Applied Anatomy

The ulnar nerve sensory and motor fibers are derived from the spinal nerves C8 and T1. Before arising from the plexus in the proximal axilla, the ulnar nerve fibers pass through the lower trunk and the medial cord of the brachial plexus (Figure C15-2). In the axilla and proximal arm, the ulnar nerve is closely related to the radial and median nerves and the brachial artery. Around the midarm, the ulnar nerve pierces the intermuscular septum and lies in close contact with the medial head of the triceps and humerus. The ulnar nerve develops no branches in the arm.

Figure C15-2 Course of the ulnar nerve and its branches. 1. Palmar ulnar cutaneous branch; 2. dorsal ulnar cutaneous branch; 3. terminal cutaneous superficial branch; 4. terminal deep palmar motor branch.

(From Haymaker W, Woodhall B. Peripheral nerve injuries. Philadelphia, PA: WB Saunders, 1953, with permission.)

At the elbow level, and in contrast to most major human peripheral nerves, the ulnar nerve traverses the extensor, rather than the flexor, surface of the elbow joint. This renders the nerve more vulnerable to trauma around the elbow. At the elbow, the ulnar nerve crosses the ulnar groove (also called the condylar or retroepicondylar groove) behind the medial epicondyle and then passes the aponeurotic arch of the FCU to enter the cubital tunnel (Figure C15-3). This tunnel, also called the humeroulnar arcade, Osborne ligament, or arcuate arcade, is formed by the attachment of the muscle to the olecranon and medial epicondyle. Its proximal edge is variable but usually is approximately 1 cm distal to an imaginary line drawn between these two insertional points. With flexion of the elbow, the distance between the olecranon and medial epicondyle increases by approximately 1 cm, which results in tightening of the FCU aponeurosis over the nerve. In addition, the medial elbow ligament bulges, flattening the concave surface of the ulnar groove.

Figure C15-3 View of the medial surface of the elbow, showing the course of the ulnar nerve through the ulnar groove and cubital tunnel.

(From Kincaid JC. The electrodiagnosis of ulnar neuropathy at the elbow. Muscle Nerve 1988;11:1005–1015, with permission.)

In the forearm, the ulnar nerve gives off its first branches. These are the motor branches to the FCU and FDP. These branches arise approximately 10 cm distal to the medial epicondyle. The ulnar nerve continues in the forearm deep to the FCU but superficial to the FDP to become superficial in the distal forearm, lying between the tendons of these two muscles. Two cutaneous sensory branches arise in the forearm, without passing through Guyon canal at the wrist, to innervate the skin in the hand. The first is the palmar ulnar cutaneous branch, which takes off at midforearm and innervates the proximal part of the ulnar border of the palm. The second is the dorsal ulnar cutaneous branch, which arises 6 to 8 cm proximal to the ulnar styloid, winds around the ulna, and innervates the dorsal surfaces of the little finger and half of the ring finger, along with the ulnar side of the dorsum of the hand (Figure C15-4).

Figure C15-4 The cutaneous distribution of the three sensory branches of the ulnar nerve.

(From Stewart JD. The variable clinical manifestations of ulnar neuropathies at the elbow. J Neurol Neurosurg Psychiatry 1987;50:252–258, BMJ Publishing Group.)

At the wrist, the ulnar nerve enters the distal ulnar tunnel (Guyon canal), where it divides into superficial (primarily sensory) and deep palmar (pure motor) branches (Figure C15-5). The superficial branch innervates the palmaris brevis muscle and the palmar aspects of digit V and half of digit IV. The deep branch innervates the hypothenar muscles, including the ADM, and travels through the palm to the dorsal and palmar interossei, the third and fourth lumbricals, the adductor pollicis, and a portion of the flexor pollicis brevis.

Figure C15-5 Anatomy of the ulnar nerve within Guyon canal at the wrist. 1 = ulnar artery, 2 = superficial branch of the ulnar nerve, 3 = hamulus, 4 = fibrous arch of the hypothenar muscles (see also Figure C15-2), 5 = pisiform, 6 = transverse carpal ligament, 7 = palmaris brevis, 8 = palmar carpal ligament.

Rights were not granted to include this figure in electronic media. Please refer to the printed book.

