The pathogenesis of human H5N1 influenza remains poorly understood and controversial. 'Cytokine storm' has been hypothesized to be the main cause of the severity of this disease. However, the significance of this hypothesis has been called into question by a recent report, which demonstrates that inhibition of the cytokine response did not protect against lethal H5N1 influenza infection in mice. Here we showed discrepant findings in two adult H5N1 autopsies and a fetus obtained at autopsy which also raise doubt about this hypothesis. Antigens of 10 cytokines/chemokines which were found to be significantly elevated in previous H5N1-infected patients and in vitro experiments, and mRNA of eight of these, were absent from the lungs of a pregnant woman and her fetus. In contrast, antigens of seven cytokines/chemokines and mRNA of six of these were found to be increased in the lungs of a male autopsy. The cells expressing these cytokines and chemokines were identified as type II pneumocytes, bronchial epithelial cells, macrophages and vascular endothelial cells. Levels of cytokines and chemokines in the serum of the male case were also significantly higher than those of infectious (infection other than by H5N1) and non-infectious controls. In comparison with results from our previous study, it appeared that the male case had increased cytokine/chemokine expression but reduced viral load, while the pregnant female had diminished cytokine/chemokine expression but a significantly increased viral load in the lungs. These disparate findings in these two cases suggest that 'cytokine storm' alone could not be a sufficient explanation for the severe lung injury of this newly emerging disease.
(c) 2008 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.