Quantitative Research Of Salvageable Zone Of Acute Myocardial Infarction Based On MR-IVIM Model

Objective:To investigate the feasibility of applying intravoxel incoherentmotion（IVIM）to heart examination of guizhou miniature pigs.To investigate the feasibility of MRI technologybasedonintravoxelincoherentmotiontoquantitativelyevaluate microcirculation in guizhou miniature pig acute myocardial infarction model.Methods:1.CMR-IVIM scans were performed on eight healthy miniature pigs.Before MRI examination,experimental animals should fast for 12 hours and abstain from water for 6 hours.2.After successful intramuscular anesthesia,the experimental animals were cleaned,skinned and scanned by GE3.0TMR（Discovery MR750W）.FIESTA sequence（fast imaging employ steady state quisition）was used to perform the standard second,third and fourth entricular and left ventricular short axis（covering the apex to the bottom of the heart）cine.IVIM imaging was performed on the left ventricular apex,middle and bottom of the left ventricle at the short axis level.Trigger delay（TD）is the time value of the relative minimum amplitude of myocardial motion on the left ventricular short axis cine.3.The grade of IVIM images of the heart was evaluated by two senior physicians who have been engaged in cardiovascular imaging diagnosis for a long time with a double-blind method.Raw data were transferred to GE post processing station（AW 4.6）for quantitative measurement of corresponding parameters,and the related parameters were measured for the second time at intervals of more than 1 month for those with higher seniority.4.Ten healthy miniature pigs were selected and divided into two groups before and after AMI modeling:group A was the pre-modeling,and group B was the post-modeling.Group A followed part 1 of the cardiac MRI scan.Raw data were transferred to GE post-processing workstation（AW 4.6）for parameter measurement.After selecting ROI,ADC_slow（D）,ADC_fast（D*）and f were automatically generated to establish the normal standard.5.After CMR-IVIM scanning,the remote end or branch of LCX was embolized to establish AMI model.After successful modeling,cardiac IVIM imaging scan was performed within 1-2hours.The FIESTA sequence was the same as the first part.6.Raw data were transferred to GE post-processing workstation（AW 4.6）,and pseudo-color images of each parameter were obtained.After delineating the myocardial infarction and the myocardial interest area around the infarction,ADC_slow（D）,ADC_fast（D*）and f were generated automatically.Results:1.The Kappa coefficient between image analysts was 0.79 on the IVIM image.A total of 24 layers of IVIM imaging images were obtained for 8 miniature pigs,and the image quality meeting level II and III was deemed as the success of IVIM imaging.The image quality of 6 layers in the 24 layers can not meet the requirements,and the image quality of 18 layers can meet the requirements of level II or level III image quality.The total success rate was 75.0%（18/24）,level II 54.17%（13/24）and level III 20.83%（5/24）.2.The left ventricular myocardial parameters obtained by two quantitative measurements and two quantitative measurements by the same physician were consistent（ICC>0.75）.3.A total of 48 layers of IVIM imaging images were obtained for the 10 miniature pigs,and a total of 25 layers of IVIM images reached level II or level III image quality requirements.Image quality grade evaluation has good consistency（Kappa=0.86）.4.The slower the heart rate,the higher the image success rate（P<0.05）.There was no significant difference in body weight and gender at all levels（P>0.05）.5.The same physician repeated two quantitative measurements of left ventricular myocardial parameters with good consistency（ICC>0.75）.Quantitative measurement of left ventricular myocardial parameters by two physicians showed good consistency（ICC>0.75）.6.Pairwise comparison of ADC_fast（D*）in myocardium supplied by different coronary arteries:there were statistical differences between RCA and LCX,LAD and LCX,among which the ADC_fast（D*）of LCX was the smallest,90.96±70.95×10^-3.There was a statistically significant difference in f between RCA and LCX（P<0.05）,but no statistically significant difference in f between LAD and RCA,LAD and LCX（P<0.05）.There was no significant difference in ADC_slow（D）among the three groups（P>0.05）.7.Both the ADC_fast（D*）of the infarcted myocardium and the peripheral area of the infarcted myocardium were lower than the normal area（P<0.05）,while the ADC_fast（D*）of the infarcted myocardium was lower than the peripheral area of the infarcted myocardium,the difference was statistically significant（P<0.05）.The value of f in the normal group was higher than that in the peripheral area of myocardial infarction,and the difference was statistically significant（P<0.05）.There was no statistically significant difference in f between the infarcted myocardium and the surrounding area（P>0.05）,and no statistically significant difference in f between the infarcted myocardium and the normal group（P>0.05）.There was no significant difference in ADC_slow（D）among the three groups（P>0.05）.ROC curve analysis showed that the accuracy of ADC_fast（D*）in evaluating the disease group was higher than that of ADC_slow（D）and f.Conclusions:The cardiac IVIM imaging technology is feasible for the heart scanning of miniature pigs.The quantitative parameters of IVIM of myocardium supplied by the three branches of coronary artery in healthy miniature pigs were different.The evaluation efficiency of ADC_fast（D*）on microcirculation perfusion information of acute myocardial infarction is better than other parameters in the quantitative evaluation of cardiac IVIM imaging,which can provide a new diagnostic idea for clinical evaluation of acute myocardial infarction or coronary artery thrombolysis.