Myocardial Extracellular Volume Fraction From Late Iodine Enhancement With Spectral Detector Computed Tomography For Risk Stratification In Non-ischemic Heart Failure

ObjectivesClinically,myocardial extracellular volume fraction(ECV)with computed tomography(CT)is rarely used for the quantification and risk stratification of diffuse myocardial fibrosis in patients with non-ischemic heart failure(NIHF)as compared with cardiac magnetic resonance(CMR)-ECV.To validate the feasibility of using late iodine enhancement(LIE)-derived ECV on iodine density images using spectral detector computed tomography(SDCT;CT-ECV)and to assess the potential of CT-ECV for risk stratification among patients with NIHF.MethodsThis study prospectively enrolled collected 235 patients with known or suspected heart disease who underwent cardiac SDCT examination in our hospital between December 2020 and March 2023.Forty-five subjects who underwent cardiac SDCT and CMR were included in the validation group to calculate and compare CT-ECV with CMR-ECV to validate CT-ECV feasibility.Another 117 subjects(82patients with NIHF,35 controls)who underwent SDCT were further included in the experimental group,and patients with NIHF were first divided into three subgroups based on LV ejection fraction(LVEF),including HFr EF,HFmEF,and HFp EF.In addition,per the NYHA Classification system,we also divided the patients with NIHF into four subgroups: NYHA Classes Ⅰ,Ⅱ,Ⅲ,and Ⅳ.In addition,enrolled patients with NIHF were followed up and the primary clinical endpoint of follow-up was establish,the first occurrence of a major adverse cardiovascular event(MACE),which included hospital admission for heart failure and all-cause mortality,in order to explore the potential of CT-ECV for risk stratification.CT-ECV and CMR-ECV was measured via iodine density images and CMR T1 mapping in accordance with American Heart Association 16-segment models.CT-ECV and CMR-ECV were compared using the paired t-test.Agreement between CT-ECV and CMR-ECV was assessed using Bland-Altman analysis.Differences in CT-ECV between the groups were assessed with one-way analysis of variance and the unpaired t-test.ROC curve and Yoden index determine the optimal cut-off value of CT-ECV.Kaplan-Meier curve and log-rank test were used to analyze the relationship between MACE and CT-ECV of patients with NIHF.P < 0.05 defined as statistically significant.Results validation group:1.Segment-based analysis showed that there was no significant difference between CT-ECV and CMR-ECV from the first to the sixteenth segment of the LV myocardium(all P > 0.05).2.Layer-based analysis showed that the differences between CT-ECV and CMR-ECV with respect to the basal,mid,and apical layers of the LV myocardium were not statistically significant(basal: 33.27 ± 4.37% vs.32.53 ± 5.47%,P > 0.05;mid: 33.07 ±4.32% vs.32.46 ± 4.82%,P > 0.05;apical: 32.54 ± 5.08% vs.33.04 ± 5.35%,P > 0.05).3.Global-based analysis showed that there was no significant difference between CT-ECV and CMR-ECV(33.10 ± 4.22% vs.32.80 ± 4.90%;P = 0.293)4.CT-ECV was well correlated with CMR-ECV(r = 0.926;P < 0.05)with a small bias(–0.30%;95% confidence interval [CI]: –3.97% to 3.38%)5.CT-ECV was well correlated with CMR-ECV(r = 0.926;P < 0.05)with a small bias(–0.30%;95% confidence interval [CI]: –3.97% to 3.38%).Experimental group:1.CT-ECV in patients with NIHF was higher than in controls(32.08 ± 4.06% vs.27.63± 2.55%;P < 0.05)2.Subgroup-based analysis of patients with NIHF showed that CT-ECV in HFr EF(HFr EF: 34.04 ± 4.01%)patients was higher than that of HFmEF(HFmEF: 31.80 ±3.18%)and HFp EF(HFp EF: 30.72 ± 3.79%)patients,there were significant difference,P< 0.05.3.CT-ECV differences between patients with NIHF with varied NYHA Classes(NYHA Ⅰ: 28.33 ± 2.28%,NYHA Ⅱ: 30.60 ± 2.65%,NYHA Ⅲ: 32.87 ± 3.24%,and NYHA Ⅳ:38.30% ± 2.76%;all P < 0.05).4.CT-ECV was negatively correlated with LVEF(r = –0.402;P < 0.05)and positively correlated with NYHA Classification(r = 0.687;P < 0.05).5.The ROC curves demonstrated CT-ECV ≥ 31.29 % to be the optimal cut-off point for MACE with 83.9% sensitivity,75% specificity and the area under the ROC curve= 0.863(95% CI 0.782 to 0.944).6.Kaplan–Meier survival curves and Log-rank test demonstrated that NIHF patients with CT-ECV ≥ 31.29% had higher probability of MACE than NIHF patients with CT-ECV <31.29%(P < 0.001).Conclusions1.LIE-derived CT-ECV on iodine density images using SDCT is a promising practical alternative to CMR-ECV.2.LIE-derived CT-ECV on iodine density images using SDCT have the potential to assist with risk stratification among patients with NIHF.