Sciatic Neuropathy

Published on 03/03/2015 by admin

Filed under Neurology

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 2 (2 votes)

This article have been viewed 6589 times

33 Sciatic Neuropathy

Sciatic neuropathies are uncommon in the electromyography (EMG) laboratory. When they occur, patients often present in a manner similar to that of peroneal neuropathy. Indeed, a footdrop from an early sciatic neuropathy may be difficult or impossible to distinguish clinically from a footdrop from peroneal neuropathy at the fibular neck. It often falls to the electromyographer to make this differentiation. Demonstration of a sciatic neuropathy on EMG has important diagnostic implications because the differential diagnosis is distinctly different from that of other peripheral nerve entrapment syndromes.

Anatomy

The sciatic nerve is derived from the L4–S3 roots, carrying fibers that eventually will become the tibial and common peroneal nerves. It leaves the pelvis through the sciatic notch (greater sciatic foramen) under the piriformis muscle accompanied by the other branches of the lumbosacral plexus (inferior and superior gluteal nerves and posterior cutaneous nerve of the thigh). In some individuals, fibers destined to become the common peroneal nerve run through the piriformis muscle before joining the sciatic nerve. Covered by the gluteus maximus, the sciatic nerve next runs medial and posterior to the hip joint between the ischial tuberosity and the greater trochanter of the femur (Figure 33–1). The knee flexors, including the medial hamstrings (semimembranosus and semitendinosus) and lateral hamstrings (long and short heads of the biceps femoris), and the lateral division of the adductor magnus are all supplied by the sciatic nerve.

image

FIGURE 33–1 Sciatic nerve anatomy.

(From Haymaker, W., Woodhall, B., 1953. Peripheral nerve injuries. WB Saunders, Philadelphia. with permission.)

Within the sciatic nerve, fibers that eventually form the common peroneal nerve often are segregated from those that distally become the tibial nerve. The peroneal division of the sciatic nerve runs lateral to the tibial division. The two divisions physically separate from each other in the mid-thigh to form their respective nerves. All sciatic innervated muscles in the thigh are derived from the tibial division of the sciatic nerve, with the important exception of the short head of the biceps femoris, which is derived from the peroneal division. In essence, the short head of the biceps femoris is the only peroneal-innervated muscle above the level of the fibular neck. This muscle assumes special importance in the EMG evaluation of peroneal palsy, sciatic neuropathy, and other more proximal lesions. As the sciatic nerve terminates in the common peroneal and tibial nerves, it supplies all motor and sensory innervation below the knee, with the exception of sensation over the medial calf and foot (saphenous sensory territory).

Clinical

Sciatic neuropathies caused by trauma, injection, infarction, or compression present acutely. Otherwise, most sciatic neuropathies present in a progressive, subacute fashion. Patients with a complete sciatic neuropathy have paralysis of knee flexion and all movements about the ankle and toes. Sensation is lost in several areas (Figure 33–2), including the lateral knee (lateral cutaneous nerve of the knee), lateral calf (superficial peroneal nerve), dorsum of the foot (superficial peroneal nerve), web space of the great toe (deep peroneal nerve), posterior calf and lateral foot (sural nerve), and sole of the foot (distal tibial nerve). Pain may be perceived in the proximal thigh, radiating posteriorly and laterally into the leg, but it usually does not affect the back. The ankle reflex is depressed or absent on the involved side.

image

FIGURE 33–2 Sensory loss in sciatic neuropathy (in green).

(Adapted from Haymaker, W., Woodhall, B., 1953. Peripheral nerve injuries. WB Saunders, Philadelphia, with permission.)

This complete deficit is seen only in severe lesions or late in the course of sciatic neuropathy. Initially, the clinical presentation most often mimics peroneal neuropathy. It has long been recognized that the peroneal fibers are preferentially affected in most sciatic nerve lesions. Thus, it is not unusual for a patient with sciatic neuropathy to present with a footdrop and sensory disturbance over the dorsum of the foot and lateral calf. Indeed, early sciatic nerve lesions may be nearly impossible to differentiate clinically from peroneal nerve lesions at the fibular neck (Table 33–1).

