No influence of smoking status on the performance of urine markers for the detection of bladder cancer

The performance of urinary markers for detecting bladder cancer (BC) is influenced by various factors. The aim of the present study was to evaluate the influence of smoking habits on the performance of four commonly used urine markers. Urine samples of 723 patients with suspected BC were analysed using urine cytology, fluorescence in situ hybridization (FISH), immunocytology (uCyt+ test), and quantitative nuclear matrix protein 22 (NMP22) immunoassay. The smoking habits of all patients were recorded and a cystoscopy performed within 2 weeks after urinary marker testing. Rates of false negative and false positive results were compared between non-smokers, former smokers, and current smokers by contingency analyses. We included 723 patients in this study, 431 (59.6%) of which were non-smokers, 215 former smokers (29.7%), and 77 (10.7%) current smokers. 148 patients (20.5%) had a tumour at the time of urinary marker testing. Respective rates of false positive test results among non-smokers, former smokers, and current smokers were: 16.3, 19.1, and 11.5% (p = 0.81) for urine cytology; 36.8, 42.0, and 32.7% for the uCyt+ test (p = 0.88); 18.0, 19.1, and 13.5% for FISH (p = 0.66); and 69.5, 71.6, and 71.2% for NMP22 (p = 0.67). Respective rates of false negatives among non-smokers, former smokers, and current smokers were: 31.4, 15.1, and 28.0% for cytology (p = 0.34); 21.4, 22.6, and 16.0% for uCyt+ test (p = 0.67); 24.3, 13.2, and 28.0% for FISH (p = 0.88); and 10.0, 18.9, and 8.0% for NMP22 (p = 0.80). Our results strongly suggest that smoking habits do not affect performance characteristics of urinary markers in the diagnostics of BC.