Validation of Noninvasive Methemoglobin and Carboxyhemoglobin Measurements Using Pulse Co-Oximeter in Healthy Dogs.

Abstract

Purpose: To assess the agreement between pulse co-oximeter and blood co-oximeter in measuring methemoglobin (MetHb) and carboxyhemoglobin (COHb) in healthy, awake dogs.

Patients and methods: Forty-five healthy dogs were enrolled in the study. Pulse co-oximetry was performed using the Masimo Radical 7 Pulse Co-Oximeter with a Rainbow^® adhesive sensor. Simultaneously, venous blood samples were collected, and MetHb and COHb were immediately measured using a Stat Profile Prime Plus VET Critical Care Analyzer. Paired measurements of MetHb and COHb were evaluated via Spearman correlation, intra-class correlation (ICC), and Bland-Altman plots to evaluate the degree of agreement between the pulse co-oximeter and the blood co-oximeter.

Results: A total of 45 paired MetHb and COHb measurements were collected. There was a weak correlation between the pulse co-oximeter and the blood co-oximeter readings. The correlation coefficients for MetHb and COHb were 0.0 (95% CI, -0.3 to 0.3) and 0.03 (95% CI, -0.27 to 0.32), respectively. The ICC indicated poor agreement between the pulse and blood co-oximeter for MetHb (ICC = 0.00, 95% CI: -0.12 to 0.15) and COHb (ICC = 0.03, 95% CI: -0.27 to 0.33). Bland-Altman plots revealed low mean bias but wide limits of agreement, indicating that the pulse co-oximeter overestimated MetHb by on average of 0.7% (P < 0.0001) (95% LoA: -0.5 to 2.0) and COHb by on average 0.2% (P = 0.59) (95% LoA: -4.6 to 5.0).

Conclusion: Obtaining MetHb and COHb measurements with the Masimo Radical 7 Pulse Co-Oximeter is straightforward in healthy, awake dogs. However, the device does not provide accurate measurements compared to the blood co-oximeter, specifically in the range of MetHb and COHb in healthy dogs, based on the wide LoA.