Oral sampling and human papillomavirus genotyping in HIV-infected patients.

Background: Oral human papillomavirus (HPV) is associated with several health complications especially in combination with HIV infections. Screening may be useful, but methodologies and results have varied widely in previous studies. We conducted a pilot study in an HIV-positive population to evaluate HPV detection in four different oral sample types.

Methods: Upon enrollment, an oral-rinse (OR) sample was collected in 10 ml saline. Additional samples of the buccal mucosa, tonsils, and oral lesion if present were collected with cytology brushes. DNA was extracted using LC-MagNAPure, and the Linear Array HPV genotyping Assay (Roche) was used for HPV genotyping.

Results: In samples from 100 HIV-positive participants, HPV was detected in 39 (%) of the oral rinses, 13 (%) mucosal and 11 (12.9%) tonsil brushings. Of seven lesion brushings collected, four were HPV positive. All participants with HPV detected in mucosal, tonsil, or lesion brushings were also positive in the OR sample. Among the rinse samples, 27 different genotypes were detected with HPV84 (n = 6), HPV55 (n = 5), and HPV83 (n = 5) being the most common. Multiple infections were detected in 17 samples (range 2-9, mean 1.9 types). As potential cofactors, only receptive oral sex was significantly associated with HPV (P = 0.018, odds ratio 2.9, 95% CI 1.2-6.9).

Conclusion: Sampling is a significant factor for oral prevalence studies. Oral rinse provides the best representation for HPV in the oral cavity. To evaluate associated cofactors other than receptive oral sex, larger studies with case-control design are necessary.