Evaluation of a portable glucometer for blood glucose measurement in juvenile green sea turtles (Chelonia mydas).

Blood glucose is a key physiological indicator in sea turtles, but its measurement in field settings can be challenging. Portable point-of-care (POC) glucometers offer practicality, yet their agreement with reference laboratory methods has not been established for juvenile green sea turtles (Chelonia mydas) from the Southwestern Atlantic. We compared glucose concentrations from a portable glucometer with a reference enzymatic method using blood samples from 30 free-ranging juvenile green turtles captured in Rio de Janeiro, Brazil. Agreement was assessed by Wilcoxon signed-rank test, Passing-Bablok regression and Bland-Altman analysis. The influence of fibropapillomatosis and packed cell volume was also examined. The glucometer significantly underestimated glucose compared to the reference method (median 70 vs. 90 mg/dL, respectively, P < 0.001), with a mean bias of -17.9 mg/dL and wide limits of agreement. Passing-Bablok regression showed no significant systematic or proportional errors (95% CI for intercept included 0; 95% CI for slope included 1). Only 43% of glucometer readings fell within ±20% of the reference ranges reported for sea turtles. Packed cell volume correction slightly reduced the bias but did not improve agreement. Fibropapillomatosis status did not affect absolute glucose concentrations but influenced relative bias when packed cell volume was within the manufacturer's validated range. The portable glucometer is not interchangeable with the reference method for this species. While it may provide rapid field estimates, its results require cautious interpretation and, whenever possible, laboratory validation.