1. The serum ferritin level provides a valuable index of the body iron store. An increase in serum ferritin has often been observed in patients with neoplastic disease and correlates well with the stage of cancer. A few studies have suggested the potential of urinary ferritin as a marker for transitional cell carcinoma. The rationale of the measurement, however, has not been investigated in detail. 2. Urinary ferritin levels were evaluated in patients with diverse urological diseases to investigate their potential clinical implications. 3. Analysis of logarithmic transformed values (ng/mg creatinine) showed that patients with both neoplastic and non-neoplastic urological diseases had significantly higher ferritin levels than normal control subjects ( P = 0.02). There was no apparent difference between subgroups of patients with urological disease ( P > 0.5). For patients with urothelial carcinoma, univariate analysis revealed a strong positive relationship between urinary ferritin levels and the density of lymphoid cells in tumour stroma ( P = 0.0001), while no important association was observed with tumour grade ( P = 0.32), stage ( P = 0.29) or urinary cytology detection ( P = 0.33). Patients with muscle-invasive tumour had significantly higher ferritin levels than those with papillary, superficial cancer ( P < 0.05). For patients with non-neoplastic urological disease ( n = 19), urinary ferritin levels tend to correlate with the severity of tissue inflammation ( P = 0.03). 4. The results suggest that urinary ferritin may reflect the degree of local inflammatory reaction in the urinary tract. For patients with urothelial carcinoma, the measurement may provide objective information regarding the amplitude of the local host response to tumour development. The potential of this test in determination of tumour treatment strategy needs further investigation.