Precision and accuracy of a point-of-care glucometer in horses and the effects of sample type.
Authors: Rendle D I, Armstrong S K, Heller J, Hughes K J
Journal: Veterinary journal (London, England : 1997)
Summary
# Editorial Summary Point-of-care glucometers are widely deployed in equine practice and research, yet uncertainty persists about their reliability compared to laboratory analysers. Rendle and colleagues tested a portable glucometer across 62 horses using four sample types—plain whole blood, EDTA tubes, fluoride oxalate (FO) tubes measured immediately, and FO samples measured after 3 hours—comparing results to laboratory analysis within one hour of collection. Plain whole blood proved unsuitable for glucometry, exhibiting poor precision (coefficient of variation 5.5–9.4%), whilst EDTA and FO samples demonstrated acceptable precision (0.7–3.5%); however, all glucometer methods underestimated glucose concentrations relative to laboratory analysis, with limits of agreement ranging from 2.75–4.65 mmol/L depending on sample type. The practical implication is that practitioners using point-of-care glucometers in horses should verify their specific device's performance and establish method-specific reference ranges rather than relying on standard laboratory values, and should prioritise FO or EDTA samples over whole blood draws. Given that glucometer accuracy varies considerably between devices and sample handling protocols, validation should be completed before implementing any portable glucose monitoring in clinical or research settings.
Read the full abstract on PubMed
Practical Takeaways
- •Do not use plain whole blood for point-of-care glucometry in horses; switch to EDTA or fluoride oxalate tube samples for reliable results
- •Establish method-specific reference ranges for your glucometer and sample type before clinical use, as this device systematically under-estimates glucose concentrations
- •Fluoride oxalate samples provide stable glucose measurements for up to 3 hours, allowing flexibility in timing of analysis
Key Findings
- •Whole blood samples showed inadequate precision (coefficient of variation 5.5-9.4%) compared to EDTA and fluoride oxalate tubes (0.7-3.5%)
- •Limits of agreement were ≥4.65 mmol/L for whole blood vs laboratory methods but ≤2.75 mmol/L for EDTA and fluoride oxalate comparisons
- •Fluoride oxalate and EDTA samples provided adequate precision and accuracy when method-specific reference ranges were used to account for systematic under-estimation
- •Sample type significantly influenced glucometer performance, with whole blood unsuitable for point-of-care glucose measurement in horses