Piriformis Syndrome Is Overdiagnosed 11 Robert A

Home , Carpal tunnel syndrome, Piriformis syndrome, Plantar nerve, Sciatica, Tarsal tunnel, Tarsal tunnel syndrome

American Association of Neuromuscular & Electrodiagnostic Medicine AANEM

CROSSFIRE: CONTROVERSIES IN NEUROMUSCULAR AND ELECTRODIAGNOSTIC MEDICINE

Loren M. Fishman, MD, B.Phil Robert A.Werner, MD, MS Scott J. Primack, DO Willam S. Pease, MD Ernest W. Johnson, MD Lawrence R. Robinson, MD

2005 AANEM COURSE F AANEM 52ND Annual Scientific Meeting Monterey, California

CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine

Loren M. Fishman, MD, B.Phil Robert A.Werner, MD, MS Scott J. Primack, DO Willam S. Pease, MD Ernest W. Johnson, MD Lawrence R. Robinson, MD

2005 COURSE F AANEM 52nd Annual Scientific Meeting Monterey, California AANEM

PRINTED BY JOHNSON PRINTING COMPANY, INC. ii CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine

Faculty

Loren M. Fishman, MD, B.Phil Scott J. Primack, DO Assistant Clinical Professor Co-director Department of Physical Medicine and Rehabilitation Colorado Rehabilitation and Occupational Medicine Columbia College of Physicians and Surgeons Denver, Colorado New York City, New York Dr. Primack completed his residency at the Rehabilitation Institute of Dr. Fishman is a specialist in low back pain and sciatica, electrodiagnosis, Chicago in 1992. He then spent 6 months with Dr. Larry Mack at the functional assessment, and cognitive rehabilitation. Over the last 20 years, University of Washington. Dr. Mack, in conjunction with the Shoulder he has lectured frequently and contributed over 55 publications. His most and Elbow Service at the University of Washington, performed some of the recent work, Relief is in the Stretch: End Back Pain Through Yoga, and the original research utilizing musculoskeletal ultrasound in order to diagnose earlier book, Back Talk, both written with Carol Ardman, were published shoulder pathology. After leaving the University of Washington, Dr. in 2005 and 1997, respectively, by W. W. Norton. At Albert Einstein Primack founded Colorado Rehabilitation and Occupational Medicine. College of Medicine, Dr. Fishman was Chief Resident in the Department He has collaborated on multiple research projects analyzing musculoskele- of Rehabilitation Medicine, and completed his General Medical tal ultrasound utility for muscle and nerve injuries. He is completing his Internship, in the Joint Tufts-Harvard Program, at Lemuel Shattuck requirements for ultrasound certification. Currently, he is the medical Hospital, in Boston. He received his MD degree in 1979 from Rush director for Special Olympics Colorado. Dr. Primack is also a clinical Presbyterian St. Lukes Medical College, Chicago, Illinois; earned his instructor in rehabilitation at Western University College of Osteopathic Bachleor of Philosophy of Mathematics from Christ Church, Oxford Medicine. University, England in 1966; and a Bachelor of Science degree with high honors from the University of Michigan, in 1962. Dr. Fishman is present- William S. Pease, MD ly Associate Editor of Topics in Geriatric Rehabilitation. He is an assistant clinical professor at Columbia College of Physicians and Surgeons in New Chair York, and at the Medical College of Wisconsin in Milwaukee. He is also in Department of Physical Medicine and Rehabilitation private practice in Manhattan. Ohio State University Columbus, Ohio Robert A. Werner, MD, MS Dr. Pease earned his medical degree in 1981 from the University of Professor Cincinnati College of Medicine, and performed a residency in physical Department of Physical Medicine and Rehabilitation medicine and rehabilitation at The Ohio State University Medical Center in 1984. He presently serves as the Ernest W. Johnson Professor and University of Michigan Chairperson in the Department of Physical Medicine and Rehabilitation Ann Arbor, Michigan at The Ohio State University College of Medicine. He is also Medical Dr. Werner earned his medical degree from the University of Connecticut Director of Dodd Rehabilitation Hospital at the Ohio State University and his masters of science at the University of Michigan. He is currently a Medical Center. Dr. Pease is a member of the Editorial Board of The professor in the Department of Physical Medicine and Rehabilitation at American Journal of Physical Medicine and Rehabilitation and has authored the University of Michigan in Ann Arbor. He is also Chief of the Physical over 40 publications, including the textbook Practical Electromyography, Medicine and Rehabilitation Service at Ann Arbor Veterans which is in its third edition. Administration Medical Center in Ann Arbor. Dr. Werner is active in several professional organizations including the American Academy of Physical Medicine and Rehabilitation, the Association of Academic Physiatrists, and the Michigan Academy of Physical Medicine and Rehabilitation. Dr. Werner is also active in the AANEM and has served on the AANEM Board of Directors and the Research, Program, and Professional Practice committees.

Course Chair: Jeffrey A. Stakowski, MD

The ideas and opinions expressed in this publication are solely those of the specific authors and do not necessarily represent those of the AANEM. iii

Ernest W. Johnson, MD Lawrence R. Robinson, MD Emeritus Professor Professor Department of Physical Medicine and Rehabilitation Department of Rehabilitation Medicine Ohio State University University of Washington Columbus, Ohio Seattle, Washington Dr. Johnson received his medical degree from Ohio State University in Dr. Robinson attended Baylor College of Medicine and completed his res- Columbus, Ohio, interned at Philadelphia General Hospital, and complet- idency training in rehabilitation medicine at the Rehabilitation Institute of ed his residency in physical medicine and rehabilitation at Ohio State Chicago. He now serves as professor and chair of the Department of University under the sponsorship of the National Foundation of Infantile Rehabilitation Medicine at the University of Washington and is the Paralysis. He has edited the textbook Practical EMG, and authored over Director of the Harborview Medical Center Electrodiagnostic Laboratory. 143 peer-reviewed articles. He established the Super EMG continuing He is also currently Vice Dean for Clinical Affairs at the University of medical education course in 1978, and is still involved in planning and Washington. His current clinical interests include the statistical interpreta- teaching this course. Currently, Dr. Johnson is an emeritus professor at tion of electrophysiologic data, laryngeal electromyography, and the study Ohio State University. He has conducted research on electrodiagnostic of traumatic neuropathies. He recently received the Distinguished medicine in recurrent carpal tunnel syndrome and the use of H waves in Academician Award from the Association of Academic Physiatrists and this upper limb radiculopathies. Dr. Johnson is a past-president of the year is receiving the AANEM Distinguished Researcher Award. AANEM, AAPM&R, AAP, former chair of the American Board of Electrodiagnostic Medicine, and has been editor of the American Journal of Physical Medicine and Rehabilitation. iv AANEM Course v

CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine

Contents

Faculty ii Objectives iii Course Committee vi

Piriformis Syndrome is Underdiagnosed 1 Loren M. Fishman, MD, B. Phil

Piriformis Syndrome is Overdiagnosed 11 Robert A. Werner, MD, MS

Musculoskeletal Ultrasound has Potential Clinical Utility 15 Scott J. Primack, DO

Musculoskeletal Ultrasound has Limited Clinical Utility 19 Willam S. Pease, MD

Tarsal Tunnel Syndrome is Underdiagnosed 23 Ernest W. Johnson, MD

Tarsal Tunnel Syndrome is Overdiagnosed 29 Lawrence R. Robinson, MD CME Self-Assessment Test 33 Evaluation 35 Member Benefit Recommendations 37 Future Meeting Recommendations 39

Please be aware that some of the medical devices or pharmaceuticals discussed in this handout may not be cleared by the FDA or cleared by the FDA for the specific use described by the authors and are “off-label” (i.e., a use not described on the product’s label). “Off-label” devices or pharmaceuticals may be used if, in the judgement of the treating physician, such use is medically indicated to treat a patient’s condition. Information regarding the FDA clearance status of a particular device or pharmaceutical may be obtained by reading the product’s package labeling, by contacting a sales representative or legal counsel of the manufacturer of the device or pharmaceutical, or by contacting the FDA at 1-800-638-2041.

O BJECTIVES—Attending this course will provide the participant the opportunity to discuss (1) whether tarsal tunnel syndrome is overdiagnosed or underdiagnosed, (2) whether piriformis syndrome is overdiagnosed or underdiagnosed, and (3) the clinical utility of musculoskeletal ultrasound. P REREQUISITE—This course is designed as an educational opportunity for residents, fellows, and practicing clinical EDX physicians at an early point in their career, or for more senior EDX practitioners who are seeking a pragmatic review of basic clinical and EDX prin- ciples. It is open only to persons with an MD, DO, DVM, DDS, or foreign equivalent degree. A CCREDITATION S TATEMENT—The AANEM is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education (CME) for physicians. CME CREDIT—The AANEM designates attendance at this course for a maximum of 3.25 hours in category 1 credit towards the AMA Physician’s Recognition Award. This educational event is approved as an Accredited Group Learning Activity under Section 1 of the Framework of Continuing Professional Development (CPD) options for the Maintenance of Certification Program of the Royal College of Physicians and Surgeons of Canada. Each physician should claim only those hours of credit he/she actually spent in the activity. The American Medical Association has determined that non-US licensed physicians who participate in this CME activity are eligible for AMA PRA category 1 credit. CME for this course is available 9/05 - 9/08. vi

2004-2005 AANEM COURSE COMMITTEE

Kathleen D. Kennelly, MD, PhD Jacksonville, Florida

Thomas Hyatt Brannagan, III, MD Dale J. Lange, MD Jeremy M. Shefner, MD, PhD New York, New York New York, New York Syracuse, New York

Timothy J. Doherty, MD, PhD, FRCPC Subhadra Nori, MD T. Darrell Thomas, MD London, Ontario, Canada Bronx, New York Knoxville, Tennessee

Kimberly S. Kenton, MD Bryan Tsao, MD Maywood, Illinois Shaker Heights, Ohio

2004-2005 AANEM PRESIDENT

Gary Goldberg, MD Pittsburgh, Pennsylvania Piriformis Syndrome Is Underdiagnosed

Loren M. Fishman, MD, B. Phil.(oxon.) Assistant Clinical Professor Department of Physical Medicine and Rehabilitation Columbia College of Physicians and Surgeons New York City, New York

