Stable patients who have a diagnostic cardiac catheterization for multivessel disease or two-vessel proximal left anterior descending disease often have percutaneous coronary intervention (PCI) in the same session, possibly without input from a multidisciplinary heart team, a new study suggests.
The study, a retrospective analysis of more than 8,000 catheterization procedures in New York State during 2018 and 2019, was published in JACC: Cardiovascular Interventions.
Among the stable patients with multivessel disease or left main (LM) disease who had PCI, 78.4% of PCIs were performed in the same session as their diagnostic catheterization procedure, known as ad hoc PCI, a “surprisingly high rate,” the authors wrote.
The 2011 clinical guidelines in place during the study period advised coronary artery bypass graft (CABG) surgery as a class 1 recommendation for LM disease, whereas PCI is a lower-class recommendation (J Am Coll Cardiol. 2011;58:e44-e-122), they noted.
“Ad hoc PCI can be inadvisable when guidelines indicate that patients can realize better outcomes with CABG surgery,” lead study author Edward L. Hannan, PhD, MS, said in an interview. “The issue is that ad hoc PCI eliminates the opportunity for a multidisciplinary heart team to evaluate the patient.”
Dr. Hannan is principal investigator for the cardiac services program at the New York State Department of Health in Albany and distinguished professor emeritus at the University of Albany School of Public Health.
The researchers analyzed data from two mandatory New York State PCI and cardiac surgery registries, the Percutaneous Coronary Interventions Reporting System and the Cardiac Surgery Reporting System. A total of 91,146 patients had an index PCI from Dec. 1, 2017, to Nov. 30, 2019.
The study included patients who had two-vessel disease with proximal left anterior descending (PLAD) disease, three-vessel disease or unprotected LM disease. Exclusion criteria included a previous revascularization, among a host of other factors. The analysis also identified 10,122 patients who had coronary artery bypass graft (CABG) surgery in addition to the 8,196 patients who had PCI.
The percentage for ad hoc PCI ranged from 58.7% for those with unprotected LM disease to 85.4% for patients with two-vessel PLAD. Among the patients who had PCI for three-vessel disease, 76.7% had an ad hoc PCI.
Selected subgroups had lower ad hoc PCI rates. When patients who had a myocardial infarction within 1-7 days were excluded, the ad hoc percentage decreased slightly to 77.2%. PCI patients with diabetes were also less likely to have ad hoc PCI (75.7% vs. 80.4%, P < .0001), as were patients with compromised left ventricular ejection fraction (< 35%; 64.6% vs. 80.5%, P < .0001).
When all revascularizations – PCI plus CABG – were taken into account, the rate of ad hoc PCIs was 35.1%. Rates were 63.9% for patients with two-vessel PLAD disease, 32.4% for those with three-vessel disease, and 11.5% for patients with unprotected LM disease.
One potential disadvantage of ad hoc PCI, the authors noted, is that it doesn’t allow time for a multidisciplinary heart team to evaluate the patient for a different treatment, such as CABG or medical therapy. “This multidisciplinary team can evaluate all the pros and cons of different approaches, such as PCI vs. CABG surgery in this case,” Dr. Hannan said.
The study findings imply a potential overutilization of PCI and a greater likelihood of forgoing a more appropriate intervention, he said, “given that we have chosen for the study groups of patients who in general benefit more with CABG surgery.”
The results also showed variability in ad hoc PCI rates among hospitals and physicians. “They are large enough to suggest that there is a fairly large variation across the state in the use of heart teams,” he said.
For unprotected LM disease, the risk adjusted rate for hospitals of ad hoc PCIs among all PCIs ranged from 25.6% in the lowest quartile to 93.7% in the highest. Physician rates of ad hoc PCIs for the same indication, which were ranked by tertile, ranged from 22% for the lowest to 84.3% for the highest (P < .001).
One strength of the study, Dr. Hannan said, is that it is a large population-based study that excluded groups for whom an ad hoc PCI would be appropriate, such as emergency patients. One limitation is that it did not account for legitimate reasons for ad hoc PCI, including contraindications for CABG surgery and patient refusal of CABG surgery.
In an invited editorial comment, James C. Blankenship, MD, and Krishna Patel, MD, wrote that this study shows that “past criticisms of ad hoc PCI have had seemingly little effect.”
“The article provides a striking example of a difference between guideline-directed practice and real-life practice,” Dr. Blankenship said in an interview. “Guideline recommendations for the heart team approach are well known by interventionalists, so the findings of this study do not reflect ignorance of cardiologists.” Dr. Blankenship, a cardiologist and professor of medicine at the University of New Mexico in Albuquerque, is a coauthor of the 2011 PCI guidelines.
It’s more likely the study findings “reflect unconscious biases and sincere beliefs of patients and interventionalists that PCI rather than CABG is in patients’ best interests,” Dr. Blankenship said.
He noted the variation in practice across hospitals and individuals suggests an opportunity for improvement. “If the guidelines are correct, then perhaps interventionalists should be held accountable for making sure the heart team approach is followed,” he said. “Alternatively, perhaps a modified approach that guarantees patient-centered decision making and is ethically acceptable could be identified.”
The study received funding from the New York State Department of Health. Dr. Hannan and Dr. Blankenship and Dr. Patel have no relevant disclosures.