Refining prediction of stroke in sinus node dysfunction patients without atrial fibrillation using a P-combined score: a multi-centre study

Abstract Aims Isolated sinus node dysfunction (ISND) is a sinus node dysfunction without atrial fibrillation. A high risk of ischaemic stroke (IS) has been reported in ISND populations. However, current guidelines do not recommend anticoagulation in ISND management. P-wave indicates ISND-related atrial remodelling. P-wave indices and the CHA2DS2-VASc score may contribute to risk stratification for ISND-related IS. Methods and results In this multi-centre longitudinal cohort, ISND patients were divided into development (n = 1185) and external validation (n = 988) cohorts. Ischaemic stroke prediction capacity of the P-combined score was assessed with regard to discrimination, calibration, and clinical effectiveness. The cut-off value of the score was confirmed by using a restricted cubic spline curve. One hundred and twenty-four (10.46%) ISND patients developed IS [1.63%/year; 95% confidence interval (CI): 1.49–1.78%/year] after a median 3.02-year follow-up in the development cohort. The P-wave terminal force in electrocardiogram-lead V1 (PTFV1) was the only significantly abnormal P-wave index (adjusted hazard ratio: 2.56; 95% CI: 1.72–3.80). Therefore, we incorporated the PTFV1 with the CHA2DS2-VASc score to generate a P-combined score. For a 5-year IS risk, the P-combined score improved Harrell’s C-statistic (95% CI) from 0.678 (0.618–0.738) to 0.716 (0.657–0.774) and 0.747 (0.677–0.816) to 0.808 (0.747–0.868) in the development and validation cohorts, respectively, along with calibration and decision curve analyses. The cut-off value of the score was 3 in the development cohort and well-discriminated in the validation cohort. Conclusion Chinese ISND patients have a higher IS risk than the general population. Compared with the CHA2DS2-VASc score, the PTFV1-combined CHA2DS2-VASc score shows a better risk-stratification capacity for ISND-related IS.