PELVIC INFLAMMATORY DISEASE AND HUMAN-IMMUNODEFICIENCY-VIRUS INFECTION

Objective: To identify the effect of human immunodeficiency virus (HIV ) infection on the clinical course of pelvic inflammatory disease (PID ). Methods: Women hospitalized with PID at an urban hospital serving a population at high risk for HIV were studied cross-sectionally. Data abstracted from medical records of 349 women, admitted between July 19 92 and April 1994 were linked anonymously to HIV serology. Main outcom e measures were length of hospital stay, prolonged fever, tube-ovarian abscess, surgery, and change in antibiotics. Results: Among the 349 w omen with PID, 27 were HIV-positive. These HIV-positive women had lowe r mean white blood cell counts at admission (7411 versus 11,266, P < . 01), lower mean lymphocyte counts (1411 versus 1928, P < .01), greater febrile morbidity (54 versus 28.3%, P < .01), and longer hospital sta ys (10.5 versus 6.4 days, P < .01) than HIV-negative women. Women who were HIV-positive required more time for defervescence and needed to c hange their antibiotic regimen more frequently (41 versus 12.7%, P < . 01); differences in tube-ovarian abscesses (19 versus 14%, P = .52) or surgery (15 versus 6.2%, P = .10) were not significant. The differenc es in hospital course between HIV-positive and HIV-negative women were modest, and they were resolved largely by the fourth or fifth hospita l day. All HIV-positive women were treated successfully with first- or second-line antibiotic regimens. Conclusion: Despite more severe init ial presentation and a prolonged hospital course, HIV-positive women w ith PID, but without other acute illnesses, were treated successfully with standard therapeutic regimens. These observations support current recommendations for hospitalization of HIV-positive women with PID an d treatment according to current standards. Copyright (C) 1997 by The American College of Obstetricians and Gynecologists.