Closed Fractures Complicated by Peripheral Nerve Injury Abstract Closed fractures may be complicated by associated peripheral nerve injury. However, because clinical information is limited, determining the best course of treatment is difficult. Most patients with closed fractures have a local nerve injury without nerve division; their prognosis for recovery is favorable. In the acute setting, immediate surgery is usually unwarranted because of the difficulty in accurately defining the severity and extent of nerve injury. When débridement of an open fracture or repair is not required, peripheral nerve injuries are best observed and the extremity treated with splinting and exercise to prevent loss of joint motion. Patients who fail to demonstrate signs of recovery at 6 months, either clinically or with electrodiagnostic testing, should undergo exploration to maximize the likelihood for return of function. When, during exploration, the nerve is in continuity, intraoperative measurement of nerve action potentials should be done. Measuring nerve action potentials will determine whether nerve grafting, local neurolysis, or excision of the injured segment, accompanied by primary repair, is the most appropriate treatment. A lthough peripheral nerve inju- ries are associated with almost every type of fracture, little consen- sus exists on the best methods for evaluation and management of these injuries. Few clinical studies demon- strate consistent results to guide treatment. Further complicating treatment choice is the fact that many widely accepted strategies are poorly substantiated by the available literature. A framework for the ap- propriate approach to these injuries should be based, whenever possible, on current understanding of the pathophysiology and natural history of peripheral nerve injuries. Incidence The overall incidence of peripheral nerve injuries associated with closed fractures is difficult to discern be- cause of the lack of prospectively ac- quired data. Noble et al 1 reported on a prospectively collected database of 5,777 multiply injured patients treated at a large regional trauma center. Patients were primarily young (mean age, 34.6 years) and male (83%). Most experienced high- energy trauma during motor vehicle accidents (51.6%) or motorcycle ac- cidents (9.9%). Humeral fractures were associated with radial nerve in- L. Randall Mohler, MD Douglas P. Hanel, MD Dr. Mohler is Fellow, Section of Hand and Microvascular Surgery, Department of Orthopaedics and Sports Medicine, University of Washington, Seattle, WA. Dr. Hanel is Professor, Section of Hand and Microvascular Surgery, Department of Orthopaedics and Sports Medicine, University of Washington, Seattle. None of the following authors or the departments with which they are affiliated has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article: Dr. Mohler and Dr. Hanel. Reprint requests: Dr. Hanel, Department of Orthopaedics and Sports Medicine, University of Washington, Box 359798, 325 Ninth Avenue, Seattle, WA 98104- 2499. J Am Acad Orthop Surg 2006;14: 32-37 Copyright 2006 by the American Academy of Orthopaedic Surgeons. 32 Journal of the American Academy of Orthopaedic Surgeons jury in 9.5% of cases, with ulnar nerve injury in 3.8%, and with medi- an nerve injury in 1.4%. Fractures of the radius and ulna were associated with ulnar nerve injury in 2.4% of cases and with median nerve injury in 1.3%. Pelvic fractures were asso- ciated with sciatic nerve injury in 1.1% of cases and with femoral nerve injury in 0.16%. Femoral frac- tures were associated with sciatic nerve injury in 1.1% of cases. Tibial and fibular fractures were associated with peroneal nerve injury in 2.2% of cases and with tibial nerve injury in 0.5%. Overall, the radial nerve was the most frequently injured nerve; in the lower limb, the perone- al nerve was most commonly in- jured. Not included in this series were nerve root avulsions and inju- ries to the brachial and lumbosacral plexus (primarily the result of trac- tion mechanisms). Because the pa- tient population was not limited to closed injuries, the finding of a low- er incidence of peripheral nerve inju- ry in this group was expected. 1 Generally, there is a higher inci- dence of peripheral nerve injuries as- sociated with fractures in the upper extremity than with fractures in the lower extremity. In a prospective clinical study of 101 patients who sustained shoulder dislocations and humeral neck fractures, 45% had evidence of peripheral nerve injury on physical examination, confirmed by electrodiagnostic study. 