OBJECTIVE
The purpose of this study was to compare CT with conventional and simulated reduced–tube current in the evaluation for acute appendicitis in children.
MATERIALS AND METHODS
Validated noise-addition (tube current–reduction) software was used to create 50% and 75% tube current reductions in 60 CT examinations performed for suspected appendicitis, resulting in 180 image sets. Three blinded pediatric radiologists scored the randomized studies for the following factors: presence of the normal appendix or appendicitis (5-point scale; 1 = definitely absent and 5 = definitely present), presence of alternate diagnoses, and overall image quality (1 = nondiagnostic and 5 = excellent). Truth was defined by the interpretation of the conventional examination.
RESULTS
For conventional examinations, the total number of reviews (60 cases × 3 readers = 180) in which the normal appendix was identified was 120 of 180 (66.7%), compared with 108 of 180 (60%) in the 50% (p = 0.19) and 91 of 180 (50.6%) in the 75% (p = 0.002) tube current–reduction groups. Appendicitis was identified in a total of 39 of 180 (21.7%), 38 of 180 (21.1%), and 37 of 180 (20.6%) examinations, respectively (p > 0.05). This translates to sensitivities of 97% and 95% for the 50% and 75% tube current–reduction groups, respectively. Alternate diagnoses were detected in 14%, 16%, and 13% of scans, respectively. Compared with conventional–tube current examinations, reader confidence and assessment of image quality were significantly decreased for both tube current–reduction groups.
CONCLUSION
Simulated tube current–reduction technology provides for systematic evaluation of diagnostic thresholds. Application of this technology in the setting of suspected appendicitis shows that tube current can be reduced by at least 50% without significantly affecting diagnostic quality, despite a decrease in reader confidence and assessment of image quality.