Search for


TEXT SIZE

search for



CrossRef (0)
Comparison of Red Blood Cell, White Blood Cell and Differential Counts between UF-5000 System and Manual Method
J Lab Med Qual Assur 2019;41:172-178
Published online September 30, 2019
© 2019 Korean Association of External Quality Assessment Service.

Mo Sae Koo, Jinsook Lim, Seon Young Kim, Sun Hoe Koo, and Gye Cheol Kwon

Department of Laboratory Medicine, Chungnam National University College of Medicine, Daejeon, Korea
Correspondence to: Gye Cheol Kwon
Department of Laboratory Medicine, Chungnam National University Hospital, Chungnam National University College of Medicine, 282 Munhwa-ro, Junggu, Daejeon 35015, Korea
Tel: +82-42-280-7799 Fax: +82-42-280-5365 E-mail: kckwon00@naver.comkckwon00@naver.com
Received May 28, 2019; Revised July 8, 2019; Accepted July 8, 2019.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
 Abstract

Background:

Analysis of body fluids provides important information for assessing various medical conditions. We aimed to validate the analytical and diagnostic performance of the Sysmex UF-5000 (Sysmex, Japan) system for the analysis of different body fluids.

Methods:

Eighty body fluid samples were analyzed using the UF-5000 system in the body fluid mode and light microscopy. Body fluids included ascitic, pleural, and cerebrospinal fluid (CSF), as well as other fluid samples.

Results:

A comparison between the UF-5000 system and manual counting demonstrated good correlations with regard to red (r=0.6555) and white blood cell (r=0.9666) counts. The UF-5000 system also demonstrated good performance for differential cell counting (r=0.9028). CSF particularly showed a good correlation.

Conclusions:

The use of the UF-5000 system for cell counting and differential analysis of body fluid samples might be an effective and automated alternative to chamber counting in laboratory routine analysis, thereby enhancing laboratory workflow and clinical effectiveness.

Keywords : Sysmex UF-5000, Automation, Laboratory, Body fluids
INTRODUCTION

Cell counting in body fluids, including ascitic fluid, pleural fluid, and cerebrospinal fluid (CSF), provides important information for clinicians in patients with various medical conditions. For example, increased white blood cell (WBC) counts in CSF might assist in the diagnosis of meningeal inflammation-related disease (meningitis, encephalitis, brain abscess, etc.). Furthermore, the elevation of WBC counts in ascitic fluid and pleural fluid might help in the diagnosis of peritonitis and parapneumonic effusion, respectively. High WBC counts in other fluids, such as synovial fluid, urine, and bronchoalveolar lavage, might help diagnose synovitis, urinary inflammation, and pneumonia, respectively.

Through body fluid examination, malignant or inflam-matory cells can be detected, and an increase in cell counts relative to the reference range can facilitate further evaluations by differential cell counting [1]. When the WBC counts in bronchoalveolar lavage are above the reference range, the proportion of neutrophils, lymphocytes, and eosinophils can be determined by differential cell counting. Moreover, differential cell counting is helpful for cirrhotic patients with ascites. Such patients are susceptible to spontaneous bacterial peritonitis (SBP). According to clinical diagnostic guidelines, SBP is present when >250×106 polymorphonuclear (PMN) cells are counted per liter of ascitic fluid [2]. Peritonitis is a serious disease, necessitating tests that yield rapid and reliable laboratory results.

The reference method for cell counting in body fluids is light microscopy (LM) using a counting chamber [1,3]. However, this method is time-consuming and requires experienced laboratory technicians [3]. Furthermore, this method has high intra-and inter-operator variability [3,4]. To overcome these problems, many automatic analyzers have emerged in recent years [4-9], which have been demonstrated to be rapid, reliable, and accurate for counting red blood cells (RBCs) and WBCs (differential) in CSF and other body fluids. These analyzers also help improve workflow in a routine laboratory. However, studies indicate that these analyzers cannot replace microscopic analysis of samples with abnormal findings. In particular, CSF from patients with leukemia or lymphoma should be processed using the microscopic reference method to detect abnormal leukemic cells.

