.
H5N1 Bird Flu Deaths in Indonesia Raise Pandemic Concerns
Recombinomics Commentary
July 20, 2005
Although some health officials raised suspicions about bird flu shortly after Rafei and his daughters died, the health ministry had declined to blame the deaths on the virus until tests were completed at a specialized laboratory in Hong Kong. A sample from Rafei tested positive for the virus, health officials said, citing the results received Wednesday. Another specimen from his older daughter showed antibodies, indicating she had been exposed to bird flu, but it tested negative for the virus itself. No test was done for the younger daughter because she was buried without a sample being taken.
The above description matches the H5N1 bird flu pattern in each country that has reported clusters of bird flu deaths. Typically there is no testing on the first to die, testing of relatives is initially negative or inconclusive, and eventually one of more family members is laboratory confirmed H5N1 positive.
These cluster demonstrate human-to-human transmission, but highlight the number of false negatives on suspected cases and the lack of testing on patients where the suspicion level is not as high. This cluster had a bimodal distribution of onset dates. The older daughter was the index case and was admitted into the hospital on June 29. Her father who frequently visited her while she was hospitalized was not admitted until July 7, the same day as his youngest daughter. No sample was collected from the youngest daught.
The initial samples from the father and index case were positive for bacteria and negative for virus, so the government declared them to be bacterial pneumonia cases. Since the cases were fatal, bacterial pneumonia was more likely due to a secondary infection. In later tests the index case still tested negative for the virus. This was probably due to clearance of the virus from the site of collection or poor sample collection. Eventually the father tested positive for the virus and his older daughter was positive for antibody, leaving little doubt that all three died from H5N1. However, if the three had not presented as a cluster, it is unlikely that any would have been declared positive for H5N1,
The same pattern happened in Vietnam, Thailand, and Cambodia. In Vietnam one of the first reported clusters was in Vietnam in January of 2004. A groom developed symptoms shortly after his wedding. Several days later his two sisters, who cared for him developed symptoms. The groom was never tested, but the cluster had a typical bimodal distribution and all three had bird flu symptoms. The two sisters who were infected by their brother had a typical common exposure history. The two sisters developed symptoms on the same day, were hospitalized the same day, test inconclusive, eventually tested H5N1 positive, and died within one hour of each other.
In Thailand the cluster was written up in the New England Journal of Medicine. It is frequently cited as the best example of human-to-human transmission. However, it is cited because when the index case was infected, her mother was several hundreds of miles away in an office in Bangkok. The mother visited her daughter and had no contact with birds. The mother developed symptoms after her daughter died, but still has not tested. Because a nurse notified investigators of another case, the mother was found just before cremation. Tissues from the mother were H5n1 positive and the aunt eventually tested positive, although initial tests were negative.
The same series was repeated in Cambodia. The index case died in Cambodia without being tested. However, shortly after his death his sister developed symptoms and was hospitalized in Vietnam. After her death she tested positive for H5N1. She had not been tested in Cambodia even though her brother had already died and both had bird flu symptoms.
Thus, all four clusters have bimodal distributions and only the aunt in Thailand survived out of the 11 cases in the 4 clusters. It is unclear if any of the 11 would have been tested if not in a cluster where all cluster members had bird flu symptoms.
Although all four clusters above clearly involve human-to-human transmission, WHO has kept the pandemic level of 3 indicating there is no human to human transmission. There are about another dozen familial clusters in Vietnam and virtually all have bimodal distributions of disease onset dates indicating the all likely involve human-to-human transmission of H5N1. Stage 4 is defined as small dead-end clusters of cases, which is what each of these familial clusters is, yet the official stage for the 2005 flu pandemic is 3.
The larger clusters in northern Vietnam have been laboratory confirmed. Larger cluster define stage 5. Third party reports out of China suggest H5N1 human-to-human transmission has become sustained, which is the definition of stage 6, the final stage. China has denied human cases and WHO has kept the stage at 3.
Unfortunately, H5N1 does not read press releases.
Media Resources