Procedural MI rates during revascularization vary widely based on definition
In two trials comparing outcomes of PCI with CABG, rates of procedural MI varied greatly depending on which definition was used, according to two analyses published in the Journal of the American College of Cardiology.
In the EXCEL trial, rates of procedural MI were higher in the PCI group when the prespecified protocol definition was used, but higher in the CABG group when the third universal definition of MI was used. In the SYNTAXES long-term study, rates of procedural MI varied across five definitions. The association between procedural MI and mortality also varied across definitions and studies.
“There might need to be acceptance that the same procedural MI definition cannot be fit to both procedures,” Donald E. Cutlip, MD, vice chair of the department of medicine at Beth Israel Deaconess Medical Center and professor of medicine at Harvard Medical School, wrote in an editorial related to both papers.
In an analysis from the EXCEL trial comparing PCI and CABG in patients with left main CAD and low or intermediate SYNTAX scores, John Gregson, PhD, research associate at the London School of Hygiene and Tropical Medicine, and colleagues compared procedural MI rates between the prespecified protocol definition, which was creatine kinase-MB rise to more than 10 times the upper reference limit as a stand-alone measure, or more than five times the upper reference limit with supporting electrocardiographic, angiographic or imaging evidence of myocardial ischemia, and the third universal definition of MI types 4a and 5.
As Healio previously reported, the EXCEL investigators were accused of improper conduct for not reporting data based on the third universal definition of MI. Healio published a response from EXCEL trial leadership.
According to the prespecified protocol definition, procedural MI was lower in the PCI group compared with the CABG group (3.6% vs. 6.1%; difference, –2.4 percentage points; 95% CI, –4.4 to –0.5; P = .015), but according to the third universal definition of MI, procedural MI was higher in the PCI group than the CABG group (4% vs. 2.2%; difference, 1.8 percentage points; 95% CI, 0.2-3.4; P = .025), the researchers wrote.
Procedural MI according to the prespecified protocol definition was associated with CV mortality at 5 years (adjusted HR = 2.18; 95% CI, 1.13-4.23), as was procedural MI according to the third universal definition (aHR = 2.87; 95% CI, 1.44-5.73), Gregson and colleagues wrote.
However, they wrote, procedural MI according to the prespecified protocol definition was consistently associated with CV mortality regardless of which procedure patients received (P for interaction = .86), whereas procedural MI according to the third universal definition was associated with CV mortality in the CABG group (aHR = 11.94; 95% CI, 4.84-29.47) but not in the PCI group (aHR = 1.14; 95% CI, 0.35-3.67; P for interaction = .004).
In addition, they wrote, “Only large postprocedure biomarker elevations (peak CK-MB 10 times the upper reference limit or more and troponin 70 times the upper reference limit or more) were associated with subsequent mortality.”
Hironori Hara, MD, cardiologist at Academic Medical Center, University of Amsterdam, and colleagues compared procedural MI rates and long-term outcomes of patients from the SYNTAXES extended study following the SYNTAX trial according to five definitions; those used in the EXCEL, ISCHEMIA and SYNTAX trials; the fourth universal definition; and the Society for Cardiovascular Angiography and Interventions definition.
Procedural MI rates were similar based on the SYNTAX trial definition (PCI, 2.7%; CABG, 2.4%; P = .756) and the fourth universal definition (PCI, 3%; CABG, 2.1%; P = .281), but were slightly higher in the CABG arm based on the ISCHEMIA trial definition (PCI, 6%; CABG, 8.8%; P = .03) and much higher in the CABG arm based on the EXCEL and SCAI definitions (PCI, 5.7% for both; CABG, 16.5% for both; P < .001), the researchers wrote.
At a median follow-up of more than 11 years, procedural MI according to the SYNTAX definition was associated with higher all-cause mortality in both groups, but after adjustment for baseline characteristics, was a predictor for 1- and 10-year mortality in the PCI group but only 1-year mortality in the CABG group. Similar trends occurred for procedural MI based on the ISCHEMIA definition and the fourth universal definition.
However, procedural MI based on the EXCEL and SCAI definitions was associated with elevated 1- and 10-year mortality risk in the PCI group but not in the CABG group, according to the researchers.
“The SYNTAX and EXCEL analyses support a requirement for the supporting criteria proposed by the universal definition of MI after CABG,” Cutlip wrote in his editorial. “However, the supporting criteria as proposed by the universal definition of MI after PCI are not useful and should be removed in favor of a stand-alone biomarker threshold that is meaningful. Determining that threshold remains a topic of debate.”
- Cutlip DE. J Am Coll Cardiol. 2020;doi:10.1016/j.jacc.2020.08.024.
- Hara H, et al. J Am Coll Cardiol. 2020;doi:10.1016/j.jacc.2020.08.009.