 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. The Operative Treatment of Peroneal Nerve Palsy*  BALTIMORE, MARYLANDInvestigation performed at the Department of Orthopaedic Surgery, The Johns Hopkins University School of Medicine, Baltimore
We retrospectively reviewed the results of operative decompression for peroneal nerve palsy in thirty-one patients who had been managed between 1980 and 1990. All patients had been managed non-operatively for at least two months after they had initially been seen. Intraoperatively, we found epineurial fibrosis and bands of fibrous tissue constricting the peroneal nerve at the level of the fibular head and at the proximal origin of the peroneus longus muscle. At a mean of thirty-six months (range, twelve to seventy-two months) postoperatively, thirty (97 per cent) of the thirty-one patients reported subjective and functional improvement and were able to discontinue the use of the ankle-foot orthosis. In contrast, only three of nine patients who had been managed non-operatively reported subjective and functional improvement (p < 0.01). Peroneal nerve palsy does not always resolve spontaneously; if it is left untreated, the loss of dorsiflexion of the ankle and persistent paresthesias can result in severe functional disability. Therefore, if non-operative measures do not lead to improvement within two months, we believe that operative decompression should be considered.
Peroneal nerve palsy is the most frequently encountered mononeuropathy in the lower extremity8. The cumulative reported prevalence after total knee arthroplasty has ranged from less than 1 to 4 per cent (seventy-four [0.58 per cent] of 12,784 procedures1,5,9,12,16,17,21,23,29,35), and the prevalence after proximal tibial osteotomy has ranged from 3 to 13 per cent (thirty-eight [10 per cent] of 395 procedures2,8,14,15,18,22,32,33,42). Operative injury of the peroneal nerve has many causes, including ischemia, mechanical irritation, traction, crushing injury, and laceration13. Some investigators have suggested that operative correction of excessive valgus or flexion deformities during knee arthroplasty can stretch the peroneal nerve1,9,11,12,17,24,29. We found a peroneal nerve palsy after a total knee arthroplasty for a fixed valgus deformity of 15 degrees or more in four (3 per cent) of 138 knees (unpublished data). Spontaneous resolution of peroneal nerve palsy, over a period ranging from months to years, has been reported by authors who have advocated non-operative treatment11,16,21,35. Operative treatment is controversial even though decompression of the common peroneal nerve has been well described in the literature26,27,38. The authors who have recommended operative treatment have not delineated the indications for or the timing of such intervention38. In the current study, we describe the results after operative decompression of the peroneal nerve in patients who had severe palsy that did not resolve after a minimum of two months of observation. The results of non-operative treatment are described for a separate group of patients.
We retrospectively reviewed the results of operative treatment of peroneal nerve palsy in thirty-one patients who had had no improvement for at least two months after the onset of the symptoms. The mean age of the twenty female and eleven male patients was forty-seven years (range, sixteen to seventy-eight years). The mean interval between the initial presentation of the nerve palsy and the operative decompression was twenty-one months (range, two to eighty-four months) (Table I).
Twenty-six of the thirty-one patients had initially been managed elsewhere for the underlying problem and were referred to us for treatment of the peroneal nerve palsy. The hospital and clinic records were reviewed with respect to the demographic data; the possible etiology, the onset, and the initial treatment of the palsy; the clinical findings; the results and timing of any diagnostic studies; the operative findings; the pathological reports; the postoperative management; and any complications. In six patients, the palsy had developed after a total knee arthroplasty; in four, after a proximal tibial osteotomy; in three, after a fracture of the tibia; in six, after a fracture of the knee that had been treated with open reduction; in three, after a fracture of the ankle that had been treated with a cast; in one, after a distal femoral osteotomy; in one, as a complication of traction after a total hip arthroplasty; in one, after a fracture of the foot; and in six, as a result of a soft-tissue injury (Table I). The preoperative and postoperative clinical evaluations and the operations were performed by one of the senior ones of us (M. A. M., A. L. D., K. A. K., or D. S. H.). At the most recent follow-up examination, the patients were questioned by four of us (M. A. M., F. C., A. L. D., and M. W. H.) about their satisfaction with the result of the procedure, and a palsy score was calculated on the basis of the findings of the motor and sensory examination. The motor examination included a manual test of the muscles of the great toe and ankle with use of the unaffected side as a control, and the sensory examination identified areas of numbness, paresthesias, and pain. Scoring of sensory and motor function was on a scale ranging from 0 to 5 points, as described previously24. The need for an orthosis and function with regard to walking were also assessed. The severity of the palsy in all patients was assigned a score according to a 5-point scale that was originally used to assess peripheral neuropathies associated with hip arthroplasty24,41. This scale describes functional deficits more precisely than the previously used terms complete or incomplete for motor loss and mild or profound for sensory loss. A score of 1 point indicates only electromyographic evidence of neural damage and no symptoms; 2 points, electromyographic and physical findings of neuropathy but no symptoms; 3 points, electromyographic changes and slight signs and symptoms of neuropathy but no need for walking aids; 4 points, moderate motor and sensory loss, electromyographic changes, and some problems