Hello,
In a study of a very old population I have one subject who has extremely high lactate in occipital cortex (~4 mmol/kg) measured with HERCULES and processed using Osprey. Anatomically, they also have a significant amount of gray matter atrophy. The actual spectra and model fits look reasonable to me (attached are figures of the Osprey model fits). I’m extracting the lactate concentration value from the tissue corrected Diff2 table.
The lactate value is more than 2.5 standard deviations greater than the mean of other older adults in the study. How do others decide whether to include or exclude these subjects in their analysis?
That rather depends on the research question, and how your subject groups are defined. As you mention, the fit looks reasonable: I’d say that’s a fairly convincing Lactate peak, so there’s no reason to reject this on the grounds of fit/quality.
If this is a group of nominally healthy older adults, you could consider whether the significant grey matter atrophy and significantly elevated Lactate is consistent with that grouping.
Another consideration though: elevated CSF lactate is expected in older subjects, so I do wonder if the CSF-corrected values from OspreyQuantify are appropriate in this case. It could well be that reduced GM/WM volume (gray matter atrophy) is leading to an over-estimate of Lactate from OspreyQuantify. What happens if you scale the Lactate values by (1 - fCSF)?
