Accelerated biological aging in bipolar disorder as determined by artificial intelligence-based electrocardiographic assessment

Background Individuals with bipolar disorder (BD) are at increased risk for major adverse cardiovascular events. Recent evidence suggests that the difference between chronological and physiological age, as determined by artificial intelligence (AI)-based electrocardiographic (ECG) assessment, may reflect biological aging and independently predict all-cause and cardiovascular mortality in unselected populations. We aimed to explore physiological aging using ECG-Age in subjects with BD. Methods We used a previously validated AI-ECG algorithm to assess physiological age in 12‑lead ECG signals from patients aged ≥30 years who sought primary care between 1998 and 2000 in Olmsted County, Minnesota, followed up using the Rochester Epidemiology Project. Delta-Age (DA) was defined as ECG-Age minus chronological age; accelerated aging was defined as DA ≥ 1SD above the mean. Results We included 278 subjects with BD (56.8% female, mean chronological age = 49.23 ± 10.56 years, mean ECG-Age = 52.47 ± 10.99) and 29,341 controls (53.7% female, mean chronological age = 54.01 ± 12.33 years, mean ECG-Age = 54.70 ± 11.35). DA was 2.5 years higher in subjects with BD than controls (3.23 ± 7.88 vs. 0.70 ± 7.62, p < 0.001). Logistic regression revealed that subjects with BD were more likely to have DA ≥ 1SD compared to controls (OR 1.34, 95% CI = 1.01–1.76, p = 0.03), independent of chronological age, sex, and established cardiovascular risk factors. Conclusion Subjects with BD show accelerated aging compared to controls, which appears to be independent of cardiovascular risk factors. Future studies should explore mechanisms of accelerated aging in BD, its generalizability to other major mental illnesses, and the potential impact of pharmacological treatment and comorbidity patterns as mitigating or further accelerating factors to biological aging.