Coracoid Abnormalities and Their Relationship with Glenohumeral Deformities in Children with Obstetric Brachial Plexus Injury

BACKGROUND Patients with incomplete recovery from obstetric brachial plexus injury (OBPI) usually develop secondary muscle imbalances and bone deformities at the shoulder joint. Considerable efforts have been made to characterize and correct the glenohumeral deformities, and relatively less emphasis has been placed on the more subtle ones, such as those of the coracoid process. The purpose of this retrospective study is to determine the relationship between coracoid abnormalities and glenohumeral deformities in OBPI patients. We hypothesize that coracoscapular angles and distances, as well as coracohumeral distances, diminish with increasing glenohumeral deformity, whereas coracoid overlap will increase. METHODS 39 patients (age range: 2-13 years, average: 4.7 years), with deformities secondary to OBPI were included in this study. Parameters for quantifying coracoid abnormalities (coracoscapular angle, coracoid overlap, coracohumeral distance, and coracoscapular distance) and shoulder deformities (posterior subluxation and glenoid retroversion) were measured on CT images from these patients before any surgical intervention. Paired Student t-tests and Pearson correlations were used to analyze different parameters. RESULTS Significant differences between affected and contralateral shoulders were found for all coracoid and shoulder deformity parameters. Percent of humeral head anterior to scapular line (PHHA), glenoid version, coracoscapular angles, and coracoscapular and coracohumeral distances were significantly lower for affected shoulders compared to contralateral ones. Coracoid overlap was significantly higher for affected sides compared to contralateral sides. Significant and positive correlations were found between coracoscapular distances and glenohumeral parameters (PHHA and version), as well as between coracoscapular angles and glenohumeral parameters, for affected shoulders. Moderate and positive correlations existed between coracoid overlap and glenohumeral parameters for affected shoulders. On the contrary, all correlations between the coracoid and glenohumeral parameters for contralateral shoulders were only moderate or relatively low. CONCLUSIONS These results indicate that the spatial orientation of the coracoid process differs significantly between affected and contralateral shoulders, and it is highly correlated with the glenohumeral deformity. With the progression of glenohumeral deformity, the coracoid process protrudes more caudally and follows the subluxation of the humeral head which may interfere with the success of repositioning the posteriorly subluxed humeral head anteriorly to articulate with the glenoid properly.