JTCVS Open (Dec 2023)

New-onset postoperative atrial fibrillation after mitral valve surgery: Determinants and the effect on survivalCentral MessagePerspective

  • Rishav Aggarwal, MD,
  • Chesney Siems, MD,
  • Koray N. Potel, BS,
  • Austin Hingtgen, BS,
  • Qi Wang, MS,
  • Prabhjot S. Nijjar, MD,
  • Stephen J. Huddleston, MD, PhD,
  • Ranjit John, MD,
  • Rosemary F. Kelly, MD,
  • Rochus K. Voeller, MD

Journal volume & issue
Vol. 16
pp. 305 – 320

Abstract

Read online

Objective: Mitral valve surgery (MVS) carries substantial risk of postoperative atrial fibrillation (PAF). Identifying patients who benefit from prophylactic left atrial appendage amputation (LAAA) or maze is ill-defined. To guide such interventions, we determined preoperative predictors of PAF and investigated 3-year survival of patients with PAF. Methods: We performed a retrospective analysis of patients undergoing isolated MVS (N = 670) between 2011 and 2021. Patients with preoperative atrial fibrillation, LAAA or pulmonary vein isolation were excluded. Patient characteristics were compared between those without PAF and those who developed transient or prolonged PAF. Predictors of any PAF and prolonged PAF were identified using multivariable regression analysis. Results: In total, 504 patients without preoperative atrial fibrillation underwent isolated MVS. Of them, 303 patients (60.2%) developed PAF; 138 (27.3%) developed transient and 165 (32.7%) developed prolonged (beyond 30 days) PAF. Patients with PAF were older (65.7 vs 54.3 years, P < .001), with larger left atria (4.8 vs 4.3 cm, P < .001), greater prevalence of hypertension (60% vs 47.8%, P < .05), and were New York Heart Association class III/IV (36% vs 8.5%, P < .001). Independent predictors of PAF included left atria volume index (odds ratio [OR], 1.02; P < .003), older age (OR, 1.04; P < .001), heart failure (OR, 6.73; P < .001), and sternotomy (OR, 2.19; P < .002). Age, heart failure, and sternotomy were independent predictors of prolonged PAF. Patients with PAF had greater mortality at 3 years compared with those without PAF (5.3% vs 0.5%, P < .005). On multivariable analysis, PAF was associated with increased mortality (hazard ratio, 7.81; P < .046). Conclusions: PAF is common after MVS and associated with late mortality. Older age, advanced heart failure, and sternotomy are associated with prolonged PAF. These factors may identify patients who would benefit from prophylactic LAAA or ablation during MVS.

Keywords