Michael H. Kwon and Christopher R. Morse
First described by Wood in 1916 and first performed successfully by Morrison in 1923, diaphragmatic plication as a surgical treatment for diaphragmatic paralysis secondary to phrenic nerve dysfunction remains an effective option for both pediatric and adult patients in whom the associated mechanical effects of a paralyzed and elevated hemidiaphragm have led to chronic symptoms of dyspnea and/or orthopnea. It is also indicated for well-selected mechanically ventilated patients in whom chronic diaphragmatic paralysis continues to preclude effective weaning from mechanical ventilation after other contributory factors have been optimized. Understanding the etiology and pathophysiology of diaphragmatic paralysis is critical in understanding the indications and contraindications of this procedure and thereby the ability to appropriately select patients who stand to benefit from the procedure with minimal risk.
Diaphragmatic paralysis can be congenital or acquired although in many cases it is difficult to identify an obvious cause, many suspecting a viral etiology. Congenital paralysis or paresis is known as diaphragmatic eventration and results from a failure of normal diaphragmatic muscular development in the setting of normal phrenic nerve function. Plication in infants and children with eventration is beyond the scope of this chapter, which will instead focus on adults with acquired disease. In recent years the most common acquired etiology in adults is iatrogenic injury to the phrenic nerve after cardiac (and to a lesser extent thoracic) surgical procedures and usually leads to unilateral diaphragmatic parlaysis. In order of increasing severity and chronicity of the paralysis itself, injury to the nerve in this setting results from hypothermic injury from ice or cold packs placed in the chest (usually leading to temporary paralysis/paresis), traction injury, thermal injury from local electrocautery, and complete laceration (usually but not always permanent). Other etiologies include invasion of tumor, penetrating trauma, blunt trauma, as well as other more rare conditions including neuralgic amyotrophy. Meanwhile, bilateral diaphragmatic dysfunction usually develops gradually on a spectrum from paresis to paralysis and is usually limited to patients with progressive neuromuscular disorders such as amyotrophic lateral sclerosis, multiple sclerosis, and myasthenia gravis.
When the diaphragm becomes paralyzed, normal caudal movement of the diaphragm during inspiration becomes impaired, leading to decreased ventilation. The diaphragm takes on a more cephalad baseline position throughout the respiratory cycle. With ongoing disuse, the muscle fibers become atrophic, and the surface area of the diaphragm expands as the connective tissue remodels, ultimately leading to a highly redundant and floppy overall architecture. As this progresses, the negative intrapleural pressure created by contraction of the accessory muscles of inspiration can lead to paradoxical motion of the diaphragm in the cephalad direction during inspiration (reduced inspiratory capacity), and in the caudad direction during expiration when the intrapleural pressure becomes positive (increased functional residual capacity), thereby further limiting tidal volumes and overall ventilation. For patients with underlying lung disease, this can create significant hypoxia as well and exacerbate dyspnea. Furthermore, when patients are in the supine position, the noncontractile and redundant diaphragm is unable to exert the normal tonic opposition to the gravitational pressure exerted by the intra-abdominal viscera, leading to further cephalad displacement of the diaphragm and the development of orthopnea as a progressive symptom.
Understanding these mechanical disadvantages, one can then appreciate that although plication does not restore contractile function of the diaphragm per se, many, if not all of the deleterious mechanical effects of chronic paralysis can be significantly ameliorated simply by respositioning and surgically remodeling the diaphragm into a flat, taut, nonredundant, and more caudally oriented planar structure by means of plication. In doing so, plication can allow for the resolution of symptoms (namely dyspnea and orthopnea) by achieving the following specific mechanical goals:
1. Increased total lung capacity by virtue of more caudal displacement of the diaphragm.
2. Decreased redundancy of diaphragmatic surface area and minimization of paradoxical motion.
3. Decreased diaphragmatic compliance that allows intercostal and other accessory muscles of respiration to more effectively create negative and positive intrapleural pressures necessary for effective inspiration and expiration, respectively.
