Does objective shoulder impairment explain patient-reported functional outcome? A study of proximal humerus fractures.

Abstract

Following a healed proximal humerus fracture, patient-reported outcome is predicted by objective physical examination findings. Range of motion and strength thresholds can correctly identify subjects with normal functioning shoulders. Subjects were age ≥55 years with a healed proximal humerus fracture. Impairment was measured with a focused physical examination of the shoulder; patient-reported outcome was assessed using 4 outcome questionnaires. Linear regression was used to determine how well variations in shoulder impairment explain differences in patient-reported outcomes. ROC curves were constructed to determine the physical examination thresholds that would correctly identify normally functioning shoulders. Thirty-one subjects with a mean age of 70 ± 8 years participated. The mean patient-reported outcome scores were: DASH 21 ± 19, ASES 82 ± 17, SST 8 ± 3, and Oxford 20 ± 8. The linear regression analysis suggests that shoulder impairment explains 38% of the Oxford, 50% of the DASH, 58% of the SST, and 70% of the ASES variance. Abduction strength was a significant predictor for all functional outcomes. Based on ROC analysis, forward flexion had the best discriminatory ability for identifying normally functioning shoulders (area under curve, 0.93). A threshold of 120° forward flexion correctly classified 90% of the shoulders. Range of motion and strength thresholds can identify subjects with normal shoulder function. A discordance between shoulder impairment and patient-reported outcome has been demonstrated and further work to identify patient, injury, or treatment factors to minimize this discrepancy is still needed.