BACKGROUND: Assessment of hemoglobin is one of the most reliable indicators for anemia, and is widely used to screen for anemia among pregnant women. The HemoCue® has been widely used for as a point-of-care device for hemoglobin estimation in health facilities. Previous studies showed contradictory results regarding the accuracy of HemoCue®. METHODS: This was a hospital-based cross sectional study carried- out among pregnant women at Khartoum hospital in Sudan to find out whether the measurement of hemoglobin concentration by HemoCue® using venous or capillary samples was comparable to that of the automated hematology analyzer as standard. Bland and Altman method was used to compare the measurements with an acceptable difference of ± 1.0 g/dl. RESULTS: Among the 108 subjects in this study the mean (SD) level of hemoglobin level using HemoCue® venous sample, HemoCue® capillary sample and automated hematology analyzer were 12.70 (1.77), 12.87 (2.04) and 11.53 (1.63) g/dl, respectively. Although the correlations between the measurements were all significant there was no agreement between HemoCue® and automated hematology analyzer. The bias + SD (limits of agreement) for HemoCue® venous versus hematology analyzer was 1.17 ± 1.57 (-1.97, 4.31) g/dl, HemoCue® capillary versus hematology analyzer was 1.34 ± 1.85 (-2.36, 5.04) g/dl, and HemoCue® venous versus HemoCue® capillary samples was 017 ± 1.90 and (3.97-3.63) g/dl. CONCLUSION: Hemoglobin concentration assessment by HemoCue® using either venous or capillary blood samples has shown unacceptable agreement with automated hematology analyzer. VIRTUAL SLIDES: The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/8797022296725036.