Publications

Tricuspid annular plane systolic excursion (TAPSE) is usually measured with M-mode using sector line, however, this may not align with the anatomical shortening of the right ventricular (RV). In this study, we compared the different methods to measure TAPSE using three different reference lines (sector line, anatomical line, and apico-annular line). We included 148 patients diagnosed with pulmonary arterial hypertension (PAH) who underwent TTE and right heart catheterization within 2 weeks of each other. TAPSE was measured by M-mode (sector, anatomical), 2D (sector, anatomical), or as tricuspid apico-annular displacement (TAAD). Agreement between measures was assessed using coefficient of variation (COV), Spearman's correlation, and Bland-Altman analysis. Receiver-operating characteristics and Kaplan-Meier analysis were used to explore associations with the combined outcome of death or lung transplantation at 5 years. There was a good concordance between anatomical and sector M-mode with a COV of 15.5 ± 1.6% and a bias of -0.6 ± 3.2 mm. In contrast, anatomical M-mode TAPSE and TAAD differed significantly with the mean difference of 3.3 ± 3.8 mm (COV 30.5 ± 6.1%;  < 0.0001). Among the different 2D methods, anatomical 2D agreed well with anatomical M-mode TAPSE (COV of 11.8 ± 2.0%;  = 0.89;  < 0.0001). Among the five methods, TADD had the strongest association with the combined endpoint of death or transplantation at 5 years (C-statistic 0.64, 95% confidence interval [CI] 0.57-0.71). We concluded that different measures of TAPSE are not interchangeable.

BACKGROUND

Myocardial blood flow (MBF) and myocardial perfusion reserve (MPR) using stress cardiovascular magnetic resonance (CMR) have been shown to identify epicardial coronary artery disease. However, comparative analysis between quantitative perfusion and conventional qualitative assessment (QA) remains limited.

OBJECTIVES

The aim of this multicenter study was to test the hypothesis that quantitative stress MBF (sMBF) and MPR analysis can identify obstructive coronary artery disease (obCAD) with comparable performance as QA of stress CMR performed by experienced physicians in interpretation.

METHODS

The analysis included 127 individuals (mean age 62 ± 16 years, 84 men [67%]) who underwent stress CMR. obCAD was defined as the presence of stenosis ≥50% in the left main coronary artery or ≥70% in a major vessel. Each patient, coronary territory, and myocardial segment was categorized as having either obCAD or no obCAD (noCAD). Global, per coronary territory, and segmental MBF and MPR values were calculated. QA was performed by 4 CMR experts.

RESULTS

At the patient level, global sMBF and MPR were significantly lower in subjects with obCAD than in those with noCAD, with median values of sMBF of 1.5 mL/g/min (Q1-Q3: 1.2-1.8 mL/g/min) vs 2.4 mL/g/min (Q1-Q3: 2.1-2.7 mL/g/min) (P < 0.001) and median values of MPR of 1.3 (Q1-Q3: 1.0-1.6) vs 2.1 (Q1-Q3: 1.6-2.7) (P < 0.001). At the coronary artery level, sMBF and MPR were also significantly lower in vessels with obCAD compared with those with noCAD. Global sMBF and MPR had areas under the curve (AUCs) of 0.90 (95% CI: 0.84-0.96) and 0.86 (95% CI: 0.80-0.93). The AUCs for QA by 4 physicians ranged between 0.69 and 0.88. The AUC for global sMBF and MPR was significantly better than the average AUC for QA.

CONCLUSIONS

This study demonstrates that sMBF and MPR using dual-sequence stress CMR can identify obCAD more accurately than qualitative analysis by experienced CMR readers.