Purpose: We prospectively evaluated the diagnostic efficacy of the BTA TRAK assay according to the stage, grade and hematuria in detecting primary and recurrent bladder cancer, and compared results with voided urine cytology. Materials and Methods: Urinalysis, cytology and BTA TRAK assay were performed simultaneously with the single voided fresh urine samples from 130 subjects. The sensitivity and specificity of the BTA TRAK assay were compared to those of urine cytology and analyzed according to the stage or grade. The subjects were also divided into 4 groups according to the degree of hematuria and the influence of hematuria on the result of the BTA TRAK assay was evaluated. Results: The overall sensitivity and specificity of the BTA TRAK assay for detecting bladder cancer were 82.8% and 65.3%, respectively and those of urine cytology were 44.8% and 100%. The sensitivity of the BTA TRAK assay was significantly higher than that of urine cytology in bladder cancer with lower stage and grade. On univariate and multivariate analysis, gross hematuria and the presence of bladder cancer affected the results of the BTA TRAK assay significantly. In cases following after transurethral resection of bladder tumor (TURB), the sensitivity and specificity of the BTA TRAK assay for detecting recurrent bladder cancer were 100% and 79.5%, respectively. Conclusions: The BTA TRAK assay was more sensitive but less specific than voided urine cytology. Because gross hematuria affected the result of the BTA TRAK assay independently, it appears reasonable to delay the BTA TRAK assay until gross hematuria subsides in cases with gross hematuria. In cases following after TURB, the BTA TRAK assay appears to be useful for detecting recurrent bladder cancer.