 |
 |
 |
Postoperative Decline in Ejection Fraction After Aortic Valve Intervention: A Marker of Ventricular Remodeling Rather Than Myocardial Injury
Erin Dwyer, Arjune S. Dhanekula, Catherine Eccleston, Bret G. DeGraaff, Makenna G. Girard, Amanda Lee, Dariga Tugan, Grihith Parreddi, Rachel Flodin, Craig Morris, Yonaton Buber, David Mauchley, Scott C. DeRoo, Christopher Burke
University of Washington, Seattle, WA
Background: A subset of patients undergoing surgical aortic valve replacement experience a decline in left ventricular ejection fraction (dEF) postoperatively. Risk factors and clinical impact remain unclear. The purpose of this study is to explore risk factors for dEF and compare clinical outcomes between dEF and non-dEF patients.
Methods: Patients aged >18 years that underwent aortic valve replacement (AVR), Bentall, David, or Ross operations at a single academic medical center from 2019-2024 were included. Exclusion criteria included acute type A dissection and transcatheter aortic valve replacement. dEF was defined as a 10-percentage-point or 20% relative reduction from preoperative to one-month postoperative echocardiogram. Differences in baseline, peri-operative, and outcome variables between dEF and non-dEF patients were assessed with either Chi-square test, Fisher's exact test, Student's t-test, or Wilcoxon rank-sum test depending on quantifier. Bonferroni correction was applied to P-values in multiple comparisons. All analysis was completed in R Version 2024.04.2.
Results: 628 patients were included in the final analysis, and 90 patients had the outcome of dEF. Mean follow-up time was 23 months. Baseline demographic characteristics and comorbidities including race, BMI, coronary artery disease, atrial fibrillation, and hypertension were similar. The dEF group had a higher baseline EF (63% vs 59%, p=0.001). Preoperatively, the median LV end-diastolic volume index (LVEDVi 91 (IQR: 57-114) vs 73 (IQR: 56-96) mL/m2, p = 0.03) and the frequency of severe aortic regurgitation (44% vs 20%, p < 0.001) were higher in the dEF group. One-month postoperatively, the median LVEDVi values remained higher in the dEF group (dEF 80 (IQR: 57-103) vs non-dEF 60 (IQR: 48-74) mL/m2, p<0.001) while median EF was lower (dEF 45% vs non-dEF 60%, p<0.001). While LVEDVi decreased from preoperative to one-month postoperative echocardiograms in both the dEF and non-dEF groups, normalization of median LVEDVi was observed only in the non-dEF group. By one-year postoperatively, LVEDVi continued to decline toward normalization in the dEF group, while remaining stable in the non-dEF group (dEF 71 (IQR: 52-89) vs non-dEF 61 (IQR: 50-72) mL/m2, p=0.027) (Figure 1). Among patients in the dEF group, 69% recovered their LVEF by the one-year echocardiogram. There was no significant difference in mortality between groups during the follow-up period (p= 0.103).
Conclusion: In this contemporary cohort of patients undergoing surgical aortic valve intervention, postoperative decline in LVEF occurred in 14% of patients. Identified risk factors included preoperative severe aortic regurgitation and higher baseline LVEDVi. The dEF group showed signs of positive LV remodeling that continued through one-year echocardiograms with a significant decline in LVEDVi over the study period. Further, nearly 70% of dEF patients recovered to their baseline EF over the first year. Importantly, postoperative LVEF decline was not associated with increased mortality, suggesting that this phenomenon may be benign and reflective of a remodeling process rather than intrinsic left ventricular injury.
Back to 2026 Abstracts
