**2. Treatment modalities**

The age of the patient, the mechanism of injury (blunt or penetrating), the location (proximal or distal in the upper limb), and the extent of the lesion will influence the type and timing of the treatment algorithm as well as the final result [46, 47]. When all brachial plexus roots are affected, particularly if avulsed, there will be very limited treatment options, and the end results will be a severe upper limb disability with a very limited chance of a useful functional recovery [48–50].

Particularly, it is important to find out if the lesion is pre- or postganglionic as the first one has no chance of spontaneous recovery [51]. Magnetic resonance imaging has proven very useful in this respect [52]. Waiting for spontaneous recovery will entail an inexcusable waste of time that will lead to an unsatisfactory recovery [50, 53]. Thus, once the diagnosis of the nerve root avulsion is confirmed, the repair will have to be done as soon as the patient is able to tolerate the surgical procedure needed to be done [54, 55]. The "urgent" repair, a few days after injury, has been reported by some in cases of confirmed avulsion and in clean nerve sections (i.e., glass) [56].

The treatment strategy is based on the mechanism of injury [54], the findings of the physical and neurological examinations [57], and the results of the complementary diagnostic tests (electrodiagnostic studies [58], magnetic resonance imaging [59], and ultrasonography [60]). This last one is relatively inexpensive and can be made available to places with very limited resources [61]. It can also been used intraoperatively to see the anatomy of the damaged nerves, helping to decide if the lesioned nerve segment has to be removed and the gap grafted or a neurolysis will solve the problem [62]. The evolution of their results overtime is particularly useful to locate the lesion(s), assess its severity, and control the response to the treatments (physiotherapy, observation, surgical repair, electrostimulation, etc.) [63]. Computerized myelo-tomography was used in the past to diagnose the nerve root avulsions, but nowadays it has been replaced by magnetic resonance imaging [59, 64–66].

Spontaneous recovery can be expected in most brachial plexus injuries [67], particularly in the case of obstetric patients [68]. Among them the rate of spontaneous recovery is particularly high (66–92%) [69]. Physical therapy is essential to correct muscle contractures and avoid neglect of the damaged limb while waiting for spontaneous recovery [69]. In the case of inadequate recovery, on-time surgical treatment might be indicated [3, 68].

Progressive improvement of the surgical techniques with direct nerve repair, nerve grafting, and particularly with nerve transfers has greatly improved the results in the brachial plexus injuries [47, 70–72]. Direct repair, when at all possible, is still the first choice, provided that there is no tension in the suture line [73]. Nerve grafts are required to cover the gaps, but the results are often not as good as expected [74, 75]. Meanwhile, the nerve transfers have expanded our treatment capabilities with excellent results [72, 76]. They are particularly useful in nerve injuries affecting the distal parts of the upper limb, as other techniques like the nerve repair, direct or with nerve grafts, yield poor results [47, 70, 77]. The growing axons coming through the nerve repair take so long to reach the hand intrinsic muscles that when they do it find them atrophied and fibrotic [78–81]. Meanwhile, the nerve transfers provide new axons close to the injured muscles with an early and efficient repair [72, 76]. At times an end-to-side nerve transfer can be added to keep the muscles viable, while the growing axons from the direct primary nerve repair to reach their final destination in the motor end plates [82]. Nerve transfers solve the problem of a long distance between the lesion site and the motor end plates to be reinnervated [6, 49, 72, 76]. They can also be used in case of delayed patient referral [83] or dense scar at the primary injury site [84]. Sensory nerve transfer is another very promising area [85, 86], particularly in tetraplegic patients [87, 88], and can also help to control the neuropathic pain [89].
