A Prospective Comparison of Standard-Dose CT Enterography and 50% Reduced-Dose CT Enterography With and Without Noise Reduction for Evaluating Crohn Disease

OBJECTIVE: The purpose of this study was to prospectively compare standard-dose CT enterography (CTE) and 50% reduced-dose CTE, obtained with and without an image noise reduction method, in the evaluation of Crohn disease. SUBJECTS AND METHODS: Ninety-two patients (69 men and 23 women; mean age ± SD, 31.2 ± 9.5 years) with Crohn disease underwent CTE. Using a dual-source scanner equipped with a proprietary noise reduction method (iterative reconstruction in image space IRIS), three sets of CTE images were obtained: standard-dose filtered back projection (FBP) (i.e., weighted FBP), low-dose (i.e., 50% reduction) FBP, and low-dose IRIS CTE. Image noise was measured. Two independent radiologists evaluated subjective image quality (1 worst to 4 best) and findings of active Crohn disease in the terminal small-bowel segment, including mural hyperenhancement, thickening and stratification, comb sign, and increased perienteric fat attenuation (1 definitely absent to 5 definitely present). RESULTS: The mean (± SD) volume CT dose index (CTDI(vol)) was 7.0 ± 0.9 mGy and 3.5 ± 0.5 mGy for standard-dose and low-dose CTE examinations, respectively. The mean (± SD) image noise for standard-dose FBP, low-dose FBP, and low-dose IRIS CTE was 10.6 ± 1.7 HU, 13.9 ± 2.1 HU, and 9.7 ± 1.7 HU, respectively (p < 0.001 for all comparisons). Both assessors found that image quality was poorer with low-dose (mean grade (± SD), 2.3 ± 0.4-2.7 ± 0.5) than in standard-dose (3 ± 0) CTE (p < 0.01), and one found that image quality was poorer with low-dose IRIS (2.3 ± 0.4) than with low-dose FBP (2.7 ± 0.5) CTE (p < 0.01). Low-dose (with or without IRIS) and standard-dose CTE showed ≥ 85% agreement (one-sided 95% CI ≥ 77%) in interpretation of bowel findings. CONCLUSION: Low-dose CTE using 50% reduced-dose performed similarly to standard-dose CTE in identifying findings of enteric inflammation of Crohn disease. Although a noise reduction method markedly reduced image noise in half-dose examinations, its effect on image quality was not as great and was reader dependent.