Current issues of ACP Journal Club are published in Annals of Internal Medicine


Diagnosis

A rapid enzyme assay test was accurate for detecting HIV infection

ACP J Club. 1997 Jan-Feb;126:19. doi:10.7326/ACPJC-1997-126-1-019


Source Citation

Irwin K, Olivo N, Schable CA, et al., and the CDC-Bronx-Lebanon HIV Serosurvey Team. Performance characteristics of a rapid HIV antibody assay in a hospital with a high prevalence of HIV infection. Ann Intern Med. 1996 Sep 15;125:471-5.


Abstract

Objective

To evaluate the performance of a rapid, synthetic peptide enzyme assay that measures antibodies to HIV-1 and HIV-2 transmembrane glycoproteins in a routine hospital laboratory setting.

Design

Blinded comparison of Genie HIV-1/HIV-2 assay with a licensed whole viral lysate nonrapid enzyme assay and Western blot testing.

Setting

The emergency department and wards of a general hospital in the Bronx, New York, USA.

Patients

837 adults (53% Hispanic, 69% women) who were not known to be infected with HIV. Exclusion criteria were age < 18 or > 44 years, seeking HIV testing at the hospital, admission for conditions related to AIDS, inability to complete an interview or speak English or Spanish, being part of another study, or having been previously offered HIV testing by another hospital staff member.

Description of test and diagnostic standard

A trained technician tested fresh serum specimens with the Genie HIV-1/HIV-2 assay (Genetic Systems, Seattle, Washington, USA). The test was completed within 10 minutes. A technician who was unaware of the results of the rapid assay tested serum samples 2 more times using a licensed whole viral lysate nonrapid assay (Abbott Laboratories, Abbott Park, Illinois, USA). Reactive samples were confirmed by Western blot assay at the hospital; discordant samples were tested at the Centers for Disease Control and Prevention using frozen samples. Patients with indeterminate results on blot testing (n = 2) were excluded from the analysis.

Main outcome measures

Sensitivity and specificity.

Main results

45 of 837 adults (5.4%) were infected with HIV-1. Discordant test results occurred in 9 patients. No false-negative results occurred. The rapid assay had a sensitivity of 100% (lower 95% CI 92.1%), a specificity of 99.1% (CI 98.2% to 99.6%), {a likelihood ratio for a positive test of 111, and a likelihood ratio for a negative test of 0.0}.* These results did not vary appreciably after analysis of demographic and HIV risk factors: age, sex, race or ethnicity, birthplace (USA or other), time of enrollment in the study (technician experience), history of intravenous drug use, history of man-to-man sex, history of high-risk heterosexual activity, reactive serologic tests for syphilis, or history of any risk behavior for HIV infection.

Conclusion

The rapid enzyme assay (Genie HIV-1/HIV-2) done in the routine hospital laboratory was accurate for detecting HIV infection in an unselected hospital population.

Source of funding: Centers for Disease Control and Prevention.

For article reprint: Dr. K. Irwin, Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, 1600 Clifton Road, MS E-45, Atlanta, GA 30333, USA. FAX 404-639-6118.

*Numbers calculated from data in article.


Commentary

Better prevention and treatment of HIV infection has increased the importance of voluntary testing and counseling. Most physicians send specimens to laboratories that do nonrapid immunoassays and confirmatory tests. Results are usually available days to weeks later. The hiatus may lead to lost opportunities for counseling patients and targeted treatment of health care workers after occupational exposure.

Irwin and colleagues have made an important contribution to the literature on rapid HIV immunoassays. The study has no serious methodologic flaws. The sensitivity of the test was excellent, comparable with current reference standard tests. Similar results were reported in an evaluation of the only rapid assay that is currently available in the United States (Single Use Diagnostic System HIV-1 or SUDS test, Murex Corporation, Norcross, Georgia, USA) (1). Clinicians can use these rapid assays to quickly reassure almost all patients who test negative and counsel them on future prevention practices.

False-positive results with the rapid assays pose more problems and may be increased when tests are done by less experienced laboratory technicians. All results of the rapid assays require confirmation by standard nonrapid tests. Counseling strategies for positive results of rapid assay screening tests must take into account the psychological cost of the infrequent false-positive test results, which occurred in 0.8% of patients tested in this study.

Further evaluations of rapid HIV immunoassays in varied patient populations would be welcome. Currently, however, sufficient data exist to recommend a rapid HIV assay when the wait for nonrapid assay results obstructs HIV counseling and treatment efforts. Their use must be determined on a local level and should include consideration of a rapid assay when health care workers have a percutaneous exposure to blood. Attention to laboratory technician training and quality control will help ensure that test results agree with published test characteristics.

Howard A. Holtz, MD
Saint Barnabas Medical CenterLivingston, New Jersey, USA


Reference

1. Kassler WJ, Haley C, Jones WK, et al. Performance of a rapid, on-site human immunodeficiency virus antibody assay in a public health setting. J Clin Microbiol. 1995;33:2899-902.