Neurosurgical forum Letters to the editor

Piriformis Syndrome could not be established by routine spine imaging, discog- raphy would be performed and, if the study were positive, TO THE EDITOR: I congratulate Filler, et al., for their lumbar fusion would proceed. Before subjecting a patient chapter-length original work on the piriformis syndrome. to an operation of such magnitude, it seems reasonable to (Filler AG, Haynes J, Jordan SE, Prager J, Villablanca JP, rule out piriformis syndrome unless the symptoms of me- Farahani K, et al: of nondisc origin and pirifor- chanical far outweigh the symptoms of leg pain. mis syndrome: diagnosis by magnetic resonance neurogra- My only criticism of their work is the clinical maneuver phy and interventional magnetic resonance imaging with they describe for testing pain status. Their maneuver is outcome study of resulting treatment. J Neurosurg Spine 2: conducted while the patient is supine and the painful leg 99–115, February 2005). is crossed over the contralateral leg with the foot lateral to Abstract the knee of the normal leg. The patient adducts the knee against the resistance of the examiner’s hand. This maneu- Object. Because lumbar magnetic resonance (MR) imaging fails to ver stretches the either with or without identify a treatable cause of chronic sciatica in nearly 1 million pa- resistance, and I do not understand what the resistance ac- tients annually, the authors conducted MR neurography and interven- complishes. Both the muscle and the are tional MR imaging in 239 consecutive patients with sciatica in whom stretched, forcing the nerve against the osseous margin of standard diagnosis and treatment failed to effect improvement. the ischium, which appears to be to be a nonspecific sciatic Methods. After performing MR neurography and interventional MR imaging, the final rediagnoses included the following: piri- nerve test. Finally, this maneuver unnecessarily compress- formis syndrome (67.8%), distal foraminal nerve root entrapment es pelvic contents, produces some shearing stress to the (6%), ischial tunnel syndrome (4.7%), discogenic pain with referred , and distorts the hip joint and capsule in an leg pain (3.4%), pudendal nerve entrapment with referred pain (3%), unnatural configuration, all of which introduce potential al- distal sciatic entrapment (2.1%), sciatic tumor (1.7%), lumbosacral ternative explanations for pain.2 plexus entrapment (1.3%), unappreciated lateral disc herniation A number of years ago I described a maneuver by which (1.3%), nerve root due to spinal surgery (1.3%), inadequate to diagnose the piriformis syndrome. It is performed with spinal nerve root decompression (0.8%), lumbar stenosis (0.8%), the patient on his or her side with the painful leg superior sacroiliac joint inflammation (0.8%), lumbosacral plexus tumor 1 (0.4%), sacral fracture (0.4%), and no diagnosis (4.2%). and the thigh and knee slightly flexed. In this position the piriformis muscle, which is usually a thigh external rotator, Open MR–guided Marcaine injection into the piriformis muscle 3 produced the following results: no response (15.7%), relief of great- becomes a thigh abductor. The patient is asked to lift the er than 8 months (14.9%), relief lasting 2 to 4 months with continu- leg against gravity or against resistance of the examiner’s ing relief after second injection (7.5%), relief for 2 to 4 months with hand. This maneuver contracts the swollen piriformis mus- subsequent recurrence (36.6%), and relief for 1 to 14 days with full cle and produces beep buttock pain and usually leg pain recurrence (25.4%). Piriformis surgery (62 operations; 3-cm inci- as described by Filler, et al. It is this maneuver by which sion, transgluteal approach, 55% outpatient; 40% with local or epi- we turn in bed from side to back and, in fact, has been dural anesthesia) resulted in excellent outcome in 58.5%, good out- a useful historic clue in my experience. Filler and col- come in 22.6%, limited benefit in 13.2%, no benefit in 3.8%, and leagues have shown that muscle spasm produces sciatic worsened symptoms in 1.9%. Conclusions. This Class A quality evaluation of MR neurography’s nerve compression, and I would agree that contracting the diagnostic efficacy revealed that piriformis muscle asymmetry and swollen muscle in abduction does precisely this. sciatic nerve hyperintensity at the sciatic notch exhibited a 93% speci- ROBERT A. BEATTY, M.D. ficity and 64% sensitivity in distinguishing patients with piriformis Hinsdale, Illinois syndrome from those without who had similar symptoms (p Ͻ 0.01). Evaluation of the nerve beyond the proximal foramen provided eight additional diagnostic categories affecting 96% of these pa- References tients. More than 80% of the population good or excellent function- al outcome was achieved. 