 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
The Journal of Bone and Joint Surgery (American). 2007;89:1108-1116. doi:10.2106/JBJS.F.00594 © 2007 The Journal of Bone and Joint Surgery, Inc.
The Ulnar Nerve in Elbow Trauma1 Massachusetts General Hospital, Yawkey Center, Suite 2100, 55 Fruit Street, Boston, MA 02114. E-mail address for D. Ring: dring{at}partners.org
Disclosure: In support of their research for or preparation of this
work, one or more of the authors received, in any one year, outside funding or
grants in excess of $10,000 from the AO Foundation, Small Bone Innovations,
Smith and Nephew, Wright Medical Technology, and Biomet. Neither they nor a
member of their immediate families received payments or other benefits or a
commitment or agreement to provide such benefits from a commercial entity. No
commercial entity paid or directed, or agreed to pay or direct, any benefits
to any research fund, foundation, division, center, clinical practice, or
other charitable or nonprofit organization with which the authors, or a member
of their immediate families, are affiliated or associated.
The ulnar nerve is susceptible to compressive neuropathy at several anatomical sites at the elbow joint1-4. Elbow trauma places the nerve at risk for direct injury, operative injury, and subsequent compressive neuropathy. Optimal management of the ulnar nerve during operative treatment of an elbow injury is commonly discussed and debated, particularly with respect to the operative treatment of fractures of the distal part of the humerus5-7. The role of ulnar nerve dysfunction in posttraumatic stiffness and pain is also being increasingly recognized8-10. This paper reviews the published scientific data and current opinion available to guide patient care, with particular emphasis on areas where more data are needed.
The prevalence of acute and chronic ulnar nerve dysfunction after elbow trauma has not been documented definitively. Acute injury-related ulnar nerve palsy is uncommon in association with fractures of the distal part of the humerus and the olecranon (including fracture-dislocations of the olecranon), but it has been described as occurring in up to 10% of elbow dislocations11,12. This number seems too high, and the 1% prevalence of acute ulnar nerve palsy associated with elbow dislocation reported by Galbraith and McCullough13 is more consistent with our experience. A possible explanation for the difference in these numbers is the failure to distinguish among acute ulnar nerve palsy, postoperative ulnar nerve palsy, and delayed-onset ulnar nerve dysfunction, with the 10% figure reflecting all types of ulnar nerve dysfunction encountered from the time of injury through the long-term follow-up period. Robinson et al. reviewed several series of adult patients with a fracture of the distal part of the humerus (320 patients in total) and found an overall rate of ulnar neuropathy of 12.3% (range, 0% to 50.9%), with an average of 5.4% (range, 0% to 15%) of these deficits being permanent14. The wide ranges and the failure to distinguish among the different types of ulnar nerve dysfunction limit the utility and interpretation of these findings. We cannot accurately describe the prevalence of acute injury-related ulnar nerve palsy, postoperative ulnar nerve palsy, and delayed-onset ulnar nerve palsy on the basis of the available published data. Future researchers should distinguish among these types of ulnar nerve dysfunction and define the prevalence of the problem according to these specific injury types.