2 The ax- illary nerve was most commonly in- jured (37%), followed by the su- prascapular (29%), radial (22%), musculocutaneous (19%), and ulnar (8%) nerves. Nerve injury was more frequent in patients aged ≥65 years (54%) than in those aged <65 years (26%). Fifty-seven patients had hu- meral neck fractures, but this group was not assessed independently of those with concomitant disloca- tions. The radial nerve is the most com- monly injured nerve in association with fractures of the humerus. Frac- tures of the middle and distal thirds of the humerus are particularly apt to have accompanying nerve damage because this is where the radial nerve is in closest contact with the bone. 3-8 Although the radial, or mus- culospiral, groove of the humerus is frequently described as containing the radial nerve and deep brachial ar- tery, this groove is actually the ori- gin of the brachialis muscle. The ra- dial nerve is separated proximally from the humerus by fibers of the medial head of the triceps and of the brachialis. Only as the nerve ap- proaches the lateral supracondylar ridge is it in direct contact with the humerus. 9 Supracondylar humerus fractures in skeletally immature patients also are frequently associated with neu- rologic complications. Any of the three major nerves of the forearm may be involved. In a review of 162 displaced supracondylar fractures in children, traumatic injury to the ra- dial nerve occurred in 7% of cases, median nerve injury in 3%, and ul- nar nerve injury in 1%. 10 Two retro- spective reviews 11,12 noted a high in- cidence of isolated injury to the anterior interosseous branch of the median nerve with supracondylar humerus fractures in children. Cramer et al 11 noted that in a cohort of 101 pediatric supracondylar frac- tures, 12 of the 15 patients with identifiable nerve involvement had involvement of the anterior in- terosseous nerve. Similarly, Dor- mans et al 12 found that among 200 pediatric patients treated for supra- condylar humerus fractures, 19 had associated nerve injuries: seven in- volved the anterior interosseous branch of the median nerve alone, and four were complete median nerve lesions. Five radial nerve and three ulnar lesions were also identi- fied. In two other studies, Brown and Zinar 13 and Mehlman et al 14 report- ed on the iatrogenic complications of treating pediatric supracondylar humerus fractures. These studies in- dicate that the ulnar nerve was in- jured, albeit transiently, in 2% of pa- tients. Natural History Most reports of peripheral nerve in- jury associated with closed fracture suggest that nonsurgical treatment leads to recovery of nerve function in nearly all patients. In a prospective clinical and electrodiagnostic study of nerve injuries associated with shoulder dislocations and humeral neck fractures, de Laat et al 2 found that 82% of 45 patients with neuro- logic injury recovered well within 4 months. Pollock et al 4 reviewed the data of 23 patients with radial nerve injury associated with closed hu- meral shaft fractures, all treated with closed management. Complete spon- taneous recovery of radial nerve function occurred in all but one pa- tient (96%). The patient without spontaneous return of function at 3 months underwent electrodiagnostic testing, which showed evidence of distal denervation. Exploration 14 weeks after injury revealed that the nerve was trapped in the callus of the healed fracture. Neurolysis provided a complete recovery 8 months later. Brown and Zinar 13 identified 23 neural injuries among 162 displaced supracondylar fractures in children. Eighteen patients sustained the nerve deficits at the time of injury; there were twelve radial, six ulnar, and five median neuropathies. An additional five nerve injuries were iatrogenic: four ulnar nerve injuries and one radial nerve injury. All of the deficits resolved spontaneously in 2 to 6 months (mean, 2.3 months). In two other studies dealing with pe- diatric supracondylar humerus frac- tures, McGraw et al 10 and Cramer et al 11 followed 138 and 101 patients, respectively. They identified a 12% to 15% incidence of injury, with complete recovery in all but one pa- tient in each series. Recovery oc- curred as late as 9 months after in- jury. L. Randall Mohler, MD, and Douglas P. Hanel, MD Volume 14, Number 1, January 2006 33 Results of Treatment The role and timing of surgical ex- ploration are controversial aspects of managing these nerve injuries. No prospective or comparative stud- ies exist to help delineate the appro- priate method of treatment. How- ever, Sonneveld et al 6 reviewed 17 cases of humeral fracture associated with radial nerve paralysis, 16 of which were closed. The radial nerve was explored acutely in the 14 frac- tures that were treated surgically. Thirteen of these nerves appeared to be undamaged; the remaining nerve was contused and showed division of a small number of fibers. Clinical recovery was complete in 12 of the 14 patients (including the patient with gross nerve damage) and in- complete in the remaining two pa- tients. One of the radial nerves that on initial inspection appeared to be uninjured failed to show any evi- dence of clinical recover y. During re-exploration 8 months after injury, the nerve was found trapped in the intermuscular septum; the ner ve was freed, and complete recovery was noted 3 years after injury. As noted previously, in 14 of the 17 ra- dial nerve injuries explored, 6 the re- maining three were