Proving H2H2H H5N1 Transmission in Pakistan
Recombinomics Commentary 12:48
April 9, 2008
This is the final installment of the rather tortuous and conflicting story
laid out in the ProMED-mail posts listed below, which is best forgotten.
The final analysis summarized in the table above now indicates clearly that
person-to-person transmission of avian influenza virus infection occurred
within a family of brothers in Peshawar, Pakistan.
The above comments from a ProMed commentary acknowledge the obvious human to human transmission of H5N1 in Pakistan, but since it is one of the longest transmissions for confirmed H5N1 reported to date, it is not “best forgotten.” ProMed would probably like to forget about the number of times their commentaries, linked below, suggested that the human to human to human (H2H2H) transmission was not an example of H2H transmission, and current media reports and blogs are still suggesting that the H2H2H was just H2H, involving the index case infecting three of his brothers.
However, it was the length of the chain which lasted for more than a month that created much of the confusion in media reports, coupled with the delay in reporting the outbreak and the degradation or absence of samples, which created more confusion. The latest report raises the number of confirmed cases to three, which more accurately reflect the H2H2H transmission, which was confusing earlier, when there was only one confirmed case in Pakistan.
Proving H2H or H2H2H transmission is straightforward and is heavily dependent on disease onset dates. Although there may be extenuating circumstances to invalidate the simplest interpretation of the data, such data rarely surfaces, other than the fact that many index cases have an association with poultry, and family members, who have the closest contact with the index case, also live in proximity to poultry. Moreover, since most patients are not tested unless they have a linkage with poultry or a confirmed case, the association with the poultry is not likely to change.
However, the likelihood of a confirmed H5N1 infection by an infected family member is several orders of magnitude higher than infection by infected poultry. The number of confirmed human cases is in the hundreds, wile the number of birds culled is in the hundreds of millions. However, clusters of human cases are common, and most have a significant gap in disease onset dates, signaling H2H infections, because there is close contact.
Clusters of H5N1 infections involving the current versions of H5N1 go back to early 2004. Human clusters alone are cause for concern, because they signal a more efficient transmission of H5N1 to humans. However, most of these clusters have significant gaps in disease onset dates, implicating H2H transmission. Although the gaps in disease onset dates are obvious, such gaps were not considered as “proof” and most write-ups in the literature included epidemiological data which required that the subsequent infections were not linked to poultry.
There have been several such clusters, including the example in Thailand in 2004 involving a child living with her aunt, and her mother, an office worker in Bangkok. This was considered H2H because the mother had no poultry contact, and the cluster also involved the aunt. Clusters of 3 or more are more easily proved, because infections of humans by H5N1 is rare and sequential infections are likely to be H2H or H2H2H. However, only the lack of involvement with poultry distinguishes these clusters from the more common set of circumstances, where additional family members are physically close to poultry. However, the difficulty in infections of human by H5N1 greatly reduces the likelihood of two independent B2H transmissions.
Although unlikely, some clusters of three are not in the proven H2H category, even though the above circumstances surrounding the cluster have been met. On such cluster in early 2004 in Vietnam involved a groom and his two sisters, who cared for him. He developed bird flu symptoms and died prior to the collection of samples. However, his two sisters subsequently developed symptoms on the same day, indicating they were exposed to a common source, their brother. Testing on the sisters was initially inconclusive, but H5N1 was subsequently confirmed. They were hospitalized on the same day and both died a week later within one hour of each other. However, H5N1 sequences differed, indicating their brother was infected with multiple H5N1 strains, but these differences were used to exclude H2H, even though one of the sisters had no prior exposure to poultry.
Many other H2H clusters are excluded because samples are not collected from the index case. However, most are excluded from the proven H2H category because family members who are subsequently infected are living in areas where there is poultry.
In the case of the four brothers in Pakistan, the transmission chain was long, and the brothers were not in contact with poultry. Only the index case had a direct link to H5N1 positive birds. Moreover, the disease onset dates had large gaps. The index case developed symptoms on October 29, while one brother developed symptoms on November 12. He died on November 19 and two days later two brothers developed symptoms. The development of symptoms within a few days of the death of the index case is common. The viral load of the index case is likely to be high at the time of death, and the death creates additional transmission opportunity to family members, including hugging and kissing, preparing the body for burial, transport of the body for burial, or other close contact at the funeral. In the cases of the two brothers who developed symptoms on November 21, they were exposed to a common source, their dead or dying brother. Thus, the transmission in Pakistan was H2H2H.
Although most H5N1 clusters have a gap in disease onset dates signaling H2H, but the vast majority of such H2H transmissions are not proven because of circumstances surrounding the H2H transmission, which includes some linkage of poultry with infected family members.
Avian influenza, human (05): China , Pakistan
Avian influenza, human (03): Egypt , Pakistan
2007
---
Avian influenza, human (177): Pakistan , Viet Nam , Egypt
Avian influenza, human (172): China , Pakistan
Avian influenza, human (171): Pakistan
Avian influenza, human (169): Pakistan
Avian influenza, human (168): Pakistan
Avian influenza, human (167): Pakistan
Avian influenza, human (166): Pakistan
Avian influenza, human (165): Pakistan
Avian influenza, human (163): Pakistan
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings