Postload glucose spike but not fasting glucose determines prognosis after myocardial infarction in patients without known or newly diagnosed diabetes.

Background: The effect of postload glucose spikes (PGS), the difference between 2 hour post-load plasma glucose (2hPLPG) and fasting plasma glucose (FPG), on post-myocardial infarction (post-MI) prognosis in nondiabetic patients is unexplored.

Methods: This is a retrospective cohort analysis of 847 nondiabetic post-MI survivors who underwent a predischarge oral glucose tolerance test (median PGS: 2.4 mmol/L). Patients were divided into the unmatched groups 1 and 2 (PGS ≤ and > 2.4 mmol/L) and the propensity score-matched groups 1M and 2M (355 pairs assembled from the overall cohort), and these groups were compared. Major adverse cardiac events (MACE: death and nonfatal reinfarction) were recorded during follow-up (median: 3.4 years). Event-free survival was compared by the Kaplan-Meier method. Multivariate Cox proportional hazards regression determined the predictors of MACE. C-statistics (change in area under the curve, δAUC), continuous net reclassification improvement (NRI ), and integrated discrimination improvement (IDI) were used to compare models.

Results: The number of MACE was higher in groups 2 (27.3% vs 14.2%, P < .001) and 2M (24.5% vs 15.5%, P < .001). Event-free survival was worse in groups 2 (hazard ratio [HR] 2.01; 95% CI, 1.49-2.71; P < .001) and 2M (HR 1.63; 95% CI, 1.17-2.27; P = .004). PGS independently predicted MACE-free survival in the whole (HR 1.16; 95% CI, 1.06-1.26; P = .002) and matched cohort (HR 1.12; 95% CI, 1.02-1.24; P = .021). PGS, but not FPG or 2hPPG, improved the predictive performance of the base model (δAUC 0.013, P = .046), with greater improvement seen when PGS was added and compared to 2hPPG (δAUC 0.005, P = .034; NRI 0.2107, P = .013; IDI 0.0042, P = .046).

Conclusion: PGS is a better predictor of post-MI prognosis than 2hPPG in nondiabetic patients.