1. Beatty RA: The piriformis muscle syndrome: a simple diagnostic maneuver. Neurosurgery 34:512–514, 1994 2. Garvin KL, McKillip TM: History and physical examination, in The usefulness of MR neurography of the sciatic nerve Callaghan JJ, Rosenberg AG, Rubash HE (eds): The Adult Hip. and of open MR imaging–guided injections of the piri- Philadelphia: Lippincott-Raven, 1998, pp 315–332 formis muscle with Marcaine and celestone for diagnos- 3. Salmons S: Muscle, in Bannister LH, Berry MM, Collins P, Dy- tic purposes and treatment are convincingly presented. son M, Dusek JE, Ferguson MWJ (eds): Gray’s Anatomy, ed Their minimally invasive section of the piriformis mus- 38. New York: Churchill-Livingston, 1995, pp 737–900 cle is nicely described. Their most important contribution, however, is bringing to our attention the frequency of this RESPONSE: In his letter, Dr. Beatty points out a useful previously underdiagnosed syndrome whose existence has physical examination test for piriformis syndrome that he been questioned. Considering the number of patients with had published previously1 and underscores some possible sciatica in whom imaging results are nondiagnostic, they pitfalls that it may avoid. We have had the opportunity to believe the piriformis syndrome may be as common a validate fully the maneuver described in our paper. In our cause of sciatica as herniated lumbar discs. One suspects, study, more than 95% of patients were able to place their in many past cases, that when the presence of sciatica leg into the crossed-leg position shown in our Fig. 7 with-

J. Neurosurg: Spine / Volume 5 / July, 2006 101 Neurosurgical forum out difficulty or immediate symptoms. The test is only pos- imaging, clinical, electrodiagnostic, and intraoperative findings. itive if the usual pain commences upon resisted abduction Neurosurgery 37:1097–1103, 1995 or adduction. This maneuver appears to do an excellent job 3. Britz GW, Haynor DR, Kuntz C, Goodkin R, Gitter A, Maravilla of isolating and relaxing the spine so that the test is di- K, et al: entrapment at the elbow: correlation of mag- netic resonance imaging, clinical, electrodiagnostic, and intraop- rected primarily at the piriformis muscle. It also provides a erative findings. Neurosurgery 38:458–465, 1996 good contrast with the finding of pain evoked by passive 4. Jarvik JG, Yuen E, Haynor DR, Bradley CM, Fulton-Kehoe D, movement of the hip into positions of internal and external Smith-Weller T, et al: MR nerve imaging in a prospective cohort rotation. In our study, patients were subjected to the test of patients with suspected . Neurology with passive motion to check for any hip joint disorder im- 58:1597–1602, 2002 mediately before testing for pain with resisted abduction and adduction. TO THE EDITOR: The article by Filler, et al., (Filler AG, As we reported, in only 63% of our patients with proven Haynes J, Jordan SE, et al: Sciatica of nondisc origin and piriformis syndrome was the test positive for the resisted piriformis syndrome: diagnosis by magnetic resonance adduction or abduction maneuver, and thus the sensitiv- neurography and interventional magnetic resonance imag- ity does not seem overly high. In a review of our data we ing with outcome study of resulting treatment. J Neurosurg found a false-positive rate for this maneuver of less than Spine 2:99–115, February 2005) represents a renewed and 5%, although the causes for the false positives are not clear. technologically revitalized but ultimately flawed effort to In any case, this examination maneuver must be consid- demonstrate that the piriformis muscle is the cause of all ered together with the patient’s history, other examination sciatica unrelieved by a disc surgery or undiagnosed by maneuvers, imaging findings, and diagnostic injections to routine examination and imaging evaluation. determine whether to perform surgical decompression. The piriformis muscle has an unfortunate