanal approach (see page 454) • endoscopic techniques
lumbar spinal stenosis	• simple decompressive laminectomy • laminectomy plus fusion: may be indicated for patients with degenerative spondylolisthesis, stenosis and radiculopathy

PRACTICE GUIDELINE 18-5 LUMBAR FUSION FOR DISC HERNIATION

Level III⁸⁰:

• lumbar fusion is not routinely recommended following disc excision in patients with HLD or recurrent HLD causing radiculopathy

• lumbar fusion is a potential adjunct to disc excision in cases of a HLD or recurrent HLD:

with evidence of preoperative lumbar spinal deformity or instability

in patients with chronic axial LBP associated with radiculopathy

Instrumentation as an adjunct to fusion

PRACTICE GUIDELINE 18-6 PEDICLE SCREW FIXATION

Level III⁸¹: pedicle screw fixation is recommended as a treatment option for patients with LBP treated with posterolateral fusion who are at high risk for fusion failure*

* routine use of pedicle screws is discouraged because of conflicting evidence of benefit, together with considerable evidence of increased cost and complications

The use of instrumentation increases the fusion rate⁸². Hardware used in the absence of fusion will eventually fatigue, especially in the region of the lumbar lordosis. Therefore, instrumentation must be viewed as a temporary internal stabilizing measure while awaiting the fusion process to complete.

CHRONIC LOW BACK PAIN

Rarely can an anatomic diagnosis be made in patients with chronic LBP ≥ 3 months⁸³. Also, see Psychosocial factors, page 436. Patients with chronic pain syndromes (CPS) refer to their problems with affective or emotional terms with a higher frequency than those with acute pain⁸⁴. The amount of time that a patient has been out of work due to low back problems is related to the chances of the patient getting back to work as shown in Table 18-8.

Table 18-8 Chances of patients going back to work

Time out of work	Chances of getting back to work
< 6 mos	50%
1 yr	20%
2 yrs	< 5%

18.1.1. Post-op clinic visits - lumbar fusion

Patients are seen in the clinic at intervals depending on the preference of the surgeon. A typical follow-up schedule with studies routinely performed is shown in Table 18-9. For specific problems, additional investigations are usually needed.

Post-op x-rays: Items to check on post op x-rays include:

1. alignment

2. position of grafts if used (e.g. inter-body grafts)

3. integrity of hardware (look for screw or rod breakage, screw pullout, rod disconnection)

4. lucencies around screws which may indicate motion and implies non-union

5. any evidence of fusion (may be difficult, e.g. with synthetic interbody fusions)

6. flexion/extension films: look for motion across fused segments (sometimes absence of motion is the only evidence of fusion on plain x-rays) and the development of abnormal motion at adjacent segments

Table 18-9 Sample post-op lumbar fusion clinic visit schedule

Time post-op	Agenda
7-10 d	wound check, D/C sutures/staples if used
6 wks	AP & lateral LS-spine x-ray in brace
10-12 wks	• AP & lateral LS-spine x-rays with flexion/extionsion views out of brace • if x-rays look good and patient is doing well, begin weaning brace
6 months	• AP & lateral LS-spine x-rays with flexion/extension views • some surgeons release patients at this time if they are doing well
1 year (optional)	• AP & lateral LS-spine x-rays with flexion/extension views • release patient if they are doing well

18.2. Sagittal balance

The normal curvature of the spine in the sagittal plane (cervical lordosis, thoracic kyphosis and lumbar lordosis) permits standing posture with the minimum of muscle activity and soft tissue deformity. Abnormalities in any component of this sagittal balance (SB) result in compensatory changes in other segments.

Hardacker et al⁸⁵ proposes the following (angles measured using Cobb’s technique (see page 430)):

1. cervical lordosis: normally 40° ± 9.7°

2. the occipital cervical junction is actually in kyphosis

3. the majority of lordosis occurs between C1-C2

4. the lower cervical spine (C4-C7) is relatively flat with an average lordosis of 6°

5. thoracic kyphosis: normally 20-50°. Increases with age (especially significant in age > 40 yrs) and the rate of increase is higher in females⁸⁶ (see Table 18-10 for normal values)

6. lumbar lordosis: normally 31-79°

To assess sagittal balance, on a standing lateral full-spine xray: a plumb line is drawn straight down from the center of the C7 VB (see Figure 18-2). There are various definitions for where the line should cross for normal sagittal balance, including:

1. the plumb line should pass within the L5-S1 disc space or within 2 cm of the sacral promontory and through or behind the axis of the hip joint

2. the mean sagittal vertical axis fell 3.2 ± 3.2 cm behind the front of the sacrum in asymptomatic patients > 40 years old⁸⁷

A plumb line can also be drawn from the center of C2 down, and this line should pass through the C7 VB and then on to the same lumbosacral targets as above.

Patients with compensated sagittal balance can temporarily correct for an abnormal plumb line using back muscles and pelvic tilt which leads to pain and fatigue.

Figure 18-2 Sagittal balance plumb line Lateral spine schematic

Etiologies of loss of sagittal balance

1. congenital: Scheuermann’s thoracic hyperkyphosis

2. aging: cervical lordosis and thoracic kyphosis increase, lumbar lordosis decreases

3. traumatic

4. iatrogenic: post-op flat-back syndrome

5. pathologic: osteoporosis

6. neuromuscular disorders

Clinical features:

1. pain

2. deformity

3. difficulty ambulating

Table 18-10 Normal values for thoracic kyphosis⁸⁶*

Age (yrs)	Kyphosis (± SD)
	Males	Females
< 10	20° (± 7.85)	24° (± 6.67)
10-19	25° (± 8.16)	26° (± 7.43)
40-49	30° (± 6.93)	33° (± 6.72)
50-59	33° (± 6.46)	41° (± 9.88)
70-79	41° (± 7.57)	42° (± 9.00)

* measured using a modified Cobb technique in asymptomatic patients

Flat back syndrome: AKA lumbar degenerative kyphosis. More common in Asia than Western countries. May also occur following lumbar fusion. Management: controversial. Restoring lost sagittal balance may cause improvement of symptoms⁸⁸.

Surgical management: for flexible deformity, the spine may be fused after sagittal balance is restored either manually or posturally. For fixed deformities, options to increase lumbar lordosis:

1. Smith-Peterson (facet) osteotomy: shortens posterior elements and lengthens anterior elements. Anterior elements may be lengthened using interbody grafts placed from an anterior approach

2. pedicle subtraction osteotomy: decreases pedicle and posterior VB. Does not produce an anterior defect

18.3. Intervertebral disc herniation

18.3.1. Lumbar disc herniation

Key concepts:

• typical disc herniation → radiculopathy in the nerve exiting at the level below

• massive disc herniations can → cauda equina syndrome (a medical emergency). Typical symptoms: saddle anesthesia, urinary retention, LE weakness (page 446)

• most patients do as well with conservative treatment as with surgery, initial conservative treatment should be considered for the vast majority

• surgery indications: cauda equina syndrome, progressive symptoms or neurologic deficits despite conservative treatment, or severe radicular pain > ≈ 6 weeks

PATHOPHYSIOLOGY

Intervertebral discs may undergo degenerative changes (see Table 18-1, page 429 for description) which increases the risk of herniation (see same table for definition).

Variants:

1. intravertebral disc herniation (Schmorl’s node): see page 455

2. intradural disc herniation: see page 455

3. limbus fracture: traumatic separation of a segment of bone from the edge of the vertebral ring apophysis at the site of anular attachment. May accompany HLD

CLINICAL ASPECTS

The posterior longitudinal ligament is strongest in the midline, and the posterolateral annulus may bear a disproportionate portion of the load. This may explain why most herniated lumbar discs (HLD) occur posteriorly, slightly off to one side, characteristically compressing the nerve root en passage → severe radicular pain.

Characteristic findings on the history often include:

1. symptoms may start off with back pain, which after days or weeks gradually or sometimes suddenly yields to radicular pain often with reduction of the back pain

2. precipitating factors: various factors are often blamed, but are rarely identified⁸⁹ with certainty

3. pain relief upon flexing the knee and thigh

4. patients generally avoid excessive movements, however, remaining in any one position (sitting, standing, or lying) too long may also exacerbate the pain, sometimes necessitating position changes at intervals that range from every few minutes to 10-20 minutes. This is distinct from constant writhing in pain e.g. with ureteral obstruction

5. “cough effect”: ↑ pain with coughing, sneezing, or straining at the stool. Occurred in 87% of patients with HLD in one series⁹⁰

6. bladder symptoms: the incidence of voiding dysfunction is 1-18%^{91 (p 966)}. Most common: difficulty voiding, straining, or urinary retention. Reduced bladder sensation may be the earliest finding. Later it is not unusual to see “irritative” symptoms including urinary urgency, frequency (including nocturia), increased post-void residual. Less common: enuresis, and dribbling incontinence⁹² (NB: frank urinary retention may indicate cauda equina syndrome, see page 446). Occasionally a HLD may present only with bladder symptoms which may improve after surgery⁹³. Discectomy may improve bladder function, but this cannot be assured

PHYSICAL FINDINGS IN RADICULOPATHY

Back pain per se is usually a minor component (only 1% of patients with acute low back pain have sciatica⁴), and when it is the only presenting symptom, other causes should be sought (see Low back pain, page 1192). Sciatica has such a high sensitivity for disc herniation, that the likelihood of a clinically significant disc herniation⁹⁴ in the absence of sciatica is ≈ 1 in 1000. Exceptions include a central disc herniation which may cause symptoms of lumbar stenosis (i.e. neurogenic claudication) or a cauda equina syndrome.

Nerve root impingement gives rise to a set of signs and symptoms present to variable degrees. Characteristic syndromes are described for the most common nerve roots involved (see Nerve root syndromesbelow).

In a series of patients referred to neurosurgical outpatient clinics for radiating leg pain, 28% had motor loss (yet only 12% listed motor weakness as a presenting complaint), 45% had sensory disturbance, and 51% had reflex changes⁹⁵.

Findings suggestive of nerve root impingement include the following. Table 18-11 shows the sensitivity and specificity of some findings on the exam among patients with sciatica.

1. signs/symptoms of radiculopathy (see Table 18-13, page 445)

A. pain radiating down LE

B. motor weakness

C. dermatomal sensory changes

D. reflex changes: mental factors may influence symmetry⁹⁶

2. positive nerve root tension sign(s): including Lasègue’s sign (see below)

3. tenderness over the sciatic notch

Nerve root tension signs

Includes⁹⁷:

1. Lasègue’s sign: AKA straight leg raising (SLR) test. Helps differentiate sciatica from pain due to hip pathology. Test: with patient supine, raise afflicted limb by the ankle until pain is elicited⁹⁸ (should occur at < 60°, tension in nerve increases little above this angle). A positive test consists of leg pain or paresthesias in the distribution of pain (back pain alone does not qualify). The patient may also extend the hip (by lifting it off table) to reduce the angle. Although not part of Lasègue’s sign, ankle dorsiflexion with SLR usually augments pain due to nerve root compression. SLR primarily tenses L5 and S1, L4 less so, and more proximal roots very little. Nerve-root compression produces a positive Lasègue’s sign in ≈ 83% of cases⁹⁰ (more likely to be positive in patients < 30 yrs age with HLD⁹⁹). May be positive in lumbosacral plexopathy (see page 795). Note: flexing both thighs with the knees extended (“long-sitting” or sitting knee extension) may be tolerated further than flexing the single symptomatic side alone

2. Cram test: with patient supine, raise the symptomatic leg with the knee slightly flexed. Then, extend the knee. Results similar to SLR

3. crossed straight leg-raising test AKA Fajersztajn’s sign: SLR on the painless leg causes contralateral limb pain (a greater degree of elevation is usually required than the painful side). More specific but less sensitive than SLR (97% of patients undergoing surgery with this sign have confirmed HLD¹⁰⁰). May correlate with a more central disc herniation

4. femoral stretch test¹⁰¹, AKA reverse straight leg raising: patient prone, examiner’s palm at popliteal fossa, knee is maximally dorsiflexed. Often positive with L2, L3, or L4 nerve root compression (e.g. in upper lumbar disc herniation), or with extreme lateral lumbar disc herniation (may also be positive in diabetic femoral neuropathy or psoas hematoma); in these situations SLR (Lasègue’s sign) is frequently negative (since L5 & S1 not involved)

5. “bowstring sign”: once pain occurs with SLR, lower the foot to the bed by flexing knee, keeping the hip flexed. Sciatic pain ceases with this maneuver, but hip pain persists

6. sitting knee extension test: with patient seated and both hips and knees flexed 90°, slowly extend one knee. Stretches nerve roots as much as a moderate degree of SLR

Other signs useful in evaluation for lumbar radiculopathy

1. FABER: an acronym for Flexion ABduction External-Rotation, AKA FABERE test (the trailing “e” is for extension), AKA Patrick’s-test. A test of hip motion. Method: the hip and knee are flexed and the lateral malleolus is placed on the contralateral knee. The ipsilateral knee is gently displaced downward towards the exam table. This stresses the hip joint and does not usually exacerbate true nerve-root compression. Often markedly positive in the presence of hip joint disease (e.g. trochanteric bursitis, see page 478), sacroiliitis or mechanical low-back pain

2. Trendelenburg sign: examiner observes pelvis from behind while patient raises one leg while standing. Normally the pelvis remains horizontal. A positive sign occurs when the pelvis tilts down toward the side of the lifted leg indicating weakness of the contralateral thigh adductors (primarily L5 innervated)

3. crossed adductors: in eliciting patellar reflex (knee jerk (KJ)), the contralateral thigh adductors contract. In the presence of a hyperactive ipsilateral KJ it may indicate an upper motor neuron lesion, in the presence of a hypoactive ipsilateral KJ it may be a form of pathological spread, indicating nerve root irritability

4. Hoover sign 102: to distinguish unilateral functional weakness of iliopsoas from organic weakness using synergistic contraction of the contralateral gluteus medius. The supine patient is asked to lift one leg off the bed against resistance from the examiner’s hand. The examiner simultaneously places the palm of his/her other hand under the heel of the unlifted leg and gently lifts. Test 1: when the patient lifts the normal leg, if the paretic leg pushes down with more force than was exhibited on manual testing of the limb beforehand, the weakness is judged functional, if the force is equally weak the weakness is judged organic. Test 1 cannot be used if the hip extensor was normal beforehand. Test 2: (the better known test) the patient is asked to lift the weak leg. If the heel on the normal side lifts passively by the examiner, it suggests the weakness is functional (i.e. the patient is not trying). Not totally reliable^{103, 104}

5. abductor sign: an alternative to the Hoover test, to differentiate functional from organic weakness in the thigh abductors using synergistic contraction of the contralateral thigh abductors¹⁰⁴. With the patient supine, the examiner places a hand on the lateral aspect of both legs. The patient is asked to abduct legs. The patient is asked to abduct one leg, and then the other while the examiner applies resistance with his/her hand. The examiner mentally notes the response of the non-abducting LE. The results are as noted in Table 18-12

Table 18-12 Abductor sign

Abducting LE	Contralateral (nonabducting) LE
	Organic weakness	Functional weakness
weak LE	maintains position	hyperadducts
normal LE	hyperadducts	maintains position

NERVE ROOT SYNDROMES

Due to the facts listed below, a herniated lumbar disc (HLD) usually spares the nerve root exiting at that interspace, and impinges on the nerve exiting from the neural foramen one level below (e.g. a L5-S1 HLD usually causes S1 radiculopathy). This gives rise to the characteristic lumbar nerve root syndromes shown in Table 18-13.

Important facts in lumbar disc disease:

1. in the lumbar region, the nerve root exits below and in close proximity to the pedicle of its like-numbered vertebra

2. the intervertebral disc space is located well below the pedicle

3. not all patients have 5 lumbar vertebrae (see Localizing levels in spine surgery, page 173)

RADIOGRAPHIC EVALUATION

See Radiographic evaluation on page 433 under Low back pain.

NONSURGICAL TREATMENT

For nonsurgical treatment, see “Conservative” treatment, page 437.

SURGICAL TREATMENT

INDICATIONS

No predictive factors have been identified that can determine which patients are likely to improve on their own and which would be better served with surgery.

Surgical indications in patients with a radiographically identified herniated disc that correlates with findings on the history and physical exam:

1. failure of non-surgical management to control pain after 5-8 weeks: over 85% of patients with acute disc herniation will improve without surgical intervention in an average of 6 weeks¹⁰⁶ (70% within 4 weeks¹⁰⁷). Most clinicians advocate waiting somewhere between 5-8 weeks from the onset of radiculopathy before considering surgery (assuming none of the items listed below applies)

2. “EMERGENT SURGERY”: (i.e. before the 5-8 weeks have lapsed). Indications:

A. cauda equina syndrome (CES): (see below)

B. progressive motor deficit (e.g. foot drop). NB: paresis of unknown duration is a doubtful indication for surgery^{89, 108, 109} (no study has documented that there is less motor deficit in surgically treated patients with this finding¹¹⁰). However, the acute development or progression of motor weakness is considered an indication for rapid surgical decompression

C. “urgent” surgery may be indicated for patients whose pain remains intolerable in spite of adequate narcotic pain medication

3. ± patients who do not want to invest the time in a trial of non-surgical treatment if it is possible that they will still require surgery at the end of the trial

Cauda equina syndrome

The clinical condition arising from dysfunction of multiple lumbar and sacral nerve roots within the lumbar spinal canal. Usually due to compression of the cauda equina. See Table 21-86, page 744 for features to help differentiate CES from a conus lesion.

Possible findings in CES:

1. sphincter disturbance:

A. urinary retention: the most consistent finding. Sensitivity ≈ 90% (at some point in time during course)^{111, 112}. To evaluate acutely: have patient empty bladder and check post-void residual (by catheterization or with bladder ultrasound). In a patient without retention, only 1 in 1000 will have a CES. Cystometrogram (when done) shows a hypotonic bladder with decreased sensation and increased capacity

B. urinary and/or fecal incontinence¹¹³: some patients with urinary retention will present with overflow incontinence

C. anal sphincter tone: diminished in 60-80%

2. “saddle anesthesia”: the most common sensory deficit. Distribution: region of the anus, lower genitals, perineum, over the buttocks, posterior-superior thighs. Sensitivity ≈ 75%. Once total perineal anesthesia develops, patients tend to have permanent bladder paralysis¹¹⁴

3. significant motor weakness: usually involves more than a single nerve root (if untreated, may progress to paraplegia)

4. low back pain and/or sciatica (sciatica is usually bilateral, but may be unilateral or entirely absent, prognosis may be worse when absent or bilateral¹¹²)

5. bilateral absence of Achilles reflex has been noted¹¹⁵

6. sexual dysfunction (usually not detected until a later time)

Etiologies of CES includes:

1. compression of cauda equina

A. massive herniated lumbar disc: see below

B. tumor

1. from compression: e.g. with metastatic disease to the spine with epidural extension

2. intravascular lymphomatosis (B-cell lymphoma): a circulating lymphoma without solid mass (see page 674). Often presents with CNS findings: dementia, enhancing meninges on MRI, lymphoma cells in CSF, and CES

C. free fat graft following discectomy¹¹⁶

D. trauma: fracture fragments compressing cauda equina

E. spinal epidural hematoma

2. infection: may cause neurologic deficit from

A. compression: typically from spinal epidural abscess complicating discitis or vertebral osteomyelitis

B. a significant number of cases of CES from infection may be due to vascular compromise resulting from local septic thrombophlebitis. This may carry a worse prognosis as surgical decompression cannot correct this mechanism

3. neuropathy:

A. ischemic

B. inflammatory

4. ankylosing spondylitis: etiology is often obscure (see page 502)

CES from HLD: May be due to massive herniated disc, usually midline, most common at L4-5, often superimposed on a preexisting condition (spinal stenosis, tethered cord…)¹¹³.

Prevalence of CES:

1. 0.0004 in all patients with LBP⁹⁴

2. only ≈ 1-2% of HLD that come to surgery⁹⁴

Time course: CES tends to develop either acutely, or (less typically) slowly (prognosis is worse in the acute onset group, especially for return of bladder function, which occurred in only ≈ 50%)¹¹¹. 3 patterns¹¹⁷:

• Group I - sudden onset of CES symptoms with no previous low back symptoms

• Group II - previous history of recurrent backache and sciatica, the latest episode combined with CES

• Group III - presentation with backache & bilateral sciatica that later develop CES

Surgical issues: some advise a bilateral laminectomy¹¹³ (but this is not mandatory). Occasionally, when it is difficult to remove a very tense midline disc, transdural removal may be helpful¹¹⁵.

Timing of discectomy in CES: controversial, and the point of contention in numerous law suits. In spite of early reports emphasizing rapid decompression¹¹⁵, other re-ports found no correlation between the time to surgery after presentation and the return of function^{111, 112}. Some evidence supports the goal of performing surgery within 48 hours (although performing surgery within 24 hours is desirable if possible, there is no statistically significant proof that delaying up to 48 hrs is detrimental)^{118, 119}.

BOOKING THE CASE - LUMBAR DISCECTOMY

Also see defaults & disclaimers (page v).

1. position: prone

2. equipment: microscope (if used), minimally invasive retractors (if used)

3. consent (in lay terms for the patient - not all-inclusive):

A. procedure: through the back to go between the bones and remove the piece of disc that is pressing on the nerve(s)

B. alternatives: nonsurgical management

C. complications: (usual spine surgery complications - see page v) plus the disc can herniate again in the same place in ≈ 6% of cases, it is possible that a fragment of disc can be missed at the time of surgery, there might not be the amount of pain relief desired (back pain does not respond as well to surgery as nerve-root pain),

SURGICAL OPTIONS FOR LUMBAR RADICULOPATHY

Once it is decided to treat surgically, options include:

1. trans-canal approaches

A. standard open lumbar laminectomy and discectomy: 65-85% reported no sciatica one year post-op compared to 36% for conservative treatment¹²⁰. Long-term results (> 1 year) were similar. 10% of patients underwent further back surgery during the first year¹²⁰

B. “microdiscectomy”^{121, 122}: similar to standard procedure, however smaller incision is utilized. Advantages may be cosmetic, shortened hospital stay, lower blood loss. May be more difficult to retrieve some fragments^{123 (p 1319), 124}. Overall efficacy is similar to standard discectomy¹²⁵

C. sequestrectomy: removal of only the herniated portion of disc

2. intradiscal procedures (see below)

A. chemonucleolysis: using chymopapain (see below)

B. automated percutaneous lumbar discectomy: utilizes a nucleotome

C. percutaneous endoscopic discectomy: see below

D. intradiscal endothermal therapy (IDET or IDTA): see below

E. laser disc decompression

Chemonucleolysis

Acceptable treatment, but less efficacious than discectomy¹ (“open” or micro). Chymopapain (Chymodiactin®) is injected intradiscally. Proven more effective than placebo injection^{126, 127}. Typical success rates: at 1 year 85% of patients undergoing discectomy had good or excellent results compared to 44%¹²⁸ to 63%¹²⁹ for chemonucleolysis (CNL). Although sciatica improves in both groups, only the discectomy group had significant improvement in back pain¹²⁸. In one study, at 6 months 56% of patients initially having CNL had undergone surgery for unrelieved symptoms¹³⁰.

Risks^{131, 132}:

Risk of the significant complication of anaphylaxis (sometimes fatal) may be reduced by skin-tests for allergic sensitivity to the agent. Other complications reported include: discitis¹³³, neurologic injury, vascular injury, thrombophlebitis, PE, transverse myelitis¹³⁴ (very uncommon).

Intradiscal surgical procedures (ISP)

ISPs (see below for specific procedures) are among the most controversial procedures for lumbar spine surgery. The theoretical advantage is that epidural scarring is avoided, and that a smaller incision or even just a puncture site is used. This is also purported to reduce postoperative pain and hospital stay (often performed as an outpatient procedure). The conceptual problem with ISPs is that they are directed at removing disc material from the center of the disc space (which is not producing symptoms) and rely on the reduced intradiscal pressure to decompress the herniated portion of the disc from the nerve root. Only ≈ 10-15% of patients considered for surgical treatment of disc disease are candidates for an ISP. ISPs are usually done under local anesthetic in order to permit the patient to report nerve root pain to identify impingement on a nerve root by the surgical instrument or needle. Overall, ISPs are not recommended until controlled trials prove the efficacy¹.

Indications utilized by proponents of intradiscal procedures:

1. type of disc herniation: appropriate only for “contained” disc herniation (i.e. outer margin of anulus fibrosus intact)

2. appropriate level: best for L4-5 HLD. May also be used at L3-4. Difficult but often workable (utilizing angled instruments or other techniques) at L5-S1 because of the angle required and interference by iliac crest

3. not recommended in presence of severe neurologic deficit¹³⁵

Results:

“Success” rate (≈ pain free and return to work when appropriate) reported ranges from 37-75%^136-138.

Automated percutaneous lumbar discectomy: AKA nucleoplasty. Utilizes a nucleotome¹³⁹ to remove disc material from the center of the intervertebral disc space. Significantly less efficacious than chymopapain¹³⁸, with 1 year success rate of 37% (compared to 66% for CNL). Complications include cauda equina syndrome from improper nucleotome placement¹⁴⁰. In another study, nucleoplasty (with or without IDET (see below)) for HLD showed only modest reduction in pain at 9 months¹⁴¹.

Laser disc decompression: Insertion of a needle into the disc, and introduction of a laser fiberoptic cable through the needle to allow a laser to burn a hole in the center of the disc^{142, 143} (with or without endoscopic visualization).

Percutaneous endoscopic lumbar discectomy (PELD): This term refers to an essentially intradiscal procedure indicated primarily for contained disc herniations, although some small “noncontained” fragments may be treatable¹⁴⁴. No large randomized study has been done to compare the technique to the accepted standard, open discectomy (with or without microscope). In one report¹⁴⁵ of 326 patients with L4-5 HLD, only 8 (2.4%) met study criteria (no previous operation, failure of conservative treatment, imaging study proving disc protrusion followed by discography to R/O “disc perforation”) for PELD. Of these 8, only 3 were reported as having a good result. This study is not adequate for evaluating the technique.

Intradiscal endothermal therapy (IDET): AKA intradiscal (electro)thermal anuloplasty (IDTA). Efficacy: 23-60% at 1 year for treating “internal disc disruption”¹⁴⁶ (radial fissures in the nucleus pulposus extending into the anulus fibrosus) which is purported to account for 40% of patients with chronic low back pain of unknown etiology¹⁴⁷.

ADJUNCTIVE TREATMENT IN LUMBAR LAMINECTOMY

Epidural steroids following discectomy

In a single-blinded non-randomized study of the use of epidural steroids (methylprednisolone acetate (Depo-Medrol®), dose not specified) irrigation of the thecal sac and nerve root following discectomy prior to wound closure found no statistically significant evidence of benefit in terms of amount of post-op analgesic medication needed, duration of hospital stay, or time to return to work¹⁴⁸. However, the combination of systemic steroids at the start of the case (Depo-Medrol® 160 mg IM and methylprednisolone sodium succinate (Solu-Medrol®) 250 mg IV) combined with infiltration of 30 ml of 0.25% bipuvicaine (Marcaine®) into the paraspinal muscles at incision and closure, may reduce hospital stay and post-op narcotic requirements¹⁴⁹.

Methods to reduce scar formation

Epidural free fat graft: The use of an autogenous free fat graft in the epidural space has been employed in an attempt to reduce post-op epidural scar formation. Opinion varies widely as to the effectiveness, some feel it is helpful, others feel it actually exacerbates scarring¹⁵⁰. In some patients, no evidence of the graft will be found on reoperation years later. The fat graft can very rarely be a cause of nerve root compression¹⁵¹ or cauda equina syndrome¹¹⁶within the first few days post-op, and there is a case report of compression 6 years following surgery¹⁵².

Other measures: Other measures include the placement of barrier films or gels. There are numerous products available, none has been shown to have reproducible benefit.

RISKS OF LUMBAR LAMINECTOMY

Overall risk of mortality in large series^{153, 154}: 6 per 10,000 (i.e. 0.06%), most often due to septicemia, MI, or PE. Complication rates are very difficult to determine accurately¹²⁰, but the following is included as a guideline.

Common complications

(consider discussing these as part of informed consent)

1. infection:

A. superficial wound infection: 0.9-5%¹⁵⁵ (risk is increased with age, long term steroids, obesity, ? DM): most are caused by S. aureus (see Laminectomy wound infection, page 348 for management)

B. deep infection: < 1%(see below under Uncommon complications)

2. increased motor deficit: 1-8% (some transient)

3. unintended “incidental” durotomy^A: (see below) incidence is 0.3-13% (risk increases to ≈ 18% in re-do operations)¹⁵⁶. Possible sequelae include those listed in Table 18-14

A. CSF fistula (external CSF leak): the risk of a CSF fistula requiring operative repair is ≈ 10 per 10,000153

B. pseudomeningocele: 0.7-2%¹⁵⁶ (may appear similar radiographically to spinal epidural abscess (SEA), but post-op SEA often enhances, is more irregular, and is associated with muscle edema)

4. recurrent herniated lumbar disc (same level either side): 4% (with 10 year followup)¹⁵⁷ (see page 460)

A. the term “unintended durotomy” has been recommended in preference to “dural tear” (see below)

Uncommon complications

1. direct injury to neural structures. For large disc herniations, consider a bilateral exposure to reduce risk

2. injury to structures anterior to the vertebral bodies (VB): injured by breaching the anterior longitudinal ligament (ALL) through the disc space, e.g. with pituitary rongeur. The depth of disc space penetration with instruments should be kept ≤ 3 cm, since 5% of lumbar discs had diameters as small as 3.3 cm¹⁵⁸. Asymptomatic perforations of the ALL occur in up to 12% of discectomies. Breach of the ALL risks potential injuries to:

A. great vessels¹⁵⁹: risks include potentially fatal hemorrhage, and arteriovenous fistula which may present years later. Most such injuries occur with L4-5 discectomies. Only ≈ 50% bleed into the disc space intraoperatively, the rest bleed into the retroperitoneum. Emergent laparotomy is indicated, preferably by a surgeon with vascular surgical experience, if available. Mortality rate is 37-67%

1. aorta: the aortic bifurcation is on the left side of the lower part of the L4 VB, and so the aorta may be injured above this level

2. below L4, the common iliac arteries may be injured

3. veins (more common than arterial injuries)

a. vena cava at and above L4

b. common iliac veins below L4

B. ureters

C. bowel: at L5-S1 the ileum is the most likely viscus to be injured

D. sympathetic trunk

3. wrong site surgery: incidence in self-reporting survey was 4.5 occurrences per 10,000 lumbar spine operations¹⁶⁰. Factors identified as potential contributors to the error: unusual patient anatomy, not performing localizing radiograph. 32% of responding neurosurgeons indicated that they removed disc material from the wrong level at some time in their career

4. rare infections:

A. meningitis

B. deep infection: < 1%. Including:

1. discitis: 0.5% (see page 383),

2. spinal epidural abscess (SEA): 0.67% (see page 376)

5. cauda equina syndrome: may be caused by post-op spinal epidural hematoma (see below). Incidence was 0.21% in one series of 2842 lumbar discectomies¹⁶¹ and 0.14% in a series of 12,000 spine operations¹⁶². Red flags: urinary retention, anesthesia that may be saddle or bilateral LE

6. postoperative visual loss (POVL)¹⁶³:

A. ischemic optic neuropathy¹⁶⁴: the most common cause of POVL. Commonly bilateral. Usually associated with significant blood loss (median: 2 L), and/or prolonged operative time (≥ 6 hrs). All cases had anesthetic time > 5 hrs or blood loss > 1 L. Blood loss can cause hypotension (may cause release of endogenous vasoconstrictors in addition to reduced blood flow due to low hemodynamic pressure) and increased platelet aggregation. Is not due to direct pressure on the globe in most cases, and can occur at any age and even in otherwise healthy patients. Blindness can be extensive and is often permanent, although early aggressive treatment may result in some improvement

1. posterior ischemic optic neuropathy (PION)¹⁶⁴: may follow surgery (surgical PION). Risk factors as above, plus:

a. surgery in the prone position (can cause peri-orbital edema, and rarely, direct pressure on the orbit)

b. lack of tight glycemic control

c. use of Trendelenburg position

d. hemodilution or overuse of crystalloid vs. colloid (blood) fluid replacement

e. prolonged hypotension

f. cellular hypoxia

g. decreased renal perfusion

2. anterior ischemic optic neuropathy (AION): divided into arteritic (as with GCA) and nonarteritic (common with DM)

B. central retinal artery occlusion

C. cortical blindness

7. complications of positioning:

A. compression neuropathies: ulnar, peroneal nerves. Use padding over elbows and avoid pressure on posterior popliteal fossa

B. anterior tibial compartment syndrome: due to pressure on anterior compartment of leg (reported with Andrew’s frame). An orthopedic emergency that may require emergent fasciotomy

C. pressure on the eye: corneal abrasions, damage to the anterior chamber

D. cervical spine injuries during positioning due to relaxed muscles under anesthesia

8. post op arachnoiditis: risk factors include epidural hematoma, patients who tend to develop hypertrophic scar, post op discitis, and intrathecal injection of Pantopaque®, anesthetic agents or steroids. Surgical treatment is disappointing. Intrathecal depo-medrol may provide short-term relief (in spite of the fact that steroids are a risk factor for the development of arachnoiditis). Also see page 458

9. thrombophlebitis and deep-vein thrombosis with risk of pulmonary embolism (PE)¹⁵³: 0.1% (see Thromboembolism in neurosurgery, page 42)

10. complex regional pain syndrome AKA reflex sympathetic dystrophy (RSD): reported in up to 1.2% of cases, usually after posterior decompression with fusion, often following reoperations¹⁶⁵ with onset 4 days to 20 weeks post-op. See page 576 for a critique of RSD. Treatment includes some or all of: PT, sympathetic blocks, oral methylprednisolone, removal of hardware if any

11. very rare: Ogilvie’s syndrome (pseudo-obstruction (“ileus”) of the colon). Usually seen in hospitalized/debilitated patients. May be related to narcotics, electrolyte deficiencies, possibly from chronic constipation. Also reported following spinal surgery/trauma, spinal/epidural anesthesia, spinal metastases, & myelography¹⁶⁶

Unintended durotomy

Unintentional opening of the dura during spinal surgery has an incidence of 0-14%¹⁶⁷.

Terminology: The terms “unintended durotomy”, “incidental durotomy”¹⁶⁷, or even just “dural opening”, have been recommended in preference to “dural tear” which may imply carelessness¹⁵⁶ when none was present. Dural openings have been associated with one or more alleged complications or sequelae in medical malpractice suits involving surgery on the lumbar spine.

The injury: By itself, opening the dura intentionally or otherwise is not expected to have a deleterious effect on the patient^{156, 168}. In fact, dural opening is often a standard part of the operation for intradural disc herniation¹⁶⁹, tumors, etc. Although not frequent, (for incidence, see above) unintended durotomy is not an unusual occurrence, and alone, is not considered an act of malpractice. However, it may result from an event or events that produce more serious injuries. These events and injuries should be dealt with on their own merits.

Possible sequelae include those listed in Table 18-14. A CSF leak may produce “spinal headache” with its associated symptoms (see page 58), and if it breaches the skin it may be a risk factor for meningitis. Pain or sensory/motor deficits may be associated with injuries to nerve roots or delayed herniation of nerve roots through the dural opening.

Etiologies: For incidence, see above. Potential causes are many, and include¹⁵⁶: unanticipated anatomic variations, adhesion of the dura to removed bone, slippage of an instrument, an obscured fold of dura caught in a rongeur or curette, thinning of the dura in cases of long-standing stenosis, and the possibility of a delayed CSF leak caused by perforation of the dura when it expands onto a surgically created spicule of bone¹⁷⁰. The risk may be increased with anterior decompression for OPLL, with revision surgery, and with the use of high-speed drills¹⁶⁷.

Table 18-14 Possible sequelae of dural opening

Well documented

1. CSF leak A. contained: pseudomeningocele B. external: CSF fistula 2. herniation of nerve roots thorough opening 3. associated nerve root contusion, laceration or injury to the cauda equina 4. CSF leak collapses the thecal sac and may increase blood loss from epidural bleeding

Less well documented

1. arachnoiditis 2. chronic pain 3. bladder, bowel and/or sexual dysfunction

Treatment: If the opening is recognized at the time of surgery, watertight primary closure (with or without patch graft) should be attempted with nonabsorbable suture if at all possible to prevent pseudomeningocele and/or CSF fistula. A cottonoid placed over the opening prevents aspiration of nerve roots¹⁷¹. Care must be taken to avoid incorporating a nerve root into the closure. Most repairs will be accomplished with no complication or sequelae to the patient. When the opening is in the far (anterior) side of the dura, consideration may be given to intradural repair accessed through a posterior durotomy which is subsequently closed (this may risk additional injury to the nerve roots). Biocompatible fixatives (e.g. fibrin glue¹⁶⁷) may be used to supplement primary closure.

Primary repair may be impossible in some situations (e.g. when the opening cannot be found or accessed, as is sometimes the case when it occurs on the nerve root sleeve) and alternatives here include placement of a fat or muscle graft over the suspected leak site, use of a the patient’s own blood for a “blood patch” (one technique is to have the anesthesiologist draw ≈ 5-10 ml of the patient’s blood from an arm vein, keeping it in the syringe for several minutes until it starts to coagulate, and then to have the anesthesiologist inject the blood onto the dura), use of gelfoam, fibrin glue… Some recommend that the wound not be drained post-op, with a water-tight closure of fascia, fat, and skin to add to the barrier. Others use a subcutaneous drain or epidural catheter. CSF diversionary procedures (e.g. through a drain inserted 1 or more levels away) may also be used.

Although bed rest x 4-7 days is often advocated to reduce symptoms and facilitate healing, when watertight closure has been achieved, normal post-op mobilization is not associated with a high failure rate (bed rest is recommended if symptoms develop)¹⁶⁷.

In one report of 8 patients with leaks that appeared post-op, re-operation was avoided when treated by resuturing the skin under local anesthesia, followed by bed rest in slight Trendelenburg position (to reduce pressure on the leakage site), broad spectrum antibiotics and antibiotic ointment over the skin incision, and daily puncture and drainage of the subcutaneous collection¹⁷².

For other treatment measures for H/A associated with CSF leak, see page 59.

POST-OP CARE

Post-op orders

The following are guidelines for post-operative orders for a lumbar laminectomy without intraoperative complications; variations between surgeons and institutions must be taken into consideration:

1. admit post-anesthesia recovery (PAR) unit

2. vital signs on the nursing unit: q 2° x 4 hrs, q 4° x 24°, then q 8°

3. activity: up with assist, advance as tolerated

4. nursing care

A. I’s & O’s

B. intermittent catheterization q 4-6° PRN no void

C. optional: TED hose (may reduce risk of DVT) or PCB

D. optional (if drain used): empty drain q 8° and PRN

5. diet: clear liquids, advance as tolerated

6. IV: D5 1/2 NS + 20 mEq KCl/l @ 75 ml/hr, D/C when tolerating PO well (after antibiotics D/C’d if prophylactic antibiotics are used)

7. meds

A. laxative of choice (LOC) PRN

B. sodium docussate (e.g. Colace®) 100 mg PO BID when tolerating PO (stool softener, does not substitute for LOC)

C. optional: prophylactic antibiotics if used at your institution

D. acetaminophen (Tylenol®) 650 mg PO or PR q 3° PRN

E. narcotic analgesic

F. optional: steroids are used by some surgeons to reduce nerve-root irritation from surgical manipulation

8. labs

A. optional (if significant blood loss during surgery): CBC

Post-op check

In addition to routine, the following should be checked:

1. strength of lower extremities, especially muscles relevant to nerve root, e.g. gastrocnemius for L5-S1 surgery, EHL for L4-5 surgery…

2. appearance of dressing: look for signs of excessive bleeding, CSF leak…

3. signs of cauda equina syndrome (see page 446), e.g. by post-op spinal epidural hematoma:

A. loss of perineal sensation (“saddle anesthesia”)

B. inability to void: may not be not unusual after lumbar laminectomy, more concerning if accompanied by loss of perineal sensation

C. pain out of the ordinary for the post-op period

D. weakness of multiple muscle groups

Any new neurologic deficit should prompt rapid evaluation for spinal epidural hematoma¹⁶² (EDH). Delayed deficits may be due to EDH or epidural abscess. Post-op films in the recovery room can rule out graft or hardware malposition for fusions or instrumentation procedures, or changes in alignment. The diagnostic test of choice is MRI. If contraindicated or not available, CT/myelography may be indicated. An extradural defect immediately post-op suggests EDH.

OUTCOME OF SURGICAL TREATMENT

In a series of 100 patients undergoing discectomy, at 1 year post-op 73% had complete relief of leg pain and 63% had complete relief of back pain; at 5-10 years the numbers were 62% for each category⁹⁰. At 5-10 years post-op, only 14% felt that the pain was the same or worse than pre-op (i.e. 86% felt improved), and 5% qualified as having a failed back surgery syndrome (a heterogeneous not-precisely defined term, here meaning not returned to work, requiring analgesics, receiving worker’s compensation, see Failed back surgery syndrome, page 457).

Attempts to measure relative merits of conservative treatment vs. surgery have failed. The recent SPORT study^{173, 174} suffered from significant selection bias since patients were allowed to crossover to the other arm of the study and thus more nearly approximated the current methodology of surgical selection than an actual RCT¹⁷⁵. Earlier attempts at randomized trials also suffered from methodological flaws⁸⁹. Conclusions that can be drawn from these studies¹⁷⁵: most patients with manageable or improving pain and less disability typically choose conservative treatment and most have improvement in symptoms, whereas patients with severe, persistent or worsening pain and/or neurologic deficit are more likely to choose surgery with a resultant excellent outcome.

In patients with a diminished knee-jerk or ankle-jerk pre-op, 35% and 43% (respectively) still had reduced reflexes 1 year post-op⁹⁵; reflexes were lost post-op in 3% and 10% respectively. The same study found that motor loss was improved in 80%, aggravated in 3%, and was newly present in 5% post-op; and that sensory loss was improved in 69% and was worsened in 15% post-op.

Foot drop: severe or complete paralysis of ankle dorsiflexion occurs in 5-10% of HLD, and about 50% of cases recover with or without treatment. Discectomy does not improve the outcome, especially in cases of painless foot drop¹¹⁰.

Recurrent disc herniation: (see page 460)

HERNIATED UPPER LUMBAR DISCS (LEVELS L1-2, L2-3, & L3-4)

L4-5 & L5-S1 herniated lumbar discs (HLD) account for most cases of HLD (realistically ≈ 90%, possibly as high as 98%⁹⁴). 24% of patients with HLD at L3-4 have a past history of a HLD at L4-5 or L5-S1, suggesting a generalized tendency towards disc herniation. In a series of 1,395 HLDs, there were 4 at L1-2 (0.28% incidence), 18 at L2-3 (1.3%), and 51 at L3-4 (3.6%)¹⁷⁶.