treated nonsur- gically, and all three experienced complete nerve recovery. The au- thors concluded that routine explo- ration was not warranted. Similarly, Böstman and col- leagues 7,8 examined 75 patients with radial nerve palsy complicating a fracture of the humeral shaft: 59 with immediate palsy (occur ring at the time of injury) and 16 with sec- ondary palsy (occurring as a result of fracture manipulation). No distinc- tion was made between closed and open injuries. Early ner ve explora- tion and internal fracture fixation (within 3 weeks) was performed in 37 patients (27 immediate and 10 secondary palsies). Thirty-eight pa- tients (32 immediate and 6 second- ary palsies) were treated with initial observation alone. Of the latter group, 26 patients experienced spon- taneous recovery and were not ex- plored; the remaining 12 patients who failed to show early spontane- ous recovery underwent delayed ex- ploration (at an average of 17 weeks postinjury). Because the patients treated with early exploration neces- sarily included a certain number of patients who would have recovered spontaneously, the authors com- pared the final outcome of those who underwent early exploration with the outcome of those initially treated expectantly. Complete re- covery was documented in 73% of patients who underwent early explo- ration and in 87% of the group treat- ed with initial expectance (including 12 of 38 with late exploration). Al- though the more severe fractures in this study may have prompted early exploration, it is also possible that some nerves potentially able to re- cover spontaneously were addition- ally damaged during exploration. Böstman and colleagues 7,8 concluded that routine early exploration could not be supported, suggesting that the choice between open and closed treatment is dictated by the nature of the fracture and not by the func- tion of the nerve. A recent study by Ring et al 15 reinforces these conclu- sions. In any series of humeral fractures, a subset of patients has normal radi- al nerve function following fracture but subsequently develops radial nerve palsy during closed treatment. Even surgeons who advocate obser- vation of primary radial nerve palsy have recommended early explora- tion in these instances of secondary peripheral nerve injuries. 5 Shah and Bhatti 5 identified 16 patients with humeral fractures who developed secondary paralysis during closed treatment. Eight patients were treat- ed closed, and eight underwent sur- gical exploration. In all cases, the nerve was found to be intact, and each of the16 patients had complete neurologic recovery. Because of the small number of patients in this se- ries, definitive conclusions cannot be drawn; nevertheless, results sug- gest that even with this clinical sce- nario, aggressive early exploration may not yield improved results. Most large series of peripheral nerve repairs have been per formed during times of war. A recent pro- spective evaluation of factors influ- encing outcome was done in 490 pa- tients with complete peripheral nerve disruptions caused by projec- tile injuries during the Yugoslav civ- il war. 16 All repairs were performed in a single treatment center, and fi- nal outcome assessment, including motor function, sensory function, electrodiagnostic testing, and pa- tients’ subjective evaluation of the quality of recovery, was measured 24 to 30 months postoperatively. Out- come was correlated with regenera- tive potential of the damaged nerve, level of the nerve injury, surgical technique (ie, direct suture, nerve graft, or denatured muscle graft), lo- cal nutritive state, length of nerve defect, duration of interval from in- jury to repair, and patient age. Inter- estingly, certain nerves appeared to have greater “regenerative poten- tial,” 16 which seemed to be the fac- tor that most strongly affected out- come. Radial, musculocutaneous, and femoral nerves had the best po- tential for recovery of function; me- dian, ulnar, and tibial nerves had only moderate potential; and the peroneal nerve had poor potential for recovery. Although outcome was better in the more distal nerve inju- ries, the level of injury significantly (P < 0.001) affected outcome only for nerves with moderate regenerative potential (ie, median, ulnar, and tib- ial nerves). Similarly, the length of the nerve defect did not significant- ly influence outcome for nerves with the best potential for recovery (ie, musculocutaneous and femoral). For other, less resilient nerves, however, a linear relationship did exist be- tween the length of the defect and repair results. Final outcome was not affected by the state of local Closed Fractures Complicated by Peripheral Nerve Injury 34 Journal of the American Academy of Orthopaedic Surgeons blood supply (vascularized soft- tissue bed) and scar tissue, by the ap- plied surgical technique, or by pa- tient age. This latter finding is probably reflective of the limited number of children in this study population. Of the variables discussed, only time from