CORONARY REVASCULARIZATION STRATEGIES AND THE EFFECTS OF DIABETES COMPLICATIONS ON POOR HEALTH
Abstract
Objectives: This research assessed the effect of comorbid diabetes complications on short-term adverse outcomes: in-hospital mortality, postoperative stroke, postoperative renal failure and readmissions within 30 days of discharge after coronary artery bypass graft (CABG) or percutaneous coronary intervention (PCI) among patients age 45 and older who have diabetes. The analysis compared differences in outcomes for CABG and PCI and for off-pump and on pump CABG. The analysis also focused on assessing associations between structure and process factors and the study outcomes. Specifically for readmissions, the effect of discharge disposition was evaluated, including discharge to home without home health care (HHC), to home with HHC, and to a transitional care facility. Methods: In-hospital mortality, postoperative stroke, and postoperative renal failure were assessed using the 2007 Nationwide Inpatient Sample (NIS) database. Readmission was categorized into early (¡Ü10 days) and late (11 to 30 days), and assessed using the 2007 State Inpatient Databases (SIDs) for Arizona, California, and Florida. Analyses included chi-square, t-test, propensity adjusted multivariate logistic regression, multilevel regression, and Cox proportional hazard regression. Analyses using the NIS data accounted for the survey design and were weighted for national representation. Covariates included age, race/ethnicity, health insurance, median household income, gender, 30 comorbidities, illness acuity measure, procedure volume and hospital characteristics. Results: In adjusted analyses, patients with comorbid diabetes complications were more likely to have in hospital mortality with both CABG (Odds Ratio, OR 1.60; 95% Confidence Interval, CI 1.22-2.10) and PCI (OR 1.59, CI 1.27-1.99) than those without diabetes complications. Their odds of renal failure were higher with PCI than CABG (OR 4.27 vs. 2.25, p<0.05). Further, they were more likely to have postoperative stroke with off-pump CABG (OR 1.69, CI 1.10-2.60). For readmissions, the adjusted hazard of early (Hazard Ratio, HR 1.24, CI 1.08-1.42) and late (HR 1.27, CI 1.11-1.39) readmissions for those with comorbid diabetes complications and early (HR 4.16, CI 3.53-4.91) and late (HR 1.88, CI 1.69-2.10) readmissions for patients discharged to a transitional care facility were higher. Discharge to home with HHC (HR 1.24, CI 1.12-1.36) was associated only with late readmission. Discussion: Regardless of the revascularization strategy, patients with comorbid diabetes complications may have higher risks of in-hospital death after coronary revascularization. Effects of comorbid diabetes complications on poor outcomes should be considered when making clinical decisions about CABG and PCI for patients with diabetes. Comprehensive discharge planning may be needed to identify potential vulnerabilities, such as a predisposition to poor diabetes management for patients with comorbid diabetes, to reduce their risk for readmission. Further research should examine the association between termination of HHC services and late readmission for patients using HHC services after discharge.