Introduction: Carbohydrate antigen (CA 19-9) is a tumor marker that had been used for many types of cancer, and especially for gastrointestinal system. Studies directed for the diagnostic significance of CA 19-9 levels have been reported, under the light of high serum levels reported in some sporadic cases in urothelial cancers also. We evaluated the serum CA 19-9 levels of the patients with bladder tumor by comparing the characteristics of the disease in this study. Materials and methods: Fifty-seven patients who were diagnosed with bladder tumor between June 2001 and May 2004, and 20 patients who applied with hematuria and no malign tumors were found and received various treatments directed to prostate hyperplasia were recruited in the study. Patients with bladder tumor were studies in two groups, namely, superficial and invasive. Patients were grouped according to stage (superficial and invasive), grade (Grade I-II and Grade III), number of tumor foci (single or multiple foci), size of the tumor (>3 cm and <3 cm), and whether or not there were recurrences, and levels of serum CA 19-9 were examined. The limit value for CA 19-9 was taken as 37 U/ml. Intra-group comparisons were performed for all the groups and the two groups with and without bladder tumor were compared according to CA 19-9 levels with independent t-test. Values of P<0.05 were taken as significant statistically. Results: Twenty of the total 77 patients (22%) consisted the control group with benign pathologies. Twenty-two (35%) of the 57 patients with bladder tumors were in invasive stage (pT2 and higher stages), and 35 (65%) were in superficial (pTa and pT1) stages. The average age of the group with tumors was 64.35±6.28 years. The average age of the group applying to urology department with hematuria was found to be 62.34±10.9 years. There was no difference between the groups in terms of average age (p>0.05). Average follow up time of these patients was 10.1 months. While the serum CA 19-9 levels were found over the limit value in none of the patients in the control group, it was found to be positive in 10 (17.5%) of the patients with bladder tumor. 14 patients with tumor recurrence had higher CA 19-9 value than the patients with no recurrence disease (p=0.01). Conclusion: A tumor marker with suitable characteristics in terms of the diagnosis and follow-up of the bladder tumor is required. Our data indicate that CA 19-9 can be an indicator concerning the existence of the tumor; however, it has no adequate prognostic value when the known risk factors are considered. However, larger series and longer follow up time are required, especially for the analysis of the risk factors.