Another Observational CTO Study Confirms Procedural Success of PCI but Doesn't Definitively Show Effect on Outcomes

Although there is no randomized trial evidence showing that PCI for patients with chronic total occlusion (CTO) lesions improves hard outcomes or even quality of life, another observational study shows that the procedure can be successfully performed in a large group of veterans with obstructive CAD who have identified CTOs.

While procedural success increased over the study period and with greater operator experience, successful CTO PCI was associated with significantly improved survival at 2 years compared with unsuccessful intervention.

The complex procedure, which unlike most is generally done to improve symptoms and not prolong life, has taken flack for potentially being overused, especially by eager operators who might not have the level of expertise needed to perform optimally. However, in a study published in the May 8, 2017, issue of JACC: Cardiovascular Interventions, Thomas Tsai, MD, MSc (Denver VA Medical Center, CO), and colleagues showed that CTO PCI is only performed in fewer than 10% of veteran patients with at least one CTO lesion.

“I think every single epidemiologic analysis that we have ever done has shown that CTO [PCI] is not the place where we're over utilizing. If anything, these patients consistently receive much less PCI than any other population,” Robert Yeh, MD (Beth Israel Deaconess Medical Center, Boston, MA), who was not involved in the study, told TCTMD. “So those interested in focusing on an area of overutilization probably should turn their attention elsewhere.”

Following the presentation at the American College of Cardiology Scientific Session of the highly anticipated, randomized DECISION-CTO trial, which failed to show a benefit of CTO PCI over medical therapy alone for all-cause death, MI, stroke, and repeat revascularization at 5 years and quality of life at 1 year, many proponents of CTO PCI have been left with results that don’t enable them to champion the procedure as perhaps they had hoped.

Causality Cannot be Inferred

In this analysis of 111,273 veterans with obstructive CAD in the VA CART registry, 26.4% were diagnosed with at least one CTO lesion. The CTO PCI success rate was 79.7% among the 2,394 patients who underwent the procedure. Overall, as noted, successful CTO PCI was associated with a lower mortality risk at 2 years (HR 0.67; 95% CI 0.47-0.95). Additionally, patients with failed procedures had higher rates of CABG at 2 years (P < 0.001), although procedural success had no impact on MI occurrence.

While the study showed “a really dramatic difference” in mortality between successful and unsuccessful CTO PCIs, “these data cannot be interpreted in a causal way,” Yeh warned. “Those patients who get unsuccessful CTO PCI are fundamentally different patients than those patients for whom we succeed at a procedure, and to compare their outcomes and suggest that somehow if we’d only succeeded in PCI for those who were unsuccessful, we would have normalized their mortality and made it equivalent to those who had successful PCI, that is probably not true.”

In an editorial accompanying the study, J. Aaron Grantham, MD (Saint Luke’s Mid America Heart Institute, Kansas City, MO), agrees, although he cites the large study population as the main benefit of the study. “This is particularly important with regard to their descriptive analysis of CTO PCI utilization in the VA system, where outcomes may be different than the general population due to the unique makeup of the population,” he writes.

Because the study’s findings are consistent with other similar contemporary analyses, continues Grantham, the researchers “provide the CTO PCI ‘believers’ with further reassurances that treating asymptomatic CTO patients with PCI might provide some benefits. Unfortunately, for the ‘nonbelievers,’ who might also call themselves ‘evidence-based’ clinicians, yet another retrospective, potentially confounded analysis, no matter how big, is not better.”

Before dismissing the entire study, however, Grantham says the data will be beneficial in confirming or refuting future datasets and the investigators “may be poised to map the appropriateness of CTO PCI procedures, evaluate the appropriateness in relation to attempt rate to help identify over- or underutilization, and describe the clinical outcomes according to AUC designation.”

Symptoms and Medication

In the absence of reliable randomized data, Yeh said the community needs to “talk less about this observational mortality data and we need to talk more about the symptomatic benefits of CTO PCI.” Also, more and better studies are needed “to convince those who are not convinced that it actually improves symptoms that it does,” he added.

It will be important to better understand which CTO patients benefit from PCI and which do not, Yeh observed. “If the risks are different between different people, and if the benefits are different between different people, then you have a classic case where we need to do better individualization of treatment and we need to develop better tools,” he said, adding that data from the Open CTO registry may help do so eventually.

As for a randomized trial, besides the ongoing EuroCTO study of 450 patients set to look at both quality of life and MACE, Yeh said the reason to do one would be more for the benefit of the scientific community.

“I don’t think we need to have a CTO trial that looks at hard endpoints. That would fail, for one, and it would actually just arm the naysayers of CTO PCI with more literature to say that this is a procedure that shouldn't be done when it's not really the primary reason that most of us are doing the procedure,” he said. “When a patient has angina and they are on maximal medical therapy and they have intractable chest pain, I know from clinical experience, and many of us who do the procedure regularly know, that it’s effective. It effectively reduces their angina.”

For any study of CTO patients, however, Yeh stressed that the participants need to be both symptomatic and on optimal medical therapy before the study begins as “those are really the patients who we should be doing the procedure on anyway.

“If you have a trial which takes people who are sort of marginally symptomatic and aren’t on good medical therapy and in one arm they do finally get their good medical therapy when they get into the trial, then you are going to diminish the impact of CTO PCI,” he continued.

What About Collaterals?

In the same issue of the journal, researchers led by José P.S. Henriques, MD, PhD (University of Amsterdam, the Netherlands), also investigated CTO lesions, looking closely at the survival effects of well- versus poorly-developed collaterals on 413 patients with STEMI undergoing PCI between 2000 and 2012.

Over 5 years, the researchers found lower mortality among those with well-developed collaterals (P = 0.01), who also happened to present more often in cardiogenic shock. Additionally, cardiac death was less likely in this cohort.

“This interesting data, while limited in its observational nature, suggests that cardiac outcomes may be significantly affected by the extent of collateralization of a nonculprit CTO artery after STEMI,” write C. Michael Gibson, MD, and Serge Korjian, MD (Harvard Medical School, Boston, MA), in an accompanying editorial. “This finding may be related to improved perfusion of the myocardium supplied by the CTO leading to preservation of myocardial tissue and in turn better myocardial tissue reserve.”