PRESENTATION

Typically presents with LBP, onset following trauma or strain in 51%. With progression, paresthesias and pain in the anterior thigh occur, with complaints of leg weakness (especially on ascending stairs).

SIGNS

Quadriceps femoris was the most common muscle involved, demonstrating weakness and sometimes atrophy.

Straight leg raising was positive in only 40%. Psoas stretch test was positive in 27%. Femoral stretch test may be positive (see page 444).

50% had reduced or absent knee jerk; 18% had ankle jerk abnormalities; reflex changes were more common with L3-4 HLD (81%) than L1-2 (none) or L2-3 (44%).

EXTREME LATERAL LUMBAR DISC HERNIATIONS

Definition: herniation of a disc at (foraminal disc herniation) or distal to (extraforaminal disc herniation) the facet (some authors do not consider foraminal disc herniation to be “extreme lateral”). See Figure 18-3.

Incidence (see Table 18-15): 3-10% of herniated lumbar discs (HLD) (series with higher numbers¹⁷⁷ include some HLD that are not truly extreme lateral).

Figure 18-3 Zones of lumbar disc herniation

Differs from the more common central and subarticular HLD in that:

• the nerve root involved is usually the one exiting at that level (c.f. the root exiting at the level below)

• straight leg raising (SLR) is negative in 85-90% of cases ≥ 1 week after onset (excluding double herniations; ≈ 65% will be negative if double herniations are included); may have positive femoral stretch test (see page 444)

• pain is reproduced by lateral bending to the side of herniation in 75%

• myelography alone is rarely diagnostic (usually requires CT^{178, 179} or MRI) subarticular

• higher incidence of extruded fragments (60%)

• higher incidence of double herniations on the same side at the same level (15%)

• pain tends to be more severe (may be due to fact that the dorsal root ganglion may be compressed directly) and often has more of a burning dysesthetic quality

Table 18-15 Incidence of extreme lateral HLD by level*

Disc level	No.	%
L1-2	1	1%
L2-3	11	8%
L3-4	35	24%
L4-5	82	60%
L5-S1	9	7%

* series of 138 cases¹⁷⁷

Occurs most commonly at L4-5 and next at L3-4 (see Table 18-15), thus L4 is the most common nerve involved and L3 is next. With a clinical picture of an upper lumbar nerve root compression (i.e. radiculopathy with negative SLR), chances are ≈ 3 to 1 that it is an extremely lateral HLD rather than an upper lumbar disc herniation.

PRESENTATION

Quadriceps weakness, reduction of patellar reflex, and diminished sensation in the L3 or L4 dermatome are the most common findings.

Differential diagnosis includes:

1. lateral recess stenosis or superior articular facet hypertrophy

2. retroperitoneal hematoma or tumor

3. diabetic neuropathy (amyotrophy): see page 796

4. spinal tumor

A. benign (schwannoma or neurofibroma)

B. malignant tumors

C. lymphoma

5. infection

A. localized (spinal epidural abscess)

B. psoas muscle abscess

C. granulomatous disease

6. spondylolisthesis (with pars defect)

7. compression of conjoined nerve root

8. on MRI, enlarged foraminal veins may mimic extreme lateral disc herniation

RADIOGRAPHIC DIAGNOSIS

Radiographic diagnosis may be elusive, and up to one third are initially missed¹⁸⁰. However, if actively sought, many asymptomatic far-lateral disc herniations may be demonstrated on CT or MRI.

Myelography: fails to disclose the pathology even with water soluble contrast in 87% of cases due to the fact that the nerve root compression occurs distal to the nerve root sleeve (and therefore beyond the reach of the dye)¹⁸¹.

CT scan¹⁷⁹: reveals a mass displacing epidural fat and encroaching on the intervertebral foramen or lateral recess, compromising the emerging root. Or, may be lateral to foramen. Sensitivity is ≈ 50% and is similar with post-myelographic CT¹⁸¹. Post-discography CT¹⁸² may be the most sensitive test (94%)¹⁸¹.

MRI: similar sensitivity to post-myelographic CT. Sagittal views through the neural foramen may help demonstrate the disc herniation¹⁸⁰. MRI may have ≈ 8% false positive rate due to presence of enlarged foraminal veins that mimic extreme lateral HLD¹⁸³.

SURGICAL TREATMENT

NB: compression of the dorsal root ganglion may result in a slower recovery from discectomy and overall less satisfying outcome than with the more commonplace para-median disc herniation.

Foraminal discs

Usually requires mesial facetectomy to gain access to the region lateral to the dural sac without undue retraction on nerve root or cauda equina. Caution: total facetectomy combined with discectomy may result in a high incidence of instability (total facetectomy alone causes ≈ 10% rate of slippage), although other series found this risk to be lower (≈ 1 in 33^{184, 185}). An alternative technique is to remove just the lateral portion of the superior articular facet below¹⁸⁶. Endoscopic techniques may be well suited for herniated discs in this location¹⁸⁷.

Discs herniated beyond (lateral to) the foramen

Numerous approaches are used, including:

1. traditional midline hemilaminectomy: the ipsilateral facet must be partially or completely removed. The safest way to find the exiting nerve root is take the laminectomy of the inferior portion of the upper vertebral level (e.g. L4 for a L4-5 HLD) high enough to expose the nerve root axilla, and then follow the nerve laterally through the neural foramen by removing facet until the HLD is identified

2. lateral approach (i.e. extra-canal) through a paramedian incision¹⁸⁸. Advantages: the facet joint is preserved (facet removal combined with discectomy may lead to instability), muscle retraction is easier. Disadvantages: unfamiliar approach for most surgeons and the nerve cannot be followed medial to lateral. A localizing x-ray is taken with a spinal needle. A 4-5 cm vertical skin incision is made 3-4 cm lateral to the midline on the side of the disc herniation. The incision is taken down to the thoracolumbar fascia and the subcutaneous tissue is dissected off the fascia. Above L4, one may palpate the groove between multifidus (medial) and long-issimus (lateral), where the fascia is incised. The facet joint is palpated, and blunt dissection is used to gain access to the lateral facet joint and transverse processes above and below the level of the disc herniation. The correct level is confirmed on x-ray using a probe as a marker. The intertransversarius muscle and fascia are divided. Care must be taken to avoid mechanical and electrocautery injury to the nerve and dorsal root ganglion (which lies immediately beneath the intertransverse ligament). The radicular artery, vein, and nerve root are located just beneath the transverse processes, usually slightly medial to this position. The nerve root hugs the pedicle of the level above as it exits the neural foramen (palpating this e.g. with a dental dissector helps locate it), and it may be splayed over the herniated disc fragment. If more medial exposure is necessary, the lateral facet joint may be resected. The HLD is removed. Additional removal of disc material from the disc space may be performed with down-biting pituitary rongeurs. Extracanal approach to L5-S1 requires removal of part of the sacral ala in order to access the space caudal to the L5 transverse process

DISC HERNIATIONS IN PEDIATRICS

Less than one percent of surgery for herniated lumbar disc is performed on patients between the ages of 10 and 20 yrs (one series at Mayo found 0.4% of operated HLD in patients < 17 yrs age¹⁸⁹). These patients often have few neurologic findings except for a consistently positive straight leg raising test¹⁹⁰. Herniated disc material in youths tends to be firm, fibrous and strongly attached to the cartilaginous end-plate unlike the degenerated material usually extruded in adult disc herniation. Plain radiographs disclosed an unusually high frequency of congenital spine anomalies (transitional vertebra, hyperlordosis, spondylolisthesis, spina bifida…). 78% did well after their first operation¹⁸⁹.

INTRADURAL DISC HERNIATION

Herniation of a fragment of disc into the thecal sac, or into the nerve root sleeve (the latter sometimes referred to as “intraradicular” disc herniation) has been recognized with a reported incidence of 0.04-1.1% of disc herniations^{169, 191}. Although it may be suspected on the basis of pre-op myelography or MRI, the diagnosis is rarely made preoperatively¹⁹¹. Intraoperatively, it may be suggested by the impression of a tense firm mass within the nerve root sleeve or by the negative exploration of a level with obvious clinical signs and clear cut radiographic abnormalities (after verifying that the correct level is exposed).

Surgical treatment:

Although a surgical dural opening may be utilized¹⁶⁹, others have found this to be necessary in a minority of cases¹⁹².

INTRAVERTEBRAL DISC HERNIATION

AKA Schmorl’s node or nodule. AKA Schmor’s (no “l”) nodule AKA Geipel hernia¹⁹³. Disc herniation through the cartilaginous end plate into the cancellous bone of the vertebral body (VB) (AKA intraspongious disc herniation). Often an incidental finding on xray or MRI. Clinical significance is controversial. May produce low back pain initially that lasts ≈ 3-4 months after onset. Diffuse displacement (as may be seen in osteoporosis) is sometimes referred to as a balloon disc⁵.

Clinical findings

During the acute (symptomatic) phase, patients may exhibit LBP that is aggravated by weight bearing and movement. There may be tenderness to percussion or manual compression over the involved segment.

Table 18-16 MRI signal intensity in Schmorl’s nodes*

Lesion	T1WI	T2WI
symptomatic (acute)	low	high
asymptomatic (chronic)	high†	low†

* signal intensity in surrounding marrow

^† the same as normal marrow

Radiographic findings

Plain x-ray: ≤ 33% may be seen on plain x-rays¹⁹⁴. They may not be detectable acutely until sclerotic osseous bone casting develops.

MRI: the extrusion of disc material into the VB is easily appreciated on sagittal images. It has been suggested¹⁹⁵ that acute (symptomatic) lesions may appear differentiated from chronic (asymptomatic) lesions by the presence of MRI findings of inflammation in the bone marrow immediately surrounding the node as outlined in Table 18-16.

Treatment

Conservative treatment is indicated, usually consisting of non-steroidal anti-inflammatory drugs (NSAIDs). Occasionally stronger pain medication and/or lumbar bracing may be required. Surgery is rarely indicated.

Outcome

With conservative treatment, symptoms generally resolve within 3-4 months of onset (as with most vertebral body fractures).

JUXTAFACET CYSTS OF THE LUMBAR SPINE

The term juxtafacet cyst (JFC) was originated by Kao et al.¹⁹⁶ and includes both synovial cysts (those having a synovial lining membrane) and ganglion cysts (those lacking synovial lining) adjacent to a spinal facet joint or arising from the ligamentum flavum. Distinction between these two types of cysts may be difficult without histology (see below) and is clinically unimportant¹⁹⁷.

JFC occur primarily in the lumbar spine (although cysts in the cervical^198-200 and thoracic²⁰¹ spine have been described). They were first reported in 1880 by von Gruker during an autopsy²⁰², and were first diagnosed clinically in 1968²⁰³. The etiology is unknown (possibilities include: synovial fluid extrusion from the joint capsule, latent growth of a developmental rest, myxoid degeneration and cyst formation in collagenous connective tissue…), increased motion seems to have a role in many cysts, and the role of trauma in the pathogenesis is debated^{199, 204} but probably plays a role in a small number (≈ 14%)²⁰⁵. JFC are relatively rare, only 3 cases were identified in a series of 1,500 spinal CT exams²⁰⁶, but the frequency of diagnosis may be on the rise due to the widespread use of MRI and an increasing awareness of the condition.

Clinical

The average age was 63 years in one series²⁰⁵ and 58 years in a review of 54 cases in the literature²⁰⁷ (range: 33-87) with a slight female preponderance in both series. Most occur in patients with severe spondylosis and facet joint degeneration²⁰⁸, 25% had degenerative spondylolisthesis²⁰⁵. L4-5 is the most common level^{205, 209}. They may be bilateral. Pain is the most common symptom, and is usually radicular. Some JFC may contribute to canal stenosis and can produce neurogenic claudication²¹⁰ (see page 477) or on occasion a cauda equina syndrome. Symptoms may be more intermittent in nature than with firm compressive lesions, such as HLD. A sudden exacerbation in pain may be due to hemorrhage within the cyst. Some JFC may be asymptomatic²¹¹.

Differential diagnosis (also see Differential diagnosis, Sciatica on page 1188). Differentiating JFC from other masses relies largely on the appearance and location. Other distinguishing features include:

1. neurofibroma: unlikely to be calcified

2. free fragment of HLD: not cystic in appearance

3. epidural or nerve root metastases: not cystic

4. dural subarachnoid root sleeve dilatation: see Spinal meningeal cysts, page 509

5. arachnoid cyst (from arachnoid herniation through a dural defect): not associated with facet joint, margins thinner than JFC²¹²

6. perineurial cysts (Tarlov’s cyst): arise in space between perineurium and endoneurium, usually on sacral roots²¹³, occasionally show delayed filling on myelography. Usually associated with remodelling of adjacent bone

Pathology

Cyst walls are composed of fibrous connective tissue of varying thickness and cellularity. There is usually no signs of infection or inflammation. There may be a synovial lining²⁰⁷ (synovial cyst) or it may be absent²⁰⁸ (ganglion cyst). The distinction between the two may be difficult¹⁹⁷, possibly owing in part to the fact that fibroblasts in ganglion cysts may form an incomplete synovial-like lining²¹⁴. Proliferation of small venules is seen in the connective tissue. Hemosiderin staining may be present, and may or may not be associated with a history of trauma²⁰⁵.

Evaluation

Identifying a JFC pre-op helps the surgeon, as the approach differs slightly from that for HLD, and the cyst might otherwise be missed or unknowingly deflated and unnecessary time wasted afterwards trying to find a compressive lesion. Or, the unwitting surgeon may misinterpret the cyst as a “transdural disc extrusion” and needlessly open the dura. Pre-op diagnoses were incorrect in 30% of operated cases of JFC²⁰⁵.

Myelography: posterolateral filling defect (whereas most discs are situated anteriorly, an occasional fragment may migrate posterolaterally, whereas a JFC will always be posterolateral), often with a round extradural appearance.

CT scan: shows a low density epidural cystic lesion typically with a posterolateral juxtaarticular location. Some have calcified rim²¹¹, and some may have gas within²¹⁵. Erosion of bony lamina is occasionally seen^{209, 216}.

MRI: variable findings (may be due to differing composition of cyst fluid: serous vs. proteinaceous²¹⁷). Unenhanced signal characteristics of non-hemorrhagic JFC are very similar to CSF. Hemorrhagic JFC are hyperintense. May be missed on sagittal imaging without contrast. Axial images may better demonstrate JFC. Gadolinium enhancement increases the sensitivity²⁰⁸. MRI usually misses bony erosion.

Treatment

Optimal treatment is not known. There is one case report of a cyst that resolved spontaneously²⁰⁶. If symptoms persist with conservative treatment, some promote cyst aspiration or facet injection with steroids²¹⁸, while most advocate surgical excision of the cyst.

Surgical treatment considerations: The cyst may be adherent to the dura. The cyst may also collapse during the surgical approach and may be missed. A JFC may serve as a marker for possible instability and should prompt an evaluation for the same. Some argue for performing a fusion since JFC may arise from instability, however, it appears that fusion is not required for a good result in many cases²¹⁸. Therefore it is suggested that consideration for fusion be made on the basis of any instability and not merely on the basis of the presence of a JFC.

Minimally invasive spine surgery (MISS) has also been used for removal²¹⁹, long-term follow-up is lacking. A 15 mm entry incision is made 1.5 cm lateral to midline.

Following surgical treatment, symptomatic JFCs may recur or may develop on the contralateral side²⁰⁵.

FAILED BACK SURGERY SYNDROME

Definition: failure to improve satisfactorily following back surgery (for herniated intervertebral disc, laminectomy for stenosis…). These patients often require analgesics and are unable to return to work. The failure rate for lumbar discectomy to provide satisfactory long-term pain relief is ≈ 8-25%²²⁰. Pending legal or worker’s compensation claims were the most frequent deterrents to a good outcome¹⁵⁷.

Factors that may cause or contribute to the failed back syndrome:

1. incorrect initial diagnosis

A. inadequate pre-op imaging

B. clinical findings not correlated with abnormality demonstrated on imaging

C. other causes of symptoms (sometimes in the presence of what was considered to be an appropriate lesion on imaging studies which may have been asymptomatic): e.g. trochanteric bursitis, diabetic amyotrophy…

2. continued nerve root or cauda equina compression caused by:

A. residual compression (retained disc material, osteophytes…)

B. recurrent pathology at same level: disc reherniation at the same level (usually have pain-free interval > 6 mos post-op (see page 460)) or restenosis (over many years²²¹ - was more common with midline fusions))

C. adjacent level pathology: disc herniation or stenosis²²¹

D. compression of nerve root by peridural scar (granulation) tissue (see below)

E. pseudomeningocele

F. epidural hematoma

G. conjoined nerve roots with compression at another level or in atypical location

H. segmental instability: 3 patterns^{222, 1}) lateral rotational instability, 2) postop spondylolisthesis, 3) post-op scoliosis

3. permanent nerve root injury from the original disc herniation or from surgery, includes deafferentation pain which is usually constant and burning or ice cold

4. adhesive arachnoiditis: responsible for 6-16% of persistent symptoms in post-op patients²²³ (see below)

5. discitis: usually produces exquisite back pain 2-4 weeks post-op (see page 383)

6. spondylosis

7. other causes of back pain unrelated to the original condition: paraspinal muscle spasm, myofascial syndrome… Look for trigger points, evidence of spasm

8. post-op reflex sympathetic dystrophy (RSD): see page 450

9. “non-anatomic factors”: poor patient motivation, secondary gains, drug addiction, psychological problems… (see Psychosocial factors, page 436)

ARACHNOIDITIS (AKA ADHESIVE ARACHNOIDITIS)

Inflammatory condition of the lumbar nerve roots. Actually a misnomer, since adhesive arachnoiditis is really an inflammatory process or fibrosis that involves all three meningeal layers (pia, arachnoid, and dura). Many putative “risk factors” have been described for the development of arachnoiditis, including²²⁴:

1. spinal anesthesia: either due to the anesthetic agents or to detergent contaminants on the syringes used for same

2. spinal meningitis: pyogenic, syphilitic, tuberculous

3. neoplasms

4. myelographic contrast agents: less common with currently available water soluble contrast agents

5. trauma

A. post-surgical: especially after multiple operations

B. external trauma

6. hemorrhage

7. idiopathic

Radiographic findings in arachnoiditis

NB: Radiographic evidence of arachnoiditis may also be found in asymptomatic patients²²⁴. Arachnoiditis must be differentiated from tumor: the central adhesive type (see below) may resemble CSF seeding of tumor, and myelographic block may mimic intrathecal tumor.

Myelogram: May demonstrate complete block, or clumping of nerve roots. One of many myelographic classification systems²²⁵ for arachnoiditis is shown in Table 18-17.

MRI:³ patterns on MRI^{226, 227}:

1. central adhesion of the nerve roots into 1 or 2 central “cords”

2. “empty thecal sac” pattern: roots adhere to meninges around periphery, only CSF signal is visible intrathecally

3. thecal sac filled with inflammatory tissue, no CSF signal. Corresponds with myelographic block and candle-dripping appearance

Enhancement: acute arachnoiditis may enhance. Chronic arachnoiditis usually does not enhance with gadolinium as much as tumor.

Table 18-17 Myelographic classification of arachnoiditis

Type	Description
1	unilateral focal filling defect centered on the nerve root sleeve adjacent to disc space
2	circumferential constriction around thecal sac
3	complete obstruction with “stalactites” or “candle guttering”, “candle-dripping”, or “paint-brush” filling defects
4	infundibular cul-de-sac with loss of radicular striations

PERIDURAL SCAR

Although peridural scar is frequently blamed for causing recurrent symptoms^{228, 229}, there has been no proof of correlation²³⁰. Peridural fibrosis is an inevitable sequelae to lumbar disc surgery. Even patients who are relieved of their pain following discectomy develop some scar tissue post-op²³¹. Although it has been shown that if a patient has recurrent radicular pain following a lumbar discectomy there is a 70% chance that extensive peridural scar will be found on MRI²³⁰, this study also showed that on post-op MRIs at 6 months, 43% of patients will have extensive scar, but 84% of the time this will be asymptomatic²³². Thus, one must use clinical grounds to determine if a patient with extensive scar on MRI is in the 16% minority of patients with radicular symptoms attributable to scar²³².

For a discussion of measures to reduce peridural scarring, see page 448.

RADIOLOGIC EVALUATION

Patients with only persistent low back or hip pain without a strong radicular component, with a neurologic exam that is normal or unchanged from pre-op, should be treated symptomatically. Patients with signs or symptoms of recurrent radiculopathy (positive SLR is a sensitive test for nerve root compression), especially if these follow a period of apparent recovery, should undergo further evaluation.

It is critical to differentiate residual/recurrent disc herniation from scar tissue and adhesive arachnoiditis as surgical treatment has generally poor results with the latter two (see Treatment of failed back surgery syndrome below).

MRI WITHOUT AND WITH IV GADOLINIUM

Diagnostic test of choice. The best exam for detecting residual or recurrent disc herniation, and to reliably differentiate disc from scar tissue. Pre-contrast studies with T1WI and T2WI yields an accuracy of ≈ 83%, comparable to IV enhanced CT^{233, 234}. With the addition of gadolinium, using the protocol below yields 100% sensitivity, 71% specificity, and 89% accuracy²³⁵. May also detect adhesive arachnoiditis (see above). As scar becomes more fibrotic and calcified with time, the differential enhancement with respect to disc material attenuates and may become undetectable at some point, ≈ 1-2 years postop²³⁴(some scar continues to enhance for > 20 yrs).

Recommended protocol²³⁵

Get pre-contrast T1WI and T2WI. Give 0.1 mmol/kg gadolinium IV. Obtain T1WI images within 10 minutes (early post-contrast). No benefit from post-contrast T2WI.

Findings on unenhanced MRI

Signal from a HLD becomes more intense as the sequence is varied from T1WI → T2WI, whereas scar tissue becomes less intense with this transition. Indirect signs (also applicable to CT):

1. mass effect: a nerve root is displaced away from disc material, whereas it may be retracted toward scar tissue by adherence to it

2. location: disc material tends to be in contiguity with the disc interspace (best seen on sagittal MRI)

Findings on enhanced MRI

On early (≤ 10 mins post-contrast) T1WI images: scar enhances inhomogeneously, whereas disc does not enhance at all. A nonenhancing central area surrounded by irregular enhancing material probably represents disc wrapped in scar. Venous plexus also enhances, and may be more pronounced when it is distorted by disc material, but the morphology is easily differentiated from scar tissue in these cases.

On late (> 30 mins post-contrast) T1WI: scar enhances homogeneously, disc had variable or no enhancement. Normal nerve roots do not enhance even on late images.

CT SCAN WITHOUT AND WITH IV (IODINATED) CONTRAST

Unenhanced CT scan density measurements are unreliable in the postoperative back²³⁶. Enhanced CT is only fairly good in differentiating scar (enhancing) from disc (un-enhancing with possible rim enhancement). Accuracy is about equal to unenhanced MRI.

MYELOGRAPHY, WITH POST-MYELOGRAPHIC CT

Postoperative myelographic criteria alone are unreliable for distinguishing disc material from scar^{224, 237}. With the addition of CT scan, neural compression is clearly demonstrated, but scar still cannot be reliably distinguished from disc.

Myelography (especially with post-myelographic CT) is very capable of demonstrating arachnoiditis²³⁷ (see above).

PLAIN LS X-RAYS

Generally helpful only in cases of instability, malalignment, or spondylosis²³⁷. Flexion/extension views are most helpful when trying to demonstrate instability.

TREATMENT OF FAILED BACK SURGERY SYNDROME

For treatment of intervertebral disc-space infection, see Discitis, page 383.

Symptomatic treatment

Recommended for patients who do not have radicular signs and symptoms, or for most patients demonstrated to have scar tissue or adhesive arachnoiditis on imaging. As in other cases of non-specific LBP treatment includes: short-term bed rest, analgesics (non-narcotic in most cases), anti-inflammatory medication (non-steroidal, and occasionally a short course of steroids), and physical therapy.

Surgery

Reserved for those with recurrent or residual disc herniation, segmental instability, or patients with a pseudomeningocele. Patients with post-op spinal instability should be considered for spinal fusion²²² (see page 440).

In most series with sufficient follow-up, success rates after reoperation are lower in patients with only epidural scar (as low as 1%) compared to those patients with disc and scar (still only ≈ 37%)²²⁰. An overall success rate (> 50% pain relief for > 2 yrs) of ≈ 34% was seen in one series²²⁹, with better results in patients that were young, female, with good results following previous surgery, a small number of previous operations, employment prior to surgery, predominantly radicular (cf axial) pain, and absence of scar requiring lysis.

In addition to the absence of disc material, factors associated with poor outcome were: sensory loss involving more than one dermatome, and patients with past or pending compensation claims^{220, 238}.

Arachnoiditis: Surgery for carefully selected patients with arachnoiditis (those with mild radiographic involvement (Types 1 & 2 in Table 18-17), and < 3 previous back operations)²²⁵ has met with moderate success (although in this series, no patient returned to work). Approximate success rate in other series^{239, 240}: 50% failure, 20% able to work but with symptoms, 10-19% with no symptoms. Surgery consists of removal of extradural scar enveloping the thecal sac, removing any herniated disc fragments, and performing foraminotomies when indicated. Intradural lysis of adhesions is not indicated since no means for preventing reformation of scar has been identified²⁴⁰.

RECURRENT HERNIATED LUMBAR DISC

Rates quoted in the literature range from 3-19% with the higher rates usually in series with longer follow-up²⁴¹. In an individual series with 10 year mean F/U, the rate of recurrent disc herniation was 4% (same level, either side), one third of which occurred during the 1st year post-op (mean: 4.3 yrs)¹⁵⁷. A second recurrence at the same site occurred in 1% in another series²⁴¹ with mean F/U of 4.5 yrs. In this series²⁴¹, patients presenting for a second time with disc herniation had a recurrence at the same level in 74%, but 26% had a HLD at another level. Recurrent HLD occurred at L4-5 more than twice as often as L5-S1241.

It is often possible for a smaller amount of recurrent herniated disc to cause symptoms than in a “virgin back”, due to the fact that the nerve root is often fixated by scar tissue and has little ability to deviate away from the fragment¹⁵⁰.

TREATMENT

Initial recommended treatment is as with a first time HLD. Nonsurgical treatment should be utilized in the absence of progressive neurologic deficit, cauda equina syndrome (CES) or intractable pain.

Surgical treatment

Disagreement occurs regarding optimal treatment. See PRACTICE GUIDELINE 18-5, page 440.

Surgical outcome:

As with first time HLD, the outcome from surgical treatment is worse in worker’s compensation cases and in patients undertaking litigation, only ≈ 40% of these patients benefit^{241, 242}. A worse prognosis is also associated with: patients with < 6 mos relief after their first operation, cases where fibrosis without recurrent HLD is found at operation.

Spinal cord stimulation

One study actually showed a better response rate to spinal cord stimulation than to reoperation²⁴³. Since surgery for recurrent HLD carries a higher risk of dural and nerve root injury, and a lower success rate than first time operations, this may be a viable option for some patients.

18.3.2. Cervical disc herniation

CLINICAL ASPECTS

The following facts explain the findings in herniated cervical disc (HCD):

1. in the cervical region, the nerve root exits above the pedicle of its like-numbered vertebra (opposite to the situation in the lumbar spine, due to the fact that there are 8 cervical nerve roots and only 7 cervical vertebrae)

2. each root exits passes through its neural foramen in close relation to the undersurface of the pedicle

3. the intervertebral disc space is located close to the inferior portion of the pedicle (unlike the lumbar region)

CERVICAL NERVE ROOT SYNDROMES (CERVICAL RADICULOPATHY)

Due to the facts listed above, a HCD usually impinges on the nerve exiting from the neural foramen at the level of the herniation (e.g. a C6-7 HCD usually causes C7 radiculopathy). This gives rise to the characteristic cervical nerve root syndromes shown in Table 18-18.

Some clinical specifics:

C4 radiculopathy is not common, and may produce nonradiating axial neck pain. Left C6 radiculopathy (e.g. from C5-6 HCD) occasionally presents with pain simulating an MI (pseudo-angina). C8 and T1 nerve root involvement may produce a partial Horner’s syndrome.

The most common scenario for patients with herniated cervical disc is that the symptoms were present upon awakening in the morning, without identifiable trauma or stress²⁴⁴.

Differential diagnosis: see page 1197.

CERVICAL MYELOPATHY AND SCI DUE TO CERVICAL DISC HERNIATION

Acute cord compression presenting with myelopathy or SCI (including complete SCI and incomplete syndromes, especially central cord syndrome (see page 948) and some-times Brown-Sequard syndrome²⁴⁵(see page 950)) is well described in association with traumatic cervical disc herniation²⁴⁶. Less commonly, these findings may occur in non-traumatic cervical disc herniation.

SIGNS USEFUL IN EVALUATING CERVICAL RADICULOPATHY

Almost all herniated cervical discs cause painful limitation of neck motion. Neck extension usually aggravates pain when cervical disc disease is present (a minority of patients instead exhibit pain with flexion). Some patients find relief in elevating the arm and cupping the back or the top of the head with the hand (abduction relief sign, see below for shoulder abduction test). Lhermitte’s sign (electrical shock-like sensation radiating down the spine) may be present (see page 1198 for DDx).

Miscellaneous

The following tests were found to be specific, but not particularly sensitive in detecting cervical root compression²⁴⁷:

1. Spurling’s sign 248: radicular pain reproduced when the examiner exerts down-ward pressure on vertex while tilting head towards symptomatic side (sometimes adding neck extension). Causes narrowing of the intervertebral foramen and possibly increases disc bulge. Used as a “mechanical sign” analogous to SLR for lumbar disc herniation

2. axial manual traction: 10-15 kg of axial traction is applied to a supine patient with radicular symptoms (pull up on patient’s mandible and occiput). The reduction or disappearance of radicular symptoms is a positive finding

3. shoulder abduction test 249: a sitting patient with radicular symptoms lifts their hand above their head. The reduction or disappearance of radicular symptoms is a positive finding. Moderately sensitive, fairly specific²⁵⁰

EVALUATION

MRI

The study of choice for initial evaluation for herniated cervical disc (HCD). Accuracy is less than water soluble contrast myelogram/CT (≈ 85-90% accuracy for MRI because of only fair to good imaging of neural foramen), but is non-invasive. For myelopathy, MRI is > 95% effective in diagnosing.

CT AND MYELOGRAM/CT

Indications: when MRI cannot be done, when resolution or image quality on MRI is inadequate, or when more bony detail is required. Also, to evaluate for ossification of the posterior longitudinal ligament (OPLL) in suspicious cases.

Plain CT: is usually good at C5-6, is variable at C6-7 (due to artifact from patient’s shoulders, depending on body habitus), and is usually poor at C7-T1.

Myelogram/CT (water soluble intrathecal contrast): invasive, on rare occasions requires overnight hospitalization. Accuracy is ≈ 98% for cervical disc disease.

TREATMENT

Over 90% of patients with acute cervical radiculopathy due to cervical disc herniation can improve without surgery²⁵¹. The recovery period may be made more tolerable by adequate pain medication, anti-inflammatory medication (NSAIDs or short-course tapering steroids) and intermittent cervical traction (e.g. 10-15 lbs for 10-15 minutes, 2-3 x daily).

Surgery is indicated for those that fail to improve or those with progressive neurologic deficit while undergoing non-surgical management.

Management of myelopathy/central cord syndrome associated with acute cervical disc herniation is controversial, since the natural history is favorable in most cases. However, some patients have poor recovery and experience permanent deficits even with emergency surgery²⁵².

Surgical options:

1. anterior cervical discectomy: see below

A. without any prosthesis or fusion

B. combined with fusion: the most common approach

1. without anterior cervical plating

2. with anterior cervical plating

C. with artificial disc (arthroplasty)

2. posterior approaches

A. cervical laminectomy

B. keyhole laminotomy

For practice guidelines regarding intra-op electrophysiologic monitoring for surgery for cervical radiculopathy, see PRACTICE GUIDELINE 18-18, page 491.

ANTERIOR CERVICAL DISCECTOMY WITH FUSION (ACDF)

Without special modifications, a routine anterior approach is limited ≈ to levels C3-7.

Advantages over posterior (nonfused) approach:

1. safe removal of osteophytes

2. fusion of disc space affords immobility (up to 10% incidence of subluxation with extensive posterior approach)

3. only viable means of directly dealing with centrally herniated disc

Disadvantages over posterior approach: immobility at fused level may increase stress on adjacent disc spaces. If a fusion is performed, some surgeons prescribe a rigid collar (e.g. Philadelphia collar) for 6-12 weeks. Multiple level ACDF can devascularize the vertebral body (or bodies) between discectomies.

Booking the case - ACDF

Also see defaults & disclaimers (page v).

1. position: supine, some use halter traction with this

2. equipment:

A. microscope (not used by all surgeons)

B. C-arm

3. implants: graft (e.g. PEEK, cadaver bone, titanium cage…) and anterior cervical plate (optional, especially on single level ACDF)

4. neuromonitoring: (optional) some surgeons used SSEP/MEP

5. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the front of the neck to remove the degenerated disc and bone spurs, and to place a graft where the disc was, and possibly place a metal plate on the front of the spine. Some surgeons take bone from the hip to replace the removed disc

B. alternatives: nonsurgical management, surgery from the back of the neck, artificial disc (in some cases)

C. complications: swallowing difficulties are common but usually resolve, hoarseness of the voice (< 4% chance of it being permanent), injury to: food-pipe (esophagus), windpipe (trachea), arteries to the brain (carotid), spinal cord with paralysis, nerve root with paralysis, possible seizures with MEPs

TECHNIQUE

A summary of the steps involved is included here. For C5-6, the skin incision is made at level of cricoid cartilage, for other levels, appropriate adjustments up or down may be made, sometimes with the assistance of fluoroscopy. The incision is approximately 5-6 cm horizontally, centered on the SCM. Many right handed surgeons prefer operating from the right side of the neck, although the risk to the recurrent laryngeal nerve (RLN) is lower with a left sided approach (the RLN lies in a groove between the esophagus and trachea). The skin may be undermined off the platysma to permit a vertical incision in the platysma in the same orientation as its muscle fibers. Alternatively, some incise the platysma horizontally with scissors horizontally.

Dissect in tissue plane medial to SCM. For the C5-6 interspace, angle slightly cranially during dissection. For the C6-7 disc, proceed almost straight down to spine. Sweep omohyoid medially (to stay out of it and to protect the RLN). The trachea + esophagus are retracted medially. The carotid sheath + SCM are retracted laterally.

After verification of level with lateral C-spine x-ray with spinal needle in the interspace, bipolar the prevertebral fascia and medial edges of the longus coli muscles longitudinally in the midline. Self-retaining retractor blades are inserted underneath the fascia to retract the longus coli muscles laterally. The anesthesiologist is asked to deflate the cuff on the endotracheal tube and then to re-inflate it using minimal leak technique to reduce the risk of compression injury from the retractor. The disc space is incised with a 15 scalpel blade. The discectomy is performed with curettes and pituitary rongeurs; a vertebral body spreader aids the exposure. The posterior longitudinal ligament is incised, one technique is to elevate it with a sharp nerve hook and then incise it with a #11 scalpel. The subligamentous space is probed with a blunt nerve hook. The posterior lip of the VB above and below are removed with a Kerrison rongeur with a small foot-plate. Decompression of the roots is verified with the blunt nerve hook. Fusion is performed at this time if desired by placing the graft in the interspace.

For redo operations (same or different levels): approach is usually from the same side as previous operation(s) since many patients have swallowing issues post-op, and some may be due to partial recurrent laryngeal nerve injury (which can be subclinical) and which could result in a permanent need for a feeding tube if a contralateral injury occurs. If for some reason it is desired to go to the opposite side, an evaluation by an ENT physician is recommended, and should include scoping the patient to rule-out subclinical problems that could turn into major difficulties if bilateral.

To fuse or not to fuse?

Table 18-19 compares ACD (no fusion) to ACDF for radiculopathy in a number of aspects²⁵³.

Choice of graft material:

Autologous bone (usually from iliac crest), non-autologous bone (cadaveric), bone substitutes (e.g. hydroxyapatite²⁵⁴) or synthetics (e.g. PEEK or titanium cage) filled with osteogenic material. Substitutes for autologous bone eliminate problems with the donor site (see page 465), but may have a higher rate of absorption. There were also cases of HIV transmission from cadaveric bone grafts in 1985, however, no further cases have been reported.

Anterior cervical plating: Recommendations for plating following ACDF are shown in PRACTICE GUIDELINE 18-7.

Table 18-19 Comparison of ACD to ACDF for radiculopathy²⁵³

Component	Assessment
Clinical outcome measures*	ACD & ACDF are equivalent (Level C, Class I)
Relief of arm pain
Relief of neck pain associated with 1-level disc degeneration	Conflicting evidence which is better (Class II)
More rapid relief of neck and arm pain	ACDF is better than ACD (Level D, Class III) (functional outcomes are similar)
Maintaining foraminal or disc space height	ACDF cannot be recommended over ACD (Level C, Class II)
Risk of post-op kyphosis	ACDF is better than ACD (Level C, Class II)
Fusion rate

* VAS, Odom’s criteria, McGill pain questionnaire, SF-36

PRACTICE GUIDELINE 18-7 ANTERIOR CERVICAL PLATING

1 level ADCF: The addition of an anterior plate to an ACDF is recommended to reduce the pseudarthrosis rate and graft problems (Level DClass III) and to maintain lordosis (Level CClass II) but it does not improve clinical outcome alone (Level BClass II)²⁵³

2 level ADCF: Plating is recommended to improve arm pain. Plating does not improve other outcome parameters (Level CClass II)²⁵³

Use of bone morphogenic proteins (BMP):

PRACTICE GUIDELINE 18-8 USE OF BMP IN CERVICAL INTERBODY GRAFTING

Current evidence does not support the routine use of rhBMP-2 for cervical arthrodesis (Level C Class II)²⁵⁵*

* italics added. Use with precautions (see text) may be indicated in cases with high risk of nonunion

Use of BMP in anterior cervical discectomies is not FDA approved, but has been used off-label. Complication rates as high as 23-27% have been reported (including postop swallowing or respiratory difficulties as a result of edema which is usually temporary) compared to 3% without BMP²⁵⁵. If used, it is recommended that a smaller dose be employed than in the lumbar spine (25% has been advocated) and to avoid contact of BMP with soft tissues in the neck.

POST-OP CHECK

In addition to routine, the following should be checked

1. evidence of significant post-op wound hematoma: should be first consideration in patient with airway obstruction post-op. Wound may need to be emergently opened on floor if airway is compromised, see Carotid endarterectomy, disruption of arteriotomy closure, management on page 1154. Also consider swelling from IJV thrombosis (rare) in differential diagnosis (see below)

A. respiratory distress

B. extreme difficulty swallowing: alternatively may indicate anterior extrusion of bone graft impinging upon esophagus (check lateral C-spine x-ray)

C. tracheal deviation: may be visible or may be seen on AP C-spine x-ray

2. weakness of nerve root of level operated: e.g. biceps for C5-6, triceps for C6-7

3. long tract signs (Babinski sign…) which may indicate cord compression by spinal epidural hematoma

5. hoarseness: may indicate vocal cord paresis from recurrent laryngeal nerve injury: hold oral feeding until this can be further assessed

ACDF COMPLICATIONS

Common ones listed below, see references^{256, 257} for more details. The most common complication following ACDFs: swallowing difficulties (may be multifactorial).

• exposure injuries

A. perforation of pharynx, esophagus and/or trachea: minimize by blunt retraction until longus colli is separated from its attachment to vertebrae.

• Esophageal injuries are very difficult to manage, and require ENT involvement. The incidence may be higher with use of anterior cervical plate, and injury may not manifest until years after the fusion (may be due to repetitive motion of esophagus over the plate). Treatment of esophageal perforation is usually facilitated by plate removal

B. vocal cord paresis: due to injury of the recurrent laryngeal nerve (RLN) or vagus. Incidence: 11% temporary, 4% permanent paresis. Symptoms include: hoarseness, breathiness, cough, aspiration, mass sensation, dysphagia, and vocal cord fatigue²⁵⁸. Avoid sharp dissection in paratracheal muscles. Some cases may be due to prolonged retraction against trachea and not to nerve division; to reduce this risk, after the self-retaining retractor is placed, have the anesthesiologist deflate the cuff on the ET tube and then inflate it to minimal leak pressure. More common with right sided approaches, primarily in the lower cervical spine (C5-6 and below) where the RLN is more vulnerable²⁵⁸

C. vertebral artery injury: thrombosis or laceration. 0.3% incidence²⁵⁷. Treatment alternatives include: packing, direct repair by temporary clipping with aneurysms clips and repair with 8-0 prolene²⁵⁹ and endovascular trapping. Risks of treating hemorrhagic complications with packing include: recurrent bleeding, AV fistula, pseudoaneurysm, arterial thrombosis²⁵⁷, distal embolic CVA (primarily in cerebellum)

D. carotid injury: thrombosis, occlusion, or laceration (usually by retraction)

E. CSF fistula: usually difficult to repair directly. Place fascial graft beneath bone plug. Keep HOB elevated post op. Consider: dural sealant (fibrin glue, DuraSeal®…), lumbar drain

F. Horner’s syndrome: sympathetic plexus lies within longus coli, thus do not extend dissection far laterally into these muscles

G. thoracic duct injury: in exposing lower cervical spine, primarily on left

H. thrombosis of internal jugular vein²⁶⁰: rare. Carries 2-3% risk of PE²⁶¹. Treatment options: anticoagulation (oral or IV) may lower the mortality²⁶², SVC filter if anticoagulation is contraindicated²⁶³, percutaneous thrombectomy²⁶⁴

• spinal cord or nerve root injuries

A. spinal cord injury: especially risky in myelopathy due to narrowed canal. Minimize risk by penetrating the osteophyte at the lateral margin of interspace (however, this increases risk to nerve root)

B. avoid hyperextension during intubation: anesthesiologist may need to determine patient’s tolerance pre-op. Consider fiberoptic guided or awake nasotracheal intubation in extreme stenosis

C. bone graft must be shorter than interspace depth. Exercise caution in tapping graft into position

D. sleep induced apnea: rare but serious complications of C3-4 level operations²⁶⁵. May be associated with bradycardia & cardiorespiratory instability. Possibly due to disruption of the afferent component of the central respiratory control mechanism

• bone fusion problems

A. failure of fusion (pseudarthrosis): see below

B. anterior (kyphotic) angulation deformity: may be as high as 60% with Cloward technique (may be reduced by collar immobilization). May develop in Hirsch technique with excessive bone removal

C. graft extrusion: 2% incidence (rarely requires re-operation unless compression of cord posteriorly, or esophagus or trachea anteriorly occurs)

D. donor site complications: hematoma/seroma, infection, fracture of ilium, injury to lateral femoral cutaneous nerve, persistent pain due to scar, bowel perforation

• miscellaneous

A. wound infection: incidence < 1%

B. post-op hematoma: see above. Placing cervical collar in O.R. may delay recognition

C. dysphagia and hoarseness: common. Usually transient (see below)

D. adjacent level degeneration: controversial whether this represents a sequelae to altered biomechanics from surgery, or a predisposition to cervical spondylosis²⁶⁶. Many (≈ 70%) are asymptomatic²⁶⁷

E. postoperative discomfort:

1. globus: the sensation of a lump in throat (see below)

2. nagging discomfort in neck, shoulder, and very commonly in interscapular regions (may last several months). May correlate with amount of distraction of the disc space

F. complex regional pain syndrome AKA reflex sympathetic dystrophy (RSD): rarely described in the literature²⁶⁸, possibly due to stellate ganglion injury (see page 576)

G. angioedema: massive edema of the tongue and neck²⁶⁹. A dramatic hypersensitivity reaction (not really a direct complication of ACDF, but superficially can mimic some findings of post-op hematoma). If limited to the tongue, the airway is not compromised. For treatment, see page 125

H. pneumothorax or hemothorax²⁷⁰: accessing C7-T1 or lower may expose the pleural apex

Dysphagia following ACDF

Symptoms: Include: difficulty swallowing (solids, liquids including saliva), pain with swallowing (odynophagia), globus (sensation of a lump in throat) and compromise of ability to protect against aspiration. Food may stick in the throat (or feel as if it is) and there may be coughing or choking.

