Brachial plexus injury in adults is a devastating injury that results in significant disability and pain in the involved extremity. The severity of function and disability depends on the number of nerve roots involved in the affected extremity. Approximately 53% of patients have a complete plexus lesion (C5–T1), 39% have upper plexus lesions including C5 and C6 with or without C7, and 6% have lower plexus lesions (C8–T1). Depending on the time of the presentation, the patient might be a candidate for nerve reconstruction. If the patient presents within the first 6 to 8 months from the time of the injury, they might benefit from nerve reconstruction; however, based on my experience, the likelihood of regaining shoulder function in patients with more severe injuries such as subtotal or total brachial plexus injury is less predictable regardless of the time of presentation. In our experience at Mayo Clinic, nerve reconstruction for patients with complete brachial plexus injury restored less than 20% of shoulder function. The standard nerve transfers to reconstruct shoulder function in complete brachial plexus injury includes the transfer of the spinal accessory to suprascapular nerve to restore the function of the posterior superior rotator cuff and additional nerve grafting using either sural nerve autograft or allograft to reconstruct postganglionic nerve injuries (if nerve roots are available) to restore additional shoulder and upper extremity function.

Patients with an upper trunk type of injury or C5, C6, and/or C7 injuries may be candidates for additional types of nerve transfer. Most of these patients may still have radial, median, and ulnar nerve function, and branches of these nerves can be transferred to restore deltoid and elbow motion.

Patients who do not respond to nerve reconstruction or patients who present too late to be a candidate for nerve reconstruction may be candidates for tendon transfers around the shoulder. Although shoulder fusion is an option, we mostly prefer to leave it as a salvage option for patients with complete brachial plexus injury or patients who strongly desire to have it performed. Shoulder fusion stabilizes the shoulder and relies on scapulothoracic motion for mobility of the arm. Stabilizing the glenohumeral joint can also improve the outcome of nerve and muscle transfers around the elbow.

The purpose of this editorial is to discuss the tendon transfer options in patients with brachial plexus injuries.

When managing patients with brachial plexus injury, my philosophy is always to try to partly reconstruct the rotator cuff function, followed by reconstruction of the deltoid function. The number of muscles available for transfer will dictate the number of transfers that may improve shoulder function.

Patients with upper trunk injuries have the largest number of donor muscles available for transfer. If the patient did not have a prior nerve transfer (spinal accessory nerve transfer to suprascapular nerve), then all the trapezius muscle is available for transfer. The deeper muscles to the trapezius, including levator scapulae, rhomboid minor, and rhomboid major, are rarely involved in brachial plexus injury and are rarely paralyzed because they receive significant innervation higher than the C5 nerve root.

Most of the pectoralis major as well as the latissimus dorsi have good strength and are amenable for transfer in upper trunk injuries. In addition, the serratus anterior muscle is preserved in upper trunk injuries and can be also preserved in more complete postganglionic nerve injuries.

The potential available muscles that we use for tendon transfers in patients with upper trunk injuries therefore include the upper, middle, and lower trapezius, the levator scapulae muscle, serratus anterior, pectoralis major, and latissimus dorsi.

When planning reconstruction of upper extremity function in patients with brachial plexus injury, the most important function to restore is elbow flexion, followed by shoulder and hand function. Restoration of elbow flexion is very essential to position the hand in the space. Nerve reconstruction with or without free muscle transfer has been shown to reliably improve elbow flexion in patients with brachial plexus injury. However, without shoulder function, specifically external rotation, the improvement of elbow flexion becomes less meaningful. The patient will have a tendency to rub the abdomen when trying to use his or her hand or put the hand on his or her face, because the patient cannot move the hand away from the midline if he or she lacks shoulder external rotation. In addition, patients who are overweight, or in women with larger breasts, the hand will be blocked by the abdomen or the breast from reaching the face.

For this reason, after restoration of elbow flexion, the most important shoulder function to restore in patients with brachial plexus injuries is shoulder external rotation. This is based on our observation and on asking brachial plexus patients, “What would be the single most important function we can improve to make you happier?” More than 90% would say shoulder external rotation. To restore shoulder external rotation in the setting of brachial plexus where the deltoid and rotator cuff musculature are paralyzed, the single option available that would be reliable to improve shoulder external rotation is lower trapezius transfer. Because the rotator cuff tendons are intact in most cases of brachial plexus injury, direct transfer of the lower trapezius to the infraspinatus is possible without the need for augmentation with tendon allograft or autograft.