(From Gross MS, Gelberman RH. The anatomy of the distal ulnar tunnel. Clin Orthop 1985;196:238–247, with permission.)

The flexor brevis digiti minimi (or quinti), a hypothenar muscle, has two separate attachments, at the hook of the hamate and at the pisiform bone. These musculotendinous attachments form a fibrous arch and create the superficial boundary of the pisohamate hiatus (PHH), an opening through which the deep motor branch of the ulnar nerve passes. The posterior boundary of the PHH, the pisohamate ligament, extends from the pisiform bone to the hook of the hamate (Figure C15-6). The origin of the major motor branch to the ADM is proximal to this hiatus in the majority of hands.

Figure C15-6 The pisohamate hiatus (PHH) in the distal portion of Guyon canal. U. = Ulnar; F.C.U. = flexor carpi ulnaris; ABD.D.Q. = abductor digiti quinti (or minimi); F.B.D.Q. = flexor brevis digiti quinti (or minimi); Opp.D.Q. = opponens digiti quinti.

(Modified from Uriburu IJF, Morchio FJ, Marin JC. Compression syndrome of the deep motor branch of the ulnar nerve [pisohamate hiatus syndrome]. J Bone Joint Surg 1976;58A:145–147, with permission.)

Clinical Features

Ulnar mononeuropathy across the elbow is a common entrapment neuropathy. It is second only to carpal tunnel syndrome in the incidence of entrapment neuropathy in general. Ulnar neuropathies across the elbow are usually caused by compression, although isolated ulnar mononeuropathy resulting from nerve infarction (such as in vasculitic neuropathy) or associated with leprosy, may occur infrequently. Causes of compressive ulnar lesions across the elbow are shown in Table C15-1. Compression of the ulnar nerve in the elbow region occurs frequently at one of the two following sites: the ulnar (condylar) groove or cubital tunnel (humeroulnar (arcuate) aponeurotic arcade). In some patients with unequivocally ulnar nerve lesions around the elbow, it is difficult to identify the exact cause, even during surgery. Most surgeons presume that the lesion is within the cubital tunnel and treat it as such.

Table C15-1 Causes of Compressive Ulnar Nerve Lesions Across the Elbow

Modified from Dimitru D. Electrodiagnostic medicine. Philadelphia, PA: Hanley and Belfus, 1995.

Patients with ulnar mononeuropathy at the elbow usually present with numbness and tingling of the little and ring fingers and variable degrees of hand weakness. Less commonly, patients present with weakness and wasting, with no clear sensory symptoms. Elbow pain, particularly around the medial epicondyle, is not uncommon. Weakness of hand with loss of dexterity and pinch strength are common symptoms in moderate entrapment. Patients may report that their little finger gets caught when trying to put their hand in their pocket (due to weakness of the third palmar interosseous muscle). Occasionally, patients may present because of intrinsic muscle hand atrophy.

Neurologic examination may reveal a positive Tinel sign. This is produced by percussion of the ulnar nerve at the elbow; however, this sign may be positive in many healthy subjects. Sensory examination may be normal despite sensory symptoms. More often, there is a relative sensory loss in the tips of both the little and ring fingers on the palmar surface only. Unfortunately, this finding is a poor localizing sign because it is abnormal with ulnar nerve lesions (anywhere at, or proximal to, the wrist), lower brachial plexus lesions (lower trunk or medial cord), and C8 radiculopathy. However, other sensory findings are more helpful in localization. These include the following:

Inspection of the hand at rest in patients with moderate or severe ulnar mononeuropathy may reveal ulnar clawing. An ulnar claw hand, also called Benediction posture (Figure C15-7), is caused by: (1) hyperextension of the metacarpophalangeal joints of the little and ring fingers caused by weakness of the third and fourth lumbricals, thus allowing the extensor digitorum communis to exert an unopposed pull; and (2) flexion of the interphalangeal joints of the same fingers resulting from an inherent flexion muscle tone of the FDP and superficialis muscles, whose tendons are stretched over the metacarpophalangeal joints because of the above hyperextension. In ulnar lesions, this clawing is more noticeable when the FDP is spared. The Wartenberg sign is recognized as abduction of the little finger at rest due to weakness of the third palmar interosseous muscle.

