Shoulder internal rotation contracture in brachial plexus birth injury: proximal or distal subscapularis release?

BACKGROUND

In cases of brachial plexus birth injury with internal rotation contracture of the shoulder, the subscapularis muscle may be released proximally, from the subscapular fossa, or distally, along with periarticular soft tissues arthroscopic to the glenohumeral joint. We hypothesized that the indication for each procedure would rely primarily on patients' bone remodeling potential and periarticular soft-tissue contractures, performing proximal releases in patients aged < 4 years and periarticular distal releases in older patients. The purpose of this study was to analyze the outcomes such a strategy could provide.

METHODS

All patients presenting with brachial plexus birth injury-related shoulder internal rotation contractures who underwent a subscapularis release were included; in addition, to restore the joint axial balance, the infraspinatus was systematically reanimated with a tendon transfer (ie, latissimus dorsi or lower trapezius) during the same operating time. Chart review yielded preoperative and postoperative clinical and radiographic measurements, including active and passive range of motion of the shoulder in external rotation (ER) with the arm at the side of the body, the modified Mallet score, glenoid version, and the percentage of the humeral head arthroscopic to the middle of the glenoid process.

RESULTS

Between July 2012 and January 2018, 28 children were operated on at our institution. In patients who underwent proximal subscapularis release (n = 13), significant improvements were observed regarding active shoulder ER, passive shoulder ER, the modified Mallet score, glenoid version, and the percentage of the humeral head arthroscopic to the middle of the glenoid process, averaging 58° ± 32° (P < .0001), 56° ± 20° (P < .0001), 9.7 ± 3.1 points (P = .0006), 15° ± 10° (P = .0034), and 24% ± 20% (P = .0113), respectively, after a mean follow-up period of 4 years. Following distal release procedures (n = 15), these improvements averaged 26° ± 29° (P = .0024), 27° ± 28° (P = .0011), 3.3 ± 4.1 points (P = .0049), 2° ± 17° (P = .4086), and 4% ± 18% (P = .215), respectively, after a mean follow-up period of 3 years.

CONCLUSION

When combined with axial rebalancing of the joint, the proximal release of the subscapularis muscle appears to be sufficient to provide satisfactory functional outcomes in patients with great bone remodeling potential and supple periarticular soft tissues. In older patients, a more comprehensive release of the glenohumeral joint's arthroscopic aspect seems to provide lower but still significant clinical improvements.