INTRODUCTION anterior divisions deriving from the lumbrosacral plexus that make up the sciatic nerve do not unite proximal to the sciat- New information is now available since the controversy of ic nerve about 30% of the time, there are a number of differ- piriformis syndrome (PS) began in 2002.5,14 Very little of it ent conditions that might result in entrapment in this loca- is statistically based—the type of evidence that would actual- tion, including the common nonanomalous condition. ly settle the question whether there are more cases of PS than Typically, the complete nerve leaves the pelvis distal to the clinicians find, or whether there are less than meet the eye. piriformis muscle, between it and the gemellus superior, adja- These new findings include a study of nondisc-originating cent and rostral to the ischiofemoral ligament. However, in a sciatica,2 new methods of identifying PS, sharper definitions minority of cases, the muscle may be bifid, and the sciatic resulting from these methods, and consequently, new esti- nerve compressed between its parts; when the posterior and mates of the prevalence of the syndrome. anterior divisions of the lumbosacral plexus do not join, they may pass the piriformis muscle separately and be compressed Piriformis syndrome is an entrapment of the sciatic nerve or separately. When there is nonunion of the posterior and ante- its branches as they leave the pelvis in relation to the piri- rior divisions, fibers of one or both divisions may perforate formis muscle. This excludes more proximal entrapment of the body of the piriformis muscle. In the anatomy laborato- the fibers that make up the sciatic nerve within the ry using imaging studies, the sciatic nerve’s divisions enclos- intramedullary space, at the neuroforamina, or at the brachial ing muscle fibers have even seen, parting above them, and plexus. It is also different from more distal pathology as the rejoining below.2,4 nerve and/or its components round the ischial tuberosity at the ischial tunnel. In addition, sciatic nerve’s fibers may be Piriformis syndrome has been variously ascribed to some or pinioned just north of the sciatic notch in the distal pelvis, all of the anatomical variations noted earlier12—vascular, from conditions such as endometriosis, fistulae, pelvic traumatic, and mechanical causes—however, after a review of inflammatory disease, tumors, postsurgical adhesions, and a the 85 conservative treatment failures that this author has number of other causes. sent to surgery over the last 17 years, anatomical anomalies have been seen in approximately the same proportion as Since the muscular anatomy surrounding the intersection of found in the cadaverous population.4 In addition, while these the sciatic nerve and the piriformis muscle is anomalous abnormalities are almost always bilateral, PS is unilateral 90% approximately 15% of the time,7,12 and the posterior and of the time.4 Cases of sciatica in which the pathogenetic 2 Piriformis Syndrome is Underdiagnosed AANEM Course mechanism derives from pressure-dependent stress on the sci- Piriformis syndrome may be underdiagnosed by individuals atic nerve by the piriformis muscle are generally accepted as or groups at one time, and overdiagnosed by the same indi- PS. viduals or groups at another time. The key is to find systematic error. It is important to determine whether the diagnosis It is in no way paradoxical for the same condition to be over- of is a “fad,” or a valid and viable means of identifying and diagnosed and underdiagnosed at the same time (Figure 1). treating patients with sciatica. “Underdiagnosed” simply means there are examples of the condition that are unrecognized. A condition is overdiag- The extent to which PS is under- and overdiagnosed may be nosed to the extent that clinicians claim there are people who due to its being considered a “diagnosis of exclusion.” have the condition who actually do not. Almost every condi- Diagnoses of exclusion are not considered unless and until all tion is probably simultaneously under- and overdiagnosed other likely diagnoses are ruled out. If PS is a genuinely since it is unlikely that either set—the set of actual cases or occurring condition that medical ignorance has relegated to the set of diagnosed cases—completely contains the other set. the “diagnosis of exclusion” category, then patients who pres- If the set of diagnosed cases is smaller than the set of actual ent with PS and another condition are systematically diag- cases, then there are more underdiagnosed than overdiagnosed and treated for that other condition without the physi- nosed cases. In that situation, for every diagnosed case that is cian considering the diagnosis of PS. It would be systemati- not an actual case (overdiagnosis), there is necessarily one cally underdiagnosed. On the other hand, a spurious but more actual case that is undiagnosed (underdiagnosis). popular “pathognomonic” sign would lead to systematic Boundaries of the two sets converge as diagnostic tests more overdiagnosis. If suitable tests for PS can be established how- faithfully reflect the pathogenetic mechanism as shown by ever, it could at times be found to coexist with other condi- the dotted line in Figure 1. tions and be diagnosed concurrently with other diagnoses,

Figure 1 Conditions can be simultaneously under- and overdiagnosed when neither the set of actual cases nor the set of diagnosed cases completely contains the other set. If the set of diagnosed cases is smaller than the set of actual cases, then every overdiagnosed case actually implies an underdiagnosed case, by virtue of the lack of overlap. The two sets tend to converge as diagnostic tests more faithfully reflect the pathogenetic mechanism (curved arrow).

PS = piriformis syndrome. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 3 just as a person might have a fever from pneumonia and to have PS (Figure 2). The next most common diagnosis was dehydration at the same time. ischial tunnel entrapment (4.7%).

Recently Stewart,14 a well-respected physician who did not Patients’ piriformis muscles were injected with 10 cc bupivi- believe in the existence of PS, proposed five criteria for PS: caine under close MRI observation, imaging the patient 15- 25 times per injection. Patients that responded positively to (1) Symptoms and signs of sciatic nerve damage; the injection or to subsequent surgery on the piriformis muscle were considered to have PS. (2) Needle electromyographic (EMG) evidence of sciatic nerve damage with normal paraspinal EMG; Eighty-four percent of these patients had positive results with injection, but unfortunately only 23% had lasting relief. (3) Unremarkable imaging studies from the paraverte- Seventy-six percent of the 64 patients sent for PS surgery had bral region to the sciatic notch; excellent or good results.

(4) Surgical exploration identifying piriformis or fibrous The clinical criteria this author has used to identify PS are the compression of the sciatic nerve, without mass following: (1) pain in the buttock and usually some part of lesions; the course of the sciatic nerve distal to it, (2) tenderness in the region of intersection of the piriformis muscle and the (5) Symptomatic and neurological improvement follow- sciatic nerve, (3) positive straight leg raise at 15 degrees less ing surgical decompression. than on the unaffected side (or less than 60 degrees when symptoms were bilateral), (4) weakness in resisted abduction Stewart admits that surgical decompression in other situa- (after Pace11), and (5) buttock pain with passive adduction of tions does not always lead to symptomatic relief. Perhaps he the flexed thigh (after Solheim13). fails to recognize the requirement that other conditions be absent returns to considering PS a diagnosis of exclusion. The electrophysiological criterion, prolongation of the poste- Would a physician define pneumonia as a pathogenic bacte- rior tibial or peroneal H reflexes through the Flexion rial or viral inflammation of the lungs without the presence Adduction and Internal Rotation (FAIR) test, has been of dehydration, tumor, or asthma? More than one condition described earlier.4,6 Essentially, while patients without piri- can coexist in a patient. formis symptoms show no mean change in their H reflexes when utilizing the FAIR test, patients with the five clinical Nevertheless, to prove the point, other causes ought to be criteria described earlier reliably show three standard devia- excluded from a sample of patients with sciatica. This has tion prolongations of the H-reflex latency utilizing the FAIR been performed by three independent groups using two dif- test (Figures 3-5). Although not always mirroring clinical ferent methods. Childers and colleagues recently excluded progress, electrophysiological criteria have shown greater patients with imaged herniated disc, nerve root impinge- than 83% sensitivity and specificity when matched against ment, or needle EMG evidence of denervation proximal to the presence of at least three of the aforementioned clinical the sciatic notch.1 Filler and colleagues2 studied 239 consec- criteria.4,6 utive sciatica patients in whom either no diagnosis had been made, or for whom lumbar surgery had been ineffective. Re- In the early days, only patients who had negative computer- diagnosis brought negative imaging studies or, in equivocal ized tomography (CT) imaging studies and who had no cases, negative disc, foraminal, and/or facet-injections in all EMG evidence of denervation in the paraspinal musculature, but 14. The remaining 225 had histories and physical exam- the tensa fascia latae, or any muscles whose nerve supply is inations that were consistent with PS. They then conducted not part of the sciatic nerve distal to the piriformis muscle neuroimaging studies of the relevant areas, from the lumbar were considered to have PS. As in pronator syndrome, the spine to beyond the buttock. piriformis muscle itself had to be electrophysiologically normal. Magnetic resonance imaging (MRI) studies of these patients highlighted neurological tissue of the buttock region, giving In later studies, a strong correlation was demonstrated detailed information of the sciatic nerve’s course, muscle vol- between the electrophysiological and clinical criteria. In addi- ume, inflammation, and the location of pathology along the tion, FAIR-test prolongation of the H reflex was shown to sciatic nerve. Neuroimaging has the ability to identify far lat- correspond to the level of patients’ pain (Figure 6). This sug- eral disc disease, schwannomas, and other abnormalities.10 gested a mechanical and generally reversible compression of While distal neuroforaminal entrapment was found in 14 of the nerve by the muscle, a straightforward pathogenesis that these patients, 162 (67.8%) of these 239 patients were found could coexist with other types of pathology. At that point, 4 Piriformis Syndrome is Underdiagnosed AANEM Course

Figure 2 A: Asymmetrically large left piriformis muscle. B. and D: Axial and coronal sections. Arrows indicate sciatic nerves. B. Curved neurographic image of hyperintense left sciatic nerve with loss of fascicular detail. D. From Filler, AG with permission of the publisher.

this author and colleagues began considering PS in patients with herniated discs, spondylolistheses, spinal stenoses, and neuropathies. A number of cases of PS was diagnosed in thesse patients.4,6

Because of exposure in the lay press, this author’s office has seen over 9000 patients with suspected PS. Some have been self referred while that have bee referred by physicians who suspected PS—resulting in a biased patient sample. After examining, treating, and following these patients on average more than 1 year, slightly more than half of them are believed to actually have PS. Reasonable criticism recently prompted a review of these records. Of the 1014 legs with sciatica Figure 3 The independent variable, Flexion Adduction and 4 Internal Rotation, increases approximately as the solid angle reviewed in an earlier study, lumbar MRIs could be verified alpha increases; the dependent variable, compression of the in 449 patients. Roughly one-fourth (129 patients) of the sciatic nerve beneath the piriformis muscle, is reflected in delay MRIs had relevant positive findings; 320 were normal. Of of the H reflex. See figures 4, 5. the positive MRI group, 74.7% of patients improved 50% or more with injection of bupivicaine and physical therapy. The AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 5

Figure 4 The H reflex in the anatomical position is compared with the H-reflex in Flexion Adduction and Internal Rotation (FAIR) test. Since this conduction crosses the piriformis muscle twice, in afferent and efferent limbs of the reflex arc, any slowing due to compression will be amplified times 2 in the reflex’s latency.

SD = standard deviation.

Figure 5 The mean increase in H-reflex latency in patients with clinical symptoms of piriformis syndrome (> 5 standard deviations beyond mean values seen in normal subjects) is easily distinguished from what is seen in normal subjects and contralateral limbs.

