Low-kilovoltage multi-detector row chest CT in adults: Feasibility and effect on image quality and iodine dose

PURPOSE: To assess the feasibility of low-kilovoltage (ie, 80-kV) chest computed tomography (CT) protocols for adults and the effect of such protocols on image quality and iodine dose. MATERIALS AND METHODS: Preliminarily, 90 patients (30 women, 60 men; mean age, 59 years) requiring contrast material-enhanced chest CT were randomly assigned to one of three protocol groups: protocol A, with use of 80 kV and 135 mAs; protocol B, with use of 80 kV and 180 mAs; or the standard protocol, with use of 120 kV and 90 mAs. Contrast material injection protocols were standardized in all groups. Image noise was calculated and plotted against patient weight. Subsequently, another 52 consecutive patients (11 women, 41 men; mean age, 57 years) were assigned to one of the protocols according to their weight: Patients weighing less than 60 kg were assigned to protocol A; patients weighing 60-75 kg, to protocol B; and patients weighing more than 75 kg, to the standard protocol. Two readers evaluated the CT images qualitatively by using a five-point scale. Statistical analyses were performed by using analysis of variance, kappa, and Fisher exact tests. RESULTS: In the preliminary study, the mean image noise values with protocols A (24 HU) and B (20 HU) were significantly higher (P < .001) than that with the standard protocol (12 HU). With protocols A and B, in the patients weighing more than 60 kg and more than 75 kg, respectively, the noise increased exponentially with patient weight. In the subsequent study, qualitative analysis revealed no significant difference between the low-kilovoltage examinations and the standard examination. Compared with use of the standard protocol, use of protocols A and B resulted in the iodine-based contrast material dose being reduced by 54% and 39%, respectively. CONCLUSION: Weight-adapted low-kilovoltage contrast-enhanced chest CT examinations can be routinely performed with 80 W. Use of these protocols results in good diagnostic image quality and makes it possible to reduce contrast material use by more than 50%. (C) RSNA 2004.