I have been told that the PCR test is not supposed to be used for diagnostic purposes. I had a PCR test done for HIV a month after I had sex with someone who turned out to be a different person than I thought he was. The PCR test said “not detected”, but if it isn’t a diagnostic test, what good is it? Do I have to start over now?

A “not detected” HIV-1 DNA PCR at one month after exposure is an extremely reliable result, and there doesn’t seem to be any compelling reason for you to retest. The actual diagnosis of HIV infection is currently given on the basis of a positive ELISA antibody test which has been confirmed by a positive Western Blot test. Positive or reactive results on an antibody test are actually diagnostic because they confirm the presence of antibodies and therefore infer the presence of the virus itself. The PCR is actually diagnostic as well, but at the present time it has not yet been given this status by the CDC. The most likely reason that the PCR is not considered to be diagnostic at the present time is that it is so sensitive that there is still a possibility of a false “detected” result, with no immediate way to confirm its veracity. False positive ELISA results are possible as well, but the Western Blot is there to weed those out immediately. It should be noted that false “detected” results on the PCR have become fewer and fewer in recent years and are now quite rare.

Back to your original question, remember that a diagnosis requires that some disease or condition be present. As an example, one is “diagnosed “ as having diabetes, but is not “diagnosed” as not having diabetes. In my example, determining that someone does not have diabetes “rules out” diabetes. It does not actually “diagnose” that person as not having diabetes. This might seem like a picky detail, but it is actually critical to the understanding of the problem posed by your question. You might not want to diagnose HIV with a PCR, but there is no better or faster way to rule it out, and ruling it out is really what we want to do.

Take, for example, a chest x-ray done because there is a family history of lung cancer. We know that many chest x-rays show some dark spots, but only a few of these turn out to be cancer. We would not begin radiation and chemotherapy upon discovery of one of those dark spots without doing a biopsy and making a diagnosis of lung cancer first. This doesn’t make the x-ray a problem, just a very sensitive type of test. Likewise, we would not begin HIV treatment without doing an ELISA/Western Blot confirmation and giving a diagnosis of HIV. This doesn’t make the PCR a problem, just a very sensitive type of test. If you are worried about lung cancer, a negative x-ray will tell you that most likely you don’t have it, not that you definitely do have it. If you are worried about recent HIV exposure, and you do not want to wait three months for a conclusive result, the HIV DNA PCR can tell you that you were not infected, and that is primarily what you want to know.