Association of HbA1C with mortality and cardiovascular outcome in patients with type 2 diabetes using drugs with different hypoglycemic risk

Abstract Background Previous clinical trials showed that intensive glucose control did not reduce macrovascular events and mortality. It was speculated that severe hypoglycaemia contributed to these adverse outcomes. More contemporary clinical trials using newer anti-diabetic agents with less hypoglycaemic potential showed a linear relationship between major adverse cardiovascular events (MACE) and HbA1c reduction according to a meta-regression analysis. Purpose The aim of this study was to examine the association between clinical outcomes and HbA1c under different treatment backgrounds in the real-world setting. We hypothesised that when using drugs with minimal hypoglycaemic risk, lower HbA1c would be associated with better outcomes. Methods This was a retrospective population-based cohort study using a multi-centre electronic medical record database. Eligible patients were those diagnosed with type 2 diabetes between 2009 and 2020 who received non-insulin antidiabetic drugs. Patients were stratified into two groups based on the drugs received: drugs with high hypoglycaemic risk (sulphonylurea, meglitinide) and drugs with low hypoglycaemic risk (thiazolidinediones, acarbose, DPP4 inhibitors, GLP1 agonists, SGLT-2 inhibitors). Patients were excluded if they had received these drugs for less than 180 days or had received both high and low hypoglycaemic risk oral drugs simultaneously. HbA1c levels were assessed from the index date to the date of the interested outcome or the end of the study period (31 December 2021), whichever came first. The relationship between HbA1c and clinical outcomes was examined using a Cox proportional hazards model. The outcomes of interest included all-cause mortality and MACE. Results The study population consisted of 6,923 patients, including 3,639 high hypoglycaemic risk drug users and 3,284 low hypoglycaemic risk drug users. Patients with lower HbA1c levels during the follow-up period were older and frailer. After adjusting for baseline covariates, the association between the risk of death and HbA1c presented a U-shaped pattern, and this U-shaped association was similar in the two treatment cohorts (Table 1). On the other hand, the association between the risk of MACE and HbA1c presented a J-shaped pattern. In the high hypoglycaemic risk drug cohort, patients with the highest HbA1c levels (mean HbA1c 9.6%) were associated with a higher risk of MACE compared to the group with a mean HbA1c of 7.2% (hazard ratio [HR] 4.16, 95% confidence interval [CI] 2.07-8.37); lower HbA1c levels were not associated with a lower risk of MACE (Table 2). In contrast, in the low hypoglycaemic risk drug cohort, lower HbA1c levels were associated with a higher risk of MACE (HR 3.82, 95% CI 1.29-11.28). Conclusion Lower HbA1c levels were not associated with a lower risk of all-cause mortality and MACE. Furthermore, this phenomenon was independent of the hypoglycaemic risk of anti-diabetic drugs.