Received: 30 March 2022 | Revised: 7 December 2022 | Accepted: 13 January 2023 DOI: 10.1111/vcp.13241 ORIGINAL ARTICLE Performance of the Sysmex XN-­V hematology analyzer in determining the immature platelet fraction in dogs: A preliminary study and reference values Oriol Jornet-­Rius1 | Montse Mesalles-­Naranjo1 | Josep Pastor1,2 1 Departament de Medicina i Cirurgia Animal, Facultat de Veterinària, Abstract Universitat Autònoma de Barcelona, Background: Immature platelets (IPs) are newly formed platelets released into circu- Bellaterra, Spain 2 Hospital Clínic Veterinari, Universitat lation that have been demonstrated as good markers of thrombopoiesis. Although Autònoma de Barcelona, Bellaterra, Spain many flow cytometric and fully automated-­based methods are available, the latest Correspondence Sysmex XN-­V hematology analyzer for veterinary use is equipped with a specific fluo- Josep Pastor, Departament de Medicina rescent platelet channel (PLT-­F) that detects platelets using a platelet-­specific dye. i Cirurgia Animal, Facultat de Veterinària, Universitat Autònoma de Barcelona, Objectives: The aims of this study were to evaluate the performance of the Sysmex Bellaterra, 08193 Barcelona, Spain. XSN-­1000 V in determining the IPF (immature platelet fraction) and other selected Email: josep.pastor@uab.es PLT-­F channel parameters and to propose IPF reference intervals (RIs) for canine blood samples. Methods: Canine EDTA blood samples were analyzed on the Sysmex XN-­1000 V to assess linearity, imprecision, carryover, stability, and the effect of platelet clumping on selected platelet parameters from the PLT-­F channel. We also reported the de novo generated RIs for the IPF in dogs. Results: Imprecision was acceptable (CV <10%) for all parameters except for the absolute IPF values (IPF#), in which the reproducibility was 12.15% for the normal-­ low concentration samples. Linearity and carryover were excellent for all variables. Relative IPF values (IPF %) and IPF# remained stable for both storage conditions for up to 48 hours; however, a nonsignificant progressive increase in these parameters was observed from 12 hours at 4°C. We observed a statistical increase in IPF% and IPF# and a statistically significant decrease in PLT-­F counts after intentional in vitro platelet aggregation. RIs were generated for all reference samples (n = 69) and for samples with (n = 25) or without (n = 44) platelet clumps. Conclusions: The performance of the new PLT-­F channel-­derived variables for dogs was excellent. Specific RIs for IPF should be used when platelet aggregates are present. KEYWORDS canine, immature platelets, method validation, reticulated platelets This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2023 The Authors. Veterinary Clinical Pathology published by Wiley Periodicals LLC on behalf of American Society for Veterinary Clinical Pathology. Vet Clin Pathol. 2023;52:433–442.  wileyonlinelibrary.com/journal/vcp | 433 | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 434 JORNET-­RIUS et al. 1 | I NTRO D U C TI O N Two cohorts of dogs were recruited, healthy dogs (Group 1) and dogs with thrombocytopenia and increased thrombopoiesis (Group Immature platelets (IPs) or reticulated platelets are young platelets 2). To be included in the study, dogs needed a complete medical recently released into circulation. IPs were first reported in 1969 history and physical examination and a minimum hematologic and using light microscopy and stained with methylene blue during a biochemical assessment. Group 1 dogs were admitted for wellness 1 study of acute blood loss in dogs. Therefore, they were consid- care or preanesthetic examinations for elective surgical procedures. ered analogous to reticulocytes, and they were termed “reticulated Inclusion criteria included an unremarkable physical examination platelets.”1 IPs are larger and more reactive than mature platelets, 2,3 and the absence of abnormalities in the complete blood count and and contain larger amounts of RNA, mainly used for protein syn- basic biochemistry panel, the latter including a minimum of alanine 4 thesis. IPs circulate as the immature form for 24-­36 hours during aminotransferase activity, serum creatinine, total proteins, and glu- which their RNA is progressively degraded, and their volumes are cose. Group 2 dogs were defined by platelet counts (PLT-­F ) less than decreased.5,6 IP relative counts have been demonstrated as a good 171 × 109/L and increased relative IPF (IPF % > 10.4%) based on our marker of megakaryopoiesis. Several studies have been conducted results.14,15 Exclusion criteria included hemolytic, lipemic, and icteric in human and veterinary medicine to support this finding.7–­10 samples. All blood samples were obtained with the dog owners' per- Moreover, in human