Background This study intended to investigate irisin levels in bladder cancer patients and healthy controls. Objective Our aim was to evaluate if serum irisin could be used as a diagnostic tool in bladder cancer and further, if it could differentiate muscle-invasive and non-muscle-invasive bladder cancer patients. Methods In this study, 90 primary bladder cancer patients in addition to 30 age-matched healthy individuals for the control group were prospectively included. Bladder cancer patients were divided into two subgroups as non-muscle-invasive (60 patients) and muscle-invasive (30 patients). Blood samples were obtained before the diagnosis of the disease. Serum irisin levels were measured using ELISA. Demographic data as well as tumor grade and stage were noted. Results Mean serum irisin level was significantly lower in the bladder cancer patients compared to the control group (4.53 +/- 2.55 vs. 16.5 +/- 5.67, p < 0.001). Also, serum irisin level was statistically lower in the muscle-invasive bladder cancer group compared to the non-muscle-invasive counterparts (3.19 +/- 1.47 vs. 5.18 +/- 2.73, p < 0.001). Serum irisin could differentiate bladder cancer patients from healthy individuals with a sensitivity of 86.2% and a specificity of 89.7% at a cut-off value of 8.689 (AUC = 0.859). Moreover, to discriminate between NMIBC and MIBC, the sensitivity was 75% and the specificity was 73.7% at a cut-off value of 3.97 (AUC = 0.732). Conclusion Our results showed that serum irisin levels can be used for the diagnosis of bladder cancer. Also, it can help distinguish high-grade and stage tumor.