Prevalence of human papillomavirus in oral gargles and tonsillar washings

Abstract

Human papillomavirus (HPV) infection drives carcinogenesis in the oropharynx. No standard sampling or HPV detection methods for evaluating oropharyngeal HPV infection exist. The prevalence of oral HPV infection in Japan is unknown. We examined 435 healthy Japanese individuals to address whether adding tonsillar washing to oral gargling would improve HPV detection. We compared HPV assessment using GENOSEARCH HPV31 versus nested PCR and direct sequencing. Associations between HPV infection and demographic and behavioral characteristics were examined. Most participants who were HPV-positive based on oral gargles were also HPV-positive based on tonsillar washings: 11 (64.7%) of 17 on nested PCR and 12 (70.6%) of 17 on GENOSEARCH HPV31. Although HPV infection was more prevalent in oral gargles followed by tonsillar washings than in oral gargles alone, the difference was not statistically significant (nested PCR, 4.8% vs. 3.9%, P=0.46; GENOSEARCH HPV31, 5.3% vs. 3.9%, P=0.33). The overall agreement between nested PCR and GENOSEARCH HPV31 was 98.6%, with 76.0% positive agreement. The overall prevalence of oral HPV infection in Japan was 5.7% (95% confidence interval, 3.9-8.3%). Men had a significantly higher prevalence of oral HPV infection than women (8.3% vs. 2.6%, P=0.02). Infection increased with number of lifetime sexual partners (P<0.001 for trend). The oropharynx is probably the major source of HPV-infected cells in oral gargles. Oral gargling could be a standard sampling method for evaluating oropharyngeal HPV infection. GENOSEARCH HPV31 could be an option for oral HPV detection.