with gait; and 5 points, severe motor and sensory loss, marked electromyographic changes, and the need for an orthosis in order to walk at all times. Only patients who had a score of 4 or 5 points were included in the study. Patients who had diffuse peripheral neuropathy (without isolated compromise of the peroneal nerve) secondary to diabetes or lead poisoning were excluded. All patients had had electromyographic and nerve-conduction-velocity studies at some point in time, with documented peroneal motor-nerve-conduction blocks (recording electrodes were placed in the tibialis anterior muscle3,20) at the level of the fibular head. Whenever possible, the studies were performed within one month after the onset of the symptoms; they were done in this period in twenty-two of the thirty-one patients. Electromyographic and nerve-conduction-velocity studies were performed immediately preoperatively in twenty-seven patients; four patients did not have the preoperative tests for personal reasons. Eighteen of the twenty-seven patients had had the earlier studies (the studies were repeated after a three-month or longer period of no clinical improvement), and none had evidence of any changes compared with the initial findings. The studies in the nine patients who initially had not had such studies all demonstrated delays in the conduction of the peroneal nerve at the level of the fibular head. None of the patients had nerve-conduction-velocity studies postoperatively. External neurolysis of the peroneal nerve was performed at the level of the fibular head24,27,39. Postoperatively, the extremity was immobilized in a bulky Robert Jones-type dressing for two days, followed by early range-of-motion exercises. All patients were allowed to bear weight as tolerated immediately after the procedure. An additional group of nine patients who had severe peroneal-nerve palsy (a score of 4 or 5 points) were managed non-operatively during the same time-period (Table II). There were three men and six women, with an average age of fifty-six years (range, twenty-six to eighty-nine years). In four patients, the peroneal nerve palsy had developed after a total knee arthroplasty; in one, after a total hip arthroplasty (because of pressure from balanced suspension); in two, after blunt trauma to the knee during a motor-vehicle accident; in one, after a hysterectomy that involved the use of leg stirrups for four hours; and in one, after a proximal tibial osteotomy. One of the nine patients was managed non-operatively because of an acute myocardial infarction; the eight remaining patients refused operative intervention for personal reasons.
At the time of follow-up (mean duration, thirty-six months; range, twelve to seventy-two months), the mean preoperative palsy score was 4.5 points (4 or 5 points), compared with a mean of 1.5 points (range, 1 to 4 points) postoperatively. Of the thirty-one patients, twenty-one (68 per cent) had no remaining motor or sensory symptoms (a palsy score of 1 point), four (13 per cent) had complete or nearly complete relief of the sensory symptoms (a score of 2 points), five (16 per cent) had some decrease in the symptoms but still had residual sensory or motor deficits (a score of 3 points), and one had little improvement (a score of 4 points). We did not find any relationship between the age of the patient and the recovery of neural function. Thirteen of the nineteen patients who were less than fifty years old had complete resolution of the sensory and motor symptoms, compared with eight of the twelve who were more than fifty. The time-interval between the onset of the symptoms and the decompression appeared to affect the outcome of the operation. Full recovery of motor function (grade 5) was seen in all eight patients who had had the decompression within six months after the onset of the palsy, in four of the five who had been operated on between seven and twelve months after it, in seven of the eleven who had been operated on between thirteen and twenty-four months after it, and in six of the seven who had been operated on more than twenty-four months after it. There was no relationship between the severity of the preoperative palsy and recovery of motor function; twelve of the fourteen patients who had had a score of 5 points and thirteen of the seventeen who had had a score of 4 points had full recovery. A longer duration of follow-up did not alter the clinical outcome. Thirteen of the seventeen patients who had been followed for more than two years (mean, fifty-three months; range, twenty-seven to seventy-two months) had full motor recovery, compared with twelve of the fourteen who had been followed for less than two years (mean, sixteen months; range, twelve to twenty-three months). The six patients in whom the palsy had developed after a total knee arthroplasty had a decrease in the symptoms after the operative decompression. Five of these patients had complete relief of the motor and sensory symptoms. The sixth patient regained complete motor function but had some residual sensory paresthesias. All four patients in whom the palsy had developed secondary to a proximal tibial osteotomy had a decrease in the symptoms. There were no intraoperative or postoperative complications. Intraoperative examination of the peroneal nerve at the fibular head revealed a variable range of scarring and edema. There was no relationship between the extent of scarring or edema and the clinical result. We found epineurial fibrosis and bands of fibrous tissue constricting the peroneal nerve at the level of the fibular head and at the proximal origin of the peroneus longus muscle. Three of the nine patients who had been managed non-operatively recovered motor and sensory function, at a mean of fifty months (range, twenty-four to eighty-four months). The palsy scores for these nine patients improved from a mean of 4.3 points (4 or 5 points) preoperatively to a mean of 3.1 points (range, 1 to 5 points) postoperatively. The difference in the rate of recovery between the patients who had been managed operatively and those who had been managed non-operatively was significant (p < 0.01, Fisher exact test).
