Objective: Our purpose was to assess the impact of human immunodeficiency virus infection on pelvic inflammatory disease.
Study design: A case-control study was performed in Abidjan, Ivory Coast, women with pelvic inflammatory disease, 57 seropositive and 113 seronegative for the human immunodeficiency virus. Women underwent an interview, physical examination, pelvic ultrasonography, and laboratory testing.
Results: Seropositive women more often had an oral temperature > or = 38 degrees C (odds ratio 2.5, confidence interval 1.0 to 6.4), a genital ulcer (odds ratio 7.8, confidence interval 1.8 to 45.4), and a tuboovarian mass on ultrasonography (odds ratio 2.6, confidence interval 1.1 to 6.4) and were more likely to require surgery (odds ratio 6.5, confidence interval 1.1 to 67.5) and hospitalization (odds ratio 3.5, confidence interval 0.9 to 14.3). The mean clinical severity score was significantly higher among seropositive than among seronegative patients (17.4 vs 15.4 p = 0.01). Gonorrhea was detected in 50 (29.4%) and chlamydia in 16 (9.4%) of the 170 patients, and neither infection was significantly correlated with human immunodeficiency virus infection. After therapy similar proportions of seropositive and seronegative patients (95% and 93%) reported symptomatic improvement within 4 days, and none had persistent fever at day 4 or 14 of follow-up.
Conclusions: Human immunodeficiency virus infection was associated with more severe clinical manifestations of pelvic inflammatory disease but did not affect microbial cause or response to therapy.
PIP: During October 1992 to July 1993 in Abidjan, Ivory Coast, health workers conducted interviews, physical examinations, pelvic ultrasonography, and laboratory testing with 170 women with pelvic inflammatory disease (PID) at the University Hospital of Treichville and four primary care clinics. They compared clinical and microbiological characteristics and the response to PID therapy in 57 HIV seropositive women (cases) and in 113 HIV seronegative women (controls). Cases were more likely than controls to have a temperature of at least 38 degrees Celsius (odds ratio [OR] = 2.5), a genital ulcer (OR = 7.8), and a tuboovarian mass on ultrasonography (OR = 2.6) and to need surgery (OR = 6.5) and hospitalization (OR = 3.5). They also had a higher clinical severity score than did the controls (17.4 vs. 15.4; p = 0.01). Cases with a lower CD4 count (14%) were significantly more likely than cases with a higher CD4 count to have a temperature of at least 38 degrees Celsius (56% vs. 13-19%; p = 0.03) and dysuria (78% vs. 33-41%; p = 0.05). They also tended to have a genital ulcer and a tuboovarian mass, but not significantly so. Among all 170 women, the most common pathogenic organisms responsible for PID were Neisseria gonorrhoeae (29.4%) and Chlamydia trachomatis (9.4%). Neither infection was significantly related to HIV infection. Yet, the cause of PID in cases with the highest CD4 count was less likely to be N. gonorrhea than that of cases with lower CD4 counts (13% vs. 44%; p = 0.07). Among the 162 women who received oral antibiotics, 95% of the 40 cases and 93% of the controls responded to antibiotic therapy within four days. On days 4 and 14, none of these women still had a fever. These findings suggest that HIV infection affected clinical manifestations of PID but did not affect the cause of PID or response to therapy.