Diagnostic evaluation of upper tract urothelial carcinoma: can we safely omit diagnostic ureteroscopy?

Abstract

To identify clinicopathological or radiological factors that may predict a diagnosis of upper urinary tract urothelial cell carcinoma (UTUC) to inform which patients can proceed directly to radical nephroureterectomy (RNU) without the delay for diagnostic ureteroscopy (URS). All consecutive patients investigated for suspected UTUC in a high-volume UK centre between 2011 and 2017 were identified through retrospective analysis of surgical logbooks and a prospectively maintained pathology database. Details on clinical presentation, radiological findings, and URS/RNU histopathology results were evaluated. Multivariate regression analysis was performed to evaluate predictors of a final diagnosis of UTUC. In all, 260 patients were investigated, of whom 230 (89.2%) underwent URS. RNU was performed in 131 patients (50.4%), of whom 25 (9.6%) proceeded directly without URS - all of whom had a final histopathological diagnosis of UTUC - and 15 (11.5%) underwent RNU after URS despite no conclusive histopathological confirmation of UTUC. Major surgery was avoided in 77 patients (33.5%) where a benign or alternative diagnosis was made on URS, and 14 patients (6.1%) underwent nephron-sparing surgery. Overall, 178 patients (68.5%) had a final diagnosis of UTUC confirmed on URS/RNU histopathology. On multivariate logistic regression analysis, a presenting complaint of visible haematuria (hazard ratio [HR] 5.17, confidence interval [CI] 1.91-14.0; P = 0.001), a solid lesion reported on imaging (HR 37.8, CI = 11.7-122.1; P < 0.001) and a history of smoking (HR 3.07, CI 1.35-6.97; P = 0.007), were predictive of a final diagnosis of UTUC. From this cohort, 51 (96.2%) of 53 smokers who presented with visible haematuria and who had a solid lesion on computed tomography urogram had UTUC on final histopathology. We identified specific factors which may assist clinicians in selecting which patients may reliably proceed to RNU without the delay of diagnostic URS. These findings may inform a prospective multicentre analysis including additional variables such as urinary cytology.