Incidence: Early dysphagia is common. Incidence: 60%²⁷¹ in retrospective survey after noninstrumented fusion^A, 50% in prospective study²⁷². At 6 months, only ≈ 5% reported moderate or severe dysphagia²⁷². Surgery at multiple levels increased the risk at 1 & 2 months²⁷². Decreases significantly in most cases by 6 months²⁷².

Etiologies: Etiologies of post-op dysphagia include:

1. post-op hematoma. If severe, may cause tracheal obstruction (see above)

2. post-op edema, due in part to retraction of esophagus

3. effects of general anesthesia: e.g. irritation from ET tube. Accounts for up to 23% of early symptoms^A. Usually subsides within ≈ 24-72 hours

4. recurrent laryngeal nerve dysfunction:

A. temporary: usually due to traction on the nerve

B. permanent: 1.3% at 12 months²⁷²

5. esophageal injury

A. at time of surgery

B. delayed: possibly from repetitive abrasion on surgical site/hardware

6. cervical collar

A. prevents patient from lowering jaw during swallow phase, which compromises effective glottic closure of airway

B. may be too tight thereby directly squeezing the throat

7. protrusion of graft/hardware anterior to the vertebral bodies

A. some protrusion is present with most anterior hardware. This may be minimized with “zero profile” instrumentation

B. hardware failure (screw-pullout/backout/breakage, plate pullout)

C. interbody graft migration: without anterior plate, or in conjunction with anterior plate displacement

8. excessive adhesions²⁷³

9. denervation of the pharyngeal plexus²⁷³

10. rare conditions: swelling from IJV thrombosis, angioedema

A. dysphagia occurred in 23% in a control group undergoing unrelated lumbar spine surgery²⁷¹

Management:

1. initial management: rule-out emergent/serious conditions (severe edema, hematoma with airway compromise, risk of aspiration)

A. if there is significant stridor or dysphonia, especially if tracheal deviation is obvious, someone must stay with the patient as efforts are made to emergently take the patient to the O.R. for wound exploration evacuation. Consider opening the wound at the bedside if delays occur or if symptoms are severe (see Carotid endarterectomy, disruption of arteriotomy closure, management on page 1154). Emergent anesthesia consultation for airway protection - alert them to the likelihood of deviated trachea which challenges even the most expert at intubating

2. once emergent conditions are ruled out, early management is geared towards amelioration of symptoms

A. advise patient to eat softer foods (temporarily avoiding steak or bread), to chew food well, to wash down dry foods with a drink. Reassure patient that most cases largely resolve with 6 months²⁷²

B. if significant symptoms persist > 2 weeks

1. refer patient to ENT for laryngoscopy to rule out vocal cord paralysis (from RLN injury) or other etiologies

2. modified barium swallow

3. persistent symptoms may be amenable to surgical intervention, including hardware removal and lysing of adhesions²⁷³, management of esophageal perforation usually requires consultation with ENT

Pseudarthrosis (or pseudoarthrosis) following ACDF

PRACTICE GUIDELINE 18-9 ASSESSMENT OF SUBAXIAL FUSION

> 2 mm movement between spinous processes on dynamic (flexion-extension) cervical spine x-rays is recommended as a criteria for pseudarthrosis (Level BClass II), this measurement is unreliable when performed by the treating surgeon (Level CClass II)²⁷⁴.

Visualization of bone trabeculation across the fusion on static films is a less reliable marker for fusion (Level DClass III) (2D reformatted CT increases the accuracy (Level DClass III))²⁷⁴

Pseudarthrosis may occur with or without supplemental anterior cervical plating.

Incidence: Difficult to assess because of lack of validated criteria. Estimate: 2-20%. Higher with dowel technique (Cloward) than with keystone technique of Bailey & Badgley or with interbody method of Smith-Robinson (10%) or with non-fusion advocated by Hirsch. One criteria: motion > 2 mm between the tips of the spinous processes on lateral flexion/extension x-rays^{275, 276}. Other criteria: lucencies around the screws of an anterior plate, toggling of the screws on flexion/extension x-rays.

Presentation: Not uniformly associated with symptoms or problems^{275, 277}. Some patients may have chronic or recurrent neck pain, some may present with radicular symptoms. (NB: when DePalma’s data is analyzed with patients reclassified as failures if neck and/or arm symptoms persist, the success rate of surgery is lower with pseudarthrosis²⁷⁸).

Management: (Guidelines are shown in PRACTICE GUIDELINE 18-10). No treatment is required for asymptomatic pseudarthrosis. Options for symptomatic patients include re-resection of the bone graft with repeat fusion²⁷⁹(some recommend using autologous bone if allograft was used, a plate may be considered if one was not used previously), cervical corpectomy with fusion²⁷⁹, or posterior cervical fusion.

PRACTICE GUIDELINE 18-10 MANAGEMENT OF ANTERIOR CERVICAL PSEUDARTHROSIS

Revision of symptomatic* pseudarthrosis should be considered (Level DClass III)²⁸⁰. Posterior approaches may be associated with higher fusion rates on revision than anterior approaches (Level DClass III)²⁸⁰

* italics added

CERVICAL DISC ARTHROPLASTY

PRACTICE GUIDELINE 18-11 CERVICAL DISC ARTHROPLASTY

Cervical arthroplasty is a recommended alternative to ACDF in selected patients for control of arm and neck pain (Level BClass II)²⁵³

An alternative to fusion. Uses an artificial disc to preserve motion at the level of the discectomy. Some of the available cervical disc replacement (CDR) models are shown in Table 18-20²⁸¹.

Booking the case - cervical disc arthroplasty

Also see defaults & disclaimers (page v).

1. position: supine, some use halter traction with this

2. equipment:

A. microscope (not used by all surgeons)

B. C-arm

3. implants: schedule vendor to provide desired artificial disc

4. neuromonitoring: (optional) some surgeons used SSEP/MEP

5. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the front of the neck to remove the degenerated disc and bone spurs, and to place an artificial disc

B. alternatives: nonsurgical management, surgical fusion (from the front or the back of the neck)

C. complications: swallowing difficulties are common but usually resolve, hoarseness of the voice (< 4% chance of it being permanent), injury to: food-pipe (esophagus), windpipe (trachea), arteries to the brain (carotid) with stroke, spinal cord with paralysis, nerve root with paralysis, possible seizures with MEPs (if used). The disc may eventually wear out and further surgery may be needed

Post-op orders:

1. no cervical collar (the goal is to preserve motion at the operated level)

2. NSAIDs around the clock for ≈ 2 weeks (this inhibits bone growth which theoretically helps avoid undesirable fusion at the operated level)

POSTERIOR CERVICAL DECOMPRESSION (CERVICAL LAMINECTOMY)

Not necessary for unilateral radiculopathy (use either ACD or keyhole laminotomy). Consists of removal of cervical lamina (laminectomy) and spinous processes in order to convert the spinal canal from a “tube” to a “trough”.

Usually reserved for the following conditions:

1. multiple cervical discs or osteophytes (anterior cervical discectomy (ACD) is usually used to treat only 2, or possibly 3, levels without) with myelopathy

2. where the anterior pathology is superimposed on cervical stenosis, and the latter is more diffuse and/or more significant (see Cervical spinal stenosis, page 485)

3. in professional speakers or singers where the 4% risk of permanent voice change due to recurrent laryngeal nerve injury with ACD may be unacceptable

Booking the case - cervical laminectomy

Also see defaults & disclaimers (page v).

1. position: prone, some use pin headholder

2. equipment:

A. C-arm

B. high-speed drill

3. implants: cervical lateral mass screws and rods if fusion is being done

4. neuromonitoring: some surgeons used SSEP/MEP

5. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the back of the neck to remove the bone over the compressed spinal cord and nerves and possibly to place screws and rods to fuse the boned together

B. alternatives: nonsurgical management, surgery from the front of the neck, posterior surgery without fusion, laminoplasty

C. complications: nerve root injury (C5 nerve root is the most common), may not relieve symptoms necessitating further surgery, possible seizures with MEPs. If fusion is not done, risk of progressive bone slippage which would require further surgery

POSTERIOR KEYHOLE LAMINOTOMY

PRACTICE GUIDELINE 18-12 CERVICAL LAMINOFORAMINOTOMY

Cervical laminoforaminotomy is recommended as a surgical treatment option for symptomatic cervical radiculopathy caused by disc herniation or lateral recess narrowing (Level DClass III)²⁸²

AKA “keyhole foraminotomy”. Decompresses only individual nerve roots (but not the spinal cord) by creating a small “keyhole” in the lamina to access the nerve root.

Indications for keyhole approach (as opposed to anterior discectomy):

1. monoradiculopathy with posterolateral soft disc sequestration (small lateral osteophytic spurs may also be addressed)

2. radiculopathy in patients who are professional speakers or singers (see above)

3. for lower (e.g. C7, C8 or T1) or upper (e.g. C3 or C4) cervical nerve root compression, especially in a patient with a short thick neck, making an anterior approach more difficult

Booking the case - cervical keyhole laminectomy

Also see defaults & disclaimers (page v).

1. position: prone, some use pin headholder

2. equipment:

A. microscope (not used by all surgeons)

B. C-arm

3. instrumentation: some surgeons use a tube retractor system

4. neuromonitoring: some surgeons used SSEP/MEP

5. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the back of the neck to remove the bone over the compressed nerve root and possibly remove fragment of herniated disc

B. alternatives: nonsurgical management, surgery from the front of the neck, posterior surgery with fusion

C. complications: nerve root injury, may not relieve symptoms necessitating further surgery, possible seizures with MEPs

Technique^283-285

1. position

A. prone, on chest rolls. Adhesive tape is used to retract shoulders down for any level below about C4-5. The head is stabilized on a horse-shoe headrest or in a Mayfield head holder.

B. sitting position: generally abandoned. May be used with proper precautions (see page 153)

“Open” keyhole foraminotomy

The desired level is localized with intra-op x-ray or fluoroscopy before making the skin incision, a 2-3 cm midline incision is adequate. A unilateral exposure suffices. Periosteal elevators are used to dissect muscles off the lamina and facet joint in the sub-periosteal plane. A Kocher clamp may be placed on the spinous process to permit confirmation of the correct level on intraoperative x-ray. A Scoville retractor or equivalent is employed.

A high-speed drill (e.g. with diamond burr) is used to make an opening in the medial one-third to one-half of the inferior facet of the vertebra above the desired disc space, extending slightly medially into the junction with the lamina. Once the inferior facet is penetrated, the superior facet of the inferior vertebral level will be visualized. This is also thinned with the drill (it is critical to remove the bone of the superior facet of the level below caudally to where it meets the pedicle). A small Kerrison rongeur may be used to slightly enlarge the laminectomy. An opening is made in the ligamentum flavum overlying the lateral aspect of the spinal cord dura. The nerve root can be identified as it exits from the thecal sac, and can be followed as it travels between the pedicles of the vertebrae above and below. Soft tissues (including ligamentum flavum) form fibrous bands across the dorsum of the nerve, and are removed to further expose the dura of the nerve root. The venous plexus around the nerve root is coagulated with bipolar cautery and then divided to mobilize the nerve. The nerve may then be gently moved a few millimeters rostrally using a micro nerve hook. The dura overlying the spinal cord should not be manipulated, and the disc space need not be entered. Inspection for free disc fragments should begin in the nerve root axilla using a probe (e.g. blunt nerve hook). Next, the space anterior to the root (the region of the disc) may be palpated. Any disc fragments that are dislodged are removed with a small pituitary rongeur. If the disc fragment is contained anterior to the posterior longitudinal ligament (PLL), the PLL may be incised in the region of the nerve root axilla with a #11 scalpel blade in a motion that is directed downward and laterally, away from the nerve root and spinal cord. The foraminotomy may be extended slightly laterally if the foramen still feels tight when probed. Small osteophytes can potentially be reduced using a small reversed-angled curette, although some surgeons believe that the need for this is obviated by the decompression provided by the keyhole opening. In some cases, simple posterior decompression of the nerve root (without removing a disc fragment) may be adequate to relieve compression. Spinal stability is usually preserved if less than half the facet joint is removed.

MIS keyhole foraminotomy

Positioning as described above.

1. skin incision (use fluoro to locate the correct level for the incision)

A. incision 1 cm off midline on the side of the pathology at the level of the disc space

B. remove adhesive plastic barrier (e.g. Ioban®) from around the opening to prevent pieces from being dragged into the incision

2. avoid using a guidewire to reduce the risk of penetrating the interlaminar space. STAY LATERAL and insert the thinnest dilator. Dock the dilator on the lateral mass and insert progressively sized dilators

3. use Bovie to expose lateral lamina and medial facet joint. Start laterally where bone is more easily felt and there is little danger of penetrating the interlaminar space and injuring the spinal cord

4. use a straight curette to expose the inferior edge of the superior lateral lamina and the medial facet joint

5. drill off the medial inferior facet, to expose the superior facet of the level below

6. drill the medial superior facet until you are flush with the superior aspect of the pedicle below

7. this completes the bony work, the soft tissue work proceeds as above under open keyhole foraminotomy

Outcome

A number of large series have reported good or excellent outcome in the range of 90-96%²⁸⁴.

18.3.3. Thoracic disc herniation

Key concepts:

• comprise only 0.25% of herniated discs, and < 4% of operations for herniated disc

• usually occur at or below T8 (the more mobile portion of the thoracic spine)

• frequently calcified get CT through disc (may affect choice of surgical approach)

• primary indications for surgery: refractory pain, progressive myelopathy

• surgical treatment: laminectomy is usually not appropriate

Account for 0.25-0.75% of all protruded discs²⁸⁶. 80% occur between the 3rd and 5th decades. 75% are below T8 (the more mobile portion of the thoracic spine), with a peak of 26% at T11-12. 94% were centrolateral and 6% were lateral²⁸⁷. A history of trauma may be elicited in 25% of cases.

Most common symptoms: pain (60%), sensory changes (23%), motor changes (18%). With thoracic radiculopathy, pain and sensory disturbance is in a band-like distribution radiating anteriorly and inferiorly along the involved root’s dermatome. Motor involvement is difficult to document.

SURGICAL TREATMENT

INDICATIONS

Herniated thoracic discs requiring surgery are rare²⁸⁷. Indications: refractory pain (usually radicular, bandlike) or progressive myelopathy. Uncommon: symptomatic syringomyelia originating at level of disc herniation.

APPROACHES

Surgery for thoracic disc disease is problematic because of: the difficulty of anterior approaches, the proportionately tighter space between cord and canal compared to the cervical and lumbar regions, and the watershed blood supply which creates a significant risk of cord injury with attempts to manipulate the cord when trying to work anteriorly to it from a posterior approach. Herniated thoracic discs are calcified in 65% of patients considered for surgery²⁸⁷ (more difficult to remove from a posterior or lateral approach than non-calcified discs).

Open surgical approaches^{287, 288}:

1. posterior (midline laminectomy): primary indication is for decompression of posteriorly situated intracanalicular pathology (e.g. metastatic tumor) especially over multiple levels. There is a high failure and complication rate when used for single-level anterior pathology (e.g. midline disc herniation)

2. posterolateral

A. lateral gutter: laminectomy plus removal of pedicle.

B. transpedicular approach²⁸⁹

C. costotransversectomy (see below)

D. transfacet pedicle sparing

3. anterolateral (transthoracic)

4. lateral extracavitary

An option to open surgery is thoracoscopic surgery.

CHOICE OF APPROACH

For anterior approaches to the thoracic spine, see sections beginning on page 178.

Intraoperative SSEPs and MEPs may be helpful for cases of myelopathy.

For a laterally herniated thoracic disc without myelopathy: posterolateral approach with medial facetectomy is technically simple, and has generally good results. For a central disc herniation, or when myelopathy is present: transthoracic approach has the lowest incidence of cord injury with the best operative results (see Table 18-21). For anterior access, unless pathology is predominantly left-sided, a right-sided thoracotomy is preferred because the heart does not impede access.

COSTOTRANSVERSECTOMY

Indications: in the past this was often used to drain tuberculous spine abscess. It may be used for lateral disc herniation, biopsy of VB or pedicle, limited unilateral decompression of spinal cord from tumor or bone fragments, sympathectomy. Can be used at ≈ any T-spine level. Limitations: difficult to visualize anterior canal to access midline anterior pathology. Better for soft disc than for calcified central disc.

Involves resection of the transverse process and at least ≈ 4-5 cm of the posterior rib. A serious risk of this approach is interruption of a significant radicular artery which may compromise spinal cord blood supply (see Spinal cord vasculature, page 95). There is also a risk of pneumothorax which is less grave.

Booking the case - costotransversectomy

Also see defaults & disclaimers (page v).

1. position: prone, usually on chest rolls

2. equipment:

A. microscope (not used for all cases)

B. C-arm

3. implants: if post-op instability is anticipated, thoracic pedicle screws and possibly a cage (e.g. for fracture or tumor, not typically for disc herniation)

4. neuromonitoring: SSEP/MEP

5. blood availability: type and cross 2 U PRBC

6. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the back of the chest to remove a small piece of rib to permit removal of the herniated/calcified disc

B. alternatives: nonsurgical management, surgery from the side through the chest

C. complications: spinal cord injury with paralysis, lung complications including pneumothorax or hemothorax (blood or air outside lungs), possible seizures with MEPs

Surgical technique

The approach can be somewhat difficult due to the infrequent encounter with the anatomy by most neurosurgeons. Be prepared for a “deep, red hole, where everything initially looks the same and the bony anatomy is not easy to define”. With patience and persistence and the help of an anatomic model in the O.R., the surgeon can get his/her bearings. One of the most helpful landmarks is following the NVB (or just the nerve root) medially to the neural foramen.

In the O.R. before the skin incision, localizing x-rays are obtained; a spinal needle inserted between 2 spinous processes may be used as a marker.

Patient position: the approach is from the side of the pathology/symptoms; for central disc herniations a right-sided approach reduces risk of injury to artery of Adamkiewicz (located on the left in 80%, see page 96). Options:

1. lateral oblique, ≈ 30° elevated from straight prone, a “bean-bag” is good for stabilization. For a thin patient, the surgeon may stand in front of the patient (gives more horizontal angle of view - does not work as well with heavier patients due to mass of skin/muscle in the way laterally)

2. prone on chest rolls: the chest roll on the side of the pathology should be more medial to allow the shoulder and scapula to fall forward out of the way

Skin incision: options

1. curved paramedian skin incision: apex oriented away from the midline along the slight depression demarcating the junction of the lateral border of the paraspinal muscles with the ribs (≈ 6-7 cm lateral to midline) centered over the interspace of interest extending ≈ 3 vertebral bodies (VB) above and below. The incision is carried through the skin, subcutaneous fat, trapezius, and (for lower 6 thoracic levels, where most thoracic disc herniations occur) the latissimus dorsi, down to the ribs, and this musculocutaneous flap can be reflected medially as a unit

2. midline incision: need to extend 3-4 levels above and below the level of pathology to get an angle low enough to visualize posterior to the facet in order to access the posterior vertebral body. The inferior aspect can be curved laterally towards the side of pathology. Advantage: a laminectomy can more easily be performed if needed (if the angle does not provide adequate visualization, as a “bail-out” contingency, a facetectomy may be performed, and pedicle may even be removed to access inferior to the disc space. This usually permits easy decompression of the entire thecal sac. In the thoracic spine, stabilization is optional, and if chosen, unilateral pedicle screws and fusion are usually adequate)

Rib removal and thoracic exposure: for a simple biopsy or drainage of a small abscess, removal of only 1 rib may suffice. The rib to be removed is from the level inferior to the disc space to be accessed²⁹¹(e.g. remove the T4 rib to access T3-4). For most other pathologies, 2 or 3 ribs are often removed²⁹². To access a VB, the like-numbered rib and the rib below are removed.

There are a number of ligaments attached to the rib: the intercostal neurovascular bundle (NVB) courses medial to the superior costotransverse ligament which extends from the superior aspect of the rib to the transverse process of the level above. This ligament and the lateral costotransverse ligament are divided and the transverse process is rongeured off (the base of which lies on the lamina directly posterior to the pedicle). This exposes the rib anterior to the transverse process. The periosteum is incised on the rib from the angle of the rib to the costovertebral articulation, and by subperiosteal dissection around its circumference the pleura is dissected off the anterior surface of the rib. The NVB is dissected from the deep-inferior surface along with the periosteum. The rib is then transected laterally at the angle (≈ 5 cm lateral to the rib head) with rib shears, it is gripped with a clamp, and is rotated while the ligaments (including the radiate ligaments which attach the rib to the both the VB above and the VB below the disc space at the superior and inferior costal facet, respectively, except T1, 11 & 12 which only articulate with their like-numbered VB) are sharply dissected off the rib which is then removed. The removed rib material may be used for fusion substrate except in cases of tumor or infection. The pleura is then dissected from the deep surface of the adjacent ribs and VB (taking care not to injure the segmental vessels and to dissect the sympathetic trunk off the VB with the pleura). The pleura is then retracted laterally with a malleable ribbon or Deaver retractor.

The intervertebral foramen of interest may be located by following the NVB of the rib above proximally, the intercostal nerve (the ventral ramus of the nerve root at that level) enters between the two pedicles. The dura may then be exposed by enlarging the neural foramen by removing part of the pedicles with a high-speed drill and Kerrison rongeurs.

Instrumentation/fusion are rarely required for simple discectomy. Instability due to fracture, tumor, or extensive resection (e.g. with total facet takedown) necessitates surgical stabilization, typically with pedicle screws/rods extending 2 levels above and 2 levels below. Prior to closure, check for air leak by filling the opening with saline and having the anesthesiologist apply a valsalva maneuver. If an air leak is identified, a Cook catheter may be placed into the pleural space through the surgical exposure, or alternatively a chest tube is placed through a separate intercostal incision after the laminectomy wound is closed. A post-op CXR is obtained regardless of whether an air leak is identified.

TRANSPEDICULAR APPROACH

Drilling down the pedicle and removing a small amount of bone from the vertebral body, then pushing material from the epidural space into the defect created and removing it. Requires removal of just the rib head. Advantages: minimal risk of pneumothorax, more familiar anatomy. Disadvantages: requires instrumentation especially if done bilaterally, the angle is not very obliques so visualization of epidural space is minimal, may need to be done bilaterally if there are extensive bilateral components to the pathology.

Booking the case - transpedicular approach

Same as for costotransversectomy (see page 472).

TRANSTHORACIC APPROACH

Indications: thoracic disc disease, burst fractures of the thoracic spine, etc.

Advantages²⁹³:

• excellent anterior exposure (especially advantageous for multiple levels)

• little compromise of stability (due to supporting effect of rib cage)

• low risk of mechanical cord injury

Disadvantages:

• requires thoracic surgeon (or familiarity with thoracic surgery)

• some risk of vascular cord injury (due to sacrifice of intercostal arteries)

• definitive diagnosis may not be possible if it is uncertain prior to procedure

Possible complications:

• pulmonary complications: pleural effusion, atelectasis, pneumonia, empyema, hypoventilation

• CSF-pleural fistula

Booking the case - transthoracic spine surgery

Also see defaults & disclaimers (page v).

1. position: typically on the side, often on a beanbag

2. equipment:

A. microscope (not used for all cases)

B. C-arm

3. anesthesia: double lumen tube

4. implants: if post-op instability is anticipated, thoracic pedicle screws and possibly a cage (e.g. for fracture or tumor, not typically for disc herniation)

5. neuromonitoring: SSEP/MEP

6. blood availability: type and cross 2 U PRBC

7. some surgeons use chest surgeon for the approach, closure and for follow-up

8. consent (in lay terms for the patient - not all-inclusive):

A. procedure: surgery through the chest with removal of a small piece of rib to permit removal of the herniated/calcified disc

B. alternatives: nonsurgical management, surgery from the side or through the back

C. complications: spinal cord injury with paralysis, pneumothorax, possible seizures with MEPs

Key technical points

1. the services of an experienced thoracic surgeon are usually engaged

2. position: true lateral (facilitates intra-op localizing x-rays); approached from the more involved side. For the upper thoracic midline region, some prefer right-side-up to eliminate thoracic aorta from obstructing exposure and to reduce the possibility of encountering the artery of Adamkiewicz²⁹⁴, others prefer left-side-up to use the aorta as a landmark²⁹³ (for levels below the cardiophrenic angle, a left-sided approach is preferred because the inferior vena cava is difficult to mobilize)

3. usually one rib is resected; most often the rib of the vertebra immediately above the disc space desired (facilitates exposure). Multiple ribs may be resected to increase exposure

4. when removing the vertebral body (VB) (corpectomy, e.g. for osteomyelitis, especially Pott’s disease or for kyphoscoliosis)

A. the posterior cortex of the VB must be pulled anteriorly (e.g. with angled curettes) to avoid mechanical cord trauma

B. anterior fusion may be performed using the removed rib. If inadequate, fibula or iliac crest may be used

5. sizeable radicular arteries are spared. The intercostal nerve is used as a guide to the intervertebral foramen (nerve enters foramen superiorly and posteriorly)

6. the disc space is situated off the caudal aspect of the intervertebral foramen for most thoracic levels

7. one or two intervertebral arteries and veins usually have to be sacrificed; to minimize the risk of ischemic cord injury, cut them as close to the midline of the spine as possible (collaterals tend to lie on the lateral aspect of the spine)

8. the sympathetic chain is dissected off the VBs and is pushed posteriorly

LATERAL APPROACH

The same instrumentation used for lateral lumbar interbody fusion (see page 194) may be used to access the lateral thoracic bodies for thoracic disc herniation.

Check the pre-op MRI for the location of the aorta and to rule-out aortic aneurysm. Above T11, enter on the right side. The access retractor is “reversed” so that the center blade is positioned anteriorly and the shim is place so that as the retractor is expanded in the AP direction, the lateral blades move posteriorly, giving more access to the posterior disc space. In general, do not penetrate the contralateral anulus because of proximity of aorta. At or just below L12-L1, the diaphragm attached to the VB. A dual lumen endotracheal tube is not required. A chest tube is mandatory if there is an air leak, otherwise it is optional (a pigtail catheter may suffice).

18.4. Degenerative disc/spine disease

Since structures outside of the disc are usually also involved, the term degenerative spine disease (DSD) may be preferable to degenerative disc disease. Spondylosis is a non-specific term which may include degenerative spine disease. “Cervical spondylosis” is occasionally used synonymously with cervical stenosis (see Cervical spinal stenosis, page 485).

DSD is a progressive deterioration of the structures of the spine including:

1. disc abnormalities:

A. the proteoglycan content of the disc nucleus decreases with age

B. disc desiccation (loss of hydration) occurs

C. tears develop in the disc annulus and progress to internal disruption of the lamellar architecture. Herniation of the nucleus may occur from increased nuclear pressure under mechanical loads

D. mucoid degeneration and ingrowth of fibrous tissue ensues (disc fibrosis)

E. subsequently disc resorption occurs

F. there is a loss of disc space height and increased susceptibility to injury

2. facet joint abnormalities: hypertrophy and capsular laxity

3. osteophytes often form on the edges of the VB bordering the degenerated disc

4. spondylolisthesis: subluxation of one VB on another (see Spondylolisthesis below)

5. spondylolysis: alternative term for isthmic spondylolisthesis (see below), a failure of the neural arch due to a defect in the pars interarticularis which may present as spondylolisthesis. There may be a fibrous mass from the nonunion

6. hypertrophy of the ligamentum flavum

Clinical presentation:

1. the above abnormalities may produce spinal stenosis which can lead to neural compromise producing the following symptoms

A. radicular symptoms (more common in cervical spine than lumbar)

B. neurogenic claudication (lumbar) or spinal myelopathy (cervical)

2. discogenic pain (controversial) may be less prevalent in the late stages of DSD. May contribute to “musculoskeletal low back pain” but the actual pain generators here are not definitively identified

ETIOLOGY

The etiology of DSD is multifactorial and includes:

1. cumulative effects of microtrauma and macrotrauma to the spine

2. osteoporosis

3. cigarette smoking: several epidemiologic studies have shown that the incidence of back pain, sciatica and spinal degenerative disease is higher among cigarette smokers than among nonsmokers^{295, 296}

4. in the lumbar spine:

A. stresses on the spine including effects of excess body weight

B. loss of muscle tone (primarily abdominals and paraspinals) resulting in increased dependence on the bony spine for structural support

SPONDYLOLISTHESIS

Anterior subluxation of one vertebral body on another. Usually L5 on S1, the next most common is L4 on L5. The Meyerding^{297, 298} grading of subluxation in the sagittal plane is shown in Table 18-22.

Disc herniation and nerve root compression: It is rare for a herniated lumbar disc to occur at the level of the listhesis, however the disc may “roll” out as it is exposed and produce findings on MRI that have been termed a “pseudodisc”. It is more common to see a herniated disc at the level above the listhesis. If the listhesis does cause nerve root compression, it tends to involve the nerve exiting below the pedicle of the anteriorly subluxed vertebra. The compression is usually due to upward displacement of the superior articular facet of the level below together with disc material, and symptoms typically resemble neurogenic claudication, although true radiculopathy may sometimes occur.

Table 18-22 Spondylolisthesis grading

Grade	% subluxation*
I	< 25%
II	25-50%
III	50-75%
IV	75%-complete
spondyloptosis	> 100%

* % of the AP diameter of the VB

CLASSIFICATION OF SPONDYLOLISTHESIS

Type 1: dysplastic: congenital. Upper sacrum or arch of L5 permits the spondylolisthesis. No pars defect. 94% are associated with spinal bifida occulta. Some of these may progress (no way to identify these)

Type 2: isthmic spondylolisthesis AKA spondylolysis: a failure of the neural arch manifesting as a defect in the pars interarticularis (the neck of the “Scotty dog” on oblique LS-spine x-ray). May be seen in 5-20% of spine x-rays⁴. Three subtypes:

C. lytic: fatigue fracture or insufficiency fracture of pars. In the pediatric age group may occur in athletes (especially gymnasts or football players); in some this may be an exacerbation of a pre-existing defect, in others it may be a result of repetitive trauma

D. elongated but intact pars: possibly due to repetitive fractures and healing

E. acute fracture of pars Type 3: degenerative: due to long-standing intersegmental instability. Usually at

L4-5. No break in the pars. Found in 5.8% of men and 9.1% of women (many of whom are asymptomatic)⁴

Type 4: traumatic: due to fractures usually in areas other than the pars

Type 5: pathologic: generalized or local bone disease, e.g. osteogenesis imperfecta

ISTHMIC SPONDYLOLISTHESIS (SPONDYLOLYSIS) - PARS INTERARTICULARIS DEFECT

Presentation

Isthmic spondylolisthesis rarely produces central canal stenosis since only the anterior part of the spinal canal is shifted forward. May present with radiculopathy, with the nerve exiting under the pedicle at that level being the most vulnerable. May also present with low back. Many cases are asymptomatic.

Management⁴

1. lesions with sclerotic borders are usually well established with little chance of healing. Surgery is reserved for patients with neurologic deficit or incapacitating symptoms

2. lesions without sclerosis that show increased uptake on bone scan (indicating active lesion with potential for healing) or MRI high signal changes on T2WI²⁹⁹ or

STIR may heal in a rigid orthosis such as the Boston brace for ≥ 3 months

3. management of symptoms:

A. LBP only: treat with NSAIDs, PT

B. LBP with myelopathy, radiculopathy, or neurogenic claudication: surgical treatment³⁰⁰ (see Table 18-23 for surgical options)

4. in pediatrics: may be managed with TLSO and long course of PT (e.g. 6-9 months) for symptoms. Resumption of sports may be considered when symptoms subside, but recurrence should prompt elimination of athletics or consideration of surgery

SURGICAL CONSIDERATIONS

When surgery is indicated, Table 18-23 serves as a guide to the type of procedure.

Table 18-23 Surgical recommendations for spondylolisthesis

Nature of spondylolisthesis	Nature of problem	Type of procedure needed
degenerative	nerve root compression within confines of spinal canal	decompression (preserving facets)
	spinal stenosis at the level of spondylolisthesis	decompression; some advocate with intertransverse-process fusion301
	nerve root compression far lateral, outside confines of spinal canal	radical decompression (Gill procedure, see below) plus fusion
traumatic	(does not matter)	decompression plus fusion

Reduction of high-grade (grade III or IV) spondylolisthesis carries a risk of radiculopathy (e.g. L5 radiculopathy in cases of L5-S1 spondylolisthesis) in 50% of cases (some permanent) and may produce a cauda equina syndrome, probably from stretching nerve roots by distraction. The risk of nerve root injury with reduction of grade I or II spondylolisthesis is low.

Gill procedure: This procedure, and its modifications³⁰², consist of radical decompression of nerve roots including removal of the loose posterior elements and total facetectomy. This is often followed by fusion (posterolateral or interbody). Fusion rate may be enhanced with the use of internal fixation (e.g. transpedicular screw-rod fixation)³⁰³.

18.4.1. Spinal stenosis

Classified as:

1. central canal stenosis: narrowing of the AP dimension of the spinal canal. The reduction in canal size may cause local neural compression and/or compromise of the blood supply to the spinal cord (cervical) or the cauda equina (lumbar)

2. foraminal stenosis: narrowing of the neural foramen. May be the result of any combination of: foraminal disc protrusion, spondylolisthesis, facet hypertrophy, disc space collapse, hypertrophy of uncovertebral joints (cervical), synovial cyst

3. lateral recess stenosis (lumbar spine only): see page 484

Central canal stenosis

May be congenital (as in the achondroplastic dwarf), acquired, or most commonly acquired superimposed on congenital.

In the lumbar region, the syndrome of neurogenic claudication (see below) is well recognized. In the cervical region, cervical myelopathy and ataxia (from spinocerebellar tract compression) may be present. In 5%, lumbar and cervical stenoses are symptomatic simultaneously³⁰⁴. Symptomatic spinal stenosis in the thoracic region is rare³⁰⁵.

18.4.1.1. Lumbar spinal stenosis

Unless indicated otherwise, this discussion refers primarily to central canal stenosis

Key concepts:

• caused by hypertrophy of facets and ligamentum flavum, may be exacerbated by disc bulging or spondylolisthesis, may be superimposed on congenital narrowing

• most common at L4-5 and then at L3-4

• symptomatic stenosis produces gradually progressive back and leg pain with standing and walking that is relieved by sitting or lying (neurogenic claudication)

• symptoms differentiated from vascular claudication which is usually relieved at rest regardless of position

• usually responds to decompressive surgery (sometimes with fusion) or interspinous spacer

Symptomatic lumbar stenosis is most common at L4-5, then L3-4, L2-3 and lastly L5-S1306. It is rare at L1-2. Generally occurs in patients with congenitally shallow lumbar canal (see Normal LS spine measurements, page 480) with superimposed acquired degeneration in the form of some combination of facet hypertrophy, hypertrophy of the ligamentum flavum, protruding (and often calcified) intervertebral discs, and spondylolisthesis. First recognized as a distinct clinical entity producing characteristic symptoms in the 1950’s and 60’s^{307, 308}.

May be classified as³⁰⁹:

1. stable form of lumbar spinal stenosis: hypertrophy of facets and ligamentum flavum accompanied by disc degeneration and collapse

2. unstable: have the above with superimposed

A. degenerative spondylolisthesis: (see page 475) the unisegmental form

B. degenerative scoliosis: the multisegmental form

CLINICAL EVALUATION

PRESENTATION

Often presents as neurogenic claudication (NC), (claudicate: from Latin, claudico, to limp) AKA pseudoclaudication. To be differentiated from vascular claudication (AKA intermittent claudication) which results from ischemia of exercising muscles (see Table 18-24). NC characteristics: unilateral or bilateral buttock, hip, thigh or leg discomfort that is precipitated by standing or walking and characteristically relieved by a change in posture.

NC is thought to arise from ischemia of lumbosacral nerve roots, as a result of increased metabolic demand from exercise together with vascular compromise of the nerve root due to pressure from surrounding structures. NC is only moderately sensitive (≈ 60%) but is highly specific for spinal stenosis³¹¹. Pain may not be the major complaint, instead, some patients may develop paresthesias or LE weakness with walking. Some may complain of muscle cramping, especially in the calves.

Relief from symptoms: occurs with positions that decrease the lumbar lordosis which increases the diameter of the central canal (by reducing inward buckling of the ligamentum flavum) and distracts the facet joints (which enlarges the neural foramina. Favored positions include sitting, squatting and recumbency. Patients may develop “anthropoid posture” (exaggerated waist flexion). “Shopping cart sign” patients often can walk farther if they can lean forward e.g. as on a grocery cart. Riding a bicycle is also often well tolerated.

Table 18-24 Clinical features distinguishing neurogenic from vascular claudication³¹⁰

Feature	Neurogenic claudication	Vascular claudication
distribution of pain	in distribution of nerve (dermatomal)	in distribution of muscle group with common vascular supply (sclerotomal)
sensory loss	dermatomal distribution	stocking distribution
inciting factors	variable amounts of exercise, also with pro-longed maintenance of a given posture (65% have pain on standing at rest); coughing produces pain in 38%	reliably reproduced with fixed amount of exercise (e.g. distance ambulated) that decreases as disease progresses; rare at rest (27% have pain on standing at rest)
relief with rest	slow (often > 30 min), variable, usually positional (stooped posture or sitting often required, standing and resting is usually not sufficient)	almost immediate; not dependent on posture (relief of walking induced symptoms with standing is a key differentiating feature)
claudicating distance	variable day-to-day in 62%	constant day-to-day in 88%
discomfort on lifting or bending	common (67%)	infrequent (15%)
foot pallor on elevation	none	marked
peripheral pulses	normal; or if ↓ usually reduced only unilaterally	↓ or absent; femoral bruits are common
skin temp of feet	normal	decreased

NEUROLOGIC EXAM

The neurologic exam is normal in ≈ 18% of cases (including normal muscle stretch reflexes and negative straight leg raising). Absent or reduced ankle jerks and diminished knee jerks is common³¹¹. Pain may be reproduced by lumbar extension.

DIFFERENTIAL DIAGNOSIS

1. vascular insufficiency: (AKA vascular or intermittent claudication) see above

2. hip disease: trochanteric bursitis (see below), degenerative joint disease

3. disc herniation (lumbar or thoracic)

4. facet joint pain (controversial): may respond to medial branch block (therapeutic & diagnostic)

5. Baastrup’s syndrome³¹²: AKA arthrosis interspinosa. Radiographically: contact of adjacent spinous processes (“kissing spines”) with enlargement, flattening and reactive sclerosis of apposing interspinous surfaces. Produces localized midline lumbar pain & tenderness on back extension relieved by flexion, local anesthetic injection or partial excision of the involved spinous processes

6. juxtafacet cyst: see page 456

7. arachnoiditis

8. intraspinal tumor

9. Type I spinal AVM (spinal dural AVM): see page 507

10. diabetic neuritis: with this, the sole of the foot is usually very tender to pressure from the examiner’s thumb

11. delayed onset muscle soreness (DOMS): onset usually 12-48 hours after beginning a new activity or changing activities (NC occurs during the activity). Symptoms typically peak within 2 days and subside over several days

12. inguinal hernia: typically produces groin pain

13. functional etiologies

Trochanteric bursitis (TBS) and degenerative arthritis of the hip are also included in the differential diagnosis of NC^{313, 314}. Although TBS may be primary, it can also be secondary to other conditions including lumbar stenosis, degenerative arthritis of the lumbar spine or knee, and leg length discrepancy. TBS produces intermittent aching pain over the lateral aspect of the hip. Although usually chronic, it occasionally may have acute or subacute onset. Pain radiates to lateral aspect of thigh in 20-40% (so called “pseudoradiculopathy”), but rarely extends to the posterior thigh or as far distally as the knee. There may be numbness and paresthesia-like symptoms in the upper thigh which are usually not dermatomal in distribution. Like NC, the pain may be triggered by prolonged standing, walking and climbing, but unlike NC it is also painful to lie on the affected side. Localized tenderness over the greater trochanter can be elicited in virtually all patients, with maximal tenderness at the junction of the upper thigh and greater tro-chanter. Pain increases with weight bearing (and is often present from the very first step, unlike NC) and with certain hip movements, especially external rotation (over half the patients have a positive Patrick-FABERE test, see page 444), and rarely with hip flexion/extension. Treatment includes NSAIDs, local injection of glucocorticoid (usually with local anesthetic), physical therapy (with stretching and muscle strengthening exercises) and local application of ice. No controlled studies have compared these.

ASSOCIATED CONDITIONS

1. congenital:

A. achondroplasia

B. congenitally narrowed canal

2. acquired:

A. spondylolisthesis

B. acromegaly

C. post-traumatic

D. Paget’s disease (see Paget’s disease, page 498)

E. ankylosing spondylitis: see page 502

F. ossification of the ligamentum flavum: more common in East Asians, rare in Caucasians³¹⁵. Often, but not always, associated with OPLL³¹⁶

DIAGNOSTIC EVALUATION

RADIOGRAPHIC EVALUATION

Comparison of modalities:

Lumbosacral spine x-rays: may disclose spondylolisthesis. AP diameter of canal is usually narrowed (congenitally or acquired) (see Normal LS spine measurements below) whereas the interpediculate distance (IPD) may be normal³¹⁰. Oblique films may demonstrate pars defects.