On physical examination, close attention must be paid to muscles that receive non-peroneal innervation, especially ankle inversion (tibialis posterior–tibial nerve), toe flexion (flexor digitorum longus–tibial nerve), and knee flexion (hamstring muscles–sciatic nerve). Weakness in any of these muscles in a patient with a footdrop suggests dysfunction beyond the peroneal nerve distribution. Likewise, on sensory examination, any sensory disturbance over the lateral knee, lateral foot, or sole of the foot suggests a lesion of the sciatic or tibial nerves or more proximally. Isolated sciatic nerve lesions spare sensation over the medial calf and foot (saphenous nerve) and posterior thigh (posterior cutaneous nerve of the thigh). Any involvement of these territories in a patient with a footdrop suggests a more widespread lesion, either in the lumbosacral plexus or proximally.

It is important to remember that in addition to sciatic neuropathy and peroneal neuropathy, a footdrop with sensory disturbance over the lateral calf and dorsum of the foot may occur in lumbosacral plexopathy, radiculopathy (especially L5), or even a central lesion, such as a frontal meningioma or anterior cerebral artery infarct.

Etiology

Sciatic neuropathy is distinctly uncommon and is associated with a limited differential diagnosis (Box 33–1). As the sciatic nerve runs posterior to the hip joint, one of the most common presentations occurs following hip or femur fracture (especially posterior dislocation) or as a complication of the subsequent surgery to repair the fracture. As a complication of surgery, sciatic neuropathy may occur due to retraction or stretch, as well as a result of methylmethacrylate cement forming spurs and then eroding into the nerve months to years later, which has been well documented in several case reports.

Another common cause of sciatic neuropathy is tumor (neurofibroma, schwannoma, neurofibrosarcoma, lipoma, and lymphoma). Tumors affecting the sciatic nerve usually can be imaged quite well as a mass lesion on computed tomography or magnetic resonance imaging (MRI) scanning (Figure 33–3). Other rare mass lesions also may affect the sciatic nerve. An enlarged Baker’s cyst in the popliteal fossa may compress the distal sciatic nerve as it bifurcates into the tibial and common peroneal nerves. Several unusual vascular abnormalities, including aneurysms of the inferior gluteal, iliac, or persistent sciatic arteries and arteriovenous malformations near the piriformis muscle, have been associated with sciatic neuropathy.

Damage to the sciatic nerve can occur from trauma or as a result of a penetrating injury, such as gunshot and knife wounds. Sciatic neuropathy also may occur as a complication of immobilization and external compression, such as during anesthesia, coma, or intoxication. In the hospital setting, damage to the sciatic nerve may occur iatrogenically from misplaced intramuscular buttock injections, especially in thin patients.

Disorders that result in a mononeuritis multiplex syndrome (see Chapter 26) may affect the sciatic nerve. For example, vasculitic neuropathy commonly results in infarction of the sciatic nerve in the proximal thigh, which is a watershed area for nerve ischemia. The neuropathy often is acute and begins with prominent pain. Until additional nerve lesions develop, recognition of the underlying mononeuritis multiplex pattern is difficult or impossible.

Piriformis Syndrome

As the sciatic nerve leaves the pelvis, it runs under or through the piriformis muscle (Figure 33–4). The piriformis muscle originates from the sacrum, the sciatic notch and the sacrotuberous ligament, and then runs through the greater sciatic foramen to attach to the greater trochanter of the femur. The main action of the piriformis is to externally rotate the hip. When the hip is in a flexed position, it also acts as a partial hip abductor. Theoretically, a hypertrophied piriformis muscle could compress the sciatic nerve (piriformis syndrome), somewhat comparable to compression of the median nerve by the pronator teres muscle in pronator teres syndrome. In the past, many cases of “sciatica” were attributed to piriformis syndrome. However, most, if not all, cases of sciatica are due to lumbosacral radiculopathy and not sciatic neuropathy from piriformis syndrome. Piriformis syndrome is considered by many to be a controversial entity. There are very few reported cases of patients who meet the criteria for definite piriformis syndrome, which include (1) sciatic neuropathy clinically, (2) electrophysiologic evidence of sciatic neuropathy, (3) surgical exploration showing entrapment of the sciatic nerve within a hypertrophied piriformis muscle, and (4) subsequent improvement following surgical decompression.