4. Increased opposition to cephalad displacement of abdominal viscera while in supine position to decrease orthopnea.
However, unless patients are symptomatic, improving the respiratory mechanics in this manner is not indicated, that is, the presence of an elevated hemidiaphragm alone does not in and of itself warrant plication unless symptoms are present. Furthermore, depending on the etiology, the decision to operate should be delayed to ensure that the paralysis is indeed permanent. Several authors have suggested a 1- to 2-year period of observation for patients with postcardiac surgery diaphragmatic paralysis if the symptoms are mild, chronic, and stable. A much shorter period of observation may be appropriate in patients with more severe symptoms or for patients in whom symptoms have progressed and are not attributable to other forms of lung disease or heart failure. A significantly shorter period of 2 to 6 weeks may be suitable for patients in whom the goal of plication is to assist in effective weaning from mechanical ventilation.
Meanwhile, plication is typically deferred in morbidly obese patients for several reasons including (1) overall technical difficulty due to body habitus and the potential for arriving at suboptimal mechanical results, and (2) the possibility that with aggressive weight loss/bariatric surgery, the associated improvement in overall chest wall compliance may be sufficient to improve symptoms without plicating the diaphragm. Finally, those with progressive neuromuscular disorders are usually not candidates because it is not just the diaphragm that is dysfunctional but also the accessory muscles of respiration. The benefit of plication is, therefore, limited, and such patients are better candidates for an eventual transition to a tracheostomy and mechanical ventilation as palliation near the end of life.
PREOPERATIVE PLANNING
The three critical concepts in the preoperative planning for a diaphragmatic plication are to (1) make the diagnosis, (2) determine the presence and duration of symptoms, and (3) medically optimize other factors that may be contributing to dyspnea.
In symptomatic patients, diagnosis is usually established by the finding of an elevated hemidiaphragm on a chest radiograph. Portable radiographs taken in critically ill patients are significantly less specific for the diagnosis, especially if the degree of elevation is mild (less than 2 cm), but a high-quality film taken of a standing patient at full inspiration demonstrating significant elevation (e.g., several rib spaces) is usually sufficient to make the diagnosis without additional studies. When the diagnosis is in doubt, a fluoroscopic sniff test or ultrasound can be used to demonstrate limited or paradoxical diaphragmatic movement with full respiratory effort.
We obtain standing and supine pulmonary function tests where a restrictive pattern may worsen when supine, evidenced by a drop in vital capacity. However, PFTs are often nonspecific for the diagnosis and are rather used as an adjunct to a carefully taken history and physical examination in determining how the patient will tolerate single-lung ventilation during the planned procedure.
Ruling out and optimizing other concomitant sources of respiratory insufficiency (e.g., heart failure, interstitial lung disease, pneumonia, smoking, morbid obesity, reactive airway disease, etc.) is critical in the observational period as symptoms may resolve without the need for surgery when these factors are improved.
Standard preoperative planning otherwise includes a full chemistry panel, complete blood count, coagulation profile, and a recent chest radiograph (assists in determining the rib space for entry into the chest).
SURGERY
Positioning
After intubation with a dual-lumen endotracheal tube, a nasogastric tube is placed to decompress the stomach. The patient is then positioned in the standard decubitus position with the break of the operating room table near the level of the xiphoid or slightly lower. Steep Trendelenburg positioning aids in caudal displacement of the abdominal viscera and, therefore, the diaphragm by gravity, allowing for less tension on the sutures as the operation proceeds. In addition, placement of a thoracic epidural catheter for postoperative pain control is recommended, especially for patients undergoing an open procedure.
Open Transthoracic Diaphragmatic Plication
A posterolateral thoracotomy is made in the sixth, seventh, or eighth intercostal space depending on the degree of diaphragmatic elevation on preoperative imaging. The inferior pulmonary ligament is taken down sharply to improve the working space. The standard technique, thought to distribute tension most evenly over the diaphragm, is termed the accordion technique, in which multiple rows of nonabsorbable, pledgeted horizontal mattress sutures are placed in the diaphragm (Fig. 18.1). Using a double-armed suture of relatively large caliber (e.g., size 2-0 or 0 polypropylene monofilament or braided polyester), preloaded with a pledget midlength, multiple consecutive full-thickness bites are taken with each arm of the suture running in a straight line from posterior to anterior along the medial aspect of the diaphragm to create pleats, which are oriented in a medial-to-lateral axis. Once both arms of each suture are taken all the way to the anterior aspect of the diaphragm, the needles are placed through another pledget and the suture tied down. Additional sutures are placed parallel to the first row, working medial to lateral with each successive suture until the redundancy has been completely taken up and the diaphragm pleated down with an accordion-like topography. Care is taken to prevent inadvertently taking bites into hollow abdominal viscera and some surgeons advocate for opening the diaphragm at the central tendon, which allows direct visualization of the abdomen. The thoracotomy is closed in layers and a thoracostomy tube is left in place.
