To The Editor:
We read the article entitled "Arthroscopic Release and Latissimus
Dorsi Transfer for Shoulder Internal Rotation Contractures and Glenohumeral
Deformity Secondary to Brachial Plexus Birth Palsy" (2006;88:564-74), by
Pearl et al., with interest. We congratulate the authors for demonstrating the
potential for glenohumeral remodeling in children with brachial plexus birth
injuries. We appreciate their attempts to clarify the surgical indications for
tendon transfer as opposed to release of the internal rotation contracture.
However, we take issue with their belief that arthroscopic release adds
anything, and we believe that it may, in fact, be somewhat inadequate. The
authors state that "releasing the subscapularis from its origin failed
in one of five children." The authors do not clarify whether those
failures were in patients with posterior dislocation or subluxation or in
patients in whom the humeral head was centered. When the glenohumeral joint is
centered, we have never encountered such a failure. For children with
long-standing subluxation or dislocation, we have taken an individualized
approach. After performing our subscapular slide, we release tight structures
anteriorly by means of intramuscular lengthening of the pectoralis, partial
release of the coracobrachialis tendon, partial coracoidectomy, and/or release
of the coracohumeral ligament. Using this approach, we have never failed to
achieve equivalent full external rotation of the affected shoulder. We do not
immobilize the shoulder in full external rotation postoperatively for fear of
overstretching these structures and causing too much weakness and loss of
internal rotation power. We wonder whether the authors are immobilizing
shoulders in too much external rotation postoperatively.
Finally, we see no logic in performing releases through the arthroscope.
One would not release a heel cord contracture with ankle arthroscopy. In
addition to the risk to the axillary nerve (with one patient in the study
losing 40° of elevation), four patients "had severe functional loss
of internal rotation, and this prompted consideration of additional
intervention, such as an internal rotation osteotomy." Alain Gilbert has
abandoned anterior release at the insertion of the subscapularis for this very
reason1. We believe that anterior tenotomy of the subscapularis
renders too much functional loss of internal rotation.
Because five (15%) of thirty-three patients had a serious complication, we
believe that this approach must be reconsidered.