The ulnar nerve is derived from the C8 and T1 nerve roots. It is a terminal branch of the medial cord of the brachial plexus. In the midportion of the arm, it passes from the anterior to the posterior part of the arm, piercing the intermuscular septum approximately 10 cm above the medial epicondyle15. The nerve then passes through a fascial tunnel bounded laterally by the internal brachial ligament and inferiorly by an accessory origin of the medial head of the triceps from the medial intermuscular septum. The latter structure is commonly referred to as the arcade of Struthers, although it appears that Struthers never described it16-18. The arcade of Struthers averages 3.75 cm in length, ending approximately 3 to 10 cm above the medial epicondyle19. The reported prevalence of an arcade of Struthers is extremely variable, ranging between 25% (sixteen of sixty-four)20 and 70% (fourteen of twenty)21, indicating that its nature as a structure is disputed and difficult to characterize. The ulnar nerve passes under the medial epicondyle and enters the cubital tunnel beneath the fibrous arch bridging the two heads of the flexor carpi ulnaris (or the Osborne ligament)1,2,4. In the forearm, the ulnar nerve passes beneath the aponeurosis between the ulnar and humeral heads of the flexor carpi ulnaris. The blood supply to the ulnar nerve at the elbow is derived from three main pedicles, the superior ulnar collateral artery proximally, the inferior ulnar collateral artery in the midportion, and the posterior recurrent ulnar artery distally. These three vessels form an anastomotic extraneural network intimately associated with the ulnar nerve22. The superior ulnar collateral artery arises from the brachial artery about 18 cm proximal to the medial epicondyle. The inferior ulnar collateral artery arises from the brachial artery about 6.6 cm proximal to the medial epicondyle. The posterior recurrent ulnar artery arises from the ulnar artery about 7.3 cm distal to the medial epicondyle. In the study by Yamaguchi et al., twenty of twenty-two elbows did not have a direct anastomotic connection between the superior ulnar collateral artery and the posterior recurrent ulnar artery and fourteen of the twenty-two elbows did not have a communication between the superior ulnar collateral artery and the inferior ulnar collateral artery22. This is an important finding because division of the inferior ulnar collateral artery during anterior transposition of the ulnar nerve may devascularize a portion of the nerve. A study of ulnar nerve transposition in monkeys demonstrated a significant decrease (p < 0.05) in regional blood flow that took up to three days to return to normal23. Some have claimed that careful dissection can often preserve this blood supply during transposition24. Others have emphasized that there is an internal blood supply that is preserved during transposition22,25. The ulnar nerve does not innervate any structures in the upper arm. There are zero to three articular branches to the elbow, and these may begin proximal to the medial epicondyle26. The first branch to the flexor carpi ulnaris is found an average of 15 mm distal to the epicondylar axis27 and usually arises from the radial aspect of the nerve26. Typically, there are two to five branches to the flexor carpi ulnaris and the flexor digitorum profundus combined, occasionally with only one branch to the flexor carpi ulnaris27. Care must be taken to protect the first motor branch, in case it is the only branch to the flexor carpi ulnaris. Intraneural dissection of the first motor branch to the flexor carpi ulnaris can be performed safely and is occasionally required to fully mobilize the ulnar nerve. The dimensions of the ulnar nerve change with elbow flexion and extension. The cross-sectional shape changes from round to elliptical with increasing flexion, as the nerve flattens against the medial epicondyle28. The cross-sectional area of the nerve decreases up to 50%29 and the nerve elongates about 5 mm with full flexion30,31. The mean intraneural pressure is approximately 45% greater than the mean extraneural pressure with full elbow flexion, suggesting that intraneural pressure is the result of both traction and flattening (external pressure) of the nerve29.
The ulnar nerve can be damaged by the initial injury or iatrogenically during operative treatment, or it may become symptomatic in a delayed fashion secondary to postoperative swelling, scarring, and thickening in its fibroosseous tunnel. It is important to evaluate peripheral nerve function carefully prior to any operation for the treatment of an elbow injury in order to distinguish injury-related from surgery-related palsies. Postoperative nerve palsies are often the result of traction on the nerve. Traction on the nerve should be carefully monitored and minimized. Retractors and loops around the nerve should be used with care, and assistants should be kept attentive. Although it can be argued that a wider loop such as a Penrose drain will apply less force per unit area than a smaller loop, our opinion is that it is not the pressure from the loop that places the nerve at risk, but rather the overall traction on the nerve. We believe that one is less likely to apply excessive traction through a smaller, more flexible loop such as a vessel loop. Postoperative palsy can occur even when the nerve is handled gently. Furthermore, we have observed several postoperative ulnar nerve palsies after the nerves were transposed in the injury setting but none after transposition for treatment of idiopathic ulnar neuropathy. Given that some anatomical studies have suggested that a segment of the ulnar nerve may be devascularized during transposition, we have speculated that a traumatized nerve may be more susceptible to injury, by means of devascularization, than a chronically compressed nerve. More research is needed to better understand the pathogenesis of, and risk factors for, postoperative ulnar nerve palsy in the setting of acute elbow trauma. Delayed compressive neuropathies may be related to (1) implants placed on the medial column of the distal part of the humerus; (2) osteophytes, and exuberant fracture callus32, or heterotopic ossification (in which the ulnar nerve is occasionally entrapped)33,34; (3) constriction of the nerve by an incompletely divided flexor pronator aponeurosis or arcade of Struthers, or over an incompletely released medial intermuscular septum or medial triceps fascia (arcade of Struthers) after transposition; and (4) thickening or scarring of the normal fibroosseous tunnel in which the ulnar nerve lies35-40. Delayed ulnar neuropathies have also been described in association with varus or valgus malunion of humeral fractures in children. In this setting, the ulnar nerve dysfunction has been referred to as a tardy ulnar nerve palsy41,42. Treatment recommendations may include humeral osteotomy in addition to ulnar nerve transposition, but the influence of the osteotomy on the ulnar neuropathy is debatable. Given the relationship between elbow trauma and subsequent ulnar neuropathy in the absence of varus or valgus deformity, the role of such deformity in the pathogenesis of ulnar nerve dysfunction can be questioned.
