Reliability of the HemoCue^® hemoglobinometer in critically ill patients

Introduction

In the ICU, the hemoglobin concentration is currently estimated by portable hemoglobinometer and in emergency transfusion a decision can be based on that result. The aim of the study was to compare the HemoCue^® with laboratory determination of the blood hemoglobin concentration in critically ill patients with a high risk of bleeding.

Methods

A prospective observational study was conducted in three adult surgical ICUs of a university hospital. After ethic committee approval, 198 consecutive patients requiring 1,166 determinations of hemoglobin concentration were included. Hemoglobin was measured from an arterial blood sample at the laboratory (Hb_m) and estimated at the bedside (H_be,a) using the same blood sampling and a portable hemoglobinometer (HemoCue^® Hb201+). Simultaneously, a capillary measurement (Hb_e,cap) was performed at the bedside using the same device.

Results

The mean difference (bias, Figure 1) between H_be,cap and Hb_m was -0.2 g/dl (95% CI, -0.3 to -0.1) and limits of agreement -1.3 g/dl (95% CI, -1.4 to -1.2) to 1.7 g/dl (95% CI, 1.6 to 1.9). Discrepancies between H_be,cap and Hb_m were greater than 1 g/dl in 30.8% of cases. The accuracy of HemoCue^® was not affected by the unit of hospitalization, the patients' severity assessed by SAPS II score on admission, the site of lancet puncture (finger or ear), the administration of norepinephrine, the presence of hypothermia (body temperature <36.3°C) or by a hemoglobin level below 10 g/dl or 8 g/dl. The bias between H_be,a and Hb_m was -0.1 g/dl (95% CI, -0.2 to 0.2) and limits of agreement -1.1 g/dl (95% CI, -1.2 to -1.0) and 1.0 g/dl (95% CI, 0.9 to 1.1).

Figure 1

(abstract P2)

Conclusions

Capillary HemoCue^® is not accurate enough to base therapeutic decisions such as blood transfusion on. The performance of the method is improved with the use of arterial blood.