Journal of Perinatology (2010) 30, 765–769 r 2010 Nature America, Inc. All rights reserved. 0743-8346/10 www.nature.com/jp STATE-OF-THE-ART Increasing without transfusion: is it time to introduce novel thrombopoietic agents in neonatal care?

H Sallmon1,2, RK Gutti1, F Ferrer-Marin1,3, Z-J Liu1 and MC Sola-Visner1 1Division of Newborn Medicine, Children’s Hospital Boston and Harvard Medical School, Boston, MA, USA; 2Department of Neonatology, Charite´FUniversita¨tsmedizin Berlin, Berlin, Germany and 3Department of Haematology and Medical Oncology, Centro Regional de Hemodonacio´n, University of Murcia, Murcia, Spain

of recombinant (Epo) and granulocyte colony- The Food and Drug Administration recently approved two novel stimulating factor prompted a flurry of studies using these agents thrombopoiesis-stimulating agents, Romiplostim (AMG-531, Nplate) and in neonates with or neutropenia, respectively. In fact, (Promacta), for the treatment of adults with immune although not as widespread as before, Epo is still used to treat thrombocytopenic purpura. For physicians taking care of critically ill anemia of prematurity4 and granulocyte colony-stimulating factor neonates, this offers the opportunity of decreasing transfusions and continues to be used in neonates with severe neutropenia of potentially improving the outcomes of neonates with severe and prolonged different etiologies.5 . However, several developmental factors need to be taken In contrast to Epo and granulocyte colony-stimulating factor, into consideration. First, the population of thrombocytopenic neonates likely the most potent stimulator of platelet production was not identified to benefit from these agents needs to be carefully selected. Second, the until 1994, when (Tpo) was clonedFalmost mechanisms underlying neonatal and adult thrombocytopenia differ from simultaneouslyFby five independent groups around the world. each other and are incompletely understood, and pre-clinical evidence The importance of this discovery was evidenced by the speed with suggests that the response of neonates to thrombopoietic factors might be which the new molecule moved from bench to clinical studies. different from that of adults. Finally, the potential non-hematopoietic effects These centered around two recombinant forms of Tpo: a full-length of thrombopoietin have not been well established. Here, we will discuss these Tpo (rhTpo) and a pegylated form containing only the receptor- issues in detail, and will highlight the critical developmental differences binding domain, named pegylated recombinant human between neonates and adults that need to be considered as we think about -growth-and-development factor (PEG-rHuMGDF). introducing these agents into neonatal care. When Food and Drug Administration approval of these compounds Journal of Perinatology (2010) 30, 765–769; doi:10.1038/jp.2010.50; seemed imminent, however, a number of subjects treated with published online 22 April 2010 PEG-rHuMGDF developed cross-reactive neutralizing antibodies Keywords: thrombocytopenia; neonate; thrombopoietin; thrombopoietic against their endogenous Tpo, which led to severe hyporegenerative 6,7 growth factors thrombocytopenia and aplastic anemia. The seriousness of this side effect eventually caused the discontinuation of all clinical development efforts for Tpo. Clinical trials evaluating the effect of rhTpo in thrombocytopenic neonates were, therefore, never Introduction carried out. Neonatal cytopenias (anemia, thrombocytopenia, and neutropenia) The termination of all studies involving rhTpo triggered an are common problems among infants admitted to the neonatal 1–3 active search for so-called ‘second generation’ thrombopoietic intensive care unit (NICU). If treatment is required, cytopenias factors. This culminated with the discovery of several ‘Tpo- can be corrected by either transfusing the cells that are deficient, or mimetics’, molecules that do not share any sequence homology by stimulating the endogenous production of the affected cell line with endogenous Tpo, but stimulate thrombopoiesis by binding with the use of specific growth factors. Historically, the availability and activating the Tpo receptor.8 This new family of agents Correspondence: Dr MC Sola-Visner, Division of Newborn Medicine, Children’s Hospital includes Tpo-mimetic peptides (AMG-531 and Fab-59), Tpo- Boston, 300 Longwood Avenue, Enders Research Building, Room 961, Boston, MA 02115, mimetic non-peptides (Eltrombopag, LGD-4665, AKR-501, and USA. Butyzamide), and Tpo agonist antibodies (VB22B sc(Fv)2 and E-mail: [email protected] 9 Received 15 September 2009; revised 8 January 2010; accepted 1 February 2010; published MA01G4G344). Among those, AMG-531 (Nplate, Romiplostim) online 22 April 2010 and Eltrombopag (Promacta) were recently approved by the Food Novel thrombopoietic agents H Sallmon et al 766 and Drug Administration for the treatment of adults with immune commitment of hematopoietic stem cells toward megakaryocytic thrombocytopenic purpura refractory to conventional therapy. differentiation and the proliferation of megakaryocyte progenitors Initial clinical trials evaluating Eltrombopag and Romiplostim in (the cells that multiply and give rise to ). adult patients consistently showed high efficacy and a favorable Immature megakaryocytes then undergo a maturational process, safety