RTB: The PCR or Viral Load tests do not diagnosis infection, or conform with results from HIV antibody tests, as is seen in the case below. The M.D.s are clear that the test must be “cautiously interpreted within the context of the patient’s entire picture.” And that the test’s accuracy is interpretive, according to concepts of “probability.” Probability, not of being infected with a single or particular particle, but probability of being like other people who have the AIDS diagnosis.
A 59-year-old man known to have type 2 diabetes mellitus, alcoholism, hypertension, and chronic obstructive lung disease was hospitalized for headache and confusion.
The result on enzyme-liked immunosorbent assay for serum HIV was negative, whereas the result on testing for serum HIV-1 RNA by using reverse-transcriptase PCR was positive.
- [That is, he was negative on an antibody (protein) test, and positive on a genetic test. Results of repeated subsequent tests “were all negative”]
The HIV-1 PCR assay was designed to monitor HIV therapy, not to diagnose HIV infection. For diagnostic tests, prior probability of a positive test result needs to be considered. in patients (like ours) with a low prior probability of ideas, almost all positive test results are false positive.
Our urge to confirm the cause of acute encephalopathy rather than accept a diagnosis of exclusion resulted in inappropriate use of HIV-1 PCR.
This case confirms the importance of prior probability in diagnostic assays. We concur with Rich and colleagues that low-level positive results on HIV-1 PCR must be interpreted within the context of the patient’s entire picture.