Hematological monitoring of oncology patients using the HemoScreen point of care analyser

Introduction

Complete blood count (CBC) analyses the cellular composition of the blood, including red blood cells, white blood cells and platelets. It provides valuable insight into overall health and is used in the diagnosis and monitoring of several medical conditions, including anaemia, infection, bleeding disorders and hematological malignancies. Among other things, CBC assessment is essential for monitoring patients with hematological disorders in order to asses disease progression, treatment efficacy and overall patient management.

Traditionally, CBC monitoring has involved laboratory-based techniques that require time-consuming specimen transport and processing, leading to potential delays in clinical decision-making. In addition, there is a growing trend in modern healthcare to treat and monitor patients closer to home. This shift is driven by the desire to improve patient comfort while optimising resource allocation within healthcare systems. The use of advanced telemedicine technologies and home monitoring models enables healthcare providers to deliver effective interventions and continuous care while minimising the need for frequent hospital visits. This approach contributes to more efficient healthcare delivery and improved patient outcomes (1,2). As these strategies evolve, there is a growing need for rapid and accurate diagnostic tools that enable real-time monitoring, particularly at the point of care (POC).

The HemoScreen POC device introduces an innovative approach to hematological analysis by combining flow cytometry, digital imaging and artificial intelligence (AI)-driven algorithms. This compact device uses microfluidic viscoelastic focusing and single plane imaging to identify blood cell types based on morphological features (3). Alerts are triggered by abnormal cells improving accuracy. The portable HemoScreen, developed by PixCell Medicals, performs a CBC on capillary or venous blood in approximately 5 minutes using disposable cartridges with pre-loaded reagents (4). Its portable nature makes it suitable not only for general outpatient use, but also for eventual home use. The performance of the HemoScreen has been compared in several publications in the past with other full hematology analysers such as the Sysmex XN and Abbott Cell/Dyn Sapphire. These studies have shown that the HemoScreen analyser is accurate and precise in the analysis of CBC parameters (3,5,6).

To date, HemoScreen has only been compared with other hematology analysers using the same type of blood, either venous or capillary. As a POC device, it is assumed that HemoScreen is mostly used with capillary blood close to the patient. In this study, we aimed to assess the performance of HemoScreen using capillary blood compared to the Sysmex XN-9000 using venous blood, the current gold standard. We performed tests on blood samples from 40 adult patients undergoing chemotherapy at Isala Hospital in The Netherlands. We also evaluated the performance of the Sysmex XN-9000 using capillary blood. Our results show a good correlation between HemoScreen and Sysmex and indicate that the use of capillary blood with HemoScreen provides similar accuracy in CBC measurements as Sysmex does with capillary blood.

Methods

Study population and sample collection

Two samples (capillary and venous blood) were collected simultaneously from 40 adult leukaemia patients at Isala Hospital in The Netherlands by a trained phlebotomist during October and November 2021. Capillary blood (500 microlitres in total) was collected by the finger prick method into minicontainers containing ethylenediaminetetraacetic acid (EDTA) (BD, Plymouth, UK) using a Safe T Pro lancet. All collections were successful. Venous blood was collected by venepuncture into EDTA tubes (BD Vacutainer, Plymouth, UK). Approximately 30 minutes after collection both types of blood were analysed in the laboratory.

The Isala Ethical Committee approved the method comparison study. All patients signed an informed consent before blood sampling.

Instrumentation

Venous blood was analysed on the Sysmex XN-9000 (Sysmex Co., Kobe, Japan). The Symex XN-9000 uses impedance technology, together with flow cytometry and fluorescent flow cytometry, to perform automated completed hematological analyses. Capillary blood was first analysed on the portable hematology HemoScreen device (PixCell Medical, Yokneam Ilit, Israel) and immediately afterwards on the Sysmex XN-9000. The blood sample (40 microlitres) was applied to the HemoScreen disposable test cartridge containing the reagent chamber and inserted into the device for automatic analysis. Results were displayed on the instrument screen within 5 minutes and stored for later analysis. The device does not require calibration or maintenance. The HemoScreen uses impedance technology for rapid analysis of blood samples and can measure a complete CBC with a 5-part differential.

