School of Medicine, King’s College London
Ulnar neuropathy at the elbow is the second most common mononeuropathy of the upper limb (Latinovic et al., 2006), after carpal tunnel syndrome. It is caused by compression of the ulnar nerve at sites of anatomical narrowing before it enters the forearm. Alongside a brief overview of the course of the ulnar nerve from its origin in the axilla to its termination in the hand, this review provides a comprehensive description of the four potential sites for compression that are encountered by the ulnar nerve. Anatomical variations of the cubital tunnel and the mechanisms for ulnar nerve compression at the site are emphasised given that this is the most common site of compression of the ulnar nerve. Finally, the review also discusses the pathophysiology of peripheral nerve injury and the symptomatology of ulnar neuropathy at the elbow.
An upper limb mononeuropathy is a lesion affecting a single peripheral nerve in the upper limb. Although there exists an extensive range of potential causes of upper limb mononeuropathy, the most common cause is nerve compression or entrapment (Andreisek et al., 2006). In theory, peripheral nerves may be compressed anywhere along their course. However, there are sites where peripheral nerves are at particular risk of compression. In the upper limb, these include the elbow and the wrist.
At the wrist, the peripheral nerves may be compressed where they pass through a tunnel to gain access to the hand. For instance, carpal tunnel syndrome, the most common mononeuropathy of the upper limb, is caused by compression of the median nerve in a fibro-osseous tunnel formed on the volar aspect of the wrist (Latinovic et al., 2006). At the elbow, the peripheral nerves may be compressed where they penetrate through a muscle to gain access to the forearm. For instance, a common site of compression of the ulnar nerve is where it passes between the humeral and ulnar heads of the flexor carpi ulnaris. Ulnar neuropathy at the elbow is the second most common mononeuropathy of the upper limb (Latinovic et al., 2006).
Anatomy (proximal to the elbow)
The ulnar nerve is a terminal branch of the medial cord of the brachial plexus. It contains C7, C8 and T1 fibres.
From the axilla, the ulnar nerve descends in the anterior compartment of the arm, closely related to the brachial artery. At around the level of the insertion of the coracobrachialis, the ulnar nerve deviates away from the path of the brachial artery and, in the middle third of the arm, enters the posterior compartment by passing through a narrow opening in the medial intermuscular septum. This is the first site of compression encountered by the ulnar nerve, during its course into the forearm (Posner, 1998). Altogether, there are four potential sites of compression which may cause ulnar neuropathy at the elbow (Figure 1).
Figure 1: A schematic diagram showing the four potential sites of compression of the ulnar nerve at the elbow. [Adapted from Grant’s atlas of anatomy].
1. Internal brachial ligament
The first site of compression of the ulnar nerve has been described by Struthers (Struthers, 1854) and Kane et al (Kane et al., 1973).
Struthers described a fibrous cord, the so-called “internal brachial ligament”, which separated from the medial intermuscular septum at around the level of the insertion of the coracobrachialis, and then reunited proximal to the medial epicondyle of the humerus. Note that Struthers did not consider this ligament to be part of the septum.
In contrast, Kane et al. described an “arcade of Struthers”, formed by the aforementioned internal brachial ligament and reinforced by superficial fibres of the medial head of the triceps and a thickening of the deep investing fascia of the arm. The arcade was bound anteriorly by a thickening of the medial intermuscular septum and laterally by the deep fibres of the medial head of the triceps. The group also speculated that the internal brachial ligament was a tendinous attachment of part of the medial head of the triceps. In this cadaveric study, the arcade of Struthers (identified based on this description) was found to be present in 70% of upper limb specimens.
More recently, this concept of an arcade, formed around the ulnar nerve in the posterior compartment of the arm, has been dismissed. For instance, in another cadaveric study, although the internal brachial ligament was found to support the attachment of the superficial-most fibres of the medial head of the triceps, no local thickening in the deep investing fascia of the arm was observed (Wehrli and Oberlin, 2005). In this study, the internal brachial ligament was found to be present in 73% of upper limb specimens. Overall, Wehrli and Oberlin concluded that their observations were, for the most part, in favour of the description afforded by Struthers. However, unlike Struthers, they suggested that the internal brachial ligament is derived from the medial intermuscular septum. More specifically, Wehrli and Oberlin defined the ligament as the dissociated part of the medial intermuscular septum, in the region where the ulnar nerve penetrated the septum, to enter the posterior compartment. In support of this definition, in the upper limb specimens in which the ulnar nerve did not penetrate the medial intermuscular septum, and was located already within the posterior compartment of the arm, the internal brachial ligament was absent (27%). According to their observations, the internal brachial ligament separated and then re-joined the medial intermuscular septum, on average, 11.5 (± 1.4) cm and 8.2 (± 1.9) cm proximal to the medial epicondyle, respectively.