CT scan (either routine, or following water-soluble myelography): classically shows “trefoil” canal (cloverleaf shaped, with 3 leaflets). CT also demonstrates AP canal diameter, hypertrophied ligaments, facet arthropathy, and bulging annulus or herniated disc. CT is poor for demonstrating spondylolisthesis although the pars defect may be seen.

Myelogram: lateral films often show “washboard pattern” (multiple anterior defects), AP films often show “wasp-waisting” (narrowing of dye column), may also show partial or complete (especially in prone position) block. May be difficult to perform LP if stenosis is severe (poor CSF flow and difficulty avoiding nerve roots).

MRI: demonstrates impingement on neural structures and loss of CSF signal on T2WI at severely stenotic levels. MRI is poor for visualizing bone which contributes significantly to the pathology (may be helpful for surgical planning). Good for evaluating nerve impingement due to spondylolisthesis (possibly better than myelogram/CT) and juxtafacet cysts. Asymptomatic abnormalities are demonstrated in up to 33% of asymptomatic patients 50-70 years old³⁰⁶.

Table 18-25 Normal AP diameter on lateral plain film

(from spinolaminar line to posterior vertebral body)³¹⁷

average (normal)	22-25 mm
lower limits of normal	15 mm
severe lumbar stenosis	< 11 mm

Table 18-26 Normal measurements on CT³¹⁸

AP diameter	≥ 11.5 mm
interpediculate distance (IPD)	≥ 16 mm
canal cross-sectional area	≥ 1.45 cm2
ligamentum flavum thickness319	≤ 4-5 mm
height of lateral recess (see below)	≥ 3 mm

NORMAL LS SPINE MEASUREMENTS

Normal dimensions of the lumbar spine are shown in Table 18-25 for plain film and Table 18-26 for CT.

Interpediculate distance (IPD): The transverse diameter of the spinal canal. On plain AP x-ray of lumbar spine, an IPD < 25 mm suggests stenosis. Average normal IPDs in the lumbar and lower thoracic spine appears in Table 18-27. An approximation for the lumbar spine is given in Eq 18-1.

ADJUNCTS TO RADIOGRAPHIC EVALUATION

“Bicycle test”: patients with NC can usually tolerate longer periods of exercise on a bicycle than patients with intermittent (vascular) claudication because the position in bicycling flexes the waist.

Ratio of ankle to brachial blood pressure (A:B ratio): > 1.0 is normal; mean of 0.59 in patients with intermittent claudication; 0.26 in patients with rest pain; < 0.05 indicates impending gangrene.

Vascular lab studies (e.g. Doppler) may assist in identifying vascular insufficiency.

EMG with NCV may show multiple nerve-root abnormalities bilaterally.

Table 18-27 Normal interpediculate distance (IPD) on AP LS film³²⁰

Level	IPD (mm)*
T10	16-22
T11	17-24
T12	19-27
L1	21-29
L2	21-30
L3	21-31
L4	21-33
L5	23-37

TREATMENT

In one study of 27 unoperated patients, 19 remain unchanged, 4 improved, and 4 worsened (mean follow-up: 49 months; range: 10-103 months)³²¹. NSAIDs (recent evidence suggests acetaminophen may be as effective) and physical therapy are the mainstays of nonsurgical management.

Surgical decompression is undertaken when symptoms become severe in spite of medical management. The goals of surgery are pain relief, halting progression of symptoms, and possibly reversal of some existing neurologic deficit. Most authors do not consider surgery unless the symptoms have been present > 3 months, and most patients who have surgery for this have symptoms of > 1 year duration.

SURGERY

Surgical options

1. laminectomy: posterior (direct) decompression of central canal and neural foramina without or with fusion. Fusion options:

A. posterolateral fusion ± pedicle screw/rod fixation

B. interbody fusion: generally not done as a “stand-alone” (i.e. usually requires additional stabilization, options here include: pedicle screws, facet screws, facet dowels, spinous process clamp…)

1. posterior lumbar interbody fusion (PLIF): usually bilateral graft placement (see page 193)

2. transforaminal lumbar interbody fusion (TLIF): unilateral graft placement though a facet take-down on that side (see page 193)

2. procedures to increase disc space height and thereby indirectly decompress neural foramina without direct decompression

A. anterior lumbar interbody fusion (ALIF): through laparotomy (see page 195)

B. lateral lumbar interbody fusion: some techniques trademarked as extreme lateral interbody fusion (XLIF™) or direct-lateral (DLIF™): see page 194

C. axial lumbar interbody fusion (Ax-LIF): L5-S1 only (see page 195)

3. limitation of extension by interspinous spacer: e.g. X-Stop® (see below)

BOOKING THE CASE - LUMBAR LAMI ± FUSION FOR STENOSIS

Also see defaults & disclaimers (page v).

1. position: prone

2. implants: for fusions, schedule with the vendor for the desired implants and associated instrumentation

3. consent (in lay terms for the patient - not all-inclusive):

A. procedure: through the back to remove bone, ligament and any other tissue that is pressing on the nerve(s). If a fusion is to be done, then typically this will be accomplished using screws, rods and small cages, as required

B. alternatives: nonsurgical management

C. complications: (usual spine surgery complications - see page v) plus there might not be the amount of pain relief desired (back pain does not respond as well to surgery as nerve-root pain). If implants are used, then there can be problems with them including breakage, migration (slippage), or undesirable positioning which may require further surgery

Choosing which procedure to use

Although beyond the scope of this book, items that factor into consideration when choosing which procedure to use include:

1. consider indirect decompression (lateral interbody fusion (e.g. XLIF® or DLIF®), interspinous decompression (e.g. X-Stop):

A. when foraminal stenosis appears to be the dominant problem (e.g. with loss of disc space height, facet hypertrophy, on the concave side of a scoliotic curve)

B. previous spine surgery that might make exposure of the nerves more difficult or risky

2. consider direct decompression (e.g. laminectomy)

A. “pinpoint” central canal stenosis especially when disc height and neural foramina are well preserved

B. where the majority of the compression is due to a focal, correctable lesion, such as a herniated disc, synovial cyst, intraspinal tumor

3. consider motion-preservation surgery

A. when a fusion is undertaken at a level and the adjacent level is already starting to show some degenerative changes that have not yet reached a surgical magnitude. Motion preservation at this adjacent segment theoretically shield it from some of the transmitted stresses from the fused level

4. situations where a fusion should be considered in addition to direct or indirect decompression of the nerves:

A. spondylolisthesis (especially > Grade I)

B. dynamic instability on flexion/extension lateral lumbar spine x-rays

C. expectation that the decompression will destabilize the spine (e.g. facet takedown for a TLIF)

D. multiply recurrent herniated disc (when this is the third or more operation for the same disc

E. controversial situations:

1. e.g. a “black disc” on MRI with positive concordant discogram at this level: fusion without decompression has been advocated when there is no neural compression

Laminectomy/laminotomy - surgical technique

Posterior approach with removal of the spines and lamina of affected levels (surgical “unroofing”), along with the associated ligamentum flavum. Individual nerve roots are palpated for compression within their neural foramen, with foraminotomies performed at appropriate levels. Doing a total L4 laminectomy for stenosis allows access to the L4-5 foramen, and the upper part of the L5-S1 foramen. If, in addition, the lower part of L3 is also removed, access is gained to the inferior pedicle of L3 and thus the L3-4 neural foramen. Undercutting the superior articular facet is often necessary to decompress the nerves in the foramen (see Lateral recess syndrome, page 484). Treatment of moderate stenosis at adjacent levels appears warranted as these levels have been shown to have a significant likelihood of becoming symptomatic later²²¹.

Alternatively, laminotomies (as opposed to laminectomies) may be performed in cases where the central canal has a normal AP diameter, but the lateral canal gutter is stenotic^{322, 323}. Multilevel subarticular fenestrations are another slight variation on this theme³²⁴.

Position: either of the following is acceptable

1. prone: on a frame or chest rolls or knee-chest position to decompress the abdomen to decrease venous pressure and thus reduce bleeding

2. lateral decubitus position: if there is no laterality to symptoms, right lateral decubitus (left-side-up) is easier for most right-handed surgeons to use angled Kerrison rongeur parallel to nerve roots

Minimally invasive spine surgery (MISS) decompression

Usually a “mini-open” technique using ≈ 1” incisions and expandable retractors.

1. options include bilateral laminotomies (see above)

2. bilateral decompression through a unilateral laminotomy

A. entry site: 3.5-4 cm off the midline to permit the needed angle

B. when using a retractor with an “open side” orient the retractor with the open side facing laterally (e.g. with the Nuvasive Maxcess® place the handles medially) to permit the angle needed for contralateral decompression

C. the laminectomy and facet takedown (usually for a TLIF) are done

D. open the ligamentum flavum on the side you’re working on, to visualize the posterior extent of the spinal canal, to permit finding the plane between the posterior part of the ligamentum flavum and the undersurface of the bone

E. the ligamentum flavum is left in place on the contralateral side to protect the dura during drilling

F. complete the decompression and disc removal on the side you’re working on

G. the undersurface of the bone (spinous process and contralateral lamina) are then drilled to decompress the contralateral side

H. once the undersurface of the contralateral posterior canal has been drilled, the ligamentum flavum is removed with pituitary rongeurs. It is possible to even do a contralateral foraminotomy at this point (curved Kerrison rongeurs are very helpful for this)

I. pedicle screws are placed through the open side, and then percutaneously through the contralateral side

Interspinous process decompression/stabilization/fusion

Interspinous spacers (e.g. X-Stop™ (Medtronic)) limit extension at 1 or 2 levels (without fusion), preventing narrowing of the associated neural foramen, and may also off-load the facet joints and even the disc. “Success rate”: 63% at 2 years. This device may be used as a standalone.

Interspinous plates (e.g. Aspen® (Lanx), Affix™ (Nuvasive), Spire® (Medtronic)) clamp across two spinous processes to fixate them (unlike X-Stop™ which just limits extension). The Aspen® clamps have a space for a graft which may optionally be used to promote fusion between the spinous processes. Interspinous plates may be used to augment other constructs e.g. lateral interbody fusion³²⁵, but are notintended for standalone use. Biomechanical stability is reported to be similar to bilateral pedicle screws in flexion, and unilateral pedicle screws in lateral bending³²⁶.

Contraindications (includes exclusionary criteria from the IDE study):

1. instability at level considered for procedure: spondylolisthesis > Grade 1 or scoliosis with Cobb angle ≥ 25° (see page 430)

2. cauda equina syndrome

3. acute fracture of the spinous process

4. bilateral pars defects (disconnects spinous process from the anterior elements)

5. osteoporosis. Contraindications per the IDE: DEXA scan (see page 992) with spine or hip T-score < –2.5 (i.e. more than 2.5 SD below the mean for normal adults) in the presence of ≥ 1 fragility fractures. Concerns: spinous process fracture at the time of insertion, or late subsidence due to microfractures. However, Kondrashov³²⁷ interprets a T-score < –2.5 anywhere as indicative of osteoporosis (even without fragility fractures). Options here include:

A. augmenting the spinous processes by injecting ≈ 0.5-1 cc of PMMA into each spinous process (SP) with a 13 Ga needle inserted ≈ halfway into the SP on lateral fluoro³²⁷ prior to dilating the interspace or placing the X-Stop. Verify central position within SP on AP fluoro, and monitor injection on fluoro

B. use of an X-Stop made of PEEK (modulus of elasticity of PEEK is closer to bone than titanium) - available now in Europe, soon in the U.S.

6. ankylosed level (i.e. already fused)

7. L5-S1 level: the spinous process of S1 is usually too small (not usually an issue since symptomatic stenosis at L5-S1 is rare)

8. age < 50 years: not studied in IDE investigation

Surgical pointers:

1. it is critical that the spacer sit in the anterior third of the spinous process

2. results may be better with the patient awake, under local anesthesia, lying on their side in a position that they feel is relieving their pain (thus opening up the critical levels). This may reduce the risk of undersizing the prosthesis

Post-op (based on manufacturer’s recommendations):

1. to avoid spinous process stress fracture: build-up physical activity gradually

2. 1st 6 weeks post-op: no spine hyperextension, no heavy lifting. Minimize stair climbing

3. initially, walking (for < 1 hour) is recommended as long as it is comfortable

4. at 2 weeks post-op: cycling (stationary or bicycle) may be added

5. 6 months post-op: may add sports such as swimming, golf, racquetball, tennis, running or jogging

Progression of spondylolisthesis

May occur without decompression, but is more common following surgery³²⁸. However, lumbar instability following decompressive laminectomy is rare (only ≈ 1% of all laminectomies for stenosis will develop progressive subluxation). Fusion is rarely required to prevent progression of subluxation with degenerative stenosis³²⁹.

Stability (without need for instrumentation) is thought to be maintained if > 50-66% of the facets are preserved during surgery and the disc space is not violated (maintains integrity of anterior and middle column). Younger or more active patients are at higher risk of subluxing.

One approach is to obtain flexion/extension x-rays pre-op, and follow patients after decompression. Those who develop symptomatic slippage post-op are treated by fusion, possibly in conjunction with spinal instrumentation.

INSTRUMENTATION AND/OR FUSION

PRACTICE GUIDELINE 18-13 FUSION IN PATIENTS WITH LUMBAR STENOSIS WITHOUT SPONDYLOLISTHESIS

Level III³³⁰:

• in situ posterolateral fusion is not recommended following decompression in patients with lumbar stenosis in whom there is no evidence of preexisting spinal instability or likely iatrogenic instability due to facetectomy

• in situ posterolateral fusion is recommended in patients with lumbar stenosis in whom there is evidence of spinal instability

• the addition of pedicle-screw instrumentation is not recommended in conjunction with posterolateral fusion following decompression

PRACTICE GUIDELINE 18-14 FUSION IN PATIENTS WITH LUMBAR STENOSIS AND SPONDYLOLISTHESIS

Level II³³¹: posterolateral fusion is recommended for patients with stenosis and associated degenerative spondylolisthesis who require decompression

Level III³³¹: pedicle screw fixation as an adjunct to posterolateral fusion should be considered in patients with stenosis and spondylolisthesis in cases where there is pre-op evidence of spinal instability* or kyphosis* at the level of the spondylolisthesis or when iatrogenic instability is anticipated

* the definition of “instability” and “kyphosis” varies, and has not been standardized

Fusion may accelerate degenerative changes at adjacent levels. Some surgeons recommend fusion at levels of spondylolisthestic stenosis^{221, 309}. Patients with combined degenerative spondylolisthesis, stenosis, and radiculopathy may be reasonable candidates for fusion¹.

BRACE THERAPY

PRACTICE GUIDELINE 18-15 BRACE THERAPY AS AN ADJUNCT TO OR INSTEAD OF LUMBAR FUSION

Level II³³²:

• short-term use (1-3 weeks) of a rigid lumbar support is recommended for treatment of LBP of relatively short duration (< 6 months)

• bracing in patients with LBP > 6 months duration is not recommended because it has not been shown to have long-term benefit

PRACTICE GUIDELINE 18-15 BRACE THERAPY AS AN ADJUNCT TO OR INSTEAD OF LUMBAR FUSION

Level III³³²:

• lumbar braces may reduce the number of sick days due to LBP among workers with a previous lumbar injury. Braces are not recommended for LBP in the general working population

• the use of pre-op bracing or transpedicular external fixation as tools to predict outcome for lumbar fusion is not recommended

OUTCOME

Morbidity/mortality

Risk of in-hospital mortality is 0.32%³¹¹. Other risks include: unintended durotomy (see page 450) (0.32%³¹¹ to ≈ 13%^{329, 333}), deep infection (5.9%), superficial infection (2.3%), and DVT (2.8%) (see also Risks of lumbar laminectomy on page 449).

Nonunion

Risk factors for nonunion (does not necessarily correlate with success of operation):

1. cigarette smoking delays bone healing and increases the risk of pseudoarthrosis following spinal fusion procedures, especially in the lumbar spine²⁹⁵

2. number of levels: in lumbar fusions, fusing 2 levels resulted in increased non-union rates compared to fusing 1 level³³⁴

3. NSAIDs: controversial

A. short-term (≤ 5 days) post-op use: high-dose ketorolac (120-240 mg/d) was associated with increased risk of nonunion, but low-dose ketorolac (≤ 110 mg/d), and celecoxib (200–600 mg/d) or rofecoxib (50 mg/d) were not³³⁴

B. some feel that long-term NSAID use does lower fusion rate³³⁵

Success of operation

No randomized study exists comparing surgery to “conservative” treatment. Patients with a postural component to their pain had much better results (96% good result) than those without a postural component (50% good results), and the relief of leg pain was much more successful than relief of back pain³³⁶. Surgery is most likely to reduce LE pain and improve walking tolerance¹.

Surgical failure may be divided into two groups:

1. patients with initial improvement who develop recurrent difficulties. Although short-term improvement after surgery is common, many patients progressively deteriorate over time³³⁷. One study found a 27% recurrence of symptoms after 5 years follow-up²²¹ (30% due to restenosis at the operated level, 30% due to stenosis at a new level; 75% of these patients respond to further surgery). Other etiologies include: development of herniated lumbar disc, development of late instability, coexisting medical conditions

2. patients who fail to have any post-op pain relief (early treatment failures). In one series of 45 such patients³³⁸:

A. the most common finding was a lack of solid clinical and radiographic indications for surgery (e.g. non-radicular LBP coupled with modest stenosis)

B. technical factors of surgery had less influence on outcome, with the most common finding being failure to decompress the lateral recess (which requires judicious medial facet resection or undercutting the superior articular facet)

C. other diagnoses (e.g. arachnoiditis), missed diagnosis (e.g. spinal AVM)

Long term outcome: Literature review³¹¹ with long-term follow-up found good or excellent outcome after surgery with a mean of 64% (range: 26-100%). A patient satisfaction survey indicated that 37% were much improved and 29% somewhat improved (total: 66%) post-op³³⁹. A prospective study found a success rate of 78-88% at 6 wks and 6 months, which dropped to ≈ 70% at 1 year and 5 yrs³⁴⁰. Success rates were slightly lower for lateral recess syndrome (see below).

LATERAL RECESS SYNDROME

A variant of lumbar stenosis³⁴¹. Lateral recess: the “gutter” alongside the pedicle which the nerve root enters just proximal to its exit through the neural foramen (see Figure 18-4). It is bordered anteriorly by the vertebral body, laterally by the pedicle, and posteriorly by the superior articular facet of the inferior vertebral body. Hypertrophy of this superior articular facet compresses the nerve root. L4-5 is the most commonly involved facet.

PRESENTATION

Patients develop unilateral or bilateral leg pain predominantly when walking or standing, and usually obtain relief by squatting, sitting with the waist flexed, or lying in the fetal position. Painful burning paresthesias of the lower extremities are also described. Valsalva maneuvers usually do not exacerbate the pain. The time course is usually gradually progressive over many months to years.

In comparison, a HLD usually causes increased pain on sitting, has a more abrupt onset, has pain on straight leg raising, and is worsened by Valsalva maneuvers.

The neurologic exam may be normal (including straight leg raising). Achilles reflexes may be absent.

Figure 18-4 Schematic axial CT through the L4-5 facet joint showing the lateral recesses (normal on patient’s right, stenotic on left)

EVALUATION

High resolution CT scan best defines the bony anatomy of the lateral recess (see Figure 18-4 and Table 18-28).

MRI or water soluble contrast myelography is recommended when surgery is contemplated. Characteristic finding: flattening of nerve root as it passes beneath the hypertrophied facet joint.

Table 18-28 Dimensions of lateral recess on CT (bone windows)

Lateral recess height	Degree of lateral recess stenosis
3-4 mm	borderline (symptomatic if other lesion co-exists, e.g. disc bulging)
< 3 mm	suggestive of lateral recess syndrome
< 2 mm	diagnostic of lateral recess syndrome

TREATMENT

Conservative treatment with lumbosacral brace may be attempted.

Surgical therapy

Indicated for unresponsive cases. Consists of laminectomy and partial (typically medial one third) facetectomy. Requires removal of the hypertrophied portion of the facet dorsal to the involved nerve root, either by undercutting, or by reducing overhanging hypertrophied facet elements until they are flush with the pedicle.

18.4.1.2. Cervical spinal stenosis

“Cervical spondylosis” is occasionally used synonymously with cervical spinal stenosis. However, spondylosis usually implies a more widespread age-related degenerative condition of the cervical spine including various combinations of the following:

1. congenital spinal stenosis (the “shallow cervical canal”³⁴²)

2. degeneration of the intervertebral disc producing a focal stenosis due to a “cervical bar” which is usually a combination of:

A. osteophytic spurs (“hard disc” in neurosurgical jargon)

B. and/or protrusion of intervertebral disc material (“soft disc”)

3. hypertrophy of any of the following (which also contributes to canal stenosis):

A. lamina

B. dura

C. articular facets

D. ligaments, including

1. increased stenosis in extension is more common than with flexion (based on MRI studies³⁴³ and cadaver studies), largely due to posterior inbuckling of ligamentum flavum³⁴⁴

2. posterior longitudinal ligament: may include ossification of the posterior longitudinal ligament (OPLL)³⁴⁵ (see page 504). May be segmental or diffuse. Often adherent to dura

3. ossification of the ligamentum flavum³¹⁶ (yellow ligament)

4. subluxation: due to disc and facet joint degeneration

5. altered mobility: severely spondylotic levels may be fused and are usually stable, however there is often hypermobility at adjacent or other segments

6. telescoping of the spine due to loss of height of VBs → “shingling” of laminae

7. alteration of the normal lordotic curvature³⁴⁶ (NB: the amount of abnormal curvature did not correlate with the degree of myelopathy)

A. reduction of lordosis: including

1. straightening

2. reversal of the curvature (kyphosis): may cause “bowstringing” of the spinal cord across osteophytes

B. exaggerated lordosis (hyperlordosis): the least common variant (may also cause bowstringing)

Although the majority of individuals > 50 yrs old have radiologic evidence of significant degenerative disease of the cervical spine, only a small percentage will experience neurologic symptoms³⁴⁷.

EVALUATION

Cervical spinal stenosis is suggested on plain films when the spinolaminar line is close to the posterior margin of the lateral masses.

CLINICAL

The condition generally tends to produce three types of clinical problems²⁴⁴:

1. nerve root compression may cause radicular complaints

2. spinal cord compression may cause myelopathy. Some stereotypical syndromes are presented below (see Cervical spondylotic myelopathy (CSM) below)

3. pain and paresthesias in the head, neck and shoulders with little or no suggestion of radiculopathy nor abnormal physical findings. This group is the most difficult to treat, and often requires a good physician-patient relationship to decide if surgical treatment should be undertaken in an attempt to provide relief

CERVICAL SPONDYLOTIC MYELOPATHY (CSM)

Cervical spondylosis is the most common cause of myelopathy in patients > 55 yrs of age³⁴⁸. Cervical spondylotic myelopathy (CSM) develops in almost all patients with ≥ 30% narrowing of the cross-sectional area of the cervical spinal canal³⁴⁹ (although some patients with severe cord compression do not have myelopathy^{350, 351}).

PATHOPHYSIOLOGY

Pathogenesis is controversial. Theories include the following alone or in combination:

1. direct cord compression between osteophytic bars and hypertrophy or infolding of the ligamentum flavum, especially if superimposed on congenital narrowing or cervical subluxations

2. ischemia due to compression of vascular structures³⁵² (arterial deprivation³⁵³ and/or venous stasis³⁵⁴)

3. repeated local cord trauma by normal movements in the presence of protruded discs and/or osteophytic (spondylotic) bars (cord and root injuries³⁵⁵)

A. cephalad/caudad movement with flexion extension³⁵⁶

B. anterior/posterior traction on the cord by dentate ligaments³⁵⁷ & nerve roots

C. diameter of spinal canal varies during flexion and extension

1. increased stenosis is more common in extension (see above)

2. unstable segments may sublux (so-called pincer mechanism)³⁵⁸

Histologically³⁵⁹, there is degeneration of the central grey matter at the level of compression, degeneration of the posterior columns above the lesion (particularly in the anteromedial portion), and demyelination in the lateral columns (especially the corticospinal tracts) below the lesion. Anterior spinal tracts are relatively spared. There may be atrophic changes in the ventral and dorsal roots and neurophagia of anterior horn cells.

CLINICAL

Gait disturbance, often with LE weakness or stiffness, is a common early finding in CSM³⁶¹. Cervical pain and mechanical signs are uncommon in cases of pure myelopathy. See Table 18-29 for the frequency of symptoms in CSM in one series. In most cases the disability is mild, and the prognosis for these is good. CSM is rare in age < 40 years.

Table 18-29 Frequency of symptoms in CSM (37 cases³⁶⁰)

Finding	%
pure myelopathy	59%
myelopathy + radiculopathy	41%
reflexes
hyperreflexia	87%
Babinski	54%
Hoffman	13%
sensory deficits
sensory level	41%
posterior column	39%
dermatomal arm	33%
paresthesias	21%
positive Romberg	15%
motor deficits
arm weakness	31%
paraparesis	21%
hemiparesis	18%
quadriparesis	10%
Brown-Séquard	10%
muscle atrophy	13%
fasciculations	13%
pain
radicular arm	41%
radicular leg	13%
cervical	8%
spasticity	54%
sphincter disturbance	49%
cervical mechanical signs	26%

Table 18-30 Modified JOA score for cervical myelopathy³⁶²*

Score	Description
Upper extremity (UE) motor dysfunction
0	unable to feed self
1	unable to use knife & fork; can eat with spoon
2	can use knife & fork with much difficulty
3	can use knife & fork with slight difficulty
4	none (normal)
Lower extremity (LE) motor dysfunction
0	unable to walk
1	can walk on flat surface with walking aid
2	can walk up and/or down stairs with handrail
3	lack of smooth and stable gait
4	none (normal)
Sensory deficit
0	UE	severe sensory loss or pain
1		mild sensory loss
2		none (normal)
0	LE	severe sensory loss or pain
1		mild sensory loss
2		none (normal)
0	trunk	severe sensory loss or pain
1		mild sensory loss
2		none (normal)
Sphincter dysfunction
0	unable to void
1	marked voiding difficulty (retention)
2	some voiding difficulty (urgency or hesitation)
3	none (normal)

* total score ranges from 0 to 17 (normal)

Grading

1. the modified Japanese Orthopaedic Association scale (mJOA) (Table 18-30) is a valid and reliable grading system, although it is non-specific

2. Neck Disability Index 535: a 10 question survey similar to the Oswestry Disability Index for the lumbar spine (see page 430). Mild disability is defined as a score of 10-28%, moderate = 30-48%, severe = 50-68%, complete ≥ 72%

3. other commonly used scales (not tested for validity or reliability):

A. Nurick³⁶³ (see page 505)

B. Harsh

Natural history

The time course of symptoms is highly variable and unpredictable. In ≈ 75% of cases of CSM, there is progression either in a stepwise fashion (in one third) or gradually progressive (two thirds)³⁶⁴. In some series, the most common pattern was that of an initial phase of deterioration followed by a stabilization that typically lasts for years and may not change thereafter^{365, 366}. In these cases, the degree of disability may be established early in the course of CSM. Others disagree with such a “benign” outlook and cite that over 50% of cases continue to deteriorate with conservative treatment³⁴⁷. Sustained spontaneous improvement is probably rare³⁴⁸.

In patients < 75 years age and mJOA score > 12 (mild myelopathy), the clinical condition remains stable in 3-years of follow-up (Class I)³⁶⁷ (however, these patients can still have significant disability that can respond to surgery). Patients with stenosis without myelopathy who have electrodiagnostic abnormalities or clinical radiculopathy are at risk of developing myelopathy (Class I)³⁶⁷. Longstanding severe stenosis over many years may cause irreversible deficit due to necrosis in gray and white matter (Class III)³⁶⁷.

Motor

Findings can be due to cord (UMN) and/or root (LMN) compression. The earliest motor findings are typically weakness in the triceps and hand intrinsics³⁶². There may be wasting of the hand muscles³⁶⁸. Slow, stiff opening and closing of the fists may occur³⁶⁹. Clumsiness with fine motor skills (writing, buttoning buttons…) is common.

There is often proximal weakness of the lower extremities (mild to moderate iliopsoas weakness occurs in 54%) and spasticity of the LEs.

Sensory

Sensory disturbance may be minimal, and when present are often not radicular in distribution. There may be a glove-distribution sensory loss in the hands³⁷⁰. A sensory level may occur a number of levels below the area of cord compression.

LEs often exhibit loss of vibratory sense (in as many as 82%), and occasionally have reduced pinprick sensation (9%) (almost always restricted to below the ankle). Compression of the spinocerebellar tract may cause difficulty running. Lhermitte’s sign was present in only 2 of 37 cases. Some patients may present with a prominence of posterior column dysfunction (impaired joint position sense and 2 point discrimination)³⁷¹.

Reflexes

In 72-87%, reflexes are hyperactive at a varying distance below the level of stenosis. Clonus, Babinski’s sign (see page 116) or Hoffman’s sign (see page 116) may also be present. Dynamic Hoffman’s sign³⁷² may be more sensitive: test for Hoffman’s sign during multiple cervical flexion and extension movements as tolerated by the patient. 94% of asymptomatic individuals with Hoffman’s reflex will have significant spinal cord compression on MRI³⁷³. Inverted radial reflex: flexion of the fingers in response to eliciting the brachioradialis reflex, said to be pathognomonic of CSM³⁷⁴.

A hyperactive jaw jerk indicates upper motor neuron lesion above the midpons, and distinguishes long tract findings due to pathology above the foramen magnum from those below (e.g. cervical myelopathy): not helpful if absent (a normal variant). Primitive reflexes (grasp, snout, rooting) are not reliable localizing signs (except perhaps the grasp reflex) of frontal lobe pathology.

Sphincter

Urinary urgency and frequency are common in CSM as well as in the aging population. Urinary incontinence is rare. Anal sphincter disturbances are uncommon.

SYNDROMES

Clustering of CSM into these 5 clinical syndromes has been described³⁶⁹:

1. transverse lesion syndrome: involvement of corticospinal and spinothalamic tracts and posterior columns, with anterior horn cells segmentally involved. Most frequent syndrome, possibly an “end-stage” of the disease process

2. motor system syndrome: primarily corticospinal tract and anterior horn involvement with minimal or no sensory deficit. This creates a mixture of lower motor neuron findings in the upper extremities and upper motor neuron findings (myelopathy) in the lower extremities which can mimic ALS (see below). Reflexes may be hyperactive below the area of maximal stenosis (including the upper extremities), occasionally beginning several levels below the stenosis

3. central cord syndrome: motor and sensory deficit affecting the UEs more than the LEs. This syndrome is characterized by dysfunction of the watershed areas located centrally within the cord, which may be responsible for prominence of hand symptoms³⁷⁵ (results in “numb-clumsy hands”³⁷⁶). Lhermitte’s sign may be more common in this group

4. Brown-Séquard syndrome: often with asymmetric narrowing of the canal with the side of greater narrowing producing ipsilateral corticospinal tract (upper motor neuron weakness) and posterior column dysfunction with contralateral loss of pain and temperature sensation

5. brachialgia and cord syndrome: radicular UE pain with lower motor neuron weakness, and some associated long tract involvement (motor and/or sensory)

DIFFERENTIAL DIAGNOSIS

See Myelopathy on page 1185 for other possible causes. Some of these (e.g. spinal cord tumor, OPLL) may be demonstrated radiographically. Asymptomatic cervical spondylosis is very common, and ≈ 12% of cases of cervical myelopathy attributed to spondylosis are later found to be due to another disease process including:

1. ALS: see below

2. multiple sclerosis (MS): spinal cord demyelination may mimic CSM. With MS, remissions and exacerbations are common, and patients tend to be younger

3. herniated cervical disc (soft disc): patients tend to be younger than with CSM. Course is more rapid

4. subacute combined system disease: abnormal vitamin B₁₂ level and possibly macrocytic anemia (see page 1187)

5. hereditary spastic paraplegia: family history is key. Diagnosis of exclusion³⁷⁷

6. (spontaneous) intracranial hypotension (see page 305)

Amyotrophic lateral sclerosis (ALS)

AKA (anterior horn) motor neuron disease (also see Amyotrophic lateral sclerosis, page 65). Can mimic the motor system syndrome of CSM (see above), and spinal cord compression may be seen on MRI in > 60% of patients with ALS³⁷⁸.

“Triad” of ALS: atrophic weakness of hands and forearms (early), mild LE spasticity, diffuse hyperreflexia. Inevitably, some cases of demyelinating disease will be misdiagnosed initially as CSM until some features suggestive of ALS occur (in one series of 1500 ALS patients, 4% underwent spine surgery (56% cervical, 42% lumbar, 2% thoracic)³⁷⁸ before ALS was correctly diagnosed).

Features that may help differentiate ALS from CSM:

1. ALS: sensory changes are conspicuously absent

2. ALS: bulbar symptoms (dysarthria, hyperactive jaw-jerk…)³⁷⁹

3. ALS: extensive weakness/muscle atrophy of hands, usually with fasciculations³⁸⁰

4. ALS: lower-motor neuron (LMN) findings in the tongue (visible fasciculations, or positive sharp waves on EMG) or in the LEs (e.g. fasciculations and atrophy) favors the diagnosis of ALS over CSM (however, LMN findings in the LEs may occur in CSM if there is coincidental lumbar radiculopathy)

5. CSM or herniated cervical disc: usually includes neck and shoulder pain, limitation of neck movement, sensory changes, and LMN findings restricted to 1 or 2 spinal cord segments

EVALUATION

Plain x-rays

Plain cervical spine x-rays demonstrates osteophytic spurs, and malalignment if any. See Canal diameter on page 136 for normal dimensions and measurement techniques. Patients with CSM have an average minimal AP canal diameter of 11.8 mm³⁸¹, and values ≤ 10 mm were likely to be associated with myelopathy³⁸². Patients with an AP diameter < 14 mm may be at increased risk³⁸³, and CSM is rare in patients with a diameter > 16 mm, even with significant spurs³⁴⁸.

Pavlov ratio (AKA Torg ratio^{384, 385}): the ratio of the AP diameter of the spinal canal at the mid VB level to the VB at the same location. A ratio < 0.8 is sensitive for transient neuropraxia, but has been shown to have poor positive predictive value for CSM.

MRI

MRI provides information about the spinal canal, and can also show intrinsic cord abnormalities (demyelination, syringomyelia, spinal cord atrophy, edema…). MRI also rules out other diagnostic possibilities (Chiari malformation, spinal cord tumor…). Bony structures and calcified ligaments are poorly imaged. This shortcoming and the difficulties in differentiating osteophytes from herniated discs on MRI are overcome with the addition of plain cervical spine films³⁸⁶ or thin-section CT bone windows.

Findings that correlate with poor outcome (Class III)³⁸⁷:

1. multilevel T2WI hyperintensity within the spinal cord parenchyma

2. single level T2WI hyperintensity with corresponding T1WI hypointensity^A

3. spinal cord atrophy (transverse area < 45 mm²)

A. single level T2WI hyperintensity without T1WI changes are of uncertain prognostic significance

Other MRI findings seen with CSM:

1. reduced transverse area of the spinal cord (TASC) at the level of maximum compression. A “banana” shaped cord on axial images has a high correlation with the presence of CSM³⁸³. There is conflicting evidence whether the degree of canal stenosis predicts outcome³⁸⁷. Sagittal T2WIs tend to exaggerate the magnitude of spinal cord compression by osteophytes and/or discs, and therefore axial images and T1WIs also need to be considered in the evaluation. Narrowing is not specific for CSM: ≈ 26% of asymptomatic individuals > 64 years of age have spinal cord compression on MRI³⁸⁸

2. “snake eyes” (AKA “owl’s eyes”) within the spinal cord on axial T2WI (see Figure 18-5) may be related to cystic necrosis of the cord³⁸⁹ and may correlate with poor outcome (Class III)³⁸⁷

CT/myelogram

Plain CT scans may demonstrate a narrow canal, but do not provide adequate information regarding soft tissues (discs, ligaments, spinal cord and nerve roots). Cervical myelography followed by high-resolution CT scanning provides sagittal and axial information (including spinal cord atrophy), and delineates bony detail better than MRI³⁸⁶. Unlike MRI, CT/myelogram involves ionizing radiation and does not provide information about changes within the spinal cord parenchyma.

Figure 18-5 Snake eyes (two foci of high signal) within a slightly flattened and mildly atrophic spinal cord on axial T2WI MRI

Sensory evoked potentials (SEPs) - pre-op

Normal pre-op SEPs or normalization of SEPs in the early post-op period are associated with better outcome³⁹⁰.

PRACTICE GUIDELINE 18-16 PRE-OP SEPS IN CSM

Pre-op SEPs should be considered if the additional prognostic information would help treatment decisions (Level BClass II)³⁹⁰

EMG

Not routinely useful in CSM. EMG has poor sensitivity in cervical radiculopathy and is not reliable in predicting outcome from surgery for CSM (Class III)³⁸⁷. EMG is most helpful in suspicious cases to eliminate etiologies such as peripheral neuropathy or ALS.

TREATMENT

Indications for surgery

See PRACTICE GUIDELINE 18-17.

PRACTICE GUIDELINE 18-17 SURGICAL VS. NONSURGICAL MANAGEMENT

Mild myelopathy (mJOA score > 12)*: in the short-term (3 years) patients may be offered the option of surgical decompression or nonoperative management (prolonged immobilization in a rigid cervical collar, anti-inflammatory medications, and “low-risk” activities or bed rest (Level CClass II)³⁹¹

More severe myelopathy: should be treated with surgical decompression with benefits maintained at 5 and 15 years post-op (Level D Class III)³⁹¹

Level B Class I392 Degenerative cervical radiculopathy: patients do better with anterior decompression ± fusion (compared to conservative management) for

• rapid relief (within 3-4 months) of arm & neck pain and sensory loss

• relief of longer term (≥ 12 months) symptoms of weakness of wrist extension, elbow extension, shoulder abduction and internal rotation

* patients with mJOA scores > 12 (see page 487) may not always be mildly impaired, they may derive significant improvement from surgery, and deterioration from this point may be ominous

NONOPERATIVE MANAGEMENT

Measures include: prolonged immobilization with rigid cervical bracing in an attempt to reduce motion and hence the cumulative effects of trauma on the spinal cord, modified activity to eliminate “high-risk” activities or bed rest, and anti-inflammatory medications³⁹³.

SURGICAL TREATMENT

Intraoperative electrophysiologic monitoring

PRACTICE GUIDELINE 18-18 INTRA-OPERATIVE ELECTROPHYSIOLOGIC MONITORING DURING SURGERY FOR CSM OR RADICULOPATHY

Use of intra-op EP monitoring during routine surgery for CSM or cervical radiculopathy is not recommended as an indication to alter the surgical plan or administer steroids since this paradigm has not been observed to reduce the incidence of neurologic injury (Level DClass III)³⁹⁴

Choice of approach

The debate between anterior approaches (anterior cervical discectomy or corpectomy) and posterior approaches (decompressive cervical laminectomy) dates back to the time that both became widely practiced²⁴⁴. General sentiment is to treat anterior disease at the disc level (e.g. osteophytic bar, herniated disc…) usually limited to ≤ 3 levels (or occasionally 4) with an anterior approach, and to use a posterior approach as the initial procedure in the situations outlined below. Considerations of spinal curvature may need to enter into the decision process.

PRACTICE GUIDELINE 18-19 CHOICE OF SURGICAL APPROACH FOR CSM

There was not enough evidence to recommend any of the following techniques over the other (in terms of short-term success in treating CSM): ACDF, anterior corpectomy and fusion, laminectomy (with or without fusion) and laminoplasty ((Level DClass III)³⁹⁵

Laminectomy without fusion, however, is associated with a higher incidence of late kyphotic deformity* (Level DClass III)³⁹⁵

* incidence 14-47%, not all cases are symptomatic, not all cases need treatment (see text)

Posterior approach: Options include:

1. laminectomy alone

2. laminectomy/arthrodesis (i.e. laminectomy + lateral mass fusion) (Class III^{A 396})

3. laminoplasty (Class III^{A 397}): methods include unilateral (“open door”) and midline enlargement (“French door”)

4. multilevel foraminotomies: usually not adequate for central canal stenosis

A. this procedure was found to be effective, the class shows the strength of the evidence

Situations where a posterior approach would generally be the initial approach:

1. congenital cervical stenosis where removing osteophytes will still not provide at least ≈ 12 mm of AP canal diameter

2. disease over ≥ 3 levels (although up to 4 may occasionally be dealt with anteriorly)

3. primary posterior pathology (e.g. infolding of ligamentum flavum)

4. some cases of OPLL (anterior approach has higher risk of dural tear)

Disadvantages of the posterior approach:

1. laminectomy without fusion

A. degeneration and osteophytes continue to progress following surgery

B. risk of subsequent subluxation or progressive kyphotic angulation (“swan neck” deformity)

1. quoted incidence: 14-47%^398-400 (risk may be minimized by careful preservation of facet joints)

2. not all cases need to be treated: in one series, 31% (18/58) developed post-op kyphosis, and 16% of these (3/18) required surgical stabilization⁴⁰¹

3. the development of kyphotic deformity does not appear to diminish the clinical outcome⁴⁰⁰ and does not correlate with neurologic deterioration when deterioration occurs⁴⁰².

2. more painful initially post-op and sometimes more prolonged rehabilitation

3. long-term complaints of a heaviness of the head possibly associated with atrophy of the paraspinal muscles

4. ✖ contraindicated with pre-existing swan neck deformity, and not recommended in the presence of reversal of the normal cervical lordosis (i.e. kyphotic curve)⁴⁰³ where the spinal cord won’t tend to move away from the anterior compression or in the presence of ≥ 3.5 mm subluxation or > 20° rotation in the sagittal plane³⁸³ and caution must be exercised in hyper lordosis (see below)

Anterior approach: Also shown to be effective (Class III391).

Instrumentation options: in terms of fusion rates for 2-level anterior operations (i.e. 2 disc spaces) (Class III)³⁹⁵:

* however, the graft extrusion rate is higher for corpectomy than ACDF

Worsening of myelopathy has been reported in 2-5% of patients after anterior decompression^{404, 405} (intraoperative SSEP monitoring may reduce this rate⁴⁰⁵) and C5 radiculopathy may occur (see below).

