Invasive cervical cancer in human immunodeficiency virus-infected and uninfected hospital patients

Objective: To compare the prevalence of invasive cervical cancer in women with, and in women without, human immunodeficiency virus (HIV) infection, so as to evaluate the inclusion of invasive cervical cancer in the AIDS surveillance case definition. Methods: The Sentinel Hospital Surveillance System for HIV Infection collected data and serum specimens that remained after clinical testing of persons who received inpatient or outpatient care at 14 hospitals with high HIV prevalence. We analyzed data on invasive cervical cancer obtained from medical record review and HIV serostatus from white, black, and Hispanic women in the age groups 20-34, 35-44, and 45-54 years. Results: In 1991 and 1995, 2684 (6.6%) of the 40,524 women sampled were HIV infected. Of the HIV-positive women, 28 had invasive cervical cancer (10.4 per 1000 women) and of the HIV-negative women, 236 had invasive cervical cancer (6.2 per 1000 women, relative risk [RR] 1.7, 95% confidence interval [CI] 1.1, 2.5). The prevalence of invasive cervical cancer was higher for HIV-positive than for HIV-negative black women aged 20-34 (RR 3.8; CI 1.7, 8.5) and Hispanic women aged 20-34 (RR 7.3; CI 1.1, 37.1) and 35-44 (RR 3.9; CI 1.1, 14.7) years. Twenty-six of the 28 cases of invasive cervical cancer in HIV-positive women were in women known to be HIV-positive during admission. Conclusion: The prevalence of invasive cervical cancer was higher for women who were HIV positive than for women who were HIV negative. This lends support to the inclusion of invasive cervical cancer in the revision of the surveillance case definition for AIDS in 1993.