Journal of Perinatology (2011) 31, 258–262 r 2011 Nature America, Inc. All rights reserved. 0743-8346/11 www.nature.com/jp ORIGINAL ARTICLE Association of cord count and volume with hemoglobin in healthy term infants

M Eskola1, S Juutistenaho1, K Aranko1, S Sainio2 and R Kekoma¨ki1 1The Finnish Red Cross Blood Service, Helsinki, Finland and 2Department of Obstetrics and Gynecology, Helsinki University Central Hospital, Helsinki, Finland

a single type of hematology analyzer.11 Typically, the type of cell Objective: The aim of this study was to investigate relationships of cord analyzer and its principles of measurement are given in association blood cells in healthy term infants both from vaginal and Cesarean sections. with the results for cellular characteristics because the results are, Study Design: The study sample comprised 167 consecutive cord blood to some extent, depending on the principles and software used in 12–14 collections accepted for processing in an accredited . The the analysis. Some measurements are independent of each effect of varying anticoagulant-to-blood ratio was excluded by standardizing other, such as hemoglobin (Hb) concentration and cell counts the cell concentrations to reflect the values in native blood. Statistical analysis other than red blood cells (RBCs). Relationships between CB included descriptive statistics, simple linear regression analysis, Mann– cellular characteristics other than (WBC) Whitney U-test, cumulative frequency plots and Smirnov two-sample test. subpopulations have been rarely described. In mice, after the injection of recombinant megakaryocyte growth and development Result: As expected, hemoglobin correlated with factor, Hb decreased with rising platelet (PLT) count.15 concentration. Interestingly, mean platelet volume was associated with Recently, CB mean platelet volume (MPV) and PLT count have hemoglobin, red blood cell concentration and hematocrit. The platelet been reported to correlate with the concentrations of hematopoietic count was inversely associated with the parameters. progenitors.16 The objective of this study was to test whether an Conclusion: The observed associations of cord blood hemoglobin with association between Hb and PLT, as well as other cellular mean platelet volume and platelet count reflect the physiology of fetal characteristics, exists in CB. If such a physiological association hematopoiesis at term. existed in neonatal blood, this could facilitate the understanding of Journal of Perinatology (2011) 31, 258–262; doi:10.1038/jp.2010.95; the physiological variation even in healthy newborn infants. published online 5 August 2010

Keywords: mean platelet volume; platelet; fetal hematopoiesis Materials and methods CB collections All CB collections and measurements were performed in a Introduction Foundation for the Accreditation of Cellular Therapy accredited CB Cord blood (CB) has been shown to be an alternative source of bank (Finnish Cord Blood Bank; www.factwebsite.org). The study hematopoietic stem cells for transplantation.1–4 A number of CB sample included 167 consecutive CB collections accepted for banks collect and store placental blood from healthy infants after processing between September 2004 and August 2005 (Table 1). Any uncomplicated pregnancy and delivery.5,6 Data from CB banks abnormal clinical or laboratory data caused exclusion of the have offered an acceptable alternative for studying blood cell collection. Perinatal data were obtained from the hospital records as physiology, such as CD34 þ cells,7–9 because it can be demanding part of the CB banking program. As a part of the acceptance criteria, to ethically justify blood sampling from healthy newborn infants. hemoglobinopathies and other RBC diseases were excluded by The cross-sectional hematological values of CB from healthy separate examination of the samples at the Central Middlesex term infants have been well characterized.10 Recently, neonatal Hospital, North West London Hospitals (London, UK) in addition to reference ranges were defined in a large multi-hospital system with family history review. CB was collected ex utero by CB bank nurse- midwives immediately after uncomplicated term delivery with a Correspondence: M Eskola, Finnish Red Cross Blood Service, Kivihaantie 7, FI-00310 special collection bag (CB-Collect, double blood bag 300/20 ml for Helsinki, Finland. CB; T2950; Fresenius HemoCare, Emmer-Compascuum, the E-mail: mikko.eskola@veripalvelu.fi Received 11 March 2010; revised 2 June 2010; accepted 17 June 2010; published online 5 Netherlands). Collections were obtained 4 days a week in one daily August 2010 shift. Elective Cesarean sections were performed during office hours, Cord blood platelet characteristics and hemoglobin M Eskola et al 259

Table 1 Obstetric data

Characteristic N Maternal age (years) Length of gestation (days) Birth weight (g) Relative birth weight (s.d.)

