Clinical manifestations

Presentation and course

Ulnar neuropathy is the second most common entrapment neuropathy in the United States. In contrast to carpal tunnel syndrome, in which sensory impairment is generally the most significant disability, motor loss is usually the most important problem in ulnar nerve entrapment, most notably in the intrinsic muscles of the hand.

In the majority of patients with ulnar neuropathy at the elbow, the initial symptoms are intermittent ulnar distribution numbness and tingling, often associated with elbow flexion. Similar to carpal tunnel syndrome, these symptoms may be most noticeable at night. With mild involvement, such intermittent symptoms may persist for months or years. Patients may not seek medical care until significant intrinsic hand muscle atrophy develops. Some patients present primarily with weakness and atrophy, but with little in the way of sensory symptomatology or pain. In patients with more severe compression, persistent sensory symptoms and motor weakness may develop over a short period of time. Inquiry should be made regarding any remote elbow trauma, especially fracture or dislocation, repetitive elbow motion, or a tendency to elbowing. Inquiry into sleep positions, particularly those involving tight flexion at the elbows, may also shed light on the diagnosis.

The ulnar sensory distribution typically includes the palmar surface of the small finger and the ulnar half of the ring finger. The ulnar aspect of the palm is innervated by the palmar cutaneous branch that arises proximal to the wrist. The dorsal cutaneous branch supplies the dorsal ulnar aspect of the hand and fingers. Sensory loss is usually easiest to detect over the tip of the little finger, and diminution of sensibility is usually more pronounced for light touch and 2-point discrimination as opposed to pinprick and temperature. Though there may be some variability in the ulnar sensory territory, the following details are often useful. On the palmar surface, the median and ulnar distributions usually split the ring finger, and such splitting fairly reliably excludes plexopathy and radiculopathy. The ulnar distribution does not extend more than a few centimeters proximal to the wrist crease. The medial antebrachial cutaneous nerve passes anterior to the medial epicondyle to supply the skin of the medial forearm; abnormality in this region excludes ulnar neuropathy at the elbow because this nerve does not traverse the ulnar groove. Impaired sensation over the dorsum of the hand establishes the location of the lesion as proximal to the takeoff of the dorsal cutaneous branch, but sparing of the dorsal cutaneous territory does not exclude ulnar neuropathy at the elbow because of possible selective sparing of its fascicles. Involvement of the palmar cutaneous branch distribution likewise suggests a lesion proximal to the distal forearm.

The ulnar nerve provides fine motor control, coordination, and dexterity to the hand, in contrast to the median nerve that provides more in the way of raw power. The median nerve has been called the "workman's nerve" and the ulnar nerve the "musician's nerve." A frequent early manifestation of ulnar neuropathy is a tendency of the small finger to get "hung up," as when placing the hand in a pocket, and an abducted posture of the small finger on examination (Wartenberg sign). It is particularly useful to test small hand muscles against the strength of the examiner's like muscles, after the methods described by Wolf (73). Observing the patient's ability to touch the index to the small finger is a good test of interosseous function. Another useful maneuver is to have the patient press their first dorsal interosseous or abductor digiti minimi together to see if one side overpowers the other.

The motor disability from ulnar neuropathy is related primarily to loss of coordination in tasks requiring precision and to loss of thumb pincer strength due to weakness of the adductor pollicis, the ulnar portion of the flexor pollicis brevis, and the first dorsal interosseous. Froment sign results from substituting the flexor pollicis longus to adduct the thumb and provide a pincer function. The ulnar griffe, or claw deformity, is due to weakness of the third lumbrical and fourth lumbrical. The lumbricals flex the metacarpophalangeal joints and extend the proximal interphalangeal joints. In ulnar lesions, unopposed extensor tone at the metacarpophalangeal joints and unopposed flexor tone at the proximal interphalangeal joints draw the ring and small fingers into a position with extension at the metacarpophalangeal joints and flexion at the proximal interphalangeal joints. Clawing varies depending on the amount of muscle weakness, the laxity of the metacarpophalangeal joints, and the level of the lesion. A "low" (distal) ulnar lesion with preserved function of the flexor digitorum profundus induces more clawing than a "high" (proximal) ulnar lesion, where the accompanying long finger flexor weakness creates less of the unopposed flexor pull deforming the ring and small fingers. Conditions other than ulnar neuropathy can produce a hand deformity mimicking ulnar clawing (17).