PS = piriformis syndrome; SD = standard deviation. 6 Piriformis Syndrome is Underdiagnosed AANEM Course average improvement was 62.3%. The mean previous dura- One study estimates that of 1.5 million cases of sciatica seri- tion of sciatica for this group was 6.9 years. In the negative ous and long-lived enough to warrant MRIs, only 200,000 MRI group, 74.0% of patients improved 50% or more with “proved to have a treatable herniated disc.”2 But in this study, a mean improvement of 61.5%. The mean duration of sciat- two-thirds of the nondisc cases of sciatica were PS, and the ica for this group was 5.8 years. Perhaps the greater time to authors conclude that PS may be as common a cause of sci- determine a diagnosis for the positive MRI group was exact- atica as herniated disc. ly because of these “dual diagnoses.” Jensen and colleagues9 reviewed MRIs of 100 asymptomatic There were 179 patients with confirmed negative MRI and patients and found more than 30 revealed lumbar pathology normal paraspinal needle EMGs. Just over 76% of these of the types frequently cited as causes of sciatica. As already patients improved 50% or more with conservative therapy in implied, if a patient can have this pathology without sciatica, a mean time of 11 months of follow-up. Mean improvement then a patient can have this pathology and sciatica from was 61.8%. Mean duration of symptoms was 5.9 years in this another source. This can probably be applied, mutatis group. Ninety-seven (54%) of the 179 patients in this group mutandis, to the MRI studies of the buttock that support PS did not receive bupivicaine injections, but only received as well. In each case, concerning herniated disc and PS, sup- physical therapy (PT). plementing MRI with needle EMG demonstrates the actual pathogenetic mechanism of pain, while supplementing nee- Physical therapy alone focused fairly narrowly on removing dle EMG with MRI secures the pathology’s structural under- the mechanical compression, although it did include myofas- pinning and exact location. cial release of lumbosacral paraspinals in order to free up the nerve roots. Theoretically this provides more mobility to the There is a difference in the nature of the signs and prominent lumbosacral plexus in order to increase the sciatic nerve’s symptoms of PS found in different studies. Filler’s group mobility in the notch that bears its name. This ought to allow emphasizes asymmetrical piriformis muscle volumes, the it, like any physical object not under constraint, to be moved absence of ischial tunnel entrapment (structural criteria), and by pressure to regions of lower pressure. This author’s PT enhanced sciatic nerve brightness, ostensibly an indication of protocol can be seen on www.sciatica.org. Performed under inflammation. This is a functional criterion. Needle EMG needle EMG guidance, injection of lidocaine and corticos- criteria are three standard deviation prolongation of the H- teroid, or botulinum toxin3 has brought the percentage of reflex latency, the dependent variable, in response to stretch conservatively treated cases achieving at least 50% improve- of the piriformis muscle over the sciatic nerve, the independ- ment from the low 70% to the low 90%. ent variable (Figure 3). This is a functional criterion. The FAIR test can be used to monitor patient progress during A few general points are not evident from this data. This conservative treatment. A previously unpublished surgical author’s largest study of 958 patients had symptoms for an series monitored patients utilizing the FAIR test (Table 1). average of 6.2 years.4 Patients were seen, on average, by more than 6.5 clinicians before receiving what appears to have been One group delineates the structural situation, the other cap- the proper diagnosis. Childer’s and Filler’s patients had also tures the pathological events that occur. Piriformis muscle been undiagnosed for significant lengths of time.1,2 asymmetry and sciatic nerve hyperintensity had 93% specificity and 64% sensitivity. Prolongation of the H reflex with Moreover, analysis of the 1014-leg study found that these flexion, adduction, and internal rotation had 83.2% speci- 958 patients had a large number of studies performed includ- ficity and 88.1% sensitivity. One study using H-reflex pro- ing 1190 MRI scans, 1380 radiograph studies, and 860 other longation as the identifying criterion for PS and EMG guid- imaging studies (bone scan, ultrasound, etc.). In addition, ance for injection of botulinum toxin type B into the piri- over 400 total surgeries, (spinal, hip, and gynecological, in formis muscle showed a 77% correlation coefficient between that order) and a large number of other procedures such as the visual analogue scale and the FAIR-test results over a 3- prolotherapy, dilatation and curettage, and epidurals were month period (Figure 6).3 The electrodiagnostic abnormali- performed. Forty-nine percent of Filler’s group had previous- ties and the patients’ symptoms disappeared at the same rate ly undergone lower back surgery without benefit.2 All of this during treatment that focused on the piriformis muscle. suggests that many of the previous clinicians considered PS a diagnosis of exclusion, or did not consider it at all. Yet more The results of different authors mostly agree, but there are than 79% of all these patients improved 50% or more with differences: one group2 finds involvement of all the toes char- the focused treatment described previously, with more than acteristic of PS, while others4,6 find tenderness at the intersec- 66% of the conservative treatment failures improving 50% or tion of muscle and nerve to be the most common symptom more if they subsequently chose surgery on the piriformis (Table 2). The commonness of straight leg raise positivity is muscle.2,4,6 debated. Yet almost all agree that sitting is the most aggravat- AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 7

Table 1 Pre- and Postsurgical FAIR-tests

Pre-Op FAIR-test Standard Deviation Clinical Patient FAIR-test (ms) Post-Op (ms) Change (ms) of Change Outcome RA N.O. 1.46 N.O. N.O. Excellent BO 2.29 1.04 1.25 2.00 Excellent GA 3.54 2.08 1.46 2.34 Excellent LE 0.83 0.00 0.83 1.34 Good LO N.O. 1.25 N.O. N.O. Excellent MA N.O. N.O. N.O. Fair DC 2.08 N.O. N.O. N.O. Excellent OR 2.49 0.83 1.86 3.00 Excellent PE N.O. 1.04 N.O. N.O. Poor Totals 11.23 7.70 5.40 8.68 Mean 2.25 1.28 1.35* 2.17

Decreases in Flexion Adduction and Internal Rotation (FAIR) test values and pain relief following surgery suggest that the piriformis muscle was entrapping the sciatic nerve, and that this entrapment was causing the patients’ symptoms: generally buttock pain and sciatica, with frequent paraesthesias, occasional sensory loss, and rare weakness. N.O. = none observed. * Mean difference between pre- and postsurgical FAIR-tests was 2.17 standard deviations beyond the mean seen in normal subjects.

Figure 6 Parallel course of patients’ symptoms (VAS) and FAIR-test values during physical therapy after injection suggests that the clinical symptoms were related to sciatic nerve compression by the piriformis muscle, which caused the H-reflex delay.

FAIR = Flexion Adduction and Internal Rotation; VAS = visual analog scale. 8 Piriformis Syndrome is Underdiagnosed AANEM Course

Filler emphasizes that, “The low rate of referral and frequent Table 2 Characteristics Favoring a Successful Outcome to failure to recognize the diagnosis, however, should not be 2 Conservative Therapy (Physical Therapy) mistaken for evidence of a low incidence in the population.”

Characteristic symptom or sign p-value odds Ratio SUMMARY Positive FAIR test .001 2.225 Overuse .001 2.05 Like other functional entrapments such as neurogenic tho- Tender piriformis/sciatica .003 1.97 racic outlet syndrome, PS is a challenge to diagnose and treat. Solhiem sign .005 1.84 It is fortunate that structural and functional means of identi- Sitting worse than standing .007 1.77 fying the problem exist, and that structural (surgical) and Female gender/high weight .011 ( - ) functional (medication and physical therapeutic) means are available to treat it. Some combination of the two may be the Illiotibial band syndrome .028 1.67 most effective. For example, Filler’s group and this author’s Injection .030 1.55 group’s successful surgical outcomes are both around 80% Non-sciatic denervation .035 2.07 with reasonably similar criteria. However, this author’s long- Peroneal polyphasics .041 2.02 term improvement with injection is at 79% while Filler’s is at Non-sciatic polyphasics .041 2.51 23%, with relapse of another 37% of initial successes within Years of pain .041 .910 4 months. It is possible that the difference in longevity of the benefits lies in home exercise and physical therapy. Studies These are the characteristics that statistically affect the response to the injection are clearly needed. and physical therapy treatment this author has used in more than 5000 patients to date.6 See www.sciatica.org. Clinical suspicion of PS should rise when the patient has more pain sitting than standing; a history of overuse, trauma, or unusual body habitus (obesity or cachexia); or tenderness in the mid-buttock that resembles their initial complaint. Once PS is appropriately sought by clinicians and rationally ing position, that abduction and adduction of the flexed linked to a test that images and/or replicates its pathogenetic thigh is painful, and that successful surgical revision or neu- mechanism, PS can be properly diagnosed or ruled out rolysis of the piriformis-sciatic region does confirm the diag- according to the skill and energy of the individual practition- nosis.2,4,6,13 er. At that point, a great deal of the systematic error that brings about underdiagnosis will have been eliminated. Childers used dye-emitting needles and fluoroscopy,1 Filler Reliable methods of diagnosis will reduce the numbers that used neuroimaging,2 and this author uses needle EMG guid- suffer from PS, and the length of time during which they suf- ance for injection.6 Bupivicaine, lidocaine, various steroids, fer. Thereafter, perhaps successful investigation will extend to and botulinum toxins yielded successful results from 45-92% other functional entrapments. of the time. These are signs of an exciting, developing field of inquiry. REFERENCES Olmstead County, Minnesota, where the Mayo Clinic is located, recorded 32,655 cases of lower back pain from 1976- 1. Childers MK, Wilson DJ, Gnatz SM, Conway RR, Sherman AK. 2001. The diagnosis of PS was made 220 times over this peri- Botulinum toxin in piriformis muscle syndrome. Am J Phys Med od, giving a diagnostic rate of 0.7%. In 1976-1979 the diag- Rehabil 2002;81:751-759. nosis was made in 11 of 4416 cases, a rate of 0.25%, where- 2. Filler AG, Haynes J, Jordan SE, Prager J, Villablanca JP, Farahani K, as in 2000-2001 it was made in 54 of 4349 cases (1.24%), McBride DQ, Tsurudu JS, Morisoli B, Batzdorf U, Johnson JP. nearly a five-fold rise over this quarter century. This, howev- Sciatica of nondisc origin and piriformis syndrome: diagnosis by magnetic resonance neurography and interventional magnetic reso- er, is still five-fold short of the 6% seen at the Mayo Clinic in nance imaging with outcome study of resulting treatment. J an unbiased sample in 1983.8 In addition, this 1983 figure Neurosurg Spine 2005;2:99-115. was before the advantage of any of the modern diagnostic 3. Fishman LM, Anderson C, Rosner B. BOTOX and physical therapy techniques discussed here. Walter Reed Hospital reported in the treatment of piriformis syndrome. Am J Phys Med Rehabil 155 cases of PS (1.58%) out of 9161 diagnoses of low back 2002;82:936-942. pain during the year 2002. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 9

4. Fishman LM, Dombi GW, Michaelsen C, Ringel S, Rozbruch J, 10. Moore KR, Tsuruda JS, Dailey AT. The value of MR neurography for Rosner B, Weber C. Piriformis syndrome: diagnosis, treatment, and evaluating extraspinal neuropathic leg pain: a pictorial essay. AJNR outcome–a ten-year study. Arch Phys Med Rehabil 2002;83:295- Am J Neuroradiol 2001;22:786-794. 301. 11. Pace JB, Nagle D. Piriform syndrome. West J Med 1976;124:435- 5. Fishman LM, Schaefer MP. The piriformis syndrome is underdiag- 439. nosed. Muscle Nerve 2003;28:646-649. 12. Pecina M. Contribution to the etiological explanation of the piri- 6. Fishman LM, Zybert PA. Electrophysiological evidence of piriformis formis syndrome. Acta Anat 1979;105:181-187. syndrome. Arch Phys Med Rehabil 1992;73:359-364. 13. Solheim LF, Siewers P, Paus B. The piriformis muscle syndrome. J 7. Gotlin R. Clinical correlation of an anatomical investigation into pir- Orthop Scand 1981;52:73-75. iformis syndrome. Proc N Y Soc Phys Med Rehabil 1991;24:11. 14. Stewart JD. The piriformis syndrome is overdiagnosed. Muscle Nerve 8. Hallin RP. Sciatic pain and the piriformis muscle. Postgrad Med 2003;28:644-646. 1983;74:69-72. 9. Jensen MC, Brant-Zawadzki MN, Obuchowski N, Modic MT, Malkasian D, Ross JS. Magnetic imaging of the lumbar spine in people without back pain. New Engl J Med 1994;333:69-73. 10 AANEM Course 11

Piriformis Syndrome is Overdiagnosed

Robert Werner, MD, MS Professor Department of Physical Medicine and Rehabilitation University of Michigan Ann Arbor, Michigan

