Craig J. Della Valle
Kevin J. Bozic
Total hip arthroplasty (THA) is among the most commonly performed and successful orthopaedic operations. In 2003 there were over 230,000 primary THAs performed in the United States alone, and in terms of quality-adjusted life years gained, THA is one of the most cost-effective health care interventions available. Despite the overwhelming success of THA for the treatment of end-stage hip disease, failures do occur, and the patient with a painful THA is one of the most difficult challenges for the orthopaedic surgeon to evaluate and treat. The purpose of this chapter is to review the workup and evaluation of the painful THA.
Pathogenesis
Etiology
The differential diagnosis of the painful THA is extensive (Table 14-1), and is most logically divided into causes that are intrinsic and extrinsic to the hip joint.
Intrinsic Causes of Pain
Infection should always be considered as a cause of pain because the treatment decisions rendered are fundamentally different if periprosthetic infection is identified. Mechanical loosening can occur early as a result of poor surgical technique or failure of osseointegration of a cementless implant, or it can occur late owing to failure of the cement mantle with a cemented implant. Stress fractures can be seen in the pubic rami (Fig. 14-1), are most commonly associated with female sex and osteopenia, typically occur within the first year postoperatively, and present with groin pain. Periprosthetic fractures, when not associated with obvious trauma, typically affect the greater trochanter in association with osteolysis or severe stress shielding. If a transtrochanteric approach was used for exposure, a nonunion of the greater trochanter can develop and cause pain. Osteolysis and polyethylene wear can cause pain secondary to the inflammatory response to wear debris. Occult instability or subluxation of the prosthetic hip is another potential cause of pain. Modulus of elasticity mismatch between a stiff implant (typically a large-diameter, cementless femoral stem) and the host bone can cause pain in the thigh. Trochanteric bursitis is a relatively common source of pain and has been associated with an anterolateral exposure to the hip and changes in femoral offset that occur at the time of surgery. Iliopsoas tendonitis presents as groin pain and has been associated with underanteverted, vertically oriented, and oversized acetabular components.
Extrinsic Causes of Pain
The most common cause of hip-related pain that is extrinsic to the hip joint is degenerative disease of the lumbar spine. Vascular claudication (particularly of the aortoiliac system) may also cause pain in the groin, buttock, or thigh that can be confused with pain directly related to the hip joint as can peripheral nerve dysfunction, hernias, and referred pain from intra-abdominal or genitourinary pathology. Less frequently seen causes of pain include metabolic bone disease and primary or metastatic bone cancers.
Epidemiology
Revision total hip arthroplasty accounted for approximately 15% of all total hip arthroplasties performed in 2003. With expanding indications for THA including younger and more active patients, an aging population and an ever-increasing total number of patients who have undergone THA in the past, even with improvements in surgical techniques and materials, the number of patients presenting with pain and failure following THA may rise in the future.
P.109
|
TABLE 14-1 Differential Diagnosis of the Painful Total HIP Arthroplasty |
|
|
Diagnosis
A thorough history and physical examination can provide critical information that allows the evaluating surgeon to narrow the differential diagnosis and is the basis for a focused workup of the patient with a painful THA.
|
Figure 14-1 Healed stress fractures of the pubic rami that occurred subsequent to the THA. |
The temporal onset, duration, severity, site, and character of the pain provide important clues in determining the cause of the painful THA. Persistent pain since surgery, without a painfree interval, is suggestive of infection (particularly if the pain experienced now is of a different nature than that prior to surgery), failure to obtain initial implant stability, or misdiagnosis of the original cause of the patient's pain when seeking total hip arthroplasty in the first place (particularly if the pain is unchanged when compared with preoperatively).
When interviewing patients with a painful THA, it is important to ask about any abnormal or persistent wound drainage that occurred after surgery and to determine if the patient was placed on antibiotics for an extended period of time or if he or she was returned to the operating room shortly after surgery as each of these factors should raise the suspicion for chronic infection. Acute pain following systemic illness, dental or gastrointestinal procedures, or a distant site of infection suggests the possibility of an acute hematogenous infection. Infection should also be considered more likely in patients with predisposing risk factors such as diabetes mellitus, inflammatory arthritis, immunocompromised states, renal failure, and skin disorders. Later onset of pain after a painfree interval suggests aseptic loosening, periprosthetic stress fracture, osteolysis, or late infection.
Activity-related pain that is severe but is improved by rest suggests loosening, fracture, and neurogenic or vascular claudication. Constant pain, pain at rest, or pain at night can be indicative of infection or malignancy. Pain that begins when a patient starts to walk after sitting or resting (often referred to as “start-up pain'') has been associated with loosening, micromotion, iliopsoas tendonitis (particularly groin pain when rising out of bed or out of a car), or lumbar spine disease. The onset of pain after a traumatic fall may be caused by fracture or traumatic loosening. Patients with both lumbar spine and hip disease may find that their symptoms of neurogenic claudication worsen after successful THA, secondary to increased activity levels.
