Randomized controlled trials comparing transcatheter aortic valve implantation (TAVI) and surgical aortic valve replacement (SAVR) “show serious methodological imbalances with a common selective pattern and should be considered at high risk of performance and attrition bias that may affect internal validity,” a new systematic review and meta-analysis concludes.
The study, by the INTEGRITTY (International Evidence Grading Research Initiative Targeting Transparency and Quality) group of cardiologists and cardiac surgeons, was published online in JAMA Network Open on January 3.
“This systematic review and meta-analysis found that the randomized controlled trial design does not protect from biases other than nonrandom allocation. In randomized controlled trials comparing TAVI vs SAVR, there were systematic imbalances in the proportion of deviations from assigned treatment, loss to follow-up, and receipt of additional procedures and additional myocardial revascularization that can pose a serious threat to internal validity due to high risk of performance and attrition biases,” the authors say.
“These observations should lead clinicians to consider almost all randomized trials comparing TAVI and SAVR to be at high risk of biases, and it might reduce their internal validity, although the direction of the biases may be only hypothesized without individual patient data,” they conclude.
“My take-home message from this systematic review and meta-analysis is that we have to read the TAVI trials with a big critical eye as they have a high risk of bias,” lead author Fabio Barili, MD, PhD, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, told theheart.org | Medscape Cardiology.
“There were selective imbalances in deviations from assigned treatment, loss to follow-up and also in associated procedures. So, the results may not be as reliable as we would like, and we should be concerned about that,” Barili added.
“Given our concerns about the biases, it is important that each patient has a heart team evaluation, and the best approach is decided on an individual basis.”
Barili said the INTEGRITTY group are concerned that “guidelines are pushing indications towards TAVI without enough solid long-term evidence, particularly in the lower risk population.”
“TAVI is the future. It is a device that will change our practice completely,” he added. “But we do not believe we’ve got sufficient evidence to broaden the indication to the low-risk population, and the biases in these trials heighten these concerns.”
Barili says that the current paper highlights biases in the trials that make interpretation of the data difficult.
“Our aim was not to look at the outcomes of the trials but to look at potential biases. In my opinion the clinicians need to read the trial papers with criticism, knowing that these type of trials can be highly biased,” he said, “which makes it more important that the heart team chooses what they believe to be the right option for each patient.”
The researchers included 8 trials comparing TAVI versus SAVR in their meta-analysis: PARTNER 1A, PARTNER 2A, PARTNER 3, CoreValve US Pivotal High Risk Trial, SURTAVI, Evolut Low Risk Trial, NOTION, and UK TAVI.
They found three main causes of potential bias: deviations from assigned treatment, loss to follow-up, and receipt of additional procedures.
“There is systematic imbalance in these parameters that trend in the same direction — with significantly lower proportions of deviations to the assigned treatment, loss to follow-up, and receipt of additional procedures in the TAVI groups,” the authors write.
They suggest that loss to follow-up may be the most substantial threat to trial validity.
Almost all the pivotal trials have a high imbalance of loss to follow-up, with a loss of around 10% in the TAVI groups and more than 20% in the surgery groups at 5 years, and as the clinical events contributing to the primary outcomes (ie, death and stroke) are relatively rare in the study populations, the attrition bias introduced by significant loss to follow-up can be potentially amplified, they note.
“If there is an imbalance in loss to follow-up this will mean it will be more difficult to show differences in long-term outcomes,” Barili added.
The researchers also point out that deviation from treatment assignment was also highly imbalanced between the two groups, being 6.2-fold lower in the TAVI group compared with the surgery group, which “might raise concerns regarding internal validity.”
The third point of bias — patients who received additional procedures — also shows this occurred far less in the TAVI group (4.6%) than in the surgery group (16.5%).
“Patients undergoing surgery are more likely to have other procedures at the same time. The higher number of procedures will increase short-term risk but should lower the long-term risk,” Barili added. “These trials were only designed to look at relatively short follow-up periods, with the main results at 1 or 2 years. But we need to know what happens at 5 or 10 years follow-up.
“This makes it even more important to follow all patients and again questions the large loss to follow-up in the surgery groups in these trials,” he concluded.
Trialist Responds
Commenting on the study for theheart.org | Medscape Cardiology, Michael J. Reardon, MD, distinguished chair in the Department of Cardiovascular Surgery at Methodist DeBakey Heart & Vascular Center in Houston, who was involved in several of the TAVI vs SAVR studies, was unconvinced.
“This seems to be to be statisticians looking for things to criticize without understanding the complexity of running these trials that include surgery vs a transcatheter option and providing no reasonable alternative way to do this better,” he said. “I would challenge them to show me where there is better data for this question.”
On the individual points made, Reardon said he did not believe the imbalance in deviations from assigned treatment would have favored the TAVI group. “In most of the trials the surgery group had more subjects withdrawn prior to procedure, which is not at all unanticipated and was in large part attributable to either patients at the upper end of the risk spectrum in the surgery group seeking TAVI elsewhere or progressing to a worsened health status where the heart team no longer felt comfortable operating (ie, this imbalance systematically favored surgery),” he noted.
On the number of patients lost to follow-up, Reardon said this was “well within what would be expected for a trial of these durations.” He added that additional work using CMS claims data had validated that the primary endpoint analyses were robust.
He acknowledged that the issue of concomitant procedures was “reasonable,” but pointed out that many of the trials stratified by need for revascularization. “I see our approach to the need for revascularization as a strength rather than a weakness of the trials,” he said.
Barili reports no relevant financial relationships. Reardon has reported financial relationships with Abbott, Boston Scientific, Medtronic, and Gore Medical.
JAMA Netw Open. Published online January 3, 2023. Full text
For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook
Source: Read Full Article