Comparison of haemoglobin assessments by HemoCue and two automated haematology analysers in young Laotian children

Background Haemoglobin (Hb) assessment by Hemocue is used widely for anaemia screening in both adults and children. However, few studies have compared the diagnostic accuracy of Hemocue with an automated haematology analyser in young children. Aim To compare Hb concentrations by Hemocue Hb301 and two automated haematology analysers in young children in rural communities of Lao PDR. Methods Capillary blood was collected from 6-month-old to 23-month-old children (n=1487) for determination of Hb concentration by Hemocue Hb301. On the same day, venous blood was collected for complete blood count using one of two haematology analysers (XT-1800i, Sysmex, and BC-3000Plus, Mindray Medical International). In a subsample of children (n=129), venous Hb was also measured by HemoCue Hb301. Agreement between the two methods was estimated using Bland-Altman plots. Results Mean capillary Hb by Hemocue was significantly higher than mean venous Hb by haematology analysers combined (108.410.3g/L vs 102.3 +/- 13.1g/L; P<0.001), resulting in a significantly lower anaemia prevalence (Hb <110g/L) by Hemocue (53.7% vs 73.9%; P<0.001). The Bland-Altman assessment of agreement showed a bias of 6.1g/L and limits of agreement were -11.5g/L to 23.7g/L. Mean venous Hb concentration by Hemocue Hb301 (113.6 +/- 14.0g/L) was significantly higher than mean capillary Hb concentration by Hemocue Hb301 (110.0 +/- 10.7; P=0.03g/L), which in turn was significantly higher than mean venous Hb concentration by the Mindray BC-3000Plus (102.3 +/- 17.4g/L). Conclusion Capillary and venous Hb concentrations assessed by Hemocue Hb301 showed poor agreement compared with venous Hb by automated haematology analysers, resulting in significantly different anaemia prevalences.