Accuracy and precision of a point-of-care hemoglobinometer for measuring hemoglobin concentration and estimating packed cell volume in horses.

Objective: To determine accuracy and precision of a point-of-care hemoglobinometer for measuring hemoglobin concentration and estimating PCV in horses.
Design: Prospective trial.
Animals: 55 horses.
Procedure: Blood samples were obtained from 43 horses examined at a veterinary teaching hospital. Hemoglobin concentration was measured with the hemoglobinometer and by means of the standard cyanmethemoglobin method; PCV was measured by centrifugation. Blood samples were also obtained from 12 healthy horses, and PCV of aliquots of these samples was altered to approximately 5 to 80% by removing or adding plasma. Hemoglobin concentration and PCV were then measured.
Results: For samples from the clinic patients, hemoglobin concentrations obtained with the hemoglobinometer were less than concentrations obtained with the cyanmethemoglobin method; however, there was a linear relationship between concentrations obtained with the 2 methods. Breed, sex, body weight, and duration of sample storage did not significantly affect the difference between hemoglobin concentrations obtained with the 2 methods. There was a significant linear relationship between PCV and hemoglobinometer hemoglobin concentration (PCV = [2.83 x hemoglobin concentration] - 0.62). For samples from the healthy horses, a substantial negative bias was evident with the hemoglobinometer when hemoglobin concentration exceeded 16 g/dL.
Conclusions and Clinical Relevance: Results suggest that this hemoglobinometer is reasonably accurate and precise when used to measure hemoglobin concentration in blood samples from horses with a hemoglobin concentration < 16 g/dL.