A Cross-comparison Between PMOD And HeartSee For Absolute Quantification Of Myocardial Blood Flow In PET Imaging

Objective:Absolute quantification of myocardial blood flow（MBF）by positron emission tomography（PET）has become one of the most advanced techniques in myocardial perfusion imaging（MPI）in recent years,as it is not only relevant for the diagnosis of single vessel lesions,but also valuable for the diagnosis of multivessel lesions and coronary microcirculation disorders（CMD）.¹³N-NH₃PET-MPI provides excellent accuracy for the diagnosis of CMD by absolute quantification of MBF and myocardial flow reserve（MFR）with good accuracy for the diagnosis of CMD.commonly used software for absolute quantification post-processing of MPI are PMOD and Heart See,with PMOD using a single tissue compartment model and Heart See using a simple preservation model considered to be a special single tissue compartment model.We investigated the consistency between software for absolute quantification of myocardial perfusion and the diagnostic efficacy of the software for coronary microcirculatory dysfunction by comparing the results of NOCA patients post-processed with PMOD and Heart See.Methods:In this study,69 patients with coronary angiographically confirmed non-obstructive coronary artery disease（angiography showing<70%stenosis）underwent one-day resting and stressing ¹³N-ammonia positron emission computed tomography（PET）myocardial perfusion imaging（MPI）.After completion of the images,they were processed by PMOD and Heart See software and compared for absolute quantification of MBF and myocardial flow reserve（MFR）and diagnostic performance of both software.The differences between resting and stressing MBF and MFR were compared by paired samples t-test and paired samples rank-sum test between PMOD and Heart See software.The area under the curve（AUC）of the subject’s operating characteristic curve（ROC）was used to establish the best intercept value between sensitivity and specificity of Heart See at different intercept points i.e.the intercept value of MFR versus stress MBF by PMOD’s ability to diagnose abnormal MFR≥2.0 and abnormal stress MBF≥2.0 ml/min/g.Where the optimal intercept value was determined based on the Yordon Index（YI）versus the AUC of the ROC curve.Results:1.Differences:Rest MBF was higher in PMOD than Heart See（1.02±0.22 vs.0.92±0.23,p<0.05）,but there was no statistical difference between stress MBF（2.64±0.54 vs.2.68±0.78,p=0.644）.MFR was higher in Heart See than in PMOD（2.96±0.73 vs.2.64±0.51,p<0.05）.While rest relative uptake was higher for Heart See than PMOD（77.73±3.60%vs.72.96±4.82%,p<0.05）,stress relative uptake was higher for Heart See than PMOD（78.94±2.86%vs.76.22±5.32%,p<0.05）.2.Correlation:Whereas rest MBF,stress MBF and MFR were significantly correlated between PMOD and Heart See（r:0.35-0.49,all p<0.05）,there was no correlation between rest and stress relative uptake in PMOD and Heart See（all p>0.05）.3.Diagnostic concordance:the AUC for Heart See identification of optimal MFR threshold by PMOD was 0.77（95%CI:0.64 to 0.89,p<0.001）.the ROC curve showed a Heart See intercept of 2.885 for predicting abnormal MFR by PMOD,while the identification of Heart See by stress MBF of PMOD threshold had an AUC of 0.63（95%CI,0.40 to 0.87,p>0.05）.When using the best intercept,Heart See correctly identified PMOD as 43 of 60 patients with normal MFR（MFR>2）（specificity of 72%）and correctly identified PMOD as 7 of 9 patients with abnormal MFR（MFR<2）（sensitivity of 78%）,with an accuracy of 72%.Conclusions:The absolute quantification values and the relative uptake values obtained by the PMOD and Heart See processing software are different,but there is a correlation between the absolute quantification of the two software.Furthermore,the diagnostic accuracy of Heart See for patients deemed abnormal by the PMOD software was 72%as assessed by the ROC curve and YI.