In patients with an ulnar neuropathy, numbness usually involves the small finger and the ulnar half of the ring finger. Sensory disturbance can be objectively and quantitatively evaluated with use of tests of threshold sensibility (Semmes-Weinstein monofilament testing)43 and innervation density (two-point discrimination testing). Semmes-Weinstein monofilament testing assesses the ability of the patient to sense pressure from filaments that bend at a known pressure. The normal value for this test is 2.83 MN (marking number), which is equivalent to ten times the log of the force in milligrams—in this case 68 mg. Patients who can detect only larger monofilaments have decreased sensation to light touch. Two-point discrimination can be assessed with a finger-blanching technique, in which two blunt points separated by a known distance are either held in one spot (static testing)44 or kept in motion (dynamic testing)45. The upper limit of normal has been described as 6 mm for the former46 and as 4 mm for the latter47. If sensation is normal on the dorsal ulnar aspect of the hand and wrist but not in the small finger, ulnar nerve compression at the wrist should be suspected. Atrophy and weakness reflect advanced nerve damage. Atrophy of the first dorsal interosseous muscle is usually more apparent than atrophy of the other intrinsic hand muscles innervated by the ulnar nerve, such as those in the hypothenar eminence and the other interosseous muscles. The strength of the first dorsal interosseous muscle can be tested by placing the patient's extended index finger in maximum radial deviation, having him or her resist forceful ulnar deviation, and comparing the resistance with that of the other hand. Weakness of the abductor digiti minimi and the flexor digitorum profundus to the small finger should also be assessed. The relative strength of the abductor digiti minimi can be tested conveniently with the confrontational test, in which the small fingers of both hands are held in abduction with maximal force, the small fingers are brought together, and, starting at approximately a 90° angle to each other, are pushed against one another. The test result is negative when both small fingers give way and collapse toward the ring fingers simultaneously. Unilateral giving-way is a positive result, suggesting weakness48. Static sensory or motor deficits may be absent in the early stages of compressive ulnar neuropathy. Symptoms can be provoked by sustained elbow flexion (the so-called elbow flexion test49), sustained pressure over the nerve in the cubital tunnel50, or tapping over the ulnar nerve at the cubital tunnel (to elicit the Tinel sign51). Novak et al. compared patients with electrodiagnostically confirmed idiopathic ulnar nerve compression with asymptomatic controls and found that a greater duration of elbow flexion or pressure over the nerve as well as the combination of both compression and flexion increased the sensitivity of provocative testing with little decrease in specificity (Table I)50. These findings are likely to apply to delayed posttraumatic ulnar neuropathy as well.