Anterior cervical plating:

Many systems are available, with more similarities than differences. All include some method of preventing screw back out. Some general pointers:

1. for single level fusion, typical plate length is 22-24 mm

2. screw length: rule of thumb is 12 mm for females, 14 mm for males

3. do not completely tighten a single screw (to avoid kicking up plate) until the diagonally opposite screws are placed and loosely tightened

4. most systems have fixed and variable angle screws. Variable angle screws allow for load sharing with the graft (here is where a derivative of Wolff’s law is often invoked: the weight sharing helps stimulate fusion). Avoid over-angling screws which may prevent the locking mechanism from properly engaging

5. optimal plate placement allows for contact of the plate with the VB at the screw location. This may require

A. contouring of plate to follow the lordosis of the c-spine

B. reduction of anterior osteophytes

Posterior approach: For decompression, some recommend cervical laminectomy extending one or two levels beyond the stenosis above and below^{406, 407}.

Curvature considerations: extending the laminectomy to include C2 and some-times C1 has been recommended for patients with straightening of the cervical curvature³⁴⁶. In cases of hyperlordosis, posterior migration of the spinal cord following an extensive laminectomy may put increased tension on the nerve roots and blood vessels (with possible neurologic worsening), and a limited laminectomy just where the cord is compressed is recommended (see Post-op C5 palsy below).

“Keyhole foraminotomies” or medial facetectomy with undercutting of the facets may be performed at levels involved with radiculopathy.

Position: choices are primarily: prone, lateral oblique, or sitting. The prone position has a major disadvantage of difficulty elevating the head above the heart, resulting in venous engorgement with significant operative bleeding. The sitting position has a number of inherent risks (see Sitting position, page 153) including cord hypoperfusion⁴⁰⁵. The lateral oblique position may introduce some distortion to the anatomy due to asym-metrical positioning.

The rate of post-op spinal deformity is 25-42%. Neurologic worsening has been reported in 2% in some series, higher in others. C5 radiculopathy may occur (see below).

To avoid significant destabilization of the cervical spine:

1. during the dissection, do not remove soft tissue overlying the facet joints (to preserve their blood supply)

2. take the laminectomy only as far lateral as the extent of the spinal canal, carefully preserving the facet joints³⁴⁷ (use keyhole laminotomies where necessary)

3. avoid removing a total of one facet at any given level

OUTCOME

Even excluding cases that are later proven to have demyelinating disease, the outcome from surgery for CSM is often disappointing. Once CSM is clinically apparent, complete remission almost never occurs. The prognosis with surgery is worse with increasing severity of involvement at the time of presentation⁴⁰⁶ and with longer duration of symptoms (48% showed clinical improvement or cure if operated within 1 yr of onset, whereas only 16% responded after 1 yr³⁴⁷). The success of surgery is also lower in patients with other degenerative diseases of the CNS (ALS, MS…).

Progression of myelopathy may be arrested by surgical decompression. This is not always borne out, and some early series^{363, 366} showed similar results with conservative treatment as with laminectomy which yielded improvement in 56%, no change in 25%, and worsening in 19%. Also as discussed earlier (see Natural history, page 487) some cases of CSM develop most of the deficit early and then stabilize.

Some series show good results, with ≈ 64-75% patients having improvement in CSM post-op³⁶⁰. However, other authors remain less enthusiastic. Utilizing a questionnaire in 32 post-op patients operated anteriorly, 66% had relief from radicular pain, while only 33% had improvement in sensory or motor complaints³⁶⁰. In one series, half the patients had improvement in fine motor function of the hands, but the other half worsened postoperatively⁴⁰⁸. Spinal cord atrophy as a result of continued pressure or ischemia may be partly responsible for poor recovery. Bedridden patients with severe myelopathy rarely recover useful function.

Post-op C5 palsy: Criteria: weakness of deltoid and/or biceps with no worsening of myelopathy. Follows ≈ 3-5% of extensive anterior or posterior decompression (including laminoplasty)^{404, 409}. 50% have motor involvement only (deltoid > biceps), 50% also have C5 dermatomal sensory loss and/or C5 dermatomal pain (shoulder). Most occur < 1 week post-op⁴⁰⁹. 92% are unilateral⁴⁰⁹. No pre-op risk factors have been identified⁴¹⁰. Etiology: unproven, may be related to traction on the nerve root from posterior migration of the cord after decompression or to bone graft displacement. Prognosis for spontaneous recovery is generally good; more severe deficits take longer to recover⁴⁰⁹.

Late developments:

Some patients who show early improvement will develop late deterioration (7-12 yrs after reaching a plateau)³⁸³, with no radiographically apparent explanation in up to 20% of these cases⁴¹¹. In others, degeneration at levels adjacent to the operated segments may be demonstrated.

Adjacent segment disease (ASD): degeneration that develops at a motion segment adjacent to a previous fusion. Findings include: disc degeneration, stenosis, facet hypertrophy, scoliosis, listhesis and instability. After ACDF, ASD occurred at a rate of 2.9% per year over 10 years observation⁴¹². Estimate: 25% of patients will develop symptomatic adjacent level changes within 10 years of surgery⁴¹². This rate was higher with single level fusion at C5-6 or C6-7 than it was with multilevel fusion, and natural progression of the disease was felt to be a significant contributor⁴¹² (i.e. it was not all attributable to the fusion). Most cases of ASD observed radiographically are asymptomatic

18.4.1.3. Coincident cervical and lumbar spinal stenosis

Coincident symptomatic lumbar and cervical spinal stenosis is usually managed by first decompressing the cervical region, and later operating on the lumbar region (unless severe neurogenic claudication dominates the picture). It is also possible, in selected cases, to operate on both in a single sitting^{304, 413}.

18.5. Craniocervical junction and upper cervical spine abnormalities

Also see Axis (C2) vertebra lesions, page 1231.

ASSOCIATED CONDITIONS

Abnormalities in this region are seen in a number of conditions including:

1. rheumatoid arthritis: see page 494

2. traumatic & post-traumatic: including fractures of odontoid, occipital condyles…

3. ankylosing spondylitis: (see page 502) may result in fusion of the entire spine which spares the occipitoatlantal and/or atlantoaxial joints which can lead to instability there

4. congenital conditions:

A. Chiari malformations: see page 233

B. Klippel-Feil syndrome: see page 253

C. Down syndrome

D. atlantoaxial dislocation (AAD)

E. occipitalization of the atlas: seen in 40% of congenital AAD⁴¹⁴

F. Morquio syndrome (a mucopolysaccharidosis): atlantoaxial subluxation occurs due to hypoplasia of the odontoid process and joint laxity

5. neoplasms: metastatic (see page 743) or primary

6. infection

7. following surgical procedures of the skull base or cervical spine: e.g. transoral resection of the odontoid

TYPES OF ABNORMALITIES

Abnormalities include:

1. basilar impression/invagination: as with Paget’s disease

2. atlanto-occipital dislocation

3. atlantoaxial dislocation

4. occipitalization of the atlas, or thin or deficient posterior arch of atlas⁴¹⁵

TREATMENT

Fractures of the occipital condyles, atlas or axis are usually adequately treated with external immobilization (also see Occipital condyle fractures, page 954). Because traumatic occipitocervical dislocations are usually fatal, optimal treatment is not well defined. Occipitalization of the atlas may be treated by creating an “artificial atlas” from the base of the occiput and wiring to that⁴¹⁵.

Indications and techniques are outlined in Atlantoaxial fusion (C1-2 arthrodesis) on page 183.

18.6. Rheumatoid arthritis

More than 85% of patients with moderate or severe rheumatoid arthritis (RA) have radiographic evidence of C-spine involvement⁴¹⁶.

The grading system of Ranawat et al.⁴¹⁶ for myelopathy is shown in Table 18-31 is used in RA as well as other etiologies of myelopathy.

Table 18-31 Ranawat classification of myelopathy

Class	Description
I	no neural deficit
II	subjective weakness + hyperreflexia + dysesthesia
III	objective weakness + long tract signs III A = ambulatory III B = quadriparetic & non ambulatory

Common cervical spine involvement in RA:

1. upper cervical spine: involved in 44-88% of RA cases⁴¹⁷ (often found together):

A. anterior atlantoaxial subluxation: the most common manifestation of RA in the cervical spine, found in up to 25% of patients with RA (see below)

B. basilar impression (BI): upward translocation of the odontoid process, found in ≈ 8% of patients with RA (see page 497)

C. pannus of granulation tissue: forms around the odontoid

2. subaxial C-spine (i.e. below C2): subluxation (see page 498)

Less common involvement of the cervical spine in RA:

1. posterior subluxation of the atlantoaxial joint: must have either associated fracture of or near total arthritic erosion of odontoid

2. vertebral artery insufficiency secondary to changes at cranio-cervical junction⁴¹⁸

18.6.1. Atlantoaxial subluxation (AAS) in RA

Inflammatory involvement of the atlantoaxial synovial joints causes erosive changes in the odontoid process (anteriorly at the synovial joint with the C1 arch, and posteriorly at the synovial joint with the transverse ligament) and decalcification and loosening of the insertion of the transverse ligament on the atlas. These changes lead to instability allowing a scissoring effect with anterior subluxation of C1 on C2. AAS occurs in ≈ 25% of patients with RA⁴¹⁸. Mean time between onset of RA symptoms to the diagnosis of AAS in 15 patients: 14 years⁴¹⁹.

CLINICAL

Signs and symptoms of AAS are shown in Table 18-32.

AAS is usually slowly progressive. Mean age at onset of AAS symptoms: 57 years.

Pain is experienced locally (upper cervical and suboccipital regions, often from compression of C2 nerve root) or is referred (to mastoid, occipital, temporal, or frontal regions).

VBI may occur from VA involvement (see page 1158).

Table 18-32 Signs and symptoms of AAS* (15 patients with AAS⁴¹⁹)

Finding	%
pain
local	67%
referred	27%
hyperreflexia	67%
spasticity	27%
paresis	27%
sensory disturbance	20%

* other possible findings not reported in this series: clumsiness, neurogenic bladder, Babinski sign

RADIOGRAPHIC EVALUATION

The magnitude of AAS is usually increased with neck flexion.

LATERAL C-SPINE X-RAY

Anterior atlantodental interval (ADI)

The ADI (see page 957 for details) only gives information about the stability of the C1-2 joint. The normal ADI in adults is < 3-4 mm^{420, 421}. Widening of the ADI suggests possible incompetence of the transverse ligament. However, the ADI does not correlate with the risk of neurologic injury^{422, 423} and is not predictive of progression from asymptomatic AAS to symptomatic AAS.

Posterior atlantodental interval (PADI)

The amount of room available for the spinal cord can vary for any given ADI depending on the AP diameter of the spinal canal and the thickness of any pannus. The PADI (see page 136) and the AP diameter of the subaxial canal measured on a lateral C-spine x-ray correlates with the presence and severity of paralysis⁴²².

The PADI also predicts neurologic recovery following surgery. Patients with paralysis from AAS showed no recovery if the pre-op PADI was < 10 mm⁴²².

PADI ≤ 14 mm has been proposed as an indication for surgical stabilization.

MRI

The optimal test to evaluate the source and magnitude of upper cord or medulla compression. Demonstrates location of odontoid process, extent of pannus, and effects of subluxation (may need to be performed with head flexed to evaluate this).

TREATMENT

Requires knowledge of the following information:

• natural history: AAS in most patients progresses, with a small percentage either stabilizing or fusing spontaneously. In one series⁴²⁴ with 4.5 years mean followup, 45% of patients with 3.5-5 mm subluxation progressed to 5-8 mm, and 10% of these progressed to > 8 mm

• once myelopathy occurs, it may be irreversible

• the worse the myelopathy, the higher the risk for sudden death

• the chances of finding myelopathy are significantly increased once the subluxation reaches ≥ 9 mm⁴²⁵

• associated cranial settling further decreases the tolerance for AAS

• life expectancy of patients with RA is 10 years less than the general population⁴²⁴

• the morbidity and mortality of surgical treatment (see below)

• pannus may regress some after medical treatment

When to treat?

• symptomatic patients with AAS: almost all require surgical treatment (C1-2 fusion in most cases). For management, see below

A. some surgeons do not operate if the maximal dens-C1 distance is < 6 mm

• asymptomatic patients: controversial

A. some authors feel surgical fusion is not necessary in asymptomatic patient if the dens-C1 distance is below a certain cutoff. Recommendations for this cutoff have ranged from 6 to 10 mm⁴²⁶, with 8 mmcommonly cited (an un-validated delineation)

B. these patients are often placed in a rigid cervical collar, e.g. while outside the home, even though it is generally acknowledged that a collar probably does not provide significant support or protection

C. NB: some cases of sudden death in previously asymptomatic RA patients may be due to AAS and may then be erroneously attributed to cardiac arrhythmias, etc.⁴²⁷

Surgical management

It is necessary to either reduce the subluxation or to decompress the upper cord before doing a C1-C2 or occipital-C1-C2 fusion.

Menezes assesses all subluxed patients for reducibility using MRI compatible Halo cervical traction as follows: start with 5 lbs, and gradually increase over a period of a week. Most cases reduce within 2-3 days. If not reduced after 7 days then it is probably not reducible. Only ≈ 20% of cases are not reducible (most of these have odontoid > 15 mm above foramen magnum).

Most require stabilization via posterior wiring and fusion, either of C1 to C2, or of occiput to C2. The latter is used when fusion is combined with decompression (posterior laminectomy of C1 with posterior enlargement of the foramen magnum). See Atlantoaxial fusion (C1-2 arthrodesis) on page 183.

Posterior fusion alone does not provide adequate relief if the subluxation is irreducible, or if pannus causes significant compression (however, there may be some reduction of pannus after fusion). In these cases, transoral odontoidectomy may be indicated. Performing the posterior stabilization and decompression first allows some patients to avoid a second operation, and permits the remainder to undergo the anterior approach without becoming destabilized. Still, some surgeons do the odontoidectomy first⁴²⁶ (requires the patient to remain in traction until the fusion).

Reminder: the patient must be able to open the mouth greater than ≈ 25 mm in order to perform transoral odontoidectomy without splitting the mandible.

POSTERIOR FUSION

See Atlantoaxial fusion (C1-2 arthrodesis) on page 183 for technique. In RA, erosion and osteoporosis weakens the C1 arch, and extra care is needed to avoid fracturing it.

Morbidity and mortality

Because of the frequency of simultaneous involvement of other systems in RA including pulmonary, cardiac, and endocrine, operative mortality ranges from 5-15%⁴²⁶.

The non-fusion rate for C1-2 wiring and fusion has been reported as high as 50%⁴²⁸, typical rates are lower (with 18% of patients in one series developing a fibrous union⁴²⁶). The most common site of failure of osseous fusion is the interface between the bone graft and the posterior arch of C1429.

Post-operative care

The patient is usually mobilized almost immediately post-op in halo vest traction (some use an optional period of maintained traction before mobilization). Impaired healing in RA dictates that the Halo be worn until fusion is well established, as seen on x-ray (usually 8-12 weeks). Sonntag evaluates the patient with flexion-extension lateral C-spine x-rays by disconnecting the halo ring from the vest.

18.6.2. Basilar impression in rheumatoid arthritis

AKA atlantoaxial impaction. Erosive changes in the lateral masses of C1 → telescoping of the atlas onto the body of C2 causing ventral migration of C1 with resultant ↓ in AP diameter of the spinal canal. There is concomitant upward displacement of the dens. The posterior arch of C1 often protrudes superiorly through the foramen magnum. All of these factors lead to compression of the pons and medulla. Rheumatoid granulation tissue behind the odontoid also contributes. Vertebral artery and/or anterior spinal artery compression may also cause neurologic dysfunction.

The degree of erosion of C1 correlates with the extent of odontoid invagination.

CLINICAL

See Table 18-33 for signs & symptoms.

Pain may occur as a result of compression of the C1 and/or C2 nerve roots. Compression of the medulla can cause cranial nerve dysfunction.

Motor exam usually difficult because of severe polyarticular degeneration and associated pain. Sensory findings (all non-localizing): diminished vibratory, position, and light touch.

Table 18-33 Symptoms and signs of BI (45 patients with RA⁴¹⁷)

Finding	%
headache	100%
progressive difficulty ambulating	80%
hyperreflexia + Babinski	80%
limb paresthesias	71%
neurogenic bladder	31%
cranial nerve dysfunction	22%
trigeminal nerve anesthesia	20%
glossopharyngeal
vagus
hypoglossal
miscellaneous findings
internuclear ophthalmoplegia
vertigo
diplopia
downbeat nystagmus
sleep apnea
spastic quadriparesis

RADIOGRAPHIC EVALUATION

See Basilar invagination & basilar impression (BI) on page 138 for radiographic criteria of BI. Erosion of the tip of the odontoid, commonly seen in RA, obviates use of any measurement that is based on the location of the tip of the odontoid⁴³⁰. For this reason, other measures have been developed, including the Clark station⁴²⁹, Redlund-Johnell criteria⁴³¹, and Ranawat criteria⁴¹⁶. Since even these methods will miss up to 6% of cases of BI in RA⁴³⁰, it is recommended that suspicious cases be investigated further (e.g. with CT and/or MRI).

MRI: optimal for demonstrating brain stem impingement, poor for showing bone. Cervicomedullary angle: the angle between a line drawn through the long axis of the medulla on a sagittal MRI and a line drawn through the cervical spinal cord. The normal CMA is 135-170°. A CMA < 135° correlates with signs of cervicomedullary compression, myelopathy or C2 radiculopathy⁴³².

CT: primarily done to assess bony anatomy (erosion, fractures…).

CTA should be performed when surgery is contemplated, to show VA anatomy.

Myelography (water soluble) with CT: also good for delineating bony pathology.

TREATMENT

See also Craniocervical junction and upper cervical spine abnormalities on page 494.

CERVICAL TRACTION

May attempt with Gardner-Wells tongs. Begin with ≈ 7 lbs, and slowly increase up to 15 lbs. Some may require several weeks of traction to reduce.

SURGERY

Reducible cases: posterior occipitocervical fusion ± C1 decompressive laminectomy. Irreducible cases: requires transoral resection of odontoid. May perform before posterior fusion (but then must be kept in traction while waiting for posterior fusion).

18.6.3. Subaxial subluxation in rheumatoid arthritis

The direct effects of RA on the subaxial spine involves the facet joints posteriorly. Degenerative disc disease, which is generally a late manifestation in RA, is not the result of synovitis⁴³³. Involvement is most common at C2-3 and C3-4.

18.7. Atlantoaxial subluxation (AAS) in Down syndrome

No all cases of AAS are unstable (which, by definition, needs treatment).

Incidence of AAS in Down syndrome (DS) is 20% 434, but only 1-2% of DS patients have symptomatic AAS⁴³⁵. AAS in DS appears to be due to laxity of the transverse atlantal ligament (TAL). This laxity may decrease with age as the TAL stiffens.

Management

Controversial. There have been position statements⁴³⁶ and rebuttals^{435, 437}.

Recommendations (modified⁴³⁸):

1. children who have been screened and do not have AAS: no further screening after age 10 years (since AAS does not develop later; the cutoff age is controversial)

2. os odontoideum: surgical fusion

3. symptomatic AAS

A. symptoms may include: gait difficulties, neck pain, limited neck motion, torticollis, clumsiness, sensory deficits, and other symptoms of myelopathy

B. for ADI^A > 4.5 mm or PADIA < 14 mm or spinal cord damage on cervical MRI: surgical fusion

4. asymptomatic AAS seen on lateral C-spine x-ray:

A. for ADI ≤ 4.5 mm and PADI ≥ 14 mm: no need for further testing

B. for ADI > 4.5 mm or PADI < 14 mm: cervical MRI

1. if the MRI shows spinal cord damage: surgical fusion

2. if MRI shows no spinal cord damage: surgical fusion is optional. If fusion is not done, prohibit high-risk activities and restudy in 1 year

A. ADI = atlantodental interval (see page 136), PADI = posterior atlantodental interval (see page 136)

18.8. Paget’s disease

PATHOPHYSIOLOGY

Paget’s disease (PD) (AKA osteitis deformans) is a disorder of osteoclasts (possibly virally induced) causing increased rate of bone resorption with reactive osteoblastic overproduction of new, weaker, woven bone, producing characteristic “mosaic pattern”.

Initially there is a “hot” phase with elevated osteoclastic activity and increased intraosseous vascularity. Osteoblasts lay down a soft, nonlamellar bone. Later a “cool” phase occurs with disappearance of the vascular stroma and osteoblastic activity leaving sclerotic, radiodense, brittle bone⁴³⁹ (“ivory bone”).

Malignant degeneration

A misnomer, since the malignant changes actually occur in the reactive osteoblastic cells. About 1% (reported range: 1-14%) degenerate into sarcoma (osteogenic sarcoma, fibrous sarcoma, or chondrosarcoma)^{440 (p 2642)}, with the possibility of systemic (e.g. pulmonary) metastases. Malignant degeneration is much less common in the spine than in the skull or femur.

EPIDEMIOLOGY

Prevalence: ≈ 3% of population > 55 years old in the U.S. and Europe⁴⁴¹. Slight male predominance. Family history of Paget’s disease is found in 15-30% of cases (accuracy is poor since most are asymptomatic).

Common sites of involvement

Affinity for axial skeleton, long bones and skull. In approximate descending order of frequency: pelvis, thoracic and lumbar spine, skull, femur, tibia, fibula, and clavicles.

PRESENTATION

Only ≈ 30% of pagetic sites are symptomatic⁴⁴², the rest are discovered incidentally. The overproduction of weak bone may produce bone pain (the most common symptom), predilection for fractures and compressive syndromes (cranial nerve (see page 1204), spinal nerve root…). Painless bowing of a long bone may be the first manifestation. A number of patients present due to pain from joint dysfunction related to PD.

NEUROSURGICAL INVOLVEMENT

PD may present to the neurosurgeon as a result of:

• back pain: usually not as a direct result of vertebral bone involvement (see below)

• spinal cord and/or nerve root symptoms

compression of the spinal cord or cauda equina (relatively rare)

spinal nerve-root compression

vascular steal due to reactive vasodilatation adjacent to involved areas

• with skull involvement:

compression of cranial nerves as they exit through bony foramina (8th nerve is most common, producing deafness or ataxia): see page 1204

skull base involvement → basilar invagination

• to ascertain diagnosis in unclear bone lesions of the spine or skull

EVALUATION

1. lab work (markers may be normal in monostotic involvement):

A. serum alkaline phosphatase: usually elevated (this enzyme is involved in bone synthesis and so may not be elevated in purely lytic Paget’s disease^{100 (p 1416)}); mean 380 ± 318 IU/L (normal range: 9-44)⁴⁴³. Bone-specific alkaline phosphatase may be more sensitive and may be useful in monostotic involvement⁴⁴¹

B. calcium: usually normal (if elevated, one should R/O hyperparathyroidism)

C. urinary hydroxyproline: found almost exclusively in cartilage. Due to the high turnover of bone, urinary hydroxyproline is often increased in PD with a mean of 280 ± 262 mg/24 hrs (normal range 18-38)⁴⁴³

2. bone scan: lights up in areas of involvement in most, but not all⁴⁴³ cases

3. plain x-rays:

A. localized enlargement of bone: a finding unique to PD (not seen in other osteoclastic diseases, such as prostatic bone mets)

B. cortical thickening

C. sclerotic changes

D. osteolytic areas (in skull → osteoporosis circumscripta; in long bones → “V” shaped lesions)

E. spinal Paget’s disease often involves several contiguous levels. Pedicles and lamina are thickened, vertebral bodies are usually dense and compressed with increased width. Intervening discs are replaced by bone

4. CT: hypertrophic changes at the facet joints with coarse trabeculations

18.8.1. Paget’s disease of the spine

PRESENTATION

The overwhelming majority of pagetic lesions are asymptomatic^{100 (p 1413)} with lesions detected on radiographs or bone scan obtained for other reasons or as part of a work-up for an elevated alkaline phosphatase. Although the most common complaint in patients with Paget’s disease is of back pain, this is attributable to pagetic involvement alone in only ≈ 12%⁴⁴³, in the remainder it is secondary to other factors, some of which are described below.

Symptoms that may be related to the Paget’s disease itself:

1. symptoms from the following are slowly progressive (usually present for > 12 months; rarely < 6 mos)

A. neural compression

1. causes of compression

a. due to expansion of woven bone

b. due to osteoid tissue

c. pagetic extension into ligamentum flavum and epidural fat⁴⁴⁴

2. sites of compression

a. spinal cord (see below)

b. nerve root in neural foramen

B. osteoarthritis of facet joints (Paget’s disease may precipitate osteoarthritis⁴⁴³)

2. symptoms from the following tend to progress more rapidly

A. malignant (sarcomatous) change of involved bone (rare, see above)

B. pathologic fracture (pain usually sudden in onset)

C. neurovascular (compromise of vascular supply to nerves or spinal cord) by

1. compression of blood vessels (arterial or venous)

2. pagetic vascular steal (see below)

Spinal cord symptoms

Myelopathy or cauda equina syndrome may be due to spinal cord compression or from vascular effects (occlusion, or “steal” due to reactive vasodilatation of nearby blood vessels^{100 (p 1415)}). Only ≈ 100 cases had been described as of 1981445. Characteristically, 3-5 adjacent vertebrae are involved^{446 (p 2307)}, whereas monostotic involvement is usually asymptomatic⁴⁴⁷. In case reports in the literature, progressive quadri- or paraparesis was the most common presentation⁴⁴⁸. Sensory changes are usually the first manifestation, progressing to weakness and sphincter disturbance. Pain was the only symptom in a neurologically intact patient in only 5.5%.

A rapid course (averaging 6 wks) with a sudden increase in pain is more suggestive of malignant degeneration.

TREATMENT

MEDICAL TREATMENT FOR PAGET’S DISEASE

There is no cure for Paget’s disease. Medical treatment is indicated for cases that are not rapidly progressive where the diagnosis is certain, for patients who are poor surgical candidates, and pre-op if excessive bleeding cannot be tolerated. Medical therapy reverses some neurologic deficit in 50% of cases⁴⁴⁹, but generally requires prolonged treatment (≈ 6-8 months) before improvement occurs, and may need to be continued indefinitely due to propensity for relapses. Medications used include the following.

Calcitonin derivatives

Parenteral salmon calcitonin (Calcimar®)⁴⁴⁹: reduces osteoclastic activity directly, osteoblastic hyperactivity subsides secondarily. Relapse may occur even while on calcitonin. Side effects include nausea, facial flushing, and the development of antibodies to salmon calcitonin (these patients may benefit from a more expensive synthetic human preparation (Cibacalcin®) starting at 0.5 mg SQ q d⁴⁵⁰).

Rx 50-100 IU (medical research council units) SQ q d x 1 month, then 3 injections per week for several months⁴⁴¹. If used pre-op to help decrease bony vascularity, ≈ 6 months of treatment is ideal. Doses as low as ≈ 50 IU units 3 x per week may be used indefinitely post-op or as a sole treatment (alkaline phosphatase and urinary hydroxyproline decline by 30-50% in > half of patients in 3-6 months, but they rarely normalize).

Bisphosphonates

These drugs are pyrophosphate analogues that bind to hydroxyapatite crystals and inhibit reabsorption. They also alter osteoclastic metabolism, inhibit their activity, and reduce their numbers. They are retained in bone until it is resorbed. Oral absorption of all is poor (especially in the presence of food). Bone formed during treatment is lamellar rather than woven.

Etidronate (Didronel®) (AKA EHDP): reduces normal bone mineralization (especially at doses ≥ 20 mg/kg/d) producing mineralization defects (osteomalacia) which may increase the risk of fracture but which tend to heal between courses⁴⁵¹. Contraindicated in patients with renal failure, osteomalacia, or severe lytic lesions of a LE. Rx 5-10 mg/kg PO daily (average dose: 400 mg/d, or 200-300 mg/d in frail elderly patients) for 6 months, may be repeated after a 3-6 month hiatus if biochemical markers indicate relapse.

Tiludronate (Skelid®): unlike etidronate, does not appear to interfere with bone mineralization at recommended doses. Side effects: abdominal pain, diarrhea, N/V. Rx 400 mg PO qd with 6-8 ounces of plain water > 2 hrs before or after eating x 3 months. Available: 200 mg tablets.

Pamidronate (Aredia®): much more potent than etidronate. May cause a transient acute flu-like syndrome. Oral dosing is hindered by GI intolerance, and IV forms may be required. Mineralization defects do not occur in doses < 180 mg/course. Rx 90 mg/d IV x 3 days, or as weekly or monthly infusions.

Alendronate (Fosamax®): does not produce mineralization defects (see page 994).

Clodronate (Ostac®, Bonefos®): Rx 400-1600 mg/d PO x 3-6 months. 300 mg/d IV x 5 days.

Risedronate (Actonel®): does not interfere with bone mineralization in recommended doses⁴⁵². Rx: 30 mg PO q d with 6–8 oz. of water at least 30 minutes before the first meal of the day.

Under development: ibandronate, neridronate, and others.

Plicamycin

Formerly mithramicin. A cytotoxic antibiotic that inhibits RNA synthesis with preferential toxicity for osteoclasts. Reserved for severe and extensive involvement due to dose dependent renal and hepatic impairment and possible thrombocytopenia. Not approved for treating Paget’s disease in any country.

Rx 15-25 μg/kg given IV over 8-10 hours qod x 10 infusions.

SURGICAL TREATMENT

In general, conservative treatment of fractures in PD have a high rate of delayed union.

Surgical indications for spinal Paget’s disease

1. rapid progression: indicating possible malignant change or spinal instability

2. spinal instability: severe kyphosis or compromise of canal by bone fragments from pathologic fracture. Although the collapse is usually gradual, sudden compression may occur

3. uncertain diagnosis: especially to R/O metastatic disease (osteoblastic lesions)

4. failure to improve with medications

Surgical considerations:

1. profuse bleeding is common: if significant bleeding would present an unusual problem, treat for as long as feasible pre-op with a bisphosphonate or calcitonin (see above)

A. use bone wax to help control bleeding

B. hemostasis may be difficult

2. to treat resultant spinal stenosis: decompressive laminectomy is the standard procedure in the thoracic region⁴⁴⁸. However, if most of the pathology is anterior, consideration should be given to anterior approach

3. bone is often thickened, and may be fused with obliteration of interspace landmarks. A high-speed drill is usually helpful

4. post-op medical treatment may be necessary to prevent recurrences⁴⁴⁹

5. osteogenic sarcoma

A. surgery and chemotherapy are used, cure is less likely than in primary osteosarcoma of non-pagetic origin

B. biopsy proven of the scalp requires en-bloc excision of scalp and tumor

Surgical outcome⁴⁴⁸:

In 65 patients treated with decompressive laminectomy, 55 (85%) had definite but variable degrees of improvement. Patients who had only minimal improvement were often ones with malignant changes. One patient was worse after surgery, and the operative mortality was 7 patients (10%). Survival with malignant degeneration is < 5.5 mos after admission.

18.9. Ankylosing spondylitis

Key concepts:

• a seronegative spondylarthropathy (enthesopathy)

• starts in the SI joints and progresses rostrally

• clinical: morning back stiffness, kyphoscoliosis limits chest expansion

• x-ray findings: “bamboo spine”, Andersson lesions, progressive thoracic kyphosis

• risk of SCI due to fracture is increased, and may follow minimal trauma

AKA Marie-Strümpell disease. Ankylosing spondylitis (AS): one of the so-called seronegative arthropathies (ANA and serum rheumatoid factor are negative⁴⁵³, unlike rheumatoid arthritis). The spine is the primary skeletal site involved, usually starting in the sacroiliac joints and lumbar spine and progressing rostrally. Enthesopathy: nongranulomatous inflammatory changes at the entheses (attachment points of ligaments, tendons or capsules on bones; the locus of involvement in AS) stimulates replacement of ligaments by bone ultimately resulting in osteoporotic VBs, calcified intervertebral discs (sparing the nucleus pulposus), and ossified ligaments, producing square appearing VBs with bridging syndesmophytes, the so-called “bamboo spine” or “poker spine”.

Differential diagnosis:

1. early on, AS may resemble rheumatoid arthritis. However, in AS nodules do not form in joints, and rheumatoid factor is absent in the serum

2. metastatic prostate Ca in elderly male patients with sacroiliac pain and blastic changes compatible with sacroiliitis

3. Forestier’s disease (see page 506) and DISH (see page 506): these overlapping conditions produce exuberant bony overgrowth anterior and lateral to the disc without degeneration and ossification of the disc as in AS. Both spare the facets and SI joints, do not produce flexion deformity, and tend to occur in men > 50 yrs old (older than typical AS)⁴⁵³

4. psoriasis and Reiter’s syndrome: spondylitis tends to be milder and less uniform and SI joint involvement is asymmetrical

EPIDEMIOLOGY

Incidence is ≈ 1-3 cases per 100,000. Symptoms tend to be more pronounced in males, which has resulted in an underreporting of the condition in females and an exaggeration of the estimation of the male predominance (incidence is probably ≈ equal)⁴⁵⁴. Peak incidence: 17-35 yrs age. Although AS is not hereditary, first degree relatives are at increased risk.

DIAGNOSTIC CRITERIA

Modified New York criteria (see Table 18-34) may not be helpful for diagnosis early on. SI joint involvement is the sine qua non for definite diagnosis.

CLINICAL

Typical initial presentation is with nonradiating low back pain, morning back stiffness, hip pain and swelling (due to large joint arthritis), exacerbated by inactivity and improved with exercise⁴⁵⁵. Patrick’s test (see page 444) usually positive. Compressing the pelvis with the patient in the lateral decubitus position produces pain. Schober test (measure distraction between skin marks on the back with forward flexion to detect reduced mobility of the spine due to fusion⁴⁶) is not specific for inflammatory spondylopathies⁴⁵⁶ but may be helpful for monitoring ongoing physical therapy.

Table 18-34 Modified New York Criteria for AS

Diagnosis (see criteria below)

Definite AS: radiologic criterion + ≥ 1 clinical criteria

Probable AS: radiological criterion without clinical criteria, or 3 clinical criteria without radiological criterion

Clinical criteria

low back pain > 3 months, improved by exercise, not relieved by rest

limitation of lumbar spine motion in both sagittal and frontal planes

limitation of chest expansion relative to normal values for age and sex

Radiological criterion

sacroiliitis

Neurosurgical involvement usually results from the following:

1. cauda equina syndrome (CES): etiology is frequently unclear, but is usually not due to stenosis or compressive lesion. In the absence of compression, surgical intervention is not indicated

2. rotatory subluxation: at occipito-atlantal and atlanto-axial joints. May occur as these are typically the last mobile segments of the spine. Incidence is much less than with rheumatoid arthritis. Lesions that might be stable in otherwise normal spines are often not stable in AS

3. myelopathy secondary to bow-stringing of the cord: laminectomy may aggravate

4. acute spinal cord injury (SCI): risk of SCI or CES due to fracture is increased in AS, and may occur following minimal trauma. Injuries are more common in the lower cervical spine. The rigid spine of AS when fractured acts as a long lever and is extremely unstable⁴⁵⁷. Delayed deterioration may be due to spinal epidural hematoma⁴⁵⁸

5. vertebral stress fractures: most common in the lower thoracic and upper lumbar spines, usually through the ossified intervertebral disc (“chalk-stick” fracture)

6. painful pseudarthroses

7. spinal deformity

8. spinal stenosis: rare

9. basilar impression

RADIOGRAPHIC EVALUATION

Plain x-rays: vital for diagnosis and follow-up. Sacroiliac (SI) joint involvement (on AP pelvic x-rays or on oblique views through the plane of the SI joints) is one of the earliest findings, and the often symmetric osteoporosis followed by sclerosis is characteristic. “Bamboo spine” (see above) is also classic. X-ray of the entire spine is recommended since multiple, non-contiguous (and often unsuspected) fractures are not unusual.

MRI: can rule out spinal epidural hematoma and the occasional herniated disc. Andersson lesions: pathologic changes at ligament insertion sites (MRI signal abnormalities at front and back of the endplates) are characteristic. Erosive changes due to pseudarthrosis at the disc space can mimic discitis (hi signal on T1WI & T2WI with enhancement).

Bone scan: ratio of uptake of SI joint to sacrum > 1.3:1 is suggestive of AS.

NATURAL HISTORY

Progression is slow, and patients usually remain functionally active. Thoracic kyphosis with compensatory increase in cervical and lumbar lordosis is common. The shift in center of gravity together with spine stiffness and fragility predisposes to frequent falls and further spine injuries.

TREATMENT

Management of the disease itself may involve use of immune modulators.

Surgical considerations

Aspects of AS than may increase anesthetic risk:

1. difficulty intubating due to fragile, angulated & immobile spine

2. mitral valve disease

3. myocardial conduction abnormalities

4. decreased pulmonary compliance and reduced lung volume in patients with advanced thoracic kyphoscoliosis

Spinal cord injury: Following trauma, the routine of initially securing the head to a backboard with the neck in neutral position may be deleterious in the patient with kyphotic deformity related to AS⁴⁵⁷. AS may be suspected when the patient spontaneously holds their head in significant flexion⁴⁵⁹, and in these cases the neck should be immobilized in that position⁴⁶⁰. When traction is used, the axis of traction (not the pin sites) often needs to be ventral to neutral and the use of minimal weight is recommended⁴⁵⁷.

Subsequent treatment for SCI in AS is controversial: halo immobilization alone has been advocated by many, citing similar outcome and fewer complications. Others advocate early internal fixation because of cases of nonunion and progression of deficit while in the halo brace⁴⁵⁷ and risk of skin breakdown under the vest due to severe kyphosis. Neurologic deficit related to cord or root compression is an indication for decompressive laminectomy and fusion⁴⁵⁷.

Positioning for surgery may be difficult due to the immobile kyphotic spine. Anterior approaches are difficult due to extensive bridging osteophytes, and the screws for anterior plates may not hold well in the osteoporotic VBs. Posterior cervical lateral mass plating (see page 179) is advocated for most.

Kyphotic deformity: Severe flexion deformity may be treated by spinal osteotomy⁴⁵³.

18.10. Ossification of the posterior longitudinal ligament (OPLL)

Key concepts:

• fibrosis followed by calcification and then ossification of the posterior longitudinal ligament. The process may involve the dura

• more common in Asian population

• most patients have only mild subjective complaints

• 50% of patients have impaired glucose tolerance, respiratory compromise may result from ossification of the costotransverse and costovertebral ligaments

• surgery is best for moderate neuro involvement (Nurick grade 3 & 4)

The age of patients with OPLL ranges from 32-81 years (mean = 53), with a slight male predominance. The prevalence increases with age. Duration of symptoms averages ≈ 13 months. It is more prevalent in the Japanese population (2-3.5%)^{461, 462}.

PATHOPHYSIOLOGY

The pathologic basis of OPLL is unknown, but there is an increased incidence of ankylosing hyperostosis which suggests a hereditary basis.

OPLL begins with hypervascular fibrosis in the PLL which is followed by focal areas of calcification, proliferation of periosteal cartilaginous cells and finally ossification⁴⁶³.

The process frequently extends into the dura. Eventually active bone marrow production may occur. The process progresses at varying rates among patients, with an average annual growth rate of 0.67 mm in the AP direction and 4.1 mm longitudinally⁴⁶⁴.

When hypertrophied or ossified, the posterior longitudinal ligament may cause myelopathy (due to direct spinal cord compression, or ischemia) and/or radiculopathy (by nerve root compression or stretching).

Changes within the spinal cord involve the postero-lateral gray matter more than white matter, suggesting an ischemic basis for the neurologic involvement.

DISTRIBUTION

Average involvement: 2.7-4 levels. Frequency of involvement:

1. cervical: 70-75% of cases of OPLL. Typically begins at C3-4 and proceeds distally, often involving C4-5 and C5-6 but usually sparing C6-7

2. thoracic: 15-20% (usually upper, ≈ T4-6)

3. lumbar: 10-15% (also usually upper, ≈ L1-3)

PATHOLOGIC CLASSIFICATION⁴⁶⁵

1. segmental: confined to space behind vertebral bodies, does not cross disc spaces

2. continuous: extends from VB to VB, spanning disc space(s)

3. mixed: combines elements of both of the above with skip areas

4. other variants: includes a rare type of OPLL that is contiguous with the endplates and is confined to the disc space (involves focal hypertrophy of the PLL with punctate calcification)

CLINICAL

Most patients are asymptomatic, or have only mild subjective complaints. This is probably explained by the protective effect of the fusion resulting from OPLL and the very gradual compression.

Natural history: 17% of patients without myelopathy developed myelopathy in one study⁴⁶⁶ over 1.6 years mean follow-up. Statistically, the myelopathy-free rate in patients without initially presenting with myelopathy was 71% after 30 years⁴⁶⁶.

EVALUATION

Plain x-rays

Often fail to demonstrate OPLL.

MRI

OPLL appears as a hypointense area and is difficult to appreciate until it reaches ≈ 5 mm thickness. On T1WI it blends in with the hypointensity of the ventral subarachnoid space; on T2WI it remains hypointense while the CSF becomes bright. Sagittal images may be very helpful in providing an overview of the extent of involvement, and T2WI may demonstrate intrinsic spinal cord abnormalities which may be associated with a worse outcome.

Myelography/CT

Myelography with post-myelographic CT (especially with 3D reconstructions) is probably best at demonstrating and accurately diagnosing OPLL.

TREATMENT

Treatment decisions

Based on clinical grade⁴⁶⁵ as follows:

• Class I: radiographic evidence without clinical signs or symptoms. Most patients with OPLL are asymptomatic⁴⁶². Conservative management unless severe

• Class II: patients with myelopathy or radiculopathy. Minimal or stable deficit may be followed expectantly. Significant deficit or evidence of progression warrants surgical intervention

• Class IIIA: moderate to severe myelopathy. Usually requires surgical intervention

• Class IIIB: severe to complete quadriplegia. Surgery is considered for incomplete quadriplegics showing progressive slow worsening. Rapid deterioration or complete quadriplegia, advanced age or poor medical condition are all associated with worse outcome

In moderate grade patients (Nurick grades 3 & 4363, see Table 18-35), surgery provided a statistically significant reduction in deterioration. There was no difference between surgery and conservative treatment in mild grade (Nurick 1 or 2), and surgery was ineffective in severe grade (Nurick 5)⁴⁶⁶.

Pre-op assessment

Appropriate cardiorespiratory assessment should be made knowing that:

• respiratory compromise may result from ossification of the costotransverse and costovertebral ligaments

• 50% of patients have impaired glucose tolerance with the attendant risks associated with diabetes

Table 18-35 Nurick grade of disability from cervical spondylosis³⁶³

Grade	Description
0	signs or symptoms of root involvement without myelopathy
1	myelopathy, but no difficulty in walking
2	slight difficulty in walking, able to work
3	difficulty in walking but not needing assistance, unable to work full-time
4	able to walk only with assistance or walker
5	chairbound or bedridden

Technical considerations for surgery

Severe OPLL increases the risk of spinal cord injury during neck positioning for intubation, and strong consideration should be given to awake nasotracheal intubation.

