Prognostic value of computational pressure-flow dynamics derived fractional flow reserve among diabetic versus non-diabetic patients presenting with acute coronary syndrome

Computational pressure-flow dynamics derived fractional flow reserve (caFFR) is a novel index to assess the severity of coronary artery stenoses without requirement of invasive pressure wire and hyperaemic stimulus in fractional flow reserve measurement. In this study, non-ST-elevation myocardial infarction (NSTEMI) and unstable angina are considered to be acute coronary syndromes (ACS). Approximately one-third of patients considered for coronary revascularisation have diabetes, which is a major determinant of clinical outcomes as patients are at greater risk of microvascular dysfunction. Whether prognosis differs between diabetic and non-diabetic patients presenting with ACS when revascularisation is guided by caFFR is unclear. 198 diabetic patients (mean age=71.5±10.6, 59.6% male) and 284 non-diabetic patients (mean age=65.1±12.7, 73.6% male) that presented with ACS were included and retrospectively analysed for caFFR. Patients had undergone PCI according to the angiographic and clinical data available at the time. Vessels were said to be adherent to caFFR if all ischemic vessels, defined by caFFR≤0.8 were treated by PCI and all nonischaemic vessels, defined by caFFR>0.8 were not treated by PCI. Otherwise, they were considered to be non-adherent to caFFR. The primary endpoint was the vessel-oriented composite endpoint (VOCE), defined as vessel-related cardiovascular death, vessel-related myocardial infarction (MI) and unplanned revascularisation at median follow-up time of 4.4 years. Among the total 1446 vessels, caFFR was analysable in 1020 (70.5%) vessels. The non-diabetic cohort had 449 (78.9%) vessels that were adherent and 120 (21.1%) vessels that were non-adherent to caFFR. The diabetic cohort had 353 (78.3%) vessels that were adherent and 98 (21.7%) vessels that were non-adherent to caFFR. Non-diabetic adherent vessels had significantly lower incidence rates of VOCE (4.5% vs 10.0%; P=0.019), cardiovascular mortality (2.7% vs 6.7%; P=0.035) and MI (0.9% vs 4.2%; P=0.011) than non-adherent vessels. In comparison, diabetic adherent vessels had significantly lower incidence rates of MI (3.4% vs 15.3%; P<0.001) than diabetic non-adherent vessels. After multivariate adjustment, non-diabetic adherent vessels had significantly lower risk than non-adherent vessels for VOCE (HR 0.42; 95%Cl 0.21-0.87; P=0.019), cardiovascular mortality (HR 0.38; 95%Cl 0.15-0.93; P=0.034) and MI (HR 0.21; 95%Cl 0.06-0.78; P=0.020). After multivariate adjustment, diabetic adherent vessels had a significantly lower risk of MI (HR 0.22; 95%Cl 0.10-0.47; P<0.001) than non-adherent vessels. Revascularisation adherent to caFFR guidelines was associated with superior prognosis in non-diabetic patients presenting with ACS compared to diabetic patients. Nonetheless, caFFR adherence is associated with a significant reduction in MI for diabetic patients. The findings support the potential use of caFFR to optimise treatment decision for patients presenting with ACS. Kaplan-Meier plots for VOCE Kaplan-Meier plots for vessel-related MI