MicroRNAs (miRNAs/miRs) act as markers for various cancers, including bladder cancer (BC). Plasma miRNAs are potential biomarkers for the noninvasive diagnosis and long‑term surveillance of BC. The aim of the present study was to identify diagnostically reliable miRNAs in the plasma and examine their potential monitoring capacity. miRNAs were measured by reverse transcription‑quantitative polymerase chain reaction. The study was performed as a discovery phase, which included plasma samples from each of the 10 patients with and without BC prior to transurethral resection (TURB). The results were validated in a second phase, involving 36 patients with plasma samples collected before the second TURB or radical cystectomy (RC), and after RC. During the discovery step, three elevated miRNAs (miR‑15b‑5p, miR‑590‑5p, miR‑29b‑3p) and two decreased miRNAs (miR‑10b‑5p, miR‑144‑5p) were selected as potential miRNA candidates for further validation. miR‑15b‑5p and miR‑590‑5p were finally confirmed to discriminate between cancer cases and controls; however, for disease monitoring of BC, both miRNAs were not suitable as a decline in the miRNA levels was not observed in some patients after tumor removal. Our results suggested that circulating miR‑15b‑5p and miR‑590‑5p have useful diagnostic potential for BC, but are rather unsuitable as monitoring markers for disease. The reasons of this apparent contradictory observation may be due to the aspect of biological variation of circulating miRNAs and serial measurements could be unreliable.