| PurposeTo explore the feasibility of cardiac magnetic resonance intravoxel incoherent motion imaging(CMR-IVIM)in vivo,and the repeatability of the IVIM related measurement parameters.In addition,we also wanted to preliminarily investigate the role of CMR-IVIM in evaluation of cardiac diseases.Materiel and methodsSixty-four volunteers and patients who were performed CMR-IVIM on 3.0T MR scanner from May 2015 to January 2016 in the First Affiliated Hospital of Dalian Medical University,were enrolled in this study,including thirty-four healthy volunteers(nineteen male,fifteen female,age 8-76 years old)and thirty patients with heart damage related diseases(fifteen male,fifteen female,aged 37-77 years old),following,ten patients with hypertrophic cardiomyopathy and twenty patients with diabetes and/or hypertension.They all were performed CMR-IVIM on the apex,middle and base of left ventricular short axis respectively using multi b values(0,20,50,80,100,120,200,300,500 s/mm~2).The quality of IVIM images and the repeatability of IVIM parameters,including ADCslow,ADCfast and f(fraction of ADCfast)values,were assessed by two cardiac radiologist in double-blind method.And then the more experienced cardiac radiologist quantitatively repeated measurement of relevant parameters in an interval of more than one months.All data was analyzed by GraphPad Prism 5 and SPSS 17.0 software.Interreader agreement on image quality was assessed by Kappa test.For discrete variables(rate),chi-square test was performed.Different influence factors were analyzed by independent sample t-test and Wilcoxon test.Interobserver and intraobserver repeatability on quantitative IVIM parameters were assessed with intraclass correlation coefficient.The analysis of various IVIM parameters among normal group,different patient groups were used independent sample t-test,Wilcoxon signed rank test,analysis of variance(ANOVA)and Kruskal Wallis H rank signed rank test.P < 0.05 was defined as statistical significance.Results1.Among sixty-four cases,one hundred and thirty-one CMR-IVIM images of fifty-one cases were successfully obtained finally.These images accorded with II or III grade evaluation standard and could be quantitatively measured.Based on myocardial imaging slices,the overall success rate of CMR-IVIM was 68.23%(131/192).Among them,the normal group was 74.51%(76/102),and the patient group was 61.11%(55/90).The success rate of normal group was higher than patient group(P < 0.05).The success rate of CMR-IVIM among apex,middle and base slices of left ventricular short axis had no statistical difference(P > 0.05).Heart rate was obvious higher in failure group than that in successful group(P < 0.05).2.The consistency on the evaluation of CMR-IVIM images quality between two doctors was good(Kappa = 0.80).Whether the normal group or patient group,we found that the intero-bserver and intrao-bserver consistency for every left ventricular myocardial IVIM parameters(ADCslow,ADCfast and f values)were good(ICC > 0.80).3.In normal group,the CMR-IVIM parameters had no significant difference among apex,middle and base short axis in normal group(P > 0.05).The ADCfast value in patients with hypertrophic cardiomyopathy and patients with diabetes and/or hypertension were significant lower than that in normal group(83.12±22.33×10-3 mm~2/s,103.02±21.10×10-3 mm~2/s vs 132.43±20.23×10-3 mm~2/s,P<0.05).ADCfast value hadn`t find any statistical difference between patients with diabetes and/or hypertension and patients with hypertrophic cardiomyopathy(P > 0.05).There were not any statistical differences between normal group and patient group in ADCslow value and f value(P > 0.05).ConclusionWe preliminarily verify that CMR-IVIM technique could be used in vivo heart and could be used in clinic.And ADCfast value could respond myocardial microcirculation changes in heart damage related diseases and could be used to evaluate and predict myocardial microcirculation.But,CMR-IVIM still had some challenge. |
PDF Full Text Request