Evaluation of a patient with suspected ulnar neuropathy at the elbow should include examination of the elbow both for range of motion and for deformity. Impaired range of motion, flexion contracture, valgus deformity, or other bony or joint abnormality suggests an elbow level lesion. Reproduction of symptoms with elbow flexion and ulnar groove pressure can be informative (51). Worsening of the hypesthesia or pain with sustained elbow flexion, especially if combined with digital pressure over the nerve, suggests ulnar nerve compression near the elbow (the "elbow flexion test") (13). Forced flexion for more than a minute risks false positive results (60). Examining for subluxation is seldom helpful, as this is a common phenomenon in normal individuals. Eliciting a Tinel sign can be useful, but many normal patients "Tinel" over all their nerves; only the presence of a disproportionately active Tinel sign over the clinically suspect ulnar nerve has any localizing value. Palpation of the nerve itself may disclose an area of thickening or point tenderness. Each of these tests should be compared with the opposite extremity.

A nerve conduction study is also a useful diagnostic test, with conduction delays of at least 10 m/s in the affected arm relative to the unaffected arm. When significant atrophy is noted, differences of at least 15 m/s are typically found (07). To further predict outcomes, high-resolution sonography can also be performed. More pronounced ulnar nerve thickening at the time of the diagnosis is associated with poor outcome at follow-up, especially in conservatively treated cases, whereas electrodiagnostic signs of demyelination on testing indicate favorable outcome (09).

Prognosis and complications

The prognosis varies with the location and severity of neuropathy. At over 50 years of age, severe ulnar neuropathy at the elbow and comorbidities such as diabetes or alcoholism are poor prognostic indicators for response to anterior transposition (53). Electrodiagnostic studies can predict which patients may have a progressive or deteriorating course. The likelihood of spontaneous recovery is greatest in those with the mildest electrophysiological abnormalities (27). Dunselman and Visser reported that patients with 50% or more conduction block across the elbow have a better prognosis than those with less than 20% conduction block (30). In a systematic review of the impact of preoperative electrophysiologic findings on outcome of ulnar neuropathy at the elbow, the quality was good in only two out of the eight retrospective cohort studies that met inclusion criteria, and the level of certainty was low overall (47). Within those limitations, reduced or absent preoperative ulnar SNAP may predict worse postoperative outcomes. Based on the higher quality studies, conduction block at the elbow may also predict worse postsurgical outcomes.

Biological basis

Etiology and pathogenesis

Ulnar neuropathy at the elbow is the second most common upper extremity compression neuropathy after carpal tunnel syndrome, and there are at least four potential sites of compression in the region of the elbow:

Lesions in the ulnar groove account for most cases, but humeroulnar aponeurotic arcade compression is also common (24; 29). The exit compression syndrome is infrequent but does turn up regularly if one is sensitive to its existence (04; 23). Compression at the medial intermuscular septum is rare.

Lesions occur in the ulnar groove for a number of reasons including acute or chronic external pressure or trauma, bony or scar impingement, anomalous muscles or bands, chronic stretch in the presence of a valgus deformity, and rarely, mass lesions. One of Panas's original cases was due to a sesamoid bone in the groove (55). Compression syndromes at the other locations are virtually always due to entrapment by normal pathoanatomic structures. The proportion of cases of ulnar neuropathy at the elbow related to major elbow trauma has steadily decreased over the years, whereas the proportion of "idiopathic" cases has steadily increased (29). Ulnar neuropathy at the elbow may result from immobility and sustained pressure due to anesthesia, intoxication, coma, or prolonged bed rest (69). Ulnar neuropathy at the elbow may complicate depression, as the patient sits propped on the elbow with chin in hand in a "woe is me" posture for prolonged periods.

