Mortality of H5N1 human infections might be due to H5N1 virus pneumonia and could decrease by switching receptor
This pandemic risk is a concern as the mortality was 458 (52%) of the 876 influenza A(H5N1) cases reported in Europe since 2002. The haemagglutinin protein is the host-range determinant as it mediates virus binding to the sialic acid receptors.
Here we argue that the high mortality might be due to a H5N1 virus pneumonia, and should the H5N1 switch to the upper airway receptor for human influenza (H1, H2, and H3), α2,6-sialic acid (SA α2,6), we hypothesise that the mortality would be lower because most infections would be rescricted to the upper respiratory tract infections and only in rare cases pneumonia.
The current outbreak of influenza A(H5N1) in dairy cattle in the USA has raised concerns of increased risk for sustained human-to-human transmission.3 As of July 12, 2024, 151 dairy herds and 99 million poultry are affected and H5N1 has been found in 9528 wild birds.3 Five humans cases have been reported and in three, the symptoms reported included conjunctivitis.4
The influenza virus hemagglutinin protein binds to sialic acid receptors on the host cells, which can be either SA α2,3 or SA α2,6.5,6 SA α2,3 is found on specific human tissues especially lung alveoli and conjunctiva, while SA α2,6 is predominantly found in the upper respiratory tract of humans.6 The avian influenza's uses the SA α2,3 receptor whereas the three human influenza viruses (H1N1, H2N2, and H3N2) use the SA α2,6 receptor.6
Avian influenza can occasionally cross the species barrier from animals to humans. This transmission likely requires exposure to a high number of avian influenza viruses for the virus to reach the SA α2,3 receptor in the alveoli, after which the infected person will develop diffuse, double-sided pneumonia. Receptor distribution also explains why conjunctivitis has been reported in at least three of the five reported human H5N1 cases infected from cattle in the USA.3
Our experience from the 2009 H1N1 pandemic was that admissions to intensive care were due to a H1N1 pneumonia.7 The mortality rate was five (23·8%) in 21 patients and three (33·3%) in nine patients receiving extracorporeal membrane oxygenation treatment.7 These rates might not be considerably different to the 52% mortality reported by the European Food Safety Agency,2 given the variance between centres in Europe.
Therefore, we hypothesise that if the H5N1 virus switched receptor preference from SA α2,3 to the human upper respiratory receptor SA α2,6, the virus might cause a less severe upper respiratory infection and the mortality rate would decrease because most cases would no longer be due to influenza virus pneumonia.