# 1452
Normally, once a patient tests negative twice for the H5N1 virus, medical authorities assume the patient isn't infected with the virus.
Sometimes, however, a patient's symptoms will be so `classic' that even in the absence of a positive test, doctors will continue to treat the patient under the assumption that they have bird flu. We saw this in Turkey in 2006, and numerous times in Indonesia over the past year or two.
Since we've seen cases where patients have repeatedly tested negative (up to 5 or 6 times!) before finally testing positive, one can understand why doctors sometimes follow the old medical school adage to `treat the patient, not the medical test'.
Over the past 48 hours there have been a number of news articles in the Indonesian media describing a hospitalized patient, initials AT, who while testing negative, was strongly suspected of having bird flu.
Yesterday this patient died.
This from the Vietnam News Agency.
New bird flu suspected death reported in Indonesia
08/01/2008 -- 5:20 PM
Jakarat (VNA) - One Indonesian was suspected of dying from bird flu in Bandung, West Java, on January 7, local media reported.
Arlan Tamrin, 33, who was from Kabupaten Bandung, died after being treated since January 2, the Jakarta Post cited a doctor in the Hasan Sadikin Hospital as saying.
The doctor added that the patient experienced breathless difficult and high body temperature, symptoms similar to the flattest sign of bird flu.
If confirmed, this will be the 108th bird flu case in Indonesia . –Enditem
Since no positive test was ever obtained, and since autopsies are almost never performed in Indonesia (it violates both religious and cultural conventions), this case will likely go into the record books as `not bird flu'.
But based on the news coverage, AT's doctors apparently believed they were treating avian influenza up until the moment of his death.
Of course, a really severe pneumonia with accompanying septic shock would look a good deal like avian flu. Just because it walked like a duck disease . . . doesn't make it a duck disease.
This is one of those cases where we will probably never truly know.
When a patient presents with `bird flu-like' symptoms, they are immediately placed on antivirals (oseltamivir). It has been theorized that antiviral therapy `skews the testing' by suppressing viral shedding.
Again, this is something that while plausible, is hard to prove.
There are also worries that if the H5N1 virus were to mutate significantly (which is not proven to have happened) it would no longer match the current PCR primers, and thereby escape detection.
Also possible, I suppose.
And there are always concerns that samples may not be collected, packaged, or transported properly or that local labs might simply make a mistake.
Getting a positive test requires not only that the patient be infected, but that a whole host of other steps go exactly right.
So we shouldn't be terribly surprised if some positive patients go undetected. In fact, the surprise would be if the lab always got it right.
I've no idea whether AT was, in fact, infected with the H5N1 virus. All we know for sure is that his tests came back negative.
For whatever that's worth.
