Prospective analysis of voided and barbotage urine cytology in the routine follow up of non-muscle invasive transitional cell carcinoma of bladder

INTRODUCTION: Bladder cancer is one of the commonest malignancies in humans, with an estimated 200,000 new cases per annum world wide.It is two and half times commoner in males than females. Its incidence, like most other malignancies, increases with age, though it can occur in any age. Several important risk factors have been identified for bladder cancer including cigarette smoking, exposure to chemicals such as aniline dyes, benzidine compounds, aromatic amines; “slow acetylators” metabolic phenotypes and the presence of chronic inflammation or infection of the bladder. Histologically, greater than 90% of bladder cancers are transitional cell carcinomas; squamous cell cancers and adenocarcinomas constituting 5% to 6% and 1% respectively. Around 70% of all transitional cell carcinomas are classified as superficial lesions i.e., they do not invade more extensively than into the lamina propria. They comprise of a heterogeneous group ranging in both histologic grade (low or high) and stage Ta confined to the mucosa, T1 invasive into the lamina propria, or CIS - carcinoma in situ. Even with early adequate treatment there is an overwhelming propensity of carcinoma bladder to recur. Up to 70% of superficial tumours recur within 5 years, a figure that rises to 90% in 15 years. Over 20% of the superficial tumours progress to invasive disease with a poor prognosis. This entails an intensive follow up protocol to detect recurrences at the earliest. Presently cystoscopy remains the primary diagnostic modality to detect carcinoma bladder. AIMS AND OBJECTIVE: 1. To assess the accuracy of urinary cytology of voided urine sample in predicting the recurrence of non-muscle invasive transitional cell carcinoma (TCC) of the bladder. 2. To assess the accuracy of urinary cytology of barbotage urine sample in predicting the recurrence of non-muscle invasive transitional cell carcinoma (TCC) of the bladder. MATERIAL AND METHODS: Study Design: Prospective Study. Duration of study: 1st August 2009 to 31st January 2011 (eighteen months). Patient Characteristics: One hundred and ninety two patients from the inpatients and outpatients, of the Department of Urology, CMC, Hospital, Vellore, Tamil Nadu were enrolled into the present study. Inclusion Criteria: 1. Patients with biopsy proven TCC undergoing follow up cystoscopy after initial diagnosis. 2. A functioning bladder, free of any concurrent non-bladder urological cancers. 3. Sterile urine culture. Exclusion Criteria8: 1. Patients with non-bladder urological malignancies. 2. Non-biopsy proven cases. 3. Gross haematuria. 4. Benign inflammatory or untreated infectious conditions. 5. Renal, bladder or ureteric calculi. 6. Recent history of a foreign body in the urinary tract. 7. History of bowel segment interposition in the urinary tract. Informed consent was taken in all cases. The urine was cultured to detect urinary tract infection. If the culture was positive, an appropriate culture sensitive antibiotic was administered for 5 days. The study cohorts comprised 192 patients and were followed with office based cystoscopy and voided urinary cytology and barbotage urinary cytology. Of this 192 patients 35 came for review twice, 10 came for review thrice and only one had four reviews. Hence the numbers of cystoscopies were 250 and cytology specimens were 250 of voided urine sample and 250 of barbotage urine sample each. Patients were included at various stages of follow-up and the number of recurrences before inclusion varied from none to several. Voided urine sample and barbotage urine sample for cytology was obtained before cystoscopy. RESULTS: 192 patients were enrolled in the study. There were 172 males and 20 females. The patient’s age ranged from 27 to 81 years with a mean age of 55.57 years. The majorities, 71 patients (36.97%), were between 41 -50 years of age. All patients had previous, histologically confirmed, TCC with no evidence of muscle invasion (stages Ta, T1 and/or CIS) and were followed with office based cystoscopy and voided urinary cytology. Patients were included at various stages of follow-up and the number of recurrences before inclusion varied from none to several. All 192 patients had non-muscle invasive TCC of the bladder. CONCLUSION: Cystoscopy, the current gold standard for bladder cancer detection, and follow up of TCC bladder exhibits high sensitivity and specificity will continue with its numero uno status. Urine cytology is an established method and there is sufficient evidence to show its high specificity. However the method suffers from low sensitivity. In the present study the sensitivity of 23% and 57% for the voided and barbotage urine cytology respectively strongly questions the need for these tests during the follow up. There would be no harm done if urine cytology is discontinued as a test for TCC bladder during the follow up of low grade nonmuscle invasive TCC. Barbotage urine cytology showed higher sensitivity than voided urine cytology, but was not significant enough to recommend for follow up of the TCC bladder. Also barbotage urine cytology is invasive and in it self can cause changes in the morphology of the cells affecting its sensitivity. In conclusion one could safely discontinue the protocol of urine cytology in the follow up of low grade TCC bladder as cystoscopy which is more sensitive is mandatory and is the standard of care during the follow up of TCC bladder. Barbotage urine cytology also showed poor sensitivity. In this study the barbotage urine cytology sensitivity (50%) was lower than the voided urine cytology (70%) for high grade tumours which was contrary to all the previous studies. Also it is an invasive test which can increase the morbidity by introducing infection and it is better if one can avoid this invasive test with poor sensitivity.