Figure 2: A photograph (left), taken during cadaveric dissection of the arm, and a representative schematic diagram (right) showing the ulnar nerve penetrating the medial intermuscular septum. The internal brachial ligament is the dissociated part of the medial intermuscular septum, in this region.
2. Medial epicondyle
In the posterior compartment of the arm, the ulnar nerve descends along the medial head of the triceps and towards the medial epicondyle. Delayed onset ulnar neuropathy, due to compression by the medial epicondyle, in individuals (usually children) who develop cubitus varus or cubitus valgus, is commonly referred to as “tardy ulnar nerve palsy”. In these individuals, cubitus varus or cubitus valgus usually develops following trauma, such as (malunion of) a supracondylar fracture of the humerus, which is most common in children (Uchida and Sugioka, 1990). Children may also develop cubitus valgus due to previous epiphyseal injury to the lateral condyle of the humerus (Toh et al., 2002).
3. Cubital tunnel
The ulnar nerve enters the forearm via the cubital tunnel, which guides the nerve between the humeral and ulnar heads of the flexor carpi ulnaris. This is the most common site of compression of the ulnar nerve. The floor of the cubital tunnel is formed by the epicondylar groove. This groove is bound anteriorly by the medial epicondyle and laterally by the olecranon process of the ulna and, the fibrous capsule and the medial collateral ligament of the elbow (Posner, 1998). The roof of the cubital tunnel is usually formed proximally by a ligament which spans the width of the groove (approximately 4 mm) (O’Driscoll et al., 1991). Although it is distally continuous with the deep layer of the aponeurosis between the two heads of the flexor carpi ulnaris, these two structures are anatomically and functionally distinct from each other (Wachsmuth and Wilhelm, 1968). For instance, the fibres of the ligament are aligned perpendicular to those of the aponeurosis and the specific function of this ligament is to retain the ulnar nerve in the epicondylar groove. For this reason, O’Driscoll et al. referred to it as the cubital tunnel retinaculum (CTR). (The term retinaculum is derived from the Old Latin word “retinēre”, which means to retain).
In 1957, Osborne first described the CTR as the primary cause of ulnar nerve compression in the cubital tunnel (Osborne et al., 1957). Consequently, it is also commonly referred to as Osborne’s band. However, this was not universally accepted. Other potential causes included the aponeurosis (Adelaar et al., 1984) and an aberrant muscle, called anconeus epitrochlearis (Masear et al., 1988). Overall, the literature regarding the anatomy of the cubital tunnel, and its relationship with ulnar neuropathy, was highly discrepant. Then, in a landmark study, O’Driscoll et al. identified four clinically relevant variants of the cubital tunnel, based on their observations during cadaveric dissection. This classification distinguished variants according to the nature of the roof of the cubital tunnel.
Figure 3: A series of schematic diagrams. The image furthest on the left shows the ulnar nerve entering the cubital tunnel (left). Note that the roof of this tunnel is formed proximally by the CTR, which is distally continuous with the deep layer of the aponeurosis between the two heads of the flexor carpi ulnaris. The images on the right show O’Driscoll’s classification of the anatomical variants of the cubital tunnel.
Figure 4: A photograph (left), taken during cadaveric dissection of the arm, and a representative schematic diagram (right)) showing the type 0 variant. In the absence of the CTR, the ulnar nerve is at risk of dislocation, particularly during flexion of the elbow.
Type 0 implied the complete absence of the CTR, which was observed in 1 out of 27 (3.7%) upper limb specimens. In this case, the ulnar nerve would be at risk of subluxation or dislocation, particularly with flexion of the elbow, which may lead to neuropathy caused by damage due to friction or increased susceptibility to inadvertent trauma. This is consistent with reports of ulnar nerve dislocation in 7 out of 8 patients after surgical decompression of the nerve in the cubital tunnel by incision of the roof (Chalmers, 1978). The estimated prevalence of ulnar nerve dislocation in the general population is 16.2% (Childress, 1975). Other causes of ulnar nerve dislocation include congenital or traumatic bone abnormalities, particularly of the medial epicondyle (Posner, 1998).