The humeroulnar aponeurotic arcade, the aponeurotic band responsible for the "cubital tunnel syndrome," forms an archway between the two heads of the flexor carpi ulnaris beneath which the ulnar nerve passes. Compression by the humeroulnar aponeurotic arcade should be particularly suspected when there is no elbow deformity and no history of trauma. In the two studies that have convincingly addressed the issue, there is remarkable concordance in the incidence of retroepicondylar abnormalities (69% vs. 62%), humeroulnar aponeurotic arcade abnormalities, ie, cubital tunnel syndrome (23% vs. 28%), and changes in both locations (8% vs. 10%) (29; 22). Occasionally, the ulnar nerve suffers entrapment by the deep fascial lining of the flexor carpi ulnaris at a point 5 to 7 cm distal to the medial epicondyle, where the nerve exits distally after its passage through the substance of the flexor carpi ulnaris (04; 23). A 6% incidence of perioperative ulnar neuropathy in a series of 203 patients has been reported. There was a highly significant association between a tilted body position on the operating table and development of ulnar neuropathy on the contralateral side. This position rotates the arm internally and places the ulnar nerve at risk for direct compression (42).

Anatomical factors account for much of the susceptibility of the ulnar nerve to injury in this region (21). The lack of protective covering over the nerve in its course through the ulnar groove accounts for its susceptibility to external pressure. After traversing the ulnar groove, the nerve passes beneath the humeroulnar aponeurotic arcade that typically lies 1.0 to 2.5 cm distal to the medial epicondyle. This arcade has also been referred to as the "cubital tunnel" and as the "Osborne band." After passing under the arcade, the nerve runs through the belly of the flexor carpi ulnaris, then exits from the muscle through the deep flexor pronator aponeurosis (thick connective tissue that lines the deep surface of the flexor carpi ulnaris). Both the humeroulnar aponeurotic arcade and the deep flexor pronator aponeurosis may entrap the nerve (32; 48; 23).

Important anatomical changes occur with elbow movement. In extension, the medial epicondyle and olecranon process are juxtaposed, with the humeroulnar aponeurotic arcade slack and the nerve lying loosely in the groove. With elbow flexion, the olecranon moves forward and away from the medial epicondyle; the separation increases by 5 mm for every 45° of flexion (06). The two heads of the flexor carpi ulnaris are pulled apart, progressively tightening the humeroulnar aponeurotic arcade across the nerve and dramatically increasing the pressure against the nerve (33). In addition, with elbow flexion, the ulnar collateral ligament bulges into the floor of the groove and the medial head of the triceps may impinge on the groove from behind (63). In extension, the ulnar groove is smooth, round, and capacious, but in flexion the nerve is crowded into a flattened, tortuous, narrow canal with the humeroulnar aponeurotic arcade pulled tightly across it. Repetitive flexion and extension may predispose to ulnar neuropathy at the elbow because of the dynamic changes in the nerve's passageway with motion. With elbow joint derangement due to trauma or arthritic changes, the nerve's vulnerability increases even further. Valgus deformities increase the stretch on the nerve with elbow flexion, and osteophytic overgrowth further narrows an often already narrow passageway.

The internal fascicular anatomy of nerves is complex, and some fascicles are more prone to injury than others. The intricate fascicular organization and the various susceptibilities of different fascicles to injury likely explains some of the curiosities of ulnar neuropathy at the elbow, including preferential involvement of the first dorsal interosseous and the common sparing of the forearm flexors (64; 20).

