Optimal fasting glucose levels with regard to cardiovascular and mortality outcomes in people treated with or without antidiabetic medication

Abstract Aims To investigate the optimal fasting blood glucose (FBG) levels among individuals actively treated or untreated with antidiabetic drugs. Methods In two population‐based cohorts of Korean adults extracted from the National Health Information Database, multivariable‐adjusted hazard ratios of outcomes over 10 and 8 years of follow‐up were estimated according to achieved FBG levels and antidiabetic drug use. The primary outcomes were major cardiovascular disease (CVD) events and all‐cause mortality. Results In total, 66 533 of 450 537 and 100 556 of 767 382 participants in the respective cohorts received antidiabetic treatment. For untreated FBG, the CVD risk and mortality increased linearly from an FBG threshold of 5.6 mmol/L; however, for FBG treated with antidiabetic drugs there were J‐shaped associations with the outcome risks. For treated FBG levels of 4.4 to 5.5 mmol/L, 7.8 to 8.8 mmol/L, 8.9 to 9.9 mmol/L and ≥ 10.0 mmol/L, vs 6.1 to 6.9 mmol/L, the hazard ratios for major CVD events were 1.17 (95% confidence interval [CI] 1.04‐1.32), 1.06 (95% CI 0.96‐1.18), 1.37 (95% CI 1.22‐1.53) and 1.61 (95% CI 1.46‐1.78), respectively, and those for all‐cause mortality were 1.20 (95% CI 1.11‐1.29), 1.05 (95% CI 0.99‐1.12), 1.29 (95% CI 1.10‐1.50) and 1.69 (95% CI 1.59‐1.81), respectively. Conclusions These findings indicate that pharmacological therapy achieving FBG levels of <7.8 to 8.9 mmol/L and a non‐pharmacological approach to maintaining normal glucose levels help reduce the risk of adverse outcomes, while lowering FBG to normal levels through antidiabetic drugs is not beneficial or may even be harmful.