Comparing Papanicolau smear, visual inspection with acetic acid and human papillomavirus cervical cancer screening methods among HIV-positive women by immune status and antiretroviral therapy.

BACKGROUND: A rigorous comparison of cervical cancer screening methods utilizing data on immune status, antiretroviral therapy (ART) and colposcopy-directed biopsy has not been performed among HIV-positive women. METHODS: Between June and November 2009, 500 HIV-positive women were enrolled at an HIV treatment clinic in Nairobi, Kenya, and underwent Papanicolau (Pap) smear, visual inspection with acetic acid (VIA), human papillomavirus (HPV) and colposcopy-directed biopsy (gold standard). Positive Pap smear (ASCUS+, LSIL+, HSIL+), VIA, HPV and their combinations were compared with CIN2/3+. Sensitivity, specificity and AUC (sensitivity and 1-specificity) were compared using pairwise tests and multivariate logistic regression models that included age, CD4⁺ cell count and ART duration. RESULTS: Of 500 enrolled, 498 samples were collected. On histology, there were 172 (35%) normal, 186 (37%) CIN1, 66 (13%) CIN2, 47 (9%) CIN3 and 27 (5%) indeterminate. Pap (ASCUS+) was the most sensitive screening method (92.7%), combination of both Pap (HSIL+) and VIA positive was the most specific (99.1%) and Pap (HSIL+) had the highest AUC (0.85). In multivariate analyses, CD4⁺ cell count of 350 cells/µl or less was associated with decreased HPV specificity (P = 0.002); ART duration of less than 2 years was associated with decreased HPV (P = 0.01) and VIA (P = 0.03) specificity; and age less than 40 years was associated with increased VIA sensitivity (P < 0.001) and decreased HPV specificity (P = 0.005). CONCLUSION: Pap smear is a robust test among HIV-positive women regardless of immune status or ART duration. Results should be cautiously interpreted when using HPV among those younger, immunosuppressed or on ART less than 2 years, and when using VIA among those aged 40 years or more.

Cervical squamous intraepithelial lesions among HIV-positive women on antiretroviral therapy in Kenya.

BACKGROUND: The prevalence of cervical squamous intraepithelial lesions (SIL) among HIV-infected women on antiretroviral therapy in sub-Saharan Africa has not been well described. METHODS: HIV-infected women enrolled in an HIV treatment clinic in Nairobi, Kenya were offered free cervical screening with Papanicolaou (Pap) smear testing if they were 30 to 39 years of age and on antiretroviral therapy. Women with SIL were compared to those without SIL with univariate analyses and logistic regression. RESULTS: Of 595 eligible women, 267 accepted Pap testing and had available cytology results, of whom 258 (97%) were on a non-nucleoside reverse transcriptase inhibitor (NNRTI) based regimen. Median duration of antiretroviral therapy was 13 months [interquartile range (IQR), 8-19]. Abnormal cytology was found in 123 women (46%) with 70 women (26%) having low grade squamous intraepithelial lesions (LSIL), 22 (8%) high grade squamous intraepithelial lesions (HSIL), 30(11%) atypical squamous cells of unknown significance (ASCUS) and 1 (0.4%) atypical glandular cells (AGC). Women with SIL had lower median CD4 cell count (239 vs 287 cells/mm3; P=0.02), lower income (<70 USD per month: 57% vs 38%; P=0.01), and less regular condom use (24% vs 40%; P=0.02) compared to those with no SIL. Duration and type of antiretroviral regimen were not significantly associated with SIL. CONCLUSION: SIL is prevalent among women on antiretroviral therapy and is associated with immunosuppression, low income, and less frequent condom use. Cervical cancer screening and counseling on condom use should be routinely offered to HIV-infected women in antiretroviral treatment clinics in Africa.