| |
Rapid HIV Test Misses Infections Spotted Later by Older Methods
|
| |
| |
2nd International Workshop on HIV Transmission
August 26-28, 2007
Washington, DC
Mark Mascolini
OraQuick, the rapid HIV test that first used finger-stick blood samples and now uses oral samples, missed several HIV infections among high-risk gay men in Seattle, according to results of a University of Washington study [1]. Earlier reports from California saw the opposite problem with OraQuick--an HIV diagnosis in people actually free of infection. But wider scrutiny suggested these "false-positives" were an isolated cluster [2].
In November 2005 Seattle public health authorities found an OraQuick-negative man with a positive HIV antibody test by a first-generation enzyme immunoassay (EIA). From that point all high-risk gay men with a negative OraQuick result got a first- or second-generation EIA. Public health officials also began targeted pooled nucleic acid amplification testing of high-risk gay men at public clinics in 2003. They defined "high risk" as (1) having unprotected sex with an HIV-infected or serostatus-unknown partner, (2) having a bacterial sexually transmitted disease, or (3) using methamphetamine or poppers in the last year.
Among 4208 gays who had a rapid HIV test, 104 (2.5%) had a positive result. Seven men (0.2%)--including the one originally identified in 2005--had a negative rapid test but a positive EIA. Fourteen men (0.3%) had a negative first- or second-generation EIA but a positive result from targeted pooled nucleic acid amplification. OraQuick gave a positive HIV reading for none of these 14 men.
Retrospective analysis of stored viral samples from 5 men with a negative rapid HIV test but a positive EIA rated two samples "weak positives" on the rapid test, two of them "positive," and one of them "strong positive."
Studies reported in the OraQuick ADVANCE package insert found the test has a diagnostic sensitivity of 99.3%. In other words, among 840 samples positive for HIV by standard testing ("true positives"), OraQuick identified 834 of them, and 834 divided by 840 equals 99.3%. In the University of Washington Study, OraQuick sensitivity fell to 93.7% if the "true positives" include all those positive by OraQuick plus EIA (104 divided by 111). Sensitivity drops to 83.2% if "true positives" include all those positive by OraQuick and targeted pooled nucleic acid amplification (104 divided by 125)
The University of Washington team concluded that "OraQuick may be less sensitive or more prone to operator error than first- or second-generation EIAs during early HIV infection." As a result, they added, "widespread use of OraQuick to screen populations at high risk for HIV acquisition may therefore have adverse consequences for public health unless HIV RNA screening is also performed."
References
1. Stekler J, Swenson P, Coombs R, et al. Decreased sensitivity of a rapid HIV antibody test during early HIV infection. 2nd International Workshop on HIV Transmission. August 26-28, 2007. Washington, DC. Abstract 40.
2. Jafa K, Patel P, Mackellar DA, et al. Investigation of false positive results with an oral fluid rapid HIV-1/2 antibody test. PLoS One. 2007;2:e185.
|
|
| |
| |
|
|
|
