[Dose-reduced 16-slice multidetector-row spiral computed tomography in children with bronchoscopically suspected vascular tracheal stenosis -- initial results].

PURPOSE

To evaluate the diagnostic accuracy of contrast-enhanced dose-reduced 16-slice multidetector-row CT (MDCT) in newborns and infants with fiberoptic bronchoscopically suspected vascular-induced tracheal stenosis.

MATERIALS AND METHODS

12 children (4 days to 3 years, 1.2 - 13.5 kg body weight) were examined using i. v. contrast-enhanced 16-slice MDCT (SOMATOM Sensation 16, Forchheim, Germany) without breath-hold and under sedation (11/12). All MDCTs were performed with a dose reduction. The beam collimation was 16 x 0.75 mm, except in the case of one child. MPRs along the tracheal axis in the x-, y- and z-directions and volume-rendering-reconstructions (VRTs) were calculated based on a secondary raw data set in addition to conventional axial slices. 2 radiologists used a three-point grade scale to evaluate the image quality, motion, and contrast media artifacts as well as the usefulness of the 2D- and 3D-reconstructions for determining the diagnosis. Statistical analysis was performed on the basis of a Kappa test.

RESULTS

In all cases the cause of the fiberoptic bronchoscopically suspected tracheal stenosis was revealed: compression due to the brachiocephalic trunk (n = 7), double aortic arch (n = 2), lusorian artery (n = 1), vascular compression of the left main bronchus (n = 2). In 3 patients further thoracic anomalies, such as tracheobronchial (n = 2), and vascular (n = 2) and vertebral (n = 1) anomalies were found. The attenuation in the anomalous vessels was 307 +/- 140 HU. The image noise was 9.8 +/- 1.9 HU. The mean dose reduction was 82.7 +/- 3.2 % compared to a standard adult thoracic CT. All examinations were rated as diagnostically good (median 1, range 1, k = 1). 3D images did not show any stair artifacts (median 2, range 1 - 2, k = 1). The image noise was minor to moderate and hardly any motion artifacts were seen (median 1, range 1 - 2, k = 0.8). Contrast media artifacts were rated zero to minor (median 1.5, range 1 - 2, k = 0.676). MPRs (median 1, range 1, k = 1) and VRTs (median 1, range 1, k = 1) were found to be useful for diagnosis. Subsequent vascular surgery was performed on 8 patients.

CONCLUSION

Contrast-enhanced dose-reduced 16-slice MDCT is effective for demonstrating the cause of fiberoptic bronchoscopically suspected vascular-induced tracheal stenosis even in very small and severely ill children despite the small contrast media amount and free breathing.