The Prognostic Value Of Cardiac Magnetic Resonance Imaging In Patients With Severe Dilated Cardiomyopathy

Background and purpose: Patients with severe dilated cardiomyopathy(DCM)had severely reduced left ventricular ejection fractions(LVEF<35%),their systolic function of left ventricles were severely impaired.They are at higher risk of adverse cardiac events including end-stage heart failure and malignant arrhythmia compared to dilated cardiomyopathy patients with preserved ejection fractions.It is of great significance to identify and select high-risk individuals to perform prophylactic treatment to improve the overall prognosis of the population.LVEF has shown to be of great value in predicting outcomes in dilated cardiomyopathy patients with preserved ejection fractions.However,the classical parameter is of very limited value in risk stratification of the population.Myocardial strain derived from cardiac magnetic resonance(CMR)could accurately describe the motion features of myocardium.In addition,the late gadolinium enhancement(LGE)in late enhanced scanning is sensitive marker of myocardial fibrosis.The aim of this study was to evaluate the prognostic value of myocardial strain and LGE derived from CMR in predicting adverse events in DCM patients with severely impaired left ventricular systolic function.Material and methods: We retrospectively screened patients with dilated cardiomyopathy who underwent CMR at local institute and collected those with LVEF<35%.Patients were followed for endpoint events.A composite endpoint was defined as all-cause deaths,cardiac transplantation,implantable cardioverter-defibrillator(ICD)implantation and rehospitalization for heart failure.Baseline CMR images of included patients were analyzed to generate myocardial strain and parameters of LGE.Cox proportional hazards regression modeling was used to evaluate the association between CMR parameters including myocardial strain and LGE and adverse outcomes.Results: A total of 129 patients with severely impaired left ventricular systolic function who had a median LVEF of 15.33% were included.Among them,84 were male,45 were female,and their median age was 47 years.During a median follow-up of 518 days,endpoint events were observed in 50 patients(38.8%),including 13 all-cause deaths(26%),20 heart transplantations(40%),7 ICD implantations(14%)and 10 re-hospitalizations for heart failure(20%).There was no significant difference between patients with endpoint events and those without events in LVEF(15.49% vs.14.25%,p=0.125).There was significant difference between event-group and event-free groups in global circumferential strain(GCS)([-4.65±2.52]% vs.[-5.88±2.83]%,p=0.013),while no difference was found between groups in global longitudinal strain([-3.58±2.88]% vs.[-2.88±2.79]%,p=0.176)and global radial strain([7.19±4.30]% vs.[7.36±8.75]%,p=0.879).Although patients in event-group had higher LGE extent than those without adverse events,the presence of LGE was both prevalent in both groups(79.5% vs.86.7%,p=0.458).GCS and LGE extent were significantly associated with adverse outcomes in both univariate and multivariate Cox proportional hazards regression models.In addition,adding GCS into the model including clinical risk factors and the extent of LGE resulted in significant improvements in risk stratification(C-index from 0.706 to 0.742;P<0.05).Conclusion: DCM patients with severely reduced LVEF had significantly impaired myocardial strain values and prevalent LGE presence.GCS and LGE extent were significantly associated with adverse outcome,reduced absolute value of GCS and increased LGE extent were both associated with poorer prognosis.Background and purpose: Myocardial fibrosis is an important histopathological hallmark of dilated cardiomyopathy.It presents as late gadolinium enhancement(LGE)on cardiac magnetic resonance(CMR)imaging.Previous studies demonstrate that the presence of LGE is significantly associated with adverse endpoint events in patients with dilated cardiomyopathy.However,patients with dilated cardiomyopathy at mid and late stages not only had severely impaired systolic function,myocardial fibrosis was also prevalent among them.It is of very limited value for risk stratification by determining the presence and distribution pattern of LGE.However,the method of texture analysis can extract subtle features of images by quantitative analysis that cannot be observed nakedly.It can offer a wealth of information for diseases diagnosis and treatment.This retrospective study aimed to evaluate the prognostic value of texture features derived from late enhanced CMR images in dilated cardiomyopathy patients with severely reduced ejection fraction.Material and methods: We retrospectively screened patients with dilated cardiomyopathy who underwent CMR at local institute and collected those with severely impaired systolic function.Included patients were followed for endpoint events including all-cause deaths and heart transplantation.Texture features were extracted and selected from their baseline CMR late enhanced images.Cox proportional hazards regression modeling was used to evaluate the association between texture features and adverse outcomes.Results: A total of 114 patients with severely impaired left ventricular systolic function who had a median left ventricular ejection fraction(LVEF)of 14.0% were included.Among them,77 were male,37 were female,and their median age was 47.5 years.During a median follow-up of 504.5 days,29 patients(25.4%)experienced endpoint events including 12(41.4%)deaths and 17(58.6%)heart transplantations.There was no difference between patients with endpoint events and those without events in LVEF(13.7% vs.14.9%,p=0.873).The presence LGE(93.1% vs.80.0%,p=0.149)and distribution pattern of LGE(p=0.203)neither showed no differences between groups nor showed significant association to adverse outcomes.However,three texture features selected from gray level cooccurrence matrix(GLCM,GLCM_Contrast,GLCM_Difference Average and GLCM_Difference Entropy)showed significant associations with endpoint events(univariate Cox regression analysis p=0.007,0.011,0.007 respectively).In addition,when each of the three features was analyzed using a multivariable Cox regression model that included the clinical parameter(systolic blood pressure)and LGE extent,these features were independently associated with adverse outcomes.Conclusions: In dilated cardiomyopathy patients with severely reduced ejection fraction,texture features related LGE heterogeneities and variations derived from late enhanced CMR images(GLCM_contrast,GLCM_difference average and GLCM_difference entropy)are significantly associated with adverse events.Increased LGE heterogeneity is associated with higher risk of occurring adverse cardiac events.