Comparative Evaluation of Diagnostic Quality in Native Low-dose CT without and with Spectral Shaping employing a Tin Filter in Urolithiasis with implanted Ureteral Stent

Abstract

Purpose Spectral shaping employing a tin filter can be used for dose reduction in CT of the abdomen in patients with urolithiasis. As ureteral stents may be in direct contact with the calculus, a good image quality is mandatory. The goal of this study was to obtain data of the effect of tin filtering on image quality and dose in patients with urolithiasis in direct contact with ureteral stents. Materials and Methods 84 examinations (conventional low dose vs. modified low dose protocol with tin filtering, randomized) were performed in 65 patients (48 men, 17 women, age 55.0 ± 15.2 years (18–90 years), maximum of one examination per protocol). Image quality and visibility of the calculus was rated on a 5-point-Likert scale by 2 experienced radiologists. Quantitative indicators of image quality were signal-to-noise-(SNR) and contrast-to-noise-ratios (CNR) as well as a figure-of-merit (FOM). Results With a non-inferiority margin of 0.5 points of the 5-point Likert scale, there was non-inferiority of the examinations with tin filter regarding image quality (95 % CI 4.1–4.3, rejection limit 3.5). Non-inferiority regarding visibility of the calculus could be shown (calculus size: 1–2.4 mm: 95 % CI 3.39–4.12; limit 2.73; 2.4–3.8mm: 95 % CI 4.09–4.47; limit 3.65; > 3.8mm: all maximal ratings). Average values of CNR were significantly higher using tin filters (17.0 vs. 10.6). Doses were significantly reduced in the modified protocol (effective dose 1.2 mSv vs. 1.5 mSv; size-specific dose estimate 2.33 mGy vs. 3.09 mGy) with non-significant effect in the subgroup of patients with BMI ≥ 35. Conclusion Even with direct contact between a calculus and ureteral stent, radiation reduced examinations by spectral shaping by tin filters are non-inferior to examinations without tin filtering at a concurrent significant dose reduction. Key points:  Citation Format