INTRODUCTION mentioned in many prominent textbooks such as The Essentials of Musculoskeletal Care4 by the American Society of There is a great deal of controversy regarding the existence of Orthopaedic Surgeons and Sunderland’s Peripheral Nerves piriformis syndrome (PS). Although not listed in Dorland’s and Nerve Injury11 are a strong indication that the medical Medical Dictionary or many medical encyclopedias, the community has not embraced the term. National Institute of Neurologic Disorders and Stroke within the National Institutes of Health defines the syndrome as Assuming the syndrome does exist, the lack of a method “a rare neuromuscular disorder that occurs when the piri- (electrodiagnostic or imaging) to help establish a diagnosis formis muscle compresses or irritates the sciatic nerve.” If one makes it difficult to determine the prevalence of the disorder. accepts the diagnosis as real, confusion reigns over how to It is a syndrome that is defined based upon the history and establish the diagnosis. Piriformis syndrome is most often referred to as a form of sciatic nerve entrapment causing buttock and hamstring pain. The original description of this Table 1 Causes of sciatic nerve injury. From Sunderland,11 condition dates from 1928 when Yeoman stated that “insuf- Peripheral nerve injury, series of 209 cases. ficient attention” has been paid to the piriformis muscle as a potential cause of sciatica.14 Despite this, Sunderland never mentions PS in his review of the causes of sciatic nerve Stab wound 1% injury.11 He reported that only 1.5% of all cases in his series Ganglia or tumor 0.5% of 209 cases were caused by compression (Table 1) and he did Gunshot wound 70% not attribute any of these to PS. Similarly, a review of the last Iatrogenic during surgery 1.5% 5 years at this author’s electrodiagnostic (EDX) laboratory Compression 1.5% found 96 cases of sciatic neuropathy and none were attrib- Injection injury 3% uted to PS. Dislocation of hip 9% When searching Medline, PS is not listed as a medical sub- Fracture of acetabulum 2.4% ject heading. Similar compression neuropathy “syndromes” Fracture of femur 5.3% such as thoracic outlet syndrome (TOS) and carpal tunnel Injury of the knee with traction 5.3% syndrome (CTS) have accepted medical subject heading. The lack of a consensus definition of PS and the fact that PS is not 12 Piriformis Syndrome is Overdiagnosed AANEM Course clinical presentation. Most efforts at confirming the diagno- symptoms that began after trauma to the buttocks region. In sis have not been helpful. If the syndrome involves an abnor- addition to sciatic nerve injury, these patients had needle mality of the sciatic nerve as it passes by the piriformis mus- EMG evidence of injury to the inferior gluteal nerves. The cle, physicians should be able to identify an abnormality of surgical findings included myositis ossificans of the piri- the sciatic nerve using electrophysiologic measures or demon- formis muscle as well as adhesions between the piriformis strate an abnormality on imaging studies. Stewart suggested muscle, the sciatic nerve, and the roof of the sciatic notch. the following criteria for establishing the diagnosis of PS.10 The surgical release of the piriformis tendon improved their symptoms. (1) Presence of symptoms and signs of sciatic nerve damage. The majority of the reports of PS do not demonstrate objective evidence of sciatic nerve injury that is confirmed with (2) Presence of electrophysiological evidence of sciatic EDX testing or imaging studies. This leaves the largest group nerve damage. Paraspinal muscle needle electromyo- of patients with complaints of chronic buttock pain but no graphy (EMG) must be normal, to help in excluding evidence of sciatic nerve involvement. Is this really PS? Since a radiculopathy. sciatica is a vague term that includes many etiologies, it is only by a process of exclusion that PS can be considered. If (3) Imaging of the lumbosacral nerve roots and of the the primary symptom is buttock pain (often with sciatica) paravertebral and pelvic areas must be normal to and no neurological deficits are found, PS is one of many exclude radiculopathy, or lower lumbar or sacral possible causes. The exclusion of radicular involvement and plexus infiltration or damage. Imaging of the pelvis plexus involvement does not establish the diagnosis of PS. If and sciatic notch must show the absence of mass the symptoms are a result of buttock trauma, these patients lesions. do not meet the criteria for PS.

(4) Surgical exploration of the proximal sciatic nerve The main issue is whether PS actually causes compression of should confirm an absence of mass lesions. Ideally, the sciatic nerve with resultant pain but no evidence of nerve compression of the sciatic nerve by the piriformis fiber damage. In most compressive neuropathies, if the muscle or associated fibrous bands should be identi- patient reports pain from nerve irritation, there are typically fied. However, it can sometimes be difficult to recog- sensory or motor symptoms, clinical deficits, and electro- nize a compressed nerve. physiological abnormalities. If the nerve injury cannot be documented, can the condition be confirmed? In patients (5) Relief of symptoms and improvement in neurologi- with buttock pain who also have complaints of sciatica, it cal abnormalities should follow surgical decompres- should be recognized that the term sciatica may mean differ- sion. As in other situations of chronic nerve damage, ent things to different people and the causes of sciatica can be decompression may not always lead to symptom quite varied. Most would accept a definition for this term as relief. Surprisingly, surgical division of the piriformis being pain radiating down the leg from the lower back, but- muscle has been described as relieving pain in tock, or hip. Despite the terminology, the definition does not patients with lumbosacral radiculopathies. indicate sciatic nerve involvement. The most frequent neurological etiology is a lumbosacral radiculopathy; but lumbar Many of the older studies regarding PS were published before and sacral plexopathy and proximal sciatic neuropathies are the use of modern imaging techniques, so these studies can- also common. It is also likely that they cause a musculoskele- not be evaluated using Stewart’s criteria. A few studies report tal abnormality of the lumbosacral spine, hip, or pelvis. There patients who meet some of the criteria for PS. One study may be a tendonitis of the piriformis muscle itself due to over reported a patient with a surgical finding of a hypertrophied activity (usually sports-related) without any involvement of piriformis muscle compressing the sciatic nerve.9 Other stud- the sciatic nerve. The referred pain into the buttock and the ies have described three patients with bifid piriformis muscles posterior thigh are common and frequently described by compressing the lateral trunk of the sciatic nerve.2,5 Two patients and health care providers as sciatica despite a lack of patients were reported to have nerve compression by fibrous neurologic injury. bands associated with the piriformis muscle.5,12 Establishing the diagnosis of PS rests upon excluding other Several other reports have demonstrated damage to the sciat- disorders that have overlapping symptoms and then using ic nerve that is associated with trauma at or around the piri- provocative maneuvers to demonstrate irritation of the sciat- formis muscle. Benson and Schutzer reported a series of ic nerve when the piriformis muscle is either contracted or patients with sciatic nerve lesions that were associated with stretched. The validity of these signs is suspect as is the claim trauma and called it “post-traumatic PS.”1 All subjects had that they are specific for demonstrating compression of the AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 13 sciatic nerve by the piriformis muscle. Similar to Tinel’s and function but instead used an H-reflex testing protocol before Phalen’s signs for CTS, and Adson’s maneuver for TOS, the and after a provocative posture (hip flexion, adduction, and sensitivity and specificity of these provocative tests is quite internal rotation). The H-reflex testing protocol has not been poor. The provocative tests for PS have not been critically reproduced in other EMG laboratories and even when per- evaluated. Tenderness with deep palpation in the buttock is formed by Fishman it has a reported 35% false positive rate common in patients with lumbosacral radiculopathy, pelvic in normal control subjects. This is analogous to the high muscle strain, tumors or other masses at the sciatic notch, as false-positive rate for Adson’s maneuver in the evaluation of well as posttraumatic scarring in this area. Tenderness may TOS. Fishman and colleagues also evaluated treatments indicate an abnormality of the piriformis muscle, but it does which were broad-based and had the potential to provide not mean that the sciatic nerve is injured. benefit to patients with a wide variety of musculoskeletal disorders of the lower spine, pelvis, and hips.3 Most patients in The report of pain relief following local anesthetic or corti- the study improved regardless of how many clinical criteria costeroid injections into the piriformis muscle and sciatic they met or whether the H-reflex test was abnormal. The lack notch area does not establish the diagnosis of PS. If the pain of reproducibility and the high false-positive rate of the is caused by a strain or posttraumatic scarring of the piri- provocative H-reflex protocol make this study of question- formis muscle, injections will tend to relieve local symptoms, able value in establishing the diagnosis of true neurogenic PS. but this does not imply that the sciatic nerve was involved in generating the pain. Additionally, it is recognized that a nerve block distal to a nerve lesion can still provide pain relief.6 SUMMARY One study even found that division of the piriformis muscle in patients with lumbosacral radiculopathies could affect The medical community is uncertain if there is an entity pain relief.8 Thus, it is difficult to accept that improvement called PS. There are certainly patients with unexplained but- of pain after steroid injections and even from surgical divi- tock pain but no consensus as to how to use the term PS. sion of the piriformis muscle as proof that the sciatic nerve McCrory has written: “Whether the piriformis muscle is the was compressed by the piriformis muscle. cause of the compression has not been clearly established. It is possible that the obturator internus/gemelli complex is an The situation with PS is similar to that of TOS. Thoracic alternative cause of neural compression. For this reason, I outlet syndrome is a much more common disorder but cases suggest that sports medicine clinicians consider using the of true neurogenic TOS and vascular TOS are rare (less the term ‘deep gluteal syndrome’ rather than piriformis syn- 1% of the cases considered to have TOS).13 The majority of drome.”7 This assumes that some type of nerve compression cases of TOS are classified as “disputed TOS” since there is is involved, which is debatable, since there are many muscu- little convincing evidence that nerve trunks are involved in loskeletal disorders that can mimic the symptoms. Given this the genesis of symptoms. This author believes that PS should overlap of symptoms, there will be many clinicians who have a similar classification scheme, i.e., neurogenic PS and choose to lump these disorders into one term. If this is the disputed PS. The relative frequency of true neurogenic PS is case, this author believes a classification system of true neu- likely to be less than 1% of the cases, as is the case with TOS. rogenic PS and disputed PS should be adopted to help distin- The experience at this author’s institution would tend to sup- guish the cases where there is objective electrophysiologic evi- port this since none of the cases of identified sciatic nerve dence of localized sciatic nerve involvement from the vast injury at the EMG laboratory (n=96) were suspected to have majority of cases in which there is only vague symptoms. PS. Response to treatment that is focused on injection into the piriformis muscle with subsequent relief of symptoms is not There is one novel electrophysiologic study associated with a a diagnostic confirmation of PS. Muscle strain or posttrau- provocative technique that supported the contention that matic scarring treated with massage or injection is a different there was objective evidence of sciatic nerve involvement in a entity than that proposed by Yeoman in the 1920s. high percentage of patients with disputed PS where tradition- Physicians need to differentiate a disorder of the muscle from al electrophysiologic studies and radiographic studies had that of a compression neuropathy. If it cannot be done, been normal.3 Fishman and colleagues reported a series of physicians need to be honest and use a term such as “disput- 918 patients with disputed PS.3 The entry criteria for disputed PS” or “disputed deep gluteal syndrome” to reflect the true ed PS consisted of nonspecific symptoms and signs. Their status of medicine’s understanding of these patients. exclusion criteria were not described. They did not perform Alternatively, local muscle and tendon disorders that do not the standard electrophysiological studies of sciatic nerve involve the sciatic nerve need to be redefined. If the buttock 14 Piriformis Syndrome is Overdiagnosed AANEM Course pain is referred from the lumbosacral spine or pelvis, the term 6. Kibler RF, Nathan PW. Relief of pain and paraesthesiae by nerve PS is discouraged. In many patients, a diligent search for block distal to a lesion. J Neurol Neurosurg Psychiatry 1960;23:91- alternative causes of their pain can prove helpful. 98. 7. McCrory P. The “piriformis syndrome”--myth or reality? Brit J Sports Med 2001;35:209-210. 8. Mizuguchi T. Division of the pyriformis muscle for the treatment of REFERENCES sciatica. Postlaminectomy syndrome and osteoarthritis of the spine. Arch Surg 1976;111:719-722. 1. Benson ER, Schutzer SF. Posttraumatic piriformis syndrome: diagno- 9. Stein JM, Warfield CA. Two entrapment neuropathies. Hosp Pract sis and results of operative treatment. J Bone Joint Surg Am (Off Ed) 1983;18:100A, 100E, 100H. 1999;81:941-949. 10. Stewart JD. The piriformis syndrome is overdiagnosed. Muscle Nerve 2. Chen WS. Bipartite piriformis muscle: an unusual cause of sciatic 2003;28:644-645. nerve entrapment. Pain 1994;58:269-272. 11. Sunderland S. Nerves and nerve injuries, 2nd edition. New York: 3. Fishman LM, Dombi GW, Michaelsen C, Ringel S, Rozbruch J, Churchill Livingstone; 1978. p 164. Rosner B, Wober C. Piriformis syndrome: diagnosis, treatment, and 12. Vandertop WP, Bosma NJ. The piriformis syndrome. A case report. J outcome--a 10-year study. Arch Phys Med Rehabil 2002;83:295- Bone Joint Surg Am 1991;73:1095-1097. 301. 13. Wilbourn AJ. Thoracic outlet syndrome is overdiagnosed. Muscle 4. Greene WB, editor. The essentials of musculoskeletal care, 2nd edi- Nerve 1999;22:130-136. tion. Rosemont, IL: American Academy of Orthopedic Surgeons; 14. Yeoman W. The relationship of arthritis of the sacro-iliac joint to sci- 2001. atica. Lancet 1928;ii:1119-1122. 5. Hughes SS, Goldstein MN, Hicks DG, Pellegrini VD Jr. Extrapelvic compression of the sciatic nerve. An unusual case of pain about the hip: report of five cases. J Bone Joint Surg Am 1992;74:1553-1559. 15

