Abstract 17562: Poorer Survival with Insulin Therapy at Discharge in Diabetic Patients with Acute Myocardial Infarction: The FAST-MI Registry
The role of insulin therapy after AMI is debated. We assessed 3-year survival in survivors of AMI with diabetes, according to the prescription of insulin at discharge. All patients were included in the FAST-MI registry, a French nationwide registry of patients with AMI hospitalized in 2005 in 223 institutions.
Of the 1228 diabetic patients, 467 (38%) had insulin at discharge. These were older (70±12 vs 69±12 years, p=0.004), had a higher GRACE risk score (156 ± 33 vs 147 ± 35, p<0.001), higher admission glycemia (232 ± 110 vs 190 ± 85 mg/dL, p<0.001), higher HbA1c (8.1 ± 1.6 vs 7.2 ± 1.4, p<0.001) and longer duration of diabetes (diabetes Hx >10 years: 35% vs 16.5%, p<0.001). Type of MI did not differ (STEMI: 39% vs 44%, p=0.11). Fewer patients in the insulin group had coronary angiography (78% vs 88%, p<0.001) or PCI (53% vs 67%, p<0.001) during the hospital stay. Optimal medical therapy at discharge was prescribed in 51% vs 47% of the patients, respectively (P=0.12). Finally, other antidiabetic medications at discharge were less frequent in the insulin group (metformin: 8% vs 25%, p<0.001; sulfonylureas: 9% vs 36%, p<0.001; glitazones: 0.2% vs 3.4%, p<0.001).
Three-year survival was 66% in insulin-treated versus 80% in patients without insulin (p<0.001). Subgroup analyses confirmed higher mortality in patients with insulin at discharge in both STEMI and NSTEMI patients, in those aged 75 years or less, or in those with or without PCI.
Using Cox multivariate analysis (covariates: age, sex, risk factors, comorbidities, type of AMI, CAD extent, use of PCI, use of CABG, HbA1c and admission glycemia levels, duration of diabetes, in-hospital complications, and other discharge medications), the adjusted HR for 3-year death was 1.56 (1.15-2.10), p=0.004, for patients with insulin at discharge; when treatment with insulin within 48 hours of admission was added to the model, the HR was 1.40 (1.06-1.84), p<0.02. Finally, in propensity score matched cohorts (n=240 in each group), HR for death at 3 years was 1.41 (1.01-1.98), p=0.04 for patients receiving insulin.
Conclusion: in this real-world nationwide registry, the prescription of insulin at discharge in diabetic patients was associated with poorer long-term survival, even after extensive multivariate adjustment for confounders.
- © 2011 by American Heart Association, Inc.