Direct oral anticoagulants (DOACs) are available for nonvalvular atrial fibrillation patients. The advantage of DOACs is that regular anticoagulation monitoring is not required. However, adherence to the recommended regimen is essential. We investigated the association between medication adherence and the risk of cerebral infarction in patients taking DOACs. Patients admitted to any of the participating hospitals for cerebral infarction from September 2018 to February 2020 and prescribed DOACs before admission were defined as the case group, and patients hospitalized for diseases other than cerebral infarction, except for bleeding disorders, and prescribed DOACs before admission were defined as the control group. A nested case–control study was adapted, and 58 and 232 patients were included in the case and control groups, respectively. Medication adherence was assessed by the pharmacists through standardized interviewing. The adjusted odds ratio for the risk of cerebral infarction for low-adherence patients (<80% adherence rate) against good-adherence patients (100% adherence rate) was 9.69 (95% confidence interval, 3.86–24.3; p < 0.001). The patients’ age and other background characteristics were not found to be risk factors for cerebral infarction. In conclusion, low adherence is a risk factor for cerebral infarction in patients taking DOACs. Pharmacists should focus on maintaining ≥80% adherence to DOAC therapy to prevent cerebral infarction.

Background: Repaglinide is widely prescribed to reduce postprandial hyperglycemia and elevated glycated hemoglobin (HbA1c) levels associated with type 2 diabetes, and clopidogrel is a thienopyridine antiplatelet agent and widely used in cardiovascular and cerebrovascular diseases. It has been suggested that the concomitant use of repaglinide with clopidogrel may inhibit repaglinide metabolism, because repaglinide is a substrate of cytochrome P450 2C8 (CYP2C8) and the main metabolite of clopidogrel acyl-β-D-glucuronide inhibits CYP2C8 activity. In this study, we retrospectively investigated the effect of clopidogrel with repaglinide on plasma glucose and the risk of hypoglycemia associated with the combination of both drugs. Method: Patients were taking clopidogrel (75 mg/day) and started taking glinide (1.5 mg/day repaglinide or 30 mg/ day mitiglinide) for the first time from April 2012 to March 2017. We targeted subjects who were hospitalized at the start of glinide and whose preprandial plasma glucose was measured by a nurse. The glucose levels were collected for up to 5 days before and after the glinide start date. Results: Average fasting plasma glucose levels (before breakfast) in the repaglinide and clopidogrel group before and after starting repaglinide were 180.1±35.5 and 136.5 ± 44.1 mg/dL, with a mean decrease of 43.6 ± 33.6 mg/dL. In contrast, there was only a moderate decrease of 11.6 ± 30.0 mg/dL in the mitiglinide and clopidogrel group. Minimum plasma glucose levels in the repaglinide and clopidogrel group before and after starting repaglinide were 145.2 ± 42.9 and 93.3 ± 36.3 mg/dL, respectively. Decrease in minimum levels after starting glinide in the repaglinide and clopidogrel group (51.9 ± 47.5 mg/dL) was more significant than those in the mitiglinide and clopidogrel group (only 2.1 ± 29.1 mg/dL), and the repaglinide group (without clopidogrel, 15.5 ± 20.0 mg/dL). Hypoglycemia was observed in 6 of 15 patients in the repaglinide and clopidogrel group, but only 1 of 15 patients in the mitiglinide and clopidogrel group, and no patients in the repaglinide group. Conclusion: These findings indicate that minimum plasma glucose levels were significantly decreased in patients taking repaglinide and clopidogrel. Considering the risk of hypoglycemia associated with taking repaglinide and clopidogrel, when a glinide is required in patients taking clopidogrel, mitiglinide may be a better choice.