Refocus of Qinghai H5N1 Testing in North America
Recombinomics Commentary
May 7, 2007
Interior will test about the same number of birds, 27,000, but this time it will spend more effort on dead and sick birds that are more likely to show symptoms of the disease.
"There are several theories on how (high-pathogenic H5N1) could get here, if you go by the wild migratory birds theory Alaska is the most likely spot," said Nicholas Throckmorton with the Interior Department.
Other vehicles being targeted for spread of the virus are illegal smuggling of imported birds, trade of poultry products and humans transporting it in on an airplane.
The U.S. testing program last year detected a low-pathogenic bird flu strain in six states -- Delaware, Illinois, Maryland, Michigan, Montana and Pennsylvania. It is common for mild and low pathogenic strains of bird flu to appear in the United States and other countries.
The above comments on H5N1 testing in North America will focus more on dead and dying birds, which is consistent with results in Europe and Africa, where virtually all H5N1 positives in birds were from wild birds or poultry that was dead or dying.
Detection of the Qinghai strain of H5N1 has been very rare in Europe and Africa, even when endemic areas were tested. Russia and Egypt are among the few countries that have detected Qinghai H5N1 in healthy wild birds. Russia described detection in about 25 species in its Mission report on testing in 2005. However, even those tests were somewhat clustered in time and space. NAMRU-3 in Egypt was able to obtain sequence data from a healthy Teal, collected in December, 2005, but the sequences require multiple tests.
In contrast, recent reports of H5N1 in wild birds in Germany and Burkina Faso describe detection in dead and dying wild birds, highlighting the difficulties in detection of H5N1 in healthy wild birds. Surveys of live wild birds in Europe and Africa have been uniformly negative.
These detection failures may be related to changes in the PB2 gene, which is almost always E627K in Qinghai isolates. This change is found in mammalian isolates, because it is associated with an increased polymerase activity at 33 C. The version found in other H5N1 strains is 627E, which has peak activity at 41 C, the avian body temperature. Therefore, Qinghai levels are more likely to be lower in avian isolates, and may not reach the limits of the PCR screening tests.
A focus on dead and dying birds would be more likely to find Qinghai H5N1. Last year an H5 positive was found in a dead gosling on Prince Edward Island. This bird was among 4 that had symptoms of Qinghai H5N1 infection and suddenly died the next day. Only one of the four was tested. The necropsy was "inconclusive" and the size of the insert was withheld. All birds on the backyard farm were culled and the owner was offered Tamiflu. The lone sample was held on Prince Edward Island for well over a week, and by the time it was shipped to Winnipeg to confirm the confirmed H5 PCR test(s), the sample had degraded so Winnipeg failed to confirm or isolate the H5 virus. Since low path H5 rarely kills waterfowl, it is likely that the H5 detected was Qinghai H5N1. The detection signal was weak, providing additional evidence that Qinghai H5N1 may be difficult to detect, even when samples are collected from dead or dying birds.
A more aggressive testing of such birds would be useful. Similarly, the detection of Qinghai H5N1 can be signaled by sequences in low path isolates. Although the number of H5 low path isolates were reported in Canada in 2005 and 2006, as well as the US, as noted above, only four gene sequences of one 2005 H5N2 isolate in British Columbia (which had 107 PCR H5 positives) have been made public, even though the 2005 isolates in Canada were collected in August, 2005.
Media sources
Recombinomics Presentations