Neurophysiological testing provides objective confirmation of suspected ulnar neuropathy. Electromyography is used to evaluate motor nerve dysfunction. This test measures the electrical activity of a motor unit. Evidence of fibrillations suggests denervation of the motor end plates. Positive sharp waves appear approximately ten to twenty days after nerve injury and are also characterized by rest activity and increased insertional activity. Polyphasic motor unit potentials represent reinnervation and motor nerve recovery. Transient suppression of electromyographic activity may be demonstrated by noxious digital nerve stimuli. This period of electrical silence is known as a cutaneous silent period. Its presence is normal and indicates intact small fibers (A-delta), which are typically spared in compression neuropathy. Cutaneous silent periods become elongated with more severe nerve involvement. Most importantly, cutaneous silent periods confirm residual nerve continuity, which may not be apparent with use of other electromyographic parameters52,53. Nerve conduction studies are used to evaluate sensory and motor nerve conduction amplitudes and velocities. Changes in sensory conduction are more sensitive indicators of neuropathy and correlate more directly with findings on physical examination. With both sensory and motor nerve conduction studies, the earliest sign of nerve dysfunction is a decrease in amplitude. This is closely followed by a decrease in motor velocity. Either absolute velocity across the elbow of <50 ms or slowing of velocity by >10 m/s compared with the value in the contralateral, unaffected arm suggests ulnar nerve compression. Additional features include an above-the-elbow to below-the-elbow segment being >10 ms slower than the below-the-elbow to wrist segment, a decrease in compound muscle action potential negative peak amplitude from below the elbow to above the elbow of >20%, and a substantial change in compound muscle action potential configuration at the above-the-elbow site compared with the below-the-elbow site54. Distal sensory and motor latencies of >2.4 and >2.6 ms, respectively, are also considered abnormal.
McGowan55 classified ulnar nerve dysfunction into three grades. Grade 1 represents a minimal lesion with no detectable motor weakness of the hand. Grade 2 is characterized by an intermediate lesion in which a sensory disturbance is accompanied by interosseous muscle weakness and some wasting. Grade 3 is a severe lesion with profound weakness of the interossei. In 1990, Gabel and Amadio39 reviewed the results of reoperations done because of failed cubital tunnel releases in thirty patients and devised a 9-point scoring system to rate ulnar neuropathy preoperatively and postoperatively. The score is based on the summation of scores on 0 to 3-point scales for pain, sensory deficits, and motor deficits (Table II), with 9 points representing no evidence of neuropathy and 0 points representing the most severe neuropathy.
In 1989, Dellon proposed a similar grading system56. The severity of the lesion is graded as mild, moderate or severe, depending on sensory and motor findings as well as the results of provocative maneuvers (Table III). Sensory findings include the presence of paresthesias, response to vibratory stimuli, and two-point discrimination. Motor findings include atrophy and the presence of objective weakness. Provocative tests include the elbow flexion test and the Tinel sign.
It is also useful to classify ulnar nerve dysfunction according to its time of onset in relation to the traumatic event. It is important to distinguish among acute injury-related ulnar neuropathies, postoperative surgery-related ulnar neuropathies, and delayed compressive ulnar neuropathies. Delayed ulnar neuropathy can be either subacute57 (onset within three months after the injury) or chronic (onset more than three months after the injury). Subacute ulnar neuropathy usually presents as loss of elbow motion and increasing pain with or without hand stiffness after an initial period during which pain was decreasing and motion was recovering well. Chronic ulnar neuropathy—whether posttraumatic or idiopathic—usually presents with advanced findings, with the patient having adapted to the symptoms. The surgeon should be mindful of the potential for ulnar nerve dysfunction at all stages of recovery from elbow trauma and remember to evaluate ulnar nerve function at each office visit.