The UF-5000 system (Sysmex Co., Kobe, Japan) uses fluorescent flow cytometry technology and hydrodynamic focusing to analyze body fluids and perform differential cell counts. A recent study has shown that the UF-5000 system exhibits good performance for counting nucleated cells in the CSF [10]. However, to the best of our knowledge, no studies have examined the performance of the UF-5000 system for body fluids other than CSF.

Therefore, we aimed to validate the analytical and diagnostic performance of the UF-5000 system for analyzing various body fluids. We compared the results of the UF-5000 system with the traditional chamber method for cell counts and differential counts.

MATERIALS AND METHODS

A total of 80 body fluid samples from hospitalized patients were analyzed. These included ascitic fluid (30 samples), pleural fluid (22 samples), CSF (16 samples), and other fluids (12 samples). All samples were collected in sterile tubes. The study lasted for 2 months and was approved by the ethical committee of Chungnam National University Hospital. Informed consent was obtained from all patients, and the study was carried out in accordance with the Declaration of Helsinki.

Manual microscopic cell counting was performed in a counting chamber using LM. The samples were analyzed according to the Clinical and Laboratory Standards Institute [11] and International Council for Standardization in Haematology (ICSH) guidelines [12]. To ensure standardization of the procedure for manual counting, certified pipettes and sterile chambers were used. In addition, all UF-5000 body fluid measurements were performed in accordance with the manufacturer’s instructions and recommendations. UF-5000 body fluid parameters were compared with cell counts obtained by LM counting chambers. CSF was analyzed when the WBC count was greater than 20/µL, whereas other fluids were examined by differential counts when the WBC count was greater than 40/µL. The correlation of dysmorphic RBCs was not performed, and all evaluated RBCs were morphologically normal. PMN cells were not evaluated separately (e.g., basophils, eosinophils, etc.). Epithelial cells and malignant cells were not evaluated in this study. The numbers of bacteria and epithelial cells were not evaluated because of an insufficient number of samples.

The agreement of the two methods was assessed using Passing and Bablok regression and Bland-Altman plot analyses. The slope and intercept of Passing and Bablok regression were calculated within 95% confidence intervals (CIs) to highlight any significant differences between the methods.

For the Bland-Altman plots, differences were plotted against the results of the chamber counts. Seventy-nine samples were used for RBC and WBC counts, and 51 samples were used for differential cell counting.

The correlation coefficient (r) was compared according to body fluid type. The fluid type consisted of ascitic fluid, pleural fluid, and CSF. For each sample, correlation coefficients were calculated for the RBC, WBC, and differential counts.

Statistical analysis was carried out using MedCalc Statistical Software ver. 18.2.1 (MedCalc Software bvba, Ostend, Belgium; http://www.medcalc.org; 2018).

RESULTS

A total of 80 samples were collected and evaluated in this study. All the samples were analyzed using the UF-5000 system and LM counting, and the evaluations were compared. One sample was excluded because the RBC and WBC counts were inaccurate. Table 1 shows the distribution of the gender and age of the subjects. Body fluids were divided into four categories (ascitic, pleural, CSF, and other fluids). Other fluids included bile, pericardial effusion, renal cyst, liver cyst, and percutaneous transhepatic biliary drainage. The sex ratio of the patients sampled for total body fluids was 3:2 (male:female). The distribution of ages showed a predominance of elderly individuals. In particular, pleural and other fluids showed a higher representation of elderly individuals. Meanwhile, CSF was represented by a relatively higher proportion of younger patients.

Table 1 . Comparison of gender and average age according to body fluids.

SampleSexMean of age (range)

Male (%)Female (%)
Ascitic fluid16 (53.3)14 (47.7)57 (21–80)
Pleural fluid15 (68.2)7 (31.8)75 (38–92)
Cerebrospinal fluid10 (62.5)6 (37.5)36 (0–78)
Other fluid4 (36.4)7 (63.6)68 (19–95)
Total45 (60.0)34 (43.0)58 (0–95)

Table 2 shows the distribution of patient diseases and total body fluids. Diseases were divided into three categories: inflammation, malignancy, and “others.” The inflammation category comprised various inflammatory diseases: meningitis, pneumonia, cholangitis, and urinary tract infection, among others. The malignancy category included tumors and hematologic cancers (stomach cancer, hepatocellular carcinoma, lung cancer, leukemia, malignancy of unknown origin, etc.). The “other” category included liver cirrhosis and chronic kidney disease, among others. In the total body fluids, the malignancy category formed the largest of the three categories (54.4% in total body fluids). Ascitic fluid and pleural fluid (66.7% and 63.6%, respectively) constituted a large proportion of malignancies among the three categories. The inflammation category was predominantly observed in other fluids (72.7%).

