Detection of Urothelial Carcinoma: Comparison of Reduced-Dose Iterative Reconstruction with Standard-Dose Filtered Back Projection.

PURPOSE: To prospectively assess radiation dose, image quality, and diagnostic performance of computed tomography (CT) urography for detection of urothelial carcinomas by performing reduced-dose scanning with iterative reconstruction (IR) compared with standard-dose scanning with filtered back projection (FBP).

MATERIALS AND

METHODS

The institutional review board approved the study with written informed patient consent. In total, 2163 patients at high risk for urothelial carcinomas randomly underwent standard-dose scanning with FBP (protocol A, 120 kVp for >80 kg body weight; protocol B, 100 kVp for 50-80 kg body weight) or reduced-dose scanning with IR (protocol C, 100 kVp for >80 kg body weight; protocol D, 80 kVp for 50-80 kg body weight). Objective image quality (signal-to-noise ratio and contrast-to-noise ratio) between the two groups with same weight range was measured for various regions of interest. Subjective image quality (visual image noise, artifact, ureter depiction, and overall image quality) and diagnostic accuracy (per lesion and per patient) were assessed with three- and five-point scores, respectively.

RESULTS

Size-specific dose estimate (protocol A vs protocol C, 24.2 mGy vs 19.2 mGy, respectively; protocol B vs protocol D,13.9 mGy vs 8.8 mGy, respectively) was significantly lower in reduced-dose scanning (P < .001 for both). There were significantly higher signal-to-noise and contrast-to-noise ratios in reduced-dose scanning, except for the abdominal aorta (P < .05 for all). There was no significant difference in subjective image quality, except for artifacts in protocols B and D (range, 4-5 vs 3-4; P < .05). Per-lesion diagnostic accuracy was 90.8% (89 of 98, protocol A), 91.3% (105 of 115, protocol B), 92.9% (79 of 85, protocol C), and 88.8% (111 of 125, protocol D).

CONCLUSION

Reduced-dose scanning with IR showed dose reduction and no significant difference of image quality in detection of urothelial carcinomas, except for some artifacts in 80-kVp scanning.