Diabetes Does Not Appear to Hinder FFR Insights for CAD
Concerns that the test might be unreliable or function differently in diabetic patients aren’t warranted, observational data suggest.
Fractional flow reserve (FFR) is just as safe at guiding coronary artery disease management decisions in patients with diabetes as it is in those without the disease, according to a new observational analysis by the PRIME-FFR study group.
What’s not safe, the researchers found, is ignoring the advice of physiological testing: when FFR results were disregarded, patients had higher MACE risk at 12 months irrespective of their diabetes status.
Approaches to optimize treatment decisions for diabetic patients with cardiovascular disease are needed, but “concerns about microcirculatory responsiveness and the potential for accelerated atherosclerosis have cast doubts on the usefulness of FFR among patients with diabetes,” senior author Luís Raposo, MD (Hospital de Santa Cruz–Centro Hospitalar de Lisboa Ocidental, Carnaxide, Portugal), and colleagues write in a recently published JAMA Cardiology paper. “Data on this subject are scarce, derived from a limited number of retrospective small cohorts, and findings are conflicting.”
Sunil Rao, MD (Duke University Medical Center and Durham VA Medical Center, Durham, NC), commenting on the results for TCTMD, said he suspects that FFR use is comparatively lower in this population. “Diabetes does impair the microcirculation, and thus hyperemia may be less reliable” in this context, he explained, and “patients with diabetes tend to have diffusely diseased and smaller coronary arteries.”
Rao pointed out that prior research on visual assessment versus quantitative coronary angiography has shown “that the smaller the coronary artery, the more severe a stenosis appears,” raising the possibility that here, too, with FFR in diabetic patients fewer lesions would be deemed “intermediate” and more would be classified as “severe.”
Yet the data from PRIME-FFR suggest that these concerns aren’t warranted, said Rao. “Like the other studies examining the use of coronary physiology, this study shows that FFR is a great tool to assist in decision-making” in patients who have an indication for angiography and an intermediate lesion.
Two Registries Combined
For their cross-sectional study, investigators analyzed data from the PRIME-FFR study, which itself was derived from two prospective multicenter registries sharing a common design: POST-IT (March 2012-November 2013) and R3F (October 2008-June 2010).
Among the 1,983 all-comers patients (mean age 65 years; 77% men), approximately one-third had diabetes. Patients with diabetes tended to have more comorbidities, were more likely to have an LVEF ≤ 50%, were more apt to receive secondary prevention drugs, and had more complex disease.
FFR was performed on average for 1.4 lesions per patient. In both groups, the researchers identified the same independent risk factors for lower FFR values: age, left anterior descending artery location, American College of Cardiology/American Heart Association lesion type, stenosis percentage, lesion length, and number of diseased vessels.
Similar proportions of the diabetes and no-diabetes groups were reclassified by FFR results from one treatment to another (41.2% and 37.5%, respectively; P= 0.13). Patients with diabetes, however, were more likely to shift from medical therapy to revascularization (41.5%) compared to patients without diabetes (31.5%; P = 0.001).
One-year rates of MACE (all-cause death, MI, or unplanned revascularization) were higher overall for diabetic versus nondiabetic patients (11.3% vs 9.0%).
Rates did not differ between patients who were and weren’t reclassified by FFR (9.7% vs 12.0%; P = 0.37). Patients who deferred revascularization based on FFR > 0.80 had similar MACE rates regardless of whether they did or didn’t have diabetes (8.4% vs 7.9%; P = 0.87). By comparison, diabetic patients who underwent revascularization at the outset had a 1-year MACE rate of 13.1%. Insulin treatment did not affect these patterns, nor did multivariable adjustment for potential confounders.
FFR results were disregarded for 6.6% of the overall population; this subset saw a 1-year MACE rate double that of patients whose management was informed by FFR (17.5% vs 9.2%; P = 0.002).
“Our results, in this large multicenter international cohort, are very reassuring. As such, clinicians and interventionalists managing [diabetic] patients should definitely be encouraged to use and rely on FFR for risk stratification and decision-making,” the researchers told TCTMD via email. Also noteworthy, they added, is that “these observations speak to the fact that even if hyperemia may be somewhat hampered, this may have little clinical significance in the average patient with diabetes undergoing invasive physiologic assessment using FFR.”
If you’re going to do the test, you should know how to interpret it and act on it. Sunil Rao
Four co-authors prepared the comments on behalf of the PRIME-FFR group: Raposo, Eric Van Belle, MD, PhD (CHU de Lille, France); Sergio Bravo Baptista, MD, PhD (Hospital Prof Doutor Fernando da Fonseca, Amadora, Portugal); and Patrick Dupouy, MD (Hôpital Privé d'Antony, Antony, France).
To the best of their knowledge, they said, no dedicated randomized trials have been done on the potential for FFR in diabetic patients, though these individuals have been included in the seminal RCTs of physiological testing. In terms of real-world use, around 20% to 30% of CAD patients who undergo invasive evaluation have diabetes, the investigators estimated. “But we do not know how the presence of diabetes is ultimately influencing the way operators perform FFR.”
Rao offered some advice that would apply regardless of whether patients have diabetes, given the high MACE rates seen when FFR results went unheeded. “If you’re going to do the test, you should know how to interpret it and act on it,” he stressed.
Van Belle E, Cosenza A, Baptista SB, et al. Usefulness of routine fractional flow reserve for clinical management of coronary artery disease in patients with diabetes. JAMA Cardiol. 2020;Epub ahead of print.
- Van Belle reports receiving personal fees from Abbott (St. Jude Medical) and Philips Volcano during the conduct of the study.
- Baptista reports receiving grants from Abbott during the conduct of the study and receiving personal fees from Abbott, Boston Scientific, HeartFlow, and Opsens Medical outside the submitted work.
- Raposo reports receiving grants, personal fees, and nonfinancial support from Abbott (St. Jude Medical) during the conduct of the study.
- Dupouy and Rao report no relevant conflicts of interest.