Valve-sparing versus biologic root replacement for root aortopathy with a non-stenotic bicuspid aortic valve

M. A. Thompson¹, B. P. Kramer², W. Kim³, T. Fujiyoshi⁴, A. Lowry², E. H. Blackstone², E. E.. Roselli^2
1Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio ²Cleveland Clinic, Cleveland, Ohio ³Riverside Methodist Hospital, Columbus, Ohio ⁴Tokyo Medical University, Tokyo, Tokyo

Purpose: In patients with root aortopathy and a non-stenotic bicuspid aortic valve (BAV) who desire to avoid anticoagulation, consensus has not been reached between valve-sparing reimplantation (VSRi) or biologic replacement (BR). We describe morphology, compare outcomes, and identify selection criteria for VSRi versus BR in matched cohorts.

Methods: From 01/01/2008–01/01/2020, 237 adults (220, 93% male; mean age 52 years) with root aortopathy and a non-stenotic BAV underwent VSRi (133, 56%) or BR (104, 44%) with either a composite valve-graft (98, 94%) or porcine xenograft (6, 6%). Most BAVs were Sievers type I (193, 81%) with right-left cusp fusion (222, 89%) and symmetrical orientation (132, 56%). Multivariable logistic regression was performed to identify factors associated with BR. One-to-one propensity matching produced 56 pairs. Endpoints were mean aortic valve gradient, moderate/severe aortic insufficiency (AI), aortic valve reoperation, and death. Median follow-up was 6.4 years.

Results: Postoperative mean aortic valve gradient remained constant at 10–11 mmHg for both groups through 10 years. Moderate/severe AI developed more rapidly following VSRi than BR. There were 20 (8.4%) aortic valve reoperations: 15 (11%) following VSRi and 5 (4.8%) following BR for AI (8, 40%), aortic stenosis (7, 35%), endocarditis (3, 15%), or mixed valve disease (2, 10%). Eight-year freedom from reoperation was 78% for VSRi and 91% for BR (P=.07; Figure 1). Ten-year survival was 92% and did not differ significantly between groups. Factors associated with BR versus VSRi were older age (58±11 vs. 47±12 years; P<.0001), leaflet calcification (58, 56% vs. 37, 28%; P<.0001) and aortic root diameter (5.1±0.6 vs. 4.7±5.1 cm; P<.0001). After propensity matching, matched VSRi patients developed more postoperative moderate/severe AI than unmatched VSRi patients (27/56, 48% vs. 7/77, 9%), leading to a higher frequency of aortic valve reoperations (9/56, 16% vs. 6/77, 7.8%). Matched versus unmatched VSRi patients had larger root diameters (5.0±0.4 vs. 4.6±0.5 cm; P<.0001), were more likely to have undergone leaflet debridement (13/56, 23% vs. 4/77, 5.2%; P=.002) and less likely to have undergone figure-of-eight suspension repair (18/56, 32% vs. 38/77, 49%; P=.05) (Figure 2).

Conclusion: Judicious patient selection is required to achieve durable VSRi in patients with root aortopathy and a non-stenotic BAV. Young patients, with excellent-quality valves that only require simple repairs (e.g., figure-of-eight suspension) are unlikely to develop postoperative AI or require aortic valve reoperation. However, patients with leaflet calcification and large aortic root diameters are at higher risk of VSRi failure. These patients may benefit from up-front BR due to its excellent durability regardless of patient morphology.

Identify the source of the funding for this research project: This study was sponsored in part by the High Risk Cardiovascular Research Philanthropic Fund.