Post-load glucose spike is a determinant of post-MI prognosis in patients without known or newly diagnosed diabetes

Abstract

Abstract Background Post-prandial plasma glucose spikes contribute to the progression of atherosclerosis. Glycaemic variability may predict post-ACS prognosis. A third to two-thirds of these patients had diabetes mellitus (DM). Post-ACS prognosis is worse in DM than in those without. This has not been tested in patients without DM. Purpose To test whether post-load spike in plasma glucose in patients without known or newly diagnosed DM adversely affects prognosis. Methods Retrospective cohort analysis of 847 MI survivors without known or newly diagnosed DM who were followed up for MACE (death and non-fatal MI). The median post-glucose spike (PGS, defined as the difference between the 2h-PG and FPG) was 2.4 mmol/l for the whole cohort and 1.5 mmol/l for the patients with normal glucose tolerance (NGT). Group 1: PGS ≤2.4 mmol/l and Group 2: PGS >2.4 mmom/l were compared using Mann-Whitney test for continuous variables and chi-squared test for categorical variables. Event free survival in the two groups was estimated from the Kaplan–Meier curves and compared using log-rank test. Cox proportional hazard regression identified predictors of MACE. Continuous net reclassification improvement (NRI>0) and integrated discrimination improvement (IDI) and c-statistics determined the added predictive value of glycaemic matrices Results MACE was higher in group 2 (OR 1.99, 95% CI 1.36 to 2.91, p=0.0004) compared to group 1. In patients with NGT, MACE was higher in patients with PGS ≥1.5 mmol/l vs those below (OR 2.37, 95% CI 1.31 to 4.26, p=0.0041). Event free survival was worse in pre-diabetes than in the NGT groups (HR 1.57, 95% CI 1.17 to 2.12, p=0.003). and in group 2 than 1 (HR 2.01, 95% CI 1.49 to 2.71, p<0.001). Amongst the patients with NGT, event free survival was worse in patients with PGS ≥1.5 mmol/l (HR 2.09, 95% CI 1.35 to 3.25, p<0.001). PGS independently predicted MACE in the whole cohort (HR 1.16, 95% CI 1.06 to 1.26, p=0.002) and NGT group (HR 2.06, 95% 1.51 to 2.79, p<0.000). Group 2 independently predicted MACE in the whole cohort (HR 1.75, 95% CI 1.26 to 2.42, p<0.001). In the NGT group, PGS >median, independently predicted of MACE (HR 2.67, 95% CI 1.54 to 4.61, p<0.001). The c-statistic a model containing GRS only increased on addition of PGS (δAUC 0.0134, p=0.046) but not on addition of FPG. Within the whole cohort, PGS improved the net reclassification by 28% when added to the model containing GRS only. NGT cohort had higher net improvement at 46.6%. Addition of PGS to the model containing GRS and FPG resulted in NRI>0 of 25.5% in the whole cohort and 56.3% in the NGT cohort. Similar changes were seen in the IDI. Conclusion(s) PGS predicts post MI prognosis in patients without known or newly diagnosed DM including in patients with NGT. This suggests that PGS is a more powerful indicator of post-MI prognosis than FPG Funding Acknowledgement Type of funding source: None