profile.10,11 However, all clinical studies so far have consisting of a progressive increase in nuclear ploidy and exclusively focused on adult patients with a naturally long disease cytoplasmic maturation that ultimately generates large polyploid course, and who were refractory to other established treatment cells capable of releasing platelets into the circulation.16 protocols. From a pathophysiological perspective, our understanding of the mechanisms underlying low platelet counts in neonates is still limited. We know, however, that there are important developmental Significance of neonatal thrombocytopenia differences between neonatal and adult megakaryocytes, which Thrombocytopenia, classically defined as a platelet count below likely contribute to the vulnerability of neonates to develop <150 000 per ml, is a common problem among sick neonates. In prolonged and severe thrombocytopenia. Among these differences, fact, 20–35% of all neonates admitted to an NICU have at least neonatal megakaryocytes are significantly smaller and of lower one platelet count <150 000 per ml.2 Extremely low birth weight ploidy than adult megakaryocytes,17,18 and smaller megakaryocytes infants (<1000 g birth weight) are even more likely to have are known to give rise to less platelets.19 Under normal conditions, thrombocytopenia, with approximately 70% of them developing this this apparent impairment is circumvented by the higher problem.12 proliferative potential of neonatal megakaryocyte progenitors,20,21 To this date, the only available treatment option for newborns which generate megakaryocytes faster and in higher numbers than with non-immune thrombocytopenia has been to administer adult progenitors. During thrombocytopenia, however, recent platelet transfusions. Owing mostly to a lack of solid evidence to studies have suggested that neonates might not be able to up- guide transfusion decisions in neonates, transfusion guidelines and regulate platelet production to the degree that adults do. practices differ substantially among centers and among individual Specifically, adults with consumptive thrombocytopenia can neonatologists.13 However, based on published single-institution increase platelet production by two- to eightfold, and they do so by studies, it has been estimated that approximately 32 000 (8%) of increasing the size and number of their marrow the 400 000 neonates admitted to NICUs in the USA each year megakaryocytes.22,23 Neonates, in contrast, seem to have a limited receive at least one platelet transfusion. Approximately half of these ability to increase the size of their megakaryocytes, thus abrogating infants receive two or more transfusions.14 The incidence of one of the compensatory mechanisms of adults.24,25 Furthermore, transfusion-related complications in neonates is unknown. neonates with certain varieties of thrombocytopenia have lower Tpo However, several studies in adults have shown that platelet levels than adults with similar mechanisms of transfusions, in particular, are associated with substantial risks, thrombocytopenia.20,26 Taken together, these findings imply a such as bacterial infection and transfusion-associated lung injury. generally lower potential of neonates to increase thrombopoiesis in Given the high incidence of sepsis and respiratory decompensation response to increased platelet demand, and provide a good episodes among neonates, it is likely that these transfusion rationale to consider the use of thrombopoiesis-stimulating agents complications are under-recognized in the NICU setting. in this population. Furthermore, several studies have documented a direct correlation However, a number of developmental factors need to be taken between the number of platelet transfusions and morbidity and into consideration when deciding how to introduce these agents in mortality in neonates. It is unclear whether this simply reflects the neonatal care. First, it will be critical to carefully determine which fact that sicker patients receive more platelet transfusions, but at neonates would benefit from treatment with thrombopoiesis- least one study has suggested that platelet transfusions per se stimulating agents. In this regard, we know that Tpo and Tpo might contribute to the high mortality.15 Given these facts, mimetics begin to increase the platelet count 4–6 days after the pharmacologically stimulating platelet production and decreasing start of therapy, and reach peak platelet counts at 10–14 days. the need for platelet transfusions in infants with prolonged and Thus, only infants whose disease process would predict a platelet severe thrombocytopenia might lead to improved outcomes among transfusion requirement lasting longer than 2 weeks would be such neonates. appropriate candidates for treatment.27 Given that approximately 80% of cases of severe thrombocytopenia in the NICU resolve within 14 days,28 this constitutes the minority of thrombocytopenic Use of thrombopoietic agents in neonates: special neonates. Furthermore, there are currently no markers that allow considerations us to predict which neonates will experience prolonged Platelet formation is a complex process. The first step is the thrombocytopenia, although two studies have shown an association production of a thrombopoietic stimulus, which promotes the between liver disease and prolonged thrombocytopenia in neonates.

