Low-Normal Peri-Procedural Potassium as a Prognostic Marker of Atrial Fibrillation and Stroke After CABG

Le résumé fourni par la source

Background Postoperative atrial fibrillation (POAF) and stroke are major complications following coronary artery bypass grafting (CABG). The clinical significance of peri-procedural potassium levels within the normal range remains unclear. Objective To evaluate whether low-normal peri-procedural mean potassium identifies patients at increased risk of POAF and 30-day stroke after isolated CABG. Methods We included 5,794 patients undergoing isolated CABG in an international multicenter cohort. Peri-procedural mean potassium (average of measurements 24 hours before and after surgery) was stratified into three groups: 3.5–3.9, 4.0–4.4 (reference), and 4.5–5.0 mmol/L. Primary and secondary outcomes were 30-day POAF and stroke, respectively. We used adjusted Cox proportional hazards and linear spline models. Results POAF occurred in 1,156 patients (20.0%) and 30-day stroke in 105 (1.8%). Compared with the reference group, low-normal potassium (3.5–3.9 mmol/L) was associated with a higher risk of POAF (adjusted HR 1.40, 95% CI 1.21–1.63, p<0.001) and stroke (adjusted HR 2.13, 95% CI 1.31–3.44, p=0.002). These associations remained materially unchanged in sensitivity analyses additionally adjusting for diabetes-related, potassium-modifying, and postoperative antithrombotic medications. Higher potassium (4.5–5.0 mmol/L) was not significantly associated with either outcome. A linear spline model showed an inverse association between potassium levels and POAF risk (p<0.001), without significant slope change at 4.0 mmol/L. Conclusion Low-normal peri-procedural potassium was associated with increased risks of POAF and 30-day stroke after CABG. These findings suggest that low-normal potassium may serve as a marker of increased perioperative vulnerability rather than evidence supporting routine aggressive potassium supplementation.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.