Epidemiology

A change in the spectrum of etiologies of ulnar neuropathy at the elbow has occurred gradually over the years. Improved orthopedic management of elbow injuries has led to a decline in the proportion of ulnar neuropathy at the elbow due to elbow trauma and deformity, and an increase in the proportion due to other causes, particularly humeroulnar aponeurotic arcade compression and "idiopathic" retroepicondylar neuropathies (29). Less common compressive agents include inflammation, acquired demyelinating sensory and motor neuropathy (37), rheumatoid synovitis, lipomas and other tumors, osteophytes from the ulnohumeral articulation, and a rare anomaly referred to as the “persistent epitrochleoanconeus muscle” (34). Some cases may not be related to a compressive lesion at all, but rather to “friction” neuritis. This may be especially true of those individuals with chronic and recurrent dislocation of the nerve from the ulnar groove (72).

Prevention

The cases due to external pressure and excessive elbow motion can be prevented by avoiding the precipitating factors.

Differential diagnosis

Weakness of non-ulnar C8 muscles is the usual clue to disease involving the lower brachial plexus, C8 root, or cervical spinal cord. This finding should, in turn, prompt further examination of the cervical spine and a check for Horner syndrome. Elderly patients, particularly those with dementia, Parkinson disease, and other debilitating illnesses, may have nonspecific atrophy of the hand muscles that should not be mistaken for ulnar neuropathy. An ulnar nerve lesion at the wrist can usually be identified by more significant slowing of nerve conduction across the wrist than across the elbow, a normal dorsal ulnar cutaneous sensory potential, and by lack of denervation in the forearm. A lesion of the deep palmar branch produces a characteristic, but complicated electrodiagnostic picture. True neurogenic thoracic outlet syndrome classically produces ulnar distribution paresthesias and sensory loss accompanied by median distribution weakness and atrophy. Possible diagnoses of amyotrophic lateral sclerosis and syringomyelia can be excluded by identification of sensory nerve abnormality and the finding of denervation restricted to ulnar-innervated muscles. A polyneuropathy presenting as an ulnar nerve lesion can be recognized by identifying generalized abnormalities of nerve conduction. Several neurologic and non-neurologic conditions can produce an abnormal hand posture (pseudoulnar claw) that could be confused with an ulnar griffe (17).

Dupuytren contracture is a painless thickening characterized by fibroblastic proliferation and disorderly collagen deposition of tissue beneath the skin on the palm of the hand and fingers. Subsequently, nodules will form due to contraction of fibroblasts in the superficial palmar fascia. The possibility of a T-cell mediated autoimmune disorder as a cause of Dupuytren contracture is suggested by the demonstration of the presence of CD3-positive lymphocytes and the expression of major histocompatibility complex class II proteins along the affected areas. The flexor tendons are not intrinsically involved, but invasion of the dermis occurs, resulting in the characteristic puckering and tethering of the skin. Progressive contracture results in deformity and loss of function of the hand. This disorder is usually observed in white males older than 50 years, and, curiously, it appears to have a pronounced genetic predisposition with up to 68% of male relatives of affected patients developing the disease at some time. Although the treatment of choice is surgery, indications for the timing of surgery exist. In general, operative management should be performed on the affected MCP joint if the contracture is 30 degrees or greater. Usually, a limited fasciectomy of the pretendinous cord is sufficient to establish normal function in the corresponding joint. Techniques for this are also varied. Some favor use of a regional fasciectomy of the pretendinous cord to prevent recurrence. Since its initial description during the sixteenth century, Dupuytren disease has been exhaustively researched. However, controversy corresponding to the exact cause of the disease is ongoing. Future nonoperative therapies may include percutaneous needle fasciotomy, skeletal traction, therapy with calcium channel blockers, and treatment with gamma interferon. The last of these treatments shows the most promise. Until any of these treatments is proven to be therapeutic, surgery is still the only option for a cure.

