Background This study evaluated the associations of recent sexually transmitted infections (STIs) with cervical HPV prevalence among hormonal and non-hormonal contraceptive users.

Methods Data came from a prospective study conducted in 1046 women aged 20–38 years with normal cervical cytology in Thailand. We assessed whether baseline HPV prevalence was predicted by STIs which were newly detected and laboratory-confirmed within 2 years prior to enrolment. Prevalence ratios (PRs) with 95% CIs were estimated using generalised linear models.

Results Baseline prevalence of any HPV and high-risk (HR)-HPV were 19.9% and 8.7% respectively. Having genital chlamydia (CT) or gonorrhoea (NG) in the past 2 years was associated with increased risk of any HPV as well as HR-HPV infection after controlling for current and past sexual behaviours, age, contraceptive use and other concurrent STIs [adjusted PRs (aPRs) for any HPV: CT: 1.7 (95% CI 1.1 to 2.7), NG: 1.8 (95% CI 1.1 to 3.1); aPRs for HR-HPV: CT: 2.9 (95% CI 1.3 to 6.5); NG: 3.4 (95% CI 1.7 to 6.7)]. Association between CT and prevalent HR-HPV was statistically significant only among non-hormonal contraceptive users [aPR: 2.7 (95% CI 1.2 to 6.3)] but not among those using hormonal contraceptives in the past 2 years [aPR: combined oral contraceptive (COC) users: 1.1 (95% CI 0.5 to 2.4); users of depot medroxyprogesterone acetate (DMPA): 1.1 (95% CI 0.4 to 3.3)] (Abstract P1-S5.32 table 1). Association of NG with prevalent HR-HPV was observed among those who used COC [aPR: 6.2 (95% CI 2.2 to 17.7)] or DMPA [aPR: 3.5 (95% CI 1.1 to 10.9)] during the past 2 years but not among non-hormonal contraceptive users [aPR: 1.9 (95% CI 0.3 to 10.3)] (Abstract P1-S5.32 table 1). No significant association was found between other STIs and baseline prevalence of HR-HPV in this cohort.

Conclusions The differential impact of recent hormonal contraceptive use on the associations of CT and NG with HR-HPV prevalence suggests that the observed correlations may be attributed to biologic interactions between the pathologies of HPV and CT or NG, and not merely residual confounding by shared sexual risks.