Type I implied the presence of a fibrous retinaculum, which was observed in all but 4 upper limb specimens (85.2%). This category was sub-divided into type IA and type IB. Type IA implied the presence of a “normal” fibrous retinaculum, which was not associated with the development of ulnar neuropathy. Indeed, this was the most commonly observed of all anatomical variants, present in 17 upper limb specimens (63.0%). In contrast, type IB implied the presence of an abnormal fibrous retinaculum, which was associated with the development of ulnar neuropathy and was observed in 6 upper limb specimens (22.2%).
Flexion of the elbow causes the separation between the medial epicondyle and the olecranon process, i.e. the attachments of the CTR, to increase by 5 mm for every 45° of flexion. The resultant increase in tension within the CTR (Froimson and Zahrawi, 1980) causes the capacity of the cubital tunnel to decrease (Apfelberg and Larson, 1973) by up to 55% (Watchmaker et al., 1994). This ultimately leads to dynamic compression of the ulnar nerve. In fact, the mean intraneural pressure of the ulnar nerve increases from 7 mmHg to up to 24 mmHg, during flexion of the elbow (Pechan and Julis, 1975). Interestingly, the mean intraneural pressure is significantly greater than the mean extraneural pressure with no less than 90° of flexion at the elbow (Yamaguchi et al., 1999). This is accounted for by increased traction of the ulnar nerve during flexion.
O’Driscoll et al. found that the type IA retinaculum was slack in extension and taut only at full flexion of the elbow. In comparison, the type IB retinaculum, which was noticeably thicker, was taut before full flexion of the elbow had been achieved, at 90° to 120° of flexion. Therefore, the type IB retinaculum is more likely to cause dynamic compression of the ulnar nerve. Indeed, O’Driscoll et al. found evidence of chronic nerve compression on observation of upper limb specimens with a type IB retinaculum.
Figure 5: Photographs (left), taken during cadaveric dissection of the arm, and representative schematic diagrams (right) showing the type IB, during extension (above) and flexion (below) of the elbow. Note that, with flexion of the elbow, the retinaculum becomes taut and the capacity of the cubital tunnel decreases. This leads to dynamic compression of the ulnar nerve.
Note that the medial collateral ligament, in the floor of the cubital tunnel, may also contribute to dynamic compression of the ulnar nerve. Some authors have suggested that flexion of the elbow is associated with a decrease in tension of the medial collateral ligament, which causes the ligament to bulge into the tunnel (Froimson and Zahrawi, 1980, Wadsworth, 1977). On the contrary, it has been shown that both the anterior and posterior aspects of the medial collateral ligament become taut with flexion of the elbow (Morrey and An, 1985). These findings agree with the observations made by O’Driscoll et al., who instead noticed that the groove on the medial epicondyle is shallower inferiorly than it is posteriorly. In effect, this means that the floor of the cubital tunnel “rises” with flexion of the elbow.
Type II implied the replacement of the CTR by the aforementioned anconeus epitrochlearis, which was observed in only 3 upper limb specimens (11.1%), though other studies have shown that it may be present in up to 28% of cadavers (Dellon, 1986, Wadsworth, 1977). It is hypothesised that the CTR is a vestigial remnant of this atavistic muscle (Testut, 1923), given that these two structures are mutually exclusive and anatomically indistinct. Interestingly, although the anconeus epitrochlearis is innervated by the ulnar nerve, it appears to contribute to extension of the elbow and therefore may be an accessory slip to the medial head of the triceps (O’Driscoll et al., 1991). This muscle, when present, often causes static compression (i.e. independent of movement at the elbow joint) of the ulnar nerve (Hirasawa et al., 1979). In fact, MacNicol reported it as a cause of ulnar neuropathy in 9% of individuals undergoing surgical decompression (MacNicol, 1982).
Other causes of extrinsic compression
As described above, extrinsic compression of the ulnar nerve in the cubital tunnel may be caused by certain anatomical variants of the CTR, i.e. a thickened CTR or the anconeus epitrochlearis muscle. Other causes of extrinsic compression include trauma and sustained pressure on the medial aspect of a flexed elbow. This may be observed in patients who are bedbound (Warner et al., 2000) or due to improper positioning of the arm in patients undergoing surgery with general anaesthesia (Kroll et al., 1990).
