No Safety Signal Seen for CTO PCI at 3 Years: EUROCTO

SAN FRANCISCO, CA—Three-year safety data from the randomized EUROCTO trial show that PCI in these patients does not lead to an increased adverse cardiovascular event rate over optimal medical therapy (OMT), which the researchers argue lends further justification for revascularizing chronic total occlusions (CTOs) when experienced operators are doing the procedures.

“This study was the first to demonstrate a benefit of PCI over OMT for symptomatic patients with CTO,” said Gerald Werner, MD, PhD (Klinikum Darmstadt, Germany), referring to the previously reported, 1-year efficacy results showing improved quality of life and angina frequency with PCI as compared with OMT. At TCT 2019, Werner presented that 3-year safety outcomes, which provide additional support for the earlier results, he said.

Others appeared to agree. “This reinforces that [CTO PCI] is a good option for people with stable angina or symptoms that can be improved,” Emmanouil Brilakis, MD, PhD (Minneapolis Heart Institute, MN), who was not involved with the study, told TCTMD. “It doesn't bring up any long-term concerns for CTO PCI, which is similar to all the observational studies.” For Brilakis, the next issue that needs to be addressed in this controversial space is how to bring up procedural success among all centers from around 60%, as has been seen in registries, to closer to 90%, as has been observed in randomized trials.

Even so, it seems that two distinct camps still exist favoring CTO PCI or OMT for this patient cohort, Brilakis said, with the former saying the procedure provides tangible benefit for patients’ lives and the latter arguing that revascularization adds unnecessary risk. It’s likely that the only thing that would align the proponents and naysayers would be a randomized outcomes trial of CTO PCI, but “that's probably not going to happen because the event rates are so low that to get a difference we'd have to enroll several thousand patients, which realistically is not feasible. There are a couple studies going on in Europe on long-term outcomes, but in the range of 2,000 patients, [which are] unlikely to show a significant difference.”

EUROCTO at 3 Years

EUROCTO, which did not recruit its target number of participants (1,200), ultimately enrolled just 407 patients with a CTO from 26 centers and randomized 396 of them 2:1 to PCI or OMT (aspirin, statin, ACE inhibitor where tolerated, plus at least two antianginal agents at maximum tolerated dose) between March 2012 and May 2015. PCI was successful in 86.3% of patients treated and one-third received radial access. At 1 year, 10 patients assigned to OMT (7.2%) crossed over into the PCI arm due to ongoing angina, and this number jumped to 24 (17.5%) by 3 years.

The primary safety endpoint of cardiovascular death and nonfatal MI at 3 years was similar for the PCI and OMT groups (5.0% vs 3.2%; P = 0.32), as were the rates of cardiovascular death (2.7% vs 1.5%; P = 0.42) and nonfatal MI (2.3% vs 1.5%; P = 0.56).

MACCE, which included cardiovascular death, MI, ischemia-driven revascularization, stent thrombosis, and cerebrovascular events, was lower at 3 years in the PCI arm (10.7% vs 20.1%; P = 0.019), driven by a reduction in revascularization (7.3% vs 18.2%; P = 0.0035). There rates of cerebrovascular events and stent thrombosis were low and similar between the groups. “We achieved a very durable result with just one stent thrombosis,” Werner said during his presentation.

Seven patients (2.7%) in the PCI arm died of noncardiovascular causes compared with one (0.7%) in the OMT group—including five (1.9%) who had cancer. Werner explained that he looked to see if the patients with cancer received an excess of radiation during their procedures but deemed this a chance observation given that “none of these patients had more than 5 Gy. On average it was below 3 Gy.”

Looking at these results in the context of several studies of CTO PCI, Werner explained that while all-comer registries like OPEN CTO show 1-year mortality rates of between 4% and 8% depending on PCI success, randomized trials like EUROCTO and DECISION-CTO come in with lower rates “because we of course have a selection bias of excluding severely symptomatic patients.”

Future randomized trials should seek to increase patient risk, he argued.

Need for More Cross-referrals

In a discussion following the presentation, panelist Jonathan Hill, MD (King’s College Hospital, London, England), said he believed the trial “is important for the CTO community.” He asked about the maintenance of the symptomatic benefit in terms of both patient-reported outcomes and angina frequency.

Werner replied that his team did not conduct a Seattle Angina Questionnaire Health Status survey analysis at 36 months but reported that Canadian Cardiovascular Society class improvement was “maintained over time.” He acknowledged an uptick in the patients who crossed over to PCI, “but we had a low reintervention rate in the PCI arm, so that also attests to the longevity. . . with the modest restriction of a 3-year follow up.”

Panelist Robbert De Winter, MD, PhD (Academisch Medisch Centrum, Amsterdam, the Netherlands), said it was “reassuring” to see a very low adverse event rate at 3 years but questioned whether the outcomes could be generalized to lower-volume centers and operators. “The success of the procedure is restricted to high-volume centers [and] high-volume operators, but the safety may also be low for these centers,” he said.

“This is an important point,” Werner replied. “CTO PCI is a subspecialty within PCI and should be referred and reserved to centers or operators. It's not the center, it's the operator with high experience. I fully agree. We cannot say that everybody should do CTO PCI [and] will improve. If we lose our success rate, as we know from all the registries failed patients always do worse, so we need to avoid failures.”

In agreement, panelist Anthony Gershlick, MD (University of Leicester, England), : “There's far too few cross-referrals.” If a J-CTO score is no more than 2, “it's completely acceptable that good, well-experienced interventionists—maybe not regular CTO operators—attempt [CTO PCI] according to current techniques. If you fail at that, then I think cross-referral is what is best.”