Clinically, piriformis syndrome should be suspected when a patient has more pain while sitting than standing; worsening of symptoms with flexion, adduction, and internal rotation of the hip; a history of trauma or unusual body habitus (especially very thin); and tenderness in the mid-buttock that reproduces the pain and paresthesias. Several physical examination maneuvers are reported to be useful in suspected piriformis syndrome. In each, the piriformis muscle is either stretched or voluntarily contracted. Pain from the buttock down the sciatic nerve, but without any back pain, is said to be consistent with piriformis syndrome. These maneuvers include:

Electrophysiologic Evaluation

The electrophysiologic evaluation plays a key role in the assessment of a possible sciatic neuropathy. The electrophysiologic approach is similar to the clinical approach: evaluate and exclude disorders that can mimic sciatic neuropathy, including peroneal palsy at the fibular neck, lumbosacral plexopathy, and lumbosacral radiculopathy (Table 33–2).

Nerve Conduction Studies

The nerve conduction evaluation of sciatic neuropathy is straightforward (Box 33–2). Routine peroneal and tibial motor studies should be performed bilaterally, recording the extensor digitorum brevis (EDB) and abductor hallucis brevis, respectively. Careful attention must be paid to the peroneal motor study, with the electromyographer looking for evidence of peroneal palsy at the fibular neck (either focal slowing or conduction block). In this regard, it is useful to perform peroneal motor studies recording the tibialis anterior as well as the extensor digitorum brevis. In sciatic nerve lesions with axonal loss, the amplitude of the peroneal or tibial compound muscle action potentials (CMAPs) may be reduced on the symptomatic side compared with normal control values or, more importantly, when compared with the contralateral asymptomatic leg. The peroneal fibers often are affected out of proportion to the tibial fibers. If there has been loss of the fastest conducting axons, there may be mild prolongation of the distal motor latency and some slowing of conduction velocity, but never into the demyelinating range.

Bilateral peroneal and tibial F responses and H reflexes should be obtained. In sciatic neuropathy, ipsilateral F wave responses may be prolonged compared with the contralateral side. In a sciatic nerve lesion, the H reflex may be prolonged or more difficult to elicit on the involved side. Although abnormal late responses place the lesion somewhere along the course of the nerve fibers being studied, the finding of prolonged or absent F and H responses cannot help in differentiating among a sciatic neuropathy, lumbosacral plexopathy, or radiculopathy. A proximal lesion is implied only if the distal conductions are normal.

Likewise, sensory nerve conduction studies must be performed bilaterally, comparing the superficial peroneal and sural sensory responses to the contralateral side. In sciatic neuropathy, both responses are expected to be abnormal, reflecting dysfunction of both the peroneal and tibial nerves. However, as noted earlier, the peroneal fibers are often the most affected.

Special Studies in Suspected Piriformis Syndrome

Most often, standard nerve conduction studies and needle EMG are normal in patients who are clinically diagnosed with piriformis syndrome. The one electrophysiologic test proposed to be of value is a modification of the H reflex. In piriformis syndrome, the H reflex is reported to be prolonged when performed with the hip in flexion, adduction, and internal rotation (FAIR test) compared to the normal anatomic position (Figure 33–5). This position stretches the piriformis muscle and theoretically may put pressure on the sciatic nerve.

