Purpose: The subclassification of T1 bladder tumors into T1A and T1B has an important prognostic significance and a great impact on patient management. Unfortunately, staging T1 tumors is highly subject to interpathologist variation that can be critical for patients included in randomized clinical trials. To determine the value of immunohistochemistry (IHC), such as desmin and keratin, in comparison to hematoxylin-eosin (H&E) in classifying T1 stage disease, we retrospectively examined 93 consecutive cases diagnosed at our department. Materials and Methods: The study was conducted in two phases (H&E then IHC), each in two time periods. First H&E, and then IHC slides were reviewed independently by two experienced pathologists and discrepant cases from each phase were discussed between the two pathologists to reach a final decision. Results: The two methodologies (H&E and IHC) showed total agreement in 76 out of 93 cases. IHC downstaged seven cases, that is from T1B to T1A, upstaged four cases, that is from T1A to T1B, lowered the rate of imprecision and eliminated the disagreement between the two pathologists. However, IHC failed to subclassify T1 tumors in three cases. Finally, the discussion supported by the IHC was very useful in reaching the diagnosis in some cases. Conclusions: IHC appears to be a useful tool in staging T1 bladder cancer, especially in difficult cases where specimen orientation and artifact could create a major hindrance in reaching an accurate diagnosis. (C) 2002 Elsevier Science B.V. All rights reserved.