Confirmed B2H2H2H2H H5N1 Transmission in Pakistan
Recombinomics Commentary 16:02
October 4, 2008

With respect to the chain of transmission, evidence gathered during the investigation supports the theory of initial transmission from poultry to humans followed by human-to-human transmission involving a third generation.

Case 5 was a 33-year-old brother of Case 1. He was asymptomatic but clinical specimens were collected from him owing to the close and prolonged contact with his ill brothers. Initial testing at the National Institute of Health yielded positive results for H5 RT–PCR on a throat swab collected on 29 November. When serum specimens were tested by microneutralization assay, a specimen collected on 8 December yielded an H5 antibody titre of 1:320 and a positive western blot assay.

The above comments are from WHO’s Oct 3 Weekly epidemiological record (WER), which includes a report entitled “Human cases of avian influenza A(H5N1) in North-West Frontier Province, Pakistan, October-November 2007”.  As indicated in the title, the report is coming out almost a full year since the start of the outbreak, and as indicated in the quotes above, includes an asymptomatic brother who was H5N1 confirmed by three different lab tests.  However, these results have not been released previously, and the newly described case extends the transmission chain to B2H2H2H2H, which is the longest recorded to date for H5N1.  Moreover, even though the case is acknowledged in this week's report, the case has not been added to the WHO table of confirmed H5N1 cases.  These delays and omissions extend the long list of deficiencies in the detection and reporting of this historic and important H5N1 outbreak in Pakistan.

The report provides detail that was lacking for months after the outbreak began, and clears up a subset of the long list of questions raised by the outbreak.  However, many important issues, including the sequence of the H5N1 isolated from the first confirmed case, which remains in WHO’s private password protected database, along with hundreds if not thousands of H5N1 sequences from infected patients or other hosts, remain unresolved.

As noted in the report, the transmission chain began when one of the cullers was infected in October (B2H) and developed symptoms on October 29, 2007.  He then infected one of his brothers (B2H2H) who developed symptoms on November 12 and died a week later, but not before the brother infected two more brothers (B2H2H2H) who developed symptoms on November 21.  One of these brothers died November 28 and one infected a fifth brother (B2H2H2H2H) who was asymptomatic, but as noted above, was H5N1 positive by three lab tests, including PCR on a nasal swab collected November 29.

However, none of the above was made public until there were local media reports, which were picked up by internet discussion groups in early December, 2007.  The story was subsequently picked up by wire services followed by comments by agencies in Pakistan or WHO.  However, the initial stories were largely confusing, in part because of the long delay between the start of the outbreak and the start of media coverage.  Consequently some stories noted that the outbreak began in October, while others assumed the outbreak began in December, just prior to the media stories. 

In addition to the brothers described above, other cullers and contacts were also said to be H5N1 positive based on testing done in Pakistan.  However, by the time investigators from WHO regional centers in Egypt and England arrived, the samples had largely degraded and initially the only positive was a sample from the brother who died November 28.  A sample from the brother who died November 19 was not tested (although at least one local media report indicated a sample had been collected), and samples from the brothers who were hospitalized and recovered tested negative.  The time of the testing of the asymptomatic remains unclear, because the WER indicates the sample collected November 29 was PCR positive, yet this positive result was not disclosed prior to this week.

However, in addition to the long delay in the acknowledgment of the asymptomatic case and the failure to test the first fatal case, the two cases who recovered were not reported confirmed until April, when results from neutralizing antibody tests were reported.  It is unclear if these delays were linked to the establishment of a new test using the H5N1 isolated from the second fatal case as a target, because the sequence of this isolate has been withheld.  Therefore, it is unclear if there were significant differences in sequence between the human isolate and other available targets. 

However, since the titer for the index case was 1:2560 and the recovered brother was 1:320, as was the asymptomatic brother, it seems likely that these high titers would have been detected when the samples were collected in late November or early December.  Thus, the reasons for the four month delay in reporting the confirmation of the two recovered brothers or the 10 month delay in reporting the asymptomatic case remains unclear, as is the reason for the failure to add the asymptomatic case to the list of H5N1 confirmed cases, since the brother was H5N1 positive in three lab tests (PCR, neutralizing antibody, and Western blot).

The reporting delays associated with the longest human to human H5N1 transmission recorded to date has been followed by questionable reports by other countries.  The H5N1 in Pakistan was followed by massive outbreaks in India (West Bengal) and adjacent Bangladesh.  Bangladesh has acknowledged one human case, which was also reported months after the fact.  The location of the reported case in the slums of Dhaka strongly suggests that the number of human cases in Bangladesh and India was markedly higher than one. India has yet to report any human cases, although the bird flu symptoms in villagers were wildly reported, as was the similarity in sequence between the H5N1 in India and Bangladesh, although neither country has released sequences from these outbreaks (and Bangladesh has not released sequences from any H5N1 outbreak).

Similarly, H5N1 clusters in Indonesia have been denied.  Fatally infected index cases from clusters that involve H5N1 confirmed cases have been said to have died from respiratory disease, typhus, and dengue fever, which has raised serious credibility issues with regard to reporting from Indonesia, and WHO was stopped reporting confirmed H5N1 cases in Indonesia in a timely manner.  Instead of the mandated IHR reporting time of 24-48 hours, WHO has been reporting H5N1 cases weeks or months after lab confirmation, setting the precedent for more reporting violations, including South Korea who refused to acknowledge a soldier/culler who was H5 PCR positive earlier this year.

In addition to the delays or lack of reports on H5N1 cases and clusters, WHO regional centers continue to hoard H5N1 sequences in the WHO private database.  NAMRU-3 became a WHO regional center last year and has not released any human H5N1 sequences since, even though cases were confirmed at the end of 2007 and beginning of 2008.  Sequences from West Bengal have been sequestered at Genbank for several months (see list here), and over 170 HA H5N1 sequences from Turkey have also been sequestered at Genbank for several months (see list here here here here here).

These reporting failures coupled with the hoarding of H5N1 sequence data by WHO and regional centers continue to be cause for concern and continue to be hazardous to the world’s health.

Media Links

Recombinomics Presentations

Recombinomics Publications

Recombinomics Paper at Nature Precedings