Diagnostic workup

The primary diagnostic test for evaluation of suspected ulnar neuropathy at the elbow is a nerve conduction study. In certain patients, plain films, MRI of the elbow, or a laboratory evaluation to screen for an underlying susceptibility to compression neuropathy may be useful (31; 12). High-resolution ultrasound is a noninvasive, safe, and reliable modality for imaging the ulnar nerve at the elbow and it may provide a valuable adjunct to nerve conduction studies in the diagnosis of ulnar neuropathy at the cubital tunnel (71). In patients with abnormal nerve conduction studies at the elbow, a cutoff value of 1.5/1 for the ratio of the nerve cross-sectional area (CSA; equal to the area of maximal swelling divided by the area of a normal nerve segment 2 cm proximally or distally) is highly sensitive (100%) and specific (96.7%) (74). When the nerve conduction study is normal, ultrasound may still reveal an increased ulnar nerve cross-sectional area near the elbow. However, the authors did not use more sensitive nerve conduction techniques such as motor recording from the FDI or motor inching across the elbow (75). Ulnar sensory short segment stimulation may even provide additional diagnostic confirmation and localization of the site of nerve compression in mild ulnar neuropathy at the elbow where routine nerve conduction studies are normal (68). Ulnar nerve entrapment confirmation can be optimized by using not only nerve CSA at the elbow, but also the proximal to distal CSA ratio method, while dividing the elbow area into four segments with 2-cm intervals (65). Another ultrasound study found that the highest CSA and lowest nerve conduction velocity were both noted in the same short segment (Podnar and Omejec 2017). The authors of this study also reported that in the majority of cases (54/106), the highest CSA and lowest nerve conduction velocity were noted surprisingly proximal to the humeroulnar aponeurosis. In a study of 56 patients with clinical and electrodiagnostic evidence of ulnar neuropathy, 12 cases were nonlocalizable. Ultrasound parameters assessed included cross-sectional area (cut-off value 10 mm2) and subjective measurements of echogenicity, fascicular structure, and vascularity. High resolution ultrasound added value by localizing all 12 cases; one was at the wrist and another at the retrocondylar groove with a tandem lesion higher in the proximal arm (02). The remaining 10 cases were localized to the elbow area; five at the retrocondylar groove, two at the cubital tunnel, and three at both the retrocondylar groove and cubital tunnel. In an ultrasound study of 34 patients with ulnar neuropathy at the elbow compared to 38 healthy controls, shear-wave elastography was reported to be a promising adjunct diagnostic method to confirm ulnar neuropathy. There was a weak positive correlation between the ulnar nerve cross-sectional area and nerve stiffness (54).

Diffusion-weighted imaging (DWI) and conventional MRI in patients with ulnar neuropathy at the elbow was compared to healthy controls (03). Although no abnormal signal was detected in nerves of healthy controls, there was an increase in signal on DWI. In all patients with ulnar neuropathy and on T2-weighted imaging, there was increased signal intensity in most cases. Using MR neurography, diffusion tensor imaging, and tractography at 3 Tesla to assess 46 patients with ulnar neuropathy at the elbow in comparison to 20 controls, an increase in T2 nerve signal intensity on T2 neurography was most sensitive in detecting ulnar neuropathy at the elbow (sensitivity, 91%; specificity, 79%), followed by tractography (88%; 69%) (11). Tractography showed a deranged ulnar nerve appearance in 20%, a partially disrupted course in 25%, and a completely disrupted course of the ulnar nerve in 40%. Though patients showed a significant reduction of ulnar nerve fractional anisotropy values at the retrocondylar sulcus and the deep flexor fascia, cross-sectional area and fractional anisotropy measurements were less effective in detecting ulnar neuropathy at the elbow.

For standard ulnar nerve motor conduction technique, recording electrodes are placed over an ulnar-innervated muscle in the hand (usually the abductor digiti quinti), and the nerve is stimulated at the wrist, below the elbow, and above the elbow, calculating nerve conduction velocity from below elbow to wrist, above elbow to wrist, and above elbow to below elbow. There have been several areas of controversy regarding ulnar conduction studies, including debate over the best elbow position, the ideal length of the across elbow segment, and the value of absolute slowing in the above elbow to below elbow segment in contrast to relative slowing in the above elbow to below elbow segment as compared to the below elbow to wrist segment. The difficulties with elbow position relate to the discrepancies between true nerve length and measured skin distance in different elbow positions. In extension, the nerve is redundant, and skin distance is falsely short compared to true nerve length, causing spurious and artifactual conduction slowing (25; 18). Because of differences in technique in different laboratories, normal values may vary enormously. A practice parameter on the electrodiagnosis of ulnar neuropathy at the elbow concluded that the most logical elbow position for ulnar conduction studies was moderate flexion, 70° to 90° from horizontal (Table 1) (18).