Alternative suturing schema for plicating the diaphragm exist including (1) interrupted horizontal mattress sutures without the accordion technique (requires more pledgets, more time-consuming, but sometimes necessary in an extremely thin and friable diaphragm), and (2) continuous running stitches (baseball stitch vs. horizontal mattress). Due to the small number of patients in even the largest case series, it is unknown whether the long-term results are more favorable with one technique versus the other. In our anecdotal experience, the pledgeted accordion technique described above represents the optimum balance between efficiency and distribution of tension.
Figure 18.1 “Accordion plication.” A: Mattress sutures are placed with bites oriented along the posterior to anterior axis. B: Completed repair.
Thoracoscopic Diaphragmatic Plication
Thoracosopic diaphragmatic plication is a minimally invasive method by which similar results can be achieved in comparison to the open technique. The potential advantages include decreased postoperative pain and decreased length of stay. The main disadvantages are that unlike the open procedure (which, with an adequate incision and good retraction of inflated lung, can reasonably be performed without single-lung ventilation), thoracoscopic plication requires dual-lumen intubation and decompression of the ipsilateral lung, especially given the limited space in which to maneuver in the setting of the elevated diaphragm. For patients in whom pulmonary function is adequate, thoracoscopic plication is performed using the same positioning as in the open technique. Two ports can be used if the operating port is made into a posterolateral minithoracotomy through which standard instruments can be used. A more standard approach is to use three ports, which typically include a 10-mm thoracoscopic port at the midaxillary or posterior axillary line in the eighth interspace and two operating ports in the fifth interspace, one anteriorly and one posteriorly (Fig. 18.2). Sutures can be placed and tied intracorporeally with traditional thoracoscopic instruments or, depending on the experience of the surgeon, the Endo Stitch (Covidien, Mansfield, MA) device can also be used. The type of suture used is similar to that used in the open technique. Again, as in the open technique, there are several schema for placing the plicating sutures including (1) the pledgeted accordion technique (often technically difficult due to space and angle constraints), (2) multiple rows of pledgeted interrupted U-stitches, and (3) one or more continuous running sutures (Fig. 18.3). Regardless of the method, the principle of distributing tension while taking up as much of the redundancy as possible remains the same as in open surgery.
Figure 18.2 Port site placement for thoracoscopic plication. Two ports are placed in the fifth intercostal space and a third port in the eighth intercostal space. The thoracoscope is moved from one port to another depending on the specific location of the diaphragm being sutured. Shown here with use of an Endo Stitch device.
Figure 18.3 Alternate schema for placement of diaphragmatic sutures for plication.
Laparoscopic Diaphragmatic Plication
Laparoscopic techniques have also gained in popularity in recent years. A full description of the laparoscopic approach can be found in Chapter 19.
POSTOPERATIVE MANAGEMENT
Patients are generally extubated in the operating room unless they already required mechanical ventilation in the preoperative period. A single thoracostomy tube is left to suction at a pressure of −20 cm H2O. Notwithstanding significant patient comorbidities and varying levels of comfort among staff and nursing, most patients undergoing surgery electively can be admitted to a monitored floor bed after a short stay in the postanesthesia care unit. A portable chest radiograph is obtained immediately in recovery to assess for the new baseline position of the diaphragm and adequate expansion of the lung. The thoracostomy tube is usually removed on postoperative day 1 without the need for a trial on water seal unless an air leak is present. The diet is advanced, followed by removal of the epidural catheter and Foley catheter when patients demonstrate an ability to tolerate oral intake. Virtually all patients experience a sensation of tightness in the lower abdominal area after the procedure and should be counseled as such. Attention is paid to rule out signs of peritonitis, which could represent injury to the stomach or bowel. An aggressive bowel regimen, antiemetics, and pulmonary toilet are essential to prevent and treat problems such as a postoperative ileus, wretching/vomiting, and coughing, all of which have the potential to lead to undue tension on the new repair and cause tearing or failure of the stitches. Barring these and other early complications, patients are often ready for hospital discharge in 2 to 4 days.
