The main findings of this long-term population-based retrospective cohort study are as follows. (1) Metformin was associated with a decreased risk of new onset AF in type 2 DM patients not using other anti-diabetics; (2) this protective effect remained for at least two years after the use of metformin; (3) metformin inhibited the tachypacing-induced generation of ROS in atrial myocytes; and (4) metformin attenuated tachypacing-induced myofibril degradation. The current study is the first population-based epidemiological investigation on the association between the use of metformin and development of AF. This study is also the first to suggest a protective effect of metformin against tachypacing-induced cellular injury in atrial myocytes.
DM increases the incidence of AF through unknown mechanisms –,. Anatomical (increasing atrial size), biochemical (inflammation), and clinical (coronary disease and heart failure) changes have been suggested as possible factors involved in the pathogenesis of AF –. In the current cohort study, the metformin user and non-user groups had different profiles of co-morbidities and relevant medications, which were adjusted for our analysis. However, anatomical and biochemical characteristics were not available in the dataset.
This study demonstrated that metformin protected the diabetic patients from AF. This protective effect, however, seemed to diminish two to three years after the diagnosis of DM. This weakening of the protective effect of metformin after the third year may be attributable to common and gradual anatomical and biochemical deterioration because the co-morbidities were similar between the metformin user and non-user groups after 13 years of follow-up. The exact reason for the diminishing effect, however, remains to be investigated in future studies.
The studied population might be a subset of “early stage” of DM patients since they either used metformin or non-pharmacological tools to manage their sugar, although biochemical data was not available. This study, however, strongly hints that earlier use of metformin along diet control and physical activities might offer unexpected benefits in preventing arrhythmia. On the other hand, the effect of other anti-diabetics was not been evaluated in this study. For example, thiazolidinediones, another commonly used anti-diabetic that acts by activating peroxisome proliferator-activated receptor gamma, has also been shown to decrease new onset AF in patients with type 2 DM not using insulin . This protection is thought to be through anti-inflammatory and anti-oxidant mechanisms, which have been well documented –. Metformin has also been shown to have anti-inflammatory and anti-oxidant properties –,. Since oxidative stress mediates tachycardia-induced cellular remodeling, the protection offered by metformin may be due to inhibition of inflammation and oxidation . The effect of sulfonylureas, glucosidase inhibitors, dipeptidyl peptidase-4 inhibitors, and glucagon-like peptide 1 receptor agonist were not investigated either, and their connection with AF is completely unknown.
This is a large population-based and long-term observational study. The diagnosis of AF as well as other diseases and the use of medications came from a nationwide registry in Taiwan. Because of the limitations inherent in the design of this study, it is possible that potential registration errors, bias, and confounding factors may have existed to prevent the establishment of a solid relationship between the use of metformin and the incidence of AF. For example, impaired renal function as an ascertainment bias might interfere with the prescription of metformin. Although we included diagnosis of chronic renal disease in the sub-analysis of regression model and yielded similar results (Additional file 1: Supplement 4), the data available do not eliminate this possible confounding factor entirely. A lack of laboratory data such as hemoglobin A1c also prevented further detailed analysis and classification. Nevertheless, our findings are still important in that they provide significant evidence with regard the association between metformin usage and the lower incidence of AF in DM patients in a large population over a long period of time. The findings may help to elucidate the role of metformin and lead to other potential therapeutic targets in clinically reducing the incidence of AF. Further parallel cohort observations or strict randomized studies are warranted to clarify and validate our findings.
The in vitro experiment provides the first piece of evidence that metformin may be able to attenuate tachycardia-induced oxidative stress generation and subsequent atrial myocyte remodeling, which are known to be substantially involved in the complex and multifactorial pathogenesis of AF. The results of this experiment did not directly explain the mechanism by which metformin was able to reduce the incidence of AF in DM patients, but they did suggest that metformin was protective against AF-related adverse remodeling. More comprehensive and conclusive clinical or animal experiments are needed to elucidate why or how metformin offers protection.