Recently, we blogged about a paper by Kyle & Lawrence that demonstrated poor precision of the lamellar body counts (LBC) using the Coulter Unicel DxH 800 instrument. Now we want to make you aware of a study by Beaudoin and others that demonstrates a bias in LBCs between the Sysmex XE-2100 and Sysmex XT-2000i instruments.
Previous reports have demonstrated excellent concordance between the LBC when performed on Sysmex XE-2100 and the Coulter-brand instruments. However the Beaudoin laboratory recently noticed a positive bias in LBC results using the Sysmex XT-2000i. Therefore, they conducted a study to compared LBC results in 52 specimens using two Sysmex models (Sysmex XE-2100 and Sysmex XT-2000i), with those of the Coulter LH-750.
The authors demonstrated a very small negative bias (mean = -8%) between the Coulter LH-750 and the Sysmex XE-2100. However, the Sysmex XT-2000i had significantly higher results than the Coulter LH-750 (mean bias = 56%). Interestingly, despite the difference in LBCs, similar results were obtained on all 3 instruments when platelets were counted using Sysmex e-CheckTM control material or whole blood specimens.
Using Deming regression to extrapolate the appropriate cutoff, the authors calculated that the medical decision limit for the Sysmex XT-2000i is closer to 77,000 counts/µL, which is 54% higher than the 50,000 counts/µL recommended for the Coulter LH-750. The reason for this difference is not understood, but is likely related to the different principles to count lamellar bodies on each instrument. Although, all three of the analyzers in this study use impedance principles to count platelets, each analyzer counts slightly differently. The authors speculate that the difference may be due to the fluid used for the hydrodynamic focusing of the samples.
This study once again highlights the need for manufacturer- as well as model-specific LBC clinical decision limits. Laboratories performing LBCs on hematology analyzers should be aware of this important analytical variable. Individual laboratory LBC cutoffs need to be established even when using the analyzers from the same manufacturer if the models are different.