medicine, IP levels have been demonstrated to mission, and an informed consent document was obtained for each be altered in several diseases resulting in numerous potential clinical patient to use the leftover blood for possible studies. applications, such as risk assessment in patients with cardiovascular Samples were kept at room temperature and placed on an ag- disease, response to antiplatelet drugs, predicting sepsis in critically itator for 10 minutes and then were inverted 10 times manually to ill patients, and recovery after marrow/stem cell transplantation or ensure proper homogenization. All samples were analyzed within chemotherapy, among others.11 2-­12 hours after collection using the Sysmex XN-­1000 V for veter- Detection of IPs in a laboratory setting can be achieved through inary use (Sysmex Corporation, Norderstedt, Germany). Prior to the (a) flow cytometric and (b) fully automated methods. Most flow cyto- hematologic analysis, an air-­dried blood smear was prepared and metric methods detect IPs using the staining and detection technique stained with an automated Wright Stainer (Hematek slide Stainer; with thiazole orange (TO).7,10 However, recently, alternative flow cy- Siemens, Erlangen, Germany). For all samples, a thorough examina- tometry protocols have been proposed.12,13 Over time, methods for tion of a blood smear was performed, and the presence of platelet detecting IPs on fully automated hematology analyzers have contin- aggregation was noted. Platelet histogram distribution was studied. uously evolved. The latest Sysmex XN-­1000 V hematology analyzer When a Qflag, defined as the probability of having platelet clumps for veterinary applications uses a novel optic-­fluorescent analysis of in a sample, was displayed, a value <100 indicated that platelet ag- platelets (PLT-­F channel). The stained platelets are passed through gregates were unlikely to be present, and a value ≥100 indicated a semiconductor diode laser beam. Forward scatter is recorded as that platelet aggregates were most likely present. As per the man- platelet volume, and fluorescence intensity as platelet RNA content. ufacturer's recommendations, the presence or absence of platelet From these measurements, the platelet count and immature platelet clumps was verified during the blood smear examination. The above-­ fraction (IPF) are reported. mentioned procedures were performed by the same clinical pathol- Although numerous method validation studies for the new PLT-­F ogy resident under board-­certified supervision. If conflicting results channel using the Sysmex XN-­1000 V in human medicine have been existed, they were reviewed by both a clinical pathology resident and carried out and reviewed elsewhere,11 these are lacking in veterinary boarded certified clinical pathologist. Internal quality control (QC) medicine. The aims of this article were (a) to study the performance was performed daily using three levels of commercially available QC (linearity, repeatability, reproducibility, carryover, interference material (Sysmex XN Check Level 1 or low range, Level 2 or normal studies, and stability) of the PLT-­F channel and IPF on canine blood range, and Level 3 or abnormal high range; Sysmex Corporation). The samples using the Sysmex XN-­1000 V hematology analyzer for vet- following platelet parameters were analyzed: fluorescent platelet erinary use, (b) to assess the effect of platelet aggregation on IPF count (PLT-­F ), IPF (relative [%] and absolute [#] IPF values, respec- values, and (c) to propose a reference interval (RI) for canine samples. tively), mean platelet volume (MPV), and PLT clump. Although other variables related to platelets, erythrocytes, and leukocytes were also analyzed, they were not included in this study. 2 | M ATE R I A L S A N D M E TH O DS 2.1 | Study specimens 2.2 | Repeatability and reproducibility Blood samples from dogs presented to the Fundación Hospital The repeatability coefficient of variation was studied by analyzing Clínic Veterinari of the Universitat Autónoma de Barcelona between two selected canine group samples five consecutive times (a control December 2020 and December 2021 were collected from the ce- animal from Group 1) with PLT-­F counts within reference ranges and phalic or jugular veins and placed into 0.5-­1 mL ethylenediaminetet- with low-­normal IPF (%) and IPF# values, and a thrombocytopenic raacetic acid (K3-­EDTA) (Aquisel, Abrera, Barcelona, Spain) tubes. animal from Group 2. Normal platelet counts were established using | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License JORNET-­RIUS et al. 435 previously published and validated reference ranges for the Sysmex-­ Lipemia was simulated by adding a fat emulsion (Lipofundina 20%; 14,15 XT-­2100iV and IPF levels based on our results (IPF >10.4%). Braun Medical S.A., Barcelona, Spain) to aliquots from healthy ca- Due to platelet instability, reproducibility was evaluated from five nine blood. Final triglyceride concentrations in the samples were replicates in two selected canine specimens, a control patient and 20, 10, 5, 2.5, 1.25, and 0 g/L, corresponding to extreme, marked, a patient with increased thrombopoiesis, analyzed over a 12-­hour moderate, and slight lipemia. period. 