SHORT REPORT ABSTRACT: This 16-year retrospective study reports the data on 456 consecutive patients with 557 peripheral (PNIs) between 1989 and 2004. Most patients were men (74%) and the mean age was 32.4 years. In 83% of cases the PNIs were isolated; combined lesions most commonly involved the ulnar and median . Upper-limb PNIs occurred in 73.5% of cases; the ulnar nerve was most often injured, either singly or in combi- nation. Vehicle accidents affecting the brachial plexus or radial, sciatic, facial, and peroneal nerves were the most common cause of . Pene- trating trauma commonly affected the ulnar and median nerves; falls and gunshot frequently affected the ulnar, radial, and median nerves; and sports injuries, particularly soccer, affected mainly the peroneal and tibial nerves. More than half of the brachial plexus lesions after vehicular accidents were from motorcycle crashes. Muscle Nerve 34: 785–788, 2006

PERIPHERAL NERVE INJURIES: A RETROSPECTIVE SURVEY OF 456 CASES

JOA˜ O ARIS KOUYOUMDJIAN, MD, PhD

Department of Neurological Sciences, State Medical School (FAMERP), Sa˜o Jose´ do Rio Preto, Sa˜o Paulo, Brazil

Accepted 31 May 2006

Peripheral nerve injuries (PNIs) are a major medi- after a PNI, not so early that severe axonal degener- cal problem caused mainly by motor vehicle acci- ation is missed or too late when possible surgical dents, after stabbing incidents, benefits are minimal.12 Clinical and electrophysio- gun-shots, and stretching or crushing after falls. logical changes should be carefully followed on a Fractures of adjacent bones are commonly associ- monthly basis to evaluate possible early reinnerva- ated with PNIs, such as humeral fractures with radial tion. Patients frequently have multiple injuries and neuropathy.11 In addition to penetrating injury caus- various medical specialties may be involved in their ing sharp or blunt transections/lacerations, nerves care. Thus, clinical–electrophysiological correla- may be displaced, contused, stretched, or even par- tions may be missed or misinterpreted. The purpose tially divided, leading to neuromas or lesions in con- of this study was to analyze data from 456 patients tinuity.6 The incidence of PNIs is frequently under- referred for EMG after symptomatic PNI. estimated and, surprisingly, few large general series have been published worldwide. METHODS Electromyography (EMG) is the most appropri- From January 1989 to December 2004, 456 consec- ate diagnostic method in the evaluation of the to- utive subjects were referred for electrodiagnostic pography and severity of PNIs. Magnetic resonance consultation due to PNIs; only the first EMG re- imaging also provides useful additional information quested after trauma was considered in this study. All about root avulsion after traction injury to the bra- subjects were symptomatic, reporting either motor chial plexus. Patients should be referred for electro- (weakness) or sensory (positive or negative) symp- diagnostic consultations within a 6-month period toms. Trauma to the spine with consequent acute, subacute, or chronic , , or cauda equina syndrome was not evaluated in this Abbreviations: CMAP, compound muscular action potential; CNS, central study. From the total of 456 cases included in the nervous system; EMG, electromyography; MUAP, motor unit action potential; PNI, peripheral ; PSW, positive sharp wave; SNAP, sensory nerve study, 557 nerve lesions were identified in the face action potential and limbs. Data on age, gender, referring specialty, Key words: electromyography; nerve conduction studies; nerve injury; pe- ripheral neuropathy; trauma and interval between injury and first EMG were ob- Correspondence to: J. A. Kouyoumdjian, R. Luiz Antoˆ nio Silveira 1661, tained. 15025-020 Sa˜ o Jose´ do Rio Preto, SP, Brazil; e-mail: [email protected] All patients underwent EMG performed by the © 2006 Wiley Periodicals, Inc. Published online 31 July 2006 in Wiley InterScience (www.interscience.wiley. author and the results were analyzed retrospectively. com). DOI 10.1002/mus.20624 Electrophysiological involvement was graded using

Short Reports MUSCLE & NERVE December 2006 785 Table 1. Injury categories to peripheral nerves in 437 cases involving 534 nerve trunks. Cases Nerves Category (n) Percentage (n) Percentage A Vehicle accidents 192 43.9 237 44.4 B Penetrating traumas 90 20.6 105 19.7 C Falls 58 13.3 68 12.7 D Gun-shot wounds 37 8.5 48 9.0 E Car accidents involving pedestrian 17 3.9 23 4.3 F Sports 12 2.7 16 3.0 G Miscellaneous 31 7.1 37 6.9 Total 437 534