Journal of Perinatology Novel thrombopoietic agents H Sallmon et al 767

As the liver is the main site of Tpo production in neonates and From a biological perspective, it will also be essential to adults,29 infants with liver failure and severe thrombocytopenia carefully evaluate the effects of Tpo and its mimetics specifically on might be attractive potential candidates.14,30 Many infants with neonatal megakaryocytopoiesis. We know so far that neonatal congenital viral infections might fall into this category. This megakaryocyte progenitors are more sensitive to Tpo in vitro, and approach has already been tested in adults with thrombocytopenia respond to lower Tpo concentrations than adult megakaryocyte because of HCV-related cirrhosis, in whom treatment with progenitors.37 In concordance with these in vitro findings, Eltrombopag increased the platelet counts and permitted the newborn rhesus monkeys treated with PEG-rHuMGDF achieved initiation of antiviral therapy. maximal platelet counts at lower doses than those needed to treat The neonates who receive the largest numbers of platelet adult animals.38 Qualitative differences in the response to Tpo transfusions (‘very high users’, >20 platelet transfusions), between neonatal and adult megakaryocytes have also been however, belong primarily to four groups: extracorporeal reported. Specifically, Tpo has been shown to potently stimulate the membrane oxygenation (ECMO), bacterial or fungal sepsis, nuclear maturation of adult megakaryocytes, whereas inhibiting necrotizing enterocolitis (NEC), or congenital hyporegenerative the same process in neonatal megakaryocytes.39 The molecular thrombocytopenias (such as thrombocytopenia with absent radii mechanisms underlying these findings are unknown, but it is syndrome or congenital amegakaryocytic thrombocytopenia).31 In evident that there are substantial molecular differences between the specific cases of ECMO, sepsis, and NEC, the predominant neonatal and adult megakaryocytes. mechanism of the thrombocytopenia is increased platelet activation Taken together, the available data suggest that Tpo mimetics and consumption, although recent data suggest that neonates with will stimulate platelet production in neonates, but may do so sepsis or NEC can also have an insufficient compensatory increase through different mechanisms and at different doses (per kg) than in thrombopoiesis.32 Thus, infants with these conditions might be in adults. Furthermore, all hematopoietic have been particularly appealing candidates for treatment with Tpo mimetics. shown to affect cells and tissues outside the hematopoietic system. However, two important challenges need to be dealt with when Epo, for example, has antiapoptotic properties on several cell types, considering this approach. First is the duration of including neuronal, myocardial, and hepatic cells.40 Other non- thrombocytopenia in these disease processes. In the case of ECMO, hematopoietic effects of Epo include its pro-angiogenic effect on a recent review of 22 032 neonatal runs reported a median duration vascular cells, which might contribute to the higher incidence of of ECMO of 5.4 days (mean±s.d. of 6.4±3.7 days),33 implying retinopathy of prematurity among anemic infants treated with Epo that therapy with Tpo mimetics would not benefit the majority of during the first week of life.41 The non-hematopoietic effects of Tpo neonates on ECMO. Similarly, the thrombocytopenia associated have not yet been well characterized, particularly in a developing with sepsis and NEC resolves within 1–2 weeks in the majority of organism. However, it is known that Tpo and its receptor are adequately treated neonates, although in someFfor unexplained expressed in the brain,42 and recent data suggest that Tpo might and unpredictable reasonsFthe thrombocytopenia persists for have pro-apoptotic and differentiation-blocking effects on neuronal several weeks. In this regard, the newly developed ‘immature cells.43,44 This might be particularly important as Romiplostim and platelet fraction’ (a measure of new platelet production) might Eltrombopag have lower molecular weights than Tpo, and might become a helpful parameter to predict the duration of be more likely to cross the blood–brain barrier. Given these thrombocytopenia in neonates.34 A second consideration is the fact considerations, a careful evaluation of potential non-hematopoietic that ECMO, sepsis, and NEC are associated with high levels of effects on the developing organism is warranted. platelet activation. In vitro studies have shown that Tpo and Romiplostim (but not Eltrombopag)35 increase the degree of platelet activation in response to agonists. Thus, the potential of Conclusion Tpo-mimetic agents to increase the risk of thrombosis needs to be The recent availability of novel thrombopoiesis-stimulating