13 Oct, 15 | by Bob Phillips
Please tell me that I’m not the only one, who hearing about the magical properties of S100, CD64, microbiomology or ILx (where x >8, probably prime, and is instantly forgettable) recalls the scam of the century as told by Hans Christian Andersen in 1837.
Well. There may be more to biomarkers of disease than just a tailor wanting to get away with an awful lot of cash for a very little work.
In a really nice review article by Pak Cheung Ng and colleagues in the F&N edition, they describe the ideal properties of a biomarker for detecting neonatal sepsis or necrotising enterocolitis. Even if you don’t ‘do’ neonates, if you’re involved with acute paediatrics of any or no subspeciality the themes they bring out are worth thinking on:
- Your biomarker should be able to be done on a TINY bit of blood/fluid
- That blood shouldn’t mind if it’s venous or capillary
- It should probably be interpreted in the light of the clinical setting (with prognostic scoring)
- It might be a diagnostic / early warning marker, or an organ specific damage marker, or a response to treatment marker – or ideally all of those – but your interpretation should involve understanding what the biomarker is doing
- It should ideally be able to be done without the lab (point of care)
It’s worth thinking about the biomarkers you probably do request … I’d class a full blood count, serum albumin, lactate level or CRP in this gang … and work out what they actually do, rather than what you want them to do, and how well they do it … and then when you see the next paper declaiming F64z-soluble-activation-factor-subunit-B as the perfect marker for lupus renalitis (or whatever), you’ll have an idea of what questions need to be asked of that study.