2.7 | Effect of platelet aggregation 2.3 | Linearity The effect of platelet aggregation on platelet values, including PLT-­F Linearity was performed using triplicates of a 6-­point serial dilution count, IPF (%), IPF#, and MPV, was evaluated by two different ap- (100%, 50%, 25%, 12.5%, 6.25%, and 0%) of a fresh blood specimen proaches: (a) an in vitro platelet aggregation study, where 5 μL of from a thrombocytopenic dog in Group 2. As a diluent, the Sysmex CaCl2 was added to 250 μL of previously analyzed EDTA blood spec- diluent solution, Cellpack-­DCL (Sysmex Corporation) was used. imens from 21 healthy canine blood samples with no evidence of Results below the lower limit of detection were considered as 0. platelet clumping on either the blood smear or hematology analyzer Linearity was only tested on parameters expressed as counts using evaluation. Blood samples were analyzed 1 minute after CaCl2 was linear regression analysis (correlation of Pearson, r2). added, and both initial and post in vitro aggregation results were compared, and (b) selected platelets parameters from EDTA blood specimens from 69 healthy dogs used for generating “de novo” RIs 2.4 | Carryover were subclassified and compared according to the probability of PLT clumps determined using the Sysmex XN-­1000 V (QFlag). Twenty-­ Carryover was assessed by measuring a healthy canine blood sam- five dogs likely had platelet aggregates, and 44 likely had no platelet ple in triplicate followed by Sysmex diluent solution, Cellpack-­DCL aggregates. (Sysmex Corporation), in triplicate. 2.8 | Reference intervals 2.5 | Stability IPF RIs were calculated from 69 canine blood specimens.17 Selection EDTA blood samples from five randomly selected healthy individuals of healthy subjects and sample handling and evaluation were per- were used to evaluate the effect of storage conditions at room tem- formed as previously described for this study. perature (24°C) and stored in cold conditions (4°C). Measurements were performed in duplicate immediately, 12, 24, and 48 hours after collection. Before each run, the samples were placed on a blood 2.9 | Statistical analysis mixer at room temperature for 10-­20 minutes. The linearity study was evaluated using the Pearson correlation coefficient. For the stability study, a repeated measurement non- 2.6 | Interference study parametric ANOVA (multiple effect analysis) with Dunnett's multiple comparisons test was used. The Shapiro-­Wilk test was used for all The effect of hemolysis and lipemia effect on platelet parameters, samples to study normal data distributions. If parameters followed including PLT-­F, IPF (%), and IPF#, was studied using previously a normal distribution, a paired t-­test was used, and for nonnormally published protocols.16 A healthy canine blood sample with no evi- distributed data, the Wilcoxon matched-­pairs rank test was used to dence of plasma hemolysis or lipemia was used for each interfer- compare groups in the platelet aggregation study. One-­way ANOVA ence study. A 6-­p oint dilution was used in each study, and sample with Dunnett's multiple comparison tests were used to evaluate the aliquots were analyzed in duplicates using the Sysmex XN-­1000 V. effect of hemolysis and lipemia on IPF values. ASVCP guidelines for To investigate the effects of hemolysis, a fresh hemolysate was the Novo RI determination were followed for IPF RI in canine blood prepared by mixing distilled water with packed and washed ca- samples.17 Platelet parameters in the general population, dogs with nine erythrocytes from a healthy dog. When a final hemoglobin platelet aggregates, and IPF (%) and IPF# values in dogs without concentration of 100 g/L was obtained, the hemolysate was cen- platelet aggregates followed a nonnormal distribution; therefore, a trifuged and filtered using a syringe filter equipped with a 0.2 μm robust method was used. However, PLT-­F counts in dogs without cellulose acetate filter (Sartorius AG, Göttingen, Germany) to platelet aggregates followed a normal distribution, and a parametric eliminate cellular debris. Final hemoglobin concentrations in the method was performed. The statistical program Graph Pad v 8.0.1 samples were 10, 5, 2.5, 1.25, 0.62, and 0 g/L, corresponding to (Sant Diego, CA, USA) was used for statistical comparisons. P < 0.05 extreme, marked, moderate, and slight hemolysis, respectively. were