Is a Single Nephrographic Phase Computed Tomography Sufficient for Detecting Urothelial Carcinoma in Patients with Visible Haematuria? A Prospective Paired Noninferiority Comparison

Kristina F Galtung; Peter M Lauritzen; Gunnar Sandbæk; Dag Bay; Erica Ponzi; Eduard Baco; Nigel C Cowan; Anca M Naas; Erik Rud

doi:10.1016/j.euros.2023.06.005

Is a Single Nephrographic Phase Computed Tomography Sufficient for Detecting Urothelial Carcinoma in Patients with Visible Haematuria? A Prospective Paired Noninferiority Comparison

Eur Urol Open Sci. 2023 Jul 19:55:1-10. doi: 10.1016/j.euros.2023.06.005. eCollection 2023 Sep.

Authors

Kristina F Galtung¹, Peter M Lauritzen^{1

2}, Gunnar Sandbæk^{1

3}, Dag Bay¹, Erica Ponzi^{4

5}, Eduard Baco^{3

6}, Nigel C Cowan⁷, Anca M Naas⁸, Erik Rud¹

Affiliations

¹ Department of Radiology, Oslo University Hospital, Oslo, Norway.
² Department of Life Sciences and Health, Faculty of Health Science, Oslo Metropolitan University, Oslo, Norway.
³ Institute of Clinical Medicine, Faculty of Medicine, University of Oslo, Oslo, Norway.
⁴ Department of Research Support for Clinical Trials, Clinical Trial Unit, Oslo University Hospital, Oslo, Norway.
⁵ Oslo Center for Biostatistics and Epidemiology (OCBE), Department of Biostatistics, University of Oslo, Oslo, Norway.
⁶ Department of Urology, Oslo University Hospital, Oslo, Norway.
⁷ Department of Radiology, Portsmouth Hospitals University NHS Trust, Portsmouth, UK.
⁸ Department of Pathology, Oslo University Hospital, Oslo, Norway.

Abstract

Background: There is uncertainty about the utility of multiphase computed tomography (CT) compared with single-phase CT in the routine examination of patients with visible haematuria (VH).

Objective: To compare the accuracies of single nephrographic phase (NP) CT and four-phase CT in detecting urothelial carcinoma (UC).

Design setting and participants: This was a single-centre, prospective, paired, noninferiority study of patients with painless VH referred for CT before cystoscopy between September 2019 and June 2021. Patients were followed up for 1 yr to ascertain UC diagnosis.

Intervention: All patients underwent four-phase CT (control), from which single NP CT (experimental) was extracted. Both were independently assessed for UC.

Outcome measurements and statistical analysis: The primary outcome was the difference in accuracy between the control and experimental CT using a 7.5% noninferiority limit. Histologically verified UC defined a positive reference standard. Secondary outcomes included differences in sensitivity, specificity, negative (NPV) and positive (PPV) predictive values, and area under the curve (AUC). All results are reported per patient.

Results and limitations: Of the 308 patients included, UC was diagnosed in 45 (14.6%). The difference in accuracy between the control and experimental CT was 1.9% (95% confidence interval -2.8 to 6.7), demonstrating noninferiority. Sensitivity was 93.3% versus 91.1%, specificity was 83.7% versus 81.8%, NPV was 98.7% versus 98.2%, PPV was 49.4% versus 46.1%, and AUC was 0.96 versus 0.94 for the control versus experimental CT. Limitations included a low number of UC cases and no definite criteria for selecting a noninferiority limit.

Conclusions: The accuracy of NP CT is not inferior to that of four-phase CT for detecting UC.

Patient summary: This study shows that a computed tomography (CT) examination with only one contrast phase is no worse than a more complex CT examination for detecting cancer in the urinary tract among patients presenting with visible blood in the urine.

Keywords: Haematuria; Multidetector computed tomography; Prospective study; Transitional cell carcinoma.