The impact of insulin dependence on short-term postoperative complications in diabetic patients undergoing total shoulder arthroplasty.
Academic Article
Overview
abstract
BACKGROUND: The number of total shoulder arthroplasty (TSA) procedures performed is steadily increasing, and it is important to characterize predictors of postoperative complications for risk assessment models. Whereas diabetes has been associated with increased morbidity after TSA, the impact of insulin dependence on the risk profile of diabetic patients has not been described. METHODS: The National Surgical Quality Improvement Program database from 2011-2014 was used to identify 5918 TSA cases. Patients were categorized as non-diabetes mellitus (non-DM), non-insulin-dependent diabetes mellitus (NIDDM), or insulin-dependent diabetes mellitus (IDDM). Thirty-day postoperative complication rates, length of stay (LOS), and readmission rates were compared across the diabetes groups. Multivariable logistic regression adjusting for demographics and comorbidity burden was performed to determine the independent association between insulin dependence and postoperative outcomes. RESULTS: In bivariate analysis, NIDDM and IDDM were associated with multiple postoperative complications, including stroke, sepsis, wound complications, blood transfusion, and extended LOS. With multivariable logistic regression, however, NIDDM patients did not have significantly greater odds of any study end point relative to non-DM patients. IDDM patients had significantly greater odds for having any postoperative complication (odds ratio [OR], 1.53), stroke (OR, 13.63), blood transfusion (OR, 1.67), and extended LOS (OR, 1.38). CONCLUSIONS: After adjustment for demographics and comorbidity burden, NIDDM patients had risk profiles similar to those of non-DM patients. IDDM was an independent predictor of multiple postoperative complications. Insulin dependence should be considered in the preoperative risk assessment of diabetic patients undergoing TSA.
