The Role of Urine and Washing Cytology in Primary Transurethral Resection of Bladder Tumours

Abstract

Urine cytology (UC) is a recommended tool for the diagnosis of urothelial malignancies. Thus far, no specific recommendations regarding the role of washing cytology (WC) have been included in the guidelines. The goal of our study was to analyse the relationship between the histology of transurethrally (transurethral resection of the bladder [TURBT]) resected bladder tumours (BCa) and intraoperative UC or WC findings. Materials and Methods: Five hundred consecutive primary TURBT cases conducted between November 2010 and 2015 at our department of the University Hospital Luebeck were retrospectively analysed. Sensitivity, specificity, positive (PPV) and negative predictive values (NPV) of UC and WC were evaluated to detect BCa. Multivariate logistic regression models were fit to further examine associations between patient- and tumour-related factors and a bladder UC or WC positive for BCa. Results: UC was performed in 297 patients, WC in 294 patients, and both in 261 patients. Sensitivity was 50.7% in UC, 58.1% in WC, and 62.1% for both tests combined. Specificity was 97.8% for UC, 98.0% for WC, and 96.4% for the combined tests. PPV was 98.0% for UC, 98.1% for WC, and 97.2% for combined tests. NPV was 47.8% for UC, 54.5% for WC, and 55.9% for the combined tests. The multivariate analyses revealed no association between positive UC or WC results and subsequent radical cystectomy (UC OR 1.35, 95% CI: 0.3-5.7; WC OR 2.0, 95% CI: 0.4-11.4). Neither UC nor WC was significantly correlated with local recurrence. Conclusions: Cytologic testing is an important diagnostic tool in BCa detection, showing acceptable sensitivity of around 60% and excellent specificity of over 90%. UC and WC present similar sensitivity. Our results advocate, however, against cytologic testing during primary TURBT, especially with regard to the lack of value in assessing the risk of recurrence. The clinical benefit of taking both types of samples at once is minimal. Furthermore, intraoperative WC collection does not reliably predict subsequent cystectomies.