Patients with median nerve injuries can present acutely or in delayed fashion. Generally, they are open injuries with sensory deficits involving the thumb, index, long, and radial half of the ring fingers. Partial injuries are more common than complete injuries, and lacerations at the wrist are more common than at the elbow. Depending on the level of injury, motor deficits can involve the pronator teres, flexor carpi radialis, palmaris longus, flexor pollicis longus, and flexor digitorum profundus muscles to the index and long fingers. If the entire nerve has been injured proximally, thenar muscle weakness will be observed [11]. A Tinel’s sign at the site of nerve injury will develop, and distal propagation of the Tinel’s is a good marker for axon regeneration after repair.

When nerve continuity is uncertain, one can take an observatory approach to determine if there is either clinical or EMG evidence of reinnervation. The mechanism of injury (e.g., sharp vs. blunt) will help guide treatment. Axon regrowth from the proximal stump optimally occurs at 1 mm/day after about a 1-month delay [2]. Irreversible muscle atrophy occurs anywhere from 12 to 18 months. Schwann cells and endoneurial tubes remain viable for 18–24 months after injury. An advancing Tinel’s sign can help clarify if reinnervation is occurring. If deficits persist past 3 months or if there is no evidence for reinnervation clinically, authors agree that there has typically been axonal damage [12]. Any patients without evidence of clinical recovery should undergo surgical exploration by 6 months, and some surgeons advocate for even earlier intervention for more proximal injuries [2].

Electrodiagnostic studies may provide information to help guide when to proceed with surgical exploration. The optimal timing is still debated. Although loss of amplitude of compound muscle action potential and nerve action potential is complete by 11 days after injury [2], electrical studies performed before 3 weeks after injury can be unreliable. Within the first week, electrodiagnostic tests can be useful for localization and determining complete from incomplete injuries. At 1–2 weeks, they can help distinguish axonotmesis or neurotmesis from neurapraxia. At 3–4 weeks, after fibrillation potentials have had a chance to develop, this provides the most information from a single study. At 3–4 months, they may provide information regarding reinnervation [2].

There are times when the nerve is not entirely transected and there is no universal agreement on management of these lesions. Options include conservative management, nerve grafting, and repair of the lacerated fascicular groups only. Depending on the severity of the lesion, the deficit, and the aforementioned factors affecting recovery, treatment is individualized. Early end-to-end microsurgical repair of lacerated fascicular groups has been shown to result in good motor and sensory outcomes (BMRC 4 and 3.81, respectively) [13].

The principal surgical options for complete lacerations include neurolysis, primary end-to-end repair, nerve grafting, and nerve transfer [14]. Primary nerve repair involves direct end-to-end suture of separated nerve ends [15] and has the best prognosis [16] (see Fig. 18.2). The indications for primary repair include the ability to directly approximate nerve ends without undue tension in a nerve that has not sustained either a crush injury or mechanical disruption. If there is tension on the repair, ischemia will occur, leading to dysfunction. If nerves are stretched by 8–10%, blood flow is reduced by half [17]. A basic clinical tenet is that if one cannot coapt two nerve ends with a single 9-0 nylon suture, there is too much tension [18]. However, there is little clinical literature supporting this assertion.

Tissue approximation and alignment will be easier with earlier repair [19], with most surgeons preferring to operate before 2 weeks. Furthermore, with earlier repair, there is improved neuron survival [20] and decreased fibrosis of the distal stump [21] (see Fig. 18.3). The definition of secondary repair is end-to-end suture 2–3 weeks after injury. If a tension-free repair cannot be achieved, primary repair should be abandoned for another method.

