Commentary
H5N1 H2H Transmission Dictates New Pandemic Alert System
Recombinomics Commentary 18:10
June 21, 2008
THE World Health Organisation (WHO) has released details of a new flu pandemic alert system to replace its existing one, which has been criticised for lacking clarity.
.
Currently, the world is in phase 3 of the alert system, which is defined as “no or very limited human-to-human transmission”. The six-step, three-layer alert system is a ladder going from “low risk of human cases” in phase 1 to “efficient and sustained human-to-human transmission” in phase 6, the pandemic phase.
WHO’s new system addresses such issues of interpretation with three layers. The first comprises phases 1 to 3 and is defined as “predominantly animal infections; limited transmissibility among people”.
.
Phase 4 would be “sustained human-to-human transmission of animal or hybrid animal-human influenza virus, able to cause sustained community-level outbreaks that have been verified” — this is when health authorities can consider issues such as rapid containment, discussion of phase changes, and switching to a pandemic vaccine.
.
Phases 5 and 6 would comprise the “geographical spread” of flu. There is also a post-pandemic phase.
The above comments describe the WHO’s old and new pandemic alert systems. The old system has been a problem because H5N1 has been at phase 4 since 2005, but WHO has been denying the dozens of examples of human to human (H2H) transmission, which has created significant credibility issues, leading to report violations in an increasing number of countries, with Indonesia’s recent actions and announcements as the most glaring examples.
Under the old system, phases 3-6 defined degrees of H2H. Phase 3 was no or little H2H. Phase 4 was increased H2H. Phase 5 was significant H2H. Final phase 6 was sustained H2H.
When H5N1 exploded out of China in late 2003, leading to confirmed human cases in Vietnam and Thailand in early 2004, the level had moved to phase 3, with clear examples of H2H among a small number of family members. One of the clearer examples was in early 2004 in southern Vietnam involving a groom and his two sisters, who cared for him. The index case died with bird flu symptoms prior to testing, but both sisters developed symptoms on the same day, were hospitalized on the same day, initially were inconclusive on H5N1 testing, were subsequently confirmed H5N1, and died within an hour of each other. However, WHO maintained that this was not an example of H2H because the index case wasn’t tested, and there was some sequence difference between the H5N1 from the two sisters. However, cases can be infected with multiple versions of H5N1 and in this cluster one version grew out in one sister, while the other grew out in the other sister.
Later in 2004 there was another example of H2H in Thailand. The index case was living on a farm with her aunt. When she developed symptoms and was hospitalized, her mother, and office worked in Bangkok, hundreds of miles from the farm, visited her daughter. The daughter was misdiagnosed with dengue fever, so her mother was not were PPE’s when holding her daughter when she vomited blood, which was the likely source of the mother’s infection. The daughter died and the mother developed symptoms when she returned to Bangkok. She also was not tested for H5N1 and died, but WHO personnel in Bangkok investigating another case were told of the case and retried the body just before cremation. A sample of fixed tissue was positive for H5N1 sequences. The aunt also developed symptoms and although she initially tested negative, H5N1 was also confirmed in the aunt, who survived. This cluster was written up in the New England Journal of Medicine and was consider an example of H2H.
At about the same time there was another cluster in Vietnam involving four relatives. The first three deaths were not tested, but a sister developed symptoms after caring for one of the cases, and she was H5N1 confirmed.
Thus, in 2004 H5N1 was confirmed in multiple clusters and disease onset dates indicated these clusters involved H2H transmission, although WHO only acknowledged the cluster in Thailand because the other clusters involved patients who had poultry exposures.
However, the H2H situation changed markedly in northern Vietnam in early 2005. In southern Vietnam there were a few clusters with a case fatality rate of 100%. Similarly, Cambodia reported its first confirmed case, which was also a cluster, but the index case was not tested. His sister developed symptoms when he died and was H5N1 confirmed after she died at a hospital in Ho Chi Minh City, representing an increase in H2H clusters to a new country.