Table 2 . Distribution of patients’ diseases.

SampleInflammation (%)Malignancy (%)Others (%)
Ascitic fluid5 (16.7)20 (66.7)5 (16.7)
Pleural fluid4 (18.2)14 (63.6)4 (18.2)
Cerebrospinal fluid3 (18.8)7 (43.8)6 (37.5)
Other fluid8 (72.7)2 (18.2)1 (9.1)
Total20 (25.3)43 (54.4)16 (20.3)

We compared the UF-5000 system with the LM counting method for RBC and WBC counts and differential counts (Figs. 1-4). Fig. 1A shows the comparison of RBC counts between the UF-5000 system and chamber method. For this comparison, the Passing and Bablok regression slope was 1.12 (95% CI, 1.03 to 1.19) and intercept was 1.16 (95% CI, 0.00 to 16.86). Fig. 1B shows the Bland-Altman analysis for the RBC counts. The Bland-Altman mean value was –15165.4 (95% CI, –31,873.7247 to 1,542.9045), and 79 samples were used. The overall correlation of RBC counts between the UF-5000 system and chamber method was found to be statistically significant and high (r=0.6555, P<0.0001).

Figure 1.

(A) Comparison of RBC counts between the UF-5000 system and chamber method. (B) Comparison of RBC counts based on Bland-Altman plots. Difference in RBC counts=RBC counts by UF-5000 minus RBC counts by manual method. Abbreviations: RBC, red blood cell; SD, standard deviation.


Fig. 2A shows the comparison of WBC counts between the UF-5000 system and chamber method. In this comparison, the Passing and Bablok regression slope was 1.24 (95% CI, 1.11 to 1.40) and intercept was 7.77 (95% CI, 1.10 to 34.93). Fig. 2B shows the Bland-Altman analysis for WBC counts. The Bland-Altman mean value was 304.7 (95% CI, 23.1007 to 586.2664), and 79 samples were used. The overall correlation of WBC counts between the UF-5000 system and chamber method was statistically significant and very high (r=0.9666, P< 0.0001).

Figure 2.

(A) Comparison of WBC counts between the UF-5000 system and chamber method. (B) Comparison of WBC counts based on Bland-Altman plots. Difference in WBC counts=WBC counts by UF-5000 minus WBC counts by manual method. Abbreviations: WBC, white blood cell; SD, standard deviation.


Fig. 3A shows the comparison of PMN cell counts between the UF-5000 system and chamber method. In this comparison, the Passing and Bablok regression slope was 1.07 (95% CI, 0.95 to 1.23) and intercept was –11.30 (95% CI, –17.90 to –7.17). Fig. 3B shows the Bland-Altman analysis for PMN cells. The Bland-Altman mean value was –11.2 (95% CI, –14.7405 to –7.7575), and 51 samples were used. The overall correlation of differential cell counts for PMN cells between the UF-5000 system and chamber method was statistically significant and very high (r=0.9028, P<0.0001).

Figure 3.

(A) Comparison of PMN cell counts between the UF-5000 system and chamber method. (B) Comparison of PMN cells (%) based on Bland-Altman plots. Difference (%) in PMN cells=percentage of PMN cells by UF-5000 minus percentage of PMN cells by manual method. Abbreviations: PMN, polymorphonuclear; SD, standard deviation.


Fig. 4A shows the comparison of mononuclear (MN) cell counts between the UF-5000 system and chamber method. In this comparison, the Passing and Bablok regression slope was 1.07 (95% CI, 0.95 to 1.23) and intercept was 4.20 (95% CI, –5.20 to 11.73). Fig. 4B shows the Bland-Altman analysis for MN cells. The Bland-Altman mean value was 11.1 (95% CI, 7.6575 to 14.6405), and 51 samples were used. The overall correlation of differential cell counts for MN cells between the UF-5000 system and chamber method was statistically significant and very high (r=0.9028, P<0.0001).