but well-doc- Our paper was intended to take an important step toward umented history of physicians unintentionally transform- confirming the existence of piriformis syndrome and pro- ing its normal muscular anatomy and variations into nerve viding more information about the condition. It will cer- compressive and pathological entities. Physicians have tainly be interesting to compare these two examination ma- been trying to explain and understand all the phenomena of neuvers—ours and that of Dr. Beatty—to determine which sciatica for more than a century. As the authors recognize, is more predictive of a treatable piriformis syndrome. Con- the notion of “piriformis syndrome” was postulated be- firmation of the existence of piriformis syndrome in our fore the current understanding of sciatica was described by report was ultimately defined by clinical outcome—that is, Mixter and Barr.10 In a curious twist, in his original paper good to excellent relief from piriformis surgery. in 1934, Freiberg2 acknowledged Drs. Debakey, Gage, and Magnetic resonance neurography showed a 93% spec- Ochsner for their seminal work in the release of the anteri- ificity for the findings of ipsilateral nerve hyperintensity or scalene muscle—the contested and piriformis muscle asymmetry in patients with pirifor- of today—for his own conceptualization of the piriformis mis syndrome. No other test, such as intraoperative NAP syndrome. There exists, however, a group of patients with recording, is known to correlate with the type of mild to sciatica for whom no explanation can be found. In these moderate irritative nerve disease that seems to underlie this patients normal neurological status is unremarkable, with condition. A number of researchers have documented the normal strength and reflexes and, unless the individual cas- relationship between nerve hyperintensity upon MR neu- es are culled from a group of patients with failed–back rography and electrodiagnostic abnormalities in carpal tun- surgery syndrome, (EMG) and nerve nel syndrome and cubital tunnel syndrome.2,3,4 Analysis of conduction studies usually reveal normal findings. For this these findings suggests that the two examination maneu- group, physicians have been unable to find any clear and vers could be reasonably tested in patients with an imag- convincing peripheral nerve pathological evidence in all ing-based diagnosis of piriformis syndrome. but a small number of cases, even fewer if the rigorous cri- Physical examination maneuvers exhibiting improved teria of Stewart15 are applied. sensitivity and specificity for piriformis syndrome are cer- Filler and colleagues inform us that most cases of failed– tainly desirable. The use of MR neurography imaging and back surgery syndrome are actually piriformis syndrome, injection tests, however, will most likely continue to play a which can be cured either with two MR imaging–guided critical role in the diagnosis of piriformis syndrome and in Marcaine injections or imaging–guided piriformis muscle differentiating among the several different types of pelvic excision. sciatic nerve entrapment described in our report. The cornerstone of clinical neurosurgery is the patient AARON FILLER, M.D., PH.D. history and physical examination. Filler and colleagues re- Institute for Nerve Medicine port that 239 detailed neurological examinations were per- Santa Monica, California formed, but they provide a summary of the data in only 162 Peripheral Nerve Program patients. The other 77 patients were then excluded from Institute for Spinal Disorders Cedars Sinai Medical Center evaluation but not from comparison. In summarizing data Los Angeles, California obtained in these 162 patients, the authors report on only two patients with any weakness (foot and leg). Although the authors allude to ancillary studies, they do not detail the References results of any electrophysiological abnormalities in these 1. Beatty RA: The piriformis muscle syndrome: a simple diagnostic patients. As a result, these cases represent patients with maneuver. Neurosurgery 34:512–514, 1994 pain in whom there is no significant recognized evidence 2. Britz GW, Haynor DR, Kuntz C, Goodkin R, Gitter A, Kliot of nerve injury. M: Carpal tunnel syndrome: correlation of magnetic resonance Sixty patients in the study had moderate to severe low-

102 J. Neurosurg: Spine / Volume 5 / July, 2006