Neurorrhaphy can be performed with sutures, fibrin glue , or nerve tubes. One must dissect to scar-free, healthy appearing tissue. The injured portion of the nerve must be removed to expose healthy nerve with a visible fascicular pattern. It is of paramount importance to align the proximal and distal stumps. One must utilize both surface landmarks and other indicators to properly align the nerve fascicles. Recommendations for alignment include visualizing fascicular patterns, using surface vessels as markers and any obliquity of the injury. Some authors have reported on using histologic acetylcholinesterase staining to identify motor axons and carbonic anhydrase staining to identify sensory fibers [22], but this adds a great deal of time to the operative procedure. Others have advocated “awake” electrical nerve stimulation to identify motor fibers [23]. Neither of these methods has resulted in significantly improved results.

Ends should be lined up such that the fascicular groups are gently touching. One may place sutures in the epineurium, which is less traumatic but may not adequately approximate the deep fascicles. A fascicular repair involves dissecting the epineurium and suturing the perineurium of each fascicle. Grouped fascicular repair minimizes nerve trauma and allows for improved alignment. There have been no studies documenting superiority of one technique over another [24].

Fibrin-based tissue glue is becoming more popular for coaptation of the nerve ends. Its advantages include efficiency, simplicity, minimal trauma, and creation of a barrier to invading scar tissue [25]. Furthermore, there is animal data suggesting fibrin glue can promote angiogenesis, stimulate chemotaxis and leukocytosis, enhance macrophage proliferation, and provide hemostasis [6, 26, 27]. Fibrin glue does not appear to be a barrier to regeneration [28], but may have inferior holding strength [29, 30].

In secondary reconstruction with retraction of nerve endings and a large gap, end-to-end neurorrhaphy is no longer an option. Nerve diameter, gap size, and quality of the injured nerves influence the decision-making process. When a tension-free neurorrhaphy is not feasible, nerve grafting is typically the first option. Nerve graft can be autogenous (e.g., sural) or allograft (e.g., cadaver). The gold standard for a long gap is autograft. A variety of options exist, although the most commonly utilized donor is the sural nerve. Advantages of using sural nerve include a lack of motor deficit, a fairly superficial dissection to harvest, and a relatively lengthy course of the nerve with limited branches. Disadvantages of using autograft include limited availability, sensory donor nerves instead of mixed nerves as options, the obligate loss of nerve function, scarring, and painful neuroma formation [31].

Dissection, scar removal, neuroma resection, and management of median nerve injuries with a gap rest on the knowledge of interfascicular relationships and nerve architecture. It is important to align the median nerve and place the autograft accordingly in as near anatomic position as possible. Authors have previously identified the internal topography of the median nerve [32, 33]. In the upper two thirds of the forearm, the motor branches to the extrinsic muscles lie about the periphery, on the radial and ulnar aspect [32]. The sensory branches to the hand, the thenar motor branches, and the palmar cutaneous branch are in the central and dorsal quadrant. In the distal third, the thenar motor, lumbrical and sensory components are segregated and can be isolated.

Processed nerve allografts provide decellularized and predegenerated human nerve tissues which maintain the microarchitecture including the epineurium, fascicles, endoneurial tubes, and microvasculature [34, 35]. Processed nerve allograft may provide a viable option for mixed nerves and has the advantage of avoiding donor morbidity and decreasing surgical time [34]. The thickness and length of the processed autograft are also variable, and they are available in a range of sizes. Furthermore, they can be easily obtained. It is unclear, however, if there is a limit in the efficacy based on the length of the gap.

Nerve substitutes can include vein grafts, synthetic nerve conduits, and Schwann cell-lined nerve conduits [35, 36]. Nerve conduits are also an option for small gaps or partial lacerations. Lundborg and Hansson originally presented the concept of nerve entubulation in 1980 [37]. Conduits available include those made of collagen, polyglycolic acid (PGA), and polycaprolactone. In 1997, Lundborg et al. [38] performed a prospective, randomized clinical study comparing conventional microsurgical repair of median and ulnar nerves to using silicone tubes, with gaps measuring 3–4 mm between nerve ends. They found no difference in sensory or motor function between the two groups. Lundborg et al. [39] demonstrated that the median nerve can regenerate across 5 mm gaps equivalently to direct repair. However, a recent report was published describing four cases of failed conduit-based major nerve reconstructions [40].

