False Negatives in Fatal H5N1 Case in Thailand Cause Concerns
Recombinomics Commentary
September 26, 2006
59-year-old man from Nong Bua Lamphu was yesterday declared the latest victim of bird flu
He was the first person to undergo nine PCR tests, yet the virus was still not detected until an autopsy was conducted on tissue taken from his lungs and faeces, he said.
That PCRs failed to detect the virus could be because of the effects of oseltamivir, which might have prevented the virus from being present in the respiratory-tract areas where swabs were taken for testing, according to Dr Tawee Chotpitayasundond, of the Queen Sirikit National Institute of Child Health.
The above comments describe confirmation of a fatal H5N1 infection in a patient in Thailand who died almost a month after being admitted. Although H5N1 was present during the entire time of hospitalization, it was not detected until autopsy in spite of running 9 PCR tests.
The data raise serious questions about the WHO reliance on PCR test to confirm H5N1bird flu infections, especially on samples collected after Tamiflu treatment has begun. The number of false negatives in Thailand is not known, but the number of such negatives in Indonesia is likely to be high, especially in pandemic regions where Tamiflu blankets are applied.
The recent outbreak in Garut in West Java in Indonesia is an example of such false negatives. Patients linked to three separate clusters where H5N1 confirmed, but initial cases died prior to sample collection and subsequent patients were tested after the Tamiflu blanket was applied. Consequently, only three of the more than 20 hospitalized patients were confirmed.
There are three current clusters being reported in Indonesia. In two of the clusters, H5N1 has been confirmed. However, several patients are` improving after initiation of Tamiflu treatment and have not been reported to be H5N1 positive.
The recovered patients should be tested 3-4 weeks after disease onset to determine if they have neutralizing H5N1 antibodies, signaling a recent H5N1 infection. Without these additional tests, the extent of H5N1 spread among contacts of the index cases will remain unclear and the repeated false negatives will create a significant undercount in the number of cluster and length of transmission chains.
Media Sources
Phylogenetic Trees