Sex differences in QTc prolongation and electrocardiographic changes before and after antipsychotic treatment of schizophrenia

Introduction Sex differences in ventricular repolarization are well established, with women showing longer corrected QT (QTc) intervals and greater susceptibility to drug-induced QT prolongation, but longitudinal data on sex differences in electrocardiographic (ECG) changes during antipsychotic treatment in acutely hospitalized patients with schizophrenia (SCZ) are scarce. Methods We included 586 inpatients with SCZ (314 males, 272 females) who underwent ECG and metabolic assessments at admission and before discharge. ECG parameters, lipids and fasting glucose were compared between sexes, and QTc prolongation was defined using sex-specific thresholds (males >440/450 ms, females >460/470 ms). Analyses included paired and independent t-tests, McNemar tests, ANCOVA and multivariable logistic and linear regression, together with pre-specified sensitivity analyses using four alternative rate-correction formulae (Fridericia, Framingham, Hodges and Rautaharju), mixed-effects models and Benjamini-Hochberg correction; 95% confidence intervals (CIs) were reported for proportions. Results At baseline, females had longer QTc than males (430.3 ± 23.3 vs. 416.6 ± 23.9 ms; P < 0.001), a difference consistent across all five formulae, whereas males had wider QRS duration (91.6 ± 10.2 vs. 86.0 ± 9.2 ms; P < 0.001). Marked QTc prolongation at admission was more frequent in males (5.4% [95% CI 3.4–8.5] vs. 1.5% [0.6–3.7]; Fisher exact P = 0.013). After treatment, QTc shortened significantly in males (−4.6 ms; P = 0.0004) but not in females (−2.2 ms; P = 0.165); this crude between-sex difference was not significant and its magnitude was formula-dependent, whereas the baseline-adjusted between-sex difference in ΔQTc remained significant across all formulae (female β = +6.5 to +9.8 ms; all P < 0.001). Male marked prolongation fell from 5.4% to 0.6% (exact McNemar P = 0.0007), and female sex was independently associated with lower odds of QTc prolongation tendency (odds ratio [OR] = 0.49, 95% CI 0.29–0.85). Discussion In hospitalized patients with SCZ, females had longer baseline QTc and, in adjusted models, less QTc shortening, whereas males showed a higher prevalence of threshold-defined marked prolongation at admission. Because these exploratory findings are partly dependent on the rate-correction formula, sex-specific ECG monitoring should be evaluated prospectively.