Figure C15-7 Ulnar claw hand. Note the hyperextension of the fourth and fifth metacarpophalangeal joints with flexion of the interphalangeal joints.

(From Haymaker W, Woodhall B. Peripheral nerve injuries. Philadelphia, PA: WB Saunders, 1953, with permission.)

Weakness of ulnar-innervated muscles in the hand predominates in ulnar nerve lesions across the elbow; the forearm muscles are affected less often. Weakness of the FDP to the fourth and fifth digits is assessed by flexion of the distal interphalangeal joints of these digits. Positive Froment sign is helpful in the clinical diagnosis of ulnar neuropathy because it shows the weakness of the adductor pollicis (ulnar muscle) and the normal flexor pollicis longus (median muscle), both of which are innervated by the C8/T1 roots via the lower plexus. This sign is assessed by asking the patient to grasp a piece of paper between the thumb and second digit. Because of weakness of the adductor pollicis, the patient uses the flexor pollicis longus as a substitute in an attempt to keep the paper from sliding (Figure C15-8).

Figure C15-8 Froment sign in an ulnar nerve lesion. The patient is asked to pull a piece of paper apart with both hands. Note that the right hand (affected hand with interosseous atrophy) flexes the thumb (by using the flexor pollicis longus) to prevent the paper from slipping out of the hand, thus substituting for the weakness of the adductor pollicis.

(From Haymaker W, Woodhall B. Peripheral nerve injuries. Philadelphia, PA: WB Saunders, 1953, with permission.)

Many patients with ulnar nerve lesions at the elbow have variable weakness and numbness distally. This is explained by the propensity for partial focal lesions to affect fascicles differentially within that nerve. This fascicular phenomenon is common in ulnar nerve lesions across the elbow and is demonstrated in Figure C15-9. Atrophy of ulnar muscles in the hand is common with long-standing lesions and is most apparent in the interossei, particularly the first dorsal interosseous (see Figure C15-8).

Figure C15-9 Clinical abnormalities in the distribution of three sensory and four motor branches of the ulnar nerve in 25 patients with ulnar neuropathy at the elbow. “Only” is used to denote the number of patients in whom a single sensory area was involved, as well as to denote those patients with weakness of only one of the four muscles. ADM = abductor digiti minimi, DC = dorsal cutaneous, FCU = flexor carpi ulnaris, FDI = first dorsal interosseous, FDP = flexor digitorum profundus, PC = palmar cutaneous, TD = terminal digital branches.

(From Stewart JD. The variable clinical manifestations of ulnar neuropathies at the elbow. J Neurol Neurosurg Psychiatry 1987;50:252–258, BMJ Publishing Group.)

Treatment of ulnar neuropathy at the elbow may be conservative or surgical. Identification and reversal of the cause of repetitive compression is essential. Many patients with mild symptoms and signs who demonstrate evidence of slowing or conduction block on nerve conduction studies may be treated successfully using conservative approaches (Table C15-2). Patients with substantial weakness, particularly when progressive or associated with evidence of axonal loss or ongoing denervation, benefit from surgery. There is no clear consensus on the optimal surgical procedure. Simple decompression of the cubital tunnel may be ideal for patients with cubital tunnel syndrome, while medial epicondylectomy or submuscular transposition of the ulnar nerve are more suitable for compression at the ulnar groove. Submuscular transposition has a higher success rate but is a more complex surgery and carries a risk of nerve devascularization and ischemia.

Table C15-2 Conservative Measures in the Treatment of Mild Ulnar Neuropathy at the Elbow

Sensory NCS

Motor NCS

Ulnar Motor Conduction Studies, Recording the Hypothenar Muscles (ADM)

This is a routine motor conduction study that is performed in many laboratories. The nerve is stimulated at the wrist, below the elbow, and above the elbow. Stimulation at the axilla also is sometimes useful because it allows comparison of two segments (forearm and arm) to the across-elbow segment and also evaluate for rare ulnar nerve lesions in the arm. Erb point (supraclavicular) stimulation is also occasionally helpful in excluding a high ulnar nerve lesion (Figure C15-10). The conduction abnormalities observed across the elbow are:

• Conduction block. With focal segmental demyelination at the elbow, the CMAP that is elicited with above-elbow stimulation is much lower in amplitude (>20–30% decrease in amplitude and area) than the response obtained with below-elbow stimulation without CMAP dispersion. All patients with apparent conduction block of the ulnar motor fibers should undergo further investigation to rule out the presence of an anomalous connection between the median and ulnar nerves, named Martin-Gruber anastomosis. This occurs in approximately 20% of the population and is sometimes bilateral. It may be established easily by stimulating the median nerve while recording the ulnar-innervated muscles. When present, a small CMAP is recorded and accounts for the apparent conduction block in the forearm. Occasionally, an ulnar lesion across the elbow may occur in a patient with this anastomosis (Figure C15-11).

