Rapid HIV Testing in Emergency Departments - Beware of False Positives
Carlos del Rio, MD
In a low-prevalence ED, rapid tests often ruled out HIV, but, when reactive, they often yielded false-positive results.
In 2006, the CDC recommended routine HIV testing for all adults and adolescents in all healthcare settings, including emergency departments (EDs; AIDS Clin Care Oct 6 2006). This recommendation, coupled with the availability of rapid HIV tests, has resulted in a variety of efforts to increase HIV screening in EDs. However, many concerns remain (JAMA 2008; 300:945), such as the feasibility of testing in this setting, mainly because of the resources required (AIDS Clin Care Jan 14 2008), and the specificity of rapid oral-fluid tests (AIDS Clin Care Jul 7 2008).
In the present study, investigators evaluated the performance of the OraQuick Advance Rapid HIV-1/2 Antibody Test in the context of a clinical trial conducted in a busy ED in Boston. All rapid tests were performed using oral fluid, and all reactive tests were followed by a Western blot, an EIA, and measurement of both CD4-cell count and plasma viral load. During an 8-month period in 2007, 2356 ED patients were invited to participate in the HIV testing trial; 1397 (59.3%) agreed.
Of the 854 who actually underwent rapid testing, 849 had interpretable results. Thirty-nine patients (4.6%) had reactive tests. Of the 31 who agreed to confirmatory testing, only 5 were found to be HIV-infected, for an overall HIV prevalence of 0.6%; the others were catalogued as not infected. In this setting, the estimated specificity of the test was 96.9% (95% confidence interval, 95.7%-98.1%), which was significantly lower than the specificity reported by the manufacturer (99.8%; 95% CI, 99.6%-99.9%).
Comment
This HIV testing program was considered a success, because many patients who otherwise would not have been tested left the ED knowing that they were not HIV-infected. However, we clearly need better ways to confirm or rule out the presence of HIV infection in patients whose rapid tests are reactive. The authors suggest including viral-load testing in the panel of confirmatory tests, along with Western blot and EIA.
Carlos del Rio, MD
In 2006, the CDC recommended routine HIV testing for all adults and adolescents in all healthcare settings, including emergency departments (EDs; AIDS Clin Care Oct 6 2006). This recommendation, coupled with the availability of rapid HIV tests, has resulted in a variety of efforts to increase HIV screening in EDs. However, many concerns remain (JAMA 2008; 300:945), such as the feasibility of testing in this setting, mainly because of the resources required (AIDS Clin Care Jan 14 2008), and the specificity of rapid oral-fluid tests (AIDS Clin Care Jul 7 2008).
In the present study, investigators evaluated the performance of the OraQuick Advance Rapid HIV-1/2 Antibody Test in the context of a clinical trial conducted in a busy ED in Boston. All rapid tests were performed using oral fluid, and all reactive tests were followed by a Western blot, an EIA, and measurement of both CD4-cell count and plasma viral load. During an 8-month period in 2007, 2356 ED patients were invited to participate in the HIV testing trial; 1397 (59.3%) agreed.
Of the 854 who actually underwent rapid testing, 849 had interpretable results. Thirty-nine patients (4.6%) had reactive tests. Of the 31 who agreed to confirmatory testing, only 5 were found to be HIV-infected, for an overall HIV prevalence of 0.6%; the others were catalogued as not infected. In this setting, the estimated specificity of the test was 96.9% (95% confidence interval, 95.7%-98.1%), which was significantly lower than the specificity reported by the manufacturer (99.8%; 95% CI, 99.6%-99.9%).
Comment
This HIV testing program was considered a success, because many patients who otherwise would not have been tested left the ED knowing that they were not HIV-infected. However, we clearly need better ways to confirm or rule out the presence of HIV infection in patients whose rapid tests are reactive. The authors suggest including viral-load testing in the panel of confirmatory tests, along with Western blot and EIA.
Carlos del Rio, MD
0 comments
Post a Comment