An anterior approach is generally favored, although laminectomy may be acceptable. SSEP monitoring has been recommended by some⁴⁶³. Distraction should be avoided until the spinal cord has been decompressed from the OPLL.

Some authors advocate complete removal of bone from the dura, while others feel it is permissible to leave a thin rim of bone adherent to the dura. Care must be taken in removing bone because it tends to blend imperceptibly with dura and the next thing one may see is bare spinal cord.

Depending on the distance of vertical involvement, vertebral corpectomy with strut grafting may be required. Internal plate fixation is often used as an adjunct. Post-operative immobilization for at least 3 months is employed with rigid collars for single level ACDF or 1-2 level corpectomies, or halo-vest traction for corpectomies > 2 levels.

Results with surgery

The incidence of pseudarthrosis after vertebral corpectomy and strut graft ranges from 5-10% and increases with the number of levels fused.

In one series there was a 10% incidence of transient worsening of neurologic function following anterior surgery⁴⁶⁴ which may have been related to distraction.

The risk of dural tear with CSF leak following an anterior approach depends on the aggressiveness with which bone is removed from the dura, and ranges ≈ 16-25%.

Other risks of anterior approaches (see ACDF complications, page 465) also pertain.

18.11. Ossification of the anterior longitudinal ligament (OALL)

OALL of the cervical spine and/or hypertrophic anterior cervical osteophytes may produce dramatic radiographic findings and minimal clinical symptoms. Distinct from Forestier’s disease (see below). Cervical involvement may produce dysphagia⁴⁶⁷.

18.12. Diffuse idiopathic skeletal hyperostosis

Key concepts:

• usually asymptomatic, but may present with globus

• W/U: speech therapy consult for dysphagia evaluation (usually includes modified barium swallow), CT of cervical spine, ± digital video esophagoscopy

AKA “DISH”, AKA spondylitis ossificans ligamentosa, AKA ankylosing hyperostosis, among others. A condition characterized by flowing osteophytic formation of the spine in the absence of degenerative, traumatic, or post-infectious changes. Affects Caucasians and males more commonly, and usually seen in patients in their mid 60s.

97% of cases occur in the thoracic spine, also in the lumbar spine in 90%, cervical spine in 78%, and all three segments in 70%. Sacroiliac joints are spared (unlike anky-losing spondylitis (AS), see page 502). As with AS, unfused levels may be very unstable.

Usually does not produce clinical symptoms. Patients may have early morning stiffness and mild limitations of activities. Cervical involvement may present with dysphagia or globus (globus pharyngis: a sensation of a lump in the throat) due to compression of the esophagus between the osteophytes and the rigid laryngeal structures⁴⁶⁸ (part of Forestier’s disease⁴⁶⁹).

Plain x-rays and CT scan demonstrates the pathology. In cases of dysphagia, evaluation should include speech therapy consult for dysphagia evaluation, modified barium swallow to help localize the site of obstruction, and DVE (digital video esophagoscopy) to rule-out intrinsic esophageal disease.

Treatment: Cases that do not respond satisfactorily to dietary modifications in patients who are losing weight or are having recurrent episodes of choking or pneumonia should be considered for surgery. An anterior cervical approach, and utilization of a high-speed drill with careful protection of soft-tissue structures (esophagus, carotid sheath) without need for discectomy nor spine stabilization has been recommended⁴⁶⁸. Patients need to be made aware that post-op they are likely to be worse initially (from manipulation of esophagus and possibly disruption of some of the autonomic innervation of the esophagus) and will probably need a gastrostomy feeding tube. By 1 year post-op there may be some improvement.

18.13. Scheuermann’s kyphosis

AKA Scheuermann juvenile kyphosis or kyphoscoliosis AKA Scheuermann disease AKA juvenile osteochondrosis of the spine.

Definition: anterior wedging of at least 5° of ≥ 3 adjacent thoracic vertebral bodies.

Other findings include: Schmorl nodes (see page 455) and endplate narrowing.

Presentation

Adolescents: often present as a result of the cosmetic deformity associated with progressive kyphosis which may be mistaken for “slouching”.

Adults: often present with pain.

Radiographic findings

Anterior wedge deformities at multiple levels. End plate irregularities and Schmorl’s nodes.

Treatment

Bracing may be used in adolescents.

Adults presenting with pain often respond to nonsurgical treatment including: physical therapy and NSAIDs.

Surgical indications: refractory pain, progressive kyphosis, or neurologic deficit.

18.14. Spinal vascular malformations

Often also referred to by the term spinal AVMs which technically refers to a subset of spinal vascular malformations (SVMs). Incidence of SVM is about 4% of primary intraspinal masses. 80% occur between age 20 and 60 years^{470 (p 1850-3)}.

CLASSIFICATION

Classification is evolving (see review by Black⁴⁷¹). 3 current era systems:

The “American/English/French Connection^A classification

References for classification: include^472-479

• Type I: dural AVM AKA AV-fistula (AVF). The most common type (80%) of SVM in the adult⁴⁸⁰. Fed by radicular artery which forms an AV shunt (fistula) at the dural root sleeve (located in the intervertebral foramen)⁴⁷⁷, drains into an engorged spinal vein on posterior cord. Usually in lumbar or lower thoracic spine. Slow flow. High pressure in draining vein may cause venous congestion of the cord. Cord involvement may be distant to the fistula. Symptoms: LBP and progressive myeloradiculopathy or cauda equina syndrome (due to venous congestion) with urinary retention usually in middle-aged patients, 90% males. Up to 35% have pain. 15-20% are associated with other AVMs (cutaneous or other). Rarely bleed

Type IA: single arterial feeder

Type IB: two or more arterial feeders

• intradural AVMs (high flow): 75% present with acute onset of symptoms, usually from hemorrhage (SAH or intramedullary)

Type II: AKA spinal glomus AVM. Intramedullary. True AVM of the spinal cord. 15-20% of all SVMs. Compact nidus fed by medullary arteries with the AV shunt contained at least partially within the spinal cord or pia. May be associated with feeding artery aneurysms. Worse prognosis than dural AVM⁴⁷⁷. Fed by 1, or at most 2-3 feeders 80% of the time

Type III: AKA juvenile spinal AVM. Essentially an enlarged glomus AVM which occupies the entire cross-section of the cord and invades the vertebral body which may cause scoliosis

Type IV⁴⁷⁶: intradural perimedullary AVM (also called arteriovenous fistulae - (AVF)). Direct fistula between artery supplying spinal cord (usually anterior spinal artery, often artery of Adamkiewicz) and draining veins. Typically occur in younger patients than Type I, and may present catastrophically with hemorrhage into the subarachnoid space⁴⁸¹. Table 18-36 shows the 3 subtypes⁴⁷⁸.

• miscellaneous spinal vascular lesions:

spinal cord cavernomas

spinal cord venous angiomas: extremely rare. Difficult to visualize angiographically

vertebral body hemangiomas (see page 738)

A. this descriptive label was coined by Black⁴⁷¹

Hôpital Bicêtre classification⁴⁸²

A. AVMs

B. fistulae: micro- or macrofistulae

C. genetic classification of spinal cord AV shunts

a. genetic hereditary lesions: macrofistulae and hereditary hemorrhagic telangiectasias

b. genetic nonhereditary lesions: multiple lesions with metameric or myelomeric associations

c. single lesions: incomplete associations of categories a or b

Spetzler, et al. classification⁴⁸³

(this system reincorporated vascular spinal neoplasms)

1. neoplastic vascular lesions

A. hemangioblastoma

B. cavernous malformation

2. spinal aneurysms (rare)

3. arteriovenous lesions

A. AVFs

• extradural

• intradural: dorsal or ventral

B. AVMs

• extradural-intradural

• intradural

• intramedullary

• intramedullary-extramedullary

• conus medullaris

PRESENTATION

85% present as progressive neuro deficit (back pain associated with progressive sensory loss and LE weakness over months to years). Yet, SVMs account for < 5% of lesions presenting as spinal cord “tumors”. 10-20% of SVMs present as sudden onset of myelopathy usually in patients < 30 yrs age^{484, 485}, secondary to hemorrhage (causing SAH, hematomyelia, epidural hematoma, or watershed infarction). Coup de poignard of Michon = sudden excruciating back pain with SAH (clinical evidence of SVM).

Foix-Alajouanine syndrome (subacute necrotic myelopathy): acute or subacute neurologic deterioration in a patient with a SVM without evidence of hemorrhage. Presents as spastic → flaccid paraplegia, with ascending sensory level and loss of sphincter control. Initially thought to be due to spontaneous thrombosis of the AVM causing sub-acute necrotizing myelopathy⁴⁸⁶ which would be irreversible. However, more recent evidence suggests that the myelopathy may be due to venous hypertension with secondary ischemia, and there may be improvement with treatment⁴⁸⁷.

CLINICAL

Auscultation over spine reveals a bruit in 2-3% of cases. Cutaneous angioma over back is present in 3-25%; valsalva maneuver may enhance the redness of the angioma⁴⁸⁵.

EVALUATION

Spinal angiography: necessary for treatment planning. Best performed at centers that do this study regularly. For Type I dural AVMs, angiography must encompass all dural feeders of the neuraxis, which includes:

1. ICAs: because of the artery of Bernasconi & Cassinari (see page 99)

2. every radicular artery including the artery of Adamkiewicz (see page 96)

3. internal iliac arteries: for sacral feeders

MRI: detects some SVMs with greater sensitivity and safety than angiography⁴⁸⁸, but is inadequate for treatment planning. 82% show extramedullary flow voids. Variable degree of cord enhancement (from venous congestion or venous infarction). Negative MRI does not rule out diagnosis.

Myelography: classically shows serpiginous intradural filling defects. Generally superseded by MRI. If done, patient should be imaged prone and supine (to avoid missing a dorsal AVM) ✖ Risk of bleeding from puncture of a dilated artery/vein with myelography needle.

TREATMENT

Type I (dural AVMs): usually require treatment. Usually amenable to endovascular techniques using glue, in which case the proximal vein must be taken as well. If you don’t completely eliminate a dural fistula (spinal or intracranial) it will come back!

Type II (spinal glomus AVMs): may be amenable to interventional neuroradiologic procedures including embolization⁴⁸⁹, especially type IIA (single feeder). However recurrence may be higher with endovascular treatment than surgery, and surgery is often preferred for Type IIB (≥ 2 feeders).

Surgical strategy: similar to intracranial AVMs except that the parenchyma cannot be retracted, bleeding is rarely life threatening, and arteries of passage must be preserved to avoid devastating deficits. Intraoperative ICG angio is often helpful. The nidus is compact, and the hemosiderin ring around the nidus on MRI often represents a plane that can be exploited.

Type III (juvenile spinal AVMS): the natural history is probably better than the prognosis with any type of treatment.

Type IV (perimedullary fistulae): see Table 18-37 for suggested management⁴⁷⁹.

18.15. Spinal meningeal cysts

Spinal meningeal cysts (SMC): diverticula of the meningeal sac, nerve root sheath or arachnoid. May have familial tendency.

Terminology in literature is confusing. One classification system is shown in Table 18-38. Previously AKA Tarlov’s perineural cysts, spinal arachnoid cysts, and extradural diverticula, pouches or cysts. Only congenital lesions are considered here.

• Type I SMCs above the sacrum usually have a pedicle adjacent to entrance of dorsal nerve root

• Type II SMCs: formerly called Tarlov’s cysts and were differentiated from nerve root diverticula because the former were defined as communicating with subarachnoid space, and the latter not. However, intrathecal contrast CT (ICCT) shows both communicate. Often multiple, occur on dorsal roots anywhere, but are most prominent and symptomatic in sacrum

• Type III SMCs: may also be multiple and asymptomatic. More common along posterior subarachnoid space. Attributed to proliferation of arachnoid trabeculae

Table 18-38 Types of spinal meningeal cysts⁴⁹⁰

Type	Description
Type I	extradural meningeal cysts without spinal nerve root fibers
IA	“extradural meningeal/arachnoid cyst”
IB	(occult) “sacral meningocele”
Type II	extradural meningeal cysts with spinal nerve root fibers (“Tarlov’s perineural cyst”, “spinal nerve root diverticulum”)
Type III	spinal intradural meningeal cysts (“intradural arachnoid cyst”)

PRESENTATION

May be asymptomatic (i.e. incidental finding). May cause radiculopathy by pressure on adjacent nerve root (may or may not cause symptoms of nerve root from which it actually arises). Symptom complex depends on size of SMC, and proximity to spinal cord and nerve roots.

• Type I SMCs: in thoracic and cervical region, may present with acute myelopathy (spasticity and sensory level); lumbar region → LBP and radiculopathy; sacral region → sphincter disturbance

• Type II SMCs: often asymptomatic, but sacral lesions may → sciatica and/or sphincter disturbance

• Type III SMCs: may also be multiple and asymptomatic; more common along posterior subarachnoid space

EVALUATION

MRI to identify the mass, then water-soluble ICCT scan to evaluate communication of cyst with subarachnoid space.

• Type II SMCs: all 18 cases had bony erosion (demonstrated by canal widening, pedicle erosion, foraminal enlargement, or vertebral body scalloping)

• Type III SMCs: may also cause bony erosion; appear on myelogram as intradural defect, may not appear on ICCT if they communicate with subarachnoid space which causes them to blend with adjacent subarachnoid space

TREATMENT

• Type I SMCs: close ostium between cyst and subarachnoid space. Above sacrum, can usually be dissected from dura; occasionally fibrous adhesions prevent this

• Type II SMCs: no pedicle, thus either partially resect and oversew cyst wall, or excise cyst and involved nerve root. Simple aspiration is not recommended

• Type III SMCs: excise completely unless dense fibrous adhesions prevent this, in which case marsupialize cyst. Tend to recur if incompletely excised

18.16. Syringomyelia

Key concepts:

• AKA syrinx. Cystic cavitation of the spinal cord

• 70% are associated with Chiari I malformation, 10% with basilar invagination. May also be posttraumatic or associated with tumor, infection…

• symptoms: progressive neurologic deterioration over months to years, usually affecting UE first

• diameter > 5 mm + associated edema predict a more rapid deterioration

• preferred treatment is directed at correcting the causative pathophysiology

AKA syrinx. Cystic cavitation of the spinal cord. Other terms not precisely defined include: hydrosyringomyelia, communicating or noncommunicating syringomyelia.

Syringobulbia: Rostral extension into brainstem (usually medulla). May present with (bilateral) peri-oral tingling and numbness, due to compression of the spinal trigeminal tracts as the fibers decussate.

DISTINGUISHING FROM SIMILAR ENTITIES

1. tumor cyst:

A. especially with intramedullary spinal cord gliomas. Tumors may secrete fluid, or may cause microcysts that eventually coalesce. Most (but not all) intramedullary tumors will enhance with IV contrast on MRI

B. tumor cyst fluid is usually highly proteinaceous, syrinx fluid usually has the same MRI characteristics as CSF (NB: true syrinx can occur with tumor)

2. central spinal canal

A. residual central spinal canal: the central canal is present within the spinal cord at birth and normally gradually involutes with age⁴⁹¹. Persistence of the canal is a normal variant. Characteristic imaging features:

1. linear or fusiform on sagittal MRI

2. ≤ 2-4 mm in maximal width

3. may be singular, or there may be several discontinuous regions in the rostral-caudal direction

4. perfectly round in cross-section and centrally located on axial MRI

5. if IV contrast is given, there should be no enhancement

B. simple dilatation of the central canal with ependymal cell lining has some-times been called hydromyelia, but this usage is ambiguous

ETIOLOGIES

1. primary syringomyelia: this term is used differently by different authors⁴⁹². Herein, refers to syrinx in the absence of identifiable cause

2. secondary syringomyelia: Most cases are thought to be secondary to partial obstruction of the spinal subarachnoid space⁴⁹². Unanswered question: Why then do patients with varying degrees of degenerative cervical spinal stenosis generally not get syringomyelia?

A. Chiari I malformation: the most common cause of syrinx (see page 233)

B. postinflammatory

1. postinfectious

a. granulomatous meningitides (TB and fungal)

b. postoperative meningitis, especially after intradural procedure

2. chemical or other sterile inflammations

a. rarely after SAH

b. after myelography: especially with older agents no longer in use (ophendylate (Pantopaque®))

C. posttraumatic: also see below

1. with severe posttraumatic kyphotic deformity: e.g. with retropulsed bone, scarring…

2. arachnoid scarring without recognized trauma

3. severe injury to spinal cord and/or its coverings. Blood may be a contributing factor

✖ the older concept of syrinx developing as a coalescence of foci of traumatic hematomyelia has not been borne out

D. postsurgical: has been identified many years after uncomplicated intradural neoplasm removal (e.g. neurofibromas)

E. basilar arachnoiditis:

1. idiopathic

2. postinfectious: see above

F. basilar impression (with constriction of the foramen magnum, page 138)

G. associated with spinal tumors: this is distinct from a tumor cyst

H. associated with disc protrusion:

I. cerebellar ectopia

J. Dandy Walker syndrome

EPIDEMIOLOGY⁴⁹³

Prevalence of non-posttraumatic syringomyelia: 8.4 cases/100,000 population. Usually presents between ages 20-50.

Associated clinical syndromes are shown in Table 18-39.

Table 18-39 Conditions associated with syringomyelia

Condition	%*
Chiari type 1 malformation	70
basilar invagination	10
intramedullary spinal cord tumors	4

* percent of cases of syringomyelia

PATHOPHYSIOLOGY

Major theories of formation of the cyst:

• hydrodynamic (“water-hammer”) theory of Gardner: systolic pulsations are transmitted with each heartbeat from the intracranial cavity to the central canal. Has been essentially disproven using MRI⁴⁹⁴

• Williams’ (“craniospinal dissociation”) theory: maneuvers that raise CSF pressure (valsalva, coughing…) cause “hydrodissection” through the spinal cord tissue. May be more common in noncommunicating syringomyelia

• Heiss-Oldfield theory: occlusion at the foramen magnum causes CSF pulsations during cardiac systole to be transmitted through the Virchow-Robin spaces which increases the extracellular fluid which coalesce to form a syrinx⁴⁹³(i.e. through the cord parenchyma)

CLINICAL

Presentation: highly variable. Usually progresses over months to years, with a more rapid deterioration early that gradually slows⁴⁹³. Initially, pain, weakness, atrophy and loss of pain & temperature sensation in the upper extremities (with cervical syrinx) is common. Myelopathy that progresses slowly over years ensues.

Characteristic syndrome

(nonspecific for intramedullary spinal cord pathology):

• sensory loss (similar to central cord syndrome) with a suspended (“cape”) dissociated sensory loss (loss of pain and temperature sensation with preserved touch and joint position sense → painless ulcerations from unperceived injuries and/or burns)

• pain: commonly cervical and occipital. Dysesthetic pain often occurs in the distribution of the sensory loss⁴⁹³

• weakness: lower motor neuron weakness of the hand and arm

• painless (neurogenic) arthropathies (Charcot’s joints) especially in the shoulder & neck due to loss of pain & temperature sensation: seen in < 5%

EVALUATION

Prior to the CT/MRI era, diagnosis relied on myelography or on autopsy.

MRI: defines anatomy in sagittal as well as axial plane. Test of choice. Cervical & thoracic spine and brain MRI (without & with contrast, to include craniocervical junction) should be obtained. Syringomyelic cavities may be complex, with noncommunicating channels (more common with posttraumatic syrinx).

CT: low attenuation area within cord seen on either plain CT or myelogram/CT (with water soluble contrast).

Myelogram: rarely used alone (usually performed in conjunction with CT). When used alone: often normal (false negative), some → complete block at level of syrinx; iodine contrast studies may show fusiform widening of spinal cord, whereas air contrast studies may show collapse of the cord⁴⁹⁵. Dye may slowly leach into the cyst.

EMG: no characteristic findings, but may be useful to R/O other conditions that may be responsible for symptoms (e.g. peripheral neuropathy causing paresthesias).

MANAGEMENT

For an incidentally discovered syrinx (i.e. asymptomatic and no neurologic deficit) with no identified etiology, if the size remains stable over 2-3 years of observation, F/U studies at 2-3 year intervals can be done if there are no changes in symptoms.

SURGICAL TREATMENT

Intervention is considered for symptomatic lesions (not all are symptomatic). If an underlying cause cannot be determined, it may be very difficult to treat a very small syrinx directly (however, these are unlikely to be causing reversible symptoms).

Options include:

1. current philosophy is to treat the underlying pathophysiology (and to use syrinx draining procedures as second choice when this is not feasible)

A. posterior decompression: procedure of choice when posterior anomalies (e.g. Chiari malformation) are present

B. decompression if a different site of compression is identified

2. shunts:

A. disadvantages:

1. complication rate: 16%

2. clinical stabilization rate: 54% at 10 yrs

3. may produce traction on spinal cord with potential for further injury

4. prone to obstruction: 50% at 4 years

5. does not correct underlying pathophysiology and so syrinx may recur

B. indications: cases of diffuse arachnoiditis (e.g. following tuberculous or chemical meningitis) where the obstruction extends over many levels, and syrinx diameter > 3-4 mm

C. K or T tube drainage. Choice of distal sites includes:

1. peritoneum⁴⁹⁶ (difficult in cervical region)

2. pleural cavity

3. subarachnoid space (e.g. Heyer-Schulte-Pudenz system): requires normal CSF flow in subarachnoid space, therefore cannot use in arachnoiditis

3. percutaneous aspiration of the cyst⁴⁹⁷ (may be used repeatedly)

4. ✖ no longer recommended:

A. plugging the obex with muscle, teflon or other material

B. opening the subarachnoid space & removing inferior tonsils

C. syringostomy: usually fails to remain patent, therefore using a stent or a shunt (syringosubarachnoid or syringoperitoneal) is recommended

Technical considerations:

1. intraoperative ultrasound may be helpful for:

A. localizing the cyst

B. assessing for septations (to avoid shunting only part of cyst)

2. if Chiari malformation is not present, consider syringosubarachnoid shunt as the initial procedure. If this fails, then syringoperitoneal shunt may be inserted

3. Rhoton suggests performing the myelotomy in the dorsal root entry zone (DREZ), between the lateral and posterior columns (instead of the midline as with a tumor) because this is consistently the thinnest part and there is usually already an upper extremity proprioceptive deficit from the syrinx^{123 (p 1317)}. There is ≈ 10% incidence of posterior column dysfunction with shunting

4. with syringosubarachnoid shunts, be sure the distal shunt tip is subarachnoid (and not just subdural) or else it will not function

5. with syringopleural shunt, the pleural opening can be made posteriorly, adjacent to one of the ribs as described for ventriculopleural shunt (see page 210)

OUTCOME

Assessing treatment results is difficult due to rarity of the condition, variability of natural history (which may arrest spontaneously), and too short follow-up⁴⁹⁸. Enthusiasm for direct treatment (shunts, fenestration…) is low among neurosurgeons because of the perceived poor response and risk of iatrogenic neurologic worsening.

18.16.1. Posttraumatic syringomyelia

Posttraumatic syringomyelia (PTSx) may follow significant spinal trauma (with or without clinical spinal cord injury). Includes penetrating injury or non-penetrating “violent” trauma to the spinal cord (injuries such as post-spinal anesthesia or following thoracic disc herniation are not included).

EPIDEMIOLOGY

Often a late presentation following spinal cord injury, therefore incidence is higher in series with longer follow-up. Incidence increasing with increasing survival following spinal cord injury and with increasing use of MRI. Range: ≈ 0.3-3% of cord injured patients (see Table 18-40).

In a large number of patients followed via multicenter cooperative data bank, there were fewer cases of syrinx following cervical injuries than following thoracic injuries⁵⁰⁰ (may be artifactual since patients with lower lesions may be more aware of ascending levels).

Table 18-40 Incidence of posttraumatic syringomyelia

Type of injury	No./risk*	Incidence
all spinal cord injury patients	30/951	3.2%
complete quadriplegics	14/177	7.9%
incomplete quadriplegics	4/181	4.5%
complete paraplegics	4/282	1.7%
incomplete paraplegics	4/181	2.2%

* number occurring over number at risk in 951 patients followed for 11 years⁴⁹⁹

Latency following spinal cord injury:

• latency to symptoms: 3 mos to 34 yrs (mean 9 yrs) (earlier in complete cord lesions than incomplete: mean 7.5 vs. 9.9 yrs)

• latency to diagnosis: up to 12 yrs (mean 2.8 yrs) after onset of new symptoms

CLINICAL

The presentation of patients with PTSx is shown in Table 18-41. The late appearance of upper extremity symptoms in a paraplegic patient should raise a high index of suspicion of posttraumatic syringomyelia⁵⁰².

Hyperhidrosis may be the only feature of descending syringomyelia in patients with complete cord lesions⁵⁰³. For the differential diagnosis, see Delayed deterioration following spinal cord injuries, page 1000.

EVALUATION

One end of the cavity is often found at a site of spinal column fracture or abnormal angulation.

MANAGEMENT

Many authors advocate early surgical drainage of cyst as a means of reducing increased delayed deficit⁵⁰⁴. Some authors feel that aside from disturbing sensory symptoms, that motor loss was infrequent and therefore conservative management is indicated in most cases⁵⁰⁵.

MEDICAL

Managed non-surgically: 31% stable, 68% progressed over yrs (longer F/U in latter).

SURGICAL

There is probably no benefit in operating on a patient with a small syrinx⁴⁹⁹.

Surgical options:

As in Communicating syringomyelia, with the following differences:

• cord transection (cordectomy)⁵⁰⁶: an option in complete injuries only

• plugging the obex is probably not indicated (controversial in congenital syrinx)

Table 18-41 Presentation (in 30 SCI patients with syrinx⁴⁹⁹)

Symptom	Initial	At time of diagnosis
pain*	57%	70%
numbness	27%	40%
increased motor deficit	23%	40%
increased spasticity	10%	23%
increased sweating (hyperhidrosis)	3%	13%
autonomic dysreflexia	3%	3%
no symptoms	7%	7%
Signs	Frequency
ascending sensory level	93%
depressed tendon reflexes	77%
increased motor deficits	40%

* pain is often quite severe, and unrelieved with analgesics⁵⁰¹

OUTCOME

In 9 PTSx patients treated with syringosubarachnoid shunt⁴⁹⁹: pain relieved in all 9 (1 only slightly), motor recovery in 5/8, improved tendon reflex in 1/10. Some post-op complications in 9 patients included: 1 incomplete lesion became complete, 1 sensorimotor deterioration, transient pain in 3.

Most results are good for radicular symptoms, with dubious efficacy for autonomic symptoms or spasticity.

18.17. Spinal cord herniation (idiopathic)

Rare. Spinal cord herniates though a defect in the dura usually located anteriorly or anterolaterally between T2-8507. Bone erosion anterior to the dural defect may occasionally be seen. Frequently associated with a calcified disc fragment which theoretically may have gradually eroded through the dura.

Main DDx is with dorsal arachnoid cyst (see page 1186). Both result in increased subarachnoid space posterior to the cord, and a ventral kinking of the spinal cord. Contiguous CSF pulsation artifact on MRI can be seen with cord herniation, whereas an arachnoid cyst tends to interrupt this.

Commonly presents as an incomplete Brown-Séquard syndrome (with relative sparing of posterior columns). Symptoms may be due to distortion of the spinal cord, but vascular injury may also play a role.

Surgery

Requires a posterolateral or anterolateral approach to minimize spinal cord manipulation (see page 471). The dural defect is widened which usually results in reduction of the spinal cord herniation. A sling of dural substitute can then be slid anterior to the cord to prevent reherniation.

18.18. Spinal epidural hematoma

Rare. Over 200 cases of varying etiology have been reported⁵⁰⁸, although one third of recent cases have been associated with anticoagulation therapy⁵⁰⁹. NSAIDs may also be a risk factor⁵¹⁰. Etiologies include:

1. traumatic: including following LP or epidural anesthesia^{508, 511-513}, fracture (see below), spinal surgery¹⁶² or chiropractic manipulation⁵¹⁴. Occurs predominantly in patient who is: anticoagulated⁵¹⁵, thrombocytopenic, or has bleeding diathesis or a vascular lesion

2. spontaneous⁵¹⁶: rare. Etiologies: hemorrhage from spinal cord AVM (see page 507), from vertebral hemangioma (see page 738) or tumor

May occur at any level of the spine, however, thoracic is most common. Most often located posterior to spinal cord (except for hematomas following anterior cervical procedures), facilitating removal via laminectomy⁵⁰⁹.

Traumatic spinal epidural hematoma (TSEH) associated with spine fracture:

In one series⁵¹⁷, among 74 trauma patients who underwent emergent spinal MRI, ≈ half of the patients with spine fractures also had TSEH. Treatment was based solely on the fracture, and the outcome in patients with neurologic deficits was no worse in the group with TSEH than in the group without.

Presentation

The clinical picture of spontaneous spinal epidural hematoma is fairly consistent but nonspecific. Usually starts with severe back pain with radicular component. It may occasionally follow minor straining, and is less commonly preceded by major straining or back trauma. Spinal neurologic deficits follow, usually progressing over hours, occasionally over days. Motor weakness may go unnoticed when patients are bedridden with pain.

Treatment

Recovery of neurologic deficit without surgery is rare⁵¹⁰, therefore optimal treatment is immediate decompressive laminectomy in those patients who can tolerate surgery⁵⁰⁹. In one series, most patients who recovered underwent decompression within 72 hrs of onset of symptoms⁵¹⁸. In another, decompression within 6 hours was associated with better outcome¹⁶².

High-risk patients: for medically high-risk patients (e.g. acute MI) on anticoagulation, surgical mortality and morbidity is extremely high, and this must be considered when making the decision of whether or not to operate. In patients not operated, anticoagulants should be stopped, and reversed if possible (see Correction of coagulopathies or reversal of anticoagulants, page 40). Consider use of high dose methylprednisolone to minimize cord injury (see Methylprednisolone, page 936 under spinal cord injury). Percutaneous needle aspiration may be a consideration in high-risk patients.

18.19. Spinal subdural hematoma

Rare. May be posttraumatic (including iatrogenic causes) or may occur spontaneously. Spinal subdural hematomas (SSH) that occur spontaneously or following lumbar puncture usually occur in patients with coagulopathies (primary or iatrogenic)⁵¹⁹.

Conservative treatment is possible in nontraumatic SSHs with minimal neurologic impairment⁵¹⁹.

18.20. Spinal epidural lipomatosis (SEL)

Hypertrophy of epidural fat, due to prolonged exogenous steroid therapy in most (75%) cases⁵²⁰ (usually moderate to high dosage for years⁵²¹), but may also be associated with: Cushing’s disease, Cushing’s syndrome, obesity⁵²², hypothyroidism or may be idiopathic⁵²³. Male:female = 3:1521.

Back pain usually precedes all other symptoms. Progressive LE weakness and sensory changes are common. Sphincter disturbance occurs but is rare. SEL is most common in the thoracic spine (≈ 60% of cases), the rest are in the lumbar spine (no cases reported in cervical spine).

Evaluation

CT: density of adipose tissue is extremely low (-80 to -120 Hounsfield units)⁵²⁴ which distinguishes SEL from most other lesions (except lipoma).

MRI: signal follows fat (high signal on T1WI, intermediate on T2WI). Suggested diagnostic criteria: epidural adipose should be > 7 mm thick to be considered SEL^{522, 525}.

Treatment

In those patients who can be weaned off steroids and lose weight, surgery may be avoided in some cases⁵²⁶. If SEL is related to obesity, weight loss alone can be successful.

Surgery is indicated for symptomatic patients in whom the above interventions are unsuccessful or not feasible. An effort to normalize cortisol levels in those with endogenous hypercortisolism (Cushing’s disease…) should be made before laminectomy is performed. Due to potential complications and slow growth of the tissue, the decision to operate surgery should made with caution.

Surgery usually consists of laminectomy with removal of adipose tissue. Occasionally repeat surgery is needed for reaccumulation of adipose tissue.

Outcome

Surgery usually results in significant improvement⁵²⁵. Idiopathic cases may fare better than those due to steroid excess. Cauda equina compression responds better than thoracic myelopathy.

Complications rates may be higher than expected in part due to medical comorbidities. Fessler et al.⁵²⁷ reported 22% 1-year mortality.

18.21. Coccydynia

Pain and tenderness around the coccyx. A symptom, not a diagnosis. Typically, discomfort is experienced on sitting or on rising from sitting. More common in females, possibly due to a more prominent coccyx. The condition is unusual enough in males that in the absence of trauma, strong consideration should be given to an underlying condition.

Etiologies

For differential diagnosis, see Acute low back pain, page 1192. Better accepted etiologies include⁵²⁸:

1. local trauma (may be associated with fracture or dislocation):

A. 25% of patients give a history of a fall

B. 12% had repetitive trauma (rowing machine, prolonged bicycle riding…)

C. 12% started with parturition

D. 5% started following a surgical procedure (half of which were in the lithotomy position)

2. idiopathic: excluding traumatic cases, no etiology can be identified in most cases

3. neoplasms

A. chordoma

B. giant cell tumor

C. intradural schwannoma

D. perineural cyst

E. intra-osseous lipoma

F. carcinoma of the rectum

G. sacral hemangioma⁵²⁹

H. pelvic metastases (e.g. from prostate cancer)

4. prostatitis

Controversial etiologies include^{528, 530}:

1. local pressure over a prominent coccyx

2. referred pain:

A. spinal disease

1. herniated lumbosacral disc

2. cauda equina syndrome

3. arachnoiditis

B. pelvic/visceral disease

1. pelvic inflammatory disease (PID)

2. perirectal abscess

3. perirectal fistula

4. pilonidal cyst

3. inflammation of the various ligaments attached to the coccyx

4. neurosis or frank hysteria

Histological evaluation of the coccyx has not helped delineate the cause, even though avascular necrosis has been suggested⁵³¹.

Evaluation

Sacrococcygeal films are often performed to rule-out a bony destructive lesion. Often, the question of a fracture will be raised, and many times cannot be definitely ruled-in or out based on this study. There may or may not be any significance of such a fracture.

Nuclear bone scans were not helpful in 50 patients with coccydynia⁵²⁸.

CT scan: no consistent findings.

Treatment

Most cases resolve within ≈ 3 months of conservative management consisting of NSAIDs, mild analgesics, and measures to reduce pressure on the coccyx (e.g. a rubber ring (“doughnut”) sitting cushion, lumbar supports to maintain sitting lumbar lordosis to shift weight from coccyx to posterior thighs)⁵³².

Recurrence: Occurs in ≈ 20% of conservatively treated cases, usually within the first year. Repeat therapy was often successful in providing permanent relief. More aggressive treatment may be considered for refractory cases.

Management recommendations for refractory cases^{528, 532}:

1. local injection: 60% respond to corticosteroid + local anesthetic (40 mg Depo-Medrol® in 10 cc of 0.25% bipuvicaine). Recommended as initial treatment; response should be achieved by 2 injections

2. manipulation of the coccyx: usually under general anesthesia. ≈ 85% successful when combined with local injection

3. ± physiotherapy (diathermy & ultrasound): found to be of benefit only in ≈ 16%

4. caudal epidural steroid injection

5. blockade or neurolysis (with chemicals or by cryoablation⁵³³) of the ganglion impar (AKA ganglion of Walther, the lowest ganglion of the paired paravertebral sympathetic chain, located just anterior to the sacrococcygeal junction): some success has been described with this technique (traditionally used for intractable sympathetic perineal pain of neoplastic etiology⁵³⁴)

6. neurolytic techniques directed to S4, S5 and coccygeal nerves

7. coccygectomy (surgical removal of the mobile portion of the coccyx, followed by smoothening of the residual bony prominence on the sacrum): was required in ≈ 20% of patients in one series⁵²⁸, with a reported success rate of 90%. However, many practitioners do not view this as a highly effective treatment and feel that great restraint should be used in considering this form of therapy

18.22. References

1. Bigos S, Bowyer O, Braen G, et al.: Acute low back problems in adults. Clinical practice guideline no. 14. Ahcpr publication no. 95-0642. Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services, Rockville, MD, 1994.

2. Cypress B K: Characteristics of physician visits for back symptoms: A national perspective. Am J Public Health 73: 389-95, 1983.

3. Cunningham L S, Kelsey J L: Epidemiology of musculoskeletal impairments and associated disability. Am J Public Health 74: 574-9, 1984.

4. Frymoyer J W: Back pain and sciatica. N Engl J Med 318: 291-300, 1988.

5. Fardon D F, Milette P C: Nomenclature and classification of lumbar disc pathology. Recommendations of the combined task forces of the North American spine society, American society of spine radiology, and American society of neuroradiology. Spine 26 (5): E93-E113, 2001.

6. McCarron R F, Wimpee M W, Hudkins P G, et al.: The inflammatory effect of nucleus pulposus: A possible element in the pathogenesis of low-back pain. Spine 12: 760-4, 1987.

7. Modic M T: Degenerative disorders of the spine. In Magnetic resonance imaging of the spine. Year-book Medical, New York, 1989: pp 83-95.

8. Kuklo T R, Polly D W, Owens B D, et al.: Measurement of thoracic and lumbar fracture kyphosis: Evaluation of intraobserver, interobserver, and technique variability. Spine 26 (1): 61-5; discussion 66, 2001.

9. Fairbank J C, Couper J, Davies J B, et al.: The Oswestry low back pain disability questionnaire. Physiotherapy 66 (8): 271-3, 1980.

10. Fairbank J C, Pynsent P B: The Oswestry disability index. Spine 25 (22): Spine: 2940-52; discussion 2952, 2000.

11. Baker D, Pynsent P, Fairbank J: The Oswestry disability index revisited. In Back pain: New approaches to rehabilitation and education, Roland M and Jenner J, (eds.). Manchester University Press, Manchester, 1989: pp 174-86.

12. Roland M, Morris R: A study of the natural history of back pain. Part I: Development of a reliable and sensitive measure of disability in low-back pain. Spine (Phila Pa 1976) 8 (2): Spine (Phila Pa 1976): 141-4, 1983.

13. Grevitt M, Khazim R, Webb J, et al.: The short form-36 health survey questionnaire in spine surgery. J Bone Joint Surg Br 79 (1): J Bone Joint Surg Br: 48-52, 1997.

14. Kelsey J L: Idiopathic low back pain: Magnitude of the problem. American Academy of Orthopedic Surgeons Symposium on Idiopathic Low Back Pain. St. Louis, C. V. Mosby Co. 1982, pp 5-8.

15. Spengler D M, Ouellette E A, Battié M, et al.: Elective discectomy for herniation of a lumbar disc. Additional experience with an objective method. J Bone Joint Surg 72A: 230-7, 1990.

16. Braddom R L, Johnson E W: Standardization of H reflex and diagnostic use in S1 radiculopathy. Arch Phys Med Rehabil 55: 161-6, 1974.

17. Schütte H E, Park W M: The diagnostic value of bone scintigraphy in patients with low back pain. Skeletal Radiol 10: 1-4, 1983.

18. Whalen J L, Brown M L, McLeod R, et al.: Limitations of indium leukocyte imaging for the diagnosis of spine infections. Spine 16: 193-7, 1991.

19. Mills G H, Davies G K, Getty C J M, et al.: The evaluation of liquid crystal thermography in the investigation of nerve root compression due to lumbosacral lateral spinal stenosis. Spine 11: 427-32, 1986.

20. Harper C M, Low P A, Fealy R D, et al.: Utility of thermography in the diagnosis of lumbosacral radiculopathy. Neurology 41: 1010-4, 1991.

21. Deyo R A, Bigos S J, Maravilla K R: Diagnostic imaging procedures for the lumbar spine. Ann Intern Med 111: 865-7, 1989 (editorial).

22. Hitselberger W E, Witten R M: Abnormal myelo-grams in asymptomatic patients. J Neurosurg 28: 204-6, 1968.

23. Wiesel S W, Tsourmas N, Feffer H L, et al.: A study of computer-assisted tomography. I. The incidence of positive cat scans in an asymptomatic group of patients. Spine 9: 549-51, 1984.

24. Jensen M C, Brant-Zawadzki M N, Obuchowski N, et al.: Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med 331: 69-73, 1994.

25. Boden S D, Davis D O, Dina T S, et al.: Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. J Bone Joint Surg 72A: 403-8, 1990.

26. Spitzer W O, LeBlanc F E, Dupuis M, et al.: Scientific approach to the assessment and management of activity-related spinal disorders: A monograph for clinicians: Report of the quebec task force on spinal disorders. Chapter 3: Diagnosis of the problem (the problem of diagnosis). Spine 12: S16-21, 1987.

27. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 6: Magnetic resonance imaging and discography for patient selection for lumbar fusion. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 662-9, 2005.

28. Holt E P: The question of lumbar discography. J Bone Joint Surg 50A: 720-6, 1968.

29. Carragee E J, Tanner C M, Khurana S, et al.: The rates of false-positive lumbar discography in select patients without low back symptoms. Spine 25 (11): 1373-80; discussion 1381, 2000.

30. Carragee E J, Paragioudakis S J, Khurana S: 2000 Volvo award winner in clinical studies: Lumbar high-intensity zone and discography in subjects without low back problems. Spine 25 (23): 2987-92, 2000.

31. Nachemson A L: The lumbar spine: An orthopedic challenge. Spine 1: 59-71, 1976.

32. World Health Organization: A rational approach to radiodiagnostic investigations, World Health Organization, Geneva. Technical Report Series 689, 1983.

33. Scavone J G, Latschaw R F, Rohrer G V: Use of lumbar spine films: Statistical evaluation of a university teaching hospital. JAMA 246: 1105-8, 1981.

34. Modic M T, Masaryk T, Boumphrey F, et al.: Lumbar herniated disk disease and canal stenosis: Prospective evaluation by surface coil MR, CT, and myelography. AJR 147: 757-65, 1986.

35. Jackson R P, Cain J E, Jacobs R, et al.: The neuroradiologic diagnosis of lumbar herniated nucleus pulposus: II. A comparison of computed tomography (CT), myelography, CT-myelography, and magnetic resonance imaging. Spine 14: 1362-7, 1989.

36. Modic M T, Pavlicek W, Weinstein M A, et al.: Magnetic resonance imaging of intervertebral disk disease. Radiology 152: 103-11, 1984.

37. Bosacco S J, Berman A T, Garbarino J L, et al.: A comparison of CT scanning and myelography in the diagnosis of lumbar disc herniation. Clin Orthop 190: 124-8, 1984.

38. Moufarrij N A, Hardy R W, Weinstein M A: Computed tomographic, myelographic, and operative findings in patients with suspected herniated lumbar discs. Neurosurgery 12: 184-8, 1983.