In the largest reported study of this test, in patients with clinical criteria suggestive of piriformis syndrome, the mean prolongation of the H reflex in the FAIR position was 3.39 ms, which is equivalent to 5.45 standard deviations above the mean for a normal population. Compare this to the mean delay of the H reflex in 88 normal persons in the FAIR position compared to the anatomic position, which was 0.01 ms, with a standard deviation of 0.62 ms (Figure 33–6). However, the asymptomatic population was not normally distributed. Using a cutoff of 3 standard deviations (1.86) resulted in a specificity of 83% (i.e., 17% of a normal control population would be misidentified as abnormal). In addition, the contralateral, asymptomatic limbs of the patient group often demonstrated abnormalities, although they were less marked than in the symptomatic limbs.

The authors have little personal experience with the FAIR test. Other so-called dynamic nerve conduction tests generally fail to increase the yield of abnormalities in entrapment neuropathies (e.g., flexing the wrist in carpal tunnel syndrome while performing median nerve conduction studies), although this is not always true. In addition, the H reflex is well known to be affected by a variety of variables, including body and especially head position. Because the circuitry of the H reflex traverses the spinal cord, it can be modified by a variety of suprasegmental facilitatory and inhibitory inputs. For instance, the Jendrassik (reinforcement) maneuver is commonly used to “prime” the anterior horn cells and is of use in the EMG laboratory to elicit H reflexes. Presumably, head position can modify the H reflex by activating the vestibulospinal tracts. The take-home message is the following: if the FAIR test is used in patients with suspected piriformis syndrome, ensure that other variables are held constant, especially the head and body positions; and remember the possibility of false-positive results, given the distribution of the values from a control population.

Electromyographic Approach

After the nerve conduction studies are completed, EMG is used to further localize the lesion and assess its severity (Box 33–3). First, muscles innervated by the deep and superficial peroneal nerves should be sampled (e.g., tibialis anterior, extensor hallucis longus, peroneus longus). Abnormalities in these muscles are consistent with a lesion of the peroneal nerve, sciatic nerve, lumbosacral plexus, or L5–S1 nerve roots. Next, tibial-innervated muscles in the calf should be sampled, including the medial gastrocnemius and especially the tibialis posterior or flexor digitorum longus. If abnormalities are found in any of these muscles, as well as in the peroneal-innervated muscles, an isolated lesion of the peroneal nerve has been excluded. The differential at this point includes a lesion of both the tibial and peroneal nerves versus a lesion of either the sciatic nerve, lumbosacral plexus, or L5–S1 nerve roots.

Next, the hamstring muscles need to be sampled. The short head of the biceps femoris has an important role, being the only muscle supplied by the peroneal division of the sciatic nerve that originates above the fibular neck. The short head of the biceps can easily be sampled four fingerbreadths above the lateral knee, just medial to the long head of the biceps femoris tendon. Abnormalities found in the short head of the biceps femoris muscle exclude an isolated lesion of the peroneal nerve at the fibular neck and imply a more proximal lesion. After examination of the hamstring muscles, the gluteal muscles should be checked. Both the gluteus maximus (inferior gluteal nerve) and either the gluteus medius or tensor fascia latae (superior gluteal nerve) should be checked. If abnormalities are found in any of these muscles, an isolated sciatic neuropathy is excluded, and the differential diagnosis at this point is restricted to a lesion of the lumbosacral plexus or the L5–S1 nerve roots. Next, the L5 and S1 paraspinal muscles must be sampled to look for abnormalities at or proximal to the root level. Lastly, if any of the muscles studied during the needle EMG examination show borderline or equivocal abnormalities, comparison to the contralateral side is indicated.

It is important to emphasize that the EMG study can only localize a lesion at or proximal to the most proximal abnormal muscle sampled. For instance, in examining the hamstring muscles, if the semitendinosus muscle is abnormal and the semimembranosus muscle is normal, one would be tempted to assume that the sciatic nerve lesion lies between these two sites. The situation is not that simple, however. It is well known from evaluation of various compressive neuropathies that fascicles to certain muscles can be preferentially affected, whereas others are preferentially spared. Thus, in the earlier example, the lesion could even be at the level of the nerve roots, sparing fascicles to the semimembranosus. Accordingly, EMG can be used only to identify a lesion at or proximal to the most proximal muscle involved.

