| Literature DB >> 15243917 |
Philip E Castle1, Mark Schiffman, M Concepcion Bratti, Allan Hildesheim, Rolando Herrero, Martha L Hutchinson, Ana Cecilia Rodriguez, Sholom Wacholder, Mark E Sherman, Hortense Kendall, Raphael P Viscidi, Jose Jeronimo, John E Schussler, Robert D Burk.
Abstract
We compared point prevalences and determinants of human papillomavirus (HPV) DNA detection by testing enrollment vaginal specimens from hysterectomized women (n=569) and enrollment cervical specimens from nonhysterectomized women (n=6098) >or=30 years old, using MY09/MY11 L1 consensus-primer polymerase chain reaction. The subjects were participating in a population-based cohort study (n=10,049) in Guanacaste, Costa Rica, that was initiated in 1993. Non-cancer-associated HPV types, especially types 61, 71, and 72, were detected more frequently in the vaginal specimens from hysterectomized women (23.7% [95% confidence interval [CI], 20.3%-27.4%]) than in the cervical specimens from nonhysterectomized women (16.7% [95% CI, 15.7%-17.6%]) (P=.0001). There was no difference between the prevalences of cancer-associated HPV types in hysterectomized women and those in nonhysterectomized women; in both groups, the prevalence of HPV DNA was greater in women with multiple lifetime sex partners. We infer from our data that the cervical transformation zone may not be needed for cancer-associated HPV infection but may be uniquely susceptible to HPV-induced carcinogenesis; we also infer that specific phylogenetic groups of HPV (i.e., A3/A4/A15) may have a predilection for vaginal epithelium.Entities:
Mesh:
Substances:
Year: 2004 PMID: 15243917 DOI: 10.1086/421916
Source DB: PubMed Journal: J Infect Dis ISSN: 0022-1899 Impact factor: 5.226