Musculoskeletal Ultrasound Has Potential Clinical Utility

Scott J. Primack, MD Co-director Colorado Rehabilitation and Occupational Medicine Denver, Colorado

INTRODUCTION amount of time and expense. Most clinicians who treat musculoskeletal disorders would agree that because diagnoses can Providing quality care and appropriate treatment of patients be suggested in the vast majority of cases by taking a concise with musculoskeletal disorders is based upon an accurate and relevant history, a working diagnosis can be confirmed diagnostic assessment. A good history is the most important by a careful physical examination. When the diagnosis is still component of the clinical evaluation. Information such as unclear, proper selection of an objective modality becomes the chief complaint, circumstances surrounding the onset or paramount in today’s health care system. mechanism of injury, progression of the problem, response to past treatment, and factors that aggravate or improve func- Because most patients present to the clinician’s office for tion are critical. Concomitantly, age is crucial in determining musculoskeletal problems secondary to their alteration in the likelihood that the patient has a specific correlation. For function, the evaluation of demands placed upon the injured example, carpal tunnel syndrome (CTS) has been shown to body part becomes important. This can be seen through an be present in the fourth and fifth decades of life.17 Rotator understanding of the demands of a sport on the body or the cuff pathology, particularly rotator cuff tears, is rare in essential functions of a job. This knowledge allows the clini- patients younger than 30 years of age. Shoulder instability, cian to make a more complete and accurate diagnosis, to however, commonly presents in the second or third decade of direct a more specific treatment regimen, to allow for preven- life.11,12 tive intervention, and to make recommendations regarding further testing. Clinicians must sift through and analyze the extraneous information that inevitably accompanies the patient present- Advances in medical technology have included the developing for the first time. The physician is faced with the difficult ment of sophisticated imaging techniques which have task of prioritizing and objectifying the information provid- improved the physician’s diagnostic skills and the ability to ed by the patient. At times, however, this information is stage disease and/or injury. Although the ultimate goal is to laden with emotion, or it can be biased by the possibility of return a patient to his or her highest level of function, today’s secondary gain from coaches, fans, the legal system, family, health care cost-containment system has generated an inter- and friends. Aside from these potential limitations of the his- est in a cost-benefit analysis of the various applications of tory-taking process, other obstacles can include the patient’s imaging tests in musculoskeletal medicine. language skills and individual physician’s interpretive ability. Decisions regarding further investigations must be based The sequence in which studies are ordered depends on many upon the history and examination. These decisions are cru- factors. If the situation is critical, the test with the highest cial in determining the cost of the examination and the like- yield is performed despite the possibility of some risk. lihood of arriving at an accurate diagnosis in a minimal Ordinarily, the order of testing is (1) from inexpensive to 16 Musculoskeletal Ultrasound has Potential Clinical Utility AANEM Course costly, (2) from less to more risky, and (3) from simple to Okamoto and colleagues and most recently by Beekman.1,13 more complex. With constraints of time, risk, and cost, Given the ultrasound’s ability to detect motion, Jacobson and physicians attempt to order the study with the most efficien- colleagues were able to demonstrate ulnar nerve dislocation cy as soon as possible; that is, use the procedure with highest and snapping triceps syndrome.6 sensitivity, specificity, and predictive values.

A clinician is often confused by conflicting or indeterminate SUMMARY results, such as “suggestive of” or “compatible with.” In fact, tests or reports that neither exclude nor confirm a diagnosis When used as an extension of the clinical examination, stud- only provoke physicians to order yet another study. A frus- ies have shown that ultrasound imaging is sensitive, specific, trating remark that clinicians often receive from studies is and accurate.14 With its usefulness, physicians can design the “clinical correlation advised,” which for some physicians is most efficacious and accurate treatment plan that ultimately like saying “batteries not included.” Thus, in musculoskeletal should give the best outcome.15 Diagnostic ultrasound for disorders, studies that become an extension of the clinical muscle and nerve disorders is complimentary to other imag- examination are attractive. An excellent example is seen in ing modalities and electrophysiologic testing. As ultrasound the use of electromyography (EMG) and nerve conduction technology further advances, it will be important to educate studies. Likewise, there is great potential in diagnostic mus- physicians in family practice, rheumatology, sports medicine, culoskeletal ultrasound. and orthopedics about the bridge between the sonographic analysis and one’s clinical diagnostic skills. Ultrasound for musculoskeletal injuries was first described by Porter and colleagues in 1978.16 They imaged the lumbar spine of more than 700 subjects from early infancy until the REFERENCES age of 65 and looked at the spinal canal size of patients who had complaints of neurogenic claudication. Although their 1. Beekman R, Schoemaker MC, Van Der Plas JP, Van Den Berg LH, work was difficult to reproduce, they began to see the poten- Franssen H, Wokke JH, Uitdehaag BM, Visser LH. Diagnostic value tial value of this research in preventative medicine by of high-resolution in ulnar neuropathy at the elbow. Neurology 2001;62:767-773. attempting to identify those individuals most at risk for 2. Buchberger W, Judmaier W, Birbamer G, Lener M, Schmidauer C. injury. Macdonald and colleagues correlated male subjects Carpal tunnel syndrome: diagnosis with high-resolution sonography. with significant long-standing histories of low back pain with AJR Am J Roentgenol 1992;159:793-798. those who had the longest absenteeism from work with nar- 3. Buchberger W, Schon G, Strasser K, Jungwirth W. High-resolution row spinal canals.8 Seltzer and colleagues were the first to ultrasonography of the carpal tunnel. J Ultrasound Med 1991;10:531-537. describe the utility of diagnostic ultrasound for imaging of 4. Chiou HJ, Chou YH, Cheng SP, Hsu CC, Chan RC, Tiu CM, Teng 19 the shoulder. Mack subsequently demonstrated that there MM, Chang CY. Cubital tunnel syndrome: diagnosis by high-resolu- were specific criteria needed in order to diagnose rotator cuff tion ultrasonography. J Ultrasound Med 1998;17:643-648. pathology.9,10 One of the most important outcome-oriented 5. Harryman DT 2nd, Mack LA, Wang KY, Jackins SE, Richardson studies that utilized diagnostic ultrasound imaging with a ML, Matsen FA 3rd. Repairs of the rotator cuff: correlation of func- functional measure was Harryman’s work.5 The goal of the tional results with integrity of the cuff. J Bone Joint Surg Am 1991;73:982-989. study was to correlate the functional outcome of open rota- 6. Jacobson JA, Jebson PJ, Jeffers AW, Fessell DP, Hayes CW. Ulnar tor cuff repairs with the integrity of the repaired cuff imaged nerve dislocation and snapping triceps syndrome: diagnosis with at follow-up. The functional measure was called the Simple dynamic sonography – report of three cases. Radiology Shoulder Test Questionnaire. Romeo and colleagues demon- 2001;220:601-605. strated similar results as Harryman when ultrasound imaging 7. Leed, Van holsbeeck M. High-frequency sonography of the median nerve in carpal tunnel syndrome: Radiology, 1992. was utilized to look for recurrent rotator cuff defects follow- 8. Macdonald EB, Porter R, Hibbert C, Hart J. The relationship 18 ing arthroscopic rotator cuff repairs. Buchberger described between spinal canal diameter and back pain in coal miners. the utility of ultrasound for imaging the carpal tunnel.3 He Ultrasonic measurement as a screening test. J Occup Med subsequently was able to demonstrate that one could diag- 1984;26:23-28. nose CTS with ultrasound.2 These conclusions were support- 9. Mack LA, Gannon MK, Kilcoyne RF, Matsen RA 3rd. Sonographic ed by Lee and colleagues.7 A comprehensive analysis of evaluation of the rotator cuff. Accuracy in patients without prior surgery. Clin Orthop Relat Res 1988;234:21-27. peripheral nerve echotexture was described by Silvestri and 10. Mack LA, Matsen FA 3rd, Wang KY. Diagnostic ultrasound. St. 20 colleagues in 1995. The second most common nerve Louis: Mosby; 1991. entrapment is the ulnar nerve at the elbow. The benefit of 11. Matsen FA 3rd, Lippitt SB, Sidles JA, Harryman DT 2nd. Practical ultrasound for this neuropathy has been described by Chiou evaluation and management of the shoulder. Philadelphia: WB and colleagues.4 These findings have been supported by Saunders; 1994. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 17

12. Matsen FA 3rd, Zuckerman JD. Anterior glenohumeral instability. 17. Radecki P. Carpal tunnel syndrome: effects of personal and associat- Clin Sports Med 1983;2:319-338. ed medical conditions. Phys Med Rehabil Clin North Am 13. Okamoto M, Abe M, Shirai H, Ueda N. Diagnostic ultrasonography 1997;8:419. of the ulnar nerve in cubital tunnel syndrome. J Hand Surg 18. Romeo A, Primack SJ. Arthroscopic rotator cuff repairs: correlation 2000;25:499-502. of functional results. Presented at: Orthopedic Sports Medicine 14. Primack SJ, Bernton JT, Schauer LM. Diagnostic ultrasound of the Conference; Breckenridge, Colorado; 2002. shoulder: a cost-effective approach. Presented at: Annual Meeting of 19. Seltzer SE, Finberg HJ, Weissman BN, Kido DK, Collier BD. the American Academy of Occupational & Environmental Medicine; Arthrosonography: gray-scale ultrasound evaluation of the shoulder. Las Vegas, Nevada; 1995. Radiology 1979;132:467-468. 15. Primack SJ, Bernton, JT, Schauer LM. Musculoskeletal ultrasound of 20. Silvestri E, Martinoli C, Derchi LE, Bertolotto M, Chiaramondia M, the shoulder: a physiatric approach: Presented at: The 57th Rosenberg I. Echotexture of peripheral nerves: correlation between AAPM&R Assembly; Orlando, Florida; 1995. ultrasound and histologic findings and criteria differentiate tendons: 16. Porter RW, Hibbert CS, Wicks M: The spinal canal in symptomatic Radiology 1995;197:291-296. lumbar disc lesions. J Bone Joint Surg Br 1978;60:485-487. 18 AANEM Course 19

Musculoskeletal Ultrasound Has Limited Clinical Utility

William S. Pease, MD Chair Department of Physical Medicine and Rehabilitation Ohio State University Columbus, Ohio

INTRODUCTION of US to provide quick, cheap imaging is unlikely to be real- ized due to technical limitations related to depth penetration The use of imaging in problems of the musculoskeletal sys- and resolution, and because of its inability to evaluate bone. tem is part of the foundation of radiology as a discipline. The development of both magnetic resonance imaging (MRI) and ultrasound (US) in this area has allowed the original interest TENDINOPATHY in bone pathology on radiography to be expanded to the soft tissues of body’s mechanical systems. Technical advances in Tendons lend themselves particularly well to US imaging image quality are progressing rapidly in both areas as manu- evaluation because of several features including: (1) consis- facturers compete for their devices to be considered the stan- tent linear structure as seen by US, (2) relatively superficial dard for care. With present technology, MRI is the preferred locations, and (3) close proximity to bones. modality for the imaging of bone, joint, and adjacent soft tissues. Ultrasound imaging should be able to produce similar A recent investigation by Premkumar10 compared MRI and results before it can gain universal acceptance as a primary US in patients suspected of posterior tibialis tendon patholo- imaging modality. Unfortunately, increases in frequency that gy using readily available, high-quality instruments. are needed to improve image resolution will further limit the Magnetic resonance images were completed in a 1.5 T system depth of penetration of the device. with the aid of a knee coil (Figure 1). Ultrasound imaging was completed with a 10 MHz device. Both techniques The world-wide interest in sports and the large revenue demonstrated the tendency of the tendon to enlarge into a streams associated with sports have created a market for rounder than normal state in the presence of pathology. health care with deep pockets. Rapid determination of an However, tendinosis identified by MRI was frequently athlete’s fitness to compete safely is a key consideration in the missed on the US evaluation (sensitivity 83%). Peritendinosis evaluation of any method of clinical examination and testing. was detected in more of the subjects in the study with both After the need for partial or complete rest of a mechanical techniques and the sensitivity of US for this was 86% (speci- structure is determined, the diagnostic and therapeutic values ficity 80%). of imaging modalities as well as clinical examination must be considered. Clinical studies are demonstrating the potential Other foot and ankle tendons are also evaluated well with US of sonography in specific indications. However the promise and with MRI. Dynamic subluxation of the peroneal ten- 20 Musculoskeletal Ultrasound has Limited Clinical Utility AANEM Course

images were more sensitive in detecting mild injuries, and mild residual injuries in serial testing following acute onset of injury. The MRI also was more likely to detect an intermus- cular hematoma than was US. Interestingly, most of the athletes returned to full activity before their imaging abnormalities were resolved by either method.