39. Aejmelaeus R, Hiltunen H, Härkönen M, et al.: Myelographic versus clinical diagnostics in lumbar disc disease. Arch Orthop Trauma Surg 103: 18-25, 1984.

40. Herron L D, Turner J: Patient selection for lumbar laminectomy and discectomy with a revised objective rating system. Clin Orthop 199: 145-52, 1985.

41. Kortelainen P, Puranen J, Koivisto E, et al.: Symptoms and signs of sciatica and their relation to the localization of the lumbar disc herniation. Spine 10: 88-92, 1985.

42. Hirsch C, Nachemson A: The reliability of lumbar disk surgery. Clin Orthop 29: 189, 1963.

43. Slebus F G, Braakman R, Schipper J, et al.: Non-corresponding radiological and surgical diagnoses in patients operated for sciatica. Acta Neurochir 94: 137-43, 1988.

44. Walsh T R, Weinstein J N, Spratt K F, et al.: Lumbar discography in normal patients. A controlled, prospective study. J Bone Joint Surg 72A: 1081-8, 1990.

45. Waddell G, McCulloch J A, Kummel E, et al.: Non-organic physical signs in low back pain. Spine 5: 117-25, 1980.

46. McCombe P F, Fairbank J C T, Cockersole B C, et al.: Reproducibility of physical signs in low-back pain. Spine 14: 908-18, 1989.

47. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 13: Injection therapies, low-back pain, and lumbar fusion. Journal of Neurosurgery: Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 707-15, 2005.

48. Nachemson A L: Newest knowledge of low back pain. A critical look. Clin Orthop 279: 8-20, 1992.

49. Deyo R A, Diehl A K, Rosenthal M: How many days of bed rest for acute low back pain? A randomized clinical trial. N Engl J Med 315: 1064-70, 1986.

50. Vroomen P C A J, de Krom M C T F M, Wilmink J T, et al.: Lack of effectiveness of bed rest for sciatica. N Engl J Med 340: 418-23, 1999.

51. Allen C, Glasziou P, Del Mar C: Bed rest: A potentially harmful treatment needing more careful evaluation. Lancet 354: 1229-33, 1999.

52. Lindström I, Ohlund C, Eek C, et al.: The effect of graded activity on patients with subacute low back pain: A randomized prospective clinical study with an operant-conditioning behavioral approach. Phys Ther 72: 279-93, 1992.

53. Hancock M J, Maher C G, Latimer J, et al.: Assessment of diclofenac or spinal manipulative therapy, or both, in addition to recommended first-line treatment for acute low back pain: A randomised controlled trial. Lancet 370 (9599): 1638-43, 2007.

54. Chlorzoxazone hepatotoxicity. Med Letter 38: 46, 1996.

55. Deyo R A, Diehl A K: Patient satisfaction with medical care for low-back pain. Spine 11: 28-30, 1986.

56. Thomas K B: General practice consultations: Is there any point in being positive? Br Med J 294: 1200-2, 1987.

57. Keijsers J F E M, Bouter L M, Meertens R M: Validity and comparability of studies on the effects of back schools. Physiother Theory Pract 7: 177-84, 1991.

58. Di Fabio R P: Manipulation of the cervical spine: Risks and benefits. Phys Ther 79 (1): 50-65, 1999.

59. Ernst E: Life-threatening complications of spinal manipulation. Stroke 32 (3): 809-10, 2001.

60. Stevinson C, Honan W, Cooke B, et al.: Neurological complications of cervical spine manipulation. J R Soc Med 94 (3): 107-10, 2001.

61. Cuckler J M, Bernini P, Wiesel S, et al.: The use of epidural steroids in the treatment of lumbar radicular pain. A prospective, randomized, double-blind study. J Bone Joint Surg 67A: 63-6, 1985.

62. Spaccarelli K C: Lumbar and caudal epidural corticosteroid injections. Mayo Clin Proc 71: 169-78, 1996.

63. Carette S, Leclaire R, Marcoux S, et al.: Epidural corticosteroid injections for sciatica due to herniated nucleus pulposus. N Engl J Med 336: 1634-40, 1997.

64. Haimovic I C, Beresford H R: Dexamethasone is not superior to placebo for treating lumbosacral radicular pain. Neurology 36: 1593-4, 1986.

65. Meek J B, Giudice V W, McFadden J W, et al.: Colchicine confirmed as highly effective in disk disorders. Final results of a double-blind study. J Neuro & Orthop Med & Surg 6: 211-8, 1985.

66. Schnebel B E, Simmons J W: The use of oral colchicine for low-back pain. A double-blind study. Spine 13: 354-7, 1988.

67. Goodkin K, Gullion C M, Agras W S: A randomized, double-blind, placebo-controlled trial of trazodone hydrochloride in chronic low back pain syndrome. J Clin Psychopharmacol 10: 269-78, 1990.

68. Deyo R A, Walsh N E, Martin D C, et al.: A controlled trial of transcutaneous electrical stimulation (TENS) and exercise for chronic low back pain. N Engl J Med 322: 1627-34, 1990.

69. Mathews J A, Hickling J: Lumbar traction: A double-blind controlled study for sciatica. Rheumatol Rehabil 14: 222-5, 1975.

70. van Poppel N N M, Koes B W, van der Ploeg T, et al.: Lumbar supports and education for the prevention of low back pain in industry: A randomized controlled study. JAMA 279: 1789-94, 1998.

71. Bush C, Ditto B, Feuerstein M: A controlled evaluation of paraspinal EMG biofeedback in the treatment of chronic low back pain. Health Psychol 4: 307-21, 1985.

72. Carette S, Marcoux S, Truchon R, et al.: A controlled trial of corticosteroid injections into facet joints for chronic low back pain. N Engl J Med 325: 1002-7, 1991.

73. Jackson R P: The facet syndrome. Myth or reality? Clin Orthop Rel Res 279: 110-21, 1992.

74. Manheimer E, White A, Berman B, et al.: Meta-analysis: Acupuncture for low back pain. Ann Intern Med 142 (8): 651-63, 2005.

75. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 7: Intractable low-back pain without stenosis or spondylolisthesis. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 670-2, 2005.

76. Fritzell P, Hagg O, Wessberg P, et al.: 2001 Volvo award winner in clinical studies: Lumbar fusion versus nonsurgical treatment for chronic low back pain: A multicenter randomized controlled trial from the Swedish lumbar spine study group. Spine 26 (23): 2521-32; discussion 2532-4, 2001.

77. Ivar Brox J, Sorensen R, Friis A, et al.: Randomized clinical trial of lumbar instrumented fusion and cognitive intervention and exercises in patients with chronic low back pain and disc degeneration. Spine 28 (17): 1913-21, 2003.

78. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 11: Interbody techniques for lumbar fusion. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 692-9, 2005.

79. Turner J A, Ersek M, Herron L, et al.: Patient outcomes after lumbar spinal fusions. JAMA 268: 907-11, 1992.

80. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 8: Lumbar fusion for disc herniation and radiculopathy. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 673-8, 2005.

81. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 12: Pedicle screw fixation as an adjunct to postero-lateral fusion for low-back pain. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 700-6, 2005.

82. Lorenz M, Zindrick M, Schwaegler P, et al.: A comparison of single-level fusions with and without hardware. Spine 16: S455-8, 1991.

83. Gatchel R J, Mayer T G, Capra P, et al.: Quantification of lumbar function, VI: The use of psychological measures in guiding physical functional restoration. Spine 11: 36-42, 1986.

84. Morley S, Pallin V: Scaling the affective domain of pain: A study of the dimensionality of verbal descriptors. Pain 62: 39-49, 1995.

85. Hardacker J W, Shuford R F, Capicotto P N, et al.: Radiographic standing cervical segmental alignment in adult volunteers without neck symptoms. Spine 22 (13): Spine: 1472-80; discussion 1480, 1997.

86. Fon G T, Pitt M J, Thies A C, Jr.: Thoracic kyphosis: Range in normal subjects. AJR Am J Roentgenol 134 (5): 979-83, 1980.

87. Gelb D E, Lenke L G, Bridwell K H, et al.: An analysis of sagittal spinal alignment in 100 asymptomatic middle and older aged volunteers. Spine 20 (12): Spine: 1351-8, 1995.

88. Jang J S, Lee S H, Kim J M, et al.: Can patients with sagittally well-compensated lumbar degenerative kyphosis benefit from surgical treatment for intractable back pain? Neurosurgery 64 (1): Neurosurgery: 115-21; discussion 121, 2009.

89. Weber H: Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine 8: 131-40, 1983.

90. Lewis P J, Weir B K A, Broad R, et al.: Long-term prospective study of lumbosacral discectomy. J Neurosurg 67: 49-53, 1987.

91. Wein A J: Neuromuscular dysfunction of the lower urinary tract and its treatment. In Campbell’s urology, Walsh P C, Retik A B, Vaughan E D, et al., (eds.). W.B. Saunders, Philadelphia, 7th ed., 1998, Vol. 1: pp 953-1006.

92. Jones D L, Moore T: The types of neuropathic bladder dysfunction associated with prolapsed lumbar intervertebral discs. Br J Urol 45: 39-43, 1973.

93. Ross J C, Jameson R M: Vesical dysfunction due to prolapsed disc. Br Med J 3: 752-4, 1971.

94. Deyo R A, Rainville J, Kent D L: What can the history and physical examination tell us about low back pain? JAMA 268: 760-5, 1992.

95. Blaauw G, Braakman R, Gelpke G J, et al.: Changes in radicular function following low-back surgery. J Neurosurg 69: 649-52, 1988.

96. Stam J, Speelman H D, van Crevel H: Tendon reflex asymmetry by voluntary mental effort in healthy subjects. Arch Neurol 46: 70-3, 1989.

97. Scham S M, Taylor T K F: Tension signs in lumbar disc prolapse. Clin Orthop 75: 195-204, 1971.

98. Dyck P: Lumbar nerve root: The enigmatic eponyms. Spine 9: 3-6, 1984.

99. Spangfort E V: The lumbar disc herniation. A computer-aided analysis of 2,504 operations. Acta Orthop Scand 142 (Suppl): 1-93, 1972.

100. Rothman R H, Simeone F A, (eds.): The spine. 3rd ed., W.B. Saunders, Philadelphia, 1992.

101. Estridge M N, Rouhe S A, Johnson N G: The femoral stretch test: A valuable sign in diagnosing upper lumbar disc herniations. J Neurosurg 57: 813-7, 1982.

102. Hoover C F: A new sign for the detection of malingering and functional paresis of the lower extremities. JAMA 51: 746-7, 1908.

103. Archibald K C, Wiechec F: A reappraisal of Hoover’s test. Arch Phys Med Rehabil 51: 234-8, 1970.

104. Sonoo M: Abductor sign: A reliable new sign to detect unilateral non-organic paresis of the lower limb. J Neurol Neurosurg Psychiatry 75: 121-5, 2004.

105. Keegan J J: Dermatome hypalgesia associated with herniation of intervertebral disk. Arch Neurol Psychiatry 50: 67-83, 1943.

106. Fager C A: Observations on spontaneous recovery from intervertebral disc herniation. Surg Neurol 42: 282-6, 1994.

107. Weber H, Holme I, Amlie E: The natural course of acute sciatica, with nerve root symptoms in a double blind placebo controlled trial evaluating the effect of piroxicam (nsaid). Spine 18: 1433-8, 1993.

108. Weber H: The effect of delayed disc surgery on muscular paresis. Acta Orthop Scand 46: 631-42, 1975.

109. Saal J A, Saal J S: Nonoperative treatment of herniated lumbar intervertebral disc with radiculopathy: An outcome study. Spine 14: 431-7, 1989.

110. Marshall R W: The functional relevance of neurological recovery 20 years or more after lumbar discectomy. J Bone Joint Surg Br 90 (5): 554-5, 2008.

111. Kostuik J P, Harrington I, Alexander D, et al.: Cauda equina syndrome and lumbar disc herniation. J Bone Joint Surg 68A: 386-91, 1986.

112. O’Laoire S A, Crockard H A, Thomas D G: Prognosis for sphincter recovery after operation for cauda equina compression owing to lumbar disc prolapse. Br Med J 282: 1852-4, 1981.

113. Shapiro S: Cauda equina syndrome secondary to lumbar disc herniation. Neurosurgery 32: 743-7, 1993.

114. Scott P J: Bladder paralysis in cauda equina lesions from disc prolapse. J Bone Joint Surg 47B: 224-35, 1965.

115. Tay E C K, Chacha P B: Midline prolapse of a lumbar intervertebral disc with compression of the cauda equina. J Bone Joint Surg 61B: 43-6, 1979.

116. Prusick V D, Lint D S, Bruder J: Cauda equina syndrome as a complication of free epidural fat-grafting. J Bone Joint Surg 70A: 1256-8, 1988.

117. Tandon P N, Sankaran B: Cauda equina syndrome due to lumbar disc prolapse. Indian J Orthopedics 1: 112-9, 1967.

118. Shapiro S: Medical realities of cauda equina syndrome secondary to lumbar disc herniation. Spine 25: 348-51, 2000.

119. Kostuik J P: Point of view: Comment on Shapiro, S: Medical realities of cauda equina syndrome secondary to lumbar disc herniation. Spine 25: 351, 2000.

120. Hoffman R M, Wheeler K J, Deyo R A: Surgery for herniated lumbar discs: A literature synthesis. J Gen Intern Med 8 (9): 487-96, 1993.

121. Williams R W: Microlumbar discectomy: A conservative surgical approach to the virgin herniated lumbar disc. Spine 3: 175-82, 1978.

122. Caspar W, Campbell B, Barbier D D, et al.: The caspar microsurgical discectomy and comparison with a conventional lumbar disc procedure. Neurosurgery 28: 78-87, 1991.

123. Schmidek H H, Sweet W H, (eds.): Operative neurosurgical techniques. 1st ed., Grune and Stratton, New York, 1982.

124. Fager C A: Lumbar discectomy: A contrary opinion. Clin Neurosurg 33: 419-56, 1986.

125. Tulberg T, Isacson J, Weidenhielm L: Does microscopic removal of lumbar disc herniation lead to better results than the standard procedure? Results of a one-year randomized study. J Neurosurg 70: 869-75, 1993.

126. Gogan W J, Fraser R D: Chymopapain. A 10-year, double-blind study. Spine 17: 388-94, 1992.

127. Javid M J, Nordby E J, Ford L T, et al.: Safety and efficacy of chymopapain (chymodiactin) in herniated nucleus pulposus with sciatica. Results of a randomized, double-blind study. JAMA 249: 2489-94, 1983.

128. Crawshaw C, Frazer A M, Merriam W F, et al.: A comparison of surgery and chemonucleolysis in the treatment of sciatica. A prospective randomized trial. Spine 9: 195-8, 1984.

129. van Alphen H A M, Braakman R, Bezemer P D, et al.: Chemonucleolysis versus discectomy: A randomized multicenter trial. J Neurosurg 70: 869-75, 1989.

130. Ejeskär A, Nachemson A, Herberts P, et al.: Surgery versus chemonucleolysis for herniated lumbar discs. A prospective study with random assignment. Clin Orthop 174: 236-42, 1983.

131. Bouillet R: A comparative survey of complications of surgical treatment and nucleolysis with chymopapain. Clin Orthop 251: 144-52, 1990.

132. Agre K, Wilson R R, Brim M, et al.: Chymodiactin postmarketing surveillance. Demographic and adverse experience data in 29,075 patients. Spine 9: 479-85, 1984.

133. Deeb Z L, Schimel S, Daffner R H, et al.: Intervertebral disk-space infection after chymopapain injection. AJNR 6: 55-8, 1985.

134. Eguro H: Transverse myelitis following chemonucleolysis: Report of a case. J Bone Joint Surg 65A: 1328-9, 1983.

135. Hoppenfield S: Percutaneous removal of herniated lumbar discs. 50 cases with ten-year follow-up periods. Clin Orthop 238: 92-7, 1989.

136. Kahanovitz N, Viola K, Goldstein T, et al.: A multi-center analysis of percutaneous discectomy. Spine 15: 713-5, 1990.

137. Davis G W, Onik G: Clinical experience with automated percutaneous lumbar discectomy. Clin Orthop 238: 98-103, 1989.

138. Revel M, Payan C, Vallee C, et al.: Automated percutaneous lumbar discectomy versus chemonucleolysis in the treatment of sciatica. Spine 18: 1-7, 1993.

139. Maroon J C, Onik G, Sternau L: Percutaneous automated discectomy. A new method for lumbar disc removal. Technical note. J Neurosurg 66: 143-6, 1987.

140. Onik G, Maroon J C, Jackson R: Cauda equina syndrome secondary to an improperly placed nucleotome probe. Neurosurgery 30: 412-5, 1992.

141. Cohen S P, Williams S, Kurihara C, et al.: Nucleoplasty with or without intradiscal electrothermal therapy (IDET) as a treatment for lumbar herniated disc. J Spinal Disord Tech 18 Suppl: S119-24, 2005.

142. Yonezawa T, Onomura T, Kosaka R, et al.: The system and procedures of percutaneous intradiscal laser nucleotomy. Spine 15: 1175-85, 1990.

143. Choy D S J, Ascher P W, Saddekni S, et al.: Percutaneous laser disc decompression: A new therapeutic modality. Spine 17: 949-56, 1992.

144. Mayer H M, Brock M: Percutaneous endoscopic discectomy: Surgical technique and preliminary results compared to microsurgical discectomy. J Neurosurg 78: 216-25, 1993.

145. Kleinpeter G, Markowitsch M M, Bock F: Percutaneous endoscopic lumbar discectomy: Minimally invasive, but perhaps only minimally useful? Surg Neurol 43: 534-41, 1995.

146. Karasek M, Bogduk N: Twelve-month follow-up of a controlled trial of intradiscal thermal anuloplasty for back pain due to internal disc disruption. Spine 25 (20): 2601-7, 2000.

147. Schwarzer A C, Aprill C N, Derby R, et al.: The prevalence and clinical features of internal disc disruption in patients with chronic low back pain. Spine 20 (17): 1878-83, 1995.

148. Lavyne M H, Bilsky M H: Epidural steroids, postoperative morbidity, and recovery in patients undergoing microsurgical lumbar discectomy. J Neurosurg 77: 90-5, 1992.

149. Glasser R S, Knego R S, Delashaw J B, et al.: The perioperative use of corticosteroids and bipuvicaine in the management of lumbar disc disease. J Neurosurg 78: 383-7, 1993.

150. Dunsker S B: Comment on Cobanoglu S, et al.: Complication of epidural fat graft in lumbar spine disc surgery: Case report. Surg Neurol 44: 481-2, 1995.

151. Cabezudo J M, Lopez A, Bacci F: Symptomatic root compression by a free fat transplant after hemilaminectomy: Case report. J Neurosurg 63: 633-5, 1985.

152. Cobanoglu S, Imer M, Ozylmaz F, et al.: Complication of epidural fat graft in lumbar spine disc surgery: Case report. Surg Neurol 44: 479-82, 1995.

153. Ramirez L F, Thisted R: Complications and demographic characteristics of patients undergoing lumbar discectomy in community hospitals. Neurosurgery 25: 226-31, 1989.

154. Deyo R A, Cherkin D C, Loeser J D, et al.: Morbidity and mortality in association with operations on the lumbar spine. The influence of age, diagnosis, and procedure. J Bone Joint Surg 74A: 536-43, 1992.

155. Shektman A, Granick M S, Solomon M P, et al.: Management of infected laminectomy wounds. Neurosurgery 35: 307-9, 1994.

156. Goodkin R, Laska L L: Unintended ‘incidental’ durotomy during surgery of the lumbar spine: Medicolegal implications. Surg Neurol 43: 4-14, 1995.

157. Davis R A: A long-term outcome analysis of 984 surgically treated herniated lumbar discs. J Neurosurg 80: 415-21, 1994.

158. Bilsky M H, Shields C B: Complications of lumbar disc surgery. Contemp Neurosurg 17 (21): 1-6, 1995.

159. DeSaussure R L: Vascular injuries coincident to disc surgery. J Neurosurg 16: 222-39, 1959.

160. Jhawar B S, Mitsis D, Duggal N: Wrong-sided and wrong-level neurosurgery: A national survey. J Neurosurg Spine 7 (5): 467-72, 2007.

161. Mclaren A C, Bailey S I: Cauda equina syndrome: A complication of lumbar discectomy. Clin Orthop 204: 143-9, 1986.

162. Porter R W, Detwiler P W, Lawton M T, et al.: Postoperative spinal epidural hematomas: Longitudinal review of 12,000 spinal operations. BNI Quarterly 16: 10-7, 2000.

163. Lee L A, Roth S, Posner K L, et al.: The American society of anesthesiologists postoperative visual loss registry: Analysis of 93 spine surgery cases with postoperative visual loss. Anesthesiology 105: 652-9, 2006.

164. Hayreh S S: Ischemic optic neuropathy. Prog Retin Eye Res 28 (1): Prog Retin Eye Res: 34-62, 2009.

165. Sachs B L, Zindrick M R, Beasley R D: Reflex sympathetic dystrophy after operative procedures on the lumbar spine. J Bone Joint Surg 75A: 721-5, 1993.

166. Feldman R A, Karl R C: Diagnosis and treatment of ogilvie’s syndrome after lumbar spinal surgery. J Neurosurg 76: 1012-6, 1992.

167. Hodges S D, Humphreys C, Eck J C, et al.: Management of incidental durotomy without mandatory bed rest. Spine 24: 2062-4, 1999.

168. Fink L H: Unintended ‘incidental’ durotomy. Surg Neurol 45: 590, 1996 (letter).

169. Ciappetta P, Delfini R, Cantore G P: Intradural lumbar disc hernia: Description of three cases. Neurosurgery 8: 104-7, 1981.

170. Horwitz N H, Rizzoli H V: Herniated intervertebral discs and spinal stenosis. In Postoperative complications of extracranial neurological surgery, Horwitz N H and Rizzoli H V, (eds.). Williams and Wilkins, Baltimore, 1987: pp 1-72.

171. Eismont F L, Wiesel S W, Rothman R H: Treatment of dural tears associated with spinal surgery. J Bone Joint Surg 63A: 1132-6, 1981.

172. Waisman M, Schweppe Y: Postoperative cerebrospinal fluid leakage after lumbar spine operations. Conservative treatment. Spine 15: 52-3, 1991.

173. Weinstein J N, Tosteson T D, Lurie J D, et al.: Surgical vs nonoperative treatment for lumbar disk herniation: The spine patient outcomes research trial (SPORT): A randomized trial. JAMA 296 (20): 2441-50, 2006.

174. Weinstein J N, Lurie J D, Tosteson T D, et al.: Surgical vs nonoperative treatment for lumbar disk herniation: The spine patient outcomes research trial (SPORT) observational cohort. JAMA 296 (20): 2451-9, 2006.

175. McCormick P C: The spine patient outcomes research trial results for lumbar disc herniation: A critical review. J Neurosurg Spine 6 (6): 513-20, 2007.

176. Aronson H A, Dunsmore R H: Herniated upper lumbar discs. J Bone Joint Surg 45: 311-7, 1963.

177. Abdullah A F, Wolber P G H, Warfield J R, et al.: Surgical management of extreme lateral lumbar disc herniations: Review of 138 cases. Neurosurgery 22: 648-53, 1988.

178. Godersky J C, Erickson D L, Seljeskog E L: Extreme lateral disc herniation: Diagnosis by CT scanning. Neurosurgery 14: 549-52, 1984.

179. Osborne D R, Heinz E R, Bullard D, et al.: Role of CT in the radiological evaluation of painful radiculopathy after negative myelography. Neurosurgery 14: 147-53, 1984.

180. Osborn A G, Hood R S, Sherry R G, et al.: CT/MR spectrum of far lateral and anterior lumbosacral disk herniations. AJNR 9: 775-8, 1988.

181. Jackson R P, Glah J J: Foraminal and extraforaminal lumbar disc herniation: Diagnosis and treatment. Spine 12: 577-85, 1987.

182. Angtuaco E J C, Holder J C, Boop W C, et al.: Computed tomographic discography in the evaluation of extreme lateral disc herniation. Neurosurgery 14: 350-2, 1984.

183. Grenier N, Greselle J-F, Douws C, et al.: MR imaging of foraminal and extraforaminal lumbar disk herniations. J Comput Assist Tomogr 14: 243-9, 1990.

184. Garrido E, Connaughton P N: Unilateral facetectomy approach for lateral lumbar disc herniation. J Neurosurg 74: 754-6, 1991.

185. Epstein N E, Epstein J A, Carras R, et al.: Far lateral lumbar disc herniations and associated structural abnormalities. An evaluation in 60 patients of the comparative value of CT, MRI, and myelo-CT in diagnosis and management. Spine 15: 534-9, 1990.

186. Jane J A, Haworth C S, Broaddus W C, et al.: A neurosurgical approach to far-lateral disc herniation. J Neurosurg 72: 143-4, 1990.

187. Ditsworth D A: Endoscopic transforaminal lumbar discectomy and reconfiguration: A posterolateral approach into the spinal canal. Surg Neurol 49: 588-98, 1998.

188. Maroon J C, Kopitnik T A, Schulhof L A, et al.: Diagnosis and microsurgical approach to far-lateral disc herniation in the lumbar spine. J Neurosurg 72: 378-82, 1990.

189. Ebersold M J, Quast L M, Bianco A J: Results of lumbar discectomy in the pediatric patient. J Neurosurg 67: 643-7, 1987.

190. Epstein J A, Epstein N E, Marc J, et al.: Lumbar intervertebral disk herniation in teenage children: Recognition and management of associated anomalies. Spine 9: 427-32, 1984.

191. Kataoka O, Nishibayashi Y, Sho T: Intradural lumbar disc herniation: Report of three cases with a review of the literature. Spine: 529-33, 1989.

192. Schisano G, Franco A, Nina P: Intraradicular and intradural lumbar disc herniation: Experience with nine cases. Surg Neurol 44: 536-43, 1995.

193. Deeg H J: Schmorl’s nodule. N Engl J Med 298: 57, 1978 (Letter).

194. Hamanishi C, Kawabata T, Yosii T, et al.: Schmorl’s nodes on magnetic resonance imaging. Spine 19: 450-3, 1994.

195. Takahashi K, Miyazaki T, Ohnari H, et al.: Schmorl’s nodes and low-back pain. Analysis of magnetic resonance imaging findings in symptomatic and asymptomatic individuals. Eur Spine J 4: 56-9, 1995.

196. Kao C C, Winkler S S, Turner J H: Synovial cyst of spinal facet. Case report. J Neurosurg 41: 372-6, 1974.

197. Freidberg S R, Fellows T, Thomas C B, et al.: Experience with symptomatic epidural cysts. Neurosurgery 34: 989-93, 1994.

198. Cartwright M J, Nehls D G, Carrion C A, et al.: Synovial cyst of a cervical facet joint: Case report. Neurosurgery 16: 850-2, 1985.

199. Onofrio B M, Mih A D: Synovial cysts of the spine. Neurosurgery 22: 642-7, 1988.

200. Goffin J, Wilms G, Plets C, et al.: Synovial cyst at the C1-C2 junction. Neurosurgery 30: 914-6, 1992.

201. Lopes N M M, Aesse F F, Lopes D K: Compression of thoracic nerve root by a facet joint synovial cyst: Case report. Surg Neurol 38: 338-40, 1992.

202. Heary R F, Stellar S, Fobben E S: Preoperative diagnosis of an extradural cyst arising from a spinal facet joint: Case report. Neurosurgery 30: 415-8, 1992.

203. Kao C C, Uihlein A, Bickel W H, et al.: Lumbar intraspinal extradural ganglion cyst. J Neurosurg 29: 168-72, 1968.

204. Franck J I, King R B, Petro G R, et al.: A posttraumatic lumbar spinal synovial cyst. Case report. J Neurosurg 66: 293-6, 1987.

205. Sabo R A, Tracy P T, Weinger J M: A series of 60 juxtafacet cysts: Clinical presentation, the role of spinal instability, and treatment. J Neurosurg 85: 560-5, 1996.

206. Mercader J, Gomez J M, Cardenal C: Intraspinal synovial cyst: Diagnosis by CT. Follow-up and spontaneous remission. Neuroradiology 27: 346-8, 1985.

207. Liu S S, Williams K D, Drayer B P, et al.: Synovial cysts of the lumbosacral spine: Diagnosis by MR imaging. AJNR 10: 1239-42, 1989.

208. Silbergleit R, Gebarski S S, Brunberg J A, et al.: Lumbar synovial cysts: Correlation of myelographic, CT, MR, and pathologic findings. AJNR 11: 777-9, 1990.

209. Gorey M T, Hyman R A, Black K S, et al.: Lumbar synovial cysts eroding bone. AJNR 13: 161-3, 1992.

210. Conrad M, Pitkethly D: Bilateral synovial cysts creating spinal stenosis. J Comput Assist Tomogr 11: 196-7, 1987.

211. Hemminghytt S, Daniels D L, Williams M L, et al.: Intraspinal synovial cysts: Natural history and diagnosis by CT. Radiology 145: 375-6, 1982.

212. Budris D M: Intraspinal lumbar synovial cyst. Orthopedics 14: 618-20, 1991.

213. Tarlov I M: Spinal perineurial and meningeal cysts. J Neurol Neurosurg Psychiatry 33: 833-43, 1970.

214. Soren A: Pathogenesis and treatment of ganglion. Clin Orthop 48: 173-9, 1966.

215. Schulz E E, West W L, Hinshaw D B, et al.: Gas in a lumbar extradural juxtaarticular cyst: Sign of synovial origin. Am J Radiol 143: 875-6, 1984.

216. Munz M, Tampieri D, Robitaille Y, et al.: Spinal synovial cyst: Case report using magnetic resonance imaging. Surg Neurol 34: 431-4, 1990.

217. Martin D, Awwad E, Sundaram M: Lumbar ganglion cyst causing radiculopathy. Orthopedics 13: 1182-3, 1990.

218. Kurz L T, Garfin S R, Unger A S, et al.: Intraspinal synovial cyst causing sciatica. J Bone Joint Surg 67A: 865-71, 1985.

219. Sehati N, Khoo L T, Holly L T: Treatment of lumbar synovial cysts using minimally invasive surgical techniques. Neurosurgical Focus 20 (3): E2-6, 2006.

220. Law J D, Lehman R A W, Kirsch W M, et al.: Re-operation after lumbar intervertebral disc surgery. J Neurosurg 48: 259-63, 1978.

221. Caputy A J, Luessenhop A J: Long-term evaluation of decompressive surgery for degenerative lumbar stenosis. J Neurosurg 77: 669-76, 1992.

222. Markwalder T M, Battaglia M: Failed back surgery syndrome. Part 1: Analysis of the clinical presentation and results of testing procedures for instability of the lumbar spine in 171 patients. Acta Neurochir 123: 46-51, 1993.

223. Burton C V, Kirkaldy-Willis W H, Yong-Hing K, et al.: Causes of failure of surgery on the lumbar spine. Clin Orthop 157: 191-9, 1981.

224. Quencer R M, Tenner M, Rothman L: The postoperative myelogram: Radiographic evaluation of arachnoiditis and dural/arachnoidal tears. Radiology 123: 667-9, 1977.

225. Roca J, Moreta D, Ubierna M T, et al.: The results of surgical treatment of lumbar arachnoiditis. Int Orthop 17: 77-81, 1993.

226. Ross J S, Masaryk T J, Modic M T, et al.: MR imaging of lumbar arachnoiditis. AJNR 8: 885-92, 1987.

227. Delamarter R B, Ross J S, Masaryk T J, et al.: Diagnosis of lumbar arachnoiditis by magnetic resonance imaging. Spine 15: 304-10, 1990.

228. Martin-Ferrer S: Failure of autologous fat grafts to prevent post operative epidural fibrosis in surgery of the lumbar spine. Neurosurgery 24: 718-21, 1989.

229. North R B, Campbell J N, James C S, et al.: Failed back surgery syndrome: 5-year follow-up in 102 patients undergoing repeated operations. Neurosurgery 28: 685-91, 1991.

230. Ross J S, Robertson J T, Frederickson R C A, et al.: Association between peridural scar and recurrent radicular pain after lumbar discectomy: Magnetic resonance evaluation. Neurosurgery 38: 855-63, 1996.

231. Cooper P R: Comment on Ross J S, et al.: Association between peridural scar and recurrent radicular pain after lumbar discectomy. Neurosurgery 38: 861, 1996.

232. Sonntag V K H: Comment on Ross J S, et al.: Association between peridural scar and recurrent radicular pain after lumbar discectomy. Neurosurgery 38: 862, 1996.

233. Bundschuh C V, Modic M T, Ross J S, et al.: Epidural fibrosis and recurrent disc herniation in the lumbar spine: Assessment with magnetic resonance. AJNR 9: 169-78, 1988.

234. Sotiropoulos S, Chafetz N E, Lang P, et al.: Differentiation between postoperative scar and recurrent disk herniation: Prospective comparison of MR, CT, and contrast-enhanced CT. AJNR 10: 639-43, 1989.

235. Hueftle M G, Modic M T, Ross J S, et al.: Lumbar spine: Postoperative MR imaging with Gd-DPTA. Radiology 167: 817-24, 1988.

236. Braun I F, Hoffman J C, Davis P C, et al.: Contrast enhancement in CT differentiation between recurrent disk herniation and postoperative scar: Prospective study. AJR 145: 785-90, 1985.

237. Byrd S E, Cohn M L, Biggers S L, et al.: The radiologic evaluation of the symptomatic postoperative lumbar spine patient. Spine 10: 652-61, 1985.

238. Greenwood J, McGuire T H, Kimbell F: A study of the causes of failure in the herniated intervertebral disc operation. An analysis of sixty-seven reoperated cases. J Neurosurg 9: 15-20, 1952.

239. Jorgensen J, Hansen P H, Steenskov V, et al.: A clinical and radiological study of chronic lower spinal arachnoiditis. Neuroradiology 9: 139-44, 1975.

240. Johnston J D H, Matheny J B: Microscopic lysis of lumbar adhesive arachnoiditis. Spine 3: 36-9, 1978.

241. Herron L: Recurrent lumbar disc herniation: Results of repeat laminectomy and discectomy. J Spinal Disord 7: 161-6, 1994.

242. Waddell G, Crummel E G, Solts W N, et al.: Failed lumbar disc surgery and repeat surgery following industrial injuries. J Bone Joint Surg 61A: 201-7, 1979.

243. Bell G K, Kidd D, North R B: Cost-effectiveness analysis of spinal cord stimulation in treatment of failed back surgery syndrome. J Pain Symptom Manage 13 (5): 286-95, 1997.

244. Mayfield F H: Cervical spondylosis: A comparison of the anterior and posterior approaches. Clin Neurosurg 13: 181-8, 1966.

245. Kobayashi N, Asamoto S, Doi H, et al.: Brown-Sequard syndrome produced by cervical disc herniation: Report of two cases and review of the literature. Spine J 3 (6): Spine J: 530-3, 2003.

246. Dai L, Jia L: Central cord injury complicating acute cervical disc herniation in trauma. Spine 25 (3): Spine: 331-336, 2000.

247. Viikari-Juntura E, Porras M, Laasonen E M: Validity of clinical tests in the diagnosis of root compression in cervical disc disease. Spine 14: 253-7, 1989.

248. Spurling R G, Scoville W B: Lateral rupture of the cervical intervertebral discs: A common cause of shoulder and arm pain. Surg Gynecol Obstet 78: 350-8, 1944.

249. Davidson R I, Dunn E J, Metzmaker J N: The shoulder abduction test in the diagnosis of radicular pain in cervical extradural compressive miniradiculopathies. Spine 6: 441-6, 1981.

250. Rubinstein S M, Pool J J, van Tulder M W, et al.: A systematic review of the diagnostic accuracy of provocative tests of the neck for diagnosing cervical radiculopathy. Eur Spine J 16 (3): Eur Spine J: 307-19, 2006.

251. Saal J, Saal Y, Yurth E: Nonoperative management of herniated cervical intervertebral disc with radiculopathy. Spine 21: 1877-83, 1996.

252. Joanes V: Cervical disc herniation presenting with acue myelopathy. Surg Neurol 54 (2): Surg Neurol: 198, 2000 (Readers’ Comments).

253. Matz P G, Ryken T C, Groff M W, et al.: Techniques for anterior cervical decompression for radiculopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 183-97, 2009.

254. Senter H J, Kortyna R, Kemp W R: Anterior cervical discectomy with hydroxylapatite fusion. Neurosurgery 25: 39-43, 1989.

255. Ryken T C, Heary R F, Matz P G, et al.: Techniques for cervical interbody grafting. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 203-20, 2009.

256. Tew J M, Mayfield F H: Complications of surgery of the anterior cervical spine. Clin Neurosurg 23: 424-34, 1976.

257. Taylor B A, Vaccaro A R, Albert T J: Complications of anterior and posterior surgical approaches in the treatment of cervical degenerative disc disease. Semin Spine Surg 11: 337-46, 1999.

258. Netterville J L, Koriwchak M J, Winkle M, et al.: Vocal fold paralysis following the anterior approach to the cervical spine. Ann Otol Rhinol Laryngol 105: 85-91, 1996.

259. Pfeifer B A, Freidberg S R, Jewell E R: Repair of injured vertebral artery in anterior cervical procedures. Spine 19: 1471-4, 1994.

260. Karim A, Knapp J, Nanda A: Internal jugular venous thrombosis as a complication after an elective anterior cervical discectomy: Case report. Neurosurgery 59 (3): E705; discussion E705, 2006.

261. Ascher E, Salles-Cunha S, Hingorani A: Morbidity and mortality associated with internal jugular vein thromboses. Vasc Endovascular Surg 39 (4): 335-9, 2005.

262. Sheikh M A, Topoulos A P, Deitcher S R: Isolated internal jugular vein thrombosis: Risk factors and natural history. Vasc Med 7 (3): 177-9, 2002.

263. Ascher E, Hingorani A, Mazzariol F, et al.: Clinical experience with superior vena caval Greenfield filters. J Endovasc Surg 6 (4): 365-9, 1999.

264. Tajima H, Murata S, Kumazaki T, et al.: Successful interventional treatment of acute internal jugular vein thrombosis. AJR Am J Roentgenol 182 (2): 467-9, 2004.

265. Krieger A J, Rosomoff H L: Sleep-induced apnea. Part 2: Respiratory failure after anterior spinal surgery. J Neurosurg 39: 181-5, 1974.

266. Truumees E, Herkowitz H N: Adjacent segment degeneration in the cervical spine: Incidence and management. Semin Spine Surg 11: 373-83, 1999.

267. Gore D R, Sepic S B: Anterior cervical fusion for degenerated or protruded discs. A review of one hundred and fifty-six patients. Spine 9: 667-71, 1984.

268. Hawkins R J, Bilco T, Bonutti P: Cervical spine and shoulder pain. Clin Orthop Rel Res 258: 142-6, 1990.

269. Krnacik M J, Heggeness M H: Severe angioedema causing airway obstruction after anterior cervical surgery. Spine 22 (18): 2188-90, 1997.

270. Harhangi B S, Menovsky T, Wurzer H A: Hemothorax as a complication after anterior cervical discectomy: Case report. Neurosurgery 56 (4): E871; discussion E871, 2005.

271. Winslow C P, Winslow T J, Wax M K: Dysphonia and dysphagia following the anterior approach to the cervical spine. Arch Otolaryngol Head Neck Surg 127 (1): Arch Otolaryngol Head Neck Surg: 51-5, 2001.

272. Bazaz R, Lee M J, Yoo J U: Incidence of dysphagia after anterior cervical spine surgery: A prospective study. Spine 27 (22): Spine: 2453-8, 2002.

273. Fogel G R, McDonnell M F: Surgical treatment of dysphagia after anterior cervical interbody fusion. Spine J 5 (2): Spine J: 140-4, 2005.

274. Kaiser M G, Mummaneni P V, Matz P G, et al.: Radiographic assessment of cervical subaxial fusion. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 221-7, 2009.

275. Phillips F M, Carlson G, Emery S E, et al.: Anterior cervical pseudarthrosis: Natural history and treatment. Spine 22: 1585-9, 1997.

276. Cannada L K, Scherping S C, Yoo J U, et al.: Pseudoarthrosis of the cervical spine: A comparison of radiographic diagnostic measures. Spine (Phila Pa 1976) 28 (1): Spine (Phila Pa 1976): 46-51, 2003.

277. DePalma A F, Cooke A J: Results of anterior inter-body fusion of the cervical spine. Clin Orthop 60: 169-85, 1968.

278. Puschak T J, Anderson P A: Pseudarthrosis after anterior fusion: Treatment options and results. Semin Spine Surg 11: 312-21, 1999.

279. Zdeblick T A, Hughes S S, Riew K D, et al.: Failed anterior cervical discectomy and arthrodesis. Analysis and treatment of thirty-five patients. J Bone Joint Surg 79 (4): J Bone Joint Surg: 523-32, 1997.

280. Kaiser M G, Mummaneni P V, Matz P G, et al.: Management of anterior cervical pseudarthrosis. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 228-37, 2009.

281. Yi S, Lee D Y, Kim D H, et al.: Cervical artificial disc replacement. Part 1: History, design, and overview of the cervical artificial disc. Neurosurgical Quarterly 18: Neurosurgical Quarterly, 2008.

282. Heary R F, Ryken T C, Matz P G, et al.: Cervical laminoforaminotomy for the treatment of cervical degenerative radiculopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 198-202, 2009.

283. Aldrich F: Posterolateral microdiscectomy for cervical monoradiculopathy caused by posterolateral soft cervical disc sequestration. J Neurosurg 72: 370-7, 1990.

284. Zeidman S M, Ducker T B: Posterior cervical laminoforaminotomy for radiculopathy: Review of 172 cases. Neurosurgery 33: 356-62, 1993.

285. Collias J C, Roberts M P: Posterior surgical approaches for cervical disc herniation and spondylotic myelopathy. In Operative neurosurgical techniques, Schmidek H H and Sweet W H, (eds.). W.B. Saunders, Philadelphia, 3rd ed., 1995, Vol. 2: pp 1805-16.

286. El-Kalliny M, Tew J M, van Loveren H, et al.: Surgical approaches to thoracic disk herniations. Acta Neurochir 111: 22-32, 1991.

287. Stillerman C B, Chen T C, Couldwell W T, et al.: Experience in the surgical management of 82 symptomatic herniated thoracic discs and review of the literature. J Neurosurg 88 (4): 623-33, 1998.

288. Dohn D F: Thoracic spinal cord decompression: Alternative surgical approaches and basis of choice. Clin Neurosurg 27: 611-23, 1980.

289. Le Roux P D, Haglund M M, Harris A B: Thoracic disc disease: Experience with the transpedicular approach in twenty consecutive patients. Neurosurgery 33: 58-66, 1993.

