Indirectly determined reference intervals for automated white blood cell differentials of pediatric patients in Berlin and Brandenburg

OBJECTIVES: Establishing direct reference intervals for pediatric patients is a costly, challenging, and time-consuming enterprise. Indirectly established reference intervals can help to ameliorate this situation. It was our objective to establish population-specific reference intervals for automated white blood cell differentials via data mining and non-parametric percentile method. METHODS: Blood counts and automated white blood cell differentials of patients aged 0 days to 18 years, performed from the 1st of January 2018 until the 30th of June 2022, were identified in our laboratory information system. Reference intervals were established in accordance with IFCC and CLSI recommendations as well as the propositions by Haeckel et al. RESULTS: Initially, 47,173 blood counts on our SYSMEX XN-9000 were identified. 11,707 data sets were excluded, leaving 35,466 sample sets for analysis. Of these, 17,616 contained automated white blood cell differentials. Due to insufficient patient numbers, no reference intervals for automated white blood cell differentials could be established for children aged <7 months. In comparison to the corresponding reference intervals published by Herklotz et al., reference intervals determined by us showed relevant differences throughout all age groups. CONCLUSIONS: The combination of non-parametric percentile method and the propositions by Haeckel et al. utilizing conscientious data mining appears to be potent alternative to direct reference interval determination.