Zolpidem use and the risk of arrhythmia: A nationwide population-based cohort study in Taiwan.

BACKGROUND

Zolpidem is a widely prescribed hypnotic agent; however, its potential cardiovascular safety remains clinically uncertain. This study aimed to investigate the association between zolpidem use and the risk of various cardiac arrhythmias, specifically identifying which subtypes are most prevalent, using a nationwide population-based cohort in Taiwan.

METHODS

This population-based, matched-cohort study utilized the Longitudinal Generation Tracking Database (2000-2015) in Taiwan, comprising two million randomly sampled individuals. To establish a robust new-user design and minimize confounding, we included only those with zolpidem use for at least 7 days and strictly excluded individuals with any history of arrhythmia or heart failure. Users were matched 1:2 with controls by age, sex, and index year. The primary outcome was the incidence of cardiac arrhythmia, defined using International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) codes. Data were analyzed using Cox regression and Kaplan-Meier methods, adjusting for multiple comorbidities including hypertension and diabetes.

RESULTS

A total of 457 956 participants (152 652 zolpidem users and 305 304 matched controls) were analyzed. The groups were well-balanced with a mean age of 55.4 years and 49.2% being male. Zolpidem use was associated with a significantly increased risk of cardiac arrhythmia (adjusted hazard ratio [aHR], 1.983; 95% CI, 1.149-2.760; p < 0.001). Subgroup analyses revealed that zolpidem use notably increased the risk of paroxysmal tachycardia (aHR, 2.574; 95% CI, 1.514-3.998; p < 0.001) and atrial flutter (aHR, 2.491; 95% CI, 1.457-4.212, p < 0.001).

CONCLUSION

Zolpidem use is independently associated with an increased risk of incident arrhythmia, particularly tachyarrhythmias such as paroxysmal tachycardia and atrial flutter. Clinicians should incorporate cardiac risk assessment into their prescribing decisions and consider periodic monitoring for high-risk patients. Study limitations include the observational design and the unavailability of granular dosage data.