The equivalence and reliability of point-of-care devices routinely used for haemoglobin measurement compared with the laboratory standard



point-of-care, devices, haemoglobin measurement, laboratory standard


Background: Point-of-care (POC) technology aims to hasten diagnosis in the clinical setting. Reducing samples, especially invasive samples, for the same result will reduce the work volume, and potentially the cost. Faith in POC devices requires finding the most reliable POC device in the interest of cost and patient benefit, especially in an intensive care unit (ICU).

Methods: This study assessed the haemoglobin (Hb) correlation between invasive and non-invasive POC devices at a quaternary trauma ICU. A non-invasive, continuous Hb pulse oximeter was compared with two invasive POC devices, as well as the laboratory reference measurement. A 10% variance was considered a clinically significant difference.

Results: A total number of 48 arterial blood gas (ABG) measurements, 49 Mission haemoglobinometer measurements, 50 Masimo Radical-7 measurements, and 52 Inkosi Albert Luthuli Central Hospital (IALCH) laboratory Hb measurements were performed with a mean Hb of 8.8. The Masimo non-invasive haemoglobinometer failed to demonstrate interchangeability with the GEM blood gas analyser machine (p < 0.004), neither did it show interchangeability with the IALCH laboratory measurements (p < 0.001), or the Mission haemoglobinometer (p < 0.0014). Comparing the POC devices with the results from the laboratory revealed that the ABG machine (GEM 4000) was the closest to correlate with the laboratory measurements (mean device difference of 0.51, p < 0.007). The values for the Mission haemoglobinometer were the next closest, but both the devices slightly overestimate the Hb in the order of 6–7%.

Conclusion: This study compared the equivalence of the Hb level measured by two currently used POC devices (the Gem 4000 and Mission haemoglobinometer) against a novel, continuous, non-invasive device (the Masimo device) and the reference standard laboratory Hb level. The study found that the novel device did not perform equivalent or better than the existing devices.

Author Biographies

S Govender, University of KwaZulu-Natal

Department of Anaesthesia, King Edward VIII Hospital and Department of Anaesthesia and Critical Care, University of KwaZulu-Natal, South Africa

TC Hardcastle, University of KwaZulu-Natal

Department of Surgery, University of KwaZulu-Natal and Trauma and Burns Services, Inkosi Albert Luthuli Central Hospital, South Africa






Original Research