considered statistically significant. | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 436 JORNET-­RIUS et al. 3 | R E S U LT S 3.3 | Carryover 3.1 | Repeatability and reproducibility All obtained PLT-­F channel values for diluent-­only samples were 0 at all times; therefore, no carryover was observed for any of the evalu- All coefficients of variation (CV%) were <13% for all parameters ated variables. (Table 1). CVs were lower than 5% for all parameters at both con- centration levels (normal-­low and high) during the repeatability study. Furthermore, CVs for platelet parameters at high IPF levels 3.4 | Stability (Group 2 dog) were also lower than 5% during the reproducibility study. However, during the reproducibility study, all CVs were be- The effect of blood storage on PLT-­F parameters in samples stored in tween 6% and 13% at normal-­low IPF levels (Group 1), presumably K3-­EDTA at 4 and 24°C is expressed as the mean and standard devi- due to different manipulation and testing conditions and platelet ation of the animals studied (Figure 2). PLT-­F count shows a progres- stability issues. sive decrease over time. At both storage temperatures, statistically significant differences were seen at 24-­and 48-­hours postcollection compared with initial concentrations (24°C: P = 0.0006 at 24 hours; 3.2 | Linearity P = 0.0003 at 48 hours; 4°C: P = 0.0054 at 24 hours; and P = 0.002 at 48 hours) (Figure 2). By contrast, no significant differences for The linearity of the analyzer for the assessed platelet parameters IPF (%) and IPF# were observed at both temperatures during the revealed excellent linearity with correlation coefficients higher than stability study. However, in refrigerated samples, a slight steady 0.99 for all variables (Figure 1; Table 2). increase and great variability were observed on these parameters TA B L E 1 Repeatability and reproducibility results of PLT-­F, IPF (%), and IPF# variables with the Sysmex XN-­V hematology analyzer using canine blood specimens classified based on relative IPF values. Repeatability Reproducibility Normal-­low IPF High IPF Normal-­low IPF High IPF Parameters Units Mean CV % Mean CV % Mean CV % Mean CV % IPF(%) % 3.06 4.96 31.32 2.91 0.97 8.45 32.06 0.47 PLT-­F 109/L 210.60 3.88 50.80 1.65 369.83 6.86 51.20 3.21 IPF# 109/L 6.46 2.08 15.92 3.59 3.58 12.15 16.44 3.06 Abbreviations: CV %, coefficient of variation; IPF (%), relative immature platelet fraction; IPF#, absolute immature platelet fraction; PLT-­F, fluorescent platelet count. F I G U R E 1 Visual linearity results for PLT-­F and IPF# of a 6-­point serial dilution in a dog from Group 2. IPF#, absolute immature platelet fraction; PLT-­F, fluorescent platelet count. | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License JORNET-­RIUS et al. 437 from 12 hours (Figure 2). Furthermore, at 24°C, IPF (%) and IPF# val- 5 and 10 g/L hemoglobin concentrations. On the contrary, IPF (%) ues remained stable during the 48-­hour duration, although a slight and IPF# values decreased progressively, reaching minimum values decrease in IPF# was observed from 24 hours. at 10 g/L hemoglobin concentration (Figure 3). Increased triglyceride concentrations did not cause significant changes in PLT-­F values, although we observed a slight trend in de- 3.5 | Interference study creasing PLT-­F values at higher lipid concentrations (5 g/L). However, IPF (%) and IPF# values progressively increased with increasing con- Hemolysis interfered significantly with all platelet parameters. PLT-­F centrations and reached significant differences at higher lipid con- counts showed a constant increase, reaching maximum values at centrations (5 g/L) (Figure 3). higher hemoglobin concentrations with significant interferences at TA B L E 2 Linearity results for PLT-­F and IPF# variables in a dog from Group 2. 3.6 | Effect of platelet aggregation Parameters Range R After the addition of CaCL 2, an increase in platelet aggregates was 9 PLT-­F 10 /L 4.67-­50.33 0.999 observed during the blood film examination, and the Sysmex XN-­ IPF# 109/L 1.00-­15.57 0.999 1000 V analyzer was able to report a flag regarding platelet clumps in all except two of the animals (91%). Furthermore, platelet counts Abbreviations: IPF#, absolute immature platelet fraction; PLT-­F, fluorescent platelet count. decreased, and IPF (%) and IPF# increased considerably after in vitro F I G U R E 2 Changes in PLT-­F, IPF (%), and IPF# parameters after storage at room temperature (24°C) and 4°C in canine blood samples with the Sysmex XN-­1000 V analyzer. Data are shown as mean and standard deviation. Statistical significance is indicated by an asterisk (*P < 0.05). IPF (%), relative immature platelet fraction; IPF#, absolute immature platelet fraction; PLT-­F, fluorescent platelet count. | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 