Interobserver Variation in Echocardiographic Measurements and Grading of Tricuspid Regurgitation Based on a Novel Web-Based Assessment Environment.

OBJECTIVES: The primary aim of this study was to assess interobserver variability in grading tricuspid regurgitation (TR) severity. The authors' secondary goals were to delineate which transesophageal echocardiographic (TEE) parameters best correlate with severity and how consistent the participants were at grading severity. DESIGN: This was a prospective cohort study of how clinicians evaluated previously acquired TEE images and videos. SETTING: The 19 TEE studies of patients with TR were recorded by 4 senior echocardiographers across 4 US academic institutions. The participants evaluated these cases on a novel, web-based, assessment environment designed specifically for this study. PARTICIPANTS: Twenty-nine fellowship-trained and board-certified cardiologists and cardiothoracic anesthesiologists volunteered to participate in the study as observers from 19 different institutions. INTERVENTIONS: No interventions were performed on the participants. MEASUREMENTS AND MAIN RESULTS: For each case, participants measured the vena contracta (VC), proximal isovelocity surface area (PISA), and jet area before giving a final classification on the severity of TR. Variation was highest for effective regurgitant orifice area and lowest for VC and PISA. The coefficient of variation, defined as the standard deviation from the mean divided by the mean, for all cases of trace, mild, moderate and severe TR were as follows: Jet Area-111%, 46%, 48%, 76%; VC-67%, 44%, 43%, 36%; PISA-52%, 48%, 31%, 35%; and effective regurgitant orifice area-127%, 95%, 66%, 58%. CONCLUSIONS: The interobserver variation in quantifying TEE parameters for TR is high, suggesting these may be difficult to measure reliably in a busy perioperative setting. Of the parameters assessed, VC and PISA radius had the highest interobserver agreement and the highest correlation with severity.

Three-dimensional versus two-dimensional echocardiographic assessment of functional mitral regurgitation proximal isovelocity surface area.

BACKGROUND: The geometric shape of the mitral regurgitation (MR) proximal isovelocity surface area (PISA) is conventionally assumed to be a hemisphere (HS). However, in functional MR, PISA is frequently neither an HS nor a hemiellipse (HE) but is often asymmetric and crescent shaped. We used 3-dimensional transesophageal echocardiographic (3D TEE), full-volume data sets to directly measure the PISA and subsequently compared calculated values of effective regurgitant orifice area (EROA) with conventional 2D TEE techniques. EROA calculations from all PISA measurements were finally compared with the cross-sectional area at the vena contracta, a well-validated reference measure of the functional MR orifice area. METHODS: Twenty-four cardiac surgical patients with functional MR, who underwent routine intraoperative TEE examinations with a 3D matrix array probe (X7-2t; IE33; Philips Healthcare, Inc., Andover, MA) were retrospectively evaluated for MR severity using quantitative 2D and 3D TEE-derived techniques. Conventional 2D TEE methods were used to estimate PISA assuming an HS shape and an HE shape. In addition, direct measurement of the 3D PISA was obtained (QLab, Philips Healthcare, Inc.) from corresponding full-volume, color-flow Doppler data sets. EROAs calculated from HS- and HE-PISA techniques were compared with the same values obtained from 3D TEE PISAs. EROAs obtained from all 3 PISA techniques were subsequently compared with vena contracta area. RESULTS: Three-dimensional PISA was significantly larger than both HS-PISA and HE-PISA (mean ± SD: 4.65 ± 2.03 cm² vs 2.10 ± 1.58 cm² and 2.75 ± 1.42 cm²; both P < 0.0001), respectively. HE-PISA was also larger than HS-PISA (P = 0.042). In addition, 3D EROA was larger than both HS- and HE-acquired EROAs (mean ± SD: 0.44 ± 0.21 vs 0.19 ± 0.12 cm² and 0.26 ± 0.14; both P < 0.0001), respectively, while HE-EROA was larger than HS-EROA (P = 0.024). Vena contracta area correlated well with 3D EROA (Spearman r = 0.865), HS-EROA (Spearman r = 0.820; P < 0.001) and HE-EROA (Spearman r = 0.819). However, the difference between vena contracta area and 3D EROA was significantly less than the differences between vena contracta area and either 2D HS- or 2D HE-EROA (P < 0.0001). CONCLUSIONS: Quantitative assessment of functional MR severity by 3D TEE may be superior to 2D methods by permitting more direct measures of PISA. Two-dimensional TEE techniques for assessing functional MR severity that rely on an HS- or HE-PISA shape may underestimate the EROA due to geometric assumptions that do not account for asymmetry.