Since February 1990, we have attempted nonsurgical occlusion of persistent patency of the arterial duct using the Rashkind double-disk occluding device, of either 12 or 17 mm diameter. Results are presented from 31 patients (23 female and eight male, ages 15 months to 16 years). M-Mode, cross-sectional, Doppler and color Doppler echocardiographic studies were made prior to closure, in the following 24 hours, and six months later. Left ventricular diastolic dimension and shortening fraction, systolic intervals, and left atrium/aortic ratio were evaluated by M-Mode echo. We examined carefully the location of the device relative to the left pulmonary artery and descending aorta, also checking for disturbed flows and residual shunts. These studies showed a clear tendency for the left ventricular diameter to decrease, and revealed a significant normalization of the left atrium/aortic ratio. Successful closure of the duct was achieved in 27 cases. Follow-up studies six months later showed residual shunting in four cases. In one patient, a second device was implanted with an excellent result. At the immediate follow-up, turbulences were noted in the area of the device (nine cases) and in the left pulmonary artery (seven cases). Distortion of the anatomic orientation of the left pulmonary artery was observed in three cases in which we discovered 20 mm Hg systolic gradients. In our opinion, echocardiography is the best technique with which to follow-up these patients. The high sensitivity of color Doppler echocardiography revealed minimal distortion of the left pulmonary artery, probably generated by regional anatomic adjustment to the insertion of the device.

Our aim was to analyze left ventricular fractional shortening during extracorporeal membrane oxygenation under the influence of changing volume loading conditions induced by a ductal left-to-right shunt. In all patients, the fractional shortening was observed using echocardiography before, during, and after bypass, irrespective of the presence or absence of the ductal left-to-right shunt. During membrane oxygenation, there was a significant decrease in fractional shortening (p less than 0.001), with no difference before and after membrane oxygenation. A greater decrease in fractional shortening was observed in the group with a ductal left-to-right shunt when compared to patients lacking the ductal shunt (p less than 0.006). The diastolic diameter of the left ventricle also increased significantly during the membrane oxygenation in those patients with left-to-right ductal shunting. Moreover, the patients with left-to-right shunting showed a very severe decreased fractional shortening, lower than 10 per cent, with significantly greater frequency (p less than 0.05) during the course of membrane oxygenation. Conclusion: An important decrease in left ventricular fractional shortening is observed during veno-arterial extracorporeal membrane oxygenation. Left-to-right shunting during bypass, as seen in the patients with patency of the arterial duct, increases the loading conditions on the left ventricle, and produces a significant increase in left ventricular diastolic dimensions. Despite the effects of volume loading produced by the ductal shunt during bypass, the decrease in fractional shortening is significantly more pronounced for these patients. Therefore, during membrane oxygenation the volume loading produced by the ductal shunt is unable to prevent a decrease in left ventricular fractional shortening.