Correlation of Radiographic Muscle Cross-sectional Area with Glenohumeral Deformity in Children with Brachial Plexus Birth Palsy

Background: Muscle imbalance about the shoulder in children with persistent brachial plexus birth palsy is thought to contribute to glenohumeral joint deformity. We quantified cross-sectional areas of the internal and external rotator muscles in the shoulder by magnetic resonance imaging in patients with chronic brachial plexopathy and the correlation between these muscle cross-sectional area ratios and glenohumeral deformity. The purposes of this investigation were to evaluate differences in the ratios between affected and unaffected shoulders in the same individual and to assess whether an increased internal to external rotator muscle cross-sectional area correlated with greater glenohumeral deformity.

Methods: This cohort study consisted of magnetic resonance imaging of seventy-four patients with chronic neuropathic changes about the shoulder from brachial plexus birth palsy. There were at least nine patients with scans available for each of the five classified subtypes of glenohumeral deformity: type I (fifteen patients), type II (seventeen), type III (seventeen), type IV (sixteen), and type V (nine). Cross-sectional areas of the pectoralis major, teres minor-infraspinatus (external rotators), and subscapularis muscles were measured. The supraspinatus muscle cross-sectional area could not be reliably measured. The ratio of subscapularis to external rotators, the ratio of pectoralis major to external rotators, and the compound ratio of subscapularis and pectoralis major to external rotators were compared with the severity of the glenohumeral deformity. Passive range of motion, Mallet and Toronto clinical scores, and Narakas type were also compared with the severity of the glenohumeral deformity and the muscle cross-sectional area measurements.

Results: Muscle cross-sectional area ratios were significantly correlated with glenohumeral deformity type. The mean ratio of pectoralis major to external rotators for affected shoulders over all deformity types compared with that for unaffected shoulders was significantly increased by 30% (p < 0.001); the mean ratio for subscapularis and pectoralis major to external rotators, by 19% (p = 0.015), and the mean ratio for subscapularis to external rotators, by 10% (p = 0.008). There was a significant increase in the ratio of pectoralis major to external rotators in affected shoulders within each type of deformity. Analysis of variance indicated higher ratios of pectoralis major to external rotator muscle cross-sectional areas in more severe deformity types (p < 0.001). There were significant differences in external rotation measurements with the shoulder at 90 degrees of abduction only among glenohumeral deformity types I, II, and III (p < 0.05).

Conclusions: The degree of muscle imbalance between internal and external rotators about the shoulder is measurable by magnetic resonance imaging in children with persistent brachial plexopathy, and the imbalance correlates with the degree of glenohumeral deformity. Our results may provide useful information to guide the timing and the choice of operative intervention in these children.