Non-invasive determination of the systolic peak-to-peak gradient in children with aortic stenosis: validation of a mathematical model.

ABSTRACT

Doppler derived systolic pressure gradients have become widely applied as noninvasively obtained estimates of the severity of aortic valvar stenosis. There is little correlation, however, between the Doppler derived peak instantaneous gradient and the peak-to-peak gradient obtained at catheterisation, the latter being the most applied variable to determine severity in children. The purpose of this study was to validate a mathematical model based on data from catheterisation which estimates the peak-to-peak gradient from variables which can be obtained by noninvasive means (Doppler derived mean gradient and pulse pressure), according to the formula peak-to-peak systolic gradient = 6.02+/-1.49*(mean gradient)-0.44*(pulse pressure). Simultaneous cardiac catheterization and Doppler studies were performed on 10 patients with congenital aortic valvar stenosis. Correlations between the gradients measured at catheter measured, and those derived by Doppler, were performed using linear regression analysis. The mean gradients correlated well (y = 0.67 x +11.11, r = 0.87, SEE = 6 mm Hg, p = 0.001). The gradients predicted by the formula also correlated well with the peak-to-peak gradients measured at catheter (y = 0.66 x +14.44, r = 0.84, SEE = 9 mm Hg, p = 0.002). The data support the application of the model, allowing noninvasive prediction of the peak-to-peak gradient across the aortic valvar stenosis.