乌鸦传媒

PARIS -- For patients with severe aortic stenosis (AS) and concomitant obstructive coronary artery disease (CAD), a percutaneous combined intervention worked surprisingly well relative to surgery, the TCW randomized trial found.

In this study, the less invasive treatment strategy -- comprising fractional flow reserve (FFR)-guided percutaneous coronary intervention (PCI) atop transcatheter aortic valve replacement (TAVR) -- resulted in fewer instances of all-cause mortality, myocardial infarction (MI), disabling stroke, unscheduled clinically-driven target vessel revascularization, valve re-intervention, and life-threatening or disabling bleeding at 1 year compared with the usual combined surgical aortic valve replacement (SAVR) and coronary artery bypass grafting (CABG; 4.4% vs 22.9% on intention-to-treat analysis, HR 0.17, 95% CI 0.06-0.51).

The incidence of all-cause mortality alone was not different between groups at 30 days, but it came to favor the percutaneous arm at 1 year (0% vs 9.7%, P=0.002). The signal of excess deaths after surgery -- largely cardiovascular ones, at that -- led to the TCW trial being stopped prematurely with only 172 people enrolled, according to Elvin Kedhi, MD, PhD, of McGill University Health Center, Royal Victoria Hospital, in Montreal, Quebec, reporting here at the annual EuroPCR meeting.

"It is not that the results of the surgery that we see here are worse than what we know ... The difference that we see is the TAVR arm does really good," Kedhi said during a press conference.

Session discussant Hendrik Treede, MD, a cardiac surgeon at University Hospital Mainz, Germany, agreed that these were "extremely good data" for this patient population. He highlighted the near lack of deaths at 30 days in the whole study -- counting only one in the surgical arm.

Treede said he struggled to understand how patients would die in the long run after initially surviving their surgery, however.

Prior observational research had shown that the percutaneous-only approach to AS and CAD left patients with higher rates of cardiac and vascular complications compared with combined surgery.

Indeed, European guidelines combined SAVR and CABG for people with severe AS and obstructive CAD, a strong class I recommendation at diameter stenosis 70%+ and class IIa at diameter stenosis 50-70%.

Similarly, American guidelines give a of SAVR/CABG over TAVR/PCI for patients with significant AS and significant CAD consisting of complex bifurcation left main and/or multivessel CAD with a SYNTAX score >33.

Kedhi cautioned that larger randomized trials are needed to corroborate the TCW findings but suggested it would be "very difficult" to do another trial. Already, there were substantial crossovers from the surgical arm to the percutaneous arm, and it is becoming difficult to convince patients to accept going into surgery for a study, he said.

EuroPCR director and press conference moderator William Wijns, MD, PhD, of the Lambe Institute for Translational Medicine in Galway, Ireland, said the study challenges the "no-big-deal" attitude surgeons have toward adding a bypass procedure to a valve surgery. "That statement that you hear during heart team discussions needs to be challenged based on this data," he said.

"I was just a little bit stunned by the big differences you find, as we were all kind of struck," commented Matti Adam, MD, of University Hospital Cologne, Germany, during the EuroPCR discussion.

"I think the devil's in the details, isn't it? The SYNTAX score is relatively low. This is not horrible, complex disease," offered Rajesh Kharbanda, MBChB, PhD, of Oxford Radcliffe Hospitals NHS Trust, England, a fellow panelist.

TCW was conducted in over a dozen centers across Europe. The randomized trial enrolled symptomatic AS patients at least 70 years of age who met criteria for complex CAD. In particular, they had to demonstrate either at least two de novo coronary lesions of 50%+ diameter stenosis in any native coronary arteries at least 2 mm, or a single left anterior descending (LAD) artery lesion either at least 20 mm in length or involving a bifurcation.

Extensive exclusion criteria included very low ejection fraction; prior CABG, SAVR, TAVR, or thoracotomy; and bicuspid valve anatomy.

Study authors intended to have 328 people randomized 1:1 to FFR-guided PCI/TAVR or CABG/SAVR.

Ultimately, there were 172 people enrolled and randomized. Average age was 76.5 years, with 68.6% of the cohort being men. Mean Society of Thoracic Surgeons score was 3.4%. Nearly 60% had Canadian Cardiovascular Society class I angina; approximately 44% New York Heart Association (NYHA) class II and 31% NYHA class III symptoms. Half had two-vessel disease, over 80% with lesions in the LAD and over 50% in the right coronary artery. Aortic valves had a mean gradient of 43 mmHg at baseline and aortic valve areas of 0.6 cm². Left ventricular ejection fraction was 55%.

The protocol for percutaneous treatment had patients undergo upfront PCI for all lesions with FFR ≤0.80; people with persistent angina and FFR ≤0.85 could get PCI later if FFR fell to ≤0.80 at 30-day follow-up. In the end, PCI operators performed FFR-guided revascularization in 80% of cases and achieved complete revascularization in 81% of cases.

As for the comparator group, surgeons were left to pick their revascularization targets following usual practice. This resulted in FFR-guided revascularization in under 30% of cases and complete revascularization for 66% of participants. The CABG arm all got biologic aortic valve prostheses.

Treede highlighted that the early clinical difference between groups was essentially driven by bleeding, the only primary endpoint component to reach statistical significance at 30 days.

The composite endpoint of cardiovascular mortality, stroke, MI, unscheduled coronary or valve re-intervention at 1 year supported the primary results favoring PCI/TAVR over CABG/SAVR (3.3% vs 13.7%, HR 0.23, 95% CI 0.06-0.83, P=0.01 for superiority).

Of note, the PCI/TAVR arm had excess permanent pacemaker implants at 30 days (8.8% vs 1.3%, P=0.04). However, this difference was no longer significant at 1 year (9.9% vs 2.9%, P=0.07), according to Kedhi's report.

Comment