Comparison of a continuous indwelling glucometer with a point-of-care device in healthy adult horses

• Published in: Veterinary record Vol. 187; no. 3; p. e21
• Main Authors: Cunneen, Alexandra, Wood, Kelly A, Mathison, Kylie, Herndon, Aaron M, Bertin, Francois R
• Format: Journal Article
• Language: English
• Published: England BMJ Publishing Group Limited 08.08.2020; Blackwell Publishing Ltd
• Subjects: Animals; Blood Glucose - analysis; Blood Glucose Self-Monitoring - instrumentation; Blood Glucose Self-Monitoring - veterinary; Catheters; Collections; critical care; Diabetes; diagnostic testing; endocrinology; glucose; Glucose monitoring; Horses; Horses - blood; Hyperglycemia; Hypoglycemia; Insulin; Laboratories; Monitoring systems; Point-of-Care Systems; Reproducibility of Results; Sensors; Software; Veins & arteries; Veterinary medicine; Australia; United States > US; critical care; glucose; diagnostic testing; insulin; endocrinology
• ISSN: 0042-4900; 2042-7670
• DOI: 10.1136/vr.105607

BackgroundBlood glucose is tightly regulated in horses; however, since hypoglycaemia and hyperglycaemia are associated with poor prognosis, close monitoring is warranted. This study aimed at evaluating a continuous indwelling glucometer (CIG) by comparing performance with a point-of-care glucometer (POC).MethodsTen horses were equipped with CIG and an intravenous catheter. Interstitial glucose concentrations were determined by CIG every 5 min at rest, during insulin-induced hypoglycaemia and dextrose-induced hyperglycaemia, and compared with blood glucose determined by POC. Glucose concentrations were compared by two-way repeated measures analysis of variance and weighted kappa with Bland-Altman plots to determine agreement between assays.ResultsHorses tolerated CIG well; however, five devices had to be replaced. There were no statistically significant differences between assays at rest or during hyperglycaemia; however, during hypoglycaemia, glucose concentrations determined by CIG were significantly higher (P=0.01). The mean bias (95% limits of agreement) between assays ranged from −0.03 (−2.46 to 2.52) mmol/l (hyperglycaemia) to 0.97 (−1.23 to 3.16) mmol/l (hypoglycaemia). Assay agreement was ‘good’ with observed agreements of 87.04% (κ=0.67).Conclusions of the studyCIG has acceptable accuracy in horses as compared with POC but overestimates glucose concentrations during hypoglycaemia and requires frequent replacement, limiting its clinical application.