Intrinsic compression of the ulnar nerve in the cubital tunnel refers to any lesion which occupies space in the epicondylar groove. This includes bone abnormalities such as hypertrophy, osteophytes or exostoses arising from either the medial epicondyle or the olecranon process, or the presence of fracture fragments; soft tissue abnormalities such as ganglion cysts or synovitis, due to rheumatoid arthritis; and haemorrhage, due to trauma (Posner, 1998).
4. Flexor-pronator aponeurosis
On entering the cubital tunnel, the ulnar nerve passes between the humeral and ulnar heads of the flexor carpi ulnaris. It abandons this intramuscular course, approximately 2-3 cm distal to the cubital tunnel, by penetrating the “flexor-pronator aponeurosis” (Amadio and Beckenbaugh, 1986). This is the last site of compression of the ulnar nerve around the elbow.
Anatomy (distal to the elbow)
Beyond the flexor-pronator aponeurosis, the ulnar nerve remains in the anterior compartment of the forearm, descending between the flexor carpi ulnaris and the flexor digitorum profundus. Efferent branches, derived from the ulnar nerve shortly after entering the forearm, innervate these two closely related muscles. Note that the ulnar nerve innervates only the medial half of the flexor digitorum profundus.
The ulnar nerve then gives rise to two afferent branches. The dorsal branch, which arises at about the midpoint of the forearm, supplies sensory innervation to the dorsal aspect of the 5th digit, up to the base of the distal phalanx, and the dorsal aspect of the medial half of the 4th digit, up to the base of the intermediate phalanx. The palmar branch arises approximately 5 cm proximal to the wrist. It supplies sensory innervation to the medial half of the palm of the hand.
In the distal half of the forearm, the ulnar nerve becomes superficial. Then, at the level of the distal palmar crease, it passes through Guyon’s canal to enter the hand. This is a fibro-osseous tunnel, approximately 4 cm in length, which begins at the proximal aspect of the pisiform bone and extends up to the hook of hamate. Note that this is the site of compression of the ulnar nerve at the wrist.
In the hand, the ulnar nerve divides into two terminal branches. The superficial terminal branch contains predominantly afferent fibres. It supplies sensory innervation to the palmar aspect of the 5th digit, the palmar aspect of the medial half of the 4th digit and the distal and dorsal aspects of these digits, thus complementing the sensory innervation supplied by the dorsal branch. The efferent fibres of the superficial terminal branch supply motor innervation to the palmaris brevis.
In contrast, the deep terminal branch contains efferent fibres only. It supplies motor innervation to most of the intrinsic muscles of the hand. This includes the muscles of the hypothenar eminence, namely the abductor digiti minimi, the flexor digiti minimi and the opponens digiti minimi; the interossei; the third and fourth lumbricals; the adductor pollicis and the deep head of the flexor pollicis brevis.
Compression or entrapment is a common cause of peripheral nerve injury. The extent of damage caused to the different tissue components of the peripheral nerve is a determinant of the severity of the injury. As such, two widely accepted classification systems, proposed independently by Seddon and Sunderland, use the histological appearance of the damaged nerve segment to categorise peripheral nerve injuries. Seddon classified peripheral nerve injuries into three categories (Seddon, 1943). In order of increasing severity, these were neuropraxia, axonotmesis and neurotmesis. With neuropraxia, there is only focal demyelination. The axons and the stroma are preserved. Neuropraxia is associated with a transient loss of motor function and impaired sensory (and autonomic) function. With axonotmesis, there is also complete disruption of the axons. It is associated with Wallerian degeneration and a complete loss of function, distal to the site of injury. However, the stroma remains intact. Therefore, axon regeneration is possible and potentially allows complete recovery of function. Finally, with neurotmesis, there is complete disruption of the nerve segment, including all stromal elements. In this case, without surgical intervention, there is a permanent loss of function.
In contrast, Sunderland classified peripheral nerve injuries into five categories (Sunderland, 1951). This classification resembles Seddon’s classification in that first and second-degree injuries are analogous to neuropraxia and axonotmesis, respectively. However, Sunderland defined two intermediate stages, between axonotmesis and neurotmesis, based on the extent of damage to the stroma. Therefore, in terms of the severity of the injury, neurotmesis remains at the end of spectrum in Sunderland’s classification.