Repeatability and imprecision

To calculate within-run repeatability and between-run imprecision, the control material CBC-Pix was measured on the HemoScreen 5 times per day for 5 days at two different levels (low and high) as described in Clinical and Laboratory Standards Institute (CLSI) protocol EP15A3 (7).

Statistical analysis

Coefficient of variation (CV) was calculated using Excel 2016 (Microsoft, Seattle, WA, USA) for hemoglobin, hematocrit, mean corpuscular volume (MCV), erythrocytes, platelets, leukocytes, lymphocytes, monocytes, neutrophils, eosinophils and basophils. Acceptance criteria for precision were established using the European Federation of Clinical Chemistry and Laboratory Medicine biological variation database (EFML BVD) (8) (Table 1), which is the most accurate database in Europe for the study of biological variation.

Comparison between HemoScreen and Sysmex was performed with Analyse-it using Passing-Bablok linear regression (9). The minimum total allowable error (TEa) used was described in the EFML BVD, where it is calculated according to the formula: TE < 1.65 × (0.5 × CVI) + 0.25 × (CVI² + CVG²)^1/2 (8). The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics board of Isala Klinieken Zwolle (No.: 210814) and informed consent was obtained from all individual participants.

Results

Repeatability and imprecision of the HemoScreen using control samples

The within-run repeatability and between-run imprecision of HemoScreen was analysed using two control samples (low and high concentration) for all parameters over 5 days, 5 times per day (Table 1).

The CV is in accordance with the acceptance criteria for precision of the EFML BVD, except for the CV of erythrocytes for within-run repeatability [measured 3.5% and 4% versus the calculated 2.8% CV (%) within subject of the EFML BVD].

Comparison between HemoScreen for capillary blood samples and Sysmex XN-9000 for venous blood samples

Forty capillary and venous samples were collected from oncology patients following the CLSI EP09 protocol for “Measurement Procedure Comparison and Bias Estimation Using Patient Samples” (10). One of the samples had questionable Sysmex results. It is likely that the blood was not well mixed after collection given the high hemoglobin and erythrocyte concentration. Therefore, it was excluded from the analysis.

The venous and the capillary blood samples were analysed using both the Sysmex XN-9000 and HemoScreen. The performance of the Sysmex XN-9000 and of the HemoScreen on capillary blood was compared to that on venous blood using the Sysmex XN-9000. Passing-Bablok regression analysis was used to analyse and represent the CBC parameters including hemoglobin, hematocrit, MCV, erythrocytes, platelets, leukocytes, lymphocytes, monocytes, neutrophils, eosinophils and basophils (Figure 1). The minimum TEa calculated by the EFLM BVD was used for all these parameters (Table 1) and presented in the graphic as dotted lines (Figure 1).

Figure 1 Comparison of the measurement of blood parameters in Sysmex XN-9000 and HemoScreen analysers. The measurement of Sysmex XN in venous blood and HemoScreen in capillary blood are compared (A,C,E,G,I,K,M,O,Q,S,U). And the measurement in Sysmex XN-9000 in venous blood is compared with the measurement in capillary blood (B,D,F,H,J,L,N,P,R,T,V). Dotted line: minimum TEa; solid line: regression; r: correlation coefficient. L, litre; MCV, mean corpuscular volume; fL, femtolitre; TEa, total allowable error.

The correlation coefficient (r) between hemoglobin, hematocrit, MCV, platelets, erythrocytes, leukocytes, lymphocytes and neutrophils was similar between HemoScreen and Sysmex in capillary blood versus Sysmex in venous blood (Figure 1). For all these parameters, the variation of the majority of the measurements is smaller than the minimum TEa. The correlation of eosinophils, monocytes and basophils is more variable. Capillary eosinophils measured with HemoScreen correlate better with the venous eosinophils than the capillary eosinophils measured with Sysmex (r=0.975 vs. r=0.843). On the contrary the correlation coefficient of monocytes and basophils is worse when measured in capillary blood with the HemoScreen versus Sysmex (r=0.041 vs. r=0.874).

