H5N1 Detection Failures in Europe
Recombinomics Commentary
August 16, 2007
To clarify the issue: at the Speichersee near Munich 3 WILD ducks were
found infected by H5N1 avian influenza virus. Of those, 2 of them were
confirmed by the national reference laboratory for avian influenza at the
Friedrich-Loeffler-Institut on the Isle of Riems to be infected by highly
pathogenic virus H5N1 'Asia,' whereas this diagnosis is still pending with
the 3rd case due to the low amount of viral genetic material within the
sample. Virus isolation on eggs is ongoing. However, due to the
epidemiological link there is strong suspicion that this will also turn out
to be HPAIV H5N1 Asia.
The above comments from FLI in northern Germany provide additional information on the sensitivity of H5N1 surveillance program. Germany confirmed H5N1 in 343 wild birds in Germany in 2006, which was the largest number in Europe. Although of 700 H5N1 confirmations were made in Europe in 2006, all confirmations were in dead wild birds. The failure of any country to detect H5N1 in live wild birds raised serious concerns about the sensitivity of the surveillance.
This summer, those concerns were heightened by the detection of H5N1 in additional wild birds. Although long range migration in Europe is minimal in the summer, 330 H5N1 infected birds have been confirmed at Kelbra Lake at the border between Saxony-Anhalt and Thuringia. Additional confirmations in Thuringia, Saxony, and Bavaria in the past two months sets of new record for H5N1 confirmations in Germany, which again has the highest number of confirmed cases in 2007.
However, although record number of H5N1 infected wild birds were confirmed in early 2006, between June 2006 and June 2007 there was only one confirmed case, which was at the Dresden zoo in August. Similarly, almost all countries in western Europe failed to detect H5N1 since early 2006. UK reported H5N1 in turkeys, but the close similarity between the H5N1 in the UK and an outbreak in Hungary, suggests the UK infection may have been linked to poor bio-security.
The above comments on the dead ducks near Munich however, provide some insight into detection limits. The three dead ducks were positive on an H5 PCR test, but only two of the three were positive in the sequencing test for the HA cleavage site. The RNA level in the negative was low, even though the duck was dead. It is likely that RNA levels in live birds will be lower than those fatally infected by H5N1, raising the sensitivity issue in surveillance programs.
Since Germany detected far more positives in 2006 and 2007, it is likely that the sensitivity of surveillance in adjunct countries is even lower than Germany. Recent France has reported positives three times in wild birds near the German border, and the Czech Republic has reported one positive. None of Germany’s neighboring countries have reported recent positives, even though record numbers were confirmed in Germany in multiple locations in four states.
The H5 PCR positive in a dead wild bird in Germany is similar to results last year on Prince Edward Island in Canada. Four geese had displayed Qinghai H5N1 symptoms and then suddenly died. One of the four was tested and was PCR positives. Almost two weeks after the geese died, a sample was sent to the national labs in Winnipeg for confirmation. By then the sample had degraded and no H5 was isolated and no OIE report was filed. At the time the low level of RNA was cited as evidence that the H5 assay was incidental to the deaths of the four geese, but the data in Germany support fatal infections even when RNA levels are low.
Recently OIE published a five page report on H5N1 in Europe in 2007. The report cited large numbers of negatives in healthy waterfowl, but failed to address the sensitivity issues of assays that failed to detect H5N1 even detection is widespread in dead wild birds.
The detection failures, including those cited in the OIE report on H5N1 in Europe remain a cause for concern, because the widespread detection in Germany in the summer suggests H5N1 is endemic in Europe and is silently spreading beneath surveillance programs that have serious sensitivity issues.
Media Sources
Recombinomics Presentations
Recombinomics Publications
Recombinomics' Paper at Nature Precedings