The rich repertoire of Spanish proverbs, often imbued with subtle irony, aptly reflects the longstanding debate between surgical aortic valve replacement (SAVR) and transcatheter aortic valve implantation (TAVI) in patients with low surgical risk. Phrases like “haste makes waste,” “greed spoils all,” “patience is a bitter-rooted tree that bears sweet fruits,” or “when you were a hammer, you showed no mercy; now as an anvil, be patient,” could well describe the vicissitudes we are all familiar with. 

The study presented focuses on findings from the AVALON registry (Aortic Valve Replacement in Elective Patients from the Multicenter Aortic Valve Registry), a multicenter registry encompassing cases treated with TAVI and surgery across three tertiary centers in Poland. The TAVI approach was exclusively transfemoral, and only elective procedures targeting the aortic valve were performed, without any associated surgical or interventional procedures. 

Between 2015 and 2019, 2393 patients were included—1764 underwent surgery and 629 received TAVI. This registry, similar to the all-comers NOTION study, commenced when TAVI was already an established option for patients with low surgical risk. However, the surgical group had a lower predicted risk (EuroSCORE II: 1.41) compared to the TAVI group (3.43). Additionally, the nearly 1:3 ratio of patients in the surgical group relative to TAVI reflects Poland’s patient allocation policy, reserving low-risk patients for surgery and using TAVI for those at moderate to high surgical risk. To address this, the authors conducted a propensity score analysis based on a comprehensive set of preprocedural variables, resulting in two homogeneous groups of 593 patients. The standardization focused on clinical variables rather than technical aspects, which might have influenced patient selection. However, given the study’s emphasis on post-treatment survival, the chosen variables hold relevance, with mean EuroSCORE adjusted to 2.02 vs. 2.46, representing low surgical risk. 

Periprocedural complication rates were not significantly different between groups. Hospital mortality was 0.3% for surgery versus 0.9% for TAVI, with severe bleeding events at 5.2% and 2.7%, respectively; renal failure at 4.7% vs. 2.4%; neurological complications at 1.9% vs. 0.3%; and pacemaker implantation at 1.9% vs. 2.7%, all respectively. 

Starting from comparable periprocedural mortality—contrary to common clinical trial findings—and with an average follow-up of 2.7 years, extending to a maximum of 6 years, surgery demonstrated a 30% lower mortality rate (HR = 0.7; p= 0.48). This divergence emerged after the initial two years, during which both approaches yielded overlapping outcomes, but diverged thereafter. Notably, the patient profile that showed the greatest survival benefit from surgery versus TAVI included males, age <75 years, smokers, with EuroSCORE <2, hypertension, atrial fibrillation, renal insufficiency, and/or ischemic heart disease (without concomitant revascularization). 

The authors conclude that survival rates did not differ between TAVI and surgery during the first two years post-procedure. However, beyond this period, surgery was associated with improved survival. 

COMMENTARY: 

The AVALON registry results align with the five-year findings from trials like PARTNER 3, which we previously discussed on the blog, and anticipate the outcomes from the five-year Evolut Low Risk study expected this year. Initially, TAVI and surgery yield comparable survival, but this balance shifts after two years, favoring surgery. When, as in the AVALON registry, surgery does not start at a disadvantage of higher periprocedural mortality, results beyond two years appear significantly more favorable than those seen in clinical trials. This similarity in outcomes up to discharge, and subsequent divergence, could result from the meticulous selection criteria: the absence of disproportionately higher numbers of additional procedures in the surgical group compared to TAVI, as commonly seen in clinical trials. 

Several factors have been proposed to explain this “catch-up” or “overtake” by surgery: paravalvular leakage, pacemaker implantation, new bundle branch blocks, thrombosis, suboptimal patient selection with incomplete procedures in TAVI cases, tolerance for non-revascularization or existing valve dysfunctions, and prosthesis durability. 

The exact cause for these findings within this registry remains uncertain due to insufficient data. However, recent clinical trial outcomes suggest that, although considered low-risk, TAVI patients still presented higher risk profiles than the surgical group. This is evidenced, for instance, by the lower rate of concomitant procedures (primarily revascularization) not performed in TAVI cases, serving as a rationale that may help justify increasingly less favorable results. 

The only clinical trial extending beyond five years is NOTION, with a mean EuroSCORE II of 3%. Rates of moderate or severe paravalvular regurgitation were 8.2% vs. 0%, attributable to older-generation TAVI prostheses; this is reflected in the 43.7% vs. 8.7% pacemaker requirement rates. Reintervention and endocarditis rates were also numerically higher in the TAVI group, while overall survival did not significantly differ. Nonetheless, postoperative results were notably worse than in AVALON, partly due to the high rate of Abbott Trifecta® valve implants. Given the high incidence of early degeneration, surgical survival outcomes may have been compromised. By contrast, the dominant prosthesis in AVALON was Hancock II, known for durability yet unable to match the transvalvular gradients of pericardial prostheses, which in PARTNER 3’s five-year results are numerically superior to TAVI. Representation of minimally invasive techniques remains minimal, absent in NOTION and at only 3% in AVALON. 

REFERENCE: 

Kowalówka AR, Kowalewski M, Wańha W, Kołodziejczak M, Mariani S, Li T, et al. Surgical and transcatheter aortic valve replacement for severe aortic stenosis in low-risk elective patients: Analysis of the Aortic Valve Replacement in Elective Patients From the Aortic Valve Multicenter Registry. J Thorac Cardiovasc Surg. 2024 May;167(5):1714-1723.e4. doi: 10.1016/j.jtcvs.2022.10.026.