Reference Intervals for SII, NLR, PLR, and LMR: a Comparison of Nonparametric and refineR Approaches Using Big Data

Based on the widespread use of the systemic immune-inflammation index (SII), neutrophil-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), and lymphocyte-monocyte ratio (LMR), markers, we aimed to calculate and compare the reference intervals (RIs) of these indices in adults, using both nonparametric method according to the Clinical and Laboratory Standards Institute's (CLSI) EP28-A3C:2010 guideline and refineR algorithm using a large dataset.

We analyzed data from 293,585 adults (18 - 65 years) retrospectively obtained from complete blood count results (using laboratory information system). The study involved a two-stage outlier exclusion process. White blood cells, platelets, neutrophils, lymphocytes, and monocytes were measured using the Advia 2120i (Siemens Healthineers, Germany), and SII, NLR, PLR, and LMR were calculated. Subgroups were formed based on age and gender, and RIs were estimated using an indirect sampling technique, specifically the nonparametric statistical method (2.5th and 97.5th percentiles) and refineR algorithm.

Our study showed that the values of SII, NLR, PLR, and LMR varied significantly between males and females. For all the parameters, the lower and upper limits of reference intervals (RIs) for males were higher when using the RefineR algorithm. RIs based on age groups using CLSI guideline and the refineR algorithm were calculated, showing wider limits of the RIs in refineR.

We detected a difference between the RI values of the indexes according to the different RI calculation methods and observed that the reference intervals calculated according to CLSI guideline have a narrower range than by refineR algorithm.