Figure 4.

(A) Comparison of MN cell counts between the UF-5000 system and chamber method. (B) Comparison of MN cells (%) based on Bland-Altman plots. Difference (%) in MN cells=percentage of MN cells by UF-5000 minus percentage of MN cells by manual method. Abbreviations: MN, mononuclear; SD, standard deviation


In Table 3, CSF shows a very good correlation coef-ficient of 0.9 or more among RBC and WBC counts and differential counts. Ascitic fluid showed a very good correlation coefficient of 0.9775 for the WBC count, and pleural fluid showed a good correlation coefficient of 0.8952 for the differential count. Other fluids were not evaluated for correlation because of heterogeneity.

Table 3 . Comparison of correlation coefficients according to body fluid type.

SampleCorrelation coefficient (r)

Red blood cellWhite blood cellPolymorphonuclearMononuclear
Ascitic fluid (n=30)0.79380.97750.76760.7676
Pleural fluid (n=22)0.62500.80250.89520.8952
Cerebrospinal fluid (n=16)0.99780.98620.99280.9928

DISCUSSION

We sampled four types of body fluids: ascitic fluid, pleural fluid, CSF, and other fluids. For the sampled pleural fluid, 77.3% of the patients were at least 65 years old. This might be ascribed to a greater number of cases of pneumonia or lung cancer being found in older patients. In contrast, young individuals (younger than 20 years of age) accounted for 50% of the CSF fluid samples. Many babies and children were evaluated for meningitis and hematologic malignancies to rule out meningeal inflammation and metastasis of hematologic malignancies (lymphoma and leukemia), respectively.

We compared the results obtained from the UF-5000 system with those obtained by LM. Our results showed good correlations for RBC and WBC counting between these two methods. Moreover, the UF-5000 system demonstrated good performance for differential cell counting of body fluids. The evaluation of the UF-5000 system was carried out according to the ICSH guidelines [12], and the performance of the UF-5000 system was consistent with the declarations of the manufacturer.

In previous studies, WBC counting and analysis of MN and PMN cell distributions from body fluid samples have been performed using microscopic quantification in chambers. However, this method is time-consuming and labor-intensive. Therefore, automated body fluid analysis methods have been developed, particularly for hematology parameters [4-8,10,13-15]. Most recently, the UF-5000 system has been used to evaluate CSF samples [10]. Previous reports have suggested good correlations between automated WBC counts and traditional microscopic analysis, consistent with our current findings [4-7,9,10].

We also examined the correlations of each fluid. Among the three body fluids, CSF showed the best correlation coefficient. In ascitic fluid, the WBC count showed a high correlation coefficient, and pleural fluid showed a high correlation coefficient in the differential count. Overall, the three body fluids showed good correlation, and CSF was the most relevant and clinically useful.

Although the UF-5000 system in body fluid mode is capable of accurate WBC counts, it is critical to manually inspect samples with high WBC counts to identify pathological cells because the UF-5000 system cannot differentiate malignant cells from normal cells. Therefore, the manual method is still needed to differentiate and detect malignant cells and to determine specific cell types in body fluids. Moreover, although the overall agreement between the UF-5000 system and chamber method was statistically significant, there were some samples with high standard deviations, particularly for PMN and MN cell differentiation. These findings may be explained by the presence of debris or fragments in the analyzed samples, which may be interpreted by the instrument as PMN cells.

Some limitations of the study should be considered when interpreting the results. First, we used only a small number of samples. Furthermore, the distribution of samples was not even or heterogeneous (most samples were ascitic and pleural fluids or CSF). Therefore, additional studies using a larger number of samples from various sources are needed to confirm and expand upon our findings.

Although the analyzer cannot replace microscopic analysis, microscopy should be performed for samples with high proportions of abnormal cells and samples with very low or high WBC counts. Nonetheless, automation can reduce the routine workload of technicians and enable rapid analysis in laboratories with the help of nonskilled technicians.

Our findings suggest that the UF-5000 system might be useful for the initial screening of body fluids in urgent situations, when the availability of experts and trained personnel cannot be ensured.