438 JORNET-­RIUS et al. F I G U R E 3 Effects of hemolysis and triglycerides on canine PLT-­F count, IPF (%), and IPF# parameters with the Sysmex XN-­1000 V analyzer. x-­axes show increasing concentrations of hemoglobin and triglycerides; y-­axes show the mean percentage change of each parameter expressed as the final value/initial value × 100 (final value = enriched value; initial value = value without interferent). Each data point represents the mean of duplicate determination. *P < 0.05. IPF (%), relative immature platelet fraction; IPF#, absolute immature platelet fraction; PLT-­F, fluorescent platelet count. platelet aggregation. Statistically significant differences were found median age was 6 years (range from 6 months to 16 years); intact fe- in all variables studied, and none were observed for MPV (Figure 4). males represented 65.9%, whereas intact males represented 34.1% When comparing animals according to the probability of PLT of the population. Dogs in the subgroup “with platelet clumps” had clumps, as determined by the Sysmex XN-­1000 V (QFlag), animals a median age of 4 years (range from 1 to 14 years), and 64% and 36% with platelet clumps had lower PLT-­F counts and higher IPF (%) and were represented by intact females and males, respectively. Table 3 IPF# than animals without evidence of platelet aggregate, as shown shows the obtained RIs. RIs for the general population, indepen- in Table 3, matching the in vitro study. dent of the platelet aggregation, had wider ranges for all variables (PLT-­F [109/L]: 171.0-­491.5; IPF (%): 1.24-­10.42; and IPF# [109/L]: 1.68-­25.56), and dogs without platelet clumps had the narrowest 3.7 | Reference intervals ranges (PLT-­F [109/L]: 176.1-­480.3; IPF (%): 1.04-­6.85; and IPF# [109/L]: 1.31-­20.59). Furthermore, dogs with platelet clumps have Sixty-­nine healthy dogs meeting the RI inclusion criteria were com- higher ranges for IPF (%) and IPF# and lower ranges for PLT-­F counts prised of 53 purebred (76.8%) and 16 crossbred dogs (23.2%). The (PLT-­F [109/L]: 151.5-­466.63; IPF (%): 2.33-­10.44; and IPF# [109/L]: most representative breeds were Labrador Retriever (n = 6), German 4.47-­25.56). Shepherd dog (n = 5), and Maltese dog, Golden Retriever, and Pinscher with n = 3 each. Age ranged from 6 months to 16 years, with a median age of 5 years. Females represented 65.2% of the popula- 4 | DISCUSSION tion, while males accounted for 34.8% of all dogs. All dogs included in this study were intact. The IPF parameter is an important part of the hematology diagnostic Given the influence of the presence of platelet aggregates on profile in human medicine due to its utility for diagnosis, prognosti- platelet variables, RIs were calculated for all healthy dogs (n = 69) cation, follow-­up, and clinical decision-­making of many pathologic independent of platelet aggregation and for the subgroups of “dogs conditions.11 Its applicability in veterinary medicine remains largely with platelet clumps” (n = 25) and “dogs without platelet clumps” unknown, but there is some evidence that IPF could be a marker for (n = 44). In the subgroup of “dogs without platelet clumps,” the the thrombopoietic rate in canine samples.7,18,19 This study evaluates | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License JORNET-­RIUS et al. 439 F I G U R E 4 Differences in PLT-­F, IPF (%), IPF#, MPV variables, and QFlag after inducing in vitro platelet clumping with the addition of CaCL 2 in canine blood samples with the Sysmex XN-­1000 V analyzer. Data are shown as mean and standard deviation. Statistical significance is indicated by an asterisk (*P < 0.05; **P < 0.01). IPF (%), relative immature platelet fraction; IPF#, absolute immature platelet fraction; MPV, mean platelet volume; PLT-­F, fluorescent platelet count. the performance of the fluorescent platelet channel (PLT-­F ) of the they used quality control material to calculate both repeatability and Sysmex XN-­1000 V hematology analyzer. Imprecision was found to reproducibility CV for PLT parameters.15 On the other hand, another be acceptable for the selected platelet parameters (PLT-­F, IPF [%], study using a Sysmex XT-­2000iV analyzer and the PLT-­O channel and IPF#) for canine samples when compared with the human litera- to calculate IP showed higher CV on patient samples.17,22 The per- 11,20,21 ture. The CV was less than 10% for all parameters except for formance of the Sysmex XN-­1000 V is expected to be high because IPF# during the reproducibility study for normal-­low samples (Group of the exclusive channel for the detection of platelets (PLT-­F chan- 1). This result is anticipated because variability tends to be higher at nel). This channel uses reagents that specifically bind platelet mi- lower concentrations, especially when using