Abstract:

Introduction: Blood gas analyzers (BGA) have recently been widely used as a rapid testing devices for the determination of hemoglobin (Hb) in the intensive care units and emergency services of hospitals. We compared the Hb and calculated-hematocrit (Hct) values by the CO-oximetry and cyanide-free sodium lauryl sulphate (SLS) methods. Methods: Between January and June 2019, 12,049 patients who applied to the emergency department of İstanbul Training and research Hospital, for whole blood count and venous blood gas analysis were included. Samples were analyzed using SLS- Hb and CO-oximetry methods. Bland-Altman plot and Passing-Bablok regression analysis were performed to evaluate the accordance of the methods. Results: The correlation coefficients of the methods for Hb and Hct were 0.89 and 0.87, respectively (p<0.0001). Passing-Bablok regression analysis showed a significant deviation from linearity (p<0.01). Bland-Altman plot showed insufficient agreement between of the two methods for each variable. Bias % calculated as 2.5% for Hb, and 1.1% for Hct. Total error calculated as 4.08% for Hb. Total error of CO-oximetric Hb value was within the limits of allowable total error. Conclusion: Although each test shows a significant deviation from linearity, BGA’s could be used for Hb measurements since the bias and total error were still acceptable.

Keywords: