Abstract
Zhu and colleagues are to be congratulated on presenting a substantial retrospective series of patients who underwent nerve grafting to repair injury to the radial nerve between the triceps branches and the origin of the posterior interosseous nerve with a minimum follow-up of 13 months. They aimed to define factors that influenced the outcome. Overall, 24 of 33 patients gained Medical Research Council (MRC) Grade 3 or 4 (M3+) wrist extension, 16 M3+ finger extension, but only 10 had M3+ thumb extension. Better recovery was associated with delay from injury to repair of less than 6 months, a defect length of less than 5 cm and use of three of more graft cables in the repair. The outcome was known in 33 patients out of 65 cases who underwent nerve grafting during the study period.
The primary outcome was strength of wrist, finger and thumb extension. A stringent method of assessing digital extension was applied with the wrist in full passive extension, which may not have been used in all previous studies of radial nerve recovery. However, muscle strength was recorded using MRC grading, which is quite a coarse assessment. Ideally a more holistic assessment of results of nerve repair should be used including patient-reported outcome measures (PROMs) for everyday function and pain (Quick and Brown, 2020).
During exploration of the radial nerve, intraoperative electrodiagnosis was used, including recording of compound muscle action potentials and somatosensory evoked potentials in response to stimulating the proximal nerve to confirm complete nerve injury. Recording of muscle action potentials may not be an effective method of excluding spontaneous recovery if axons are regenerating but have not yet reached muscle. Therefore presence or absence of a nerve action potential across the lesion is a more certain method of predicting the chance of spontaneous recovery (Kline and Hudson, 1995). The nerves were repaired with between two and five nerve graft cables, but it is unclear how the number of cables was decided.
Twenty-five of the 33 patients had sustained humeral fractures that had been treated with plate fixation. The type of trauma sustained, high or low energy, which may influence the outcomes is not included. In ten cases the nerve was under the plate and in three it was entrapped in the fracture site, which is unfortunate since these are likely to have been avoidable situations. The nerve was ruptured in 12 cases. Six had sustained knife lacerations, where direct suture might have been possible had primary repair been performed, and two had spontaneous hourglass lesions of the radial nerve. Overall, the group of nerve injuries is quite heterogeneous.
While analysis was carried out in respect of the length of the nerve gap, this parameter is also likely to have been related to the mechanism of nerve injury, with rupture being caused by higher energy trauma and resulting in more extensive damage to the nerve extending beyond that visible on macroscopic examination. In contrast, knife lacerations and even nerve entrapment will cause more localized injury with a shorter nerve gap. In a series of repairs reported by Shergill et al. (2001), the violence of the injury had an important influence on outcome.
Patients with nerve gaps of less than 5 cm and less than 6 months delay between injury and nerve repair had significantly better recovery, and those with gaps greater than 5 cm and delay greater than 6 months having only a 20% chance of regaining finger and thumb extension. These findings provide useful support for the trend of results from other studies that repair of longer gaps and longer delay lowers the chance of good recovery. The lower chance of recovery after nerve grafting, particularly for digital extension, should help to guide management of radial nerve injuries.
In response to their findings the authors recommend that exploration should not be delayed beyond 3 months for closed radial nerve injuries in order to increase the chance of restoration of digital extension if repair is required. However, good spontaneous recovery of the radial nerve can occur later than 3 months. If managing a humeral shaft fracture with a radial nerve injury non-operatively then additional factors, including progress of a Tinel’s sign, any evidence of early reinnervation on neurophysiology tests and progress of fracture healing, should be considered when making a decision for operation. If primary operative fixation of the fracture is carried out, as in the current series, then the nerve can be explored at the same time and repaired if found to be ruptured.
The authors also discuss the place of tendon or nerve transfer rather than nerve grafting to restore radial nerve function. If the chance of a good result from nerve grafting is low, then tendon transfer is a reliable option providing early improvement in function. However, in my experience full finger and particularly thumb extension is often not achieved, this also being observed by Bertelli (2020). Concomitant injury to the forearm is not uncommon with humeral fractures and may preclude tendon and nerve transfer. The requirements of the patient should be taken into account. Some may need early restoration of function to return to work, while others may prefer to wait for recovery and undergo additional tendon transfer later if the outcome of nerve grafting is insufficient. Nerve transfer is an option, particularly if a long gap in the radial nerve is found on exploration, with Bertelli (2020) reporting better recovery than after tendon transfer. However, results are also likely to deteriorate with long delay from injury and use of median nerve branches for transfer will compromise the possibility of later tendon transfer if the nerve transfer is unsuccessful.