[Does multi-slice CT provide reliable attenuation values if measured with low slice thickness and low tube current? Results of a phantom study].

PURPOSE: To test whether CT with low slice thickness and low tube current provides reliable attenuation measurements. MATERIALS AND METHODS: Using multi-slice CT and a phantom, we measured the attenuation values of thrombi with different proportions of erythrocytes, using a slice thickness of 1.25 mm, 2.5 mm, and 5 mm with tube currents of 200 mA, 300 mA, and 400 mA and a slice thickness of 0.625 mm with tube currents of 150 mA, 175 mA, and 200 mA. Differences in attenuation values and pixel noise between the three thrombi for tube current and slice thickness were statistically analyzed. RESULTS: The attenuation values of all thrombi increased (p < 0.05) when the slice thickness decreased using a tube current of 200 mA or when the tube current decreased using a slice thickness of 1.25 mm. With higher tube currents and thicker slices, the CT values depended on the type of thrombus and the slice thickness. In slices with a thickness of 0.625 mm, the CT values decreased with the tube current in the mixed thrombus with a low proportion of erythrocytes and in the red thrombus (p < 0.05). The maximal difference in mean attenuation values was 4.3 HU with a slice thickness of 0.625 mm and 2.2 HU with a slice thickness of 1.25 mm. The pixel noise increased as the slice thickness decreased (p < 0.05) with the exception of the red thrombus, if reduced to 0.625 mm. The pixel noise also increased as the tube current decreased (p < 0.05) except in mixed thrombi measured with 0.625 mm. The maximal difference in mean standard deviation was 1.8 HU with a slice thickness of 1.25 mm. CONCLUSION: The accuracy of attenuation values as determined by CT with low slice thickness and low tube current with a maximal difference of 4.3 HU suffices for the purposes of clinical routine. A reduction of slice thickness from 1.25 mm to 0.625 mm yields the greatest differences in CT values.