The Evaluation of a Point-of-Care Hematology Analyzer Hemoscreen for Abnormal Blood Samples

Introduction:

One of the main adverse effects of modern hematooncological treatment is hematotoxicity which has to be closely monitored and managed, often by administering blood transfusions or hematopoietic growth factors. Patients undergoing active treatment have to come to the outpatient clinic frequently for blood collection, sometimes several times a week. HemoScreen is a hematology point-of-care analyzer able to measure complete blood count with 5-part differential (total of 20 parameters) from both venous and capillary blood sample in approximately 5 minutes. The device employs image-based analysis and accepts disposable cartridges containing all required reagents which makes it maintenance free and thus a potentially suitable device for home blood count monitoring in hematooncology. The aim of this part of our project was to evaluate reliability of HemoScreen compared to laboratory hematology analyzers in the examination of pathological/abnormal blood samples.

Methods:

Venous (S-Monovette 2.6 mL K3EDTA tubes) and capillary blood (Microvette 500 EDTA K3E) were obtained from 113 hematological patients and 23 healthy donors after signing the informed consent. Regarding pathological cases, there were 21% (28/136) of patients with acute leukemia,17% (23/136) with multiple myeloma, 14% (19/136) with myelodysplastic syndrome and 11% (15/136) with malignant lymphomas. The cohort of interest were patients with absolute neutrophil count (ANC) ˂ 2x10⁹/L, hemoglobin (HGB) ˂ 100 g/L and platelets (PLT) ˂ 50x10⁹/L. Venous samples were analyzed on Sysmex XN-9000, Beckman Coulter DxH 560, Yumizen H550 and HemoScreen and capillary samples on HemoScreen only. Results of white blood cells (WBC), ANC, HGB and PLT from venous and capillary sampling were compared and statistically evaluated using the statistical software RStudio 4.2.0. Sysmex XN-9000 was set as a reference analyzer due to its standard use in University Hospital Ostrava.

Results:

In the first part of the evaluation, measurement of venous and capillary collection on HemoScreen did not differ significantly and maximum deviation of measurement by capillary method was 16.16%. Subsequently, the statistically significant difference (p-value of paired t-test) was calculated for the whole cohort and for specific subgroups (normal, neutropenic, anemic and thrombocytopenic). P-value < 0.05 was identified only in thrombocytopenic cohort. This anomaly was caused by not respecting the threshold set by a manufacturer, which is PLT level of ˂ 20x10⁹/L. After excluding those values, no statistically significant differences were found. At that time, there was no low platelet flagging in HemoScreen. For the analysis of absolute mean deviation, we also evaluated the entire dataset and then pathologic subgroups separately. This time, PLT level of ˂ 20x10⁹/L was primarily excluded from the thrombocytopenic cohort. Across all parameters, the largest absolute mean deviation was found in platelet measurements. The percentage of deviation for platelet levels ranged from 29.46% in anemic group to 24.45% in neutropenic cohort. The average deviation was 6% (3.92%-6.62%) for HGB, 11% (7.49%-14.24%) for WBC and 16% for ANC (10.96%-18.29%). The absolute mean deviation was compared to Sysmex XN-9000. In general, HemoScreen showed steady results with no detected systematic errors. The outliers were identified as random errors.

Conclusions

This project showed that HemoScreen is reliable and sufficiently accurate in comparison to standard hematology analyzers even for abnormal blood samples (when the measuring limits of the device are respected) and could be eligible for home blood count monitoring for patients with blood disorders.

Popkova:Johnson and Johnson: Honoraria; Sanofi: Other: travel support. Hajek:Takeda: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria, Research Funding; Celgene: Consultancy, Honoraria, Research Funding; AbbVie: Consultancy; BMS: Consultancy, Honoraria, Research Funding; PharmaMar: Consultancy, Honoraria; Novartis: Consultancy, Research Funding.