Pain localized to the groin or the deep buttock often is associated with acetabular component loosening. Buttock pain that radiates below the knee suggests a neurogenic cause, such as lumbar disc disease or spinal stenosis. Thigh pain has been linked to femoral component loosening or a modulus mismatch as previously described. Localized pain over the greater trochanter is indicative of trochanteric bursitis or nonunion.
Physical Examination
A thorough physical examination of the patient with a painful THA should include a comprehensive musculoskeletal examination focusing on the ipsilateral and contralateral hip, knee, and spine and starts with a patient who is dressed in a gown for examination. Inspection of the skin to look for scars and signs of infection, including warmth, erythema, fluctuance, wound drainage, or sinus tracts is helpful and surprisingly oftentimes overlooked. The surgeon also should look for obvious muscle wasting around the hip or in the lower extremity. The patient's gait should be observed for evidence of antalgia, limb-length discrepancy,
P.110
muscle weakness, or a Trendelenburg gait. A Trendelenburg gait is indicative of abductor muscle weakness and is more common after the use of an anterolateral approach to the hip. True and apparent leg lengths should be measured using graduated blocks as patients with lumbar spine disease and/or fixed pelvic obliquity may present with notable differences between their true and apparent leg lengths.
A detailed neurovascular examination also should be done to rule out neurogenic and vascular causes of pain. Peripheral nerve injuries involving the sciatic or femoral nerve may result in distal muscle weakness such as a foot drop. Surgeons must be sure to establish the temporal relationship of the muscle weakness and limb-length inequality in relation to the time of surgery. Progressive limb shortening after THA documented on serial examinations suggests mechanical failure of fixation.
Palpation can be useful in the diagnosis of trochanteric bursitis, inguinal hernia, and stress fractures of the pelvic ring. Range of motion (ROM) should be assessed carefully to determine which (if any) positions reproduce the patient's pain. Pain with active ROM or at extremes of motion is indicative of loosening, whereas pain with passive ROM may indicate occult infection. Pain with passive straight leg raising should raise suspicion of sciatic nerve irritation. Pain or apprehension in certain reproducible positions, particularly at extremes of motion, is indicative of instability or impingement. Pain with resisted hip flexion and passive extension often is associated with iliopsoas tendonitis.
Radiographic Evaluation
Plain Radiographs
A thorough radiographic evaluation of the patient with painful THA begins with a critical review of serial plain radiographs that are of high enough quality to identify subtle changes and to evaluate for signs of loosening or migration of the prosthesis. Care should be taken to note differences in radiographic technique, including orientation, penetration, and rotation.
The most widely accepted criteria for radiographic loosening of cemented femoral stems have been proposed by Harris and McGann. This classification includes criteria for definite, probable and possible loosening (Table 14-2). Major signs of osseointegration of a cementless femoral component were identified by Engh et al. and include the absence of reactive, radiodense lines around the porous-coated portion of the implant and presence of endosteal spot-welds. Minor signs of osseointegration include calcar atrophy, a stable distal stem, and the absence of a pedestal. Major signs of failure of osseointegration are extensive reactive lines around the porous-surfaced portion of the implant, whereas the absence of endosteal spot-welds is considered a minor sign of failed osseointegration.
Similar to cemented femoral components, the stability of a cemented acetabular component has been divided into definite loosening (migration of >5 mm or a fracture of the cement mantle), probable loosening (100% radiolucent line at the bone/cement interface) and possible loosening (radiolucent line of 50% to 99% at the bone/cement interface). Criteria for loosening of a cementless acetabular component include migration (of >5 mm), the presence of broken screws (if present), and a complete radiolucent line around the component. Udomkiat et al. reported radiolucent lines that progressed after 2 years postoperatively, radiolucent lines of ≥1 mm in thickness that appeared after the second postoperative year, and a radiolucent line of >2 mm in any zone were particularly useful predictors of cementless acetabular component loosening.
|
TABLE 14-2 Definitions of Radiographic Loosening of Cemented Femoral Stems |
||
|
Although plain radiographs are not very sensitive for the evaluation of infection, certain radiographic findings are highly indicative of infection including periosteal new bone formation and endosteal scalloping. Component loosening that occurs within the first 5 years in an otherwise well-performed THA is particularly suggestive of infection.
