Among women enrolled in this study, the overall prevalence of hr HPV infection was 19.6% and the prevalence of multiple hr HPV infection was 5.4%. The prevalence of hr HPV infection in HIV positive women (24.5%) was significantly higher than the prevalence in HIV negative women (15.9%; OR: 1.7; 95% CI: 1.1-3.1). The most common hr HPV strains were 16 (3.9%), 35 (3.5%), 58 (3.3%) and 31(3.3%), however it was only HPV 16 (OR: 3.1; 95% CI: 1.3-6.3) and 35 (OR: 3.6; 95% CI: 1.4-6.9) genotypes that were found to have significant association with HIV infection. HIV positive women with a CD4 count of < 200 were found to be at higher risk of hr HPV infection than their counterpart with higher CD4 cell count (OR: 2.4; 95% CI: 1.7- 5.9). The use of antiretroviral drugs (OR: 0.4; 95% CI: 0.3-0.5) and CD4 cell count above 500 cells (OR: 0.7; 95% CI: 0.5-0.8) were found to be associated with a lower HPV prevalence than HIV positive women not on therapy or with CD4 count <500 cells respectively.
The overall hr HPV prevalence of 19.6% though similar to 18.3% reported from Ibadan, Nigeria , is higher than 14.6% reported from Irun, a rural and agrarian community in Osun state of Nigeria , suggesting a possible lower prevalence of hr HPV infection in rural communities. The present study further supports the reported rural–urban differences in HPV prevalence as the prevalence of hr HPV in rural communities (16.1%) was significantly lower than 23.9% in the urban settings. This finding is also confirmed by previous research in low-income countries, including Nigeria, that the prevalence of sexually transmitted infection is higher in urban than in rural areas [32, 33]. This may be related to varying sexual behavior, lifetime sexual partnership and practices in the two settings or selection bias in various studies.
HPV 16, 35, 58 and 31 were the most common hr HPV genotype detected in this study, which is similar to findings reported by previous studies in our sub region [23, 24, 34]. However, the findings differed from the hr HPV distribution patterns described in other sub-Saharan African regions, Americas, Asia and Europe [2, 3, 9, 21, 23–25, 27, 28, 33],. While HPV 16 and 18 were the commonest genotype detected in Europe and North America [3, 9, 32], in East Africa HPV 52 and 16 were the most common [2, 3, 9, 25, 27]. In southern Africa, though the predominance of HPV 16 and 18 were retained, HPV 16 lost its position as the most common next to HPV 18 [2, 27]. This confirms the assertion by Ngandwe et al. that in sub-Saharan Africa, other hr genotypes other than HPV 16 and 18 may play a more major role in evolution of cervical cancer than previously thought . However this needs to be further confirmed in samples of women with invasive cervical cancer.
In the studied sample, the prevalence of hr HPV and multiple hr HPV infection were found to be significantly higher in HIV positive women (24.6% and 8.2%) than in HIV negative women (15.9% and 3.9%) respectively (OR: 1.7; 95% CI: 1.1-3.1, OR: 2.6; 1.3-5.3). However statistically significant association with HIV infection was noted only for hr genotypes HPV 35 and 16 after controlling for confounders (see Table 4). Women infected with HIV were 3.6 and 3.1 times likely to be infected with HPV 35 (OR: 3.6; 95% CI: 1.4-6.9) and 16 (OR: 3.1; 95% CI: 1.3-6.3) respectively than HIV negative women. The reported higher hr HPV prevalence in HIV infected women noted in this study were similar to previous observations from other regions of Africa, south America and Europe connoting that HIV infected women are at increased risk of hr HPV infection [2, 3, 13, 26, 27, 34–37]. The observed increase in prevalence has been attributed to HIV related immunosuppression [6, 32, 38]. Strickler and colleagues confirmed in two separate studies an increased prevalence of HPV, persistence and decreased resolution of HPV infection in HIV positive women compared to HIV negative women [18, 39]. Clifford and colleagues showed in their meta-analysis involving 20 publications that apart from viral types, immune-suppression of HIV infection significantly increased the risk for HPV infection and invasive cancer . In addition, the reported more frequent multiple HPV types in HIV-positive women was consistent with previous reports [8–10], and may be attributable to the common mode of transmission of HPV and HIV; persistence of HPV as a result of the inability to clear HPV infections; as well as reactivation of latent HPV infections .
