Predicting operative outcomes of total shoulder arthroplasty using the model for end-stage liver disease score

Abstract

The aim of this study was to assess the efficacy of the Model for End-Stage Liver Disease (MELD) score in predicting postoperative complications following total shoulder arthroplasty (TSA). The American College of Surgeons National Surgical Quality Improvement database was queried for all patients who underwent TSA between 2015 and 2019. The study population was subsequently classified into two categories: those with a MELD score ≥ 10 and those with a MELD score < 10. Atotal of 5265 patients undergoing TSA between 2015 and 2019 were included in this study. Among these, 4690 (89.1%) patients had a MELD score ≥ 10, while 575 (10.9%) patients had a MELD score < 10. Postoperative complications within 30 days of the TSA were collected. Multivariate logistic regression analysis was conducted to explore the correlation between a MELD score ≥ 10 and postoperative complications. The anchor based optimal cutoff was calculated by receiver operating characteristic analysis to determine the MELD score cutoff that most accurately predicts a specific complication. Youden's index (J) determined the optimal cutoff point calculation for the maximum sensitivity and specificity; these were deemed to be "acceptable" if the area under curve (AUC) was greater than 0.7 and "excellent" if greater than 0.8. Multivariate regression analysis found a MELD score ≥ 10 to be independently associated with higher rates of reoperation (OR, 2.08; P=.013), cardiac complications (OR, 3.37; P=.030), renal complications (OR, 7.72; P=.020), bleeding transfusions (OR, 3.23; P<.001), and nonhome discharge (OR, 1.75; P<.001). The receiver operating characteristic analysis showed that AUC for a MELD score cutoff of 7.61 as a predictor of renal complications was 0.87 (excellent) with sensitivity of 100.0% and specificity of 70.0%. AUC for a MELD score cutoff of 7.76 as a predictor of mortality was 0.76 (acceptable) with sensitivity of 81.8% and specificity of 71.0%. A MELD score ≥ 10 was correlated with high rates of reoperation, cardiac complications, renal complications, bleeding transfusions, and nonhome discharge following TSA. MELD score cutoffs of 7.61 and 7.76 were effective in predicting renal complications and mortality, respectively.