While injury-related nerve palsies usually resolve, recovery from postoperative ulnar nerve palsies can be prolonged and incomplete. In our experience, patients with diabetes mellitus have been at particular risk for delayed or incomplete recovery58. In our opinion, subacute and chronic ulnar neuropathies and idiopathic cubital tunnel syndrome are slowly and inevitably progressive, as has been increasingly observed to be the case with idiopathic carpal tunnel syndrome. Furthermore, we believe that nonoperative treatment can help manage symptoms in the short term but cannot be relied on to prevent progression of the disease59. There is better evidence to support these concepts with regard to idiopathic carpal tunnel syndrome, but it is reasonable to expect other compressive peripheral neuropathies to behave in a similar fashion. In a clinical trial comparing operative treatment and splint immobilization in 176 patients with idiopathic carpal tunnel syndrome, 41% of those treated with the splint required surgery within eighteen months after enrollment60. In addition, several studies of idiopathic carpal tunnel syndrome have demonstrated either bilateral presentation or eventual development of nerve dysfunction on the contralateral side, suggesting that it is a bilateral disease61,62. Finally, a study of twins showed that genetics alone accounts for at least half of the risk of carpal tunnel syndrome developing63. It is plausible that chronic idiopathic peripheral neuropathies reflect progressive structural compression that is genetically mediated and inevitable. It is also plausible that posttraumatic ulnar neuropathy is structural and progressive. These speculations must be confirmed by scientific data and may be relatively academic since most cases of posttraumatic ulnar neuropathy present at a fairly advanced stage with atrophy and weakness. Nonetheless, we recommend regular follow-up of patients with posttraumatic ulnar neuropathy if nonoperative treatment is chosen.
Bicolumnar Distal Humeral Fractures Among elbow fractures, bicolumnar fractures of the distal part of the humerus are associated with the highest rate of both acute ulnar nerve palsy and delayed-onset chronic ulnar neuropathy following operative treatment5-7,64-79 (see Appendix). Surgeons handle the nerve in a variety of ways. All recommend initial exposure and protection of the nerve. Some advocate placing the nerve back into its epicondylar groove after internal fixation is completed7,76. Others transpose the nerve only when it is contused or if it lies directly on the medial plate6,77,78. Most support routine anterior transposition of the nerve5,79-84. The exact prevalence of postoperative ulnar nerve palsy cannot be stated accurately for the reasons mentioned earlier, but it may be as high as 13%5-7,67,70-79. We are beginning to analyze our experience, in which we have encountered what may be seven new iatrogenic postoperative ulnar nerve palsies in thirty-two patients (a rate of 22%) with a bicolumnar distal humeral fracture in whom a complete subcutaneous ulnar nerve transposition had been performed. If these ulnar nerve palsies were a result of the complete anterior transposition (i.e., caused by handling, retraction, or devascularization of the nerve), this would be a substantial disadvantage of that procedure, but our data are provisional and it is unclear if the rate of palsy is lower after protection of the nerve without transposition. On the other hand, it is not clear that complex fractures of the distal part of the humerus can be treated without moving the ulnar nerve out of the cubital tunnel. A few authors reporting on retrospective case series of patients who had undergone internal fixation of a bicolumnar fracture of the distal part of the humerus have commented on the handling of the ulnar nerve5-7,72,73,77,79,80. The occurrence of postoperative nerve palsy is inconsistent, and no definitive treatment recommendations can be made. Kundel et al.7 reviewed the cases of seventy-seven patients in whom transposition was performed only when implants were placed in the groove below the medial epicondyle (in thirteen cases). They reported a 27% prevalence of injury or surgery-related ulnar neuropathy with this technique. In a study of fifty-five patients in whom transposition was performed only when there were preexisting ulnar nerve symptoms, Gupta and Khanchandani6 reported that 5% had an acute postoperative ulnar nerve palsy. Wang et al.5 advocated routine subcutaneous transposition, reporting no ulnar nerve complications in twenty patients who had been so treated. The ulnar nerve is routinely transposed when bicolumnar fractures of the distal part of the humerus are treated with linked total elbow arthroplasty. Acute postoperative ulnar nerve dysfunction has been reported in 0% to 26% of patients64-66,68,69.
Capitellar and/or Trochlear Fractures and Elbow Fracture-Dislocations Fracture of the olecranon has been associated with a prevalence of postoperative ulnar neuropathy of 6% (two of thirty-one)85 and prevalences of delayed-onset ulnar neuropathy of 2% (one of fifty-two86) to 22% (four of eighteen32). We are in the process of reviewing the results of our treatment of dislocations of the elbow with associated intra-articular fractures and have noted a rate of delayed ulnar neuropathy of approximately 11% within two to five years postoperatively; the rate has been 15% after treatment of so-called terrible triad injuries (dislocation with fractures of the radial head and coronoid process). Because patients with these fracture types are at risk for delayed ulnar neuropathy, we have started to consider identification, prophylactic in situ decompression, and protection of the ulnar nerve during operative treatment of these injuries. More data are needed regarding the relative advantages and disadvantages of routine in situ release of the ulnar nerve, but we have encountered few sequelae in the short term. This situation may be analogous to that of unlinked total elbow arthroplasty, which was associated with a substantial rate of ulnar neuropathy87-95 before routine ulnar nerve decompression was added to the operative protocol.
