|Home||Founder||What's New||In The News||Contact Us|
|Paradigm Shift Intervention Monitoring||Commentary
Most H5N1 Cases Now Linked to Human to Human Transmission
February 4, 2006
Almost all cases of H5N1 human infection appear to have resulted from some form of direct or close contact with infected poultry, primarily chickens. In addition, a few persons may have been infected through very close contact with another infected person, but this type of transmission has not led to sustained transmission.
The above comments from the prepared statement for the Senate appropriations committee seriously underplay the involvement of human-to-human transmission of H5N1. The role was even more distorted in the actual testimony, which indicated that strong evidence of human-to-human existed for only two familial clusters. Familial clusters have made up an ever increasing percentage of the official H5N1 cases and virtually all such clusters involve human-to-human transmission among family members.
Although exposure to sick or dying poultry can be frequently linked to the index case in a cluster, this linkage does not necessarily extend to other family members. These members frequently have exposure to both the index case and poultry. To distinguish between a common poultry source and a common family member, the dates of disease onset are used. Since transmission from bird to human is rare, the likelihood of two independent transmission is low. Therefore, if the common source is poultry, the index case and other family members would be expected to develop symptoms over a short time course (1-2 days). If however, the index case transmitted the H5N1 to other family members, the time interval between disease onset in the index case and disease onset in other family members would be long (5-10 days).
The number of family clusters in the various countries reporting H5N1 outbreaks since 2004 has now exceeded thirty. Almost all of these clusters have a time gap of 5-10 days between disease onset of the index case and other family members. This gap indicates that most of the familial clusters involve human-to-human transmission.
The clusters date back to early 2004 in Vietnam and later in Thailand. By early 2005, these clusters account for almost one third of H5N1 cases. In Indonesia, the number of H5N1 patients in familial clusters grew to about two thirds of cases. The initial 15 clusters were described in a recent CDC/WHO publication. At that time, WHO changed wording in their characterization of the H5N1 outbreak. They had indicated that there was little evidence for human-to-human transmission. This changed to little evidence for efficient human-to-human transmission, acknowledging the growing number of familial cases which involved human-to-human transmission.
Recently, the size and number of these clusters grew, and WHO again changed their description from a lack of evidence for efficient human-to-human transmission to a lack of evidence for sustained human-to-human transmission. Although this terminology suggests the increased frequency has been noted by WHO, public comments and media reports still leave the impression that human-to-human transmission of H5N1 is rare or non-existent.
This impression is particularly misleading at the present time because a genetic change has been noted in H5N1 from the index case in Turkey. The change in the receptor binding domain of HA, S227N (also called S223N), increases the affinity of the HA for human receptors. This change coupled with another change, PB2 E627K, increases the efficiency of H5N1 infection in humans, especially in cold weather. These genetic changes have led to very large clusters in Turkey as well as linkage between clusters.
The linked cluster included the index case for Turkey. Index cases from familial clusters have in fact been the index cases for countries since 2005. The index case for Cambodia, Indonesia, China, Turkey, and Iraq all were familial index cases and all clusters included a 5-10 day gap in disease onset dates.
These data leave little doubt that human-to-human transmission of H5N1 is quite common and now represent the majority of human cases. Representations to the contrary are cause for concern.