• Focal slowing. With mild (mostly paranodal) demyelination, there is slowing of conduction across the elbow. This is confirmed by comparing the conduction velocities across the elbow, such as the above-elbow-to-below elbow segment to the below-elbow-to-wrist (forearm) segment. Comparison with the contralateral side in unilateral cases also is helpful. A combination of conduction block and focal slowing is a common presentation of acute/subacute ulnar lesions across the elbow (Figure C15-12).

• Axonal loss. In cases in which wallerian degeneration has occurred, ulnar CMAP is low in amplitude at all points of stimulation, with no significant change in configuration, amplitude, or conduction velocity. The conduction velocities may be mildly slowed diffusely if there is significant axonal loss (due to the loss of large fibers), but no focal slowing is present. Unfortunately, motor conduction studies are unable to accurately localize the site of ulnar lesion in this situation.

• Combinations of the above. This is a common finding, particularly with chronic ulnar lesions across the elbow (Figure C15-13).

Figure C15-10 A 35-year-old man awoke with severe left ulnar nerve palsy resulting in ulnar clawing and sensory loss in palmar and dorsal aspect of medial hand. Initial routine nerve conduction study, recording hypothenar muscles, was normal except for absent ulnar F wave. Subsequent, five point stimulation of the ulnar nerve that included axilla and Erb point stimulations revealed complete conduction block between these two points (A) with normal responses on the asymptomatic side (B). The patient reported later that he had slept in the bathtub, after drinking heavily that night, and draped his left arm over the edge of the tub to keep his head above water. He had complete recovery of function in 6 weeks.

Figure C15-11 Ulnar motor nerve conduction study, recording hypothenar muscle, in a patient with Martin-Gruber anastomosis and ulnar mononeuropathy across the elbow. Note that there are two drops in amplitude and area, one in the forearm and another across the elbow. The amplitude decay in the forearm (6.8 mV at wrist and 2.5 mV below elbow) is explained by Martin-Gruber anastomosis as evidenced by a large CMAP (3.0 mV) obtained by stimulating the median nerve at the elbow, while recording the hypothenar muscles. This response is not present upon stimulating the median nerve at the wrist. The block across the elbow (2.5 mV below elbow and 1.0 mV above elbow) is also associated with focal slowing and is consistent with a concomitant ulnar mononeuropathy across the elbow.

Figure C15-12 Ulnar motor nerve conduction study, recording hypothenar muscle, in a patient with several months’ history of ulnar mononeuropathy, revealing evidence of conduction block and focal slowing across the elbow. The distal (wrist) CMAP amplitude is normal (8.5 mV) which is consistent with no (or minimal) axon loss.

Figure C15-13 Ulnar motor nerve conduction study, recording hypothenar muscle, in a patient with a year history of ulnar mononeuropathy, revealing evidence of conduction block, focal slowing, and differential slowing (dispersion) across the elbow, along with axon loss. The latter is based on very low amplitude distal (wrist) CMAP (1.8 mV).

Needle EMG

Needle examination is useful in confirming an ulnar nerve lesion and in excluding a C8/T1 root lesion or a lower brachial plexopathy. It is important to establish that the C8/T1 muscles innervated by the median nerve (such as the abductor pollicis brevis or the flexor pollicis longus) and radial nerve (such as the extensor indicis proprius) are normal. In addition, in purely axonal lesions of the ulnar nerve, needle EMG is crucial in localizing the lesion to a particular segment of the nerve, which it does by establishing that muscles distal to the lesion are abnormal and muscles proximal to it are normal. Unfortunately, there are limitations to the accurate localization by needle EMG in patients with axonal ulnar neuropathies:

Electrodiagnostic Controversies

Ongoing controversies in the electrodiagnostic evaluation of ulnar mononeuropathy across the elbow include the following:

In Which Position Should the Elbow Be Placed During Performance of Ulnar Nerve Conduction Studies (Extension or Degree of Flexion)?

