Résumé
BACKGROUND: The goal of this study is to assess the usefulness of immediate-cytospin Wright stain cytology with bladder tumor markers such as BTA and NMP22 in the screening and monitoring of bladder cancer. METHODS: We evaluated immediate-cytospin (cytospin 3, Shandon, UK) Wright stain and Papanicolau stain urinary cytology with the BTA stat (Bion Diagnostic Sciences, Inc., Redmond, WA, USA) and the NMP22 (Matritech, Newton, MA, USA) in 143 urine specimens for screening and monitoring of bladder cancer from September 2001 to May 2002. The cut-off decision level of NMP22 was determined by the receiver-operating characteristics (ROC) curve and those above the cut-off were classified as positive. The Wright and PAP stained urinary cytology results were reviewed by two experts. Negative and atypia cytology results were classified as negative, and suspicions of malignancy and malignancy results were classified as positive. RESULTS: The ROC curve analysis for NMP22 demonstrated optimal sensitivity and specificity (81.8% and 83.8%, respectively) with the cut-off decision level 11.0 IU/mL. Sensitivity, specificity, false positive rate, and false negative rate for the BTA stat were 62.5%, 92.9%, 47.4%, and 4.8%, respectively, and 81.8%, 83.8%, 55.0%, and 3.4%, respectively for the NMP22. Whereas, those values for immediate-cytospin Wright stain cytology were 90.9%, 98.4%, 16.7%, and 0.8%, respectively, showing a 100% concordance rate with Papanicolau stained cytology. The area under the ROC curve of the BTA stat, the NMP, and the immediate-cytospin Wright stain cytology was 0.78, 0.81, and 0.95, respectively. CONCLUSIONS: Immediate-cytospin Wright stain cytology was useful for the decrease in false positivity and false negativity for tumor markers such as the BTA stat and the NMP22 in screening and monitoring for bladder cancer.