Initial performance evaluation of the UniCel® DxH 800 Coulter® cellular analysis system.
Bottom Line: The DxH 800 showed accurate PLT and NRBC counts in the clinically significant low range when compared to FCM.Compared to the LH 780, flagging rates were significantly reduced (NRBC flag), or equivalent (WBC, PLT flag) on the DxH 800.The DxH 800 demonstrated higher sensitivity and specificity for PLTs and NRBCs and achieved a lower NRBC false negative rate compared to the LH 780.
Affiliation: Department of Hematology, London Health Sciences Centre, London, Ontario, Canada.Show MeSH
Related in: MedlinePlus
Mentions: A subset of all the samples (n = 57) contained megathrombocytes, a cellular characteristic that potentially could influence the PLT count. The DxH 800 showed strong correlation with both reference data sets again (LH 780 and FCM, Figure 2b) with r2 values of 0.996 and 0.974, respectively. A second subset of samples containing microcytic or fragmented RBCs (n = 28) were evaluated and again the DxH 800 analyzer demonstrated good correlation with the LH 780 and FCM data (Figure 2c) with r2 values of 0.998 and 0.966, respectively. Both of these PLT subsets showed no significant difference between groups (anova, P = 0.514 and P = 0.903, respectively). PLT accuracy data at <50, <20, <10 × 109/l PLTs are shown in Table 4 show an impressive performance, particularly in the lower ranges. Of particular note, those samples with PLT counts <20 × 109/l which showed disagreement, were flagged for review by the analyzer. The number of all samples with flagged PLT values was similar (10% on the DxH 800 and 9% on the LH 780) in this group of predominantly hematologically abnormal samples.
Affiliation: Department of Hematology, London Health Sciences Centre, London, Ontario, Canada.