injury to repair was mark- edly affected by the surgeon, indicat- ing that a balance must be struck when treating closed nerve injuries. During active observation to allow spontaneous recovery of less severe injuries, the surgeon must avoid in- troducing a delay that impairs the outcome of patients who would ulti- mately benefit from surgery. Roga- novic 16 noted a linear correlation between repair outcome and pre- operative interval. According to this study, the best probability for suc- cessful median, ulnar, or tibial nerve repair exists when the preoperative interval occurs in fewer than 3 months. In cases of radial nerve inju- ry, patients had an excellent chance for success when surgery was per- formed within 15.6 months of inju- ry. For most nerves, the interval after which repair appeared to be useless was between 9 and 12 months. How- ever, for the radial nerve, such an in- terval ceiling did not exist. Approach to Treatment With an approximate 85% spontane- ous recovery rate, 7,8 peripheral nerve injuries associated with closed frac- tures probably are best followed ex- pectantly. These injuries should not be considered either an indication or a contraindication for open reduc- tion and internal fixation (Figure 1). Skeletal injury is managed at the dis- cretion of the surgeon. In cases re- quiring internal fixation, whether by plates and screws or by intramedul- lary devices, the extent of explora- tion should be limited to that neces- sary to ensure that the nerve is free of the fracture site. For example, in closed humeral fractures treated with intramedullary fixation, the fracture is exposed through a small incision, and soft tissues are mobi- lized away from the bone ends. Guidewires are passed from one frac- ture fragment to the other under di- rect visualization, and the fracture is reduced without specifically looking for the nerve. No published study to date reports a nerve lesion occurring “away” from the closed fracture site. Thus, the extent of dissection is dic- tated by fracture type, with the only indication for extensive nerve dis- section being rare cases in which a transected nerve is identified and mobilization facilitates the nerve re- pair. During the wait for spontaneous nerve recovery, joint splinting and range-of-motion exercises should be initiated as soon as fracture care al- lows, thereby minimizing stiffness and joint or muscle contracture. Most patients who recover sponta- Figure 1 Treatment algorithm for closed fracture with associated nerve injury. L. Randall Mohler, MD, and Douglas P. Hanel, MD Volume 14, Number 1, January 2006 35 neously begin to do so in the first few months. For those who do not, electrodiagnostic studies—obtained at 6 and 12 weeks—are a helpful ad- junct. 5,17,18 The studies done at 6 weeks serve as a baseline for re- examination and document the se- verity of neurologic injury. A typical finding is the identification of fibril- lation potentials, positive sharp waves, and monophasic action po- tentials of short duration. Repeat electrodiagnostic testing of patients that does not demon- strate clinical recovery at 12 weeks will likely identify two subsets of pa- tients. One group will demonstrate improved nerve function with larger polyphasic motor unit action poten- tials of longer duration compared with those in the 6-week study. This is evidence that some degree of spon- taneous recovery may be expected. The degree to which the recovery will occur is not predictable. In cases in which the recovery is limited, ten- don transfers are required to improve function. The timing of these trans- fers is not well defined. Although in- jured nerves recover over 24 to 36 months, the degree of functional re- covery is well established within 18 months of injury. The second group will show few signs of recovery, with fibrillation potentials, positive sharp waves, and diminished or absent small motor unit potentials remaining the domi- nant features of the study. This sec- ond group is further divided into three subgroups: children showing little evidence of recovery, adults with radial nerve deficits, and adults with injuries other than of the radi- al nerve. Children, defined by skele- tal immaturity, should be observed for a total of 9 months before explo- ration or reconstruction because of the reported 95% rate of recovery within this timeframe. 