Discussion

The aim of this study is to assess, for the first time, the comparability of capillary blood samples analysed on the HemoScreen POC instrument and venous blood samples analysed by the Sysmex XN-9000, specifically in the context of oncology patients at a hospital in The Netherlands.

HemoScreen is a rapid and user-friendly device for measuring CBC in venous or capillary blood (4). The last years, there has been a growing trend to monitor patients as close to home as possible, and such a device can have a major impact on the monitoring of chronically ill patients, such as those on chemotherapy, whose hematological parameters need to be monitored frequently.

Similar to other studies (5), the data from this study show generally good repeatability and precision of the HemoScreen. The intra- and inter-run coefficients of variation at a low and high concentration were within the limits established by the EFML BVD (8). Only the overall imprecision for erythrocytes was not met, which was also observed by Kristian Kur (6). This study also found that neutrophils and platelets did not meet the limits set by the selected database used to compare CV biological variation (11). The CV (%) for these parameters were slightly higher than ours in the low concentration between-day imprecision, but slightly lower for the high concentration between-day imprecision. These differences may be explained by differences in the protocol used to determine the CV (%) which involved 10 measurements instead of five.

To date, HemoScreen has only been compared with a routine hematology analyser using either venous or capillary blood in both instruments. In this study we aimed to compare the performance of HemoScreen in capillary blood with that of the Sysmex XN-9000 routine instrument in venous blood. At the same time, we looked at the correlation and performance of the Sysmex XN in capillary and venous blood to compare it with the results of the HemoScreen. We found that the HemoScreen was slightly less accurate than the standard instruments routinely used in clinical chemistry laboratories, such as the Sysmex XN-9000, but that the latter performed similarly to the HemoScreen when both were used with capillary blood. In both HemoScreen and Sysmex measurements with capillary blood, almost all measurements were within the minimum TEa (8) and showed similar correlation in both instruments for haemoglobin, haematocrit, MCV, platelets, leukocytes, lymphocytes and neutrophils. As previously described, basophils showed a very poor correlation with HemoScreen (4,5), probably a limitation due to the small volume of blood used for analysis in this device. This should not be a major problem for the use of HemoScreen in patient follow-up, as the clinical relevance of basophils during chemotherapy is low for the majority of patients. The correlation of eosinophils was better with the HemoScreen, although some differences can be seen between capillary and venous blood in the lower concentrations of eosinophils when measured with the HemoScreen. Further research is needed in this area and with different concentrations of basophils and eosinophils. Interestingly, both the HemoScreen and the Sysmex measured a lower MCV with capillary blood. This has not been seen in other studies comparing blood count parameters between venous and capillary blood, where no differences were found (12,13) or the MCV was significantly higher in capillary blood (14). Presumably, mixing with tissue fluid may occur due to skin puncture during capillary blood collection, potentially diluting cellular components like erythrocytes. This dilution effect may lead to a slightly lower MCV.

In this study the HemoScreen results have been compared across the entire measurement range simultaneously. Previously Kristian Kur et al. also compared the HemoScreen with the Sysmex XN-9000 at both the entire range and low end of the measurement range separately in either venous or capillary blood. In this study the correlation coefficient for hemoglobin, leukocytes, erythrocytes, platelets and neutrophils was greater than 0.90 in the lower range (6). It may be interesting in the future to investigate whether there are differences in the measurement characteristics at different measurement ranges in order to gain a more complete insight into the performance of the HemoScreen at low and high concentrations of the different parameters.

Our study investigated hematological parameters in 40 patients being treated for leukaemia. The selection of oncology patients provided valuable insights into device performance in this specific context, highlighting the potential benefits of patient-centred research in oncology studies. Extrapolation of these results to broader patient populations is possible but may require careful consideration of clinical factors. Therefore, future studies with a larger population and number of patients may be of great interest.

