Bird Flu Pandemic Preparedness Detection Issues
Recombinomics Commentary
March 13, 2005
>>Vietnamese health officials said Saturday they suspect a second nurse who cared for a bird flu patient has contracted the disease that's killed 46 people across the region.<<
The increased efficiency of human-to-human transmission of H5N1 in Thai Binh province has focused attention on pandemic preparedness. However, the scandalously poor monitoring of H5N1, and the lack of a definition of the diseases it causes creates significant problems in control and preparedness areas. The decrease in transparency linked to the spread of H5N1 is cause for concern, as is the misinformation of human-to-human transmission.
Bird flu cases have been narrowly defined. Many clear cases have been actively excluded because of a lack of testing and a lack of sensitivity of the tests that are run. This has led to a significant under-reporting of cases at all levels and a narrowly focused vaccine plan, which increases the likelihood of the development of an ineffective vaccine.
Although WHO has recently indicated that the definition of H5N1 diseases will be expanded, there was evidence in January of 2004 in Thai Binh that H5N1 would cause a spectrum of clinical diseases. Prior investigations of H5N1 cases in 1997 as well as infections in wild and domestic birds clearly showed that multiple versions of H5N1 could infect the same host, resulting in broad tissue tropisms and both reassorted and recombined viruses.
The results from a familial cluster in Thai Binh in January of 2004 demonstrated that these earlier observations could be extended to the 2004 H5N1 infecting large numbers of patients in Vietnam and Thailand. The infections generated a very high case fatality rate, and the familial cluster clearly demonstrated that a comprehensive approach was required.
The cluster showed human-to-human transmission, dual infections, and distinct clinical presentations with a 100% fatality rate. Instead of initiating a comprehensive program, the human-to-human transmission was discounted, most recently by the fact that the two H5N1 isolates detected in the sisters were different.
The index case for the cluster had just been married and developed typical bird flu symptoms. No samples were collected but when his two sisters, who had cared for him, also fell ill on the same day and were admitted to the hospital on the same day, samples were collected. Initial test results were inconclusive, but the sisters subsequently tested positive and died the same day. One had respiratory symptoms while the other had a gastrointestinal illness. Sequencing identified differences between the two H5N1 isolates.
Although these results were consistent with several examples of multiple H5N1 strains infecting the same animal and producing various clinical symptoms, the WHO adopted a narrow case definition that focused on the respiratory disease. They also adopted a rigorous case definition that required multiple tests and excluded cases such as the index case, because of no sample collection. This approach reduced the number of cases and clusters, which led to significant undercounts.
Recently, publication of the detection of H5N1 in another case that did not present with classical symptoms, as well as the death of a sibling, raised questions about the narrow WHO definition. The concerns were increased when samples from additional atypical patients also tested positive for H5N1. However, the concerns related to these cases were compounded by false negative results on samples from these patients. In four of seven instances the re-tests were positive, but negative results on three of the retests raised sensitivity issues.
Thus, the current status of H5N1 in Vietnam and nearby countries is unclear because of the lack of testing coupled with false negatives.
However, the most recent cases from Thai Binh appear to have reduced fatality rates in association with increased transmission efficiencies. There are now at least four patients that are linked, two of whom are health care workers. The transmission chain extends over a three week period, which is markedly longer than more limited familial transmissions described earlier.
The high case fatality rate facilitated the identification of cases. A lower case fatality rate would allow atypical cases to more easily be confused with other diseases, such as cholera, dengue fever, and typhoid, common misdiagnoses in the 1918 flu pandemic.
Transparency issues have complicated the monitoring issues. The cases in the south were well covered in media reports, but WHO was not officially notified until recently. Moreover, there has been no news on cases in the south since the beginning of February. Thus testing, contact tracing, and reporting all remain highly suspect in Vietnam, although similar problems may be present in Thailand and Indonesia, as well as other countries.
Thus, the mere presence of H5N1 in endemic areas is uncertain, creating significant issues for control, identification of relevant isolates, and identification of important genetic changes.
Media link