5,10-12 Those children who do not recover are treated with tendon transfers. Adults with radial nerve palsies may be expected to recover within 6 months in >90% of cases. For adults who do not recover, the authors of two studies 16,19 propose that explora- tion and repair remains a viable op- tion for up to 6 months after injury. This point remains controversial, however, and the general opinion, although undocumented, is that in- stead of exploring the ner ve, sur- geons should consider tendon trans- fers for radial nerve palsies rather than nerve repair. The rationale is that tendon transfers in this setting provide equal or better functional re- covery than do radial nerve repairs performed 6 months postinjury. Patients in the third subgroup include adults with nerve injuries other than of the radial nerve. As mentioned, Roganovic 16 demon- strated that certain nerves have bet- ter recovery potential that others. For instance, the femoral and mus- culocutaneous nerves demonstrate excellent recovery potential when repaired within 24 weeks of injury, whereas the median, ulnar, and tib- ial nerves demonstrate moderate re- covery when repaired within the same timeframe; all five nerves demonstrate very little recovery when repaired later than 24 weeks. The peroneal nerve has very little re- covery potential no matter when it is explored or repaired. Early exploration presents a co- nundrum: the likelihood of finding a repairable lesion in the setting of closed fractures is unlikely, especial- ly in the lower extremities, while the likelihood of notable recovery, should a repairable lesion be found, is markedly diminished when repair is delayed >4 months. 16 In other words, the delayed exploration may be unnecessary. With this in mind, the surgeon is likely to discover a nerve in anatomic, but not physio- logic, continuity; he or she is then faced with the difficult decision of choosing between neurolysis, exci- sion and repair, or doing no more than ensuring that there are no com- pressing fascial bands or bony frag- ments, or a fracture callus, along the course of the nerve. Physical conti- nuity does not ensure spontaneous recovery, nor does complete loss of distal function preclude spontane- ous recovery. Intraoperative nerve stimulation may assist with these decisions. The intraoperative technique of Kline and Happel, 20 of stimulating the in- volved nerve proximally and record- ing the nerve action potential distal to the injured segment, is useful in this difficult situation. They found that, for patients in whom only neu- rolysis was performed, good func- tional recovery occurred in 93% of nerves that had a recordable nerve action potential distal to the lesion. When resection of the injured seg- ment was based on the absence of a recordable nerve action potential distal to the lesion, histologic exam- ination uniformly confir med a le- sion with poor potential for useful recovery without repair. These studies are not simply the application of a nerve stimulator proximal to a zone of injury, delivery of an electrical stimulus, and obser- vation of a twitching muscle distally. Little information is gained from this technique, especially when there is no distal response. Intraoperative nerve evaluation is the montage of in- formation gained from somato- sensory evoked potentials, elec- tromyography, and nerve conduction velocity studies. Obtaining this infor- mation is technically demanding and requires intraoperative collaboration with a trusted neurophysiologist- electrodiagnostician. 20,21 The mechan- ics of this diagnostic technique con- sist of placing recording electrodes proximally (at the mastoid, seventh cervical region, and contralateral scalp), exposing the injured nerve, and stimulating throughout the zone of injury. Hook electrodes are placed proximal to the zone of injury and the effects of stimulation recorded prox- imally. The electrodes are advanced at 1-cm increments, and the stimu- lation response is measured. The point at which there is loss of soma- tosensory evoked potential response Closed Fractures Complicated by Peripheral Nerve Injury 36 Journal of the American Academy of Orthopaedic Surgeons designates the transition between functioning and nonfunctioning nerves. In further studies, stimulating the nerve at a constant point proximal- ly and measuring the response at 1-cm increments allows recording of nerve compound action potentials. The presence of nerve compound ac- tion potentials resulting from in- trafield stimulation demonstrates regenerating nerve fibers, which are an indication that further recovery will occur and that intraneural dis- section should be limited. Similarly, by placing an electrode in a target muscle and stimulating the injured nerve motor, compound action po- tentials reflect the response of a large group of motor units, also a positive finding. Nerve conduction velocities are measured across the zone of inju- ry and specific areas of slowing are noted. The intraoperative interpretation of this information requires peri- operative coordination, a dialogue between the neurophysiologist and the surgeon, and a clearly defined set of goals for dealing with the infor- mation provided. Most importantly, these studies require time and a pa- tient surgeon who is willing to listen to the recommendations of, and re- peat the studies requested by, the electrodiagnostician. This coordinat- ed effort, however, can at times pro- vide equivocal data. In such cases, the surgeon must make the difficult decision of resection and grafting or of leaving the neuroma in situ. In most equivocal cases, we leave the nerve intact, choosing further obser- vation