PURPOSE: The purpose of this study was to optimize high-resolution MSCT chest protocols for the evaluation of symptomatic immunosuppressed patients with suspected pneumonia using a dose-simulating program. MATERIALS AND METHODS: Using the MSCT (Siemens, Erlangen, Germany), 30 immunosuppressed patients with suspected pneumonia were examined with a low-dose HRCT of the chest (120 kV, 100 eff.mAs and collimation of 4 x 1 mm). A dose-simulating program was used to reconstruct the raw data at four different dose levels (70, 50, 35 und 25 mAs). For dose simulation for each mAs product, the correspondent noise level was added to the data. Images were generated with a slice thickness of 1 mm and 5 mm in the lung window. The images were then evaluated independently by two radiologists and graded on a scale of 1 to 3 points: 1 = no pneumonia, 2 = unclear, 3 = pneumonia. A receiver operating curve (ROC) analysis was performed to calculate the area under the curve (AUC). The actual dosage in mSv was calculated. The sensitivity and specificity were evaluated. RESULTS: Out of 30 patients, 7 had a normal chest finding and 23 had pneumonia. The area under the ROC curve (AUC) was 1.0 for every dosage and slice thickness. Infiltrates were detected correctly for all dosage levels. The sensitivity was 100 % for all dose levels and slice thicknesses. There was one false positive finding at 35 mAs and 1 mm slice thickness. At this dose level the specificity was reduced to 93 %. A reduction to 25 mAs had no influence on the detection of pneumonia. Thus, MSCT examinations of the chest can be performed with 25 mAs without missing the right diagnosis, resulting in an effective dose of 1.15 mSv (men), 1.5 mSv (women) and a CTDIvol of 2.5 mGy. CONCLUSION: For the evaluation of pneumonia in immunocompromised patients, MSCT examination of the chest can be performed with 25 mAs. Thus, radiation exposure was reduced to a quarter compared to the standard protocol.
