High Rates of Ad Hoc PCI for Left Main and Multivessel CAD

New observational data suggest that discussions amongst the heart team aren’t happening nearly as often as they should be for stable patients with multivessel or unprotected left main coronary artery disease.  

In an analysis of New York State’s PCI Reporting System and Cardiac Surgery Reporting System published in JACC: Cardiovascular Interventions, investigators found that ad hoc PCI is frequently performed in these types of patients, with roughly one-third receiving a stent immediately following a diagnostic catheterization.

“The magnitude of ad hoc PCI, and the amount of variation seen across hospitals, certainly suggests that heart teams are not being used as much as they should be used based on what the guidelines say,” lead investigator Edward L. Hannan, PhD (University at Albany—State University of New York, NY), told TCTMD.

James Blankenship, MD (University of New Mexico, Albuquerque), and Krishna Patel, MD (University of New Mexico Health Sciences Center), who wrote an accompanying editorial, say that past criticisms of ad hoc PCI have had “seemingly little effect,” adding that it continues to be performed frequently, “even in high-risk anatomical subsets where guidelines would recommend a heart team approach rather than ad hoc PCI.”

That raises several concerns, they write. “First, patients deserve the freedom to choose the strategy that provides the best odds of good outcomes. Second, in this era of patient-oriented decision-making, patients deserve to have time to evaluate options in an unhurried environment where they can consult with family and experts from various disciplines to arrive at a decision that suits them best. Both of these may be compromised by ad hoc PCI.”

But Kirk Garratt, MD, MSc (ChristianaCare, Newark, DE), a former past president of the Society for Cardiovascular Angiography and Interventions, commenting on the study, said the authors and the editorialists may be making the flawed assumption that ad hoc PCI implies no conversation about the risks and benefits of different revascularization strategies took place, a limitation acknowledged by the researchers.

“In current practice, many providers have a very thorough conversation with the patient and their families about treatment options,” he told TCTMD. “They might not have a surgeon present for the conversation, but I think interventional cardiologists are much better about presenting the benefits and liabilities of different treatment options, particularly for high-risk populations.”

Nonetheless, his hope is that rates of ad hoc PCI decline.

“It represents an existing opportunity for us as interventional cardiologists to evolve our practice,” said Garratt. The New York State data, he added, are 4 to 5 years old and interventional cardiology is changing, at times slowly. “We’re evolving away from old habits, recognizing that they’re not bad habits, but they’re just not optimal. It’s good to try to make changes where it’s appropriate.” 

Following the Guidelines

In the 2021 American Heart Association/American College of Cardiology/Society for Cardiovascular Angiography and Interventions clinical guidelines for coronary revascularization, a heart team approach that includes interventionalists, surgeons, and clinical cardiologists is a class 1 recommendation in patients for whom the optimal treatment strategy is unclear.

In both multivessel and left main disease, different revascularization strategies are possible depending on patient anatomy, comorbidities, and preferences. For example, in the 2011 guidelines for PCI, which predate the period studied by Hannan and colleagues, CABG surgery was a class 1 recommendation for patients with left main disease while PCI had a weaker recommendation (either class 2a, 2b, or 3 based on different factors). Additionally, CABG surgery was a class 1 recommendation for patients with three-vessel disease and two-vessel disease affecting the proximal LAD. Here again, PCI has a weaker recommendation.

“People are aware that the rates of ad hoc PCI are high, and have been for many years,” said Hannan. “The attempt here was to try to use the information from guidelines to limit the study to patients who would seemingly not be candidates for ad hoc PCI. According to the guidelines, bypass surgery would be, for most of them, the recommended intervention barring issues about patients being contraindicated. If the guidelines were used properly, you would expect a heart team evaluation prior to PCI” that would take place after the diagnostic angiogram.  

The magnitude of ad hoc PCI, and the amount of variation seen across hospitals, certainly suggests that heart teams are not being used as much as they should be used based on what the guidelines say. Edward L. Hannan

The analysis is based on 8,196 patients with two-vessel proximal LAD disease, three-vessel disease, or unprotected left main disease treated with PCI between December 2017 and November 2019. During this same time period, 17,573 patients with similar disease characteristics underwent CABG surgery.