The classic electrophysiologic picture of sciatic neuropathy is reduced tibial and peroneal motor amplitudes compared with the contralateral side, with normal or slightly prolonged distal motor latencies and normal or slightly slowed conduction velocities. The tibial and peroneal F responses are prolonged or absent on the symptomatic side, with similar findings for the H reflex. Both the sural and superficial peroneal sensory nerves are reduced in amplitude or absent with normal potentials on the contralateral asymptomatic side. Needle EMG findings show active denervation or reinnervation with reduced recruitment of motor unit action potentials (MUAPs) in muscles supplied by (1) the sciatic nerve in the thigh, (2) the peroneal nerve, and (3) the tibial nerve, but with sparing of the gluteal, tensor fascia latae, and lumbosacral paraspinal muscles. In both the nerve conduction and needle EMG studies, the peroneal fibers are involved more often than the tibial fibers.

image Example Case

image Case 33–1

History and Physical Examination

A 52-year-old woman was referred for further evaluation of a persistent left footdrop. She described her condition as having begun slowly 6 months previously. She initially noted a sensation of numbness over the top of the foot and the lateral calf. This was followed shortly thereafter by her left foot dropping. During the last 2 months, symptoms slowly progressed to a nearly complete footdrop. More recently, she noted a sensation of tightness and pain from her hip down to her knee and into her calf.

An orthopedic consultant advised MRI scanning of the knee to evaluate the peroneal nerve. The scan was obtained and was unremarkable. She subsequently underwent MRI scanning of the lumbar spine to look for a possible L5 radiculopathy as the cause of her footdrop. The scan was obtained and was unremarkable. Past history was notable for a left hip fracture with surgical repair 3 years previously.

On examination, there was atrophy of the anterior compartment of the left leg and wasting of the left EDB muscle. In the left lower extremity there was a complete footdrop. Toe and ankle dorsiflexion were 1/5, as was ankle eversion. Ankle inversion also was weak (4/5). In addition, toe flexion was slightly but definitely weak, as was knee flexion. Knee extension was normal. Hip flexion, extension, abduction, and adduction were completely normal. Strength testing was completely normal in the right lower extremity. Deep tendon reflexes were 2+ and symmetric in the upper extremities and 2+ at the knees and right ankle. The left ankle jerk was absent. Toes were downgoing. There was a clear sensory disturbance to light touch on the top of the foot, lateral foot and calf, lateral knee, and posterior calf on the left side. Sensation over the medial calf, anterior thigh, lateral thigh, posterior thigh, and sole of the foot was intact. There was a well-healed surgical scar over the left lateral thigh.

Summary

The initial clinical presentation is that of a footdrop with numbness over the dorsum of the foot and lateral calf. Most often, this clinical picture is the result of a peroneal neuropathy at the fibular neck. However, an early sciatic neuropathy, lumbosacral plexopathy, or lumbosacral radiculopathy (especially L5) can present in a similar fashion. The slowly progressive nature of the symptoms suggests a slowly expanding or infiltrating structural lesion. As the symptoms progressed, the patient noted a sensation of tightness and pain from the hip toward the knee into the calf. These additional symptoms would be unusual for a peroneal palsy at the fibular neck and are suggestive of a more proximal lesion. MRI scanning obtained at the usual sites of compression causing a footdrop (the fibular neck and lumbar spine) did not demonstrate any abnormality and led to further evaluation and eventually an EMG study.

Neurologic examination showed severe weakness and atrophy in the distribution of the deep and superficial peroneal nerves (ankle and toe dorsiflexion, ankle eversion). Ankle inversion (tibialis posterior) and toe flexion (flexor digitorum longus), both of which are subserved by non-peroneal-innervated L5 muscles, were also weak. In addition, there was weakness of knee flexion, which is subserved by the sciatic nerve. These findings place the lesion at or proximal to the sciatic nerve. Further testing of muscles innervated by the femoral, superior gluteal, inferior gluteal, and obturator nerves were normal. The absence of abnormalities in these muscles on clinical examination suggested that a more widespread lesion of the lumbosacral plexus or nerve roots was unlikely. Of course, early in any lesion, it may be difficult to demonstrate subtle weakness of the proximal limb muscles.