Median Range Median Range Median Range Median Range

Study sample 167 33 20–46 277 259–296 3784 2490–4975 +0.48 À2.4 to +3.4 Vaginal deliveries 63 31a 20–46 281a 261–293 3865 2490–4880 +0.36 À2.4 to +2.8 Cesarean sections 104 34a 24–45 275a 259–296 3777 2780–4975 +0.52 À1.8 to +3.4 Females 80 33 20–46 275 259–296 3695b 2490–4970 +0.47 À2.4 to +3.4 Males 87 33 20–45 278 266–294 3904b 2780–4975 +0.52 À1.8 to +2.6 aVaginal deliveries vs Cesarean sections: maternal age, P<0.01; length of gestation, P<0.01. bFemales vs males: birth weight, P<0.01. and thus, the proportion of Cesarean sections in this study does not approved the procedures of the CB bank program. The provisions of show the general section frequency in the department. the Declaration of Helsinki Principles and international laws were The samples for cell enumeration were drawn from the followed. The voluntary mothers of the donor infants provided collections within 24 h of CB collection, with the exception of one written informed consent. sample drawn after 45 h. The collected CB volume was 92 ml (median; range 42 to 150) but higher after Cesarean section Statistical analysis (P<0.0001), as expected. StatsDirect software (StatsDirect, Cheshire, UK) was used for descriptive statistics and simple linear regression analyses. The Newborn infants analyses were repeated after the removal of outliers (values outside Infants with an Apgar score of at least 7 at 5 min of age were mean ±3 s.d.). The Mann–Whitney U-test was used for group included in the study. Relative birth weight was expressed as a comparisons. Cumulative frequency plots were used to analyze the standard deviation from the mean birth weight z-score, which distribution of cell counts or other variables between groups. The 17 adjusts the birth weight for gestational age and gender. A Finnish distribution differences between the two upper or lower quartiles 18 reference population was applied in this study, and 96% of the were analyzed by a Smirnov two-sample test as indicated in the infants were appropriate for gestational age. Because one of the results. Descriptive data are expressed as median (range) unless aims in CB banking is to collect CB from larger infants to obtain otherwise stated. A two-sided P<0.05 was considered significant. larger collection volumes (with associated higher nucleated cell and progenitor numbers),19 no attempt was made to exclude macrosomic infants (N ¼ 5; relative birth weight >2 s.d.)20 from Results the study sample. The hematological values and their distribution were within the Blood cell counts published ranges (Table 2). The correlation between CB Hb and A hematology analyzer (Sysmex K-1000; Sysmex, Kobe, Japan) was RBC concentration, two independently measured values, was tight used to obtain electric impedance-based automated cell as expected (r ¼ 0.96; P<0.0001; y ¼ 0.0278xÀ0.0697). measurements and spectrophotometric measurements of Hb by MPV was associated with Hb (r ¼ 0.25, P ¼ 0.0013) and the using the sodium lauryl sulfate Hb method. The WBC count RBC concentration, as well as hematocrit (Hct) (Figure 1). PLT included nucleated RBCs. The results of the instrument, including count was inversely associated with Hb, the RBC concentration and the coefficients of variation and uncertainties of measurement, Hct. MPV was inversely associated with mean corpuscular Hb were within acceptable limits in validation studies. All cell concentration (r ¼À0.20, P ¼ 0.0081). Furthermore, plateletcrit concentrations were standardized to exclude the effect of a varying (MPV  PLT count) was associated with Hb, the RBC anticoagulant-to-blood ratio.9 concentration and Hct (r ¼À0.23, À0.20 and À0.23, respectively; P<0.01 for all). MPV of the two upper quartiles was CD34 þ cell enumeration significantly lower in male infants (n ¼ 43; 9.0 fl, 8.7 to 10.5) Enumeration of CD34 þ cells in collected CB was achieved with a flow than in female infants (n ¼ 40; 9.3 fl, 8.8 to 10.5; P ¼ 0.0288, cytometer (FACSCalibur; Becton Dickinson, San Jose, CA, USA) with a two-sample Smirnov test after removal of one outlier in each gating strategy based on ISHAGE guidelines, as previously described.21 group). An association was also observed between WBC concentration Ethical considerations and Hb (r ¼ 0.22, P ¼ 0.0041), the RBC concentration (r ¼ 0.24, The ethical committee of the collection site (Department of P ¼ 0.002) and mean corpuscular Hb concentration (r ¼À0.21, Obstetrics and Gynecology, Helsinki University Central Hospital) P ¼ 0.0068). In female infants, the WBC concentration

Journal of Perinatology Cord blood platelet characteristics and hemoglobin M Eskola et al 260

Table 2 Hematological values for cord blood

Characteristic Study sample (N ¼ 167) Vaginal delivery (N ¼ 63) Cesarean section (N ¼ 104) P-valuea

