HSV window period

[tvb773]

I keep reading that the Herpeselect test is accurate at 12 to 16 weeks, but the following study from 2003 indicates that, for secondary HSV2 infections (i.e. contracting HSV2 after already having HSV1), only 82% seroconvert by 6 months (it is better for primary HSV2 infections -- i.e. no prior HSV1 infection -- at 93%).

If only 82% of those that contract HSV2 after already having HSV1 seroconvert by 6 months, how can they say that the test is accurate at 12 - 16 weeks?  Have there been subsequent studies that show short window periods and/or has the accuracy of the test improved?  I recall seeing posts here by HHH that say that its about 95% accurate at 3 months.  Where does this number come from?

Here's the link from the study by Anna Wald and others at the University of Washington (look at table 1):

http://journals.lww.com/stdjournal/F...t_ELISA.7.aspx

[Edward W Hook, MD]

Welcome to our Forum.  There are no perfect tests and estimates of test performance are in published studies are just that, scientifically based estimates.  There are a number of published studies on the performance of the HerpesSelect and both Dr. Handsfield and I have discussed the test’s performance directly with our friend, Dr. Wald as well other experts in the field.  All such studies are estimates of test performance and are compared to the Western blot assay as determined at the University of Washington in the lab that Dr. Wald is affiliated with.  Using data generated by them, it is described that, following the onset of HSV lesions, 93% of persons who do not already have HSV-1 infection and 83% of those who do have positive HerpesSelect assays at 6 months.  Whether and when the remainder develop antibodies is not described as such studies would be cost prohibitive to perform.  The estimates I used are based on these numbers.  Also, please remember that the estimates we provide to patients on this site are based on several things including our reading of the medical literature as specialists in STDs over that past 30+ years, our conversations with the people who write the papers and most importantly, at this site are tailored to the information provided to us by our clients  on the site.  

Hope this clarification is helpful to you.  EWH

[Edward W Hook, MD]

Your analysis outlined above demonstrates your lack of understanding of medical decison making.  Test results shold not be interpreted in a vaccuum but in the context of risk assessment, clinical findings, etc.  Thus, since acquistion of HSV is a relatively rare event at any particular time for any particular person, even at 6 weeks when "only" about 2/3's of tests are postive, the added information of a negative test is strong evidence that a person did not get herpes from the exposure/event of concnern.  Reducing the probablity of a rare event by another 66% makes it very unlikely.  EWH  

[tvb773]

So, given that so many people already have HSV1 (like half he population) and given that this study shows that it's only 83% sensitive at 6 months (for all of those people...93% sensitive for those without prior HSV1), it's hardly accurate/conclusive at 12-16 weeks (3 to 4 months).  Why does the the supplier and so many medical professionals give the 12-16 week mark as the accepted window period?