Human to Human Transmission of H5N1 Is Common
Recombinomics Commentary
May 31, 2006
The initial case developed symptoms on 24 April, was hospitalized on 2 May, and died on 4 May.
Six out of the seven confirmed cases developed symptoms between 3 May and 5 May. These cases include two sons of the initial case, her brother from Kabanjahe, her sister, the sister's baby, and the son of a second brother living in an adjacent house. This second brother, the last case in the cluster, developed symptoms on 15 May. Six out of the seven cases were fatal.
The WHO updated details on the H5N1 bird flu cluster in Sumatra contain a full set of disease onset dates that leave little doubt that there was transmission from the index case to family members which developed symptoms between May 3-5, followed by transmission to the brother who developed symptoms on May 15. Thus, transmission was H2H2H. The earlier update did not include disease onset dates of the family members infected by the index case.
The disease onset dates are critical for distinguishing transmission from a common source or a contact, such as a family member. H5N1 has an incubation time of 2-5 days, and since it is inefficiently transmitted to humans, transmission frequently occurs several days after the index case develops symptoms. Consequently, the time gap between transmission from the index case to family members is 5-10 days.
The cluster described above has been studied intensely and the most recent WHO update is quite detailed, providing supporting evidence for H2H2H. However, other large clusters such as the initial clusters in Turkey and Azerbaijan have similar gaps between the index case and additional family members or contacts, and the time frame is extended, indicating several links in the transmission chain. The cases in Turkey also had exposure to H5N1 infected poultry and in Azerbaijan there was exposure to wild bird feathers, but the transmission chain involved multiple infections, which are far more common in family members than common sources.
As was noted in the prior WHO update on the small cluster in Bandung. Both siblings developed symptoms at the same time, were admitted to the hospital at the same time, and died within 5 hours of each other. That is a typical pattern for infection by a common source.
However, the vast majority of H5N1 familial clusters have the 5-10 day gap. These H2H transmissions trace back to infections in Vietnam in early 2004. The first 15 familial clusters have been described. These clusters involve cases in Vietnam, Thailand, Cambodia, and the first cluster in Indonesia. Subsequently thee have been over a half dozen of additional familial clusters in Indonesia as well as several clusters in Turkey, and smaller numbers in Azerbaijan, Iraq, and China.
Thus, since early 2004 there have been approximately 30 familial clusters and the vast majority have the 5-10 day gap indicating that most if not all were H2H.
The sequences of the H5N1 from the smaller clusters were closely related to the poultry H5N1 sequences in the area. However, the large cluster in Turkey involved the Qinghai strain, which had PB2 E627K as well as HA S227N. The Turkey cluster was the first confirm human cases involving the Qinghai strain. The Azerbaijan clusters also involved the Qinghai strain, and therefore also had E627K. The HA sequences have not been released, so the role of S227N has not been reported.
In the Sumatra cluster, the earlier WHO update described a great deal of coughing by the index case when family members were exposed on May 29. Today's Nature report also indicates the patients had high levels of H5N1 in their nose and throat suggesting that the H5N1 may have also had E627K. The sequencers failed to comment on questions on specific polymorphisms, and the sequences have been withheld.
The questions deserve answers and the sequences should be released immediately.
Media Link
Map