However, the real increase in H2H was seen in northern and central Vietnam, which the case fatality declined to about 10% while the number and size of H2H2 clusters grew. Initially WHO attempted to link clusters to common sources like duck blood pudding, but these clusters did not epidemiologically link. Disease onset dates did not fall within the typical 2-4 days of exposure and H5N1 infections or antibodies were in family members or contacts who did not eat the duck blood pudding and there were no symptoms of antibodies in others who did eat the duck blood pudding.
The increases in size and number of clusters led to the collection of approximate 1000 serum samples which were positive in testing done in Vietnam. The exact number of positives wasn’t released, but samples were sent to the CDC in Atlanta, who also found positives (but again numbers were not released). An emergency meeting was called for May in Manila. The more numerous and larger clusters were acknowledged and discussed, but the serum data was kept confidential. Although an increase of the pandemic level to phase 4 or higher was discussed, it remained at phase 3 and news from northern Vietnam declined markedly. Eventually a WHO affiliated lab in Japan declared the samples negative, but no details were ever released.
However, H5N1 was still expanding in 2005. In May H5N1 was reported in wild birds at Qinghai Lake, which was novel (clade 2.2). This was followed by the first confirmed human cases in Indonesia and China.
In Indonesia the first confirmed case was a cluster of three. All three died, but initially H5N1 was only confirmed in the father. The index cases had high H5N1 antibody titers in two serum collections, but since the samples were collected late and only 3 days apart, the serum level increased only 2 fold, and fell short of the WHO criteria of 4, but the fatalities were considered to be due to H5N1 and the disease onset dates indicated the cluster was another example of H2H. The second confirmed case in Indonesia was also a cluster. H5N1 was confirmed in the index case and her nephew and disease onset dates again supported H2H. A third cluster was also reported in Indonesia in 2005 and all three clusters were written up, providing examples of increased H2H and additional reasons for raising the pandemic level to 4. The initial cases in China were also a cluster, but disease onset dates may not have supported H2H.
However, the clade 2.2 from Qinghai Lake spread to Europe, the Middle East, and Africa in late 2005 / early 2006 leading to additional H2H clusters. In Turkey three siblings died and were H5N1 confirmed. Disease onset dates suggested one sibling infected the others. Local media also suggested that these siblings infected cousins. Nine cousins were hospitalized and two were H5N1 confirmed. Two of their cousins were then hospitalized an both were confirmed. Locally, there were 21 confirmed H5N1 cases in Turkey, and virtually all were in clusters.
The human clusters in Turkey were followed by a cluster in Iraq. The index case was initially said to have died of a heart ailment, even though her physician indicated her symptoms matched the fatal cases in Turkey. When her uncle developed symptoms and died, both were H5N1 confirmed. Similarly, the index case in Azerbaijan was also said to have died of a heart ailment, but when relatives and a friend developed symptoms and died, all were H5N1 confirmed and disease onset dates indicated these were also H2H transmission. This the cases in early 2006 in the Middle East were also H2H and increased the number of geographic distribution of H2H cases, but the phase remained at 3.
These outbreaks were followed by a large cluster in Indonesia in North Sumatra. This cluster involved eight family members and represented H2H2H. Once again, this increase in the size of the cluster and length of the transmission chain did not lead to an increase to phase 4.
In 2007 there were H2H clusters in Nigeria and Pakistan, although there were issues with testing and confirmation in fatal cases.
Most recently, three of the five most recent confirmed cases in Indonesia were H2H clusters, although in each cluster the index case, who died, was misdiagnosed.
Thus, the examples of increased H2H based on size of clusters, number of clusters, or geographical location of location, has increased dramatically since 2004, yet the pandemic phase has remained at 3.
The new system will allow WHO to acknowledge these H2H clusters without raising the pandemic level, because they will all fit the new definition of phase 3.
The new phase 4 will represent sustained transmission, which is the current requirement for phase 6. The new phase 5 and 6 will be used to control panic.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings