Skip to main content

A comparative study of blood cell count in four automated hematology analyzers: An evaluation of the impact of preanalytical factors.

Åstrand A, Wingren C, Walton C, Mattsson J, Agrawal K, Lindqvist M, Odqvist L, Burmeister B, Eck S, Hughes G, Luporini Saraiva G, Schantz A, Psallidas I, McCrae C

Publication: PLoS One (2024) • Vol 19 • Page e0301845PMID: 38787860View on PubMed

Purpose

This paper compared differential cell counts, eosinophil enumeration, and markers of eosinophil activation, death, and eosinophil-derived neurotoxin (EDN) in whole blood samples collected from healthy volunteers and patients diagnosed with eosinophilic asthma analyzed in parallel on four hematology analyzers after delays to centrifugation (3-72 h) at different temperatures (4, 20, 30, and 37°C) with and without agitation.  

Conclusion

Comparable cell counts were obtained with all four hematology analyzers when whole blood was analyzed within 24 h of collection. Delay and temperature-specific effects in cell differential counts were observed for specific hematology analyzers after a ≥24 h delay to centrifugation, the magnitude of change was especially robust for eosinophils; no hematology analyzer generated stable differential cell counts across all delays and temperatures evaluated. Differential counts and EDN levels were not affected by whether the volunteer had asthma or was healthy, or whether the blood sample experienced agitation during the delay. Analysis of eosinophils by flow cytometry revealed marker-specific effects on cell counts that were dependent upon temperature when the delay was >24 h.

Studies (1)

Purpose

This paper compared differential cell counts, eosinophil enumeration, and markers of eosinophil activation, death, and eosinophil-derived neurotoxin in whole blood samples collected from healthy volunteers and patients diagnosed with eosinophilic asthma analyzed in parallel on four hematology analyzers after delays to centrifugation (3-72 h) at different temperatures (4, 20, 30, and 37°C) with and without agitation.  Whole blood was collected from 12 healthy volunteers (5 males, 7 females; 6 with atopy and 6 without), and 6 individuals diagnosed with asthma (3 males, 3 females) into K2EDTA tubes.  Blood specimens experienced a delay to centrifugation of 3 h (baseline), 6, 24, 48, and 72 h at 4, 20, 30, and 37°C with and without agitation on an orbital shaker for 4 min at 400 rpm twice per day prior to analysis.  Differential cell counts were quantified using four hematology analyzers (Abbott CELL-DYN Sapphire, Beckman Coulter DxH900, Siemens ADVIA 2120i, and Sysmex XN-1000V), and markers of eosinophil activation and death were measured by flow cytometry. Levels of eosinophil-derived neurotoxin (EDN) were measured in plasma and serum with the IDK EDN ELISA Kit.

Summary of Findings

The number of neutrophils, eosinophils, and monocytes quantified by four different hematology analyzers did not differ significantly from one another when baseline samples (a 3 h delay to centrifugation) were compared; lymphocyte (p=0.006 and p=0.024, respectively) and platelet (p=0.027 and p=0.002, respectively) counts were slightly higher on the Abbott Cell-DYN Sapphire and Sysmex XN-10000V machines and basophil counts were slightly higher with the Beckman Coulter DxH900 machine (p=0.038) relative to readings obtained from the other hematology analyzers with baseline samples. Correlations between samples with a 3 and 6 h delay to centrifugation were strong (r=0.88 – 0.98) for a given analyzer for platelets, neutrophils, lymphocytes, and eosinophils, but the strength of the correlation was lower for platelets quantified by the XN-1000V (r=0.71), and for monocytes (r=0.72) and basophils (r=0.31) measured by the DxH900. The authors report that white blood cell counts were more stable in blood stored at 20°C when processed within 24 h of collection than longer delays or other temperatures. Differential counts did not differ significantly between blood specimens that were shaken and those that were not shaken during the delay. When delays to blood processing exceeded 24 h, specific hematology analyzers over- or under-estimated the counts of specific cell types relative to values obtained from baseline (3 h delay) specimens.  When blood specimens were refrigerated (4°C) during a delay that was >24 h, DxH900, XN-1000V, and 2120i underestimated neutrophils; DxH900 overestimated lymphocytes and monocytes; 2120i underestimated lymphocytes and overestimated basophils; and 2120i, DxH900, and XN01000V underestimated platelets.  Only the Abott CELL-DYN Sapphire did not display a significant difference in differential counts between specimens that experienced a 3 h or 48 h delay at 4°C. When blood specimens experienced a delay at room temperature that was >24 h, DxH900, the Sapphire, and XN-1000V, underestimated neutrophils; the 2120i underestimated lymphocytes and eosinophils; the Sapphire underestimated eosinophils; DxH900 and XN-1000V underestimated monocyte counts; basophil counts remained stable with all four hematology analyzers. When blood specimens experienced a delay at 30°C that was > 24h, the Sapphire underestimated platelets; the Sapphire and 2120i underestimated eosinophils; the 210i overestimated monocytes; the DxH900, XN-1000V, and the Sapphire overestimated neutrophils; DxH900 and XN-1000V underestimated monocytes; and the Sapphire and 2120i overestimated basophils.  Cell counts differed for most cell types when a delay to centrifugation occurred at 37°C regardless of analyzer.  The authors stated that eosinophils were the cell type that exhibited the biggest differences between hematology analyzers, and the magnitude of change observed between baseline (3 h delay) and 24 and 48 h samples at 20°C was greater than the difference observed between healthy and asthmatic volunteers. Analysis of eosinophils by flow cytometry revealed marker-specific effects on cell counts that were dependent upon temperature when the delay was >24 h.  A delay at 4 or 20°C for >24 h resulted in stable counts of live-dead aqua, CD62L+, CD123+, and FceR1+ cells but greater counts of CD66ba+ and CD11b+ cells. A delay at 30°C >48 h resulted in stable counts of live-dead aqua, FceR1+, and CD123+ cells while CD62L+, Cd11b+, and CD66b+ cell counts decreased. Eosinophil-derived neurotoxin levels were stable when blood was stored at 20°C for up to 72 h, but were elevated in plasma samples from blood specimens that experienced a delay of 48 h at 30 or 37°C.

  • Biospecimens

    FluidBlood
    FluidPlasma
    CellWhite Cells
  • Preservation Types

    None (Fresh)
  • Diagnoses

    Normal
    Asthma
  • Platforms

    Cell count/volumeFlow cytometry
    Cell count/volumeHematology/ auto analyzer
    ProteinELISA
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Eosinophilic asthma
    Healthy with atopy
    Healthy without atopy
    Biospecimen Aliquots and ComponentsCentrifugation
    Centrifugation delays investigated
    StorageStorage conditions
    With agitation
    Without agitation
    StorageStorage duration
    24 h
    3 h
    48 h
    6 h
    72 h
    StorageStorage temperature
    20°C
    30°C
    37°C
    4°C
    Hematology/ auto analyzer SpecificTechnology platform
    Abbott CELL-DYN Sapphire
    Beckman Coulter DxH900
    Siemens ADVIA 2120i
    Sysmex XN-1000V