Clinical evaluation of a computer simulated prediction model of contrast enhancement of the liver in spiral CT.

OBJECTIVE: A software program was developed simulating a compartmental model of blood circulation based on differential equations. The aim of this study was to compare software-simulated levels of hepatic enhancement with the true values in patients and to test how many patients reach the simulated hepatic enhancement level. METHODS: As software program the CT application software carebolus 2 (Siemens, Forchheim, Germany) was used. Hepatic contrast-enhancement curves were simulated prior to CT examinations to evaluate a patient specific time delay after contrast application. At the time delay, when the simulation curve showed an enhancement threshold of 40 Hounsfield Units (HU), the CT spiral scan was started applying 120 ml contrast media with 2 ml/s. The simulated curves were compared with the empiric curves of each patient. RESULTS: 25 of 28 patients (89%) achieved 40 HU. The mean enhancement of empiric patients curves was 46.32 +/- 11.9 HU, the mean simulated enhancement was 46.62 +/- 4.3 HU S.D. (P= 0.48). 4.4 values per patient liver could be compared with the simulation curve (122 points for 28 patients): 50% of the patient curves were within a range of 5 HU compared with the simulation curve. CONCLUSION: Software simulation of contrast enhancement curves of the liver is a feasible and valuable method to predict individual liver enhancement curves. Improvements concerning the integration of cardiovascular parameters and preexisting liver parenchymal diseases into the simulation software have to be arranged.

Multi-detector row spiral CT angiography of the thoracic outlet: dose reduction with anatomically adapted online tube current modulation and preset dose savings.

PURPOSE: To evaluate image quality obtained with anatomically adapted online tube current modulation and preset minimum dose savings at multi-detector row spiral computed tomographic (CT) angiography of the thoracic outlet. MATERIALS AND METHODS: A total of 100 patients were evaluated for thoracic outlet arterial syndrome with spiral CT angiography (collimation, 4 x 1 mm; pitch, 1.75) both with and without dose reduction by means of anatomically adapted online tube current modulation and preset minimum dose savings. Preset minimum savings of 20% and of 32% were applied in two groups of 50 patients (groups 1 and 2). In each group, low-dose scanning was performed in 25 patients in the neutral position and in 25 patients after postural maneuver. Tube current-time product, noise, presence and quality of graininess and of linear streak artifacts on transverse CT scans, and diagnostic value of sagittal reformations and volume-rendered images were evaluated and recorded for each data set. chi2 test was used to compare frequencies; paired Wilcoxon rank test, to compare subjective and objective image quality scores. P <.05 indicated a significant difference. RESULTS: In group 1, mean tube current-time product was 3225 mAs for reference scans and 2101 mAs for low-dose scans (mean reduction, 35%; range, 27%-47%). In group 2, mean was 3070 mAs for reference scans and 2068 mAs for low-dose scans (mean reduction, 33%; range, 17%-38%). In group 1, no differences in frequencies of graininess and linear streaking or in noise level were found between images acquired with or without dose reduction. In group 2, no difference was found in noise level between low-dose and reference scans. On low-dose scans, moderate linear streaking was observed with lower frequency and moderate graininess was observed with higher frequency, but artifacts did not compromise image quality or prevent confident assessment of arterial diameter in the three compartments of the thoracic outlet. CONCLUSION: Online tube current modulation with a preset minimum dose saving of 20% allowed 35% reduction in mean tube current-time product, with no loss in image quality.