Frame counting improves the assessment of post-reperfusion microvascular patency by TIMI myocardial perfusion grade: Evidence from cardiac magnetic resonance imaging

Abstract

BackgroundQuantitative modification of TIMI myocardial perfusion grade (TMPG) by the method of frame counting may improve its sensitivity and the false negative rate for post-reperfusion microvascular dysfunction (MVD) in ST segment-elevated myocardial infarction (STEMI) patients. MethodsThe durations of contrast-washout from infarction area of 139 patients were measured by counting the cine-frame numbers between the appearance and disappearance of myocardial blush. The achieved new index, TMP Frame Counting (TMP-FC) was referenced by cardiac magnetic resonance, by which MVD was defined as microvascular obstruction on gadolinium late-enhancement imaging. ResultsMedian TMP-FC differed significantly between patients with and without MVD (126 frames, IQR 105-160 vs. 86 frames, IQR 75-100, p<0.001). By receiver-operating characteristic analysis, the cutoff of TMP-FC at ≥95.5 frames represented an independent predictor of MVD (OR=11.61, p<0.001). TMP-FC had similar specificity (75%) and positive predictive value (88%), but significantly improved sensitivity (85.3%) and negative predictive value (70.2%) for MVD compared with TMPG (88.6%, 86.5%, 33.7% and 38.2%, respectively) and other traditional angiographic assessments, leading to a better overall accuracy (area under the curve: 0.801 compared with 0.612 from TMPG, p<0.001) for the evaluation of microvascular patency. TMP-FC was positively correlated with MVD extent (r=0.5, p<0.001). Abnormal TMP-FC was associated with larger infarction size (28.67±13.72% vs. 16.51±10.68% of left ventricular mass, p<0.001) and lower LVEF (49.37±11.06% vs. 56.84±9.72%, p<0.001). ConclusionFrame counting can improve the accuracy of TMPG for MVD. Moreover, TMP-FC is correlated with the degree of MVD and cardiac detriments, which is useful for risk stratification.