Nerve conduction studies can localize a lesion to the elbow by either of two criteria: (1) slowing of the absolute conduction velocity in the across-elbow segment as compared to reference values, or (2) slowing of the elbow segment in relation to the below elbow to wrist segment. After a review of the literature, a practice parameter recommended an absolute above elbow to below elbow conduction velocity of less than 50 m/sec, or slowing of greater than 10 m/sec in the above elbow to below elbow segment compared to the below elbow to wrist segment, which must be considered suggestive of ulnar neuropathy at the elbow (Table 1) (18). A reduction in compound muscle action potential amplitude of more than 20% or a significant change in compound muscle action potential configuration between the below elbow and above elbow sites are also suggestive of ulnar neuropathy at the elbow (40; 18). A Martin-Gruber anastomosis must be excluded before relying on across elbow compound muscle action potential amplitude changes for localization. Some patients with ulnar neuropathy at the elbow have no or minimal detectable conduction abnormalities, despite obvious clinical and electrical evidence of an elbow level lesion: the so-called pure axon loss ulnar neuropathy. Abnormalities of the ulnar sensory potential are sensitive for detecting ulnar nerve pathology and determining its severity but are seldom useful for localization purposes. A comparison investigation found that motor studies were more sensitive than the ulnar mixed sensory study across the elbow in localizing ulnar neuropathy of the elbow (41). Motor conduction studies have shown localizing abnormalities in symptomatic elbows with a sensitivity of 37% to 100% (18). Precise localization can often be achieved by short segment incremental studies (inching) and monitoring the compound muscle action potential when moving the stimulator in discrete, small steps in search of a telltale sudden change in amplitude, configuration, or latency (29; 22; 19). Short ulnar segmental nerve conduction studies or “inching” significantly improves detection of ulnar mononeuropathy at the elbow and should be considered when routine studies are negative and clinical suspicion remains high (36). Conduction studies recording from the first dorsal interosseous may show abnormalities not evident when recording from the abductor.

Table 1. Synopsis of the Recommendations of the American Association of Electromyography and Electrodiagnosis Practice Parameter on Ulnar Neuropathy at the Elbow

Although generally not as sensitive as nerve conduction studies, needle electrode examination to detect denervation of ulnar-innervated hand muscles is a helpful adjunct. Needle exam abnormalities may be the only manifestation of a pure axon loss ulnar lesion. The first dorsal interosseous is the most frequently involved muscle, followed by the abductor digiti minimi, flexor digitorum profundus, and flexor carpi ulnaris, respectively (64).

The forearm muscles, flexor carpi ulnaris, and flexor digitorum profundus are frequently spared in ulnar neuropathy at the elbow, so the lack of clinical or electromyographic abnormality in these muscles in no way excludes a lesion at the elbow (20).

Management

The treatment of ulnar neuropathy at the elbow is generally considered surgical if patients fail conservative management. Many patients, especially those with mild involvement, may recover spontaneously or with conservative treatment (27; 28). In a systematic review and meta-analysis of conservative treatment of ulnar neuropathy at the elbow, 20 studies were eligible for analysis. Out of the 687, 54% improved after steroid/lidocaine injection, whereas 89% responded to splinting.

The best surgical outcomes occur in patients with the least severe neuropathies and the shortest duration of symptoms; patients with end-stage neuropathies do poorly (01). Ulnar nerve surgery, however, does not have the excellent success record of carpal tunnel release; studies generally report a good outcome rate of approximately 70%, and preoperative nerve conduction studies are of prognostic value (47).

