ABSTRACT
PURPOSE: The incidence of human papillomavirus detection in the anal canal is rising. Efficient anal screening by cytology is hampered because of poor specificity. Human papillomavirus (HPV) testing is proposed in addition to Papanicolaou (Pap) testing for the detection of cervical neoplasia. The purpose of this study was to determine the usefulness of a human papillomavirus-DNA detection test to detect human papillomavirus-associated disease and to compare two different methods of sample collection. METHODS: In 555 patients, anal samples were obtained by using a cervical brush and a Dacron swab to test for high-risk and low-risk human papillomavirus-DNA. Patients positive for human papillomavirus-DNA underwent anoscopy. Biopsies were taken from visible lesions. RESULTS: Low-risk human papillomavirus-DNA was found in 325 of 555 patients (58.6 percent) and high-risk human papillomavirus-DNA in 285 of 555 patients (51.4 percent). Positive results confined to one single test method were higher for Dacron swab sampling (2.3 vs. 4.3 percent for low-risk human papillomavirus, P < 0.0001; 3.1 vs. 4.9 percent for high-risk human papillomavirus, P < 0.001). A positive correlation of relative light units was found for both sampling methods in the total cohort (P < 0.0001) as well as for patients who tested human papillomavirus-positive by both sampling techniques (P < 0.0001). Sampling with Dacron swabs yielded higher relative light units values compared with sampling with cervical brush for low-risk human papillomavirus-DNA and high-risk human papillomavirus-DNA. CONCLUSIONS: Anal screening for human papillomavirus-DNA by hybrid capture 2 is a useful method for detection of human papillomavirus-associated disease. Sample collection using Dacron swabs identifies more human papillomavirus-positive patients, and yields higher relative light unit values than using the cervical brush. Further studies are needed to determine the exact value of hybrid capture 2 in the screening for (pre)cancerous lesions of the anal canal.