Assessment of modified Blalock-Taussig shunt in children with congenital heart disease using multidetector-row computed tomography.

The purpose of this study was to assess the feasibility of multidetector-row computed tomography (MDCT) for the evaluation of modified Blalock-Taussig (B-T) shunt in children with congenital heart disease associated with reduced pulmonary blood flow. A total of 25 consecutive patients (mean age, 2.6 ± 3.6 years; range, 2 months-16 years) underwent MDCT angiography of the thorax with a 16-detector row scanner prior to cardiac catheterization. A total of 39 shunts (right, 22; left, 17) were included in the study. Conventional angiographic findings were used as the gold standard for the detection of B-T shunts. Shunt diameter was measured quantitatively and independently at four sites (the subclavian artery site, the pulmonary artery site, the widest site, and the stenotic site) on MDCT and on conventional invasive angiography. All B-T shunts were depicted on multiplanar reconstruction (MPR), maximum intensity projection (MIP), curved planar reconstruction (CPR), and three-dimensional volume-rendered (VR) images, enabling evaluation in all patients except for one with occluded shunt. There were excellent correlations between MDCT- and conventional angiography-based measurements of shunt diameter at the subclavian artery site, pulmonary artery site, and the widest site (R² = 0.46, 0.74 and 0.64, respectively; p < 0.0001 for each), although systematic overestimation was observed for MDCT (mean percentage of overestimation, 23.1 ± 32.4%). Stenotic site diameter and degree of stenosis showed a mild correlation (R² = 010 and 0.25, respectively; p < 0.01 for each). This study demonstrates that MDCT is a promising tool for the detection of lesions in B-T shunts.