[Feasibility of cervical smear in HIV-positive women living in Chad]

Cervical cancer is the leading cause of cancerrelated death in Sub-Saharan African women. HIV-infected women are at increased risk for cervical intraepithelial lesions and invasive cervical cancer. WHO guidelines for screening and treatment of precancerous cervical lesions are regularly actualized. There are no data on cervical squamous intraepithelial lesions in Chad. Between August 2013 and May 2015, screening for cervical squamous intraepithelial lesions was proposed to HIV-infected women living in Moundou (Chad). Cytology examination was performed after with Papanicolaou coloration. Three hundred and eleven HIV-seropositive women accepted the screening without refusal. Mean age of the patients was 38 years (95% Confidence Interval: 37.7-39.9). The women declared a mean of 4.1 pregnancies (range: 0-12). The patients had been followed-up for their seropositivity for 8 years (range: 0-25). All were on highly active antiretroviral therapy (HAART). Of the patients whose results were known (N = 231), 98% had a CD4 lymphocyte nadir count less than 350/mm(3). Cytological results were as follows: normal smear (N = 59; 19%), inflammatory or hemorrhagic smear (N = 139; 44%), low grade squamous intraepithelial lesion (N = 58; 19%), high grade squamous intraepithelial lesion (N = 28; 9%), epidermoid carcinoma (N = 13; 4%), and uninterpretable smear (N = 14; 5%). The inflammatory lesions were due to cervicitis (N = 54), vaginosis (N = 22), and trichomonas infection (N = 3). The patients' age, CD4 lymphocyte nadir count, and CD4 count at the time of the cervical smear were not different according to the cytological results. Only five patients had a cone biopsy. Three patients deceased during the study of whom two from a gynaecological cancer diagnosed too late. The screening of dysplasia and cervical cancer in HIV-seropositive women is possible in Chad. In our study, 13% of the women had highgrade dysplasia or carcinoma needing curative care. We also showed that simple cytology did not permit the interpretation of half of the smears. The performance of cervical smear would have increased if it had been preceded by the visualization of the cervix with coloration.