In conclusion, the results of our study provide evidence that the use of the UF-5000 system for cell counting and differential cell counting in body fluid samples can provide an effective and automated alternative to chamber counting in routine analysis in the laboratory, thereby enhancing the laboratory workflow. In contrast to many other analyzers, this system also enables differential leukocyte counting. Accordingly, this system may have high diagnostic potential, particularly in the emergency department.

References
  1. Fleming C, Russcher H, Lindemans J, de Jonge R. Clinical relevance and contemporary methods for counting blood cells in body fluids suspected of inflammatory disease. Clin Chem Lab Med 2015;53:1689-706.
    Pubmed CrossRef
  2. Angeloni S, Nicolini G, Merli M, Nicolao F, Pinto G, Aronne T, et al. Validation of automated blood cell counter for the determination of polymorphonuclear cell count in the ascitic fluid of cirrhotic patients with or without spontaneous bacterial peritonitis. Am J Gastroenterol 2003;98:1844-8.
    Pubmed CrossRef
  3. Zimmermann M, Ruprecht K, Kainzinger F, Heppner FL, Weimann A. Automated vs. manual cerebrospinal fluid cell counts:a work and cost analysis comparing the Sysmex XE-5000 and the Fuchs-Rosenthal manual counting chamber. Int J Lab Hematol 2011;33:629-37.
    Pubmed CrossRef
  4. De Jonge R, Brouwer R, de Graaf MT, Luitwieler RL, Fleming C, de Frankrijker-Merkestijn M, et al. Evaluation of the new body fluid mode on the Sysmex XE-5000 for counting leukocytes and erythrocytes in cerebrospinal fluid and other body fluids. Clin Chem Lab Med 2010;48:665-75.
    Pubmed CrossRef
  5. Fleming C, Brouwer R, Lindemans J, de Jonge R. Validation of the body fluid module on the new Sysmex XN-1000 for counting blood cells in cerebrospinal fluid and other body fluids. Clin Chem Lab Med 2012;50:1791-8.
    Pubmed CrossRef
  6. Fleming C, Brouwer R, van Alphen A, Lindemans J, de Jonge R. UF-1000i:validation of the body fluid mode for counting cells in body fluids. Clin Chem Lab Med 2014;52:1781-90.
    Pubmed CrossRef
  7. Lehto TM, Leskinen P, Hedberg P, Vaskivuo TE. Evaluation of the Sysmex XT-000i for the automated body fluid analysis. Int J Lab Hematol 2014;36:114-23.
    CrossRef
  8. Sandhaus LM. Body fluid cell counts by automated methods. Clin Lab Med 2015;35:93-103.
    Pubmed CrossRef
  9. Paris A, Nhan T, Cornet E, Perol JP, Malet M, Troussard X. Performance evaluation of the body fluid mode on the platform Sysmex XE-5000 series automated hematology analyzer. Int J Lab Hematol 2010;32:539-47.
    Pubmed CrossRef
  10. Seghezzi M, Manenti B, Previtali G, Alessio MG, Dominoni P, Buoro S. Preliminary evaluation of UF-5000 Body Fluid Mode for automated cerebrospinal fluid cell counting. Clin Chim Acta 2017;473:133-8.
    Pubmed CrossRef
  11. Clinical and Laboratory Standards Institute. User protocol for evaluation of qualitative test performance:approved guideline:second edition, EP12-A2. Wayne (PA): Clinical and Laboratory Standards Institute, 2008.
  12. Bourner G, De la Salle B, George T, Tabe Y, Baum H, Culp N, et al. ICSH guidelines for the verification and performance of automated cell counters for body fluids. Int J Lab Hematol 2014;36:598-612.
    Pubmed CrossRef
  13. Park J, Kim J. Evaluation of iQ200 automated urine microscopy analyzer. Korean J Lab Med 2008;28:267-73.
    Pubmed CrossRef
  14. Previtali G, Ravasio R, Seghezzi M, Buoro S, Alessio MG. Performance evaluation of the new fully automated urine particle analyser UF-5000 compared to the reference method of the Fuchs-Rosenthal chamber. Clin Chim Acta 2017;472:123-30.
    Pubmed CrossRef
  15. Yim J, Lee SG, Cho S, Won YC, Kim JH. Performance evaluation of the CLINITEK Novus Automated Urine Chemistry Analyzer. Lab Med Online 2016;6:147-51.
    CrossRef