Prognostic significance of quantitative perfusion by cardiovascular magnetic resonance in dilated cardiomyopathy

Abstract

Background: Patients with dilated cardiomyopathy (DCM) have altered myocardial perfusion, but its prognostic significance is unknown. Objectives: To investigate whether measures of myocardial blood flow (MBF) and myocardial perfusion reserve (MPR) by stress-perfusion CMR in patients with DCM are independently associated with future major adverse cardiovascular events (MACE). Methods: Patients with DCM were recruited from three centres. Inclusion criteria were left ventricular (LV) dilatation, left ventricular ejection fraction (LVEF) < 50% and no clinical or CMR evidence of ischaemic heart disease. All patients underwent quantitative stress perfusion CMR and were followed up for incident MACE by review of electronic health records. The primary endpoint was defined as a composite of HF hospitalisation or death, based on the timing of the first event. The two secondary endpoints were the occurrence of HF hospitalisation or mortality. Association between stress MBF, rest MBF and MPR and the primary and secondary outcomes were assessed using Cox regression. Associations were adjusted for age, LVEF and the presence of non-ischaemic fibrosis on late gadolinium enhancement, which are all recognised prognostic factors in DCM. Results: Of 549 patients, the primary endpoint occurred in 86 (15.7%) patients (first event; n=49 HF hospitalisation, n=37 death) at median follow-up of 3.1 [1.8 – 4.8] years. Higher rest MBF (adjusted HR per 1SD increase 2.87, 95% CI 1.43 – 5.75, p = 0.003) and lower MPR (adjusted HR 0.57 per 1SD increase, 95% CI 0.44 – 0.74, p < 0.0001) were associated with the occurrence of the primary outcome, even after adjusting for age, LVEF and presence of nonischaemic LGE. Higher rest MBF was found to be associated with HF hospitalisation (adjusted HR 4.23, 95% CI 1.88 – 9.53, p = 0.0005), but not with mortality. While stress MBF initially showed a borderline significant association with mortality (unadjusted HR 0.63, 95% CI 0.39 – 1.01, p=0.056), this association became insignificant after adjusting for confounders. Older age (B coefficient -0.009, 95%CI -0.013 to -0.005, p < 0.0001), increased LVEDVi (B coefficient -0.003, 95%CI -0.005 to -0.002, p = 0.0001) and reduced LVEF (B coefficient 0.008, 95%CI 0.004 to 0.013, p = 0.0002) were all associated with reduced stress MBF, but not with increased rest MBF. Non-ischaemic fibrosis and pulmonary capillary wedge pressure were both associated with stress and rest MBF. Apart from diuretic use, no medication was associated with either stress or rest MBF. Conclusion: In patients with DCM, increased resting MBF is a powerful predictor of future HF hospitalisation, independent of LVEF, age, and fibrosis. MPR is independently associated with both mortality and HF hospitalisation. Whether these parameters can be altered by medical therapy remains to be established.