Contributions
Abstract: PB1965
Type: Publication Only
Background
Assessment of erythrocyte morphology is essential in the study of anaemia. However, microscopic evaluation of peripheral blood smears is a time-consuming procedure. Automated digital cell morphology analysers could simplify and standardize this evaluation. These platforms have recently incorporated red blood cells (RBC) morphology evaluation.
Aims
Standardization of the morphological analysis of RBC performed by the CellaVision Advanced RBC Software Application.
Methods
Standardization: The values of macrocytosis and microcytosis proposed by Cellavision were recalibrated according to those obtained with XN analysers (Sysmex Corporation, Chuo-ku, Kobe, Japan) in several pathological samples. Subsequently, the reference interval for each of the different morphological anomalies of RBC was calculated in 67 subjects with normal mean corpuscular volume, RBC and reticulocyte count, and iron metabolism. Three levels of intensity were defined for each morphological anomaly according to the 2015 ICSH recommendations: 1 (mild), 2 (moderate) and 3 (intense).
The intra-day and inter-day reproducibilities were assessed. For the intra-day study, 5 smears of the same blood sample were analysed along the same day. For the inter-day study, three different blood smears were obtained for each sample and analysed at 0, 24, and 48 hours.
Validation: 100 patients with anaemia were selected. The results of the automated system were compared with those obtained through microscopic evaluation of direct blood smears (without anticoagulant). For each patient, morphology was evaluated by the automated digital system, by an experienced technician, and by a hematologist (the reference). According to results, morphology was divided into three categories: normal or without clinical relevance, compatible with the type of anaemia (e.g. hypochromia in iron deficiency anemia), and relevant for the diagnosis (e.g., schistocytes).
STATISTICS: In the reference population, intervals were calculated and morphological findings were compared using the Student's t-test. In the validation study, concordance (agreement) between results was analysed by the kappa index statistic.
METHODOLOGY: Blood was collected in EDTA-anticoagulated tubes. The automated morphological analysis was performed with the CellaVision Advanced RBC Application on the DM96 analyser (CellaVision AB, Lund, Sweden). Blood smears for the automated analysis were performed and stained using the SP-10 (Sysmex Corporation, Chuo-ku, Kobe, Japan), an automated slide maker and stainer.
Results
Statistically significant differences were found for macrocytosis and dacryocytes with regard to sex. In the analysis of the intra-day reproducibility, no differences were observed. A decrease in stomatocytes and an increase in equinocytes were detected in the inter-day analysis.
VALIDATION: Results obtained by the technician using the direct blood smear coincided with those retrieved by the haematologist in 92.5% of cases (κ = 0.918). The automated analysis agreed with the hematologist’s results in 93.7% of the cases (κ = 0.879).
Conclusion
The automated evaluation of RBC morphology using the CellaVision Advanced RBC Software Application offered reproducible intra-day but not inter-day results. Its standardization allowed the evaluation of the RBC morphology in a faster way and with decreasing variability. There was an excellent concordance between the morphological assessment by CellaVision and by optical microscopy in the setting of anaemia.
Session topic: 29. Enzymopathies, membranopathies and other anemias
Keyword(s): Anemia, Red blood cell, Standardization
Abstract: PB1965
Type: Publication Only
Background
Assessment of erythrocyte morphology is essential in the study of anaemia. However, microscopic evaluation of peripheral blood smears is a time-consuming procedure. Automated digital cell morphology analysers could simplify and standardize this evaluation. These platforms have recently incorporated red blood cells (RBC) morphology evaluation.
Aims
Standardization of the morphological analysis of RBC performed by the CellaVision Advanced RBC Software Application.
Methods
Standardization: The values of macrocytosis and microcytosis proposed by Cellavision were recalibrated according to those obtained with XN analysers (Sysmex Corporation, Chuo-ku, Kobe, Japan) in several pathological samples. Subsequently, the reference interval for each of the different morphological anomalies of RBC was calculated in 67 subjects with normal mean corpuscular volume, RBC and reticulocyte count, and iron metabolism. Three levels of intensity were defined for each morphological anomaly according to the 2015 ICSH recommendations: 1 (mild), 2 (moderate) and 3 (intense).
The intra-day and inter-day reproducibilities were assessed. For the intra-day study, 5 smears of the same blood sample were analysed along the same day. For the inter-day study, three different blood smears were obtained for each sample and analysed at 0, 24, and 48 hours.
Validation: 100 patients with anaemia were selected. The results of the automated system were compared with those obtained through microscopic evaluation of direct blood smears (without anticoagulant). For each patient, morphology was evaluated by the automated digital system, by an experienced technician, and by a hematologist (the reference). According to results, morphology was divided into three categories: normal or without clinical relevance, compatible with the type of anaemia (e.g. hypochromia in iron deficiency anemia), and relevant for the diagnosis (e.g., schistocytes).
STATISTICS: In the reference population, intervals were calculated and morphological findings were compared using the Student's t-test. In the validation study, concordance (agreement) between results was analysed by the kappa index statistic.
METHODOLOGY: Blood was collected in EDTA-anticoagulated tubes. The automated morphological analysis was performed with the CellaVision Advanced RBC Application on the DM96 analyser (CellaVision AB, Lund, Sweden). Blood smears for the automated analysis were performed and stained using the SP-10 (Sysmex Corporation, Chuo-ku, Kobe, Japan), an automated slide maker and stainer.
Results
Statistically significant differences were found for macrocytosis and dacryocytes with regard to sex. In the analysis of the intra-day reproducibility, no differences were observed. A decrease in stomatocytes and an increase in equinocytes were detected in the inter-day analysis.
VALIDATION: Results obtained by the technician using the direct blood smear coincided with those retrieved by the haematologist in 92.5% of cases (κ = 0.918). The automated analysis agreed with the hematologist’s results in 93.7% of the cases (κ = 0.879).
Conclusion
The automated evaluation of RBC morphology using the CellaVision Advanced RBC Software Application offered reproducible intra-day but not inter-day results. Its standardization allowed the evaluation of the RBC morphology in a faster way and with decreasing variability. There was an excellent concordance between the morphological assessment by CellaVision and by optical microscopy in the setting of anaemia.
Session topic: 29. Enzymopathies, membranopathies and other anemias
Keyword(s): Anemia, Red blood cell, Standardization