the Seddon classification.13 Neurotmesis was diag- the upper limbs represented 73.5% of instances; to nosed on the basis of absence of sensory nerve action lower limbs, 21.5%; and to the face, 5%. In 2 lower- potential (SNAP) or compound muscular action po- limb cases and in another 2 involving the facial tential (CMAP) after distal stimulation of the af- nerve, nerve injuries were bilateral. fected nerve, profuse fibrillation potentials and pos- PNI categories were identified in 437 of 456 cases itive sharp wave potentials (PSWs), and absence of (95.8%), or 534 nerves (Table 1). There were 377 any volitionally recruited motor unit action poten- (83%) cases with single and 79 (17%) with multiple tials (MUAPs). In , mild to profuse fi- PNIs. Multiple nerve lesions involved two nerves in brillation potentials and PSWs were present and 62 cases, three nerves in 13 cases, four nerves in 3 MUAP recruitment was reduced. Neurapraxia con- cases, and five nerves in 1 case. The most frequent sisted of absent or rare fibrillation potentials and isolated nerve injuries, in descending order, were to PSWs, normal MUAP morphology with reduced re- the ulnar, brachial plexus, radial, median, peroneal, cruitment, and normal or reduced/absent SNAP or and sciatic nerves. The most common two-nerve in- CMAP amplitude, depending on whether stimulation juries were the ulnar plus median (37.1%), peroneal was distal or proximal to the conduction block. MUAP plus tibial (14.5%), sciatic plus femoral (6.5%), and morphology from reinnervation was also evaluated. radial plus axillary (4.8%). The most common three- The brachial plexus (and lumbar and lumbosa- nerve combination was ulnar, median, and radial. cral plexus) was considered one nerve regardless of The frequencies of the most common PNIs in any associated root avulsion. All terminal hand each trauma category are given in Table 2. branches of the median or ulnar nerves were also Neurotmesis occurred in 41% cases, axonotmesis considered together as median or ulnar lesions. Sen- in 45%, and neurapraxia in 14%. There were more sory nerve injuries were considered as neurotmesis cases of complete denervation with peroneal nerves when there was no SNAP response. Patients were (65%), followed by brachial plexus (51%), supras- classified as having single or multiple PNIs; the max- capular (50%), sciatic (49%), facial (48%), and pos- imum number was five affected nerves in one pa- terior interosseous (47%) nerves. tient. Descriptive statistics used included means and DISCUSSION frequencies. The protocol for the retrospective study was approved by the institutional ethics committee. The limitations of this work are that the data came from a private practice in which EMG consultation occurred at a variable interval after trauma, lead- RESULTS ing to underestimates of severity. The series also The 456 PNI patients constituted 3.7% of all cases includes many patients without central nervous referred for electrodiagnostic consultations. There system (CNS) lesions at the time of EMG consul- were 337 men (74%) and 119 women (26%). The tation. Because patients with severe CNS involve- mean age of patients was 32.4 years, ranging from 4 ment go to rehabilitation centers, their PNIs may to 79 years; the women were slightly older than the well be misdiagnosed as secondary medical prob- men (36.0 vs. 31.1 years). EMG was performed lems.3,4,14,15 within the first 180 days after PNI in 67% of cases, Young and male patients are more likely to between 6 and 24 months in 23.5%, and at more suffer PNIs. In fact, 95% of motorcycle accidents than 2 years after injury in 9.5%. Injuries to nerves of involved men and more than half of brachial

786 Short Reports MUSCLE & NERVE December 2006 Table 2. Percentage of individual nerve injuries in the different injury categories. Trauma ABCDEFG

Total number 237 105 68 48 23 16 37

Axillary 6.8 0.0 8.8 4.2 4.3 18.8 0.0 Brachial plexus 19.8 1.9 8.8 10.4 13.0 0.0 8.1 Facial 9.3 0.0 1.5 0.0 4.3 0.0 2.7 Femoral 2.1 0.0 1.5 2.1 4.3 0.0 0.0 Inferior gluteal 0.8 0.0 0.0 0.0 0.0 0.0 0.0 Interosseous anterior 0.0 1.0 0.0 0.0 0.0 0.0 0.0 Interosseous posterior 2.1 1.9 4.4 2.1 0.0 6.3 13.5 Lumbar plexus 0.0 0.0 0.0 2.1 0.0 0.0 0.0 Lumbosacral plexus 0.4 0.0 0.0 0.0 0.0 0.0 0.0 Median 5.1 35.2 16.2 16.7 4.3 0.0 27.0 Musculocutaneous 3.0 1.0 1.5 0.0 0.0 6.3 0.0 Peroneal 9.3 1.9 2.9 8.3 21.7 37.5 10.8 Radial 16.5 2.9 19.1 14.6 8.7 6.3 16.2 Radial superficial 1.3 1.0 1.5 2.1 0.0 0.0 0.0 Sciatic 11.4 5.7 4.4 10.4 4.3 6.3 0.0 Superior gluteal 0.8 0.0 0.0 0.0 0.0 0.0 0.0 Suprascapular 1.3 1.0 1.5 2.1 4.3 6.3 0.0 Sural 0.4 0.0 0.0 0.0 8.7 0.0 0.0 Tibial 1.7 0.0 0.0 4.2 0.0 12.5 5.4 Ulnar 8.0 45.7 27.9 20.8 21.7 0.0 16.2 Ulnar, cutaneous dorsal 0.0 1.0 0.0 0.0 0.0 0.0 0.0

