RESUMO
AIM: There is debate about whether the traditional three-tiered grading of anal intraepithelial neoplasia (AIN) should be replaced by a more reproducible two-tiered system. In this study, we review our experience with AIN to determine the most suitable classification system. METHOD: We performed a retrospective review of all histological reports over a 19 year period. All specimens were graded on haemataloxin and eosin appearance and those with dysplasia had immunohistochemistry for p16 and Ki67 performed. RESULTS: Cases included 25 condyloma acuminata, 11 dysplastic cases and 24 invasive squamous cell carcinomas. On review, 18 were classified as condyloma acuminata without dysplasia. Seven had AIN I, five had AIN II and six had AIN III when using a three-tiered system. All cases classified as dysplastic (n = 18) showed an increased proliferation index as measured by Ki67. p16 positivity was seen in all AIN III, two AIN II and none of the AIN I cases. Recurrence was not observed in any of the AIN I cases. Five of eleven AIN II and AIN III cases recurred or persisted at a similar, higher or lower grade. Both of the AIN II cases which recurred or persisted were p16 positive. None of the AIN II cases that were p16 negative recurred. Three of the p16-positive AIN III cases did not recur. None of the 18 AIN cases progressed to carcinoma. CONCLUSION: The findings support the slow progression of AIN as described in the literature. In our small series, a two-tiered system with further subclassification of the traditional AIN II group using p16 appears to be clinically useful.