Routine anal cytology screening for anal squamous intraepithelial lesions in an urban HIV clinic

Objectives: The purpose of this study is to describe our experience with routine anal cancer screening using anal cytology, determine risk factors for abnormal anal cytology, and determine if an association exists between cytology and histology in patients with HIV infection. Methods: Demographics, CD4(+) T-cell count, STD history, and cytology and histology data were extracted from medical charts of patients seen between November 1, 2002, and November 30, 2004. Analysis was done using chi(2) for comparison of proportions and Student t test for continuous variables. Multivariate analysis was conducted using logistic regression controlling for age, race, sex, CD4(+) T-cell nadir, and HIV exposure category. Results: Overall, 276 of 560 of the clinic patients received a screening anal cytology during the study period. Of these patients, 11 were excluded from the analysis and 74 of 265 (27.9%) patients screened had an abnormal anal cytology. Mean age was 44 years, and 68% were men. Forty-nine percent were African American, 34% Caucasian, and 17% Hispanic. Those with an abnormal cytology were more likely to be Caucasian (P = 0.03), and be homosexual or bisexual (P = 0.02). They were also more likely to have a lower CD4(+) nadir (142 cells/mm(3) vs. 223 cells/mm(3), p = 0.005) and CD4(+) at time of anal cytology (353 cells/mm(3) vs. 497 cells/mm(3), P < 0.001). Those with an abnormal anal cytology also had higher occurrence of anal disease on perianal visual inspection (30% vs. 9%, P < 0.001) and were more likely to have a history of genital warts (23% vs. 12%, P = 0.02) or herpes (35% vs. 22%, P = 0.02). Two patients had anal intraepithelial neoplasia (AIN) 1, 2 AIN II, 3 AIN III, and 2 squamous cell carcinoma in situ on histology. There was no apparent association between cytology and histology. Conclusion: Routine anal cytology screening is a feasible too] to incorporate into HIV care for patients regardless of gender and HIV risk factors. Its impact on morbidity and mortality warrant further study.