Classification systems of injuries specifically affecting the ulnar nerve also exist. For example, McGowan proposed a clinically relevant classification which identified three categories, based on the type and extent of presenting features (McGowan, 1950). Grade I suggests a mild injury of the ulnar nerve, such as due to transient nerve compression. This causes paraesthesia of the medial half of the hand but produces no signs or symptoms indicative of a loss of motor function. Conversely, with grade II injuries, there is also weakness of the intrinsic muscles of the hand that are innervated by the ulnar nerve. Finally, grade III suggests a severe injury of the ulnar nerve, such as due to prolonged nerve compression. In this case, there is marked weakness of the hand due to paralysis and atrophy of the interosseous muscles. Note that there is an obvious correlation between Seddon’s classification, which describes the histopathological features of a peripheral nerve injury, and McGowan’s classification, which specifically describes the signs and symptoms of an ulnar nerve injury.
As suggested by McGowan’s classification, compression of the ulnar nerve at the elbow initially causes paraesthesia of the medial half of the hand. In theory, given the sensory innervation of the ulnar nerve, ulnar neuropathy should only cause paraesthesia of the medial one and half digits. However, in reality, patients may also report paraesthesia in the other digits. Anatomical variations accounting for these anomalies include Berrettini’s anastomosis, which is a communication between the median and ulnar nerves, containing only afferent fibres. Surprisingly, cadaveric studies suggest that Berrettini’s anastomosis may have a prevalence of up to 94% (Don Griot et al., 2000).
Chronic compression of ulnar nerve at the elbow also causes weakness in the hand, due to paralysis and atrophy of the intrinsic muscles that are innervated by the ulnar nerve. For example, patients may classically complain about the 5th digit becoming caught on the edge of the pocket. This is caused by resting abduction of 5th digit, called the Wartenberg sign, due to weakness of the interosseous muscles. Note that patients may also present with uncharacteristic motor symptoms due communicating efferent branches between the median and ulnar nerves. These include Martin-Grüber and Marinacci’s anastomoses in the forearm and Riche-Cannieu anastomosis in the hand (Sirasanagandla et al., 2013).
In conclusion, the most common site of compression of the ulnar nerve at the elbow is within the cubital tunnel. For this reason, this disorder is more commonly known as cubital tunnel syndrome. The aetiology of ulnar nerve compression within the cubital tunnel may be considered in terms of intrinsic and extrinsic causes. Intrinsic causes implicate space-occupying lesions derived from the related anatomical structures. More importantly, anatomical variation of the cubital tunnel retinaculum underlies the extrinsic causes of ulnar nerve compression. A thickened retinaculum may cause dynamic compression of the ulnar nerve during flexion of the elbow. In contrast, static compression of the ulnar nerve is caused by replacement of the retinaculum by the atavistic anconeus epitrochlearis muscle.
Adelaar R, Foster W and McDowell C (1984) The Treatment of the Cubital Tunnel-Syndrome. Journal of Hand Surgery-American Volume. 9, 90-95.
Amadio PC and Beckenbaugh RD (1986) Entrapment of the ulnar nerve by the deep flexor-pronator aponeurosis. Journal of Hand Surgery – American Volume. 11, 83-87.
Andreisek G, Crook DW, Burg D, Marincek B and Weishaupt D (2006) Peripheral neuropathies of the median, radial, and ulnar nerves: MR imaging features. Radiographics. 26, 1267-1287.
Apfelberg DB and Larson SJ (1973) Dynamic anatomy of the ulnar nerve at the elbow. Plastic & Reconstructive Surgery. 51, 79-81.
Chalmers J (1978) Unusual causes of peripheral nerve compression. Hand. 10, 168-175.
Childress HM (1975) Recurrent ulnar-nerve dislocation at the elbow. Clinical Orthopaedics & Related Research. 108, 168-173.
Dellon AL (1986) Musculotendinous variations about the medial humeral epicondyle. Journal of Hand Surgery – British Volume. 11, 175-181.
Don Griot JP, Zuidam JM, van Kooten EO, Prose LP and Hage JJ (2000) Anatomic study of the ramus communicans between the ulnar and median nerves. Journal of Hand Surgery – American Volume. 25, 948-954.
Froimson AI and Zahrawi F (1980) Treatment of compression neuropathy of the ulnar nerve at the elbow by epicondylectomy and neurolysis. Journal of Hand Surgery – American Volume. 5, 391-395.