This study did not test the ability of HemoScreen to detect morphological abnormalities in leukocytes. Only a few studies have looked at the differential of HemoScreen and found that the specificity for analyser flagging was higher for HemoScreen than for Sysmex XN (5). The ability of HemoScreen to detect platelet aggregation and abnormalities in the morphology of differential cells in capillary blood versus venous blood in Sysmex may certainly add value to understanding the overall performance of HemoScreen.

Conclusions

In conclusion, our findings suggest that the HemoScreen offers a promising POC solution for reliable CBC measurements comparable to those obtained with conventional hematology analysers. This could be of significant benefit for oncology patients who require frequent monitoring, potentially revolutionizing their care management in the future.

Acknowledgments

Funding: None.

Footnote

Peer Review File: Available at https://jlpm.amegroups.org/article/view/10.21037/jlpm-23-95/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jlpm.amegroups.org/article/view/10.21037/jlpm-23-95/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics board of Isala Klinieken Zwolle (No.: 210814) and informed consent was obtained from all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Pronovost PJ, Cole MD, Hughes RM. Remote Patient Monitoring During COVID-19: An Unexpected Patient Safety Benefit. JAMA 2022;327:1125-6. [Crossref] [PubMed]
2. Lippert M, Semmens S, Tacey L, et al. The Hospital at Home program: no place like home. Curr Oncol 2017;24:23-7. [Crossref] [PubMed]
3. Bransky A, Larsson A, Aardal E, et al. A Novel Approach to Hematology Testing at the Point of Care. J Appl Lab Med 2021;6:532-42. [Crossref] [PubMed]
4. Ben-Yosef Y, Marom B, Hirshberg G, et al. The HemoScreen, a novel haematology analyser for the point of care. J Clin Pathol 2016;69:720-5. [Crossref] [PubMed]
5. Linko-Parvinen AM, Keränen K, Kurvinen K, et al. HemoScreen hematology analyzer compared to Sysmex XN for complete blood count, white blood cell differential, and detection of leukocyte abnormalities. EJHaem 2022;3:1126-34. [Crossref] [PubMed]
6. Kur KD, Thøgersen D, Kjeldsen L, et al. The HemoScreen hematology point-of-care device is suitable for rapid evaluation of acute leukemia patients. Int J Lab Hematol 2021;43:52-60. [Crossref] [PubMed]
7. Carey RN, Durham AP, Hauck WW, et al. EP15 User verification of precision and estimation of bias – 3rd edition. Clinical and Laboratory Standards Institute 2022. Available online: https://clsi.org/standards/products/method-evaluation/documents/ep15/
8. Aarsand AK, Fernandez-Calle P, Webster C, et al. The EFLM Biological Variation Database. Available online: https://biologicalvariation.eu/
9. Bilić-Zulle L. Comparison of methods: Passing and Bablok regression. Biochem Med (Zagreb) 2011;21:49-52. [Crossref] [PubMed]
10. Budd JR, Durham AP, Gwise TE, Hawkins DM, et al. Clinical and Laboratory Standards Institute. Measurement Procedure Comparison and Bias Estimation Using Patient Samples. 3rd edition. CLSI guideline EP09c, 2018. Available online: https://clsi.org/standards/products/method-evaluation/documents/ep09/
11. Coskun A, Braga F, Carobene A, et al. Systematic review and meta-analysis of within-subject and between-subject biological variation estimates of 20 haemato-logical parameters. Clin Chem Lab Med 2019;58:25-32. [Crossref] [PubMed]
12. Chavan P, Bhat V, Tiwari M, et al. Comparison of Complete Blood Count Parameters between Venous and Capillary Blood in Oncology Patients. J Lab Physicians 2016;8:65-6. [Crossref] [PubMed]
13. Doeleman MJH, Esseveld A, Huisman A, et al. Stability and comparison of complete blood count parameters between capillary and venous blood samples. Int J Lab Hematol 2023;45:659-67. [Crossref] [PubMed]
14. Schalk E, Heim MU, Koenigsmann M, et al. Use of capillary blood count parameters in adults. Vox Sang 2007;93:348-53. [Crossref] [PubMed]