Blood Glucose Measurement in Rabbits - Performance Evaluation of Portable Glucometers

Abstract

Background: Blood glucose level measurement plays an important role in the diagnosis and prognosis of different clinical conditions in rabbits. Portable glucometers offer a quick and affordable tool to measure blood glucose. However, not all commercially available glucometers are suitable for this purpose in different animal species and in different clinical conditions. This study aimed to assess the performance of 4 portable glucometers compared to a biochemical analyzer using blood samples from both healthy and sick rabbits (Oryctolagus cuniculus).

Material,s Methods & Results: Blood samples were collected via jugular venipuncture from 51 rabbits. The blood glucose concentration was measured using 3 human (FO= FreeStyle Optium Neo®; AA= Accu-chek® Active; IS= Injex Sens II®) and 1 veterinarian (PGM= Wellion Vet®, Gluco Calea) glucometer, configured for dog (WD), cat (WC), and horse (WH). The results obtained from the glucometers were compared to those from the laboratory biochemical analyzer in wet biochemistry, using Spearman correlation, intraclass correlation coefficient, Bland-Altman analysis, error grid analysis and total error analysis. Blood glucose concentrations were underestimated in the human devices evaluated and overestimated in the veterinary device. Bland-Altman analysis revealed significant systematic differences in 3 of the 4 devices tested. Total error above 20% was observed in normoglycemic samples from 2 devices (FO; and PGM - WC= cat code) and in hyperglycemic samples from 1 glucometer (AA). All glucometers showed glycemic results within Zone A and B of the error grid analysis, although a glucometer (FO) present most of the results (60.8%) in Zone B. Only a single portable glucometer (IS) consistently met the objectives across all analyses.

Discussion: The sample included 26 males and 25 females, with ages ranging from 5 to 96 months. Thirty-five rabbits were healthy, while 16 had various clinical abnormalities. Significant differences in median glycemia were found between the biochemical analyzer and all PGMs, except for the IS glucometer. FO and AA showed significantly lower glycemic values, while WD, WC, and WH showed higher values. The IS glucometer demonstrated the smallest mean variation and highest intraclass correlation coefficient (ICC), indicating excellent agreement. Bland-Altman plots revealed that FO and AA underestimated glycemic values, while WD, WC, and WH overestimated them. Only IS showed agreement with the laboratory analyzer. Error grid analysis indicated no serious errors with any PGMs; however, IS had over 75% of results in Zone A, showing high agreement, while FO had the least favorable results. Total error analysis suggested that only IS consistently met accuracy objectives across various analyses. The study concluded that IS was the most accurate PGM, while others showed varying degrees of inaccuracy. Although PGMs are clinically acceptable based on ISO 15197:2013 standards, accuracy and precision issues remain, particularly for euglycemic and hyperglycemic samples. Although all glucometers showed clinically acceptable results in the error grid analysis, the accuracy varied among the devices. Collectively, the IS glucometer achieved the best results in measuring blood glucose in pet rabbits, demonstrating a better accuracy and lower tendency for systematic error compared to the laboratory analyzer. Despite the lower accuracy, the other evaluated glucometers showed potential for use in rabbits when a laboratory analyzer is not available, but should be used with caution, considering the glycemic range presented.

Keywords: glycemia, glycemic range, biochemical analyzer, Oryctolagus cuniculus, lagomorphs, diagnosis.