Search

In Reply. —We agree with Dr Ryder's conclusion that "a positive EIA and indeterminate Western blot may be merely a false alarm." Oral mucosal transudate and blood HIV testing are subject to outcomes that are uncertain and require follow-up testing to arrive at a definitive result. When indeterminate HIV test results are interpreted by experts, inferences about true HIV status may be drawn from the magnitude of the EIA test signal and from the nature (viral vs nonviral) of individual Western blot bands. Thus, the tone of the message during counseling might differ depending on whether the test result is likely to be a false alarm or is distinctly worrisome. The predictive value of the "EIA component" of HIV testing may be of academic interest, but the clinically relevant factor is the full testing algorithm. Oral mucosal transudate testing never resulted in a false-positive outcome. Just as with blood, there are occasional

Objective. —To determine accuracy of a human immunodeficiency virus type 1 (HIV-1) antibody testing system using a device to collect and stabilize oral mucosal transudate (OMT), a fluid with increased levels of IgG; an enzyme immunoassay (EIA) screening test optimized for OMT; and a Western blot confirmatory test designed for use with OMT. Design. —The OMT specimens were tested by EIA and, if indicated, confirmatory Western blot according to a standard testing algorithm. The OMT results were compared with true HIV status as determined by serum testing and/or clinical diagnosis. Patients. —Specimens from 3570 subjects (2382 at low risk, 698 at high risk, 242 with acquired immunodeficiency syndrome [AIDS], and 248 "nonspecificity" [persons with diseases associated with an increased frequency of false-positive results in HIV testing]) were collected at 11 geographically diverse sites (including blood banks, public health clinics, general medical clinics, HIV clinics, sexually transmitted disease clinics, and a hemophilia center) in the United States. Main Outcome Measures. —Overall accuracy of testing OMT for HIV-1 antibodies compared with true HIV-1 antibody status; sensitivity and specificity of OMT EIA and Western blot. Results. —Sensitivity of OMT EIA testing in 673 true-positive subjects was 99.9% (672/673). The OMT Western blot results in the 673 true-positive subjects were positive in 665 and indeterminate in 8. The EIA followed by Western blot (if EIA was repeatedly reactive) yielded a negative result in 99.9% (2893/2897) of OMT samples from true negatives and an indeterminate result in 4. The OMT testing system provided the correct result or would trigger appropriate follow-up testing in 3569 (>99.9%) of 3570 cases. Conclusion. —HIV-1 antibody testing of OMT samples is a highly accurate alternative to serum testing.

Objective. —To compare mother-to-child transmission of human immunodeficiency virus types 1 and 2 (HIV-1 and HIV-2, respectively) and to assess the impact of maternal HIV-1 and HIV-2 infections on child survival. Design. —Prospective cohort study. Setting. —Maternal and child health center in a lower socioeconomic class district of Abidjan, Ivory Coast. Participants. —A total of 18 099 women delivering between 1990 and 1992 were tested for HIV-1 and HIV-2 antibodies. A cohort of 613 pregnant women and their infants was followed prospectively (138 women reactive to HIV-1,132 reactive to HIV-2, 69 reactive to both viruses, and 274 HIV-seronegative). Main Outcome Measures.—Rates of perinatal transmission for HIV-1, HIV-2, and both viruses, determined from results of serological and polymerase chain reaction tests on children; survival of infants born to HIV-1—positive, HIV-2—positive, dually reactive, and HIV-seronegative women. Results. —Of the 18 099 women tested, 9.4% were reactive to HIV-1 alone, 1.6% to HIV-2 alone, and 1.0% to both viruses. The rate of perinatal transmission of HIV-1 was 24.7% (95% confidence interval [CI], 15.8% to 33.7%), compared with 1.2% (95% CI, 0.0% to 3.5%) for HIV-2 (relative risk, 21.3; 95% CI, 2.9 to 154.3). Overall, 19.0% (95% CI, 9.0% to 29.0%) of infants of dually reactive women became infected; of the 11 children concerned, 10 were infected with HIV-1 and one with HIV-1 and HIV-2. Infants of HIV-seropositive mothers had a reduced survival; mortality rates were 15.1, 13.0, 6.5, and 3.4 deaths per 100 child-years, respectively, for children of HIV-1—positive, dually reactive, HIV-2—positive, and HIV-seronegative women. Conclusions. —The rate of perinatal transmission of HIV-2 (1.2%) was much lower than the rate of perinatal transmission of HIV-1 (24.7%), and this was associated with more favorable survival for infants of HIV-2—infected mothers. Dually reactive women could transmit both viruses, although transmission usually involved HIV-1 only. Public health guidelines should incorporate advice that perinatal transmission of HIV-2 is rare. ( JAMA . 1994;272:462-466)