Median Range Median Range Median Range

WBC ( Â 109 per l) 15.1 5.54–39.7 18.4 12.0–34.1 13.6 8.54–39.7 <0.0001 RBC ( Â 1012 per l) 4.7 3.46–6.62 4.78 3.89–6.30 4.62 3.46–6.62 NS Hb (g lÀ1) 174 130–234 176 140–230 171 130–234 NS Hct (%) 53.6 40.1–73.1 54.7 41.9–73.1 52.6 40.1–72.2 NS MCV (fl) 112 97.7–127 114 105–127 112 97.7–125 NS MCH (pg) 36.5 31.4–41 36.5 31.4–41 36.6 32–39.9 NS MCHC (g lÀ1) 324 303–359 323 308–359 324 303–344 NS RDW (%) 17.4 14.2–23.6 17.4 14.9–23.6 17.4 14.2–23.3 NS PLT ( Â 109 per l) 270 161–607 297 169–607 254 161–424 0.0053 MPV (fl) 8.7 7.5–11.5 8.7 7.7–11.4 8.8 7.5–11.5 NS Plateletcrit (%) 0.24 0.15–0.48 0.26 0.15–0.48 0.23 0.15–0.36 0.0056 CD34+ cells ( Â 106 per l) 43.9 7.14–253 47.7 15.9–253 39.9 7.14–120 0.007

Abbreviations: Hb, hemoglobin; Hct, hematocrit; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; MPV, mean platelet volume; NS, not significant; plateletcrit, MPV Â PLT; PLT, platelet; RBC, red blood cell; RDW, red blood cell distribution width; WBC, white blood cell. aP-values of the differences between vaginal delivery and Cesarean section. The concentrations were standardized to exclude the varying effect of the anticoagulant.

Figure 1 Associations between cord blood (CB) mean platelet volume (MPV) and platelet (PLT) count and hemoglobin (Hb), red blood cell (RBC) concentration and hematocrit (Hct) (N ¼ 167). (m) Outliers. The associations remained significant after removal of the outliers: (a) r ¼ 0.22, P ¼ 0.0043; (b) r ¼ 0.26, P ¼ 0.0008; (c) r ¼ 0.26, P ¼ 0.0008; (d) r ¼À0.31, P<0.0001; (e) r ¼À0.29, P ¼ 0.0001; (f) r ¼À0.32, P<0.0001.

(15.3  109 per l, 10.0 to 31.3) was associated with Hct MPV was similar after vaginal delivery and Cesarean section, (53.5%, 43.0 to 70.8; r ¼ 0.34, P ¼ 0.0018). However, the CD34 þ whereas PLT count was higher after vaginal delivery compared to cell concentration was not associated with Hb, the RBC section (P ¼ 0.0036) (Table 2). Notably, the CB PLT count was concentration or Hct, although there was an association with inversely associated with Hct independent of the mode of delivery PLT characteristics, as described earlier.16 (vaginal delivery r ¼À0.30, P ¼ 0.0154 and Cesarean section r ¼À0.35, P ¼ 0.0002). Mode of delivery The infants born by vaginal delivery had slightly higher Hb, CB MPV was associated with Hb after both vaginal delivery (r ¼ 0.26, RBC concentration and Hct than infants born by Cesarean section P ¼ 0.0389) and Cesarean section (r ¼ 0.22, P ¼ 0.026). (P ¼ 0.0119, 0.0031 and 0.0057, respectively; differences between

Journal of Perinatology Cord blood platelet characteristics and hemoglobin M Eskola et al 261

Figure 2 Cumulative frequencies of cord blood (CB) hemoglobin (Hb), red blood cell (RBC) concentration and hematocrit (Hct) after Cesarean section (N ¼ 104; ’) and vaginal delivery (N ¼ 63; &). P-values for the differences between the two lowest quartiles of Hb (P ¼ 0.0119), RBC (P ¼ 0.0031) and Hct (P ¼ 0.0057) were determined by the two-sided Smirnov test.