Rotator cuff evaluation with US is one of its most frequent uses in the musculoskeletal realm. With careful operator training and experience the detection of tendon abnormalities can approach that of MR imaging. Associated biceps tendon and bursa pathology are also identifiable with US, as well as tendon subluxation.6 Calcific deposits can also be identified with US, however small and scattered calcifications are frequently missed.3,9

Ultrasound has combined value as a diagnostic and therapeutic tool in the management of rotator cuff pathology in the shoulder.1 In this series of cases reported by Chiou, sonography detected calcific tendonopathy in the more superficial supraspinatus and infraspinatus tendons, when compared to the teres minor. Guidance of therapeutic injections with US has advantages over other imaging modalities in the use of any standard needle or catheter system, as well as the reduced exposure to ionizing radiation for both patient and interven- tionalist.

Wrist tendonitis can also be evaluated with ultrasound.4 In addition to measurements of thickened tendons and changes in echogeneity, the US images can suggest fluid-filled collections of abscess and cystic masses. Unfortunately the US images are unable to identify bone invasion of the infectious Figure 1 Enhancement of posterior tibial tendon is seen on MRI with increased peritendon fluid. From Prekumar.10 material, which is important in planning treatment.

LIGAMENT PATHOLOGY dons can be visualized with US, while it cannot be seen with MRI (although the associated tendonopathy can be seen with An important consideration in US imaging is its ability to be MRI). used in dynamic evaluation of the musculoskeletal systems, in much the same way that joint instability can be viewed flu- A somewhat older study of jumper’s knee by Fritschy5 oroscopically. Milz8 reported that ligament injuries at the emphasizes the value of US to produce information related to ankle can be evaluated in a way that distinguishes an intact healing as longitudinal studies record changes associated with ligament from one that is torn. 0.2 Tesla MRI imaging with treatment. Abnormalities included thickening or swelling, extremity coil was compared with US scanning at 13 MHz.11 heterogenous structure, and peritendon thickening and irreg- Only six of nine injuries to the anterior tibiofibular ligament ularities. The study did not utilize MRI comparison. The were identified with US (67%). Better sensitivity was seen imaged abnormalities were seen to return to normal in most with the talofibular and calcaneofibular ligaments (17/18 subjects, especially in those with more acute and milder combined correct) when compared to MRI. In two cases, the pathology, and these changes were consistent with their clin- US suggested a partial injury of a ligament that was not con- ical improvement. The authors do not mention whether the firmed by MRI. The posterior ligaments could not be visual- images directly guided treatment. ized at all using sonography. Difficulties for the US image production included the tenderness in the area of contact of the Muscle and tendon injuries of the hamstring muscles have transducer, interference with blood and edema fluids, and also been evaluated in athletes using US.2 Overall, the MRI increased muscle tone that interfered with dynamic evaluation. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 21

Previous study of lower frequency (7.5 MHz) US had pro- the situation of chronic pain with loading is a major reason duced worse results in terms of injury detection. to use MRI in preference to US as a primary imaging modality. However, the largest barrier to the implementation of US as a primary imaging modality is the operator-dependent SOFT TISSUE EDEMA AND JOINT EFFUSION variability in image production and interpretation. There is a continuing shortage of trained personnel who can operate Ultrasound comes up short in the evaluation of possible joint these devices appropriately for their many medical uses. effusion and of possible soft tissue infection. Although the US evaluation can detect areas of fluid collection with An obvious advantage of US is its ability to visualize abnor- reduced echogenicity, the MRI clearly has greater sensitivity malities—especially fluid collections—in close proximity to for joint fluid in addition to the capability of assessing of orthopedic hardware implants and other foreign bodies. For osteomyelitis. Full evaluation of possible abcess with gadolin- this reason alone, the management of musculoskeletal prob- ium does add to the cost of this contrast agent, but there can lems requires some use of US in its routine clinical evaluation also be the added cost of analgesia or sedation for thorough and testing. evaluation of painful, swollen limb with US. In addition, the MR image will typically demonstrate the full extent of the fluid dispersion in the joint and its surrounding tissues with REFERENCES a single examination. 1. Chiou HJ, Chou YH, Wu JJ, Huang TF, Ma HL, Hsu CC, Chang CY. The role of high-resolution ultrasonography in management of PERIPHERAL NERVE IMAGING calcific tendonitis of the rotator cuff. Ultrasound Med Biol 2001;27:735-743. 2. Connell DA, Schneider-Kolsky ME, Hoving JL, Malara F, Evaluation of peripheral nerves has been performed for eval- Buchbinder R, Koulouris G, Burke F, Bass C. Longitudinal study uation of entrapment, nerve trauma, and pathologic changes comparing sonographic and MRI assessments of acute and healing in shape.7,12 Consistent patterns of enlargement of nerves hamstring injuries. AJR Am J Roentgenol 2004;183:975-984. proximal to entrapment seen in carpal tunnel syndrome and 3. Cooper G, Lutz GE, Adler RS. Ultrasound-guided aspiration of symptomatic rotator cuff calcific tendonitis. Am J Phys Med Rehabil ulnar neuropathy have been difficult to apply in patient care 2005;84:81. because of the associated normal variation in the population. 4. Daenen B, Houben G, Bauduin E, Debry R, Magotteaux P. Martinoli’s work has pointed out some cases where a nerve Sonography in wrist tendon pathology. J Clin Ultrasound was inadvertently entrapped under orthopaedic hardware in 2004;32:462-469. a position that appears difficulty to view with either MRI or 5. Fritschy D, de Gautard R. Jumper’s knee and ultrsonography. Am J 7 Sports Med 1988;16:637-640. computed tomography because of the metal. He also 6. Jacobson JA. Musculoskeletal ultrasound and MR imaging. Radiol demonstrates a situation of dynamic entrapment near the Clin North Am 1999;37:713-735. elbow to highlight this aspect of ultrasound imaging. High- 7. Martinoli C, Bianchi S, Pugliese F, Bacigalupo L, Gauglio C, Valle M, resolution scans of superficial digital nerves can now be Derchi LE. Sonography of entrapment neuropathies in the upper accomplished, but limited access to deep, small nerves such limb (wrist excluded). J Clin Ultrasound 2004;32:438-450. as the anterior interosseus represents the inherent problem 8. Milz P, Milz S, Steinborn M, Mittlmeier T, Putz R, Reiser M. Lateral ankle ligaments and tibiofibular syndesmosis: 13-MHz high frequen- with ultrasound lacking depth penetration, and losing resolu- cy sonography and MRI compared in 20 patients. Acta Orthop tion quality when it tries for depth. Scand 1998;69:51-55. 9. Nikken JJ, Oei EH, Ginai AZ, Krestin GP, Verhaar JA, van Vugt AB, Hunink MG. Acute peripheral joint injury: cost and effectiveness of CONCLUSIONS low-field-strength MR imaging—results of randomized controlled trial. Radiology 2005;236:958-967. 10. Premkumar A, Perry MB, Dwyer AJ, Gerber LH, Johnson D, A major limitation of US imaging of the musculoskeletal sys- Venzon D, Shawker TH. Sonography and MR imaging of posterior tems in evaluation of musculoskeletal and pain complaints is tibial tendinopathy. AJR Am J Roentgenol 2002;178:223-232. its depth of penetration for the evaluation of complex prob- 11. Rand T, Bindeus T, Alton K, Voegele T, Kukla C, Stanek C, Imhof lems. The identification of superficial pathology such as H. Low-field magnetic resonance imaging (0.2T) of tendons with tendinosis may result in a failure to diagnose a nearby stress sonographic and histologic correlation. Cadaveric study. Invest Radiol 1998;33:433-438. fracture when both have been caused by the same overuse 12. Walker FO. Neuromuscular ultrasound. Neurol Clin 2004;22:563- activity. Similarly, the inability to diagnose osteomyelitis in 590. 22 AANEM Course 23

Tarsal Tunnel Syndrome is Underdiagnosed

Ernest W. Johnson, MD Emeritus Professor Department of Physical Medicine & Rehabilitation The Ohio State University Columbus, Ohio

INTRODUCTION causes of foot symptoms. It is axiomatic that localized compromise is a common occurrence in vulnerable nerves such as Tarsal tunnel syndrome (TTS) has always been a controver- diabetic peripheral neuropathy. sial subject; however, patients with painful feet deserve a diagnosis from their physician or podiatrist. Tarsal tunnel This author’s analysis of the TTS controversy concludes that syndrome can be a convenient label when the health profes- the only dispute is the significance or validity of the various sional is unable to arrive at a better diagnosis. According to EDX techniques to identify abnormalities of the tibial and the 28th edition of Dorland’s Medical Dictionary, TTS is “a plantar nerves in and distal to, the tarsal tunnel. complex of symptoms resulting from compression of the tibial nerve or of the plantar nerves in the tarsal tunnel, with Some have suggested that TTS and carpal tunnel syndrome pain, numbness, and tingling paresthesias of the sole of the co-exist in the same person. This would lend credence to the foot.” “vulnerable” nerve with multiple entrapments happening in diabetic peripheral neuropathy.14 Others point out the This is a compromise of the tibial nerve in an anatomical increased pressure in the tunnel with marked eversion of the structure usually labeled the tarsal tunnel. The bony floor subtalar joint.9 One could assume that the “tunnel” com- and side is the distal tibia converted to a tunnel by the lanci- presses or irritates the tibial nerve or vascular structures in the natum ligament (Figure 1). Another site of potential entrap- vulnerable area, similar to that occurring so often to the ment is the plantar nerve within the abductor hallucis (AH) median nerve in the carpal tunnel.10 (Figure 2). Either medial or lateral plantar nerve compromise would be a reasonable cause for TTS. Another unusual TTS case reported by Watson and colleagues was caused by the pressure from an inflated ice hock- Many referrals from podiatrists to electrodiagnostic (EDX) ey skate.20 Trepman and colleagues reported their study on physicians continue today because their patients are seeking cadavers measuring tarsal tunnel compartment pressure with relief from foot pain. Some of the reported diagnoses of TTS various foot positions.17 Both extreme eversion and inversion are better characterized as local trauma to the vulnerable loca- raised the pressure. tion of the tibial nerve, e.g., fractures or sprains. Reports of postoperative results often differentiate these, using “idio- Lau and Daniels11 reviewed the literature of TTS, noting it pathic” tarsal tunnel for those patients who lack a history of was an “uncommon clinical entity” and citing published trauma. Unfortunately many reports do not distinguish reports about its anatomy, pathology, cause, clinical presenta- between these two differing presentations and presumed tion, treatment, and outcome. This article was cited 12 times. 24 Tarsal Tunnel Syndrome is Underdiagnosed AANEM Course

Figure 1 The medial aspect of the leg showing the lancinatum ligament. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 25

Figure 2 The foot with medial and lateral plantar nerves. 26 Tarsal Tunnel Syndrome is Underdiagnosed AANEM Course

Jackson and Haglund7 describe TTS in runners as a recent Mixed NCSs should also be performed as shown in Figure 3. recognition and note that conservative treatment is generally To diagnoses TTS, use medial plantar nerve—8-15 µV and successful. the lateral plantar nerve—8-15 µV.

