Improved Longitudinal Outcomes with Surgical Aortic Valve Replacement and Atrial Fibrillation Management over Transcatheter Aortic Valve Replacement Alone

J. Hunter. Mehaffey, M. Kawsara, V. Jagadeesan, D. Chauhan, J. Hayanga, C. E.. Mascio, J. Scott. Rankin, V. Badhwar
West Virginia University, Morgantown, West Virginia

Purpose: Class I guideline recommendations support atrial fibrillation (AF) treatment during surgical aortic valve replacement (SAVR). Recently, many low to intermediate risk patients with AF and aortic stenosis (AS) are managed by transcatheter aortic valve replacement (TAVR). We evaluated real-world longitudinal outcomes of TAVR vs SAVR with or without AF treatment.

Methods: The United States Centers for Medicare and Medicaid Services inpatient claims database was evaluated for all beneficiaries aged 65 and older with AF undergoing isolated TAVR or isolated SAVR with or without AF treatment between January 2018 and December 2020. Treatment of AF was defined as either the performance of operatively concomitant left atrial appendage obliteration with/without surgical ablation, or endovascular appendage occlusion or catheter ablation at any time during or after TAVR. Procedures were identified through diagnosis-related group and International Classification of Diseases 10th revision codes. Exclusions were age 85 years and older, endocarditis, prior cardiac surgery or TAVR, and patients receiving any other concomitant procedures. Elixhauser comorbidity indices further defined low to intermediate risk. Doubly robust risk-adjustment was performed using inverse probability weighting (IPW) propensity scores as well as multilevel regression and Cox Proportional Hazards time to event analysis. Primary outcomes assessed were cumulative 3-year mortality and stroke.

Results: A total of 24,902 patients were evaluated (17,453 TAVR; 7,449 SAVR). Of patients undergoing SAVR, 3,176 (42.6%) underwent AF treatment (SAVR+AF). In the TAVR population, only 656 (4.5%) underwent concomitant or post-procedural AF treatment (appendage occlusion n=491, catheter ablation n=165). Given this real-world practice distribution, SAVR alone, SAVR+AF, and isolated TAVR were examined. After comprehensive risk-adjustment with IPW propensity scores, all baseline characteristics and comorbidities were well balanced (Standard Mean Difference < 0.10). Comparing well balanced SAVR+AF to TAVR, there were no differences in index incidence of renal failure (OR 0.41, p=0.092) or bleeding (OR 0.94, p=0.712), however, TAVR was associated with increased pacemakers (OR 4.35, p< 0.0001) and major vascular injury (OR 6.67, p< 0.0001), while SAVR+AF had increased in-hospital stroke (OR 1.18, p=0.013) and mortality (OR 4.35, p< 0.0001). There were no index hospital outcome differences between SAVR alone and SAVR+AF. Compared to TAVR, SAVR+AF had lower 3-year valve reintervention (OR 0.56, p< 0.0001) and readmission for heart failure (OR 0.85, p< 0.0001). SAVR+AF was protective of cumulative 3-year stroke compared to isolated TAVR (OR 0.67 p< 0.001) or SAVR (OR 0.87 p=0.029). Finally, SAVR+AF was associated with reduced all-cause cumulative mortality compared to SAVR alone (OR 0.84, p< 0.0001) and TAVR (OR 0.69, p< 0.0001) .

Conclusion: In Medicare beneficiaries with AF requiring aortic valve replacement, SAVR with concomitant treatment of AF was associated with improved longitudinal survival and freedom from stroke compared to TAVR. Consideration should be given for SAVR with AF treatment as a first-line approach for patients with AF requiring aortic valve replacement.

Identify the source of the funding for this research project: Supported by NIH NHLBI # 2UM1 HL088925 12 (VB, JHM)