In recent years, ambitious interventional cardiologists have started to perform percutaneous coronary intervention (PCI) on chronic total occlusions (CTOs), although these lesions have long been recognized as among the most difficult to successfully treat.
Many other physicians, including some prominent interventional cardiologists, have expressed grave concerns about this expansion of the field. In the complete absence of high-quality evidence, the practice grew on the basis of strong beliefs and observational studies alone.
Now the first randomized controlled trial, DECISION-CTO, has found no benefit from CTO PCI, although even its supporters say that the trial is far from perfect. At the recent American College of Cardiology meeting in Washington, D.C., that trial was the subject of considerable scrutiny and helped fuel the controversy at several scheduled debates over the pros and cons of CTO PCI.
To its advocates, the benefits of CTO PCI are clear. Leading interventional cardiologist Gregg Stone (Columbia University) after the meeting: "I totally believe in benefits of CTO PCI for right pts. We must prove it or DECISION-CTO will become COURAGE-CTO for non-PCI docs." (It should be noted that Stone's remark might also work against his intended position, since most cardiologists -- and even many interventionalists -- now believe that COURAGE had an enormously beneficial effect on the field.)
Passionate Advocates
The interventionalists focused on the poor prospect of patients who have a CTO. "How many here would not want their epicardial arteries closed?" asked Dimitrios Karmpaliotis (Columbia University) to his audience of cardiologists at one of the debates. He pointed out that no one tells surgeons not to bypass a CTO during coronary artery bypass grafting (CABG).
At another debate, Khaldoon Alaswad (Henry Ford) set out some of the most common arguments that have been used to support CTO PCI. Across the spectrum of coronary disease, from stable angina to STEMI, patients with CTO have much higher mortality, he said. Further, CTO patients who have complete revascularization have better outcomes than CTO patients with partial or failed revascularization. He also disputed the notion that some CTOs don't benefit from revascularization. Ischemia, he said, is inevitable, since "no CTOs are adequately collateralized." And, like interventionalists and other device users since the dawn of time, he cited continuing and ongoing improvements in devices in support of his position.
William Lombardi (University of Washington) is one of the most passionate advocates of CTO PCI. He suggested the problem isn't that there's too much CTO PCI but far too little. He believes the current atmosphere stifles innovation. "What would happen to Gruentzig if he had practiced today?" he asked at another debate.
Lombardi said he wasn't bothered by the lack of randomized controlled trials (RCTs) in the field. He said it was impossible to perform such a trial since the "luminaries of the field" will refuse to randomize patients because they are convinced of the benefits they can offer their patients.
Precisely because it is so difficult to perform and because patients with CTO are at such high risk, Lombardi argued that many more interventionalists need to be trained to perform CTO PCI. High-risk patients benefit the most, but these are the patients who often are not offered treatment. "We treat what we could, not what we should," he said.
CTOs are enormously undertreated, he argued, "not because of indication but because of technique and ability." Therefore the biggest predictor of whether a patient gets a CTO PCI is the operator, he said. "Those who do and those who get better will do a better job of taking care of patients." The operators with more experience performing CTO do better than low volume CTO PCI operators. "We have to keep evolving," he concluded. "Those who do will find a way, those who don't will find excuses."
Skeptics Seek Evidence
Perhaps paradoxically, CTO PCI advocates like Lombardi see the lack of RCTs as confirmation of their support for CTO PCI. It's obvious to them that the procedure is beneficial, at least in properly-selected patients and when performed by appropriately-trained operators. But skeptics view the lack of supportive RCTs with alarm. They point to the complete lack of evidence showing benefit and remind others that time and again the highly anticipated benefits of various procedures have failed to materialize when rigorously tested.
Both Eric Bates (University of Michigan) and Frederick Feit (NYU) agreed that decreases in angina and ischemia were the only clear benefits of CTO PCI but that these occurred at the cost of an increased contrast load, increased radiation, low success rates, more complications, and no improvement in clinical outcomes. Kirk Garratt (Christiana Care) said that, "if we are honest, then we are treating for symptoms," and that it is important to recognize the high rate of complications.
