Milder Bird Flu Cases Scored As False Negatives?
Recombinomics Commentary
March 21, 2005
>> Since January 2004, an estimated 69 people, primarily in Vietnam, have contracted a type of influenza commonly referred to as bird flu. The fatality rate among those reported to have the disease is about 70 per cent. But Dr. Keiji Fukuda said he suspects there are more cases.
"All infectious diseases have cases that are milder and cases that are more severe," said Fukuda, who spoke to medical officials at a conference on immunization. "All the cases we're seeing right now are the severe cases. There's got to be less severe ones out there." <<
It seems likely that many of the less severe cases, or at least those with lower levels of virus, are in isolation wards along with the patients with similar symptoms but slightly higher levels of viral RNA. Recent data on testing of H5N1 negative patients by NIID in Tokyo indicated that approximately one third of samples that scored as negative at the Pasteur Institute in Ho Chi Minh City were in fact positive for H5N1. When these false negatives were added to the small number of positives the yotal number positive jumped almost four fold. Only 22% of the patients infected with H5N1 were scored as positive. About half of these false negatives were due to poor testing techniques and the other half were missed because the PCR assay being used lacked sensitivity.
Similarly, only about 20% of the patients with classical bird flu symptoms and exposure to sick birds tested positive, suggesting virtually all of those who clinically presented with bird flu symptoms and were placed in isolation were actually H5N1 infected. Thus, simply looking at the fatality rate of these patients would give a better approximation of the true case fatality rate associated with H5N1 infections that produced a classical clinical presentation.
A study in Thailand looked at patients with respiratory illness that did not have a classical bird flu presentation and these patients, who numbered almost 600, were negative for H5N1 RNA and /or antibody, indicating that these milder respiratory cases that survived did not have laboratory evidence of H5N1 exposure. The case fatality rate for the lab confirmed cases was 67%.
Thus, it would seem that many of the H5N1 positive cases have already been identified by clinical presentation. Compiling a case fatality rate for these patients would be a relatively easy task.
Media link