P0144 Diagnostic value of nuclear matrix protein-22 (NMP-22) compared with urine cytology in diagnosis of bladder carcinoma and muscular invasion

Abstract

Background We aimed to compare the diagnostic value of nuclear matrix protein-22 (NMP22) and urine cytology for the diagnosis of bladder carcinoma and its muscular invasion. Methods Cross-sectional studies were done with 27 urine specimens from patients with suspicion of bladder carcinoma who were admitted to Dr. Kariadi General Hospital, Semarang, between August 2010 and April 2011. Urine cytology and NMP22 tests were done on the samples, and sensitivity, specificity, positive and negative predictive values, and accuracy were compared with histology results from cystoscopy biopsy-TURBT. Table 2X2 and SPSS 17.0 for Windows were used for statistical analyses. Findings Twenty-seven urine samples were investigated; three were excluded because no further cystoscopy biopsy was performed. NMP22 tests were positive in 21 (87.5%) samples, and negative in three (12.5%). Urine cytology were positive in eight (33.3%), and negative in 16 (66.7%) samples. From cystoscopy biopsy-TURBT investigations, two samples were non-bladder carcinoma, and 22 were bladder carcinoma, of which 19 (86.36%) were muscle-invasive bladder carcinomas and three (13.64%) were non muscle invasive. Sensitivity, specificity, positive predictive value, and negative predictive values of NMP22 in detecting bladder carcinoma were: 90.9%, 50%, 95.28%, and 33.3% respectively, compared with 38.1%, 100%, 18.75%, and 45.83%, respectively for urine cytology. Sensitivity, specificity, positive predictive value, and negative predictive value for NMP22 in predicting muscular invasion were 94.73%, 66.67%, 94.73%, and 66.67%, compared with 36.84%, 66.67%, 87.5%, and 14.28% for urine cytology. Interpretation NMP22 had better ability than urine cytology in detecting bladder carcinoma. This investigation could be useful in screening for bladder carcinoma, particularly in patients with high-risk factors, and to monitor recurrences in low grade bladder carcinoma which are undetectable with urine cytology. This investigation is simple, fast, inexpensive, and non-operator dependent. We aimed to compare the diagnostic value of nuclear matrix protein-22 (NMP22) and urine cytology for the diagnosis of bladder carcinoma and its muscular invasion. Cross-sectional studies were done with 27 urine specimens from patients with suspicion of bladder carcinoma who were admitted to Dr. Kariadi General Hospital, Semarang, between August 2010 and April 2011. Urine cytology and NMP22 tests were done on the samples, and sensitivity, specificity, positive and negative predictive values, and accuracy were compared with histology results from cystoscopy biopsy-TURBT. Table 2X2 and SPSS 17.0 for Windows were used for statistical analyses. Twenty-seven urine samples were investigated; three were excluded because no further cystoscopy biopsy was performed. NMP22 tests were positive in 21 (87.5%) samples, and negative in three (12.5%). Urine cytology were positive in eight (33.3%), and negative in 16 (66.7%) samples. From cystoscopy biopsy-TURBT investigations, two samples were non-bladder carcinoma, and 22 were bladder carcinoma, of which 19 (86.36%) were muscle-invasive bladder carcinomas and three (13.64%) were non muscle invasive. Sensitivity, specificity, positive predictive value, and negative predictive values of NMP22 in detecting bladder carcinoma were: 90.9%, 50%, 95.28%, and 33.3% respectively, compared with 38.1%, 100%, 18.75%, and 45.83%, respectively for urine cytology. Sensitivity, specificity, positive predictive value, and negative predictive value for NMP22 in predicting muscular invasion were 94.73%, 66.67%, 94.73%, and 66.67%, compared with 36.84%, 66.67%, 87.5%, and 14.28% for urine cytology. NMP22 had better ability than urine cytology in detecting bladder carcinoma. This investigation could be useful in screening for bladder carcinoma, particularly in patients with high-risk factors, and to monitor recurrences in low grade bladder carcinoma which are undetectable with urine cytology. This investigation is simple, fast, inexpensive, and non-operator dependent.