Sexually transmitted infections in a basic healthcare clinic in Dhaka, Bangladesh: syndromic management for cervicitis is not justified.

Abstract

Symptoms and signs of RTI as well as laboratory findings were strictly comparable among women attending the antenatal care (AC) family planning (FP) pregnancy interruption (PI) and expanded program on immunization (EPI) groups. Therefore data were pooled and compared with those obtained among women attending the clinic with complaints of vaginal discharge (VD group). Among women with and without signs of abnormal vaginal discharge the relative frequency of T vaginalis vaginal candidosis and bacterial vaginosis was 7% v 1% (OR 7.08; p<0.001) 21% v 6% (OR 4.30; p<0.001) and 29% v 19% (OR 1.78; p<0.001) respectively. The frequency of gonococcal/ chlamydial cervicitis was 2.7% v 2.1% (p=0.60). The sensitivity of the algorithm for bacterial vaginosis/T vaginalis infection and vaginal candidosis was 84% (79/94) and 69% (42/61) respectively with corresponding positive predictive values (PPV) of 37% (79/ 215) and 55% (42/76). Fourteen (5%) of 283 women with cervical mucopus versus 26 (2%) of 1558 without mucopus had gonococcal/chlamydial cervicitis. The sensitivity specificity and PPV of the algorithm were 35% 85% and 5% respectively. A husband not living at home or spending the night outside the house (n=200) as well as a husband suspected of being unfaithful (n=88) were the only two indicators associated with gonococcal/chlamydial cervicitis. However only 6% and 8% of the corresponding women had cervicitis. A polygamous marriage (n=151) was significantly associated with cervicitis in a bivariate analysis the association did not remain after logistic regression analysis. No association was observed between cervicitis and age income occupation of woman or husband a symptomatic partner vaginal douching after sexual intercourse the number of marriages of both the husband and woman symptoms of lower abdominal pain or the use of contraceptives. (excerpt)