A combination approach can also be made. For example, there is one case report describing reconstruction of a 4 cm median nerve graft with a piece of autogenous median nerve placed in a bioabsorbable conduit [41]. At 2 years after surgery, the patient had 7 mm moving and static two-point discrimination to the thumb and had recovered palmar abduction and EMG evidence of reinnervation of the abductor pollicis brevis.

It is beyond the scope of this chapter, but branches of the radial or ulnar nerve have been transferred to median nerve branches in the forearm, hand, or even digits [42–44]. Nerve transfers utilize intact motor nerves with a minor function to reinnervate critical muscles. One can join the distal end of the cut normal nerve with the distal stump of the injured nerve. Another option is to perform neurorrhaphy of selected fascicles from a normal nerve to an injured nerve . End to side involves taking the end of a healthy donor nerve to the side of a target nerve distal to the site of injury [45].

There is a lack of consensus on postoperative immobilization and rehabilitation after median nerve repair or grafting . The rehabilitation decision is generally patient and surgeon dependent. Tactile gnosis is generally not regained in adult patients after injury to a major nerve trunk [24, 46]. Sensory reeducation has been proposed and combines techniques to help patients with sensory impairment to learn to interpret the altered neural impulses by attempting to reprogram the brain [46, 47]. In 40 patients with low median nerve complete transection and repair, 20 were rehabilitated with a sensory reeducation program, and 20 had no further treatment than the initial therapy. In the first group, locognosia (ability to localize touch) was significantly improved compared to group B, but static and moving two-point discrimination was not different [48]. In a systematic review evaluating the effects of sensory reeducation programs on functional hand sensibility after median and ulnar nerve repair, there was limited evidence to support the use of early or late sensory reeducation programs [49].

The best-known scale for sensibility and motor grading is the British Medical Research Council (MRC) scale , which is the most widely accepted classification system to score outcome of peripheral nerve injuries [50–52]. Functional outcomes have been assessed successfully using the DASH including the functional symptom score, with strong relation found with motor and sensory recovery [53]. DASH score , Rosen score , and Highet score were found to correlate significantly when evaluating outcomes of median and ulnar nerve injuries [54]. Studies have not universally utilized the same outcome assessments, making comparisons difficult.

There are multiple factors that influence recovery, such as cooperative and motivation of the patient, hand therapy, cognitive capacity, psychological stress due to the trauma, and comorbidities such as diabetes and alcoholism [55]. Age, gap length, and delay to surgery greatly influence outcome after repair of median and ulnar nerve transection injuries [55, 56]. Time of improvement can be variable as well, with one study indicating that grip and tip-pinch strength improve over a period of 3 years following median or ulnar nerve lesions [57]. Return to work after isolated median nerve injuries is influenced by level of education, type of job, and compliance with hand therapy, with 80% of workers returning to work within 1 year [58].

Primary repair produces superior results compared to those of delayed repair [59, 60]. In a series of 2181 acute nerve injuries, a primary repair was achieved in 87% of the cases with end-to-end approximation [18]. In median nerve injuries at the wrist, protective sensation in the fingertips can be reliably restored by direct suture or nerve grafting [55]. However, more proximal injuries have had less sensory recovery due to the long distance between the site of injury and the target cutaneous receptors in the fingertips.

In a long-term outcome study, 71 median and ulnar lesions were assessed 8 years after microsurgical repair and were classified according to the DASH, the Rosen’s hand protocol, and the Highet scale. Patients regained approximately 70% of their original hand function [54]. Satisfactory motor (M4/M5) and sensory (S3+/4) recovery occurred in more than 50% of patients if the delay in repair was less than 3 months and the gap was less than 6 cm [56]. In 28 patients undergoing primary repair of a sharp transection of the median nerve at the wrist, S4 was elicited in 36%, S3+ in 29%, and S3 in 14%. This study demonstrated a significant correlation between age and functional sensibility [61].