Conservative management entails avoidance of repetitive flexion and extension of the elbow, avoidance of external pressure, resting the elbow, and the use of an extension splint or elbow pad (Table 2). Elbow pads accomplish a dual purpose: they restrict elbow flexion while simultaneously protecting the nerve from external pressure. Though the addition of a local steroid injection did not significantly improve the results of simple splinting (39), ultrasound-guided steroid injection is a promising therapy for patients with mild ulnar neuropathy at the elbow (59). Six weeks after injection, four patients were improved, two were stable, but one reported an increase in symptoms. In a randomized controlled trial, 55 patients received ultrasound-guided injection of 1 ml containing 40 mg methylprednisolone acetate and 10 mg lidocaine hydrochloride or a placebo injection. Unfortunately, the success rate as measured by the subjective change of symptoms after three months was not different between the two groups (30% in the corticosteroid injection group versus 28% in the placebo injection group) (67). In a randomized controlled study of mild to moderate ulnar neuropathy at the elbow, patients receiving perineurial injections of either dextrose or triamcinolone were followed for 6 months (26). Although both agents were found to be equally “effective,” the dextrose group had surprisingly larger magnitude of clinical improvement between months 3 and 6. In yet another randomized controlled trial of ulnar nerve entrapment at the elbow, 10 sessions of shortwave diathermy was not more effective than placebo diathermy on visual analog scale, grip strength, SF-36, and quick-DASH (10). In a randomized controlled study of perineurial dextrose injection under ultrasound guidance twice at 2-week intervals with follow-up for 12 weeks, pain, disability, ulnar motor nerve velocity, and ulnar nerve cross-sectional area were improved in comparison to normal saline injections (46).

Surgery is not usually necessary as long as there is no motor involvement or objective sensory loss and symptoms do not progress. Careful follow-up of both clinical and electrodiagnostic parameters is important. In a prospective study of patients with mild to moderate cubital tunnel syndrome, 19 cases (25 elbows) were treated with activity modifications and a 3-month course of rigid night splinting maintaining 45° of elbow flexion (61). Splints were well tolerated and most extremities (88%) were successfully treated without surgery. Disability scores improved significantly as well as the physical and mental component of Short Form-12 along with an increase in average grip strength.

Table 2. Conservative Treatment of Ulnar Neuropathy at the Elbow

When surgery becomes necessary, the procedures most commonly employed are simple decompression, anterior transposition, and medial epicondylectomy. Investigators have been hard pressed to demonstrate any clear advantage of any particular procedure (01). A 2008 meta-analysis concluded that there is no statistically significant difference in the outcome of cubital tunnel syndrome treated with simple decompression versus anterior subcutaneous or submuscular nerve transposition (44). Simple decompression is done by dividing the humeroulnar aponeurotic arcade and is most appropriate for those cases where the clinical and electrodiagnostic picture points to entrapment by the bridging aponeurosis (43). The advantages of the simple release are its simplicity and low morbidity. In medial epicondylectomy, the medial epicondyle is excised, removing the major bony block causing compression of the nerve (35). An effective endoscopic surgical technique was described (76). In a randomized controlled trial of patients with mild ulnar neuropathy at the elbow, 85% of those receiving in situ decompression had symptom relief at 3 months as compared to 50% with conservative treatment. However, long-term outcomes for up to one year were no different between the two groups (57). In patients with severe ulnar neuropathy at the elbow, postsurgical electrical stimulation (PES) was shown to enhance axonal regeneration at 3 years (58). Patients randomized to postsurgical electrical stimulation did better than those allocated to placebo on several measures such as motor unit number estimates and key pinch strength.

