The flow convergence region (FCR), a zone of progressive laminar velocity acceleration, can be imaged by color Doppler proximal to stenotic and regurgitant orifices. Theoretically, FCR proximal to a discrete circular and planar orifice consists of concentric hemispheric shells of equal and accelerating velocities centered at the orifice. According to the continuity principle, flow rate across any of these isovelocity surfaces equals flow rate through the orifice. The aim of this study was to investigate whether these principles could be applied to quantify left-to-right shunting and the size of atrial septal defects after balloon mitral commissurotomy. Biplane transesophageal echocardiography (TEE) with color flow imaging was performed on 36 consecutive patients (mean age, 57 +/- 16 years; range, 14-78 years) immediately before and within 24 hours of balloon (Inoue, n = 33; Mansfield, n = 3) mitral commissurotomy. Left-to-right atrial shunting was detected by TEE in 33 patients (92%) and by oximetry in 11 patients (31%). The radius r of FCR was measured from the first aliasing limit, at a Nyquist velocity reduced to 11 cm/sec by zero-shifting, to the orifice in the atrial septum. FCR was assumed to be hemispherical. Hence, flow rate (Q) was calculated as 2 pi r2 Vr, where Vr is the velocity at a radial distance r. The velocity profile of transatrial flow was assessed by means of high pulse repetition frequency, from which the maximum flow velocity (Vp) and the velocity-time integral (VTI) were obtained. The flow area of the atrial septal defect was calculated as Qm, the maximal flow rate, divided by Vp. Hence, shunt flow was calculated as flow area x VTI x heart rate. FCR was analyzed in two orthogonal planes. Mean Qm (38.1 +/- 26.5 versus 5.3 +/- 2.7 mL/sec), flow area (22.1 +/- 11.2 versus 4.4 +/- 2.0 mm2), and shunt flow (1,590 +/- 1,070 versus 200 +/- 130 mL/min) on transverse plane imaging were all significantly higher in patients with shunts detected by oximetry than in those without. Similar results were obtained from longitudinal plane imaging. Qm correlated well with oximetric shunt flow (r = 0.89-0.94, p < 0.001) and shunt ratio (r = 0.91-0.94, p < 0.001). Flow area correlated closely (r = 0.93-0.94, p < 0.001) with area determined by direct measurement from two-dimensional echocardiography. Shunt flow determined by FCR also correlated closely (r = 0.94-0.98, p < 0.001) with that determined by oximetry and that derived from two-dimensional echocardiography and pulsed Doppler (r = 0.96, p < 0.001). The flow convergence region imaged by TEE color flow mapping provides new and accurate quantitative information on atrial shunt flow and defect size after balloon mitral valvotomy. It is a quick, reliable, and fairly simple method that can be readily incorporated into routine clinical practice.