How reliable are the H7N9 real-time RT-PCRs for low viral loads?

An interesting document that compares the effectiveness of some different diagnostic PCR methods for use in H7 or N9-based flu diagnosis. It seems to show that in a couple of instances H7-specific or H9-specific molecular methods will fail (1:10,000,000 is a pretty extreme dilution) to detect H7N9 at low levels while some assay fail to amplify the intended target altogether. 

This sort of assay variation is pretty commonplace when you compare different PCR designs for the same target. It's a major reason why everyone prefers to design their own assay; each are convinced there's is the best. From this document, the people behind the OFFLU (FLI-H7) assay happen to be right. 

How these data relate to viral load in the human (and animal) samples these may be/have been used on is unknown. This sort of variance may contribute to negative throat swabs in H7N9-cases subsequently proven positive using lower respiratory tract samples (e.g. sputum) in which the viral load is presumed to be higher.

OFFLU: a network of expertise on animal influenza established jointly in 2005 by the World Organisation for Animal Health (OIE) and the Food and Agriculture Organization of the United Nations (FAO) to support and coordinate global efforts to prevent, detect and control important influenzas in animals.
CNIC: Chinese National Influenza Center