The shifting significance of serial serum chemistries
Frustrated by what is really significant when performing serum chemistry profiles? At CVC Kansas City, Craig Ruaux, BVSc (Hons), PhD, MACVSc, DACVIM-SA, an associate professor of small animal internal medicine at Oregon State University’s College of Veterinary Medicine, discussed several reasons for variability in serum chemistry profile results in his lecture “Let’s just monitor it: The pitfalls and problems with serial serum chemistries.” He said these reasons can be divided into three different categories:
- Venipuncture technique
- Interassay variation
- Assay operator
- Diurnal variation
- Seasonal variation
Biological variation can include variation from the group as whole as well as variation from each individual. Dr. Ruaux says that for many of the compounds we measure, there appears to be little utility in comparing an individual to a population. Substantial deviation from normal for one individual patient may be lost in the reference range.
If you’re seeing variations in your results, just run through the list above of what differences might account for them.
Diving deeper into the variability of two specific tests
Some tests do have little analytic variability, says Dr. Ruaux. Two examples Dr. Ruaux discussed were C-reactive protein (C-RP) and the canine pancreas-specific lipase (Spec cPL—Idexx) tests. With the canine pancreas-specific lipase test, there is low variation from one dog to another, so it is useful to compare to a reference range, but you need a large change to be significant because individual dogs vary a lot over time. Persistently high canine pancreas-specific lipase results imply a persistent release because of a change within the pancreas. When monitoring a chronic pancreatitis patient with repeated pancreas lipase measurements, it is unknown how much change you need to see to know if things are getting better or worse. This is why Dr. Ruaux instead typically monitors clinical signs such as diarrhea, appetite, weight gain and attitude.