patient samples. In con- tochondrial and ribosomal RNA, leading to a more intense labeling trast to our study, a previous study using the same hematology ana- signal and, therefore, better identification of platelets than with the lyzer showed a lower CV during the reproducibility study; however, reagents used in the RET/PLT-­O channel on the Sysmex analyzer | 440 TA B L E 3 Reference intervals for PLT-­F, IPF (%), and IPF# in healthy dogs using the Sysmex XN-­1000 V analyzer. Platelet parameter RIs are determined for all samples and for samples with and without platelet clumps. All samples (n = 69) Samples with platelet aggregates (n = 25) Samples without platelet aggregates (n = 44) Percentile Percentile Percentile Percentile Percentile Percentile Variable Units Mean SD Median (min-­max) 2.5% 97.5% Mean SD Median (min-­max) 2.5% 97.5 Mean SD Median (min-­max) 2.5% 97.5% P-­value PLT-­F 109/L 302.68 199.40 343.00 (156-­596) 171.00 491.50 274.85 32.53 179.00 (156-­596) 151.50 466.63 318.76 64.35 438.50 (176-­514) 176.10 480.30 0,0334* IPF (%) % 3.77 1.91 5.45 (0.3-­11.6) 1.24 10.42 4.83 1.77 5.55 (2.2-­10.4) 2.33 10.44 3.19 1.84 2.80 (0.3-­11.6) 1.04 6.85 0,0016* IPF# 109/L 10.25 4.31 16.75 (1.2-­29,2) 1.68 25.56 12.19 4.95 10.20 (5.9-­29.2) 4.47 25.56 9.11 9.83 12.85 (1.2-­24.2) 1.31 20.59 0,0065* Abbreviations: IPF (%), relative immature platelet fraction; IPF#, absolute immature platelet fraction; max, maximum; min, minimum; PLT-­F, fluorescent platelet count; SD, standard deviation. *P < 0.05. JORNET-­RIUS et al. 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License JORNET-­RIUS et al. 441 XT-­2000iV. 23 The imprecision results obtained in our study agree aggregates compared with individual platelets since MPV was not af- 20,24,25 with what has been previously published in human medicine. fected after the in vitro platelet aggregation study. To the author's The linearity study was performed using blood from dogs with knowledge, this report is the first in veterinary medicine to demon- thrombocytopenia and increased IPF (Group 2) to increase the range strate a clear effect of platelet aggregates on IPF (%) and IPF# values. of values across multiple dilutions. Linearity and carryover were ex- In our study, RIs for platelet parameters consider the presence of cellent for all platelet variables studied and similar to those reported platelet clumps in the samples, especially for IPF. This strategy is not in a recent study on PLT-­F counts.15 Also, our results agree with what used in human or veterinary medicine, and further studies concern- has been previously published in the human literature. 20,21 ing the clinical utility of this approach are warranted. Also, the IPF%, PLT-­F counts were stable for up to 24 hours at both 24 and 4°C; which predicts adequate bone marrow responses to platelet con- however, a progressive decline in its values was observed right after sumption or destruction, is not defined in veterinary medicine, and extraction. By contrast, IPF (%) and IPF# remained stable during the determination of such a decision limit will need to consider the the stability study for both storage conditions for up to 48 hours. presence of platelet aggregates. Moreover, a study evaluating the However, due to the variability observed and a progressive increase effect of platelet clumping on platelet counts when measured using in these parameters from 12 hours at 4°C, these parameters would impedance and buffy coat methods observed that the presence and lack clinical utility for samples stored at 4°C for 24 hours. A recent degree of platelet aggregation were variable and not homogeneous study evaluating the performance of the Sysmex XN-­1000 V for within blood samples. 29 Thus, there is no consensus in the literature canine blood samples revealed similar stability results for PLT-­F on how to evaluate the effect of platelet clumping on platelet counts. 15 counts. This progressive increase in IPF values has been associated Consequently, RIs for dogs with platelet clumps reported in this with the nonspecific binding of the dye to RNA released from disin- study should be interpreted with caution and may not have clinical tegrated platelets and to other membrane systems. 