Regarding Test Accuracy and Kenya

[jimjams321]

Hello Dr,

My first part of the question is a little hard to explain:

If a standard 3rd Gen HIV test is said to be, let's say, 80% reliable in detecting a rare HIV strain (HIV-2, Group O, or Group N doesn't matter) and 10 samples of one of these rare strains were tested,2 of them would be unreactive.  Now my question is, would the 2 strains that were not detected ALWAYS be undetected if retested with the same exact test or would there be an 80% chance each time?  In other words, would there theoretically be an 80% chance of detection for EACH sample EACH time tested?  Or, would there be some genetic differentiation in the undetected samples that would make it impossible for that particular test to find them regardless of retesting?

Second part of my question:

About 2 years ago, I had a onetime unprotected, vaginal intercourse with a girl whom I recently found out that her partner before me was from Kenya.  Since then, I have been tested with 4 standard ELISA tests and 1 PCR test, all negative.  So could it be feasible, since her previous partner was from Kenya, that she could have been exposed to HIV-1 Group N?  One of the tests I had said it detected both Group O and HIV-2, so no worries there.  I do know that Group N is mainly from Cameroon, but could it be possible that Group N has reached Kenya at all or in any significant numbers?

All in all, if I can be sure that there would be atleast an 80% chance of detection to this rare strain each time a test was administered (since 3 of the 4 ELISA tests were the same test), I would be satisfied that no matter what possible strain I could have possibly been exposed to, it would undoubtedly been detected.  After all, it would be 1/5 x 1/5 x 1/5 x 1/5 x 1/5 = 3125 or a 99.97% chance it would have been found.  (Btw, I read the information on the PCR test and it showed 7 out 8 Group N samples reactive.)

I'm getting married in July and need to move on from this worry.

[Edward W Hook, MD]

Welcome to our Forum.  I'll try to help.  You are worrying far too much about incredibly unlikely possibilities.  It is unlikely that the partner of your former partner had HIV or that if he did, that she was infected.  In turn it is still less likely that she became infected and infected you and that in becoming infected you were infected with a test which would not have been detected by either PCR or the standard antibody tests you took.  Your risk of infection using the scenario you outline is immeasurably low.  You are far more likely to be hit by a meteorite falling from the sky while reading this reply.

If yo cannot stop worrying about your risk for HIV from the exposure you describe you should discuss this illogical fear with a trained counselor or mental health professional.  I say this out of concern for you, nothing more.  Take care. EWH

[Edward W Hook, MD]

This is a repetitive, anxiety-driven question.  This will be the final answer. There is no reason for you to worry that you got HIV from the exposure you have descirbed.

The thread is over.  EWH

[jimjams321]

Then, in your honest opinion, it would not matter, even if I actually was exposed to a very rare strain, that after 5 tests there would be absolutely no possible way I could have anything?

[Edward W Hook, MD]

Yes, if a test missed an infection once, it would likely be detected when a person was re-tested.  EWH

[jimjams321]

Thank you very much. I appreciate your reply. So it would be your opinion that if a test missed an infection once there would still be a good chance it would be detected if tested again with the same test? If so. I would be quite satisfied.