Radiation dose evaluation in multidetector-row CT imaging for acute stroke with an anthropomorphic phantom.

This study evaluated radiation dose and dose reduction in CT imaging for acute stroke. Radiation doses in three types of CT imaging (i.e. non-contrast-enhanced CT, CT perfusion (CTP) and CT angiography (CTA)) were measured with an in-phantom dosimetry system for 4-, 16- and 64-detector CT scanners in 5 hospitals. To examine the relationship between image quality and radiation dose in CTA, image contrast-to-noise ratio was evaluated. Doses to the brain, lens, salivary glands and local skin obtained with scan protocols in routine use were: 42-71 mGy, 30-88 mGy, 3.9-7.3 mGy and 40-97 mGy in non-contrast-enhanced CT; 41-75 mGy, 9.9-10 mGy, 1.5-2.1 mGy and 107-143 mGy in CTP; and 8.2-55 mGy, 26-69 mGy, 2.0-73 mGy and 32-72 mGy in CTA. For the combination of these CT examinations, on average a patient would receive 236 mGy for the maximum local skin dose and 4.2 mSv for the effective dose evaluated by the International Commission on Radiological Protection (ICRP) 103. Effective doses in CTP in this study were less than those obtained with representative protocols of Western countries. Average effective doses in each CT examination were not more than 1.5 mSv. The use of reduced kV and a narrow scan range would be effective in dose reduction of CTA and CTP, and intermittent scanning would be essential in CTP. Although lens and maximum local skin doses were far less than the thresholds for deterministic effects, since radiation risks would be increased in repeated CT examinations, efforts should be devoted to dose reduction in stroke CT examinations.