and appropriate tendon trans- fers when the nerve fails to recover. Summary Closed fractures are occasionally complicated by peripheral nerve in- jury. The number of cases is limited and incidence is sporadic, making longitudinal research difficult. Most patients recover without surgery. Those who fail to show signs of re- covery at 6 months, either clinically or with electrodiagnotic testing, should undergo exploration. Base- line electrodiagnostic studies are made 6 weeks postinjury; when there is no sign of nerve recovery, studies are repeated at 12 weeks. Adults who show evidence of recov- ery should continue to be observed. Adults without evidence of recovery and with radial nerve injury should undergo repeat nerve studies and ul- timately, if necessary, exploration, repair, and/or tendon transfers. The same procedure should be followed in adults with injuries other than those of the radial nerve. Exploration in skeletally immature children, whether exhibiting evidence of re- covery or not, should be delayed for 9 months. References 1. Noble J, Munro CA, Prasad VS, Midha R: Analysis of upper and lower ex- tremity peripheral nerve injuries in a population of patients with multiple injuries. J Trauma 1998;45:116-122. 2. de Laat EA, Visser CP, Coene LN, Pahlplatz PV, Tavy DL: Nerve lesions in primary shoulder dislocations and humeral neck fractures: A prospective clinical and EMG study. J Bone Joint Surg Br 1994;76:381-383. 3. Postacchini F, Morace GB: Fractures of the humer us associated with paral- ysis of the radial nerve. Ital J Orthop Traumatol 1988;14:455-464. 4. Pollock FH, Drake D, Bovill EG, Day L, Trafton PG: Treatment of radial neuropathy associated with fractures of the humerus. J Bone Joint Surg Am 1981;63:239-243. 5. Shah JJ, Bhatti NA: Radial nerve paral- ysis associated with fractures of the humerus: A review of 62 cases. Clin Orthop Relat Res 1983;172:171-176. 6. Sonneveld GJ, Patka P, van Mourik JC, Broere G: Treatmentof fractures of the shaft of the humerus accompanied by paralysis of the radial nerve. Injury 1987;18:404-406. 7. Böstman O, Bakalim G, Vainionpaa S, Wilppula E, Patiala H, Rokkanen P: Radial palsy in shaft fracture of the humerus. Acta Orthop Scand 1986; 57:316-319. 8. Böstman O, Bakalim G, Vainionpaa S, Wilppula E, Patiala H, Rokkanen P: Immediate radial nerve palsy compli- cating fracture of the shaft of the hu- merus: When is early exploration jus- tified? Injury 1985;16:499-502. 9. Whitson RO: Relation of the radial nerve to the shaft of the humerus. J Bone Joint Surg Am 1954;36:85-88. 10. McGraw JJ, Akbarnia BA, Hanel DP, Keppler L, Burdge RE: Neurological complications resulting from supra- condylar fractures of the humerus in children. J Pediatr Orthop 1986;6: 647-650. 11. Cramer KE, Green NE, Devito DP: In- cidence of anterior interosseous nerve palsy in supracondylar humerus frac- tures in children. J Pediatr Orthop 1993;13:502-505. 12. Dormans JP, Squillante R, Sharf H: Acute neurovascular complications with supracondylar humerus frac- tures in children. J Hand Surg [Am] 1995;20:1-4. 13. Brown IC, Zinar DM: Traumatic and iatrogenic neurological complica- tions after supracondylar humerus fractures in children. J Pediatr Orthop 1995;15:440-443. 14. Mehlman CT, Strub WM, Roy DR, Wall EJ, Crawford AH: The effect of surgical timing on the perioperative complications of treatment of supra- condylar humeral fractures in chil- dren. J Bone Joint Surg Am 2001;83: 323-327. 15. Ring D, Chin K, Jupiter JB: Radial nerve palsy associated with high- energy humeral shaft fractures. J Hand Surg [Am] 2004;29:144-147. 16. Roganovic Z: Factors influencing the outcome of nerve repair. Vojnosanit Pregl 1998;55:119-131. 17. Robinson LR: Traumatic injury to pe- ripheral nerves. Muscle Nerve 2000; 23:863-873. 18. Robinson LR: Role of neurophysiolog- ic evaluation in diagnosis. JAm Acad Orthop Surg 2000;8:190-199. 19. Amillo S, Barrios RH, Martinez-Peric R, Losada JI: Surgical treatment of the radial nerve lesions associated with fractures of the humerus. J Orthop Trauma 1993;7:211-215. 20. Kline DG, Happel LT: Penfield Lec- ture: A quarter century’s experience with intraoperative nerve action po- tential recording. Can J Neurol Sci 1993;20:3-10. 21. Slimp JC: Intraoperative monitoring of nerve repairs. Hand Clin 2000;16: 25-36. L. Randall Mohler, MD, and Douglas P. Hanel, MD Volume 14, Number 1, January 2006 37 . electrodiagnostic testing, which showed evidence of distal denervation. Exploration 14 weeks after injury revealed that the nerve was trapped in the callus of the healed fracture. Neurolysis provided a complete recovery. the domi- nant features of the study. This sec- ond group is further divided into three subgroups: children showing little evidence of recovery, adults with radial nerve deficits, and adults with. coordinat- ed effort, however, can at times pro- vide equivocal data. In such cases, the surgeon must make the difficult decision of resection and grafting or of leaving the neuroma in situ. In most equivocal