Semi-automated quantification of mitral valve regurgitation by 3D color. Comparison with cardiac magnetic resonance imaging

Objective: To evaluate the accuracy of mitral regurgi-tation (MR) volume (RVol) quantification using a new semi-automatic 3D proximal isovelocity surface area (PISA) software (QFlow) against 2D TEE and car-diac magnetic resonance imaging (cMR). MR-RVol by 2D-PISA is the preferred quantitative method for assessing MR severity using transthoracic (TTE) and transesophageal (TEE) echocardiography. However, its reliability and accuracy may be limited by underlying geometric assumptions, while 3D TEE might be more accurate.

Methods: 51 consecutive patients (mean age 63 ± 16 years, 35 male) prospectively underwent TTE, TEE and cMR for MR evaluation. RVol by Qflow-3D-TEE was compared to 2D TTE and TEE, as well as cMR, and the accuracy of evaluation of severe MR by QFlow was compared vs guideline criteria by TEE.

Results: 29 patients had severe, 16 moderate and 6 mild MR. QFlow quantification was feasible in all pati-ents including prolapse (n=37), restriction (n=9), func-tional MR (n=5), as well as eccentric or multiple jects (n=41). QFlow Rvol correlated well with TTE 2D-PISA Rvol (ICC 0.75, p<0.001), quantitative estimated RVol (ICC=0.74, p<0.001) and 2D PISA TEE (ICC=0.79, p<0.001). The agreement between 3D-Qflow and cMR (ICC=0.86, p<0.001) was better than that of TTE-2D PISA Rvol (ICC=0.66, p<0.001) and TEE-2D PISA Rvol (ICC=0.69, p<0.001) with smaller limits of agreement by Bland-Altman analysis. Qflow had high accuracy for diagnosing severe MR (AUC 0.85, p<0.001) using TEE or cMR as criteria (AUC=0.91, 95% CI, 0.82-0.99, p<0.001).

Conclusions: The new prototype software allowed the semi-automatic quantification of MR RVol from 3D TEE color image in complex mitral regurgitations with multiple and eccentric jets, and showed better agree-ment to cMR than 2D PISA TTE and TEE, suggesting that this method is more accurate than conventional 2D PISA TTE and TEE.

ISSN – online: 2734 – 6382
ISSN-L 1220-658X
ISSN – print: 1220-658X
The Romanian Journal of Cardiology is indexed by:
ESC search engine
CODE: 379
CME Credits: 10 (Romanian College of Physicians)