Lastly, sensory NCSs should be performed recording from EVALUATION OF SUSPECTED TARSAL TUNNEL SYNDROME digit 1. This is best performed with near nerve recording. The amplitude is variable. There is no question that TTS exists. The next question, and perhaps the most controversial, is how a physician reaches the Somatosensory Evoked Potentials are Controversial diagnosis. Proper diagnosis includes a complete history and physical as well as complementary EDX studies. Using somatosensory evoked potentials (SEPs) to diagnose TTS is controversial. Dumitru3 presented two cases in which SEPs identified the tibial entrapment when standard needle PHYSICAL EXAMINATION EMG and NCSs did not. This included a description of the technique and normal values. In this author’s opinion, meas- On physical examination, the physician should find tender- uring a localized entrapment in a distal limb with SEPs is ness over the tarsal tunnel, a positive Tinel’s sign over the tib- cumbersome. ial nerve, or the medial or lateral plantar nerve, and atrophy of intrinsic foot muscles with sensory deficit in the sole. Lee and Dellon12 report the value of a Tinel’s sign over the tibial THE VALUE OF ELECTRODIAGNOSIS nerve and its distal branches as a prognostic sign in both diabetic neuropathy and localized neuropathy prior to decom- Felsenthal and colleagues suggested stimulation across the pression of the tibial nerve. Over the course of 10 years, tarsal tunnel and recording the motor latencies of the AH Bailie and Kelikian1 reviewed 47 patients with TTS who had and abductor digiti minimus pedis (ADMP).4 In 1998, Ward undergone surgery. While 81% had supporting EDX studies, and Porter reviewed 22 cases of patients who had undergone they concluded “diagnosis of TTS is made primarily on the an operation for TTS and found that only 8 had successful basis of history and clinical evaluation.” outcome.19 They were unable to predict success by either clinical or EDX evaluation and stated “we question the use of conduction velocity (CV) studies in the diagnosis in TTS and IMAGING STUDIES further more, the role of surgery in TTS.”

Kerr and Frey reported 17 of 19 patients with TTS had Vogt and colleagues found the trans-tarsal motor latency had abnormal findings on magnetic resonance imaging (MRI).8 the greatest specificity but that the mixed nerve latency and Turan found only 2 out of 10 TTS patients had abnormal sensory nerve latency (for the big toe) were more sensitive but MRI scans.18 also seen in mild peripheral neuropathy and thus less specific.6 Galardi and colleagues reviewed 13 cases of TTS and concluded that mixed nerve evaluation was necessary before sur- NEEDLE ELECTROMYOGRAPHY gery.5

Needle electromyography (EMG) examination of a patient Table 1 with TTS should show positive sharp waves and fibrillation potentials in foot intrinsic muscles, reduced recruitment, and Motor NCS: Stimulate distal tibial nerve proximal to medial malleolus increased proportion of polyphasic motor unit action potentials. These are clearly the findings when the tibial nerve or Record abductor digiti mini pedis 4.3 ± 0.6 ms its branches are compromised severely with denervation. Record abductor hallucis 5.1 ± 0.6 ms Lateral and medial planter nerve distal latency difference greater than 1 ms is considered normal. NERVE CONDUCTION STUDY Nota bene (pay attention)—amplitude is important! Across tarsal tunnel motor nerve conduction technique. Stimulate tibial To diagnose TTS, a physician should perform motor nerve nerve proximal and distal to the tarsal tunnel. conduction studies (NCSs) as indicated in Table 1. AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 27

1. History and physical examination – the Tinel’s sign is the most important marker.

2. Distal motor latency

a. Record AH

b. Record ADMP

3. Conduction velocity of the tibial nerve

4. F wave of the tibial nerve (complex in diabetes mellitus)

5. Sural nerve study Figure 3 Mixed nerve stimulation – medial and lateral plantar nerves. 6. Mixed nerve study

a. Medial plantar nerve

Oh and colleagues reported three cases of TTS with motor b. Lateral plantar nerve and sensory conduction improving as well as clinical improvement, 14 months to 3.5 years after surgery.15 They 7. Needle EMG study – intrinsic muscles i.e., AH, first noted that all three cases still had minor abnormalities in sen- dorsal of the interrosseous; fourth dorsal inter- sory conduction. rosseous; and ADMP. Also explore proximal muscles for L4 - L5 and paraspinal S1. Pfeiffer and Cracchiolo reviewed 30 patients who had exploratory surgery and decompression of the tibial nerve.16 Twenty-two patients had a diagnosis of TTS and 18 had CONCLUSION EDX studies which were considered abnormal and support- ive of the diagnosis of TTS. In their summary of these stud- Tarsal tunnel syndrome is a clinical entity that can be identi- ies, they say, “however there was no correlation between the fied with electrodiagnosis. clinical outcome at the latest follow-up of these studies.” It is rare and is more likely to occur in the presence of under- Mondelli and colleagues devised an electrophysiological scale lying generalized neuropathy such as diabetes mellitus. The for the severity of TTS.13 They tested 96 TTS patients using EDX study must be abnormal to reliably confirm this entrap- this scale and found that it correlated well with the progres- ment. sion of symptoms based on the comparison of motor and sensory latencies, the former being less sensitive, and the lat- ter more sensitive. REFERENCES

Watson and colleagues evaluated distal tarsal tunnel release 1. Bailie D, Kelikian A. Tarsal tunnel syndrome: diagnosis, surgical tech- with a partial plantar fasciotomy for chronic subcalcaneal nique and functional outcome. Foot Ankle Int 1998;19:65-72. pain in 75 patients and 88% were found to have “good” 2. Dellon A, Kim J, Spaulding CM. Variations in the origin of the medi- results.20 This supports the suggestion that the medial cal- al calcaneal nerve. J Am Podiatr Med Assoc 2002;92:97-101. 3. Dumitru D, Kalantri A, Dierschke B. Somatosensory evoked poten- caneal branch from the medial plantar nerve occurs in almost tials of medial and lateral plantar and calcaneal nerves. Muscle Nerve half of all patients.2 1991;14:665-671. 4. Felsenthal G, Butler DH, Shear MS. Across tarsal tunnel motor nerve conduction technique. Arch Phys Med Rehabil 1992;73:64-69. RECOMMENDED ELECTRODIAGNOSTIC STEPS 5. Galardi G, Amadio S, Maderna L, Meraviglia MV, Brunati L, Dal Conte G, Comi G. Electrophysiological studies in tarsal tunnel syndrome. Diagnostic reliability of motor distal latency, mixed nerve and When this author performs an EDX study on a patient with sensory nerve conduction studies. Am J Phys Med Rehabil suspected TTS, the following steps are performed: 1994;73:193-198. 28 Tarsal Tunnel Syndrome is Underdiagnosed AANEM Course

6. Herbsthofer B, Vogt T, Karbowski A, Krischek O. The value of nerve 14. Mondelli M, Cioni R. Electrophysiological evidence of a relationship conduction studies in the diagnosis of tarsal tunnel syndrome. between carpal and tarsal tunnel syndromes. Neurophysiol Clin Aktuelle Neurologie 1997;24:156. 1998;28:391-397. 7. Jackson D, Haglund B. Tarsal tunnel syndrome in runners. Sports 15. Oh S, Arnold TW, Park KH, Kim DE. Electrophysiological improve- Med 1992;13:146-149. ment following decompression surgery in tarsal tunnel syndrome. 8. Kerr R, Frey C. MR imaging in tarsal tunnel syndrome. J Comput Muscle Nerve 1991;14:407-410. Assist Tomogr 1991;15:280-286. 16. Pfeiffer W, Cracchiolo A 3rd. Clinical results after tarsal tunnel 9. Kinoshita M, Okuda R, Morikawa J, Jotoku T, Abe M. The dorsiflex- decompression. J Bone Joint Surg Am 1994;76:1222-1230. ion-eversion test for diagnosis of tarsal tunnel syndrome. J Bone Joint 17. Trepman E, Kadel NJ, Chisholm K, Razzano L. Effect of foot and Surg Am 2001;83:1835-1839. ankle position tarsal tunnel compartment pressure. Foot Ankle Int 10. Kohno M, Takahashi H, Segawa H, Sano K. Neurovascular decom- 1999;20:721-726. pression for idiopathic tarsal tunnel syndrome: technical note. J 18. Turan I, Rivero-Melian C, Guntner P, Rolf C. Tarsal tunnel syn- Neurol Neurosurg Psychiatr 2000;69:87-90. drome – outcome of surgery in longstanding cases. Clin Orthop 11. Lau J, Daniels T. Tarsal tunnel syndrome: a review of the literature. Relat Res 1997;343:151-156. Foot Ankle Int 1999;20:201-209. 19. Ward P, Porter M. Tarsal tunnel syndrome: a study of the clinical and 12. Lee C, Dellon A. Prognostic ability of Tinel sign in determining out- neurophysiological results of decompression. JR Coll Surg Edinb come for decompression surgery in diabetic and nondiabetic neu- 1998;43:35-36. ropathy. Ann Plastic Surg 2004;53:523-529. 20. Watson TS, Anderson RB, Davis WH, Kiebzak GM. Distal tarsal 13. Mondelli M, Morana P, Padua L. An electrophysiological severity tunnel release with partial plantar fasciotomy for chronic heel pain: scale in tarsal tunnel syndrome. Acta Neurol Scand 2004;109:284- an outcome analysis. Foot Ankle Int 2002;23:530-537. 289. 29

Tarsal Tunnel Syndrome is Overdiagnosed

Lawrence R. Robinson, MD Professor and Chair Department of Rehabilitation Medicine Vice Dean for Clinical Affairs University of Washington School of Medicine Seattle, Washington

