In this study 95 out of 355 (26.8%) HIV infected women had cervical lesions (VIA positive) that warranted histological evaluation. This finding was lower than that observed in a study conducted by Balandya et al in Dar es Salaam whereby 42.2% were Aceto white positive . This difference can be explained by the a relatively lower median CD4 count of their population which was 351cells/mm3compared to this study where the median CD4 count was 450cells/mm3. The lower the CD4 count and advanced stage of HIV could be responsible for a higher prevalence of cervical lesions seen in their study . In the present study, the prevalence of CIN proven by biopsy was 16.0%. Our findings are consistent with a work reported by Miotti et al among 268 postpartum women of Malawi, whereby the prevalence of CIN was 15% . Although the prevalence of CIN in this population appears to be low, it is within the range described in similar publications in Africa [6, 21]. A Botswana study, among HIV infected women using VIA showed a prevalence of 11%, and another study among HIV infected women in Senegal showed a prevalence of 10% [22, 23]. In Africa, a high prevalence of CIN (76%) among HIV infected women has been observed in Lusaka Zambia . The probable explanation could be a HIV induced immune-suppression as the median CD4 count was low (165 cells/mm3) compared to our study in which the median baseline CD4 count and current CD4 count were 269 cells/mm3 and 450 cells/mm3 respectively. Other studies in Tanzania reported a higher CIN prevalence of 32% and 38% [15, 17]. In those studies, conventional cytology was used as a screening method, and biopsy was not done.
The prevalence of CIN in the present study was higher than that found in a study by Mayaud et al among antenatal clinic attendees at the Makongoro clinic Mwanza which was 7%. The difference can be explained by the fact that Mayaud’s study involved the general population at the clinic while the present study involved newly diagnosed HIV participants and those who were in follow up for HIV status in tertiary hospital . VIA was offered to 355 women and 26.8% of these women had VIA positive results. This was similar to a study by De Vuyst et al among 653 women attending Nairobi family planning clinic, whereby 27% women had VIA positive results . The histological results showed 53.4% had CIN1 and 46.6% had CIN2/3, and cervical cancer in 7.4% women, while in the study done by De Vuyst et al CIN1 was seen in 25%women, CIN 2/3 in 38% women and invasive cancer in 4% women . In our study the frequency of chronic cervicitis with feature of HPV among HIV infected women was 27.4%, this is similar to a study by Mwakigonja et al which showed a cervicitis in 28%, which was dominated by feature of Chlamydia trachomatis .
The histological results of CIN and invasive cancer seen in this setting emphasize the importance of introducing a screening programme for cervical cancer in HIV infected women. Despite of being at risk of CIN, none of the participants in this study had previously been screened for cervical cancer. This is a major concern in many developing countries including Tanzania where cervical cancer screening is largely unavailable, leading to late presentation with advance cervical cancer .
The majority of women acquired HIV infections within a few years of sexual debut. In this study, an early sexual debut appeared to be associated with the development of CIN but was not statistically significant. This concurs with other studies done in Tanzania whereby an early sexual debut was not directly associated with CIN [14, 17]. In a study by Matasha et al involving female pupils in Mwanza, 68% of girls had an early sexual debut . An earlier age of sexual debut implies a longer period of sexual activity and a higher likelihood of having many sexual partners . Having multiple sexual partners exposes a woman to factors predisposing to CIN such as HIV and STI. This was supported by a study done among Nairobi prostitutes that showed that women with multiple sexual partners had an increased risk of developing CIN . A similar trend was seen in the current study, whereby women with many sexual partners (more than two) had more CIN compared to women with few number of sexual partners with significant difference (P< 0.001).
Co-infection with a STI like Chlamydia trachomatis, herpes simplex or genital warts in the presence of HPV increased the risk of CIN by causing inflammation which facilitates HPV persistence and hence cervical lesion and carcinogenesis [28, 29]. In the present study, a previous history of STI, including genital warts, was associated with CIN. A previous history of STI was reported to be a cofactor needed for the progression of HPV infection to cervical dysplasia and subsequently into cancer. Studies suggest that HIV induced immunosuppression leads to inability to control the HPV expression, hence the persistence of HPV infection and the development of cervical lesions . This study showed that genital warts were associated with an increase risk of CIN. A higher prevalence of HPV infection in Mwanza, found in the study by Mayaud et al could be a major contributor to the increased prevalence of genital warts which was found among HIV infected women in this study. Genital warts are caused by persistent low risk types of HPV infection; a history of genital warts in the absence of detectable virus still increases the risk of CIN, as shown by other studies [16, 31].
In the current study, women whose baseline CD4 count was less than 200 cells/mm3 were more likely to have CIN than those patients with a CD4 count of more than 200cells/mm3. Previous studies have consistently demonstrated that the amount of CD4 count of less than 200 cells/mm3 is a predictor for having or developing CIN [14, 17]. A study done by Wright et al in New York in 1994 showed that an age of more than 35 years was associated with a risk of CIN. In the current study, the association between age and CIN was not statistically significant. The same conclusion was reached seen in other studies conducted in Kilimanjaro and Dar es Salaam [14, 17].
Although we did not find a significant association between marital status and risk of developing CIN in the present study, a study done in Kilimanjaro showed that single women had a tendency to have multiple sexual partners, a factor which contributed to CIN .
Smoking and contraceptive use have been reported to be established cofactors for HPV progression to cervical lesion. Carcinogens produced by tobacco are thought to decrease the immunity in the cervix, thus facilitating HPV persistence . Furthermore, contraceptive hormones increase the expression of HPV genes in the cervix and facilitate HPV persistence . In the present study, the association between CIN and those who have been ever exposed to cigarette and ever used contraception was not significant. Similarly, there was no difference in CIN prevalence between Non ART and ART user. This finding is contrary to a study done by Mwakigonja et al which noted that the frequency of CIN and invasive cancer were higher in women who were non HAART than those who had been on HAART. Initiation of HAART has shown to have an impact on reducing the incidence and progression of CIN . A study by Balandya et al showed that women who had been on ART for more than 1 year had more abnormal cervical changes on VIA compared to those who were on ART for less than a year. The probable explanation is that despite an increased CD4 count with reconstitution of immunity there is no clearance of HPV in the cervix . In the current study, the duration since HIV diagnosis, was not associated with CIN, which is contrary to a study done by Kreiss et al which showed that women living with a HIV infection for more than two years were more likely to carry high risk HPV infections, which could result in CIN, compared to women who had been HIV infected for less than a year .