.
Efficient Transmission of H5N1 to Humans in Turkey
Recombinomics Commentary
January 15, 2006
The lack of new cases and the seeming absence of secondary cases -- people infected through exposure to human cases, not poultry -- is suggestive that the H5N1 virus's ability to infect people hasn't changed, WHO officials said.
The above comments lack a scientific rationale. In Turkey a large number of patients were hospitalized with bird flu symptoms. The patients were from large familial clusters and the number of familial clusters has been large. These data alone signal an increased efficiency in the ability of H5N1 to infect people, regardless of source, which could be birds or humans.
This increased ability is supported by two genetic changes, the HA S227N (also called S223N), which allows for more efficient recognition of binding sites on epithelial cells that line the human respiratory tract, and PB2 E627K, a change that allows H5N1 to grow at lower temperature (34 C), which is the temperature inside a human nose outside in the winter.
The presence of both changes in the same virus has not been previously reported. S227N is in two isolates (Hong Kong/212/2003 and Hong Kong/213/2003) from Hong Kong visitors to Fujian Province in 2003. However, neither isolate has PB2 E627K. At least one isolate from northern Vietnam in 2005 (VN/JP12-2/05) has the change, but neither that sequence nor the associated PB2 sequence has been made public. However S227N has been limited to human isolate(s) in Vietnam.
In Turkey, the true picture of H5N1 infections is clouded by false negatives and limitations in testing of advanced patients. The first four suspects, the Kocigit siblings, all tested negative by PCR testing of nose and throat swabs. Three patients have died, and subsequent testing of lung samples have been positive. One isolate, from the oldest brother, has been obtained and fully sequenced. The sequence identified the S227N change and the explosion of cases in Turkey strongly suggests the change is present in the wild bird population.
The true number of H5N1 patients is not publicly known. The fourth Kocigit sibling has been discharged, but he is still officially H5N1 negative, even though he had the same symptoms as his siblings and was clearly infected with H5N1. However, because no lung samples were collected, he remains H5N1 negative.
These data pose serious questions about additional patients that are hospitalized with symptoms. Another family (Ozcan) from Dogubyazit, had at least 10 family members hospitalized. Two, 3 year-old Yusuf Ozcan and 9-year old Aysegul Ozcan had been placed in the ICU in critical condition and both have tested positive for H5N1. However, at least 8 other family members were also hospitalized and none have been H5N1 confirmed. Recently four additional Ozcan patients from Dogubyazit were hospitalized. Two of them, 12 year-old Fatma Ozcan and 5 year-old Muhammet Ozcan are in critical condition in the ICU and have not been confirmed to be H5N1. Their relationship to the Ozcans admitted a week earlier is unknown.
The relationship of many patients in the Van hospital are not known. Large numbers of patients have the same last name and media reports indicate that patients with teh same name are related. However, these relationships have not been disclosed in WHO updates, which usually provide disease onset dates, admission dates, and relationships to other suspect or hospitalized cases.
The report describing the H5N1 positive Ozcan siblings noted their age and gender, but provided no data on disease onset dates or hospital admission dates. Moreover, the report did not note that at least 8 family members were hospitalized in the same hospital.
The description of the five patients confirmed the following day did not even give age and gender. The only information for those five H5N1 positive patients was their province. The patient confirmed at Van was 18 year-old Nesime Mumak. The WHO report failed to indicate her cousin, 8 year0old Sumeyya Mumak, was in the same hospital and had been H5N1 confirmed a few days earlier. She was the third patient confirmed in Turkey. The first two were two of the fatal cases of the oldest Kocigit siblings.
The failure to report disease onset dates and relationships to suspect or H5N1 confirmed relatives is cause for concern. The disease onset dates are important in determining human-to-human transmission. Such horizontal transmission in families usually is associated with a gap in disease onset. In past cases in Vietnam, Thailand, Cambodia, Indonesia, and China, disease onset dates were provided in updates of H5N1 confirmed cases. That data were glaringly absent in the cases in Turkey. Data from prior clusters show a 5-10 day gap in almost every familial cluster, indicating all or most are examples of horizontal transfer between humans.
Although such data has not been provided in WHO reports, media reports suggest similar gaps between onset dates are present in the clusters from Turkey. Such gaps would be expected, since H5N1 is clearly efficiently transmitted to humans based on the number of confirmed cases, the number of suspected case, the size of the clusters, and the number of clusters.
There largest number of confirmed cases has been from the Van hospital. Seven of the eight confirmed cases were in familial (Kocigit, Ozcan, Mumak) clusters. In addition to the seven confirmed cases, there are at least 13 other patients hospitalized at Van with the above three last names.
H5N1 transmission to humans is clearly more efficient in Turkey, and the large number of family members confirmed or hospitalized is cause for concern.
WHO has announced a plan to gather extensive epidemiological and laboratory data on citizens in Turkey. This data should be collected and made public, along with relevant information on hospitalized patients, as soon as possible.
Map
Media Resources