[Question #5199] HIV Testing

23 months ago
Just curious in reading some of the post about the timing of HIV testing windows.  Why is it that a duo antigen/antibody test performed alone  is not absolutely conclusive at 28 days nor is a pcr/nat performed alone absolutely conclusive at 28 days but when performed together a negative result is absolutely conclusive.  Forgive my nonclinical mind here......
Edward W. Hook M.D.
Edward W. Hook M.D.
23 months ago
Thanks for your question.  For a number of complex mathematical reasons there is no such thing a "truth", in part because new observations on the universe of possible answers are being generated every day.  Thus all statements regarding what is conclusive are estimates.  By convention however there reach points at which the precision of estimates is so high that there is virtually no realistic possibility that the statement that "this is correct" will be wrong. thus while there is a possibility that you will be stuck by lightening before you finish reading this reply, or that you will be struck by a meteorite falling form the sky today, it could happen.  Further, the precision of answers go up when multiple different methods of looking at the same question are used.  With respect to HIV testing, current estimates are that far more than 99% of results for combination HIV antigen/antibody tests will be true and that only a tiny fraction of those tests will become positive after 4 weeks (I have never seen or heard of this happening).  Further, well over 99% of persons who recently acquired HIV will have  positive PCR tests are day 28 (again, I have never seen an exception).  When you put these two independent measure of infection together and get the same result, there is no realistic chance that a person will have HIV.

I hope this helps.  If not, sorry but I am not going to go further into the mathematics behind these statements.  EWH
---
23 months ago
Ok.  I see the logic.  Guess I was thinking that if antigen/antibody were not detectable yet then the nat/pcr would always be positive (in an infected person).  Guess I thought that it 100% full proof at 28 days.  So a person's "odds" get better the longer they wait? ( i.e. 6 weeks).
Edward W. Hook M.D.
Edward W. Hook M.D.
23 months ago
The antigen n the combination test is a part of the virus.  The PCR is more sensitive than the antigen test.  No test is perfect but when you put two very good tests (the combo test and the PCR) together that work in different way at the right time after exposure you can be effectively totally confident in your results.

Yes, all tests become more strongly positive over time in untreated, infected persons.  EWH
---
23 months ago
So for my final followup per the rules, so a negative nat-pcr and negative antibody-antigen duo 28 days is absolutely conclusive that you are not infected with hiv......correct?  If one gets the same negative result at 42 days with the same combination of tests is the result "more"  absolutely conclusive?  Again thanks for your time and your education
Edward W. Hook M.D.
Edward W. Hook M.D.
23 months ago
Correct, on both counts.  In general, because the PCR tests cost so much more than the far more widely used combination antigen/antibody tests, they are not used routinely.  If you had had both tests at day 28, you do not need additional testing.

As you point out, this will be the final reply as part of this thread.  I hope my comments have been helpful.  Take care.  EWH
---