CLINICAL PRESENTATION OF TARSAL TUNNEL SYNDROME confirmed TTS is likely rare. In a series summarizing 33 years of experience with tibial nerve lesions at one of the most pres- Anatomically, the tarsal tunnel is a fibrosseous channel that tigious peripheral nerve surgery centers in the world, only 46 extends from the ankle to the midfoot. The floor is formed cases of the condition were reported (3 of these bilateral).5 by the tibialis posterior, flexor digitorum longus, and flexor Traumatic injury of the tibial nerve at the ankle, though rel- hallucis longus tendons overlying the calcaneus. The roof is atively uncommon, is probably nevertheless a more common formed by the flexor retinaculum. Inside the tarsal tunnel are cause of focal tibial nerve lesions than idiopathic TTS. the tibial nerve, tibial artery, and tibial vein. It is the belief of this author that TTS is often overdiagnosed Symptoms of tarsal tunnel syndrome (TTS) include burning and that physicians need to be very careful in the evaluation and sharp pain on the plantar surface of the foot, as well as of patients with possible TTS. Potential pitfalls extend to dif- pins and needles on the sole of the foot.1 The symptoms are ferential diagnosis, nerve conduction studies (NCSs), needle usually unilateral. Walking or prolonged standing usually electromyography (EMG), and treatment decisions. make symptoms worse. Signs include reduced sensation on the sole of the foot; this can be in the medial plantar, lateral The first problem with the diagnosis of TTS is that there are plantar, and/or medial calcaneal distribution. Some patients many other entities that can present with foot pain; some of may have a “Tinel’s sign” at the tarsal tunnel. More debatable these can be confused with TTS, especially by the uninitiat- are signs of hypermobility of the ankle or midfoot, or a ed. Plantar fasciitis presents with plantar foot pain that is hyperpronated foot. usually worse in the morning when first getting out of bed and then slowly improves during the day. Painful stress frac- Compression of the tibial nerve at the posterior tarsal tunnel tures of the metatarsals are common in some physically (i.e., posterior TTS) is reportedly the most common entrap- demanding activities such as ballet or running, but can also ment for the tibial nerve. Unfortunately, since clinical diag- present insidiously in the less physically active individual. nostic criteria are not well established and there are no good Rheumatological conditions can also present with foot pain epidemiologic studies, the real incidence and prevalence of and foot deformity, though these are usually apparent on TTS are unknown. However, in this author’s opinion, truly radiologic studies. 30 Tarsal Tunnel Syndrome is Overdiagnosed AANEM Course

While TTS should usually present with sensory complaints, there are also a number of other conditions that can similar- Table 1 Nerve Conduction in TTS. Galardi and colleagues, 1994 ly present with foot numbness. Polyneuropathy typically starts with numbness in the feet since the longest nerves are Technique Sensitivity Specificity typically first affected. With lumbar spinal stenosis or lum- CMAP latency 22% 100% bosacral radiculopathy, the L5 and S1 roots are most com- SNAP Med Plantar 93% 92% monly affected. Foot numbness can result from impairment SNAP Lat Plantar 100% 83% of these root levels. Morton’s neuroma, an interdigital neuro- CNAP Med Plantar 64% 100% ma usually occurring between the 3rd and 4th metatarsals can also produce sensory symptoms with pain and paresthe- CNAP Lat Plantar 71% 100% siae, although this does not usually involve the entire plantar CMAP = compound muscle action potential; CNAP = compound nerve action surface of the foot. potential; SNAP = sensory nerve action potential; TTS = tarsal tunnel syndrome.

ELECTRODIAGNOSIS OF TARSAL TUNNEL SYNDROME There are, however, a number of significant problems with Although there are many papers written about the electrodi- using NCSs for diagnosis of TTS. First temperature is often agnosis of TTS, there are few well-controlled, prospective difficult to control in the foot. Since the foot is most distal it studies. There is an excellent, recent review written by a task can be very difficult to control nerve temperature—especial- force of the American Association of Neuromuscular & ly in cold weather, in those with vascular disease, or those Electrodiagnostic Medicine that those interested in the diag- with Raynaud’s phenomena. In this author’s experience, even nosis of TTS should read.6 In this review, however, only three warm water baths and heating pads have not been universal- original papers met the authors’ criteria for acceptance. These ly successful. Correction of measured values for temperature criteria did not require the use of prospectively recruited con- is problematic when the nerve is very cold, as error in correc- trols. tion values multiplied by many degrees will produce very large errors in the calculated latency. There are a number of NCSs described for diagnosing TTS. They fall into three major categories. First, there are motor Second, although there are standard values for latency at set conduction studies stimulating at the ankle and recording the distances, there is considerable variation in foot size making compound muscle action potential (CMAP) from muscles it impossible to use these set distances in all individuals. In supplied by the medial and lateral plantar nerves, e.g., abduc- this author’s size 13 feet, for example, it’s impossible to record tor hallucis (AH) and abductor digiti quinti pedis. Second, above the medial malleolus and stimulate in the sole 14 cm there are orthodromic pure sensory studies stimulating at the distally; a 14 cm distance would not reach the lateral plantar toes and recording the sensory nerve action potential above nerve and only barely reaches the medial plantar nerve. the ankle—these typically require near nerve sensory recording techniques due to the small amplitudes of the responses. A third problem is stimulation of the nerves in the soles of Finally, there are mixed nerve studies of medial and lateral the feet. Often this area is calloused and exceedingly difficult plantar nerves stimulating both motor (antidromically) and to stimulate. Stimulation with higher current produces a very sensory (orthodromically) fibers in the sole and recording the large artifact that makes it even more difficult to record the compound nerve action potential (CNAP) above the ankle. response. Use of needle stimulation through the skin of the There are also a number of variations of the above as well. sole is very painful and this author has tried it only once. Use of a needle just through the skin epidermal layer could, how- Most studies of NCSs in TTS have been unblinded and ret- ever, be useful. rospective without a concurrent control group. These studies have also been hampered by the lack of a standardized clini- A fourth problem is the waveform of the motor responses. cal definition for the presence of TTS. There has been one Often, due to the multiple foot muscles depolarizing as a controlled study examining these three techniques in 14 result of tibial nerve stimulation, these CMAPs have an ini- patients and 12 control subjects.2 The authors found the sen- tial positive deflection making determination of latency more sitivity and specificity noted in Table 1. It would appear from uncertain than usual. this table that the CNAP recording, i.e., the third category described above may have the greatest value; i.e., acceptable Finally, the small amplitudes of the responses are problemat- specificity and sensitivity. ic. With stimulation of the toes, the responses at the ankle are AANEM Course CROSSFIRE: Controversies in Neuromuscular and Electrodiagnostic Medicine 31 so small that usually near nerve recording is required. Even One study has documented magnetic resonance imaging the CNAP responses are small. When a response is absent, findings in TTS (Table 2), but it is unclear how often similar the clinician is uncertain whether a very small and possibly findings can be seen in the asymptomatic population.4 normal response is present, or no response is present.

Needle EMG has also been used for the diagnosis of TTS. TREATMENT OF TARSAL TUNNEL SYNDROME Typically, muscles supplied by the medial and the lateral plantar nerves are examined. For example, the 1st dorsal Although not the focus of this manuscript, treatment of TTS interosseous pedis and AH can be studied fairly easily. When is a consideration for the electrodiagnostic (EDX) physician. abnormalities are found in these muscles, it is usually wise to For the diagnosis of diseases that have little risk from treat- examine other muscles to look for other explanations. ment and/or generally good outcomes, it is acceptable to have Examination of other S1 innervated muscles (e.g., gluteus more frequent false positive results. When a diagnosis leads to maximus or biceps femoris) could theoretically be used to a potentially harmful treatment, an EDX physician should be rule out radiculopathy. Similarly study of other distal muscles more concerned with the specificity (i.e., false positive rate). can be used to look for distal axon loss as seen in polyneuropathy. Treatment of TTS has a number of potential approaches. First there are biomechanical interventions. Both improved While needle EMG theoretically should be useful, there are arch support and custom-molded shoe orthotics can poten- multiple problems that hinder its use. First, there are varying tially improve ankle positioning and reduce stress (stretching reports of false positive needle EMG findings in the feet, like- and/or pressure) on the tibial nerve. Oral medications such as ly representing chronic foot trauma; these muscles typically gabapentin or antidepressants can be used for pain control. have false positive rates of 10-15%.3 Second, unlike larger Some have advocated the use of steroid injections at the proximal limb muscles, it is difficult to fully sample these ankle. small intrinsic foot muscles in four different quadrants. Third, while radiculopathy and axonal polyneuropathy Surgical release has been reported to have rates of good out- should theoretically affect other muscles outside the feet, it’s comes varying from 40% (i.e., about the placebo rate) to not uncommon for these disorder to have a predilection for nearly 100%. Studies are limited by a variety of factors. First, intrinsic foot muscles. Finally, there have not been adequate- most success rates are measured by the surgeon performing ly controlled studies of the utility of needle EMG for diagno- the procedure. Second, objective evaluation has been limited sis of TTS. and most studies report patient self-assessment of outcome. Third, follow-up periods have been variable, with some studies having very short follow-up. As mentioned earlier, in 33 PROPOSED ETIOLOGIES OF TARSAL TUNNEL SYNDROME years of experience at Louisiana State University, Kline and col- leagues2 have operated on 43 patients with TTS (3 bilateral), As a clinical syndrome, TTS may result from a number of using fairly stringent criteria for diagnosis. Of these highly different types of pathophysiology. It is unclear, at least to selected patients, 79% showed improvement with primary this author, what should be included under the rubric of TTS. It is probably common, at least in major trauma centers, to see injury of the tibial nerve resulting from trauma to the foot. This can be either immediate or may possibly result Table 2 MRI in Tarsal Tunnel Syndrome from post-traumatic fibrosis. Symptoms are similar to those reported for TTS. There have also been other reported etiolo- (33 cases: Kerr and colleagues, 1991) gies in the literature including ganglion or tumor, anomalous • Varicosities 8 muscles (flexor digitorum accesorius longus), and rheuma- • FHL Tenosynovitis 6 toid arthritis. The above causes of tibial neuropathy at the • Normal 6 ankle are often agreed upon in the literature. • Fracture / soft tissue injury 5 There are also a number of, what this author considers, dis- • Mass lesions 5 puted causes of TTS. These include thrombophlebitis or • Fibrous scar 2 venous variscosities at the ankle, and joint hypermobility or • Abductor Hallucis Hypertrophy 1 foot hyperpronation. It is less clear what role these disorders might have in the development of symptoms of TTS or how FHL = flexor hallucis longus; MRI = magnetic resonance imaging; amenable these are to any type of intervention. 32 Tarsal Tunnel Syndrome is Overdiagnosed AANEM Course operation and 56% showed improvement with re-operation REFERENCES after initial operation elsewhere. It is likely that results are not as good in a less stringently chosen population operated on by 1. DeLisa JA, Saeed MA. The tarsal tunnel syndrome. Muscle Nerve someone with less peripheral nerve experience. 1983;6:664-670. 2. Galardi G, Amadio S, Maderna L, Meraviglia MV, Brunati L, Dal Conte G, Comi G. Electrophysiologic studies in tarsal tunnel syndrome. Diagnostic reliability of motor distal latency, mixed nerve and SUMMARY sensory nerve conduction studies. Am J Phys Med Rehabil 1994;73:193-198. In summary, the differential diagnosis of foot pain and 3. Gatens PF, Saeed MA. Electromyographic findings in the intrinsic numbness is complex. Tarsal tunnel syndrome is but one of muscles of normal feet. Arch Phys Med Rehabil 1982;63:317-318. 4. Kerr R, Frey C. MR imaging in tarsal tunnel syndrome. J Comput the many possible diagnoses that should be considered. Assist Tomogr 1991;15:280-286. Nerve conduction studies have significant limitations and 5. Kim DH, Cho YJ, Ryu S, Tiel RL, Kline DG. Surgical management needle EMG is difficult to interpret in intrinsic foot muscles. and results of 135 tibial nerve lesions at the Louisiana State University Health Sciences Center. Neurosurgery 2003;53:1114-1125. Since treatment can be invasive and is not always successful, 6. Patel AT, Gaines K, Malamut, R, Park TA,Del Toro DR, Holland N. it is recommended that the EDX physician maintain a rea- AANEM Usefulness of electrodiagnostic techniques in the evaluation of suspected tarsal tunnel syndrome: an evidence-based review. sonably high threshold for diagnosis. Not only should symp- Muscle Nerve 2005;32:236-240. toms and signs be consistent, but also the EDX physician should see selectively prolonged distal latencies of medial and/or plantar nerves with normal latencies in other NCSs. Temperature control is essential and needle EMG changes should be interpreted with caution. AANEM

421 First Avenue SW, Suite 300 East Rochester, MN 55902 (507) 288-0100 / Fax: (507) 288-1225 [email protected] www.aanem.org