1. Home>
2. Specialties>
3. HIV/AIDS Clinical Care>
4. Research Notes

The Best Way to Diagnose Acute HIV Infection

Viral load and p24 antigen assays each have advantages and drawbacks in the diagnosis of acute infection.

The public health benefits of diagnosing HIV infection in its acute stage are clear, and the advantages of immediate treatment seem increasingly real (see Feature, January 2001 ACC). The most efficient and cost-effective way of making a diagnosis, however, remains unclear.

California researchers compiled data on 436 patients with symptoms suggestive of acute HIV infection who were referred for evaluation between 1993 and 1999. All patients were tested for HIV antibody with standard methodology, and for HIV antigen with branched-chain DNA (bDNA) or PCR viral load testing. Most also had p24 antigen assayed. State-of-the-art commercial kits were used for antigen testing (viral load measurements on some samples from the study's early years were performed retrospectively on frozen specimens).

Overall, 303 patients were determined not to be HIV infected; 79 were determined to have previously acquired, chronic HIV infection; and 54 were determined to have acute HIV infection on the basis of a repeatedly positive antigen test and either a negative ELISA or an indeterminate Western Blot. P24 antigen assays missed 5 patients with acute infection, for a sensitivity of 88% and a specificity of 100%, while viral load assays were falsely positive in 8 uninfected patients, for a sensitivity of 100% and a specificity of 97%. All false-positive viral loads were under 2,000 copies/ml (range, 52-1950 copies/ml), while all true positives were 100,000 copies/ml or more.

When clinical symptoms in 40 patients with acute HIV infection were compared with those in 164 without, no single symptom or constellation of symptoms clearly distinguished the 2 groups.

Do these data favor p24 or viral load as the antigen test of choice for screening possible cases of acute HIV? Because p24 antigen assays are roughly a third the cost of viral loads, one might argue that the overall advantage falls to the cheaper of these 2 flawed tests. Two editorialists take this position, also emphasizing that the mental anguish of a false-positive result, and the subsequent need for counseling and further testing, reduces the overall usefulness of viral load assays.

However, the public health consequences of missing even one patient with acute HIV infection -- who may then fail to come back to medical attention for years or even decades -- would substantially detract from a policy of screening possible cases only with p24 antigen assays. At least from the data presented here, in which true-positive and false-positive viral load assays were orders of magnitude apart, viral load seems (when financially feasible) a useful adjunct to diagnosis.

— Abigail Zuger, MD

Published in AIDS Clinical Care March 1, 2001

Citation(s):

Daar ES et al. Diagnosis of primary HIV-1 infection. Ann Intern Med 2001 Jan 2; 134:25-9.

• Original article (Subscription may be required)
• Medline abstract (Free)

Flanigan T and Tashima KT. Diagnosis of acute HIV infection: It's time to get moving. Ann Intern Med 2001 Jan 2; 134:75-7.

• Original article (Subscription may be required)
• Medline abstract (Free)