26,27 Moreover, significance. Independent of the presence of platelet clumps, the RIs cold-­induced platelet aggregation could play an important role in for IPF (%) in our study are not in concordance with those previously 28 this finding, as previously demonstrated. published using either the Sysmex XT-­2000iV or flow cytometry To the authors' knowledge, the effect of hemolysis and lipemia with the TO method.7–­9,17,22 Sysmex XN-­V detects a higher percent- on PLT-­F channel parameters in canine blood samples using the age of IPF compared with those previous studies; a possible expla- Sysmex XN-­1000 V has not been previously reported. Hemolytic nation for the discrepancies observed among RIs for IPF (%) using samples impaired the detection of both mature and IPs at higher different techniques could be caused due to variations of staining hemoglobin concentrations. PLT-­F counts increased, and IPF (%) protocols and gating strategies. Discrepancies between studies that and IPF# decreased with increasing hemoglobin concentrations. compared methodologies have been found. While a study performed Although the fresh hemolysate used during this study was filtered in veterinary medicine found that the results of IPs obtained using to avoid the interference of red blood cell fragments, the remaining the Sysmex XT2000iV were comparable to those obtained with flow fragments could have been counted as mature platelets and, there- cytometry using the TO methodology, only a modest correlation fore, increasing PLT-­F counts and decreasing IPF (%) and IPF# values. was found between IPs obtained with flow cytometry using TO and By contrast, a slight decrease in PLT-­F counts and an increase in IPF IPF obtained with the Sysmex XE-­2100/5000 in the human litera- (%) and IPF# values with increasing triglyceride concentrations were ture. 24,30 Possible explanations for the differences between results seen, but only significant differences were observed with extreme obtained with the different methods available may be related to the lipemia. stains, with different affinities to bind to different cellular compo- We previously noticed a proportion of dogs with pseudothrombo- nents between dyes and different gating strategies used. 23,24 cytopenia due to platelet aggregation had moderately high IPF values. Limitations of our study include the absence of a comparison These dogs did not have diagnoses or clinical signs that would cor- study with other methodologies and the lack of assessment of in- relate with platelet consumption or destruction, such as petechiae, terference by icteric plasma on platelet variables derived from the ecchymoses, mucosal bleeding, or other bleeding sources. Thus, we PLT-­F channel. Further studies will be needed to evaluate the effect hypothesized that IPF values would be affected by platelet aggre- of icterus interferences and whether the determination of a decision gation. After promoting platelet aggregation in vitro, we observed a limit would work better for estimating the rate of thrombopoiesis. statistical increase in IPF (%) and IPF# values and a decrease in PLT-­F In conclusion, the new PLT-­F channel can be confidently used counts. A single report in human medicine evaluating the effect of in canine blood samples for detecting IPs in samples up to 24 hours platelet aggregation on IPF values found that after platelet aggre- postextraction samples at both storage conditions (24 and 4°C). gation during cold storage, IPF (%) increased significantly in a time-­ Furthermore, samples with platelet clumps and hemolytic and li- dependent manner and resolved after the addition of an excess of pemic samples should be interpreted with caution. EDTA. Moreover, they found a positive correlation between IPF and MPV, suggesting that these changes were related to a false appear- D I S C LO S U R E S TAT E M E N T ance of larger platelet sizes in the samples with platelet aggregates.28 The authors have stated that they have no affiliation to, financial However, our study suggests that these erroneous high IPF values interest in, or financial competition with the subject matter or mate- could also be related to a higher fluorescence intensity by platelet rials discussed in this article. | 1939165x, 2023, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vcp.13241 by Readcube (Labtiva Inc.), Wiley Online Library on [03/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 442 JORNET-­RIUS et al. ORCID in dogs for the Sysmex XT-­2000iV. BMC Vet Res. 2016;12(1):148. Oriol Jornet-­Rius https://orcid.org/0000-0001-7495-3961 doi:10.1186/s12917-­016-­0779-­4 18. Wilkerson MJ, Shuman W, Swist S, Harkin K, Meinkoth J, Kocan Josep Pastor https://orcid.org/0000-0003-1702-9531 AA. Platelet size, platelet surface-­association IgG, and reticulated platelets in dogs with immune-­mediated thrombocytopenia. Vet REFERENCES Clin Pathol. 2001;30(3):141-­149. 1. Ingram M, Coopersmith A. Reticulated platelets following acute 19. Zmigrodzka M, Guzera M, Winnicka A. Evaluation of reticulated blood loss. Br J Haematol. 1969;17(3):225-­229. platelets in dogs with breed-­related thrombocytopenia. Pol J Vet 2. Allen EC, Tarigo JL, LeVine DN, Barber JP, Brainard BM. Platelet Sci. 2014;17(1):137-­142. number and function in response to a single intravenous dose of 20. Briggs C, Longair I, Kumar P, Singh D, Machin SJ. Performance vincristine. J Vet Intern Med. 2021;35(4):1754-­1762. evaluation of the Sysmex haematology XN modular system. J Clin 3. Mcbane RD, Gonzalez C, Hodge DO, Wysokinski WE. Propensity Pathol. 