Diabetes mellitus is associated with a shift in the temporal risk profile of inhospital death after percutaneous coronary intervention: An analysis of 25,223 patients over 20 years

Background Numerous studies have demonstrated that patients with diabetes have higher rates of restenosis, late myocardial infarction, and late death after percutaneous coronary interventions (PCI). However, it remains unclear. whether patients with diabetes mellitus also have an increased hazard for early death after either elective or urgent PCI. Methods Patients undergoing PCI at the Mid American Heart Institute between 1980 and 1999 were identified. The main end point was inhospital death. Patients were stratified both by diabetes status and whether they underwent elective or urgent PCI. Results There were 17,341 nondiabetic patients and 4308 patients with diabetes who underwent elective PCI. There were 2946 nondiabetic patients and 628 patients with diabetes who underwent urgent PCI. Multivariate analysis demonstrated that diabetes was associated with increased inhospital mortality rate after any PCI (odds ratio 1.4, 95% Cl 1.1-1.8, P = .003). The unadjusted inhospital mortality rates for the nondiabetic patients and patients with diabetes were 0.8% and 1.4%, respectively (P < .001), after elective PCI. The mortality rate was 6.9% for the nondiabetic patients and 12.7% for the patients with diabetes (P < .001) after urgent PCI. The inhospital mortality rates among diabetic patients appear to be decreasing over time among the elective cohort (elective PCI diabetes-time interaction, P = .007) but not in the urgent cohort (urgent PCI-diabetes-time interaction, P = .68). Conclusions There has been an improvement in the inhospital survival rate among patients with diabetes in the elective PCI cohort. This improved hospital survival has yet to be realized among patients with diabetes undergoing urgent PCI.