290. Arce A C, Dohrmann G J: Thoracic disc herniation. Surg Neurol 23: 356-61, 1985 (review).

291. Ahlgren B D, Herkowitz H N: A modified postero-lateral approach to the thoracic spine. J Spinal Disord 8: 69-75, 1195.

292. O’Leary S T, Ganju A, Rauzzino M J, et al.: Costotransversectomy. In Atlas of neurosurgical techniques, Fessler R G and Sekhar L, (eds.). Thieme Medical Publishers, Inc., New York, 2006: pp 441-7.

293. Chou S N, Seljeskog E L: Alternative surgical approaches to the thoracic spine. Clin Neurosurg 20: 306-21, 1972.

294. Perot P L, Munro D D: Transthoracic removal of midline thoracic disc protrusions causing spinal cord compression. J Neurosurg 31: 452-61, 1969.

295. Hadley M N, Reddy S V: Smoking and the human vertebral column: A review of the impact of cigarette use on vertebral bone metabolism and spinal fusion. Neurosurgery 41: 116-24, 1997.

296. Fogelholm R R, Alho A V: Smoking and intervertebral disc degeneration. Med Hypotheses 56 (4): 537-9, 2001.

297. Meyerding H W: Spondylolisthesis. Surg Gynecol Obstet 54: 371-7, 1932.

298. Rothman R H, Simeone F A, (eds.): The spine. 2nd ed., W.B. Saunders, Philadelphia, 1982.

299. Sairyo K, Katoh S, Takata Y, et al.: MRI signal changes of the pedicle as an indicator for early diagnosis of spondylolysis in children and adolescents: A clinical and biomechanical study. Spine 31 (2): 206-11, 2006.

300. Weinstein J N, Lurie J D, Tosteson T D, et al.: Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 356 (22): N Engl J Med: 2257-70, 2007.

301. Herkowitz H N, Kurz L T: Degenerative lumbar spondylolisthesis with spinal stenosis: A prospective study comparing decompression with decompression and intertransverse process arthrodesis. J Bone Joint Surg 73A: 802-8, 1991.

302. Rombold C: Teatment of spondylolisthesis by posterolateral fusion, resection of the pars interarticularis, and prompt mobilization of the patient: An end-result study of seventy-three patients. J Bone Joint Surg 48A: 1282-300, 1966.

303. Dickman C A, Fessler R G, MacMillan M, et al.: Transpedicular screw-rod fixation of the lumbar spine: Operative technique and outcome in 104 cases. J Neurosurg 77: 860-70, 1992.

304. Epstein N E, Epstein J A, Carras R, et al.: Coexisting cervical and lumbar spinal stenosis: Diagnosis and management. Neurosurgery 15: 489-96, 1984.

305. Yamamoto I, Matsumae M, Ikeda A, et al.: Thoracic spinal stenosis: Experience with seven cases. J Neurosurg 68: 37-40, 1988.

306. Epstein N E: Symptomatic lumbar spinal stenosis. Surg Neurol 50: 3-10, 1998.

307. Verbiest H: A radicular syndrome from developmental narrowing of the lumbar canal. J Bone Joint Surg 36B: 230-7, 1954.

308. Epstein J A, Epstein B S, Lavine L: Nerve root compression associated with narrowing of the lumbar spinal canal. J Neurol Neurosurg Psychiatry 52: 165-76, 1962.

309. Duggal N, Sonntag V K H, Dickman C A: Fusion options and indications in the lumbosacral spine. Contemp Neurosurg 23 (1): 1-8, 2001.

310. Hawkes C H, Roberts G M: Neurogenic and vascular claudication. J Neurol Sci 38: 337-45, 1978.

311. Turner J A, Ersek M, Herron L, et al.: Surgery for lumbar spinal stenosis: Attempted meta-analysis of the literature. Spine 17: 1-8, 1992.

312. Kota G K, Kumar N K S, Thomas R: Baastrups disease an unusual cause of backpain: A case report. 2005. URL http://www.ispub.com/ostia/index.php?xmlFilePath=journals/ijra/vol4n1/baastrups.xml. Date accessed: 1.

313. Shbeeb M I, Matteson E L: Trochanteric bursitis (greater trochanter pain syndrome). Mayo Clin Proc 71: 565-9, 1996.

314. Deen H G: Diagnosis and management of lumbar disk disease. Mayo Clin Proc 71: 283-7, 1996.

315. Xu R, Sciubba D M, Gokaslan Z L, et al.: Ossification of the ligamentum flavum in a caucasian man. J Neurosurg Spine 9 (5): 427-37, 2008.

316. Miyazawa N, Akiyama I: Ossification of the ligamentum flavum of the cervical spine. J Neurosurg Sci 51 (3): 139-44, 2007.

317. Ehni G: Significance of the small lumbar spinal canal. J Neurosurg 31: 490-4, 1969.

318. Ullrich C G, Binet E F, Sanecki M G, et al.: Quantitative assessment of the lumbar spinal canal by CT. Radiology 134: 137-43, 1980.

319. Post M J D, (ed.) Computed tomography of the spine. Williams and Wilkins, Baltimore, 1984.

320. Hinck V C, Clark W M, Hopkins C E: Normal interpediculate distances (minimum and maximum) in children and adult. AJR 97: 141-53, 1966.

321. Johnsson K E, Rosén I, Udén A: The natural course of lumbar spinal stenosis. Acta Orthop Scand 61 (Suppl): 24, 1990.

322. Aryanpur J, Ducker T: Multilevel lumbar laminotomies for focal spinal stenosis: Case report. Neurosurgery 23: 111-5, 1988.

323. Aryanpur J, Ducker T: Multilevel lumbar laminotomies: An alternative to laminectomy in the treatment of lumbar stenosis. Neurosurgery 26: 429-33, 1990.

324. Young S, Veeraoen R, O’Laoire S A: Relief of lumbar canal stenosis using multilevel subarticular fenestrations as an alternative to wide laminectomy: Preliminary report. Neurosurgery 23: 628-33, 1988.

325. Wang J C, Haid R W, Jr., Miller J S, et al.: Comparison of CD HORIZON SPIRE spinous process plate stabilization and pedicle screw fixation after anterior lumbar interbody fusion. J Neurosurg Spine4 (2): J Neurosurg Spine: 132-6, 2006.

326. Wang J C, Spenciner D, Robinson J C: SPIRE spinous process stabilization plate: Biomechanical evaluation of a novel technology. J Neurosurg Spine 4 (2): J Neurosurg Spine: 160-4, 2006.

327. Kondrashov D. 2007,

328. Tuite G F, Doran S E, Stern J D, et al.: Outcome after laminectomy for lumbar spinal stenosis. Part II: Radiographic changes and clinical correlations. J Neurosurg 81: 707-15, 1994.

329. Silvers H R, Lewis P J, L A H: Decompressive lumbar laminectomy for spinal stenosis. J Neurosurg 78: 695-701, 1993.

330. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 10: Fusion following decompression in patients with stenosis without spondylolisthesis. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 686-91, 2005.

331. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 9: Fusion in patients with stenosis and spondylolisthesis. J Neurosurg Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 679-85, 2005.

332. Resnick D K, Choudhri T F, Dailey A T, et al.: Part 14: Brace therapy as an adjunct to or substitute for lumbar fusion. Journal of Neurosurgery: Spine 2 (6): Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine: 716-24, 2005.

333. Deburge A, Lassale B, Benoist M, et al.: Le traitment chirurgical des stenosis lombaires et ses resultats a propos d’une serie de 163 cas operes. Rev Rheum Mal Osteoartic 50: 47-54, 1983.

334. Reuben S S, Ablett D, Kaye R: High dose nonsteroidal anti-inflammatory drugs compromise spinal fusion. Can J Anaesth 52 (5): 506-12, 2005.

335. Thaller J, Walker M, Kline A J, et al.: The effect of nonsteroidal anti-inflammatory agents on spinal fusion. Orthopedics 28 (3): 299-303; quiz 304-5, 2005.

336. Ganz J C: Lumbar spinal stenosis: Postoperative results in terms of preoperative posture-related pain. J Neurosurg 72: 71-4, 1990.

337. Katz J N, Lipson S J, Larson M G, et al.: The outcome of decompressive laminectomy for degenerative lumbar stenosis. J Bone Joint Surg 73A: 809-16, 1991.

338. Deen H G, Zimmerman R S, Lyons M K, et al.: Analysis of early failures after lumbar decompressive laminectomy for spinal stenosis. Mayo Clin Proc 70: 33-6, 1995.

339. Tuite G F, Stern J D, Doran S E, et al.: Outcome after laminectomy for lumbar spinal stenosis. Part I: Clinical correlations. J Neurosurg 81: 699-706, 1994.

340. Javid M J, Hadar E J: Long-term follow-up review of patients who underwent laminectomy for lumbar stenosis: A prospective study. J Neurosurg 89: 1-7, 1998.

341. Ciric I, Mikhael M A, Tarkington J A, et al.: The lateral recess syndrome. J Neurosurg 53: 433-43, 1980.

342. Miller C A: Shallow cervical canal: Recognition, clinical symptoms, and treatment. Contemp Neurosurg 7: 1-5, 1985.

343. Muhle C, Weinert D, Falliner A, et al.: Dynamic changes of the spinal canal in patients with cervical spondyolosis at flexion and extension using mangnetic resonance imaging. Invest Radiol 33: 444-9, 1998.

344. Shedid D, Benzel E C: Cervical spondylosis anatomy: Pathophysiology and biomechanics. Neurosurgery 60: Diagnosis and treatment of cervical spondylosis: S1-1-11, 2007.

345. Nagashima C: Cervical myelopathy due to ossification of the posterior longitudinal ligament. J Neurosurg 37: 653-60, 1972.

346. Batzdorf U, Batzdorf A: Analysis of cervical spine curvature in patients with cervical spondylosis. Neurosurgery 22: 827-36, 1988.

347. Cusick J F: Pathophysiology and treatment of cervical spondylotic myelopathy. Clin Neurosurg 37: 661-81, 1989.

348. Cooper P R: Cervical spondylotic myelopathy. Contemp Neurosurg 19 (25): 1-7, 1997.

349. Yu Y L, du Boulay G H, Stevens J M, et al.: Computed tomography in cervical spondylotic myelopathy and radiculopathy: Visualization of structures, myelographic comparison, cord measurements and clinical utility. Neuroradiology 28: 221-36, 1986.

350. Epstein J A, Marc J A, Hyman R A, et al.: Total myelography in the evaluation of lumbar disks. Spine 4: 121-8, 1979.

351. Houser O W, Onofrio B M, Miller G M, et al.: Cervical spondylotic stenosis and myelopathy: Evaluation with computed tomographic myelography. Mayo Clin Proc 69: 557-63, 1994.

352. Taylor A R: Vascular factors in the myelopathy associated with cervical spondylosis. Neurology 14: 62-8, 1964.

353. Bohlman H H, Emery J L: The pathophysiology of cervical spondylosis and myelopathy. Spine 13: 843-6, 1988.

354. Kim R C: Spinal cord pathology. In Principals and practice of neuropathology, Nelson J S, Parisi J E, and Schochet S S, (eds.). C V Mosby, St. Louis, 1993: pp 398-435.

355. Jeffreys R V: The surgical treatment of cervical myelopathy due to spondylosis and disc degeneration. J Neurol Neurosurg Psychiatry 49: 353-61, 1986.

356. Adams C B T, Logue V: Studies in cervical spondylotic myelopathy: I. Movement of the cervical roots, dura and cord, and their relation to the course of the extrathecal roots. Brain 94: 557-68, 1971.

357. Levine D N: Pathogenesis of cervical spondylotic myelopathy. J Neurol Neurosurg Psychiatry 62: 334-40, 1997.

358. Benzel E C: Biomechanics of spine stabilization. American Association of Neurological Surgeons Publications, Rolling Meadows, IL, 2001.

359. Ogino H, Tada K, Okada K, et al.: Canal diameter, anteroposterior compression ratio, and spondylotic myelopathy of the cervical spine. Spine 8: 1-15, 1983.

360. Lunsford L D, Bissonette D J, Zorub D S: Anterior surgery for cervical disc disease. Part 2: Treatment of cervical spondylotic myelopathy in 32 cases. J Neurosurg 53: 12-9, 1980.

361. Emery S E: Cervical spondylotic myelopathy: Diagnosis and treatment. J Am Acad Orthop Surg 9: 376-88, 2001.

362. Chiles B W, A L M, Choudhri H F, et al.: Cervical spondylotic myelopathy: Patterns of neurological deficit and recovery after anterior cervical decompression. Neurosurgery 44: 762-9, 1999.

363. Nurick S: The pathogenesis of the spinal cord disorder associated with cervical spondylosis. Brain 95: 87-100, 1972.

364. Clarke E, Robinson P K: Cervical myelopathy: A complication of cervical spondylosis. Brain 79: 483-5, 1956.

365. Lees F, Aldren Turner J S: Natural history and prognosis of cervical spondylosis. Br Med J 2: 1607-10, 1963.

366. Nurick S: The natural history and the results of surgical treatment of the spinal cord disorder associated with cervical spondylosis. Brain 95: 101-8, 1972.

367. Matz P G, Anderson P A, Holly L T, et al.: The natural history of cervical spondylotic myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 104-11, 2009.

368. Ebara S, Yonenobu K, Fujiwara K, et al.: Myelopathy hand characterized by muscle wasting: A different type of myelopathy hand in patients with cervical spondylosis. Spine 13: 785-91, 1988.

369. Crandall P H, Batzdorf U: Cervical spondylotic myelopathy. J Neurosurg 25: 57-66, 1966.

370. Voskuhl R R, Hinton R C: Sensory impairment in the hands secondary to spondylotic compression of the cervical spinal cord. Arch Neurol 47: 309-11, 1990.

371. MacFadyen D J: Posterior column dysfunction in cervical spondylotic myelopathy. Can J Neurol Sci 11: 365-70, 1984.

372. Denno J J, Meadows G R: Early diagnosis of cervical spondylotic myelopathy. A useful clinical sign. Spine 16 (12): 1353-5, 1991.

373. Sung R D, Wang J C: Correlation between a positive hoffmann’s reflex and cervical pathology in asymptomatic individuals. Spine 26: 67-70, 2001.

374. Wiggins G C, Shaffrey C I: Laminectomy in the cervical spine: Indications, surgical technniques, and avoidance of complications. Contemp Neurosurg 21 (20): 1-10, 1999.

375. England J D, Hsu C Y, Vera C L, et al.: Spondylotic high cervical spinal cord compression presenting with hand complaints. Surg Neurol 25: 299-303, 1986.

376. Good D C, Couch J R, Wacasser L: “numb, clumsy hands” and high cervical spondylosis. Surg Neurol 22: 285-91, 1984.

377. Ungar-Sargon J Y, Lovelace R E, Brust J C: Spastic paraplegia-paraparesis: A reappraisal. J Neurol Sci 46: 1-12, 1980.

378. Yoshor D, Klugh A, 3rd, Appel S H, et al.: Incidence and characteristics of spinal decompression surgery after the onset of symptoms of amyotrophic lateral sclerosis. Neurosurgery 57 (5): 984-9; discussion 984-9, 2005.

379. Campbell A M G, Phillips D G: Cervical disk lesions with neurological disorder. Differential diagnosis, treatment, and prognosis. Br Med J 2: 481-5, 1960.

380. Rowland L P: Diagnosis of amyotrophic lateral sclerosis. J Neurol Sci 160: S6-24, 1998.

381. Adams C B T, Logue V: Studies in cervical spondylotic myelopathy: II. The movement and contour of the spine in relation to the neural complications of cervical spondylosis. Brain 94: 569-86, 1971.

382. Wolf B S, Khilnani M, Malis L: The sagittal diameter of the bony cervical spinal canal and its significance in cervical spondylosis. J of Mount Sinai Hospital 23: 283-92, 1956.

383. Krauss W E, Ebersold M J, Quast L M: Cervical spondylotic myelopathy: Surgical indications and technique. Contemp Neurosurg 20 (10): 1-6, 1998.

384. Pavlov H, Torg J S, Robie B, et al.: Cervical spinal stenosis: Determination with vertebral body ratio method. Radiology 164: 771-5, 1987.

385. Torg J S, Naranja R J, Pavlov H, et al.: The relationship of developmental narrowing of the cervical spinal canal to reversible and irreversible injury of the cervical spinal cord in football players. J Bone Joint Surg 78A: 1308-14, 1996.

386. Brown B M, Schwartz R H, Frank E, et al.: Preoperative evaluation of cervical radiculopathy and myelopathy by surface-coil MR imaging. AJNR 9: 859-66, 1988.

387. Mummaneni P V, Kaiser M G, Matz P G, et al.: Pre-operative patient selection with magnetic resonance imaging, computed tomography, and electroencephalography: Does the test predict outcome after cervical surgery? Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 119-29, 2009.

388. Teresi L M, Lufkin R B, Reicher M A, et al.: Asymptomatic degenerative disk disease and spondylosis of the cervical spine: MR imaging. Radiology 164: 83-8, 1987.

389. Mizuno J, Nakagawa H, Inoue T, et al.: Clinico-pathological study of “snake-eye appearance” in compressive myelopathy of the cervical spinal cord. J Neurosurg 99 (2 Suppl): J Neurosurg: 162-8, 2003.

390. Holly L T, Matz P G, Anderson P A, et al.: Clinical prognostic indicators of surgical outcome in cervical spondylotic myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 112-8, 2009.

391. Matz P G, Holly L T, Mummaneni P V, et al.: Anterior cervical surgery for the treatment of cervical degenerative myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 170-3, 2009.

392. Matz P G, Holly L T, Groff M W, et al.: Indications for anterior cervical decompression for the treatment of cervical degenerative radiculopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 174-82, 2009.

393. Kadanka Z, Bednarik J, Vohanka S, et al.: Conservative tratment versus surgery in spondylotic cervical myelopathy treated conservatively or surgically. Eur Spine J 9: Eur Spine J: 538-44, 2000.

394. Resnick D K, Anderson P A, Kaiser M G, et al.: Electrophysiological monitoring during surgery for cervical degenerative myelopathy and radiculopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 245-52, 2009.

395. Mummaneni P V, Kaiser M G, Matz P G, et al.: Cervical surgical techniques for the treatment of cervical spondylotic myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 130-41, 2009.

396. Anderson P A, Matz P G, Groff M W, et al.: Laminectomy and fusion for the treatment of cervical degenerative myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 150-6, 2009.

397. Matz P G, Anderson P A, Groff M W, et al.: Cervical laminoplasty for the treatment of cervical degenerative myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 157-69, 2009.

398. Hamanishi C, Tanaka S: Bilateral multilevel laminectomy with or without posterolateral fusion for cervical spondylotic myelopathy: Relationship to type of onset and time until operation. J Neurosurg85 (3): J Neurosurg: 447-51, 1996.

399. Matsunaga S, Sakou T, Nakanisi K: Analysis of the cervical spine alignment following laminoplasty and laminectomy. Spinal Cord 37 (1): Spinal Cord: 20-4, 1999.

400. Ryken T C, Heary R F, Matz P G, et al.: Cervical laminectomy for the treatment of cervical degenerative myelopathy. Journal of Neurosurgery: Spine 11 (2): Journal of Neurosurgery: Spine: 142-9, 2009.

401. Guigui P, Benoist M, Deburge A: Spinal deformity and instability after multilevel cervical laminectomy for spondylotic myelopathy. Spine 23 (4): Spine: 440-7, 1998.

402. Kaptain G J, Simmons N E, Replogle R E, et al.: Incidence and outcome of kyphotic deformity following laminectomy for cervical spondylotic myelopathy. J Neurosurg 93 (2 Suppl): J Neurosurg: 199-204, 2000.

403. Benzel E C, Lancon J, Kesterson L, et al.: Cervical laminectomy and dentate ligament section for cervical spondylotic myelopathy. J Spinal Disord 4 (3): J Spinal Disord: 286-95, 1991.

404. Yonenobu K, Hosono N, Iwasaki M, et al.: Neurologic complications of surgery for cervical compression myelopathy. Spine 16: 1277-82, 1991.

405. Epstein N E, Danto J, Nardi D: Evaluation of intraoperative somatosensory-evoked potential monitoring during 100 cervical operations. Spine 18: 737-47, 1993.

406. Epstein J, Janin Y, Carras R, et al.: A comparative study of the treatment of cervical spondylotic myeloradiculopathy: Experience with 50 cases treated by means of extensive laminectomy, foraminotomy, and excision of osteophytes during the past 10 years. Acta Neurochir 61: 89, 1982.

407. Epstein N E, Epstein J A: Operative management of cervical spondylotic myelopathy: Technique and result of laminectomy. In The cervical spine, The Cervical Spine Research Society Editorial Committee, (ed.). Lippincott-Raven, Philadelphia, 3rd ed., 1998: pp 839-48.

408. Gregorius F K, Estrin T, Crandall P H: Cervical spondylotic radiculopathy and myelopathy. A long-term follow-up study. Arch Neurol 33: 618-25, 1976.

409. Sakaura H, Hosono N, Mukai Y, et al.: C5 palsy after decompression surgery for cervical myelopathy: Review of the literature. Spine 28 (21): 2447-51, 2003.

410. Komagata M, Nishiyama M, Endo K, et al.: Prophylaxis of c5 palsy after cervical expansive laminoplasty by bilateral partial foraminotomy. Spine J 4 (6): 650-5, 2004.

411. Ebersold M J, Pare M C, Quast L M: Surgical treatment for cervical spondylitic myelopathy. J Neurosurg 82: 745-51, 1995.

412. Hilibrand A S, Carlson G D, Palumbo M A, et al.: Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 81 (4): 519-28, 1999.

413. Dagi T F, Tarkington M A, Leech J J: Tandem lumbar and cervical spinal stenosis. J Neurosurg 66: 842-9, 1987.

414. Sinh G: Congenital atlanto-axial dislocation. Neurosurg Rev 6: 211-20, 1983.

415. Jain V K, Mittal P, Banerji D, et al.: Posterior occipitoaxial fusion for atlantoaxial dislocation associated with occipitalized axis. J Neurosurg 84: 559-64, 1996.

416. Ranawat C S, O’Leary P, Pellicci P, et al.: Cervical spine fusion in rheumatoid arthritis. J Bone Joint Surg 61A: 1003-10, 1979.

417. Menezes A H, VanGilder J C, Clark C R, et al.: Odontoid upward migration in rheumatoid arthritis. J Neurosurg 63: 500-9, 1985.

418. Rana N A, Hancock D O, Taylor A R: Atlanto-axial subluxation in rheumatoid arthritis. J Bone Joint Surg 55B: 458-70, 1973.

419. Hildebrandt G, Agnoli A L, Zierski J: Atlanto-axial dislocation in rheumatoid arthritis: Diagnostic and therapeutic aspects. Acta Neurochir 84: 110-7, 1987.

420. Hinck V C, Hopkins C E: Measurement of the atlanto-dental interval in the adult. Am J Roentgenol Radium Ther Nucl Med 84: 945-51, 1960.

421. Meijers K A E, van Beusekom G T, Luyendijk W, et al.: Dislocation of the cervical spine with cord compression in rheumatoid arthritis. J Bone Joint Surg 56B: 668-80, 1974.

422. Boden S D, Dodge L D, Bohlman H H, et al.: Rheumatoid arthritis of the cervical spine. A long-term analysis with predictors of paralysis and recovery. J Bone Joint Surg 75 (9): J Bone Joint Surg: 1282-97, 1993.

423. Collins D N, Barnes C L, Fitz-Randolph R L: Cervical spine instability in rheumatoid patients having total hip or knee arthroplasty. Clin Orthop Relat Res (272): Clin Orthop Relat Res: 127-35, 1991.

424. Smith P H, Benn R T, Sharp J: Natural history of rheumatoid cervical luxations. Ann Rheum Dis 31: 431-9, 1972.

425. Weissman B N W, Aliabadi P, Weinfeld M S, et al.: Prognostic features of atlantoaxial subluxation in rheumatoid arthritis patients. Radiology 144: 745-51, 1982.

426. Papadopoulos S M, Dickman C A, Sonntag V K H: Atlantoaxial stabilization in rheumatoid arthritis. J Neurosurg 74: 1-7, 1991.

427. Mikulowski P, Wollheim F A, Rotmil P, et al.: Sudden death in rheumatoid arthritis with atlanto-axial dislocation. Acta Med Scand 198 (6): Acta Med Scand: 445-51, 1975.

428. Kourtopoulos H, von E C: Stabilization of the unstable upper cervical spine in rheumatoid arthritis. Acta Neurochir 91: 113-5, 1988.

429. Clark C R, Goetz D D, Menezes A H: Arthrodesis of the cervical spine in rheumatoid arthritis. J Bone Joint Surg 71A: 381-92, 1989.

430. Riew K D, Hilibrand A S, Palumbo M A, et al.: Diagnosing basilar invagination in the rheumatoid patient. The reliability of radiographic criteria. J Bone Joint Surg 83-A (2): J Bone Joint Surg: 194-200, 2001.

431. Redlund-Johnell I, Pettersson H: Radiographic measurements of the cranio-vertebral region. Designed for evaluation of abnormalities in rheumatoid arthritis. Acta Radiol Diagn (Stockh) 25 (1): Acta Radiol Diagn (Stockh): 23-8, 1984.

432. Bundschuh C, Modic M T, Kearney F, et al.: Rheumatoid arthritis of the cervical spine: Surface-coil MR imaging. AJR Am J Roentgenol 151 (1): AJR Am J Roentgenol: 181-7, 1988.

433. Kim D H, Hilibrand A S: Rheumatoid arthritis in the cervical spine. J Am Acad Orthop Surg 13 (7): J Am Acad Orthop Surg: 463-74, 2005.

434. Martel W, Tishler J M: Observations on the spine in mongoloidism. Am J Roentgenol Radium Ther Nucl Med 97 (3): Am J Roentgenol Radium Ther Nucl Med: 630-8, 1966.

435. Pueschel S M: Should children with Down syndrome be screened for atlantoaxial instability? Arch Pediatr Adolesc Med 152 (2): Arch Pediatr Adolesc Med: 123-5, 1998.

436. American Academy of Pediatrics Committee on Sports Medicine and Fitness: Atlantoaxial instability in Down syndrome: Subject review. Pediatrics 96 (1 Pt 1): Pediatrics: 151-4, 1995.

437. Cohen W I: Atlantoaxial instability. What’s next? Arch Pediatr Adolesc Med 152 (2): Arch Pediatr Adolesc Med: 119-22, 1998.

438. Brockmeyer D: Down syndrome and craniovertebral instability. Topic review and treatment recommendations. Pediatr Neurosurg 31 (2): Pediatr Neurosurg: 71-7, 1999.

439. Walpin L A, Singer F R: Paget’s disease: Reversal of severe paraparesis using calcitonin. Spine 4: 213-9, 1979.

440. Youmans J R, (ed.) Neurological surgery. 3rd ed., W. B. Saunders, Philadelphia, 1990.

441. Delmas P D, Meunier P J: The management of Paget’s disease of bone. N Engl J Med 336: 558-66, 1997.

442. Meunier P J, Salson C, Mathieu L, et al.: Skeletal distribution and biochemical parameters of Paget’s disease. Clin Orthop 217: 37-44, 1987.

443. Altman R D, Brown M, Gargano F: Low back pain in Paget’s disease of bone. Clin Orthop 217: 152-61, 1987.

444. Hadjipavlou A, Shaffer N, Lander P, et al.: Pagetic spinal stenosis with extradural pagetoid ossification. Spine 13: 128-30, 1988.

445. Douglas D L, Duckworth T, Kanis J A, et al.: Spinal cord dysfunction in Paget’s disease of bone: Has medical treatment a vascular basis? J Bone Joint Surg 63B: 495-503, 1981.

446. Wilkins R H, Rengachary S S, (eds.): Neurosurgery. McGraw-Hill, New York, 1985.

447. Dinneen S F, Buckley T F: Spinal nerve root compression due to monostotic Paget’s disease of a lumbar vertebra. Spine 12: 948-50, 1987.

448. Sadar E S, Walton R J, Gossman H H: Neurological dysfunction in Paget’s disease of the vertebral column. J Neurosurg 37: 661-5, 1972.

449. Chen J-R, Rhee R S C, Wallach S, et al.: Neurologic disturbances in Paget disease of bone: Response to calcitonin. Neurology 29: 448-57, 1979.

450. Human calcitonin for Paget’s disease. Med Letter 29: 47-8, 1987.

451. Tiludronate for Paget’s disease of bone. Med Letter 39: 65-6, 1997.

452. Risedronate for Paget’s disease of bone. Med Letter 40: 87-8, 1998.

453. Bennett G J: Ankylosing spondylitis. Clin Neurosurg 37: 622-35, 1991.

454. Vender J R, McDonnell D E: Ankylosing spondylitis. Contemp Neurosurg 22 (25): 1-7, 2000.

455. Calin A, Porta J, Fries J F, et al.: Clinical history as a screening test for ankylosing spondylitis. JAMA 237: 2613-4, 1977.

456. Rae P S, Waddell G, Venner R M: A simple technique for measuring lumbar spinal flexion. J R Coll Surg Edin 29: 281-4, 1984.

457. Detwiler K N, Loftus C M, Godersky J C: Management of cervical spine injuries in patients with anky-losing spondylitis. J Neurosurg 72: 210-5, 1990.

458. Farhat S M, Schneider R C, Gray J M: Traumatic spinal epidural hematoma associated with cervical fractures in rheumatoid spondylitis. J Trauma 13: 591-9, 1973.

459. Podolsky S M, Hoffman J R, Pietrafesa C A: Neurological complications following immobilization of cervical spine fractures in a patient with ankylosing spondylitis. Ann Emerg Med 12: 578-80, 1983.

460. Surin V V: Fractures of the cervical spine in patients with ankylosing spondylitis. Acta Orthop Scand 51: 79-84, 1980.

461. Tsuyama N: Ossification of the posterior longitudinal ligament of the spine. Clin Orthop 184: 71-84, 1984.

462. Nakanishi T, Mannen T, Toyokura Y: Asymptomatic ossification of the posterior longitudinal ligament of the cervical spine. J Neurol Sci 19: 375-81, 1973.

463. Epstein N: Diagnosis and surgical management of ossification of the posterior longitudinal ligament. Contemp Neurosurg 14 (22): 1-6, 1992.

464. Harsh G R, Sypert G W, Weinstein P R, et al.: Cervical spine stenosis secondary to ossification of the posterior longitudinal ligament. J Neurosurg 67: 349-57, 1987.

465. Hirabayashi K, Watanabe K, Wakano K, et al.: Expansive cervical laminoplasty for cervical spinal stenotic myelopathy. Spine 8: 693-, 1983.

466. Matsunaga S, Sakou T, Taketomi E, et al.: Clinical course of patients with ossification of the posterior longitudinal ligament: A minimum 10-year cohort study. J Neurosurg 100: 245-8, 2004.

467. Epstein N E, Hollingsworth R: Ossification of the cervical anterior longitudinal ligament contributing to dysphagia: Case report. J Neurosurg 90 (Spine 2): 261-3, 1999.

468. Burkus J K: Esophageal obstruction secondary to diffuse idiopathic skeletal hyperostosis. Orthopedics 11: 717-20, 1988.

469. McCafferty R R, Harrison M J, Tamas L B, et al.: Ossification of the anterior longitudinal ligament and Forestier’s disease: An analysis of seven cases. J Neurosurg 83: 13-7, 1995.

470. Youmans J R, (ed.) Neurological surgery. 2nd ed., W. B. Saunders, Philadelphia, 1982.

471. Black P: Spinal vascular malformations: An historical perspective. Neurosurg Focus 21 (6): E11, 2006.

472. Di Chiro G, Doppman J, Ommaya A K: Selective arteriography of arteriovenous aneurysms of spinal cord. Radiology 88 (6): 1065-77, 1967.

473. Djindjian R: Embolization of angiomas of the spinal cord. Surg Neurol 4 (5): 411-20, 1975.

474. Kendall B E, Logue V: Spinal epidural angiomatous malformations draining into intrathecal veins. Neuroradiology 13 (4): 181-9, 1977.

475. Oldfield E H, Di Chiro G, Quindlen E A, et al.: Successful treatment of a group of spinal cord arteriovenous malformations by interruption of dural fistula. J Neurosurg 59 (6): 1019-30, 1983.

476. Heros R C, Debrun G M, Ojemann R G, et al.: Direct spinal arteriovenous fistula: A new type of spinal AVM. Case report. J Neurosurg 64 (1): 134-9, 1986.

477. Rosenblum B, Oldfield E H, Doppman J L, et al.: Spinal arteriovenous malformations: A comparison of dural arteriovenous fistulas and intradural avm’s in 81 patients. J Neurosurg 67: 795-802, 1987.

478. Gueguen B, Merland J J, Riche M C, et al.: Vascular malformations of the spinal cord: Intrathecal perimedullary arteriovanous fistulas fed by medullary arteries. Neurology 37: 969-79, 1987.

479. Mourier K L, Gobin Y P, George B, et al.: Intradural perimedullary arteriovenous fistulae: Results of surgical and endovascular treatment in a series of 35 cases. Neurosurgery 32: 885-91, 1993.

480. Strugar J, Chyatte D: In situ photocoagulation of spinal dural arteriovenous malformations using the nd:Yag laser. J Neurosurg 77: 571-4, 1992.

481. Bederson J B, Spetzler R F: Pathophysiology of type I spinal dural arteriovenous malformations. BNI Quarterly 12: 23-32, 1996.

482. Rodesch G, Hurth M, Alvarez H, et al.: Classification of spinal cord arteriovenous shunts: Proposal for a reappraisal--the Bicetre experience with 155 consecutive patients treated between 1981 and 1999. Neurosurgery 51 (2): 374-9; discussion 379-80, 2002.

483. Spetzler R F, Detwiler P W, Riina H A, et al.: Modified classification of spinal cord vascular lesions. J Neurosurg 96: 145-56, 2002.

484. Aminoff M J, Logue V: The prognosis of patients with spinal vascular malformations. Brain 97: 211-8, 1974.

485. Tobin W D, Layton D D: The diagnosis and natural history of spinal cord arteriovenous malformations. Mayo Clin Proc 51: 637-46, 1976.

486. Wirth F P, Post K D, Di Chiro G, et al.: Foix-Alajouanine disease. Spontaneous thrombosis of a spinal cord arteriovenous malformation: A case report. Neurology 20: 1114-8, 1970.

487. Criscuolo G R, Oldfield E H, Doppman J L: Reversible acute and subacute myelopathy in patients with dural arteriovenous fistulas: Foix-Alajouanine syndrome reconsidered. J Neurosurg 70: 354-9, 1989.

488. Barnwell S L, Dowd C F, Davis R L, et al.: Cryptic vascular malformations of the spinal cord: Diagnosis by magnetic resonance imaging and outcome of surgery. J Neurosurg 72: 403-7, 1990.

489. Anson J A, Spetzler R F: Interventional neuroradiology for spinal pathology. Clin Neurosurg 39: 388-417, 1991.

490. Nabors M W, Pait T G, Byrd E B, et al.: Updated assessment and current classification of spinal meningeal cysts. J Neurosurg 68: 366-77, 1988.

491. Yasui K, Hashizume Y, Yoshida M, et al.: Age-related morphologic changes of the central canal of the human spinal cord. Acta Neuropathol (Berl) 97 (3): 253-9, 1999.

492. Batzdorf U: Primary spinal syringomyelia. Invited submission from the joint section meeting on disorders of the spine and peripheral nerves, March 2005. J Neurosurg Spine 3 (6): 429-35, 2005.

493. Heiss J D, Oldfield E H: Pathophysiology and treatment of syringomyelia. Contemp Neurosurg 25 (3): 1-8, 2003.

494. Oldfield E H, Muraszko K, Shawker T H, et al.: Pathophysiology of syringomyelia associated with Chiari I malformation of the cerebellar tonsils. J Neurosurg 80: 3-15, 1994.

495. Williams B, Terry A F, Jones F, et al.: Syringomyelia as a sequel to traumatic paraplegia. Paraplegia 19: 67-80, 1981.

496. Suzuki M, Davis C, Symon L, et al.: Syringoperito-neal shunt for treatment of cord cavitation. J Neurol Neurosurg Psychiatry 48: 620-7, 1985.

497. Booth A E, Kendall B E: Percutaneous aspiration of cystic lesions of the spinal cord. J Neurosurg 33: 140-4, 1970.

498. Logue V, Edwards M R: Syringomyelia and its surgical treatment. J Neurol Neurosurg Psychiatry 44: 273-84, 1981.

499. Rossier A B, Foo D, Shillito J, et al.: Posttraumatic cervical syringomyelia. Brain 108: 439-61, 1985.

500. Vernon J D, Chir B, Silver J R, et al.: Posttraumatic syringomyelia. Paraplegia 20: 339-64, 1982.

501. Shannon N, Symon L, Logue V, et al.: Clinical features, investigation and treatment of posttraumatic syringomyelia. J Neurol Neurosurg Psychiatry 44: 35-42, 1981.

502. Griffiths E R, McCormick C C: Posttraumatic syringomyelia (cystic myelopathy). Paraplegia 19: 81-8, 1981.

503. Stanworth P A: The significance of hyperhidrosis in patients with posttraumatic syringomyelia. Paraplegia 20: 282-7, 1982.

504. Dworkin G E, Staas W E: Posttraumatic syringomyelia. Arch Phys Med Rehabil 66: 329-31, 1985.

505. Watson N: Ascending cystic degeneration of the cord after spinal cord injury. Paraplegia 19: 89-95, 1981.

506. Durward Q J, Rice G P, Ball M J, et al.: Selective spinal cordectomy: Clinicopathological correlation. J Neurosurg 56: 359-67, 1982.

507. Darbar A, Krishnamurthy S, Holsapple J W, et al.: Ventral thoracic spinal cord herniation: Frequently misdiagnosed entity. Spine 31 (17): E600-5, 2006.

508. Tekkok I H, Cataltepe K, Tahta K, et al.: Extradural hematoma after continuous extradural anesthesia. Brit J Anaesth 67: 112-5, 1991.

509. Harik S I, Raichle M E, Reis D J: Spontaneous remitting spinal epidural hematoma in a patient on anticoagulants. N Engl J Med 284: 1355-7, 1971.

510. Silber S H: Complete nonsurgical resolution of a spontaneous spinal epidural hematoma. Am J Emergency Med 14: 391-3, 1996.

511. Neurologic complications of regional anesthesia. In Anesthesia for obstetrics, Shnider S M and Levinson G, (eds.). Williams and Wilkins, Baltimore, 2nd ed., 1987: pp 319-20.

512. Sage D J: Epidurals, spinals and bleeding disorders in pregnancy: A review. Anaesth Intens Care 18: 319-26, 1990.

513. Gustafsson H, Rutberg H, Bengtsson M: Spinal hematoma following epidural analgesia: Report of a patient with ankylosing spondylitis and a bleeding diathesis. Anaesthesia 43: 220-2, 1988.

514. Domenicucci M, Ramieri A, Salvati M, et al.: Cervicothoracic epidural hematoma after chiropractic spinal manipulation therapy. Case report and review of the literature. J Neurosurg Spine 7 (5): 571-4, 2007.

515. Dickman C A, Shedd S A, Spetzler R F, et al.: Spinal epidural hematoma associated with epidural anesthesia: Complications of systemic heparinization in patients receiving peripheral vascular thrombolytic therapy. 72, 1990.

516. Packer N P, Cummins B H: Spontaneous epidural hemorrhage: A surgical emergency. Lancet 1: 356-8, 1978.

517. Bennett D L, George M J, Ohashi K, et al.: Acute traumatic spinal epidural hematoma: Imaging and neurologic outcome. Emerg Radiol 11 (3): 136-44, 2005.

518. Rebello M D, Dastur H M: Spinal epidural hemorrhage: A review of case reports. Neurol India 14: 135-45, 1966.

519. Domenicucci M, Ramieri A, Ciappetta P, et al.: Nontraumatic acute spinal subdural hematoma. J Neurosurg (Spine 1) 91: 65-73, 1999.

520. George W E, Wilmot M, Greenhouse A, et al.: Medical management of steroid-induced epidural lipomatosis. N Engl J Med 308: 316-9, 1983.

521. Fassett D R, Schmidt M H: Spinal epidural lipomatosis: A review of its causes and recommendations for treatment. Neurosurg Focus 16 (4): E11, 2004.

522. Kumar K, Nath R K, Nair C P V, et al.: Symptomatic epidural lipomatosis secondary to obesity: Case report. J Neurosurg 85: 348-50, 1996.

523. Haddad S F, Hitchon P W, Godersky: Idiopathic and glucocorticoid-induced spinal epidural lipomatosis. J Neurosurg 74: 38-42, 1991.

524. Roy-Camille R, Mazel C, Husson J L, et al.: Symptomatic spinal epidural lipomatosis induced by a long-term steroid treatment. Review of the iterature and report of two additional cases. Spine 16: 1365-71, 1991.

525. Robertson S C, Traynelis V C, Follett K A, et al.: Idiopathic spinal epidural lipomatosis. Neurosurgery 41: 68-75, 1997.

526. Beges C, Rousselin B, Chevrot A, et al.: Epidural lipomatosis. Interest of magnetic resonance imaging in a weight-reduction treated case. Spine 19 (2): 251-4, 1994.

527. Fessler R D, Johnson D L, Brown F D, et al.: Epidural lipomatosis in steroid-treated patients. Spine 17: 183-8, 1992.

528. Wray C C, Easom S, Hoskinson J: Coccydynia. Etiology and treatment. J Bone Joint Surg 73B: 335-8, 1991.

529. Lath R, Rajshekhar V, Chacko G: Sacral hemangioma as a cause of coccydynia. Neuroradiology 40: 524-6, 1998.

530. Thiele G H: Coccydynia: Cause and treatment. Dis Colon Rectum 6: 422-35, 1963.

531. Lourie J, Young S: Avascular necrosis of the coccyx: A cause of coccydynia? Case report and histological findings in sixteen patients. Br J Clin Pract 39: 247-8, 1985.

532. Raj P P: Miscelleneous pain disorders. In Pain medicine: A comprehensive review, Raj P P, (ed.). C V Mosby, St. Louis, 1996: pp 492-501.

533. Loev M A, Varklet V L, Wilsey B L, et al.: Cryoablation: A novel approach to neurolysis of the ganglion impar. Anesthesiology 88: 1391-3, 1998.

534. Plancarte R, Amescua C, Patt R B, et al.: Superior hypogastric plexus block for pelvic cancer pain. Anesthesiology 73: 236-9, 1990.

535. Vernon H, Mior S: The Neck Disability Index: A study of reliability and validity. J Manipulative Physiol Ther 14 (7): 409-15, 1991.