2012;65(11):1024-­1030. for young reticulated platelet recruitment into arterial thrombi. J 21. Seo JY, Lee ST, Kim SH. Performance evaluation of the new Thromb Thrombolysis. 2014;37(2):148-­154. hematology analyzer Sysmex XN-­series. Int J Lab Hematol. 4. Harrison P, Goodall AH. “Message in the platelet”—­more than just 2015;37(2):155-­164. vestigial mRNA! Platelets. 2008;19(6):395-­4 04. 22. Pankraz A, Bauer N, Moritz A. Comparison of flow cytometry with 5. Dale GL, Friese P, Hynes LA, Burstein SA. Demonstration that the Sysmex XT2000iV automated analyzer for the detection of re- thiazole-­orange-­positive platelets in the dog are less than 24 hours ticulated platelets in dogs. Vet Clin Pathol. 2009;38(1):30-­38. old. Blood. 1995;85(7):1822-­1825. 23. Wada A, Takagi Y, Kono M, Morikawa T. Accuracy of a new platelet 6. Ault KA, Knowles C. In vivo biotinylation demonstrates that re- count system (PLT-­F ) depends on the staining property of its re- ticulated platelets are the youngest platelets in circulation. Exp agents. PLoS One. 2015;10(10):1-­13. Hematol. 1995;23(9):996-­1001. 24. Sakuragi M, Hayashi S, Maruyama M, et al. Clinical significance 7. Weiss DJ, Townsend E. Evaluation of reticulated platelets in dogs. of IPF% or RP% measurement in distinguishing primary immune Comp Clin Path. 1998;8(3):166-­170. thrombocytopenia from aplastic thrombocytopenic disorders. Int 8. Hanahachi A, Moritomo T, Kano R, Watari T, Tsujimoto H, Hasegawa J Hematol. 2015;101(4):369-­375. A. Thiazole orange-­positive platelets in healthy and thrombocyto- 25. Buoro S, Seghezzi M, Manenti B, et al. Biological variation of plate- penic dogs. Vet Rec. 2001;149(4):122-­123. let parameters determined by the Sysmex XN hematology analyzer. 9. Smith R, Thomas JS. Quantitation of reticulated platelets in healthy Clin Chim Acta. 2017;470(March):125-­132. dogs and in nonthrombocytopenic dogs with clinical disease. Vet 26. Albanyan A, Murphy MF, Wilcock M, Harrison P. Changes in the Clin Pathol. 2002;31(1):26-­32. immature platelet fraction within ageing platelet concentrates. J 10. Kienast J, Schmitz G. Flow cytometric analysis of thiazole orange Thromb Haemost. 2008;6(12):2213-­2215. uptake by platelets: a diagnostic aid in the evaluation of thrombo- 27. Osei-­Bimpong A. The effect of storage on the clinical utility of the cytopenic disorders. Blood. 1990;75(1):116-­121. immature platelet fraction. Hematology. 2009;14(2):118-­121. 11. Buttarello M, Mezzapelle G, Freguglia F, Plebani M. Reticulated 28. Miyazaki K, Koike Y, Kunishima S, et al. Immature platelet fraction platelets and immature platelet fraction: clinical applications and measurement is influenced by platelet size and is a useful param- method limitations. Int J Lab Hematol. 2020;42(4):363-­370. eter for discrimination of macrothrombocytopenia. Hematology. 12. Hille L, Cederqvist M, Hromek J, Stratz C, Trenk D, Nührenberg TG. 2015;20(10):587-­592. Evaluation of an alternative staining method using SYTO 13 to deter- 29. Koplitz SL, Scott MA, Cohn LA. Effects of platelet clumping on mine reticulated platelets. Thromb Haemost. 2019;119(5):779-­785. platelet concentrations measured by use of impedance or buffy 13. Hedley BD, Llewellyn-­Smith N, Lang S, et al. Combined accurate coat analysis in dogs. J Am Vet Med Assoc. 2001;219(11):1552-­1556. platelet enumeration and reticulated platelet determination by flow 30. Ibrahim H, Nadipalli S, Usmani S, DeLao T, Green LS, Kleiman NS. cytometry. Cytometry B Clin Cytom. 2015;88(5):330-­337. Detection and quantification of circulating immature platelets: 14. Bourgès-­Abella N, Geffré A, Concordet D, Braun JP, Trumel C. agreement between flow cytometric and automated detection. J Canine reference intervals for the Sysmex XT-­2000iV hematology Thromb Thrombolysis. 2016;42(1):77-­83. analyzer. Vet Clin Pathol. 2011;40(3):303-­315. 15. Grebert M, Granat F, Braun JP, Leroy Q, Bourgès-­Abella N, Trumel C. Validation of the Sysmex XN-­V hematology analyzer for canine specimens. Vet Clin Pathol. 2021;50(2):184-­197. How to cite this article: Jornet-­Rius O, Mesalles-­Naranjo M, 16. Martínez-­Subiela S, Cerón JJ. Effects of hemolysis, lipemia, hy- Pastor J. Performance of the Sysmex XN-­V hematology perbilirrubinemia, and anticoagulants in canine C-­reactive pro- analyzer in determining the immature platelet fraction in tein, serum amyloid a, and ceruloplasmin assays. Can Vet J. 2005;46(7):625-­629. dogs: A preliminary study and reference values. Vet Clin 17. Oellers DE, Bauer N, Ginder M, Johannes S, Pernecker I, Moritz Pathol. 2023;52:433-442. doi:10.1111/vcp.13241 A. Optimized gating and reference ranges of reticulated platelets