Anterior transposition is the most common ulnar nerve operation. It is done by mobilizing the nerve, freeing any attachments at the humeroulnar aponeurotic arcade or medial intermuscular septum, then moving the nerve anterior to the medial epicondyle. The two types of transposition methods now commonly employed are the subcutaneous, or superficial, and the submuscular, or deep. The nerve is placed superficial to the flexor pronator muscle mass for subcutaneous transposition, and deep to it for submuscular transposition. Complications from any of these procedures include persistent or recurrent symptoms due to inadequate surgery or to recurrent scarring of the nerve. Patients are occasionally left with painful paresthesias and significant disability due to chronic pain. Submuscular transposition is often used as a salvage procedure following failed anterior subcutaneous transposition (53; 05). There has been renewed interest in the old surgical concept of reconstruction of the ulnar groove (66).

The main advantage of decompression and medial epicondylectomy over transposition is preservation of the ulnar nerve blood supply, as extensive dissection with the possibility of devascularization of the ulnar nerve is unnecessary. Additionally, cadaver studies have shown that tethering the ulnar nerve behind the medial epicondyle is the main cause of increased intraneural pressure (38). Only one prospective, randomized study compares the outcome of medial epicondylectomy with that after anterior transposition (35), and this found a better postoperative outcome after medial epicondylectomy.

Medial epicondylectomy addresses the pulley effect of the medial epicondyle on the ulnar nerve. This procedure allows the nerve to fall forward in a new relaxed position, choosing the path of least resistance. The negative side effects can include bony tenderness, nerve subluxation, flexor-pronator weakness, flexion contracture, and valgus instability (50).

As these treatment options are being considered, it is important to bear in mind the Cochrane review that highlights the uncertainty as to when to treat these patients conservatively or surgically (15). The literature suggests that simple decompression and decompression with transposition are equally effective. In milder cases, information on movements or triggering positions to be avoided may reduce subjective discomfort, but the same does not hold true for night splinting and nerve gliding.

Table 3. Advantages and Disadvantages of the Surgical Procedures for Cubital Tunnel Syndrome

Outcomes

Dellon and colleagues prospectively evaluated 128 patients with ulnar neuropathy at the elbow managed nonoperatively (28). Only 11% of patients with symptoms alone eventually had surgery as compared with 33% with mild to moderate abnormalities on exam and 62% with more severe abnormalities. The presence of persistent paresthesias, abnormal 2-point discrimination, and any degree of muscle wasting moved patients into the category where eventual surgery was likely. A history of elbow injury significantly worsened the outcome. The best surgical approach remains a matter of dispute and likely differs from patient to patient. A 27-year literature review was unable to formulate a uniform guideline for operative treatment (08). A Cochrane database review concluded that the best therapy for ulnar neuropathy at the elbow is unknown (15). Simple decompression and decompression with transposition are equally effective in idiopathic ulnar neuropathy at the elbow, including when the nerve impairment is severe. In mild cases, information on movements or positions to avoid may reduce subjective discomfort. These findings were confirmed in a 2016 Cochrane database review update (16). We still do not know when to recommend treating this condition conservatively or surgically. Decompression with transposition is associated with more wound infections than simple decompression whereas endoscopic surgery is associated with increased risk for hematoma. From an economic standpoint, simple decompression as an initial procedure is the most cost-effective treatment strategy (62).

Complications of ulnar nerve surgery include neuroma of the medial antebrachial cutaneous nerve, resubluxation of the transposed nerve back into the ulnar groove, progression of the neuropathy, and the development of chronic pain (53). Failed ulnar nerve surgery is often due to inadequate release of tethering attachments, particularly at the medial intermuscular septum, producing secondary kinking.

Special considerations

Anesthesia

During any operation, the elbows should be adequately padded and the forearm placed in supination to avoid pressure on the ulnar groove. Unfortunately, ulnar neuropathy at the elbow may occur after anesthesia in spite of seemingly adequate padding and positioning. Postanesthetic ulnar neuropathy at the elbow has a poor prognosis (48). Some patients who develop postanesthetic palsies have pre-existent ulnar nerve pathology and some postanesthetic neuropathies are subclinical (70).