In total, 6,425 of the stable PCI patients with multivessel/left main disease underwent ad hoc PCI. The percentage ranged from 58.7% of patients with unprotected left main disease to 85.4% of patients with two-vessel proximal LAD disease. Of those with three-vessel disease, 76.7% underwent ad hoc PCI. Stable patients with diabetes and multivessel disease were less likely to undergo ad hoc PCI, as were those with LVEF ≤ 35% and multivessel/left main disease.

Additionally, patients treated at hospitals without surgery on site were more likely to undergo ad hoc PCI (86.1% vs 76.6% at centers with surgery; P < 0.0001).

When researchers focused on all patients with multivessel/left main disease, which included those undergoing PCI and CABG surgery, ad hoc PCI was performed in 35.1%. The frequency of ad hoc PCI ranged from 11.5% for patients with left main disease to 63.9% for patients with two-vessel proximal LAD disease, with ad hoc PCI done in 32.4% of patients with three-vessel disease. Patients with diabetes and impaired LVEF (≤ 35%) were again less likely to undergo ad hoc PCI than the overall group, but roughly one-third were still treated in an ad hoc manner.     

Overall, there was a large variability in the percentage of stable patients undergoing ad hoc PCI across the 65 hospitals where they were treated. When all revascularizations were considered across hospitals, the percentage of ad hoc PCI ranged from 36.9% to 75.1% for two-vessel proximal LAD disease and from 3.3% to 25.5% for left main disease. Similar results were observed at the physician level.

To TCTMD, Garratt said the rate of ad hoc PCI for left main disease isn’t completely out of line, noting that some patients “make it abundantly clear” they don’t want to surgery while others with select left main lesions will fare just as well with PCI as with surgery. That said, “the multivessel patients with diabetes or impaired pump function, the rates of ad hoc angioplasty are a bit higher than I would hope to see in today’s practice,” he added.

Garratt emphasized the importance of discussing with the patient the plan should the angiogram reveal left main or multivessel CAD, because it can lead to disappointment and dissatisfaction otherwise.

“They go into the lab, and once the anesthesia clears, you tell them ‘you have some blockages here, but we’re going to chat about it first’—it’s not what patients expect,” said Garratt. “It speaks to the need for interventional cardiologists, and other team members, to have conversations ahead of the cath lab about what might happen.”

Modified Heart Team Approach

To TCTMD, Hannan said there are patients for whom ad hoc PCI is appropriate, even within their study, who can’t be identified given limitations of the data. The researchers acknowledge that some patients may have been adamant before catheterization that they wouldn’t be interested in CABG surgery under any circumstances, while others might not have been eligible based on coronary anatomy, surgical risks, or comorbidities. It’s also possible there was an ad hoc discussion with a cardiac surgeon between the diagnostic cath and PCI, with an agreement that PCI was the only viable option.   

Still, the researchers say the magnitude of variation suggests that hospitals and referring physicians “vary substantially in their use of heart teams.” They even point out that ad hoc PCI may expose operators and centers to legal jeopardy if bad outcomes occur, since it might be argued they didn’t follow clinical guidelines and offered substandard care.

We’re evolving away from old habits, recognizing that they’re not bad habits, but they’re just not optimal. Kirk Garratt

In the editorial, Blankenship and Patel highlight a range of biases on the part of interventional cardiologists and patients that might account for the high rate of ad hoc PCI. Overconfidence in their skill set, as well as the challenges of taking a prepped patient off the table only to arrange a consultation with the heart team, can push operators toward ad hoc PCI. The editorialists advocate for a modified systematic approach to the heart team, one that includes informed consent about the pros and cons of PCI and surgery before the catheterization. There’s also a good case to be made for having surgeons able to provide table-side ad hoc discussions, as well of general, noninvasive cardiologists available, who can provide a secondary or third opinion. 

Garratt, however, said that he’s not keen on the idea of ad hoc discussions with a surgeon when the patient is on the table. One, it’s not easy to get a surgeon to the cath lab on a moment’s notice. Additionally, “if you’re sincere about adopting a heart team approach, it needs to have the patient’s voice in the conversation,” he said. “It’s not about arranging a council of wise men. It’s about talking through the options with the experts and the person undergoing the procedure. Once they’re sedated, that’s off the table.”