Hirasawa Y, Sawamura H and Sakakida K (1979) Entrapment neuropathy due to bilateral epitrochleoanconeus muscles: a case report. Journal of Hand Surgery – American Volume. 4, 181-184.
Kane E, Kaplan EB and Spinner M (1973) [Observations of the course of the ulnar nerve in the arm]. Ann.Chir. 27, 487-496.
Kroll DA, Caplan RA, Posner K, Ward RJ and Cheney FW (1990) Nerve injury associated with anesthesia. Anesthesiology. 73, 202-207.
Latinovic R, Gulliford MC and Hughes RAC (2006) Incidence of common compressive neuropathies in primary care. Journal of Neurology, Neurosurgery & Psychiatry. 77, 263-265.
MacNicol MF (1982) Extraneural pressures affecting the ulnar nerve at the elbow. Hand. 14, 5-11.
Masear VR, Hill JJJ and Cohen SM (1988) Ulnar compression neuropathy secondary to the anconeus epitrochlearis muscle. Journal of Hand Surgery – American Volume. 13, 720-724.
McGowan A (1950) The Results of Transposition of the Ulnar Nerve for Traumatic Ulnar Neuritis. Journal of Bone and Joint Surgery-British Volume. 32, 293-301.
Morrey BF and An KN (1985) Functional anatomy of the ligaments of the elbow. Clinical Orthopaedics & Related Research. 201, 84-90.
O’Driscoll SW, Horii E, Carmichael SW and Morrey BF (1991) The cubital tunnel and ulnar neuropathy. Journal of Bone & Joint Surgery – British Volume. 73, 613-617.
Osborne G, Parkes A, Apley A, Nissen K, Seddon H and Osborne G (1957) The Surgical Treatment of Tardy Ulnar Neuritis. Journal of Bone and Joint Surgery-British Volume. 39, 782-782.
Pechan J and Julis I (1975) The pressure measurement in the ulnar nerve. A contribution to the pathophysiology of the cubital tunnel syndrome. J.Biomech. 8, 75-79.
Posner MA (1998) Compressive ulnar neuropathies at the elbow: I. Etiology and diagnosis. J.Am.Acad.Orthop.Surg. 6, 282-288.
Seddon H (1943) Three types of nerve injury. Brain. 66, 237-288.
Sirasanagandla SR, Patil J, Potu BK, Nayak BS, Shetty SD and Bhat KMR (2013) A rare anatomical variation of the Berrettini anastomosis and third common palmar digital branch of the median nerve. Anatomical Science International. 88, 163-166.
Struthers J (1854) On some points in the abnormal anatomy of the arm. British and Foreign Medico-Chirurgie Revue. 14, 170-179.
Sunderland S (1951) A Classification of Peripheral Nerve Injuries Producing Loss of Function. Brain. 74, 491-516.
Testut L (1923) Anatomia Humana. 3, 314.
Toh S, Tsubo K, Nishikawa S, Inoue S, Nakamura R and Harata S (2002) Long-standing nonunion of fractures of the lateral humeral condyle. Journal of Bone & Joint Surgery – American Volume. 84-A, 593-598.
Uchida Y and Sugioka Y (1990) Ulnar nerve palsy after supracondylar humerus fracture. Acta Orthop.Scand. 61, 118-119.
Wachsmuth W and Wilhelm A (1968) [The musculus epitrochleoanconaeus and its clinical significance]. Monatsschr.Unfallheilkd.Versicher.Versorg.Verkehrsmed. 71, 1-22.
Wadsworth TG (1977) The external compression syndrome of the ulnar nerve at the cubital tunnel. Clinical Orthopaedics & Related Research. 124, 189-204.
Warner MA, Warner DO, Harper CM, Schroeder DR and Maxson PM (2000) Ulnar neuropathy in medical patients. Anesthesiology. 92, 613-615.
Watchmaker GP, Lee G and Mackinnon SE (1994) Intraneural topography of the ulnar nerve in the cubital tunnel facilitates anterior transposition. Journal of Hand Surgery – American Volume. 19, 915-922.
Wehrli L and Oberlin C (2005) The internal brachial ligament versus the arcade of Struthers: an anatomical study. Plastic & Reconstructive Surgery. 115, 471-477.
Yamaguchi K, Sweet FA, Bindra R and Gelberman RH (1999) The extraneural and intraneural arterial anatomy of the ulnar nerve at the elbow. Journal of Shoulder & Elbow Surgery. 8, 17-21.