Metformin use in type 2 diabetic patients is not associated with lower arterial stiffness: the Maastricht Study

Abstract

INTRODUCTION: Type 2 diabetes (T2D) is associated with cardiovascular disease complications such as myocardial infarction and stroke. These complications are at least partially the consequence of diabetes-associated increased arterial stiffness. Metformin, a first choice oral glucose-lowering drug, has been associated with potential cardio-protective effects. However, there are no data on the association between real-life metformin use and arterial stiffness. The objective of the current study is to investigate in a population-based sample of individuals with T2D the association between metformin use and aortic stiffness (i.e. carotid-femoral pulse wave velocity, cfPWV) and carotid stiffness [i.e. carotid distensibility coefficient and Young's elastic modulus (YEM)]. METHODS: We used data from The Maastricht Study, an ongoing observational prospective population-based cohort study (current N = 3451). All participants with T2D, based on pharmacy records (N = 672, 31.3% women, mean age 62.6 ± 7.7), were included in the current study. Linear regression analyses were used to study the association between current metformin use and cfPWV, distensibility coefficient and YEM, as compared with no metformin use. Furthermore, metformin use was stratified by cumulative dose (in grams), continuous duration of use (in days), average daily dose (in grams) and time since first prescription (in years). Regression coefficients of distensibility coefficient were multiplied by -1, consequently, for all arterial stiffness indices, a positive regression coefficient signifies increasing arterial stiffness. RESULTS: Linear regression showed that neither current metformin use was associated with cfPWV [adjusted B: -0.04 (-0.11 to 0.02)] nor metformin use was as stratified by cumulative dose, by continuous duration of use, by average daily dose or by time since first prescription. Metformin use was statistically significantly associated with higher carotid stiffness as assessed by distensibility coefficient [0.12 (0.01 to 0.23)], but not with YEM [0.10 (-0.03 to 0.22)]. However, there was no consistent pattern with the different stratifications of metformin use when further investigating the association with distensibility coefficient. CONCLUSION: We showed that there is no significant association between current metformin use and aortic stiffness, regardless of how metformin use in itself was defined. In addition, metformin use was not associated with a lower carotid stiffness. The present results showed no beneficial effect of metformin use, dosage or duration on arterial stiffness in middle-aged patients with T2D. Alternatively, metformin may exerts its cardio-protective effects via other pathways.