A longitudinal study of human papillomavirus carriage in human immunodeficiency virus–infected and human immunodeficiency virus–uninfected women

Abstract

OBJECTIVE: We sought to determine the relationship of human immunodeficiency virus serostatus to carriage of oncogenic human papillomavirus. MATERIAL AND METHODS: A total of 268 human immunodeficiency virus–infected and 265 human immunodeficiency virus–uninfected women were seen every 6 months, at which time they had laboratory tests performed including a CD4 count. Human papillomavirus deoxyribonucleic acid was analyzed by polymerase chain reaction. Statistical methods included Kaplan-Meier and Cox's proportional hazard models. RESULTS: The prevalence at baseline of any human papillomavirus type was 73% and 43% among human immunodeficiency virus–seropositive and seronegative women, respectively ( p < 0.0001) and of oncogenic types was 32.5% and 17.0% ( p < 0.001). The prevalence of oncogenic human papillomavirus was higher in women with CD4 counts <200 mm 3 ( p < 0.001). The rate of detection of new oncogenic human papillomavirus per 100 patient years of follow-up in human immunodeficiency virus–seropositive women was almost three times higher than among human immunodeficiency virus–seronegative women ( p < 0.01). The rate of loss of an oncogenic human papillomavirus was higher in the human immunodeficiency virus–seronegative women but the difference was not significant. The relative risk of a human immunodeficiency virus–infected woman who did not initially have a specific type of oncogenic human papillomavirus having one detected during follow-up was 6.6 times greater than among human immunodeficiency virus–negative women ( p < 0.001). CONCLUSIONS: Human immunodeficiency virus–seropositive women are more likely to have newly detectable oncogenic types of human papillomavirus at follow-up and to show persistent carriage of oncogenic types of human papillomavirus types. Among human immunodeficiency virus–infected women, those with higher CD4 counts were more likely to have a newly detected oncogenic human papillomavirus during follow-up. (Am J Obstet Gynecol 1998;178:982-6.)