The higher prevalence of hr HPV infection in HIV positive women with CD4 count <200 cells/mm3 and significantly lower prevalence in those with CD4 count >500 cells found in this study further confirms the results of several studies that have consistently shown a higher prevalence of HPV infection and greater persistence of HPV infections in HIV positive women with CD4 < 200 cells/mm 3 [4, 6, 18, 38, 39]. The noted associations are reportedly due to severe immune-deficiency. In the presence of severe immunodeficiency the body’s capacity to suppress latent infection is impaired and the risk of HPV infection is increased [17, 18, 39, 40]. HIV infection could also impact on the natural history of HPV by impairing the virus’s ability to escape the immune system [4, 17, 18, 40].
In this study we were unable to confirm the earlier reported role of high HIV viral load on the prevalence and burden of hr HPV infection [4, 10], as the initial association of hr HPV infection and viral load >10,000 copies (OR: 3.1; 95% CI: 1.1-9.9) was not retained after controlling for confounders in the 2nd and final models (see Table 5). The reported association in the other studies may be due to confounders of HIV treatment, CD4 cell count and sexual behaviour which were controlled for in this study [4, 10, 18, 38].
A statistically significant association was found between hr HPV and use of antiretroviral drugs (ART) as HIV positive women on antiretroviral therapy were found to be at lower risk of acquiring hr HPV infection than those not on antiretroviral drugs (OR: 0.4; 95% CI: 0.3-0.5). This finding supports previous studies and also the hypothesis that ART has the potential to restore the immune response against HPV [40–44]. The use of ART may therefore reduce HPV persistence and reduce the occurrence of HPV infection and even favour regression [8, 21, 40].
The combined relatively low prevalence of HPV vaccine related HPV 16 and 18 reported in this study were comparable both in HIV positive and negative women. Also the combined prevalence in this study corroborate other findings reported from our sub-region [23, 24, 34], suggesting that the currently available vaccine based on HPV 16 and 18 will be useful in only a minority of women irrespective of their HIV status, as genotypes other than HPV 16 and 18 were found to constitute the majority of the circulating hr HPV genotypes.
Our results need to be interpreted with caution, as it was conducted in 2 states out of 36 states in Nigeria. However, Lagos as the commercial capital of Nigeria and former administrative capital is the sociocultural, behavioural and diverse ethnic melting point of the country. As such information obtained may be generalizable to the entire country. Secondly, being a cross-sectional study, it may have missed possible variations of specific HPV types overtime. However, as the study was conducted over a one year period this concern may have been taken care of. In addition, the recruitment of about 90% HIV positive group from the clinic and majority of HIV negative women from the community may have introduced some selection bias. This method of recruitment was used because of low HIV prevalence in Lagos and Ogun states. The HIV positive women in the HIV treatment Centre were used in this study because over 65% of them reside in the same setting as the women recruited in the community. Secondly if this strategy was not adopted, sufficient number of HIV positive women and those on antiretroviral therapy will not be recruited. In addition, there were no statistically significant differences in the sociodemographic characteristics between the two groups except in their HIV status. Mindful of the possible effect of this on the validity of our results, we controlled for the possible confounders in multivariate logistic regression using three models.
The major strength of this study is the recruitment of women of diverse characteristics including HIV positive and negative women from rural as well as urban areas, those on treatment and not on treatment, thus increasing the feasibility of generalizing the findings. It is also the first study, to the best of our knowledge, in our sub-region that evaluated the effect of HIV infection and treatment on hr HPV prevalence and distribution. In addition two sample size estimates were calculated (prevalence and comparison of proportion in two groups). This assured us that the study had sufficient statistical power to detect the expected effects.