Recombinomics Commentary 00:43
December 29, 2008
The above comment, from a description of the transmission of H5N1 to humans in 1997, focuses on the early events, which was quite different that the current state of H5N1 diversity, which is much more dangerous than 1997.
In 1997 the outbreak was limited to Hong Kong and involved 18 confirmed cases, including six fatalities. The outbreak led to the culling of all poultry in Hong Kong, and no reported human cases followed. However, there were multiple poultry outbreaks which typically began about this time of the year, which led to more culling.
The repeated outbreaks in Hong Kong suggested unreported H5N1 in China. The concerns were increased when a Hong Kong family vacationed in Fujian province in early 2003. The daughter developed bird flu symptoms and died in China, The family returned to Hong Kong and the father and brother developed symptoms and H5N1 was isolated from both. The father died, but the son recovered. Thus, for that family there was human to human transmission and a high case fatality rate. The H5N1 was more closely related to the various versions now in circulation, and the isolated H5N1 had receptor binding domain changes that favored binding to human cells.
However, it was exactly five years ago when reports started coming out of countries adjacent to China. In South Korea there was a bird flu outbreak in poultry, which proved difficult to control. In northern Vietnam children were being hospitalized who were influenza A positive, but negative for seasonal flu. H5N1 was isolated from both locations, and in early 2004 exploded into poultry in adjacent countries, from Japan in the north to Indonesia on the south. Moreover, the human cases in Vietnam spread to neighboring Thailand and the case fatality rate for the two countries was 70%, more than double the rate in Hong Kong. Moreover, there were multiple familial clusters, including the groom and his two sisters at the beginning of 2004 in Vietnam, and the mother, daughter, and aunt in Thailand later in the year. The clusters not only demonstrated more human to human transmission, but the also demonstrated that even in areas where H5N1 was confirmed in poultry, humans with bird flu symptoms were not being tested. No samples were collected from the index cases in the confirmed clusters, and those cases would have gone undiagnosed had they not been in clusters.
These clusters continued when H5N1 in humans expanded in 2005. Human H5N1 was confirmed for the first time in Cambodia, China, and Indonesia. In each case the first confirmed case was part of a cluster, but in each cluster samples were not collected or false negatives were initially reported. Moreover, China and Indonesia initially denied that the clusters were H5N1 confirmed.
However, the major development in the evolution of H5N1 was actually linked to H5N1 in long range migratory birds at Qinghai Lake Nature Reserve in China in the spring of 2005. Again H5N1 was initially denied, but subsequently confirmed in multiple species of long range migratory birds. At that point H5N1 had been limited to China and countries to the east. Qinghai Lake was at the intersection of major bird flyways, threatening worldwide spread.
Although some maintained that “dead birds don’t fly” and H5N1 would burn itself out, reports of dead waterfowl on farms in northwest China in June, 2005 suggested the Qinghai strain was migrating to the north, where many migratory birds summer in Mongolia and Siberia. Most waterfowl is resistant to H5N1 and at the end of 2004 WHO sounded an alert because H5N1 that was lethal to humans did not produce serious disease in waterfowl experimental infected with H5N1 from Vietnam (clade 1). The waterfowl were asymptomatic and excreted high levels of H5N1 in feces.
However, the Qinghai strain (clade 2.2) was distinct from the clade in southeast Asia, as well as H5N1 in Indonesia (clade 2.1) or China (clade 2.3). Thus, in the spring of 2005 there were four different strains of H5N1 in circulation. At the time, three of those strains had caused fatal infections in humans with case fatality rates above 50%, and the fourth strain was migrating out of China.
The outbreaks in China were followed by an outbreak at Chany Lake in Siberia in mid July, and were quickly shown to be clade 2.2. Moreover, sequences recently released from a June, 2005 outbreak in Kazakhstan demonstrated that the strain had migrated there prior to the Russian outbreak. Like all countries west of China, neither had previously reported H5N1. These outbreaks were followed by confirmation in Mongolia at the remote Erhel Lake. These outbreaks left no doubt that H5N1 could be transported and transmitted by long range migratory birds and raised concerns that the H5N1 in Siberia and Mongolia in the summer, would migrate to the south and east in the fall and winter.
