In order for percutaneous coronary intervention (PCI) to shine compared with meds alone in patients with type-2 diabetes and stable coronary disease (CAD), it needs help from aggressive control of low-density-lipoprotein cholesterol (LDL-C) levels, suggests a patient-level meta-analysis of three major randomized trials.
Performing PCI in such patients with diabetes conferred further benefit over optimal medical therapy (OMT) for major adverse cardiac or cerebrovascular events (MACCE) only among those whose LDL-C levels had been pushed below the guidelines-specified threshold of 70 mg/dL within 1 year.
At that level of LDL-C control, PCI compared to the meds-alone strategy was followed by a nearly 40% drop in 4-year risk for the composite endpoint, which consisted of death from any cause or nonfatal myocardial infarction (MI) or stroke.
Also for patients reaching a 1-year LDL-C of <70 mg/dL, the risk of MACCE was similar for those who had been assigned to coronary bypass surgery (CABG) compared with PCI. But that risk was significantly lower for the CABG group among those reaching LDL-C levels above that threshold.
“The strategy of revascularization with the LDL lowering, that’s the combination that seems to be a winner” in such patients with diabetes and stable CAD, lead author Michael E. Farkouh, MD, MSc, told theheart.org | Medscape Cardiology.
If their LDL-C “stays above 70 mg/dL, they don’t really enjoy any benefit of PCI. It’s a message to our interventional community to really drive that LDL down,” said Farkouh, of the University of Toronto, Canada. “Not only with statins, but perhaps with PCSK9 inhibitors, ezetimibe, and other therapies to lower that LDL-C.”
The analysis, published November 2 in the Journal of the American College of Cardiology, pooled more than 4000 patients with diabetes and stable CAD randomized in the BARI 2D, FREEDOM, and COURAGE trials.
The new study adds a twist to an ongoing theme throughout some meta-analyses and clinical trials like ISCHEMIA since the results of COURAGE were unveiled 13 years ago. The latter trial famously saw no significant difference in death, MI, or stroke in patients with stable CAD assigned to OMT with or without PCI. That set off years of controversy about the relative merits of the revascularization and meds-only strategies in stable CAD that persists today.
But, Farkouh proposed, whether PCI improves clinical outcomes compared with meds alone, at least in patients with diabetes, may be tied to the success of LDL-C-lowering therapies in reaching that goal, which in the current study was below 70 mg/dL.
“In this analysis of pooled data from the three major trials, we demonstrate that attaining that level of LDL-C at 1 year portends a better outcome for PCI” in patients with diabetes and stable CAD, he said.
The findings “probably need to be studied further, but it is compelling to think that if we can drive the LDL-C down by one year after the procedure, we have better outcomes with PCI” compared with a meds-only strategy in patients with diabetes and stable CAD. “That really vindicates a lot of those who believe in PCI,” Farkouh said.
The main message from this observational analysis is that “of course, we want to get the LDL as low as possible in these patients with demonstrated cardiovascular disease and diabetes,” Donald M. Lloyd-Jones, MD, ScM, Northwestern University Feinberg School of Medicine, Chicago, Illinois, told theheart.org | Medscape Cardiology. “Every one of these patients should be shooting for as low an LDL as possible.”
Regardless of revascularization strategy, he said, “we have to get people on a high-intensity statin, or at least their maximally targeted dose, and have a careful and thoughtful conversation about whether they need additional lowering with, perhaps, ezetimibe, if they’re not below the thresholds we’d like to see them at, in this case, 70 mg/dL.”
Still, the current findings that the relative effects of PCI and CABG in these patients may vary by degree of LDL-C reduction “are interesting, but would have to be tested a little bit more directly,” said Lloyd-Jones, who is not affiliated with the analysis.
An accompanying editorial, which also acknowledges the study’s limitations, says its results “are relevant for clinical practice and may pave the way toward the generation of novel personalized medicine models that can optimize care of patients with type-2 diabetes.”
They “support the concept of an individualized treatment strategy that accounts for a patient’s LDL-C level to estimate clinical outcomes and expected treatment effects after therapeutic interventions,” say the authors, led by Eliano P. Navarese, MD, PhD, Nicolaus Copernicus University, Bydgoszcz, Poland.
“For daily practice, these results also underscore the importance of follow-up LDL-C measurements, both as a risk stratifier and as an indicator for therapy adjustments,” they write, noting that “current guidelines provide no formal recommendation on when to check LDL-C after PCI.”
The meta-analysis followed a total of 4050 patients with diabetes and stable CAD from the three randomized trials, those with evaluable baseline and follow-up LDL-C measurements, for a median of 4 years after the 1-year LDL-C assessment. At that time, at least 90% of patients in each of the trials had statin prescriptions, the group reported.
At one year, 34.5% of the total cohort had an LDL-C <70 mg/dL; their mean was 55.8 mg/dL.
And 42.2% had an LDL-C from 70 mg/dL to <100 mg/dL; their mean was 83.4 mg/dL. Compared with patients with an LDL-C <70 mg/dL, their adjusted hazard ratio (HR) for the composite endpoint was not elevated at 1.07 (95% CI, 0.86 – 1.32, P = .54).
Finally, 23.2% had an LDL-C ≥100 mg/dL; the mean was 123.0 mg/dL. Compared with the group with the lowest 1-year LDL-C, their adjusted HR for MACCE was increased at 1.46 (95% CI, 1.15 – 1.85, P = .002).
That HR among the 42.3% of patients in the PCI cohort, compared with the 33.3% assigned to meds only, climbed significantly only among those in the lowest 1-year LDL-C stratum: HR, 0.61 (95% CI, 0.40 – 0.91, P = .016). Corresponding HRs in the mid-range and highest 1-year LDL strata were close to unity and nonsignificant at P = .71 and P = .98, respectively.
On the other hand, the 24.4% of patients assigned to CABG showed better MACCE outcomes than those in the meds-only group across all three 1-year LDL-C strata.
The risk of MACCE wasn’t significantly altered by CABG compared with PCI among patients achieving a 1-year LDL-C less than 70 mg/dL. However, it fell by about one-half for CABG vs PCI in both the mid-range and highest 1-year LDL-C strata, P = .003 and P = .022, respectively.
It’s possible, Lloyd-Jones said, that there is “no longer a dichotomy between revascularization strategies,” with respect to clinical outcomes, in such patients who maintain an LDL less than 70 mg/dL, as the study suggests.
“But I wonder, if it had continued for another 4 years of follow-up, whether we would see the CABG patients start to have more events,” such that the CABG advantage goes away at higher LDL-C levels, he proposed.
Or, Lloyd-Jones speculated, if all patients had achieved LDL-C below 70 mg/dL, “would there be such a difference between the PCI and CABG groups? My bet would be that it would be small or abolished.”
Farkouh discloses receiving research grants from Amgen, Novo Nordisk, and Novartis. Disclosures for the other study authors can be found with the original article. Editorialist Navarese discloses receiving consulting fees or honoraria from Abbott, AstraZeneca, Amgen, Bayer, Sanofi, and Pfizer; and grants from Abbott and Amgen. Disclosures for the other editorialists can be found in the article. Lloyd-Jones has disclosed no relevant financial relationships.
J Am Coll Cardiol. Published online November 2, 2020. Full text, Editorial