Resumen de Anal cytology, histopathology and anoscopy in an anal dysplasia screening program: is anal cytology enough?
Marco Silva, Armando Peixoto, José Alexandre Sarmento, Rosa Coelho, Guilherme Macedo
Background and aim: The human papilloma virus is the leading cause of anal squamous cell carcinoma. Cytological screening may reduce the associated morbidity and mortality. The aim of the study was to estimate the agreement between anal cytological examination, histopathology and anoscopic visual impression. Methods: A prospective study of patients who underwent anal dysplasia screening between 2011 and 2015, in a proctology clinic of a tertiary referral center. Results: During the study period, 141 patients (91% men, 87% with HIV infection) underwent 175 anal cytology tests. Of these, 33% were negative for intraepithelial lesions or malignancy (NILM), 22% were atypical squamous cells of uncertain significance (ASCUS), 33% were low-grade squamous intraepithelial lesion (LSIL) and 12% were high-grade squamous intraepithelial lesion (HSIL). With regard to anoscopic visual impression, 46% of patients had no lesions and excision/biopsy of the identified lesions was performed in the remaining patients. The weighted kappa-agreement between abnormal cytological results and anoscopic visual impression was moderate (k = 0.48). The weighted kappa-agreement between simultaneous anal cytological examinations and anal histopathologic findings was low (kappa = 0.20). With regard to the histological examination of cases with HSIL or superficially invasive squamous cell carcinoma, 64% of patients had dysplasia of a lower grade according to the cytological analysis (6 ASCUS, 18 LSIL and 4 NILM). Conclusion: There was a poor correlation between anal cytology, histopathology and anoscopic visual impression and a high number of histological studies of HGD that were of a lower dysplastic degree according to the cytological examination. Therefore, anal cytology screening should not be used as the sole method of anal dysplasia screening
