Color Doppler velocity accuracy proximal to regurgitant orifices: influence of orifice aspect ratio

Proximal Flow Convergence Region as Assessed by Real-time 3-Dimensional Echocardiography: Challenging the Hemispheric Assumption

OBJECTIVE

Traditionally, a hemispheric assumption for the proximal flow convergence region (PFCR) is used when calculating mitral regurgitant (MR) effective orifice area (EROA). However, 2-dimensional (2D) echocardiography limits evaluation of the complete PFCR contour. Real-time 3-dimensional (3D) echocardiography (RT3D) allows direct assessment of the true PFCR contour. We hypothesized that the PFCR contour is not necessarily hemispheric, but rather hemielliptic, and aimed to apply a hemielliptic calculation, based on the 3D contour of the PFCR for more accurate MR quantification.

METHODS

In all, 50 patients with MR underwent RT3D to characterize PFCR contour as hemispheric or hemielliptic. MR EROA by RT3D-derived PFCR was calculated using a hemielliptic formula using 3D data. The 2D EROA was computed using standard hemispheric assumption. EROAs calculated from 2D and RT3D data were compared with quantitative Doppler EROA (mitral inflow--aortic outflow/MR time-velocity integral), used as an independent comparison.

RESULTS

Only 1 of 50 patients (2%) had a hemispheric PFCR contour by RT3D. The remaining had hemielliptic PFCR contours. Compared with Doppler method, 2D echocardiography significantly underestimated EROA (0.34 +/- 0.14 vs 0.48 +/- 0.25 cm(2), P < .001). RT3D EROA was not significantly different from Doppler EROA (0.52 +/- 0.17 vs 0.48 +/- 0.25, P = not significant). Of 33 patients with Doppler EROA greater than 0.3 cm(2) (> or =moderate-severe MR), 45% (15 of 33) were underestimated as having mild to moderate MR by 2D EROA.

CONCLUSIONS

The true PFCR contour as shown by RT3D is generally not hemispheric but hemielliptic, tracking the orifice contour. Based on this 3D shape, a hemielliptic approach can be used for practical clinical application with improved MR quantification.

Real-time 3-dimensional echocardiography for quantification of the difference in left ventricular versus right ventricular stroke volume in a chronic animal model study: Improved results using C-scans for quantifying aortic regurgitation

OBJECTIVE

The purpose of our study was to test the applicability of calculating the difference between left ventricular (LV) and right ventricular (RV) stroke volume (SV) for assessing the severity of aortic (Ao) regurgitation (AR) using a real-time 3-dimensional (3D) echocardiographic (RT3DE) imaging system.

METHODS

The Ao valve was incised in 5 juvenile sheep, 6 to 10 weeks before the study, to produce AR (mean regurgitant fraction = 0.50). Simultaneous hemodynamic and RT3DE images were obtained on open-chest animals with Ao and pulmonary flows derived by Ao and pulmonary electromagnetic flowmeters balanced against each other. Four stages (baseline, volume loading, sodium nitroprusside, and angiotensin infusion) were used to produce a total of 16 different hemodynamic states. Epicardial scanning was done with a 2.5-MHz probe to sequentially record first the RV and then the LV cavities. Cavity volumes from the 3D echocardiography data were determined from angled sector planes (B-scans) and parallel cutting planes (C-scans, which are planes perpendicular to the direction of the volume interrogation). AR volumes were determined from 3D images by computing and then subtracting RV SVs from LV SVs and then these were compared with electromagnetic flowmeter-derived SV and regurgitant volumes.

RESULTS

There was close correlation between RV and LV SVs of the RT3DE and electromagnetic methods (C-scans: LV, r = 0.98, standard error of the estimate [SEE] = 2.62 mL, P =.0001; RV, r = 0.89, SEE = 2.67 mL, P <.0001; and B-scans: LV, r = 0.95, SEE = 3.55 mL, P =.0001; RV, r = 0.77, SEE = 2.78 mL, P =.0003). Because of the small size of the RV in this model, the correlation was closer for C-scans than B-scans for RV SV. AR volume estimation also showed that C-scan (r = 0.93, SEE = 4.23 mL, P <.0001) had closer correlation than B-scan (r = 0.89, SEE = 4.87 mL, P <.0001). However, B-scan-derived AR fraction showed closer correlation than did C-scan (r = 0.82 vs r = 0.85, respectively).

CONCLUSION

In this animal model, RT3DE imaging had the ability to reliably quantify both LV (B- and C-scans) and RV SVs and to assess the severity of AR.