Data expressed in percent. A, vehicle accidents (car, 116 cases; motorcycle, 63 cases); B, penetrating traumas (glass wounds, 49 cases; knife, 23 cases); C, falls (stumbling, 35 cases; over animals, 17 cases); D, gun-shot wounds; E, car–pedestrian accidents; F, sports (soccer, 8 cases); G, miscellaneous. Most frequent in each category is in bold type.

plexus injuries resulting from vehicular injury re- In spite of population differences, the litera- sulted from motorcycle accidents. Trauma related ture is consistent in reporting a higher incidence to sports only affected men, particularly soccer of PNIs of the upper limbs,2,10 although the most players, with peroneal nerve lesions being the frequently affected nerves in earlier reports were most common. Most PNIs related to vehicle acci- the ulnar,1 radial,10 or digital9 nerves. For the dents. Almost 90% of facial nerve traumas were in lower limbs the most frequently affected nerves this category; half of these related to motorcycle are reportedly the peroneal10 or sciatic nerves.9 accidents, probably because a helmet was not Men were more frequently involved, in as much as worn. Ulnar nerve injury was the most frequent 96%,2 83%,10 80%,1 and 74.2%9 of cases. In the nerve lesion, occurring either in isolation or in different reported series, the most common association with median nerve involvement; it was trauma category was variable; penetrating trauma the most common nerve injury following penetrat- was responsible for 89.4% in one series,9 gun-shots ing trauma. The association of ulnar and median for 83.9% in another,2 and motor vehicle acci- nerve lesions was the most frequent due to their dents for 46% in a third series.10 Most patients close anatomical relationship, specifically in the were from 16 to 38 years old.2,9,10 Two studies distal forearm and wrist, where penetrating specifically referred to PNIs and sports accidents, trauma was very common, presumably because of with the upper limbs being the most frequently the greater exposure to knife and glass wounds in affected.5,7,8 The right and left sides suffer manual tasks. Penetrating traumas lead to transec- similar incidence rates for PNIs, as reported else- tions or lacerations; the instrument involved may where9,10 and supported by the present findings. be sharp, when little force is necessary to divide EMG was performed within the appropriate period of the nerve, or blunt, requiring a greater force for less than 180 days after PNI in 67%. EMG later than transection.6 All sural nerve lesions were caused by this time may be less helpful for surgical decision- car accidents involving pedestrian. making.

Short Reports MUSCLE & NERVE December 2006 787 This work was supported by a BAP research grant from the State 8. Krivickas LS, Wilbourn AJ. Sports and peripheral nerve inju- Medical School of Sa˜o Jose´ do Rio Preto. The author thanks David ries: report of 190 injuries evaluated in a single electromyo- A. Hewitt for help with the English. graphy laboratory. Muscle Nerve 1998;21:1092–1094. 9. McAllister RM, Gilbert SE, Calder JS, Smith PJ. The epidemi- ology and management of upper limb peripheral nerve inju- ries in modern practice. J Hand Surg [Br] 1996;21:4–13. REFERENCES 10. Noble J, Munro CA, Prasad VS, Midha R. Analysis of upper 1. Adeyemi-Doro HO. Pattern of peripheral traumatic neurop- and lower extremity peripheral nerve injuries in a popula- athy of the upper limb in Lagos. Injury 1988;19:329–332. tion of patients with multiple injuries. J Trauma 1998;45: 2. Babar SM. Peripheral nerve injuries in a Third World country. 116–122. Cent Afr J Med 1993;39:120–125. 11. Robinson LR. Traumatic injury to peripheral nerves. American 3. Cosgrove JL, Vargo M, Reidy ME. A prospective study of Association of Neuromuscular and Electrodiagnostic Medicine. peripheral nerve lesions occurring in traumatic brain-injured Syllabus. Assessment of traumatic nerve injuries; 2005. p 1–11. patients. Am J Phys Med Rehabil 1989;68:15–17. 12. Robinson LR. Traumatic injury to peripheral nerves. Muscle 4. Garland DE, Bailey S. Undetected injured in head-injured Nerve 2000;23:863–873. adults. Clin Orthop Rel Res 1981;155:162–165. 13. Seddon HJ. Surgical disorders of the peripheral nerves. New 5. Hirasawa Y, Sakakida K. Sports and peripheral nerve injury. York: Churchill Livingstone; 1975. Am J Sports Med 1983;11:420–426. 6. Kline DG. Surgical management of nerve injuries. American 14. Selecki BR, Ring IT, Simpson DA, Vanderfield GK, Sewel MF. Association of Neuromuscular and Electrodiagnostic Medi- Trauma to the central and peripheral nervous systems. Part II: cine. Syllabus. Assessment of traumatic nerve injuries; 2005. p A statistical profile of surgical treatment New South Wales 21–18. 1977. Aust NZ J Surg 1982;52:111–116. 7. Krivickas LS, Wilbourn AJ. Peripheral nerve injuries in ath- 15. Stone L, Keenan MA. Peripheral nerve injuries in the adult letes: a case series of over 200 injuries. Semin Neurol 2000; with . Clin Orthop Rel Res 1988;233: 20:225–232. 136–144.

788 Short Reports MUSCLE & NERVE December 2006