COMPLICATIONS
All manner of complications have been described in the myriad small case series written on both open and thoracoscopic procedures. In general the spectrum of complications is similar to that seen in pulmonary resections (e.g., atrial fibrillation, pneumonia, pulmonary edema, pleural effusions, acute myocardial infarction, empyema, superficial wound infection, deep venous thrombosis, pulmonary emoblism, etc.), but with a decreased risk of persistent air leak and/or bronchopleural fistula as the lung is not intentionally violated during plication. Meanwhile, there is a higher overall risk of injury to the spleen, stomach, liver, colon, and small intestine, which can occur while taking full-thickness bites of diaphragm without adequate cephalad retraction of the local tissue. As noted above, excessive coughing or wretching in the early postoperative period can cause early recurrence of diaphragmatic elevation.
RESULTS
A pooled analysis of five retrospective and one prospective case series of nonventilated adults undergoing both open and thoracoscopic diaphragmatic plication from 1979 to 2003 (total of 71 patients, mean follow-up ranging from approximately 2 to 8.5 years) demonstrated no operative deaths, resolution of symptoms in 96% of patients, radiologic improvement in 92% of patients, and functional improvement in 100% of patients. Overall short-term morbidity noninclusive of early recurrence (see complications above) occurs in about 5% to 10% of patients. To date, there are no adequately powered studies that can comment on the long-term differences in outcomes between (1) open versus thoracoscopic plication, and (2) any plication suturing schema versus another.
CONCLUSIONS
In well-selected symptomatic patients in whom other potential sources of restrictive and obstructive pulmonary disease have been ruled out or optimized, diaphragmatic plication for diaphragmatic paralysis is an effective surgical treatment with a high success rate and overall low morbidity. Although the ideal operative approach for the typical patient who is able to tolerate single-lung ventilation remains unknown, the choice between open thoracotomy and thoracoscopic plication can be reasonably informed by the training, experience, and preference of the surgeon.
Recommended References and Readings
Blackmon SH. Technique of video-assisted thoracoscopic surgery diaphragm plication. Oper Tech Thorac Cardiovasc Surg. 2012; 302–310.
Curtis JJ, Nawarawong W, Walls JT, et al. Elevated hemidiaphragm after cardiac operations: Incidence, prognosis, and relationship to the use of topical ice slush. Ann Thorac Surg. 1989;48:764–768.
Freeman RK, Van Woerkom J, Vyverberg A, et al. Long-term follow-up of the functional and physiologic results of diaphragm plication in adults with unilateral diaphragm paralysis. Ann Thorac Surg.2009;88:1112–1117.
Graham DR, Kaplan D, Evants CC, et al. Diphragm plication for unilateral diaphragmatic paralysis: A 10-year experience. Ann Thorac Surg. 1990;49:248–252.
Groth SS, Andrade RS. Diaphragm plication for eventration or paralysis: A review of the literature. Ann Thorac Surg. 2010;89:S2146–S2150.
Higgs SM, Hussain A, Jackson M, et al. Long term results of diaphragmatic plication for unilateral diaphragm paralysis. Eur J Cordiothoracic Surg. 2002;21:294–297.
Moroux J, Padovani B, Poirier NC, et al. Technique for the repair of diaphragmatic eventration. Ann Thorac Surg. 1996;62:905–907.
Morrison JMW. Eventration of diaphragm due to unilateral phrenic nerve paralysis. Arch Radiol Electrother. 1923;28:72–75.
Patterson GA, Cooper JD, Deslauriers J, et al, eds. Pearson’s Thoracic and Esophageal Surgery. 3rd ed. Philadelphia, PA: Churchill Livingstone Elsevier; 2008:1431–1444.
Versteegh MI, Braun J, Voigt PG, et al. Diaphragm plication in adult patients with diaphragm paralysis leads to long-term improvement of pulmonary function and level of dyspnea. Eur J Cardiothorac Surg.2007;32:449–456.