Moving on with the clinical examination, the left ankle reflex was absent, signifying a lesion somewhere along that reflex loop, in the tibial nerve, sciatic nerve, lumbosacral plexus, or lumbosacral nerve roots. Lastly, the sensory disturbance involved not only the distribution of the peroneal nerve but also the territories of the sural nerve and the lateral cutaneous nerve of the knee. Normal sensation was found in the medial calf, innervated by the saphenous nerve, the anterior thigh, innervated by the femoral nerve, the lateral thigh, innervated by the lateral cutaneous nerve of the thigh, and the posterior thigh, innervated by the posterior cutaneous nerve of the thigh. This distribution of sensory abnormalities again suggests a lesion at or proximal to the sciatic nerve. However, note that the entire sciatic sensory territory was not involved because sensation on the sole of the foot was spared (innervated by the plantar nerves).

Before proceeding to the nerve conduction study and EMG findings, the clinical history of a slowly progressive deficit, along with the neurologic examination as described, should suggest a slowly expanding or infiltrating structural lesion affecting the sciatic nerve, the lumbosacral plexus, or the lumbosacral roots. The history of prior hip surgery should suggest a likely connection between the surgery and a possible sciatic nerve palsy.

Reviewing the nerve conduction studies first, the motor nerve conduction studies in the left leg are abnormal, with borderline low CMAP amplitudes for both the peroneal and tibial motor studies. Furthermore, a clear asymmetry is seen when the potentials are compared with those from the contralateral, asymptomatic side. The tibial distal motor latency, minimum F response latency, and tibial and peroneal conduction velocities are slightly slowed. However, the amount of slowing is mild, within the range of axonal loss. Of note, there is no focal drop in amplitude or focal conduction velocity slowing in the peroneal nerve around the fibular neck. Note that for the peroneal motor studies, both the extensor digitorum brevis and tibialis anterior muscles were recorded. There are some cases of peroneal neuropathy at the fibular neck wherein conduction block and/or slowing is only seen when recording the tibialis anterior.

Moving next to the sensory nerve conduction studies, both the sural and superficial peroneal sensory studies are abnormal on the symptomatic side compared with the normal findings on the contralateral side. The superficial peroneal response is absent, whereas the sural response is only borderline low, reflecting greater involvement of peroneal, compared to tibial, nerve fibers. Finally, on the nerve conduction studies, the H reflex is absent on the ipsilateral side, corresponding to the absent ankle reflex on the clinical examination.

Thus, at the conclusion of the nerve conduction studies, there is a good clinical–electrophysiologic correlation. The muscle atrophy and weakness seen on clinical examination correspond to the low CMAP amplitudes on the peroneal and tibial motor studies. Likewise, the areas of sensory loss on clinical examination correspond to the distribution of reduced sensory nerve action potentials. Both clinical examination and electrophysiologic studies demonstrate that the peroneal nerve fibers are more involved than the tibial.

Moving next to the needle EMG study, there is marked active denervation and reinnervation in muscles innervated by the superficial and deep peroneal nerves. These prominent abnormalities correspond to the patient’s clinical symptoms of footdrop. In contrast, the medial gastrocnemius (tibial nerve innervated) is normal. However, the tibialis posterior, another tibial-innervated muscle, shows fibrillation potentials and large polyphasic MUAPs with decreased recruitment. These findings provide further evidence that the abnormalities are beyond the peroneal nerve territory and must be due to either separate lesions of the tibial and peroneal nerves or a more proximal lesion.

