The Application Of Automatic Quantification Of Epicardial Fat Volume On Non-enhanced Cardiac CT Scans In Risk Classification Of Cardial Vascular Disease And Evaluation Of Type 2 Diabetes Prognosis

Objective To investigate the repeatability of the intelligent quantitative software cardiac risk,and to explore the association between the epicardial adipose tissue(EAT)and coronary artery diseases(CAD)in patients with low coronary artery calcification scores(CACS),the control of the blood sugar of the type 2 diabetes mellitus(T2DM)and the myocardial infiltrationMaterial and methods 30 patients underwent coronary artery CT examination were included in part one.Exclusive criteria:1)bad images not enough for the evaluation 2)with a history of heart operation(stent or OPCAB),of severe alular disease or pericardial effusion,3)with a history of obvious consumptive disease like tumors.Two experienced radiologists measured the volume of EAT on two software twice.Kappa test was used to analyze the repeatability of the inner-observer and the inter-observer.Bland-Altman plot test was used to analysis the difference between different software.The Pearson's test was used to compare the correlation between the EAT volume measured by two software 421 patients underwent coronary artery CT examination with CACS<100 were included in part 2 and divided into 2 groups according to their coronary stenosis.Exclusive criteria was just like part1.Their age,sex,BMI,whether complicated with the type II diabetes(T2DM)and EAT volume were compared with independent-samples T test and the Cross table chi-square test accordingly.The receiver operating characteristic(ROC)curve was used to compare the diagnostic value of these factors.Adding risk factors to find out the improvement of the risk classification155 type 2 diabetes mellitus patients(T2DM)underwent coronary CT angiography(CCTA)scanning were included in part 3.Exclusive criteria was just like part 1 except for using insulin to help control the glycemic control.According to the clinical experience,the patient' s condition was considered as under control when the glycosylated globin(HbAlc)<7%.Otherwise the condition was considered as not well controlled.The EAT volume was automatically measured on unenhanced CT images.The coronary stenosis for each patient was verdicted into two groups,one as normal or mild stenosis(?50%),another one as moderate or severe stenosis(>50%)by two experienced radiologists based on the enhanced CCTA imaging.Besides,the clinical information i.e.,body mass index(BMI),the high density lipoprotein(HDL),the total triglyceride(TG),the free fatty acid(FFA)and the total cholesterol(TC)were also collected.Continuous variables were statistically analyzed using independent-samples T test.The category variables were analyzed using chi-square test.The(receiver operating characteristic curve)ROC was used to analyze the predicting value of EAT volume in glycemic control and CADs.387 patients underwent coronary CT angiography(CCTA)scanning was included.Exclusive criteria was just like part1.The EAT volume was automatically measured on unenhanced CT images.The coronary stenosis for each patient was verdicted into two groups,one as normal or mild stenosis(?50%),another one as moderate or severe stenosis(>50%)by two experienced radiologists based on the enhanced CCTA imaging.Patients were grouped into two groups according to whether having myocardial fatty infiltration based on the CT images,if he had the fatty infiltration in the right ventricular free wall(RVFW),then the radiologist would measure the mean and the max thickness of the RVFW,and assess the infiltration degree of the adipose tissue.Besides,the clinic information i.e.,body mass index(BMI),age,sex and whether to have type 2 diabetes were also collected.Continuous variables were statistically analyzed using independent-samples T test.The category variables were analyzed using chi-square test.The(receiver operating characteristic curve)ROC was used to compare the EAT volume and some clinical indexes.The spearman analysis was used to test the correlation between the indexes and infiltration degree.Results The kappa value of two observers were respectively 0.967 and 0.958 by using the software Cardiac risk,0.940 and 0.937 by using the software Volume.The kappa of inter-observer was 0.953 by using the software Cardiac risk,0.910 by using the software Volume.EAT volume measured by the Cardiac risk was 21.9/21.3ml more than the volume measured by the software Volume.Although the two were highly correlated with r=0.984.Cardiac risk takes less time than Volume.EAT volume was related with the CAD statistically(P<0.05)and was an independent risk factor of CAD in patients with low calcific scores.The EAT volume was the best index among these risk factors in predicting CAD.The CAD risk improved markedly with more risk factors.The EAT volume is significantly associated with the glycemic control(p=0.031).After excluding patients lacking insulin secretion,the correlation between the EAT and the glycemic control seems more significant(p=0.001)and the or value,the sensitivity,specificity were all improved.EAT volume is significantly associated with the glycemic control(p=0.011).The right ventricular free wall thickness(RVWT),EAT volume,age,sex(P all<0.001),and T2DM(P<0.05)were all related with the myocardial fatty infiltration(MFI).CAD was not correlated with the MFI.The RVWT was moderate correlated with the MFI degree and the EAT volume was not correlated with the degree.Conclusion1.The repeatability of Cardiac risk is as good as Volume;The volume tested by Cardiac risk is bigger than Volume.The two are highly correlated;Cardiac risk saves time.2.EAT volume is an independent risk factor of CAD in low CACS patients.It can become an supplement for CACS in risk classification.3.EAT volume is related with the glycemic control in T2DM patients,especially in patients with normal insulin function;.EAT volume is related with CAD in T2DM patients4.The age,sex,the EAT volume,T2DM,but not with the CAD is relates with the IMF;EAT is not correlated with the degree of IMF.