Philippe Gabriel Steg (Hôpitaux de Paris) said that "this is an area in dire need of RCTs." In the wake of the DECISION CTO trial, he said that "the onus is now on the interventional cardiology community -- to which I belong -- to demonstrate that CTO procedures have either symptomatic or prognostic benefit and in whom." Because of its important limitations (including smaller than planned enrollment and other design and technical issues), DECISION-CTO is unable to provide a final answer, said Steg.
Elusive Consensus?
There may be a very rough and tentative consensus emerging. All the speakers would likely agree with Karmpaliotis that CTO PCI would be indicated in "any vessel for which the expected benefits exceeds the expected harm," though it leaves unanswered the question of how to make that assessment. He said that a symptomatic patient with a high-risk stress test result and significant ischemic burden might be a reasonable candidate.
Garratt sought to find some middle ground. "We don't want to stifle innovation," he said. But the message to the public shouldn't be that this is ready for prime time. "We don't know if we have adequate justification." For most patients, he said, "we don't know if CTO PCI is the right thing to do."
Lombardi said that CTO should not be performed by everyone. He pointed to the model of transcatheter aortic valve replacement (TAVR) in the U.S. "If there are 400 TAVR centers in the U.S. then that's about the right number for CTO."
Feit said that patients with lifestyle-limiting symptoms refractory to medical therapy had an "unambiguous indication." But in the absence of refractory symptoms, even profound ischemia represents "a questionable indication at best." He also recommended extreme care in choosing patients, ensuring that patients have a short occlusion, good overall non-cardiovascular prognosis, that they can tolerate long-term anti-platelet therapy, and that they have an expert operator.
Inappropriate Use
I asked Robert Yeh (Beth Israel Deaconess Medical Center) to comment on CTO PCI and whether enthusiasm for the procedure had outpaced reasonable expectations. "Certainly, there seems to be growing enthusiasm for many to learn how to do it, and that should always raise the specter of inappropriate use," he said. But he also said that the CTO situation is "a very different story than the generalized inappropriate stenting story that is often discussed."
He noted that most interventionalists have avoided doing CTO PCI. "The starting point is not one of overuse," he said. CTO PCI is more challenging and more expensive, the success rates are lower, and the complication rates are higher. For these reasons they constitute less than 5% of the PCI procedures performed nationally, although these lesions are present in nearly 20% of patients with CAD.
Yeh said that CTO PCI does not need to be offered in all patients with CTO. "Undoubtedly many of them will respond to medical therapy." Instead, it is "the subgroup of patients who don't achieve a good result on medical therapy that we should consider CTO PCI. That's the correct population to study in an RCT if one is truly concerned about evaluating whether the procedure should be performed at all."
In that population, said Yeh, "the real question is whether the improvement in angina that is expected is worth the riskiness and difficulty of the procedure."
Yeh said he believes that most CTO operators are not motivated by commercial considerations. "Hospitals and practices, in most cases, don't see CTO PCI as a financial win. It ties up a lot of cath lab time. The extra payment to hospitals is not enough to offset the increase in the device costs. Everyone would easily make more money reading ECGs, or doing PCI on low-risk 70% [stenosis] lesions." But, he acknowledges, both doctors and hospitals may be attracted to the prestige gained by taking on these high-complexity cases.
On the other hand, device companies stand to benefit enormously from CTO PCI. "In any given procedure, we can run through more than 10 wires, two microcatheters and four stents, so on a per-patient basis, the companies sell much more product than for other types of PCI." said Yeh.
Although CTO PCI now accounts for less than 5% of PCI procedures, industry has a clear motivation to expand its use. It is certainly then no coincidence that industry sponsors training programs for interventional cardiologists. But it is worth noting that, because of the difficulty of CTO PCI, many interventional cardiologists fail to achieve mastery of the procedure.