MethodsWe included patients with BD in a cross-sectional design, confirmed by structured interview, age 18 through 80. Clinical characteristics were obtained using a structured questionnaire or medical records review. ECG intervals duration and morphology were manually assessed by cardiologists and compared among clinical subgroups using Chi-square, Mann-Whitney, and Kruskall-Wallis tests. Exploratory multivariable linear and logistic regression models were fitted to adjust for potential confounders.
ResultsWe included 117 patients (60.7% women, 76.9% bipolar I, 50% history of psychosis, 22.6% suicide attempts). We found a significantly longer QTc interval in BD patients with hypertension (difference: 9.5 ms, p=0.006), obesity (difference: 25 ms, p=0.001), and metabolic syndrome (difference: 13 ms, p=0.007). Hypertension remained a significant predictor of longer QTc after adjusting for age, gender, and antipsychotic use (estimate 17.718, p=0.018). We observed a significantly shorter PR interval in women (difference: 6 ms, p=0.029), early age of onset (difference 6 ms, p=0.025), non-users of lithium (difference 4 ms, p=0.002), and early trauma (difference 4 ms, p=0.038). Finally, we identified significant correlations between symptom severity, blood glucose and PR interval (r=0.298, p=0.001; r=0.278, p=0.003; respectively).