the two lower quartiles, Smirnov two-sample test; Figure 2). the analyses were performed to obtain cell counts to select Furthermore, the CD34 þ cell concentration was significantly high-quality collections for processing and thus, a relatively simple higher after vaginal delivery compared to Cesarean section hematology analyzer was applied. More advanced instrumentation (P ¼ 0.007), as expected. may yield additional information on the observed relationships between, for example, nucleated RBCs and other cell counts.23 Fourth, the authors are aware of the fact that not all of the Discussion measurements were performed with independent methods. CB banks aim at as high a nucleated cell count as possible in the Although some bias could have been introduced on the basis CB units processed and stored for hematopoietic of cell counting, the measurements of Hb and MPV or PLT count transplantation. The data presented in this study originate from were by independent methods and, therefore their associations a CB bank where the CB collections were performed ex utero remain valid. immediately after delivery in a specially equipped laboratory within In a population-wide study from the same hospital as this study, the obstetric department.5 Cord clamping and other perinatal PLT counts in CB from healthy term infants were similar to those practices were performed according to hospital guidelines. in this study.24 Also, this study confirmed that the CB PLT count is All cell concentrations were standardized to exclude the varying higher after vaginal delivery than after Cesarean section.25 anticoagulant-to-blood ratio inherent to CB banking.9 Cell Recently, PLT count and MPV reference ranges for the first 3 days counting was performed in the quality control unit of the national of life were reported for neonates in a large, multi-hospital study.26 blood service with the exception of the CD34 þ cell counts The association of MPV and PLT count with Hb and RBC analyzed in the CB bank. concentration of CB collections from healthy term infants is The neonatal (placental) blood cell counts are speculated to interesting. Animal studies have shown that the hematopoietic reflect the general hematopoietic activity. In an earlier study activity is strictly regulated genetically.27 Ulich et al.15 reported concerning the same samples from healthy term infants as this hypermegakaryocytosis and erythropenia in adult study,16 CB MPV and PLT count were associated with markers of and thrombocytosis with concurrent low Hb after stimulation with hematopoietic progenitor potential (CD34 þ cells and colony- recombinant megakaryocyte growth and development factor. forming cells). Further significant associations included those In conclusion, the associations between CB MPV and PLT count between MPV or PLT count and Hb in this study. and Hb observed in this study are underscored or novel, albeit Several limitations of this study may cause concerns. First, the predictable by earlier experimental evidence. Notably, the number of analyzed CB collections is relatively small. Nevertheless, associations were observed in a population of healthy term infants. the results await confirmation from larger CB banks or maternity Recognition of the associations may improve the understanding units. Second, the CB was collected in citrate anticoagulant and the of the physiological variation even in healthy term infants. samples were stored up to 24 h (except one for 45 h) before testing. Acceptable stability of the analytes was expected because the collections were stored at a controlled temperature and differential Conflict of interest counting of leukocytes was not addressed in this study.14,22 Third, The authors declare no conflict of interest.

Journal of Perinatology Cord blood platelet characteristics and hemoglobin M Eskola et al 262