The ulnar nerve is anatomically lax and redundant in the ulnar groove when the elbow is extended. This is aimed to provide the extra nerve length needed during elbow flexion. With flexion the nerve uncoil and become more taut, and when flexion bypass 90°, the ulnar nerve may sublux in up to 20% of individuals.

Hence, surface measurements of the ulnar nerve while the elbow is extended do not reflect the true extent of the underlying nerve. In fact, the distance measured over the skin is shorter than the true length of the nerve. Hence, the conduction velocity is spuriously slowed because the impulses travel longer distances than can be estimated on skin measurement (Figure C15-14). On the other hand, surface measurements of the ulnar nerve while the elbow is hypreflexed more than 90° may be longer than the true nerve length (due to potential nerve subluxation) resulting in artifactually fast conduction velocity.

Figure C15-14 Correlation between the true nerve distance and the skin surface distance of the ulnar nerve during extension (A) and flexion (B). The lack of correlation with the elbow in extension leads to spurious slowing of conduction velocity.

(From Campbell WW. Electrophysiological approaches to the diagnosis and assessment of ulnar neuropathy: a historical and literature review. In: 2002 American Association of Electrodiagnostic Medicine plenary session, Toronto, Ontario, with permission.)

Despite the above findings, advocates for flexion or extension techniques continue to debate this issue and whether elbow position influences the sensitivity and specificity of motor nerve conduction studies in the diagnosis of ulnar mononeuropathy across the elbow. Also, there is no consensus as to the optimal degree of elbow flexion, ranging from 45° to 135°, among advocates of flexion techniques. Surface measurement over the skin with the elbow flexed at 45° to 90° most closely correlates with the true length of the nerve. This is the angle range advocated by many electromyographers. Hyperflexion beyond 90° is, however, not recommended.

Are Short Segment Incremental Studies (“Inching”) Required for Accurate Diagnosis?

The “inching” technique is helpful when a conduction block or slowing is seen between the above-elbow and below-elbow stimulations. It is recommended by some for precise (point) localization of the lesion in patients with suspected ulnar nerve lesions across the elbow. This may have both clinical and surgical implications, particularly in planning the optimal surgical procedure.

The stimulation site is moved progressively along the nerve across the elbow, in several equal steps (usually 1–2 cm), between the below and above elbow (over about 6–8 cm span) and CMAP amplitude, latency or morphology are recorded. This study should be done meticulously since it is subject to measurement error and volume conduction. An abrupt change in CMAP amplitude, morphology or latency (>0.5 ms) is noted (Figure C15-15). The short increment technique could also be used to point localize lesions in other segments of the ulnar nerve, such as the arm or forearm (Figure C15-16).

Figure C15-15 Short incremental (“inching”) motor conduction study, recording hypothenar muscles, in a patient with ulnar neuropathy across the elbow. The reference point is at the medial epicondyle (ME) and the negative numbers are successive distal stimulation points while the positive points are successive proximal stimulation sites (A). Note in (B) that there is a more prominent latency change of 0.8 ms and amplitude decay of 85% between points +2 and +3 (arrow), localizing the lesion between 2 and 3 cm above the medial epicondyle. The latency change between all other successive points ranged between 0.1 to 0.3 ms while the amplitude decay ranged between 5 and 20%.

Figure C15-16 A 65-year-old woman presented with a year history of intermittent numbness of left little and ring fingers. All ulnar nerve conduction studies were normal except for slowing of ulnar motor fibers, recording hypothenar muscles (A), in the forearm segment (39 m/s) compared to across the elbow segment (68.5 m/s). Similar slowing was noted recording first dorsal interosseous muscle (not shown). Subsequent short incremental (“inching”) study, recording hypothenar muscles, starting distally at the ulnar styloid and moving in 2 cm increments proximally, revealed a significant latency shift between 14 cm and 16 cm proximal to the styloid (B) (responses are shown superimposed). Subsequent MRI revealed a mass at the site and a schwannoma was removed successfully.