Assessment of sensitivity and specificity of syndromic management of STIs when compared to laboratory diagnosis in HIV uninfected women aged 18 years and over, enrolled in the MDP301 clinical trial in the Umkhanyakude District of Northern KwaZulu Natal.
Abstract
Background: Sexually Transmitted Infections (STIs) are markers of unprotected sex, a risk factor for HIV. In the Umkhanyakude district of KwaZulu-Natal, South Africa, HIV prevalence in 25- to 29- year-old women is about 50%. In 1994 WHO recommended the use of syndromic management of STIs in resource poor settings; this was implemented in South Africa in 1995. This analysis seeks to measure the effectiveness of syndromic management in a cohort of HIV-negative women in a high HIV and STI prevalence area. Methods: At baseline, 1200 HIV-negative women enrolled in a trial of an anti-HIV vaginal microbicide were syndromically managed for STIs, but samples were also collected for laboratory testing for Chlamydia, gonorrhoea, trichomoniasis, and syphilis. False-positive syndromic management (over-treatment) and false-negative (under-treatment) results, together with positive and negative predictive values were calculated. Results: Preliminary results from the first 755 women give a laboratory diagnosed prevalence of 6.1% syphilis (95%CI:4.5-8.04%), 6.8% gonorrhoea (95%CI:5.1-8.79), 4.6% Chlamydia (95%CI:3.2-6.39%), and 13% trichomoniasis (95%CI:10-15.2%). 5% of women were treated on the basis of symptoms for Chlamydia or gonorrhoea, 5% for trichomoniasis, and 0.93% for syphilis. The sensitivity and specificity were 6.25% (95% CI:2.06-14) and 95% (95%CI:93-96.5) for Chlamydia and gonorrhoea, 0% (95%CI:0-7.71) and 99% (95%CI:98-99.6) for syphilis, and 10.5% (95%CI:5.16-18.5) and 95.6% (95% CI:93.8-97) for trichomoniasis. The positive predictive value of syndromic treatment of Chlamydia or gonorrhoea was 12.8% (95%CI:4.3-27.4%); for trichomoniasis it was 25.6% (95%CI:13-42.1%), and of syphilis 0% (95%CI:0-41%). The negative predictive value of syndromic treatment of Chlamydia or gonorrhoea was 89.5% (95%CI:87-91.7%); for trichomoniasis it was 88.1% (95%CI:85.5-90.4%), and for syphilis 93.9% (95%CI:91.9-95.5%). Conclusion: Syndromic management has low sensitivity resulting in untreated STIs, and low positive predictive value meaning women are being unduly exposed to treatment that may affect their vaginal flora. In a high HIV prevalence area we need to consider alternative STI treatment options as part of the HIV-prevention package.
Citation
Ndlovu, H.; Gafos, M.; Newell, M.L. Assessment of sensitivity and specificity of syndromic management of STIs when compared to laboratory diagnosis in HIV uninfected women aged 18 years and over, enrolled in the MDP301 clinical trial in the Umkhanyakude District of Northern KwaZulu Natal. Presented at AIDS 2008 – XVII International AIDS Conference, Mexico City, Mexico, 3-8 August 2008. (2008)