Aspiration and Arthrography
Preoperative aspiration of the hip, with fluoroscopic guidance, to diagnose infection should be performed on a selective basis as this test has been associated with a high rate of false-positives and is time consuming, expensive, and uncomfortable for the patient. It is typically recommended when the clinical suspicion for infection is high (based on the history or presence of risk factors), when the erythrocyte sedimentation rate (ESR) and/or C-reactive protein (CRP) are elevated, or when failure occurs <5 years postoperatively. When fluid is obtained, specimens should be sent for aerobic, anaerobic, acid-fast bacilli and fungal cultures as well as a cell count with differential. The precise cutoff for identifying infection with a cell count is controversial; however, the use of >50,000 white blood cells (WBC) as would be applicable to a native hip is associated with a high rate of false-negatives. At our own centers, >3,000 WBC has been found to be highly suggestive of infection, and similar values have been identified for evaluating infection at the site of a total knee arthroplasty; we typically use cell counts intraoperatively as an adjunct to or instead of an intraoperative frozen section as a screening tool for infection. Similarly, the cutoff value for the differential is somewhat unclear, with different authors suggesting values of between 65% and 90% neutrophils. In contrast to an aspiration for
P.111
identifying infection, aspiration arthrography of the hip to delineate component loosening has been found to have a low sensitivity and specificity, especially for uncemented implants, and has been abandoned at most centers.
Anesthetic Injection
Local anesthetic injections can be useful in localizing the origin of pain in a patient with a painful THA and provide information as to whether the source of pain is intracapsular or extracapsular. Braunstein et al. reported that 10 of 11 patients with an identifiable intracapsular cause for pain obtained complete relief of pain within 20 minutes of an intraarticular bupivacaine injection. If a patient does not experience relief of symptoms after intra-articular injection, extra-articular sources of pain should be sought. Similarly, if a patient is suspected of having pain derived from the lumbosacral spine, relief from a local anesthetic injection at this site can assist in ruling out pain arising from a THA.
Nuclear Medicine Studies
Nuclear medicine studies occasionally may be helpful in evaluating the patient with a painful THA in whom the history, physical examination, and plain radiographic studies are unclear and revision surgery is not otherwise indicated (Fig. 14-2). Although a highly sensitive indicator of bone turnover and activity, this technique has a low specificity and the results are generally not considered accurate within 2 years of prosthetic implantation. Increased uptake on 99Tc MDP can be seen with various conditions, including loosening, infection, heterotopic bone formation, stress fractures, modulus mismatch, tumors, metabolic bone disease, and reflex sympathetic dystrophy. Lieberman et al., in a series of 54 hips that subsequently were revised, reported that bone scans were not more helpful than serial radiographs in determining component stability; the authors recommended a bone scan only if plain radiographic studies are inconclusive.
|
Figure 14-2 Three-phase bone scan showing diffuse uptake around the femoral component indicative of loosening. |
Gallium citrate (67Ga), which preferentially is taken up by leukocytes, has been used in combination with technetium bone scans to differentiate between septic and aseptic loosening. Both sensitivity and accuracy have been reported as poor, however, and this testing modality has been abandoned in most centers.
The use of indium-111–labeled leukocyte scans to identify deep infection also has yielded disappointing results that have been reported to be only slightly better than with gallium citrate scans. Scher et al. studied 143 hips and knees that had a reoperation for a painful or loose total joint arthroplasty or a prior resection arthroplasty. The positive predictive value only was 54%; however, the negative predictive value was 95%, suggesting that a negative scan is useful in predicting the absence of infection in a patient who is being evaluated for a painful total joint arthroplasty. The addition of technetium-99m–labeled sulfur colloid marrow scanning has been used at some centers in an attempt to decrease the high rate of false-positive indium scans secondary to physiologic marrow packing that can cause an accumulation, slowing, or retardation of flow of (indium-labeled) leukocytes around a prosthesis. The method provides some improvements; however, sensitivity has remained low and thus all of these tests are used only as a second line of investigation when the results of other testing modalities are unclear.
Computed Tomography
Computed tomography (CT) has been described as a useful adjunct, when used selectively, for evaluating the painful THA. Specific software packages are required to compensate for artifact generated by metallic prosthesis, and a preprocedure direct consultation with a radiologist is recommended to optimize the result. Its greatest utility seems to be in identifying and quantifying periprosthetic osteolysis and for evaluating component positioning that may be contributing to instability; it can also be useful for identifying iliopsoas tendonitis (Fig. 14-3).
|
Figure 14-3 CT scan of the pelvis showing enlarged iliopsoas tendon (arrows). |
P.112
Magnetic Resonance Imaging
Magnetic resonance imaging (MRI) is another potentially useful tool for evaluating the painful THA; however, the input of a radiologist is imperative to ensure that appropriate pulse sequences and software are used to decrease metallic artifact. MRI has been found to be useful in evaluating the soft tissues to identify iliopsoas tendonitis, incompetence of the hip capsule, and damage to the abductor musculature; for identifying and quantitating osteolysis; and for evaluating the adjacent neurovascular structures; it has also been suggested as a potential way to evaluate for infection and component loosening or fracture.
