Associations of preoperative patient mental health status, sociodemographic and clinical characteristics with baseline pain, function and satisfaction in patients undergoing revision rotator cuff repairs

This study aimed to investigate the associations of general patient and disease-specific factors with baseline patient-reported outcome measures (PROMs) in patients undergoing revision rotator cuff repair (RCR). Baseline characteristics were also compared to those of patients undergoing primary RCR. Patients undergoing revision RCR of recurrent superior-posterior rotator cuff tears were included. Preoperative characteristics in the cohort were compared to a primary RCR cohort with identical selection criteria. Multivariable modeling and Akaike's Information Criterion comparisons were used to investigate the associations and relative importance of 21 preoperative factors in explaining variations in baseline PENN Shoulder Score (PSS) and its subscores in the revision RCR cohort. A total of 305 revision RCRs were included. Compared to the primary RCR cohort, patients undergoing revision RCR were more likely to be White (91% vs. 84%, P = .001), have a lower Charlson Comorbidity Index (0 vs. 1, P < .001), lower Veterans Rand-12 Mental Component Scores (VR-12 MCS; 50 vs. 53, P = .005), and were more likely to have a full-thickness (90% vs. 76%, P < .001), large/massive (54% vs. 37%, P < .001) tear with complete long head biceps rupture (37% vs. 10%, P < .001), and significant glenohumeral cartilage changes (19% vs. 7%, P < .001). Eight variables were significantly associated with baseline PSS or its subscores in the revision cohort. Lower VR-12 MCS was associated with lower PSS total, pain and function, while female sex was associated with lower PSS total and function. Medicare and Medicaid insurance status and higher body mass index were associated with lower PSS function, non-White race with lower PSS pain, and lower education, lower Charlson Comorbidity Index, and absence of significant glenohumeral cartilage degeneration with lower PSS satisfaction. Baseline VR-12 MCS, insurance status and sex were the top 3 factors in the baseline PSS model. Notably, rotator cuff tear type and size were not associated with baseline PSS or its subscores in the revision cohort. Both general patient and disease-specific factors were associated with baseline PROMs in patients undergoing revision RCR, with mental health status showing the strongest association. Compared to primary RCR patients, the revision cohort were more commonly White, had lower mental health status and more severe rotator cuff pathology, though tear severity was not associated with baseline PROMs. Further studies are needed to investigate if factors associated with poor baseline PROMs predict poor postoperative PROMs following revision RCR.