Valve Sparing Root Replacement versus Bio-Bentall: A Meta-Analysis

Y. Yokoyama¹, T. Kuno², H. Takagi³, S. Fukuhara⁴, T. Kaneko^5
1University of Michigan, Ann Arbor, Michigan ²Montefiore Medical center, Bronx, New York ³Shizuoka Medical Center, Shizuoka, Shizuoka ⁴University of Michigan, Michigan Medicine, Frankel Cardiovascular Cen, Ann Arbor, Michigan ⁵Washington University in St. Louis, St. Louis, Missouri

Purpose: Valve-sparing root replacement (VSRR) has emerged as a viable alternative to the Bentall procedure, although evidence comparing the outcomes between VSRR and the Bentall procedure with a bioprosthetic valve (bio-Bentall) is scarce. Therefore, we conducted a meta-analysis comparing VSRR and bio-Bentall in patients with aortic root disease.

Methods: MEDLINE and EMBASE were searched through June 2023 to identify studies that investigated outcomes in patients who underwent VRSS or bio-Bentall for aortic root aneurysms. Outcomes of interest were short-term outcomes, including operative mortality, stroke, myocardial infarction, reoperation for bleeding, renal failure, and permanent pacemaker placement, and mid-term outcomes, including mid-term mortality and aortic valve reintervention. Risk ratios for short-term outcomes and hazard ratios (HR) for mid-term outcomes were extracted. A sensitivity analysis was conducted by excluding database studies. Furthermore, phase-specific analysis was performed for aortic valve reintervention by extracting HR for the following specific phases; early phase (0 to 5 years after repair), and late phase (more than 5 years after repair).

Results: Our analysis included five observational studies that enrolled 4,990 patients who underwent VSRR (n=889) or bio-Bentall (n=4,101). In observational studies, the propensity score methodology was used in 4 studies, and the Cox regression model was used in 1 study. One was a database study. The other four were from specialized centers. The mean follow-up was 2.3 to 9.8 years. No significant differences were observed in perioperative outcomes (early mortality, rates of stroke, myocardial infarction, reoperation for bleeding, new introduction of dialysis, and permanent pacemaker placement) between the two groups. Similarly, VSRR was associated with similar mid-term mortality (HR [95% confidence interval [CI]] =0.93 [0.67-1.02]; P =0.64; I2 =54%) and the rate of aortic valve reintervention (HR [95% CI] =0.79 [0.40-1.58]; P =0.50; I2 =67%) compared with bio-Bentall. However, sensitivity analysis after excluding database studies showed that the rate of aortic valve reintervention was significantly lower with VSRR (HR [95% CI] =0.50 [0.27-0.93]; P =0.03; I2 =37%). Similarly, the rate of aortic valve reintervention in the late phase (> 5 years after repair) was significantly lower with VSRR (HR [95% CI] =0.47 [0.23-0.94]; P =0.03; I2 =25%), although that in the early phase was similar between both groups.

Conclusion: VSRR was associated with equivalent perioperative and mid-term outcomes compared to bio-Bentall in patients with aortic root disease. However, our study suggests that VSRR might be associated with fewer aortic valve reinterventions with longer follow-up or at large-volume aortic centers.

Identify the source of the funding for this research project: None