Next, the short head of the biceps femoris is sampled. This muscle assumes special significance on the EMG examination because it is the only peroneal-innervated muscle that originates above the fibular neck. This muscle is normal in peroneal palsy at the fibular neck, but it may be abnormal in lesions at or proximal to the sciatic nerve. In this case, the short head of the biceps femoris has fibrillation potentials with reduced recruitment of large polyphasic MUAPs. Similar but less marked findings are found in the long head of the biceps femoris. The semitendinosus muscle, which is also innervated by the sciatic nerve, is normal. No abnormalities are found in the more proximal hip girdle muscles, which are innervated by the superior and inferior gluteal nerves (gluteus medius and maximus). Similarly, muscles innervated by the femoral nerve (vastus lateralis and iliacus) and the L5 and S1 paraspinal muscles are normal. At this point, we are ready to formulate our electrophysiologic impression.

Although the patient’s initial symptoms suggested a simple peroneal palsy at the fibular neck, the subsequent clinical findings suggested a more proximal lesion, which was then confirmed with nerve conduction study and EMG findings. The abnormal sensory conduction studies mark the lesion as at or distal to the dorsal root ganglion, which is inconsistent with a disorder of the L5 or S1 nerve roots. Because both the superficial peroneal and sural sensory responses were abnormal, the lesion must be in the tibial and peroneal nerves, the sciatic nerve, or the lumbosacral plexus. The needle EMG findings also demonstrated abnormalities outside the peroneal distribution, involving the tibial and distal sciatic nerves. Several important questions can be addressed at this point.

Suggested Readings

Chiao H.C., Marks K.E., Bauer T.W., et al. Intraneural lipoma of the sciatic nerve. Clin Orthop. 1987;221:267.

Cusimano M.D., Shedden P.M., Hudson A.R., et al. Arteriovenous malformation of the pyriformis muscle manifesting as a sciatic nerve tumor. Neurosurgery. 1992;31:151.

Edwards M.S., Barbaro N.M., Asher S.W., et al. Delayed sciatic palsy after total hip replacement: case report. Neurosurgery. 1981;9:61.

Eusebi V., Bondi A., Cancellieri A., et al. Primary malignant lymphoma of sciatic nerve. Report of a case. Am J Surg Pathol. 1990;14:881.

Fishman L.M., Dombi G.W., Michaelsen C., et al. Piriformis syndrome: diagnosis, treatment, and outcome – a 10-year study. Arch Phys Med Rehabil. 2002;83:295–301.

Fishman L.M., Zybert P.A. Electrophysiological evidence of piriformis syndrome. Arch Phys Med Rehabil. 1992;73:359–364.

Gasecki A.P., Ebers G.C., Vellet A.D., et al. Sciatic neuropathy associated with persistent sciatic artery. Arch Neurol. 1992;49:967.

Kirschner J.S., Foye P.M., Cole J.L. Piriformis syndrome, diagnosis and treatment. Muscle Nerve. 2009;40:10–18.

Mohan S.R., Grimley R.P. Common iliac artery aneurysm presenting as acute sciatic nerve compression. Postgrad Med J. 1987;63:903.

Papadopoulos S.M., McGillicuddy J.E., Messina L.M. Pseudoaneurysm of the inferior gluteal artery presenting as sciatic nerve compression. Neurosurgery. 1989;24:926.

Pillay P.K., Hardy R.W., Jr., Wilbourn A.J., et al. Solitary primary lymphoma of the sciatic nerve: case report. Neurosurgery. 1988;23:370.

Sieb J.P., Schultheiss R. Segmental neurofibromatosis of the sciatic nerve: case report. Neurosurgery. 1992;31:1122.

Stewart J.D., Fishman L.M., Schaefer M.P. Issues & opinions: piriformis syndrome. Muscle Nerve. 2003;11:644–649.

Stillman M.J., Christensen W., Payne R., et al. Leukemic relapse presenting as sciatic nerve involvement by chloroma (granulocytic sarcoma). Cancer. 1988;62:2047.

Yeun E.C., Olney R.K., So Y.T. Sciatic neuropathy: clinical and prognostic features in 73 patients. Neurology. 1994;44:1669.

Yuen E.C., So Y.T. Sciatic neuropathy. Neurol Clin. 1999;17:617–631.

Young J.N., Friedman A.H., Harrelson J.M., et al. Hemangiopericytoma of the sciatic nerve. Case report. J Neurosurg. 1991;74:512.