Peroneal nerve palsy after trauma or an elective procedure around the knee can be functionally debilitating. A complete palsy results in weakness of dorsiflexion of the foot and toes and eversion of the foot, leading to the characteristic footdrop and slapping gait. The sensory loss or paresthesias extend over the anterolateral surface of the leg and the dorsum of the foot. Without a high index of suspicion, it is possible that the milder forms of peroneal nerve palsy may be missed. Damage to the peroneal nerve can result from musculoskeletal trauma to the hip, ankle, or foot4,25,26,28,30,31,34,36,37 as well as from direct pressure on the nerve at the knee from a constrictive dressing, splint, or cast10. The common peroneal nerve can be damaged anywhere along its course, from its origin in the pelvis as the lateral trunk of the sciatic nerve to its bifurcation just distal to the knee6,19,40. The pressure from the proximal portion of a below-the-knee cast that ends just distal to the neck of the fibula can result in peroneal nerve palsy. A hematoma in the nerve sheath may produce sufficient compression of the peroneal nerve to result in the gradual development of a palsy34,36. The cumulative reported prevalence of peroneal nerve palsy was 0.58 per cent (seventy-four of 12,784 procedures) after total knee arthroplasty1,5,9,12,16,17,21,23,29,35 and 10 per cent (thirty-eight of 395 procedures) after proximal tibial osteotomy2,8,14,15,18,22,32,33,42. After total knee arthroplasty in 138 knees that had a fixed valgus deformity, we noted a 3 per cent prevalence of peroneal nerve palsy (unpublished data). In the current study, we could not estimate the prevalence of this problem after knee or ankle injuries, as we did not know how many patients had such injuries. Others have reported a high prevalence in association with sprains of the ankle4,25,30,31,34,36,37. Bosien et al., in a follow-up study of college students, found mild weakness of the peroneal muscles after twenty-nine (22 per cent) of 133 sprains of the ankle4. However, the weakness may have been related to overstretching or disuse of the muscles rather than to injury of the nerve. Other authors have thought that persistent disability after ankle injuries may be partially due to subclinical neural injury25,31,34 and that attempts should be made to determine whether a mild peroneal-nerve paralysis is contributing to the disability. Nobel applied traction on the superficial peroneal nerve at the foot in twelve cadaver specimens and produced increased tautness of the common peroneal nerve in eight; in the other four specimens, all from older subjects, the nerve was found to be torn where it exited from the fascia near the ankle34. Nobel suggested that, as traction alone produced these changes, the much greater tensile forces on the nerve produced by spiral fractures of both bones of the leg would result in tears in the nutrient vessels and nerve along the entire length of the nerve, including the common peroneal nerve at the sciatic bifurcation. Minor trauma to the ankle can involve the terminal branches of the superficial peroneal nerve where they pierce the deep fascia of the leg proximal to the ankle36. Treatment has ranged from simple immobilization of the leg in flexion for acute palsies to the use of an ankle-foot orthosis for chronic motor palsies with footdrop, slapping gait, and sensory loss. One report suggested that spontaneous recovery does not occur1. However, most studies2,5,9,11,16,21,29,42 have included less than five patients and have been somewhat anecdotal. We are aware of only two reports on the non-operative treatment of peroneal nerve palsies after total knee replacement. Rose et al. reported on twenty-three peroneal nerve palsies after 2626 total knee arthroplasties35. The patients either had no treatment or the dressing was simply loosened. Only two patients, who initially had motor loss alone, had complete recovery at a mean of three years (range, six months to seven years). The motor deficits initially noted in all twenty-two patients (twenty-three palsies) fully resolved in six. Asp and Rand reported the results of non-operative treatment of twenty-six peroneal nerve palsies after 8998 knee arthroplasties1. The treatment of thirteen palsies led to full recovery, at a mean of five years (range, one to eleven years). The more severe palsies (complete motor and sensory deficits) were associated with a worse prognosis (treatment of only seven of nineteen led to full recovery) than the incomplete lesions (treatment of six of seven led to complete recovery)1. These authors1,35 concluded that the value of operative exploration of the nerve was not known. It is not clear how they reached this conclusion. The ideal approach to postoperative nerve palsy is prevention. An understanding of the anatomy about the knee and meticulous operative technique when performing knee operations can prevent iatrogenic nerve palsy7,22. Anatomical studies have shown that the nerve is at high risk for injury during a proximal tibial osteotomy because of the proximity of the bone to the motor branches. Kirgis and Albrecht, in a study of twenty-nine cadaver legs, showed that the peroneal nerve is closely adherent to the periosteum of the fibula, particularly at the most proximal forty millimeters of the fibula and at points located in the