These concerns were realized when H5N1 was reported in the Volga Delta in August, and the Danube Delta in October, followed by the Crimean Peninsula and Nile Delta in December.
The migration of H5N1 into these new areas, coupled with sequence data on H5N1 and H9N2, which was endemic in the Middle East lead to the prediction that clade 2.2 would acquire the same receptor binding domain change, S227N, which was present in the H5N1 from the Hong Kong family.
At the end of 2005 family members on Turkey developed bird flu symptoms and were transferred to a larger hospital. Although they were unconscious and had clear bird flu symptoms, initial tests were negative. However, as they started to die, lung aspirates were tested and were H5N1 positive. The H5N1 from the index case had the predicted receptor binding domain change S227N, and more patients in Turkey developed symptoms.
WHO accepted local lab test results which confirmed H5N1 in 21 patients. Moreover, almost all of the confirmed cases formed familial clusters. Samples were subsequently sent to England, where 12 of the 21 were confirmed, but WHO acknowledged that the lower numbers were related to shipping/degradation issues. The human clusters in Turkey were followed by clusters in Iraq and Azerbaijan and isolates from these clusters had receptor binding domain changes that increases affinity for human cells. Human cases were also confirmed in Egypt and Djibouti. Moreover, H5N1 in wild birds or poultry was reported in over 50 countries in Europe, the Middle East, and Africa.
In 2007, H5N1 reappeared in Egyptians and was reported in humans in Nigeria and Laos. A new sub-clade (2.2.3 Uvs Lake strain) appeared in wild birds in Europe in the summer and spread throughout Europe and this year was also in western Africa. At the end of the year, one of the longest sustained human transmissions was reported in Pakistan.
In 2008, human cases were reported in Myanmar and Bangladesh and the largest poultry outbreaks reported to date in India, Bangladesh at the beginning of the year, followed by South Korea in the spring, and the outbreaks in India and Bangladesh this month are on pace to eclipse the outbreaks of a year ago.
This year Indonesia also stopped reporting human outbreaks after denying three obvious clusters, where index cases were misdiagnosed with lung inflammation, dengue fever, a and typhus, even after H5N1 was lab confirmed in family members. Now lab confirmed H5N1 is denied, which highlights one of the reason “official” human cases are lower. Indonesia in fact attributed fewer cases to prompt Tamiflu usage, which should lower case fatality rates, but not cases, unless human to human transmission is widespread and efficient. However, the case fatality rate remains at 80%, because patients who recover due to Tamiflu treatment either test negative or are never tested for H5N1.
Thus, reported human cases are down because of testing / reporting issues. Similarly reports on H5N1 in poultry are down because the most hard hit countries, like Indonesia and Egypt, and declared H5N1 endemic, and therefore are only required to report outbreaks every six months, although Indonesia has not filed a report since 2006.
Thus, the difference between the H5N1 currently in circulation, and the H5N1 reported 12 years ago In Hong Kong is like night and day. Five years ago H5N1 began its global expansion, which accelerated dramatically in the 12 months following the reports if H5N1 at Qinghai Lake in 2005. The clade 2.2 expansion has produced confirmed human cases in Turkey, Iraq, Azerbaijan, Egypt, Djibouti, Nigeria, Pakistan, and Bangladesh. Clade 2.3 has now also been confirmed in long range migratory birds and has cause human cases in China, Hong Kong, Vietnam, Thailand, Myanmar, and probably Cambodia as well as the denied case in South Korea.
Recent activity has been on the rise as the birds migrate south and the weather cools. The expansion in the past five years has been dramatic, in terms of countries, infected birds culled, or confirmed human cases and clusters, which are far larger than Hong Kong in 1997.
H5N1 is on the move, in geographic and genetic terms, even if it seems to be on a decline to some.
No comments:
Post a Comment