Oliver Fiehn states that there is a central website for storing metabolomic data from studies (blood, tissue, all species). At present there are 994 studies posted.
https://www.metabolomicsworkbench.org/
You can search for Human=> Blood => Chronic Fatigue Syndrome here
https://www.metabolomicsworkbench.org/data/metsearch_MS_form2.php
This brings up studies from Robert Naviaux (2 datasets) and Oliver Fiehn/Ian Lipkin (4 datasets) for those interested in raw data number crunching.
ETA: If you leave the sample source blank and search Human => <Blank> => Chronic Fatigue Syndrome there are also three datasets for CSF fluid from Fiehn/Lipkin. Does anyone know if this study of CSF has been published?
ETA2: FYI - Raw data for CSF fluid is 6G + 4.5G + 0.5G.
To my knowledge, this is the first study on metabolites in the csf. I believe it's yet unpublished work. It's not clear how differentiated mannose was from controls, but it would be great to finally have something really out of whack. I have no idea what low mannose could mean.