Patients who had prior spinal accessory nerve transfer will not have ipsilateral lower trapezius to transfer for shoulder external rotation. In this situation, we perform a transfer of the contralateral lower trapezius. There is a major difference in the transfer between the ipsilateral and contralateral trapezius. With the ipsilateral trapezius, we transfer the insertion of the lower trapezius to the infraspinatus insertion. In patients who have no ipsilateral lower trapezius when we transfer the contralateral trapezius, we transfer the origin, not the insertion. This is performed by detaching the origin of the lower trapezius from the spine augmented with lumber fascia, while keeping its insertion intact on the medial spine of the scapula, and transferring to the contralateral infraspinatus tendon of the affected shoulder. We have done this transfer on more than 20 patients, with very good outcomes.

In patients who present with symptomatic inferior subluxation, the lower trapezius transfer may stabilize the shoulder; however, it may not be enough to reduce a fully inferiorly subluxated humeral head. In this case, the remaining upper and middle trapezius attached to a portion of the lateral acromion could be transferred to the proximal humerus to stabilize the humeral head and reduce it proximally. This also may improve or give the patient additional shoulder abduction in the range of 30- to 50-degree maximum and also approximately 30 to 60 degrees of flexion.

When we transfer the whole trapezius, including the lower trapezius for the infraspinatus, and upper and middle trapezius for proximal humerus, we also plan to transfer the levator scapulae to the lateral aspect of the scapula to stabilize the scapula. Because we have detached all the trapezius and have mobilized it laterally to attach it on different parts of the humerus, we feel it is important to add a muscle transfer to stabilize the scapula and to replace the function of the trapezius. Then our three standard muscle transfers include transfer of the lower trapezius to infraspinatus, transfer of upper and middle trapezius to the proximal humerus, and transfer of the levator scapulae to the lateral spine of the scapula.

In most patients, these three muscle transfers will provide stability of the scapula, have minimum improvement of shoulder flexion and abduction, and improve shoulder external rotation. In patients with elbow flexion, it will significantly improve the functionality of the upper extremity and positioning of the arm in front of the body.

To further improve shoulder flexion, especially in patients with upper trunk types of nerve injury, there are two additional transfers to reconstruct the anterior deltoid.

The first option is the transfer of the pectoralis major as a pedicle transfer to reconstruct the anterior deltoid. In this situation, we transfer either the upper two-thirds or the whole pectoralis major to reconstruct the anterior deltoid and potentially a portion of the anterior aspect of the middle deltoid. The advantage of transferring only the upper two-thirds is to preserve the lower third of the pectoralis major so that it can maintain the patient’s ability to adduct the shoulder, which is important in patients with brachial plexus injury, because patients rely on it for the pinch effect. Patients can hold an object between the arm and the chest, and they can squeeze on it to hold it. This function is lost in patients who undergo fusion.

On the other hand, the advantage of transferring the whole pectoralis is better power for shoulder flexion. Patients with a three-muscle transfer and the pedicle pectoralis muscle transfer to reconstruct the deltoid have a good opportunity to flex the shoulder to horizontal.

The second option is transfer of the pedicled latissimus to reconstruct the anterior deltoid. In our practice, we leave this transfer as a second option if the pectoralis is needed for another type of transfer. For instance, in patients who did not regain elbow flexion, direct pectoralis transfer can be performed to restore elbow flexion.

With these muscle transfers in place, there is the possibility of performing a reverse shoulder arthroplasty on the shoulder in the future, because now the shoulder is stable and has muscle transfers that allow shoulder external rotation, in addition to muscles that essentially mimic the anterior deltoid (pedicled pectoralis or latissimus) and even partly the middle deltoid (upper trapezius transfer to the proximal humerus).

As mentioned earlier, shoulder fusion remains a potential salvage option in patients with brachial plexus injury. For patients with complete brachial plexus injury, we offer patients either a three-muscle transfer or shoulder fusion. We try to explain to the patients the risks and benefits for both; in the United States, most patients almost uniformly refuse to have shoulder fusion early on because they really like the ability to move the shoulder passively.

Because shoulder fusion is irreversible, we give patients the option of a trial pinning of the shoulder. This gives them an approximate idea about how the shoulder fusion will feel and function. In this situation, under anesthesia, we position the shoulder in the fusion position and perform multiple pinnings of the shoulder while avoiding the central part of the glenohumeral joint. We let the patients experience how it would feel if the shoulder were fused. If the patient understands how it feels and likes it, then we perform the fusion. If they did not like it, we remove the pins and perform the three-muscle transfer. This is a very uncommon situation, but at the same time, it is an important option. If the patient is not aware of what the fusion will do to the shoulder and are disappointed after the procedure, it becomes almost impossible to reverse it, especially in the setting of brachial plexus injury.

Finally, advances continue in the field of upper extremity reconstruction in patients with brachial plexus injuries, specifically the use of myoelectric shoulder, elbow, and hand. We are hoping that these advances will combine with the different surgical reconstructions to improve patients’ outcomes.