A subanalysis of PARTNER 3 quality of life data demonstrated a modest, but significant, improvement in one-year disease-specific quality of life after transcatheter aortic valve implantation (TAVI) compared with surgical aortic valve replacement (SAVR) in low-risk patients with severe aortic stenosis (AS) at one year. And, five-year data from the PARTNER 2A trial found that patients with severe AS at intermediate surgical risk who underwent TAVI had similar rates of death and disabling stroke as those who had SAVR. However, TAVI via a transthoracic route demonstrated poorer outcomes than surgical replacement.
Both sets of findings were presented at late-breaking sessions at the Transcatheter Cardiovascular Therapeutics scientific symposium (TCT 2019; 25–29 September, San Francisco, USA). Suzanne J Baron (Lahey Hospital and Medical Center, Burlington, USA) outlined the PARTNER 3 substudy data, simultaneously published in the Journal of American College of Cardiology, and Vinod Thourani (Georgetown University School of Medicine, and Medstar Heart and Vascular Institute, Washington, DC, USA) presented PARTNER 2A.
The randomised PARTNER 3 trial demonstrated that TAVI with the Sapien 3 balloon expandable valve (Edwards Lifesciences) resulted in lower rates of death and rehospitalisation at one year compared with surgical replacement in patients with severe AS at low surgical risk. However, the effect of treatment strategy on early and late health-related quality of life in this population remained unknown.
To address this, the researchers compared health status at one, six, and 12 months in these patients treated with either TAVI or surgical replacement in the PARTNER 3 trial. Over the one-year follow-up period they found substantial improvements in both disease-specific and generic health status compared with baseline in both treatment groups, despite most patients having only New York Heart Asssociation (NYHA) class I or II symptoms at baseline. Consistent with previous studies of transfemoral TAVI, TAVI was associated with significantly better health status than SAVR at one month on all scales (mean difference in KCCQ-OS 16 points; p<0.04 for both).
TAVI also demonstrated a significant health status benefit compared with SAVR at all timepoints (p<0.05) in analyses incorporating both survival and change in health status together. Exploratory analyses demonstrated that the late health status benefits seen with TAVI were driven by a difference in the proportion of patients who experienced a large (≥20 point) improvement in the KCCQ score. These benefits may be explained, in part, by differential rates of postprocedural complications between TAVI and SAVR.
Baron cautioned that the results may not be generalisable to other types of TAVI prostheses, alternative access routes, or to other patient groups excluded from the PARTNER 3 trial. In addition, it is not yet known whether the health status differences between the cohorts persist beyond one year.
But, she added: “Taken together with the clinical outcomes of the PARTNER 3 trial, these findings further support the use of TAVI in patients with severe aortic stenosis at low surgical risk. Longer term follow-up is necessary, and ongoing, to determine whether the health status benefits of TAVI at one year are durable.”
PARTNER 2A compared the key clinical outcomes, bioprosthetic valve function, and quality-of-life measures at five years for TAVI versus surgery in patients with severe AS at intermediate surgical risk. At two years, PARTNER 2A concluded that TAVI was similar to surgery for the primary endpoint of death or disabling stroke. The investigators randomly assigned 2,032 intermediate-risk patients with severe aortic stenosis to either TAVI or surgical valve replacement at 57 centres. The two-year primary endpoint was all-cause death or disabling stroke in the intention-to-treat (ITT) population. At five years, all primary and secondary clinical and echo endpoints were analysed in both ITT and prespecified as-treated (AT) populations.
The five-year event rate for the primary endpoint of death or disabling stroke was 47.9% after TAVI and 43.4% following surgery (hazard ratio [HR] 1.09, 95% confidence interval [CI] 0.95–1.25, p=0.21). In the transfemoral cohort, there also was no difference at five years (44.5% TAVI versus 42% surgery, HR 1.02, 95% CI 0.87–1.20, p=0.80).
However, in the transthoracic cohort, the rate of death or disabling stroke was significantly higher after TAVI (59.3% versus 48.3%, HR 1.32, 95% CI 1.02–1.71, p=0.03). Thourani indicated: “In the landmark analysis within the transthoracic cohort, the higher values of the primary endpoint for transthoracic TAVI in the first two years continue to diverge from years two to five, in favour of surgery.”
In addition, early improvements in functional status and quality of life were maintained at five years for both TAVI and surgery patients. But, said Thourani: “The rates of rehospitalisations and reinterventions are higher in the TAVI group at five years. There appears to be a divergence of [Kaplan Meier] curves starting at two years, and it is higher for TAVI. Approximately half of the reasons for reintervention for TAVI were due to structural valve deterioration, while the majority of surgery patients had endocarditis as their failure mode.”
Treatment for reinterventions was better tolerated in the TAVI group, and there was also a small, but significant, difference in valve haemodynamics in favour of TAVI. He noted: “The absolute event rates for both groups are very small, and that should be very comforting for us to see as surgeons and cardiologists.”
TAVI was also associated with increased mild, moderate, and severe paravalvular leak (PVR); Thourani pointed out that the finding “may be historic in some ways, since the Sapien 3 valve has shown negligible moderate–severe PVR in the PARTNER 2 S3 substudy and the PARTNER 3 study”. Moderate–severe PVR, but not mild PVR, was associated with increased late mortality in TAVI patients.
Summing up, he said: “At five years, the results of TAVI with a Sapien XT valve, versus surgery in intermediate risk patients … for the primary endpoint—death from any cause or disabling stroke—in the overall population and the transfemoral population, was similar for TAVI and surgery. In the transthoracic cohort, unfortunately, the primary endpoints were significantly increased after TAVI compared to surgery.”
But, as Sapien XT is no longer commercially available for aortic stenosis in the USA, the clinical applicability of the findings “should be tempered”, Thourani stressed. In addition, the current generation Sapien 3 is lower in profile “which has led to a 95–96% transfemoral usage, and a small percentage of tranthoracic cases.”
And, missing data at five years, especially in some of the surgical patients, for some clinical and echo endpoints “reduces the rigor of the conclusions,” he said. Follow-up will be extended to 10 years.
TAVI should remain an alternative to surgery in intermediate-risk patients with severe aortic stenosis, “especially those patients that can be done transfemorally”. But, in patients without acceptable transfemoral access, surgery may be the preferred alternative to transthoracic TAVI.
