UCSF

Objectives

Human papillomavirus (HPV), one of the commonest sexually transmitted infections, may be a cofactor in HIV acquisition. We systematically reviewed the evidence for an association of HPV infection with HIV acquisition in women, heterosexual men and men who have sex with men (MSM).

Design

: Systematic review and meta-analysis.

Methods

Studies meeting inclusion criteria in Pubmed, Embase and conference abstracts up to 29 July 2011 were identified. Random effects meta-analyses were performed to calculate summary hazard ratios (HR). Publication bias and statistical heterogeneity were evaluated and population attributable fractions (PAFs) calculated.

Results

Eight articles were included, with previously unpublished data from five authors. Seven studies found an association between prevalent HPV and HIV acquisition. Risk of HIV acquisition in women doubled with prevalent HPV infection with any genotype [HR = 2.06 (95% CI = 1.44-2.94), I = 0%], although adjustment for confounders was often inadequate. The effect was similar for high-risk [HR = 1.99 (95% CI = 1.54-2.56), I = 8.4%] and low-risk [HR = 2.01 (95% CI = 1.27-3.20), I = 0%] HPV genotypes with weak evidence of publication bias (P = 0.06). Two studies in men were identified: both showed an association between HPV infection and HIV acquisition. Unpublished data from one of two studies in women indicated an association between genotypes targeted by HPV vaccines and HIV acquisition. PAFs for HIV attributable to infection with any HPV genotype ranged between 21 and 37%.

Conclusion

If further studies validate the association between HPV infection and HIV acquisition, HPV vaccines may reduce HIV incidence in high HPV prevalence populations, in addition to preventing cervical cancer. HIV surveillance studies during implementation of HPV vaccine programmes are warranted.