segment of bone between seventy to 153 millimeters from the fibular head22. Therefore, a fibular osteotomy should be performed between forty and seventy millimeters from the fibular head or between the middle and distal thirds of the fibula, about 160 millimeters distal to the proximal tip of the head, to avoid the risk of damage to the peroneal nerve. These high-risk locations correspond to the locations where decompression of the peroneal nerve was performed in our patients. If a peroneal nerve palsy is recognized in the immediate postoperative period, the initial treatment should involve removal of any constrictive dressing to allow for 30 to 40 degrees of flexion of the knee. If there is evidence of an expanding hematoma about the proximal aspect of the fibula, urgent exploration of the wound may be necessary. If there is no clear evidence of a compressive hematoma and dysfunction of the peroneal nerve persists despite the measures just described, some advocate observation for years, as these palsies have been found to resolve with time11,16,35. However, the results of the present study suggest that the prognosis for sensorimotor recovery deteriorates with time and that operative decompression should be considered earlier. In the current study, thirty (97 per cent) of our thirty-one patients had a decrease in the symptoms after exploration of the peroneal nerve and external neurolysis, compared with three of nine patients who had been managed non-operatively (p < 0.01, Fisher exact test). As some of the patients in the latter group might have been found to need cable grafts if exploration had been performed, caution should be exercised in interpreting these data. Although the peroneal nerve palsies in our patients may have resolved with non-operative treatment after an extended period of time, operative decompression spared these patients many months of pain and disability. Twenty-two patients (71 per cent) noted a decrease in the symptoms within three months, and six had a dramatic decrease within four weeks. In addition, early resolution of the palsy facilitated the functional recovery of six patients who had had a knee replacement. Our findings are similar to those of Vastamäeki, who evaluated the results of decompression of the peroneal nerve in twenty-seven extremities at a mean of fourteen months (range, one month to five years) postoperatively39. Relief of paresis and of sensory symptoms was noted in twenty-four of the limbs at a mean of seventeen months (range, four days to eight years) after the onset of the neuropathy. Styf evaluated the results of fasciotomy and neurolysis of the superficial peroneal nerve in twenty-two extremities (nineteen patients)38. At a mean of thirty-seven months (range, ten to fifty months), thirteen patients had unlimited or increased physical function, although only nine were satisfied with the result of the procedure. We recently reported on five patients who had a peroneal nerve palsy after a total knee arthroplasty; decompression of the peroneal nerve was performed five to forty-five months after the onset of the symptoms24. All patients had improved neurological function, and four of them had complete recovery. In the current study, patients who had a complete peroneal nerve palsy (a palsy score of 5 points) were as likely to have full recovery (twelve of fourteen patients) as those who had a partial nerve palsy (a score of 4 points) (thirteen of seventeen patients). These results differ from those of Asp and Rand1, who found that initial profound neurological loss decreased the potential for full recovery. We did not find any relationship between the severity of involvement of the peroneal nerve and the time to improvement. Postoperative electromyographic and nerve-conduction-velocity studies would have more clearly defined the progress of recovery; however, we could not justifiably subject our patients to these studies, given the continuing decrease in the symptoms. The approach to common peroneal nerve palsy should be based on the pathophysiology and natural history of nerve compression and should mirror the treatment of other compressive neuropathies of the peripheral nerves, such as median nerve entrapment at the wrist. We see no qualitative difference between a peroneal palsy associated with a knee injury or a fracture of the leg and a median nerve compression neuropathy associated with a fracture of the distal part of the radius. On the basis of our experience and our review of the literature, we recommend: (1) that motor and sensory function of the common peroneal nerve be evaluated clinically before and after any elective operation on the knee and after any trauma to the lower extremity; (2) that electromyographic and nerve-conduction-velocity studies be conducted if the sensorimotor defect persists for more than three weeks and that the studies be repeated if the defect is still present at three months; (3) that operative exploration be performed during the fourth month of persistent dysfunction of the peroneal nerve (similar to the time-frame suggested by Styf38 and by Vastamäeki39); and (4) that simultaneous decompression of the peroneal nerve be performed in a patient who has a sensorimotor deficit of the nerve and who needs operative intervention for the inciting injury.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |