Shoulder function after medial approach and derotational humeral osteotomy in patients with brachial plexus birth palsy.

BACKGROUND

The purpose of this study was to assess outcome after rotational humeral osteotomies, using a medial approach, in children with brachial plexus birth palsy.

METHODS

A retrospective review of children with brachial plexus birth palsy who underwent external rotational humeral osteotomies, using a medial approach, for the treatment of internal rotation contractures was performed. Presurgical and postsurgical range of motion, standard Mallet, and modified Mallet scores were recorded. The traditional Mallet score was modified to include a sixth subscale that further evaluated internal rotation. This was assessed by having the patients' attempts to place their palm flat on their naval.

RESULTS

Twenty-three children underwent external rotational humeral osteotomies. The mean rotational correction achieved during the procedure was 43.2+/-11.6 degrees (range: 20-70 degrees). The mean preoperative standard aggregate Mallet score was 13.8+/-2.8 and the mean postoperative score was 16.1+/-2.5 (P=0.002). When the additional internal rotation scale was added into the score, the mean preoperative aggregate score was 18.0+/-2.1 and the mean postoperative score was 19.5+/-2.8 (P=0.032). Further analysis revealed a statistically significant improvement (P<0.05) in external rotation, hand to neck, and hand to mouth functions. Internal rotation was decreased as represented by statistically significant decreases in hand to spine and hand to belly (P<0.05).

CONCLUSIONS

Rotational humeral osteotomies can be performed safely and effectively using a medial approach. These osteotomies significantly improve activities associated with external rotation. However, the degree of external rotation must be carefully balanced against the loss of internal rotation, which would impede midline function. The addition of a sixth subscale to the Mallet score that assesses hand to belly provides more clinically relevant information regarding midline function than hand to spine.

LEVEL OF EVIDENCE

This is a Level IV study.