Background: The aim of the study was to assess the reliability of hemoglobin measurements made with HemoCue®, compared with those made with the reference method in critically ill patients.
Methods: A prospective observational study was conducted in three adult surgical intensive care units of a university hospital. One hundred and ninety-eight consecutive patients were included, and a total of 1166 hemoglobin concentrations were measured using arterial blood samples in the laboratory (HbLAB) and at bedside (HbHC.art) with a portable hemoglobinometer (HemoCue® Hb201+). Simultaneously, a capillary measurement (HbHC.cap) was performed at bedside using the same device.
Results: The mean difference (bias) between HbHC.cap and HbLAB was 0.2 g/dL (95%CI, 0.1;0.3), and limits of agreement were -1.3 g/dL (95%CI, -1.4;-1.2) to 1.7 g/dL (95%CI, 1.6;1.9). The discrepancies between HbHC.cap and HbLAB were greater than 1 g/dL in 30.8% of cases. The bias between HbHC.art and HbLAB was -0.1 g/dL (95% CI, -0.2;0.2), and limits of agreement were slightly better at -1.1 g/dL (95% CI, -1.2;-1.0) and 1.0 g/dL (95% CI, 0.9;1.1). The HemoCue®'s accuracy was not affected by the hospital unit, the puncture site (finger or ear), norepinephrine administration or by hemoglobin levels below 10 g/dL or 8 g/dL.
Conclusion: Capillary HemoCue® is not sufficiently accurate to make a therapeutic decision such as whether a blood transfusion should be performed. The method's performance was moderately improved by the use of arterial blood.