.
H5N1 False Negatives in Turkey
Recombinomics Commentary
January 6, 2006
Turkish authorities had ruled out avian influenza in these cases based on preliminary test results from samples taken from the nose and throat. Subsequent tests of additional patient specimens taken from the lungs produced positive results.
The above comments form the WHO report on the H5N1 outbreak in Turkey raises concerns about false negatives. Similar concerns have been noted on cases in Indonesia, where testing at about a week after symptoms produces false negatives. This "Goldilocks testing" (too early for antibodies, to late for PCR of nasal or throat swabs, but just right for false negatives) is playing a major role in Turkey also.
Most of the patients at Van have come from the village of Dogubeyazit, which is 120 miles away via mountainous snow covered roads. Consequently, most patients are advanced case with signs of pneumonia. Therefore most of the H5N1 has already been cleared from the nose and throat and is resident in the lungs.
Recent media reports describing the third death of the four siblings note that the latest fatality was H5N1 negative. However, the negative data again is based on antibody tests of serum collected before antibody levels were detectable, and after the H5N1 had moved to the lungs. Subsequent testing has identified H5 in the third fatality, again demonstrating that the nasal swabs of advanced cases create false negatives.
The false negatives complicate screening tests. The news of the three fatalities caused 1100 people to go to the small hospital in Dogubeyazit for testing. While most of the patients are probably not infected with H5N1, identifying those that are remains a challenge. Similarly, a large number of patients were not admitted to the hospital in Van because initial screening was negative.
It seems that the clinical picture is more accurate than lab testing, when appropriate samples are not collected.
The large clusters of H5N1 infected cases in Turkey suggest that H5N1 infections of humans is more efficient than prior H5N1 isolates. The increased efficiency of bird to human transmission probably applies to human-to-human transmission, which is most easily addressed by disease onset dates. A bimodal distribution of dates may be present in a 10 member familial cluster, raising concerns of efficient human-to-human transmission.
Crowded emergency rooms, or release of infected patients based on false negatives, could accelerate the spread of H5N1 in humans, creating more cause for concern about the H5N1 transmission in Turkey.
Map
Media Resources