|
Figure 14-4 Algorithm for evaluation of the painful total hip arthroplasty. ESR/CRP, erythrocyte sedimentation rate/C-reactive protein. |
Laboratory Tests
An erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) should be obtained from any patient being evaluated for a painful THA. If these values are elevated, deep periprosthetic infection must be suspected and additional evaluation with preoperative or intraoperative testing (e.g., intraoperative frozen section or intraoperative cell count) should be done. In one series of 202 revision THAs, all patients with a deep infection had an elevated ESR (>30 mm/hour) or CRP (>1 mg/dL). In this series, the combination of a normal ESR and CRP reliably predicted the absence of infection (100% specificity). It is important to
P.113
recognize, however, that the ESR and CRP are nonspecific markers of inflammation and may be elevated by chronic inflammatory conditions (such as rheumatoid arthritis), recent surgical intervention, or systemic illness. The ESR, although usually normal by 6 months postoperatively, may be elevated for as much as 1 year after uncomplicated THA.
Diagnostic Workup Algorithm
Evaluation of the patient with a painful THA is a complex task that requires diligence and patience on the part of the surgeon and the patient (Fig. 14-4). After a thorough history and physical examination, a focused and careful radiographic and laboratory workup can be done at a reasonable cost and with a high degree of accuracy.
Treatment
Surgical Indications/Contraindications
When considering revision surgery for the patient with a painful THA, a clear diagnosis of the condition that is responsible for the patient's symptoms must be sought prior to further operative intervention as the risks of revision THA can be substantial; also, repeat surgery without a clear diagnosis of a condition that can be remedied by surgical intervention may be met with a high rate of persistent pain and an unsatisfied patient. Patients must be clearly counseled on the cause of their problem, the treatment options, and the risks of operative intervention. Nonoperative measures and observation are the mainstays of treatment until a definitive and treatable diagnosis can be confirmed.
Definite indications for revision surgery in the patient who is a reasonable risk medically for elective operative intervention include chronic infection and loose implants. Patients with a diagnosis of infection should be revised to prevent both worsening of infection and the associated bone loss that can ensue if treated nonoperatively. Similarly, patients with loose implants may sustain periprosthetic bone loss if the problem is neglected, which in turn can lead to greater operative complexity, a higher risk of complications and a poorer clinical result when revision surgery finally is performed. Further details regarding revision of surgical technique, results, and postoperative management are covered in Chapter 15.
Suggested Reading
Barrack RL, Harris WH. The value of aspiration of the hip joint before revision total hip arthroplasty. J Bone Joint Surg. 1993;75A:66–76.
Bohl W, Steffee A. Lumbar spinal stenosis: a cause of continued pain and disability in patients after total hip arthroplasty. Spine. 1979:4:168–173.
Bourne RB, Rorabeck CH, Ghazal ME, et al. Pain in the thigh following total hip replacement with a porous-coated anatomic prosthesis for osteoarthrosis. J Bone Joint Surg. 1994;76A:1464–1470.
Braunstein EM, Cardinal E, Buckwalter KA, et al. Bupivacaine arthrography of the post-arthroplasty hip. Skeletal Radiol. 1995;24:519–521.
Della Valle CJ, Zuckerman JD, DiCesare PE. Periprosthetic sepsis. Clin Orthop. 2004;420:26–31.
Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop. 1990;257:107–128.
Harris WH, McGann WA. Loosening of the femoral component after use of the medullary-plug cementing technique. Follow-up note with a minimum of five-year follow-up. J Bone Joint Surgery. 1986;68A:1064–1066.
Lieberman JR, Huo MH, Schneider R, et al. Are technetium bone scans necessary? J Bone Joint Surg. 1993;75B:475–478.
O'Neill D, Harris WH. Failed total hip replacement: Assessment by plain radiographs, arthrograms, and aspiration of the hip joint. J Bone Joint Surg. 1984;66A:540–546.
Potter HG, Nestor BJ, Sofka CM, Ho, et al. Magnetic resonance imaging after total hip arthroplasty: evaluation of periprosthetic soft tissue. J Bone Joint Surg. 2004;86A:1947–1954.
Scher DM, Pak K, Lonner JL, et al. The predictive value of indium-111 leukocyte scans in the diagnosis of infected total hip, knee, or resection arthroplasties. J Arthroplasty. 2000;15:295–300.
Spangehl MJ, Masri BA, O'Connell JX, et al. Prospective analysis of preoperative and intraoperative investigations for the diagnosis of infection at the sites of two hundred and two revision total hip arthroplasties. J Bone Joint Surg. 1999;81A:672–683.
Udomkiat P, Wan Z, Dorr LD. Comparison of preoperative radiographs and intraoperative findings of fixation of hemispheric porous-coated sockets. J Bone Joint Surg. 2001;83:1865–1875.