Determination of Hematological Reference Intervals for Healthy Adults in Northeast Ethiopia

BACKGROUND: Hematological reference interval is the range between two reference values that are used for inter-pretation of test results. It is affected by various physiological and environmental factors; thus, locally derived hematological reference values are essential for accurate diagnosis and treatment of patients. The main goal of this study was to establish hematological reference intervals for healthy adults at Kemise, Northeast Ethiopia. METHODS: A cross-sectional study was conducted from January to April, 2019, with 170 male and 159 female apparently healthy adult blood donors at Kemise Blood Bank. A structured pretested questionnaire was used for socio demographic and clinical data collection. About 4 mL of blood was collected in an EDTA test tube and analyzed using Sysmex XP-300 to enumerate the hematological parameters. The data were collected and entered into Epi-Inf7 and then transferred to SPSS version 20 for analysis. Dixon and Reed 1/3 rule was used for outlier detection. Mann-Whitney U test was used to determine reference intervals. Harris and Boyd test were used to determine the reference intervals that need partition. The 2.5th and 97.5th percentiles were determined non-parametrically with 95% confidence interval. RESULTS: The 2.5th - 97.5th percentiles reference intervals for red blood cell males: 3.9 - 6 x 1012/L, females: 3.3 - 5.3 x 1012/L; hemoglobin males: 12 - 17.9 g/dL, females: 10.5 - 16.3 g/dL; and hematocrit males: 34.8 - 51.4%, females: 30.2 - 47%. The combined reference intervals for both genders, were mean corpuscular volume: 77.1 - 95.5 fL, mean corpuscular hemoglobin: 25.6 - 34 pg, mean corpuscular hemoglobin concentration: 31.5 - 36.7 g/dL, red cell distribution width: 11.3 - 15.4%, white blood cells: 3.1 - 10.6 x 109/L, lymphocytes: 1 - 3.7 x 109/L, mixed: 0.3 - 1.7 x 109/L, neutrophils: 1.4 - 7 x 109/L, and platelets: 136 - 425 x 109/L. CONCLUSIONS: The established reference intervals in this study were different from previous studies which were conducted in different regions of Ethiopia or African countries or in Caucasian population, and in text books. The RBC, PCV, and Hgb reference intervals were different in gender. So, using of locally determined reference interval is essential.