Medication time of metformin and sulfonylureas and incidence of cardiovascular diseases and mortality in type 2 diabetes: a pooled cohort analysis

The effect of duration of medication with metformin and sulfonylurea (SUs) on cardiovascular diseases (CVD) and mortality events by duration of type 2 diabetes (DM) is unclear. This study aimed to investigate the effect of duration of treatment with metformin and SUs on CVD and mortality events based on DM duration in newly diagnosed DM patients. Data from three prospective cohorts of Tehran Lipid and Glucose Study (TLGS), Multi-Ethnic Study of Atherosclerosis (MESA), and Atherosclerosis Risk in Communities (ARIC) including 4108 newly diagnosed type 2 diabetes individuals (mean age, 59.4 ± 0.66 years) were pooled. Exposure was defined as the duration of metformin alone, SUs alone, and a combination of both since drug initiation. The Cox proportional hazards (CPH) model adjusted for confounders, including statin, aspirin, and anti-hypertensive, was used to estimate the hazard ratio (HR) (95% CI) for the outcomes. Cumulative exposure for metformin, SUs, aspirin, statin, and anti-hypertensive medication was calculated using the same method. The median follow-up was 20.33 ± 0.45 years. Cardiovascular events, all-cause mortality (ACM), and CVD mortality occurred in 767, 913, and 439 newly diagnosed DM patients, respectively. Metformin alone reduced the hazard of cardiovascular events, ACM, and CV-mortality by 7%, 4%, and 6%, respectively, for each year of use, respectively (p < 0.05); the corresponding values for SUs alone were 4%, 3%, and 4%, respectively (p < 0.05). The effect of metformin on reducing cardiovascular events, ACM, and CVD mortality continued until approximately 8, 10, and 5 years after the start of treatment, respectively, and then reached Plato. The effect of SUs on cardiovascular events, ACM, and CVD mortality continued to decline or reach Plato until approximately 6, 5, and 8 years after initiation of therapy and then was ineffective or reversed. The effect of the combination therapy on the reduction of cardiovascular events continued until 11 years after therapy initiation. Among newly diagnosed DM patients, metformin, with and without SUs, was associated with a reduced risk of cardiovascular events, ACM, and CVD mortality for up to about one decade. The combined effect of metformin + sulfonylurea was superior to the single effect of metformin or sulfonylurea alone. The combination therapy of Metformin and SUs can still be used with good safety, especially in the first years of diabetes diagnosis.