Acknowledgments 12 Eichler H, Seetharaman S, Latta M, Kurtz JW, Moroff G. Comparison of total nucleated cell measurements of UC blood samples using two hematology analyzers. Cytotherapy We thank all the participants of the CB bank program and the staff of the Finnish 2004; 6: 457–464. Cord Blood Bank. This work was funded by the Finnish Red Cross Blood Service 13 Bourner G, Dhaliwal J, Sumner J. Performance evaluation of the latest fully automated Research Fund and the Nona and Kullervo Va¨re Foundation. hematology analyzers in a large, commercial laboratory setting: a 4-way, side-by-side study. Lab Hematol 2005; 11: 285–297. 14 Imeri F, Herklotz R, Risch L, Arbetsleitner C, Zerlauth M, Risch GM et al. Stability of References hematological analytes depends on the hematology analyser used: A stability study with Bayer Advia 120, Beckman Coulter LH 750 and Sysmex XE 2100. Clin Chim Acta 2008; 1 Gluckman E, Broxmeyer HA, Auerbach AD, Friedman HS, Douglas GW, Devergie A 397: 68–71. et al. Hematopoietic reconstitution in a patient with Fanconi’s by means 15 Ulich TR, del Castillo J, Senaldi G, Kinstler O, Yin S, Kaufman S et al. Systemic of umbilical-cord blood from an HLA-identical sibling. N Engl J Med 1989; 321: hematologic effects of PEG-rHuMGDF-induced megakaryocyte hyperplasia in mice. 1174–1178. Blood 1996; 87: 5006–5015. 2 Kurtzberg J, Laughlin M, Graham ML, Smith C, Olson JF, Halperin EC et al. Placental 16 Eskola M, Rekunen S, Aroviita P, Mottonen S, Hiilesmaa V, Sainio S et al. Association blood as a source of hematopoietic stem cells for transplantation into unrelated of cord blood platelet characteristics and hematopoietic progenitor cells. Transfusion recipients. N Engl J Med 1996; 335: 157–166. 2008; 48: 884–892. 3 Rubinstein P, Carrier C, Scaradavou A, Kurtzberg J, Adamson J, Migliaccio AR et al. 17 Wilcox AJ, Skjaerven R. Birth weight and perinatal mortality: the effect of gestational Outcomes among 562 recipients of placental-blood transplants from unrelated donors. age. Am J Public Health 1992; 82: 378–382. N Engl J Med 1998; 339: 1565–1577. 18 Pihkala J, Hakala T, Voutilainen P, Raivio K. [Characteristics of recent fetal growth 4 Wagner JE, Barker JN, DeFor TE, Baker S, Blazar BR, Eide C et al. Transplantation of curves in Finland]. Duodecim 1989; 105: 1540–1546 (in Finnish). unrelated donor blood in 102 patients with malignant and 19 Ballen KK, Wilson M, Wuu J, Ceredona AM, Hsieh C, Stewart FM et al. Bigger is better: nonmalignant diseases: influence of CD34 cell dose and HLA disparity on treatment- maternal and neonatal predictors of hematopoietic potential of umbilical cord blood related mortality and survival. Blood 2002; 100: 1611–1618. units. Bone Marrow Transplant 2001; 27: 7–14. 5 Aroviita P, Teramo K, Westman P, Hiilesmaa V, Kekoma¨ki R. Associations among 20 Schwartz R, Teramo KA. Effects of diabetic pregnancy on the fetus and newborn. Semin nucleated cell, CD34+ cell and colony-forming cell contents in cord blood units Perinatol 2000; 24: 120–135. obtained through a standardized banking process. Vox Sang 2003; 84: 219–227. 21 Sutherland DR, Anderson L, Keeney M, Nayar R, Chin-Yee I. The ISHAGE guidelines for 6 Cairo MS, Wagner EL, Fraser J, Cohen G, van de Ven C, Carter SL et al. CD34+ cell determination by flow cytometry. J Hematother 1996; 5: 213–226. Characterization of banked umbilical cord blood hematopoietic progenitor cells and 22 Freise KJ, Schmidt RL, Gingerich EL, Veng-Pedersen P, Widness JA. The effect of lymphocyte subsets and correlation with ethnicity, birth weight, sex, and type of anticoagulant, storage temperature and dilution on cord blood hematology parameters delivery: a Cord Blood Transplantation (COBLT) Study report. Transfusion 2005; 45: over time. Int J Lab Hematol 2009; 31: 496–504. 856–866. 23 Larghero J, Rea D, Brossard Y, Van Nifterik J, Delasse V, Robert I et al. 7 Sparrow RL, Cauchi JA, Ramadi LT, Waugh CM, Kirkland MA. Influence of mode of Prospective flow cytometric evaluation of nucleated red blood cells in cord blood units birth and collection on WBC yields of umbilical cord blood units. Transfusion 2002; and relationship with nucleated and CD34+ cell quantification. Transfusion 2006; 46: 42: 210–215. 403–406. 8 Solves P, Moraga R, Saucedo E, Perales A, Soler MA, Larrea L et al. Comparison 24 Sainio S, Ja¨rvenpa¨a¨ AL, Renlund M, Riikonen S, Teramo K, Kekoma¨ki R. between two strategies for umbilical cord blood collection. Bone Marrow Transplant Thrombocytopenia in term infants: a population-based study. Obstet Gynecol 2000; 95: 2003; 31: 269–273. 441–446. 9 Aroviita P, Teramo K, Hiilesmaa V, Westman P, Kekomaki R. Birthweight of full-term 25 Redzko S, Przepiesc J, Zak J, Urban J, Wysocka J. Influence of perinatal infants is associated with cord blood CD34+ cell concentration. Acta Paediatr 2004; factors on hematological variables in umbilical cord blood. J Perinat Med 2005; 33: 93: 1323–1329. 42–45. 10 Orkin SH, Nathan DG, Ginsburg D, Look TA, Fisher DE, Lux SE (eds). Nathan and 26 Wiedmeier SE, Henry E, Sola-Visner MC, Christensen RD. Platelet reference ranges for Oski’s Hematology of Infancy and Childhood, 7th edn. Saunders Elsevier: neonates, defined using data from over 47 000 patients in a multihospital healthcare Philadelphia, PA, 2009. system. J Perinatol 2009; 29: 130–136. 11 Christensen RD, Henry E, Jopling J, Wiedmeier SE. The CBC: reference ranges for 27 Mahaney MC, Brugnara C, Lease LR, Platt OS. Genetic influences on peripheral blood neonates. Semin Perinatol 2009; 33: 3–11. cell counts: a study in baboons. Blood 2005; 106: 1210–1214.

Journal of Perinatology