To determine the utility of detecting telomerase activity in transitional cell bladder cancer, using the telomeric repeat amplification protocol (TRAP) assay, and thus provide a test for the detection and monitoring of bladder cancer, especially low-grade tumours. Telomerase activity was assayed in bladder cancer tissues and the exfoliated urothelial cells from 29 patients with bladder cancer, 10 with benign disease, and in 10 healthy subjects using the TRAP assay. The levels were assessed semiquantitatively by calculating the ratio to an internal telomerase assay standard. Telomerase activity was very low in the exfoliated urothelial cells from all healthy subjects and patients with benign disease, with a mean (SD) ratio of 0.25 (0.03) and 0.33 (0.04), respectively. A threshold ratio of 0.4 was calculated as the mean +/- 2 SD of the telomerase activity level of the exfoliated urothelial cells from patients with benign disease. Using this threshold, telomerase activity was negative in exfoliated urothelial cells from all benign cases (100% specificity) and positive in all 26 bladder cancer tissue samples assayed. In tumour tissue, telomerase activity was not associated with tumour grade, size or stage. Telomerase activity in exfoliated urothelial cells from patients with bladder cancer was positive in 25 of 29 samples (86% sensitivity). The sensitivity of telomerase activity in exfoliated cells was seven of nine in G1, 10 of 12 in G2 and all eight G3 tumours; the corresponding sensitivity for voided urine cytology in G1, G2 and G3 tumours was two of nine, six of 12 and six of eight, respectively. These results indicate that telomerase activation occurs as an early step in carcinogenesis and the semi-quantitative analysis of telomerase activity in exfoliated urothelial cells could be a minimally invasive and useful method for detecting bladder cancer, even in low-grade tumours.