Postinjury or Postoperative Nerve Palsy Nerve palsies identified within the first two weeks after an injury or operation are most likely related to traumatic or operative injury to the nerve. Sometimes it appears that a nerve that was intact at the initial postoperative examination becomes dysfunctional within the next few days, but in our opinion this is unlikely. It seems more likely that the initial postoperative examination was incomplete or inaccurate than that the status of the ulnar nerve changed over a few days. Consequently, if the surgeon is confident that the nerve is not lacerated or entrapped by an implant or injury, then a postoperative palsy should be managed with observation; otherwise the nerve should be explored.
Subacute Ulnar Neuropathy
Chronic Ulnar Neuropathy McKee et al. reviewed the cases of twenty patients (twenty-one elbows) in whom chronic ulnar neuropathy had been treated with ulnar nerve decompression as part of an elbow reconstruction98,99. At an average of thirty-two months after the operation, the patients frequently had a return of intrinsic muscle function and had a high rate of satisfaction. There were seventeen good-to-excellent results, two fair results, and two poor results. Ten patients had a McGowan grade-3 lesion preoperatively, and none had one postoperatively. The average Gabel and Amadio score increased from 3.2 to 6.5 points. There was no worsening of intrinsic strength. Of seventeen patients with preoperative weakness, twelve recovered grade-5 power, four recovered grade-4 power, and one still had grade-3 power. The average satisfaction score was 8.0 of 10 points. Barrios et al. directly compared the results of neurolysis and transposition of the ulnar nerve in patients with either idiopathic or posttraumatic cubital tunnel syndrome100. The results were slightly better (as measured with a scoring system and categorical ratings based on the Nerve Committee of the British Medical Research Council101) in the patients with idiopathic ulnar neuropathy. A greater number of patients with idiopathic ulnar neuropathy had complete recovery. The duration of symptoms did not affect the outcome100.
Ulnar Neuropathy Associated with Elbow Contracture
Ulnar Neuropathy Associated with an Ununited Fracture of the Distal Part of the Humerus
Recalcitrant Ulnar Neuropathy Salvage techniques such as use of silicone sheaths or vein-wrapping39,106-108 to address scarring around the nerve have been described primarily for the treatment of pain and paresthesias, but two-point discrimination and electromyographic findings may also improve. These techniques are challenging and remain controversial. In theory, they minimize adhesions between the surrounding tissues and the nerve107. Promising early results have been reported39,106-108 by the advocates of these techniques, but additional research is needed to fully assess their value. Implantable peripheral nerve stimulators have also been used for chronic pain believed to be due to recalcitrant ulnar neuropathy, but these devices should be considered experimental109-111.
Although definitive data are not available, current expert opinion supports identification and protection of the ulnar nerve during the treatment of a bicolumnar fracture of the distal part of the humerus and other types of complex elbow injury. The decision to transpose the ulnar nerve may be based in part on the injury type and be done when implants are applied to the medial epicondyle and medial side of the trochlea. Authors of future studies should systematically document nerve status and differentiate among injury-related, postoperative, subacute, and chronic ulnar nerve dysfunction. Ulnar nerve decompression is an important aspect of posttraumatic elbow reconstruction and can improve both elbow and hand function.
A table presenting information from previous studies on posttraumatic ulnar nerve palsy is available with the electronic versions of this article, on our web site at jbjs.org (go to the article citation and click on "Supplementary Material") and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM).
|
Top
Introduction
Epidemiology
The prevalence of
ulnar nerve dysfunction after elbow injury is unknown because authors of
published investigations have inadequately differentiated among acute
injury-related, acute surgery-related, and delayed (subacute or chronic) ulnar
neuropathies and these retrospective case series have not included careful
evaluation of ulnar nerve function.