agents considered, particularly during the acute phase of clinical with a favorable safety profile in adults has opened the door to a conditions already characterized by high levels of platelet new and promising therapeutic approach for severe neonatal activation. thrombocytopenia. However, neonatologists and hematologists need In regard to congenital amegakaryocytic thrombocytopenia and to be aware of developmental factors that affect the decision of thrombocytopenia with absent radii, their pathogenesis involves when and how to use these agents in neonates. First, the group(s) mutations in the Tpo receptor and defective intracellular Tpo of neonates that will likely benefit from this treatment need signaling, respectively, which renders these diseases naturally to be carefully defined. Second, the pharmacokinetic and unresponsive to Tpo. For that reason, Tpo mimetics are currently pharmacodynamic properties of these agents in neonates need to be not indicated for these conditions, although it is possible that determined. Noteworthy is the earlier experience with Epo in the patients with specific mutations in the Tpo receptor might respond treatment of anemia of prematurity, which revealed substantial to certain receptor agonists.36 pharmacokinetic differences between neonates and adults.45 Third,

Journal of Perinatology Novel thrombopoietic agents H Sallmon et al 768 the differences between neonatal and adult megakaryocytes in their 11 Bussel JB, Cheng G, Saleh MN, Psaila B, Kovaleva L, Meddeb B et al. Eltrombopag for responses to thrombopoietic agents need to be better characterized. the treatment of chronic idiopathic thrombocytopenic purpura. N Engl J Med 2007; And fourth, the potential non-hematopoietic effects of these novel 357: 2237–2247. factors should be carefully considered, particularly when used in a 12 Christensen RD, Henry E, Wiedmeier SE, Stoddard RA, Sola-Visner MC, Lambert DK et al. Thrombocytopenia among extremely low birth weight neonates: data from a developing organism. multihospital healthcare system. J Perinatol 2006; 26: 348–353. In conclusion, we believe that Tpo mimetics bring the promise 13 Josephson CD, Su LL, Christensen RD, Hillyer CD, Castillejo MI, Emory MR et al. of reducing platelet transfusions and potentially improving the Platelet transfusion practices among neonatologists in the United States and Canada: outcomes of neonates with prolonged and severe thrombocytopenia. results of a survey. Pediatrics 2009; 123: 278–285. However, we also believe that a non-judicious extrapolation of 14 Del Vecchio A, Sola MC, Theriaque DW, Hutson AD, Kao KJ, Wright D et al. Platelet transfusions in the neonatal intensive care unit: factors predicting which patients will results from adult studies to neonates could represent a significant require multiple transfusions. Transfusion 2001; 41: 803–808. pitfall. Thus, it is our opinion that further preclinical studies 15 Baer VL, Lambert DK, Henry E, Snow GL, Sola-Visner MC, Christensen RD. evaluating these issues are warranted, and that these novel Do platelet transfusions in the NICU adversely affect survival? Analysis of 1600 thrombopoietic agents should currently be used in neonates only thrombocytopenic neonates in a multihospital healthcare system. J Perinatol 2007; 27: within the context of clinical trials designed to address the 790–796. questions posed above, and that take into account the profound 16 Kaushansky K. The molecular mechanisms that control thrombopoiesis. J Clin Invest 2005; 115: 3339–3347. differences between neonatal and adult patients. 17 Levine RF, Olson TA, Shoff PK, Miller MK, Weisman LE. Mature micromegakaryocytes: an unusual developmental pattern in term infants. Br J Haematol 1996; 94: 391–399. Conflict of interest 18 Olson TA, Levine RF, Mazur EM, Wright DG, Salvado AJ. Megakaryocytes and megakaryocyte progenitors in human cord blood. Am J Pediatr Hematol Oncol 1992; The authors declare no conflict of interest. 14: 241–247. 19 Mattia G, Vulcano F, Milazzo L, Barca A, Macioce G, Giampaolo A et al. Different ploidy levels of megakaryocytes generated from peripheral or cord blood CD34+ cells are Acknowledgments correlated with different levels of platelet release. Blood 2002; 99: 888–897. 20 Murray NA, Watts TL, Roberts IA. Endogenous thrombopoietin levels and effect of Supported by a German Academic Exchange Service (DAAD) Fellowship (HS). recombinant human thrombopoietin on megakaryocyte precursors in term and preterm babies. Pediatr Res 1998; 43: 148–151. 21 Nishihira H, Toyoda Y, Miyazaki H, Kigasawa H, Ohsaki E. Growth of macroscopic human megakaryocyte colonies from cord blood in culture with recombinant human References thrombopoietin (c-mpl ligand) and the effects of gestational age on frequency of 1 Bishara N, Ohls RK. Current controversies in the management of the anemia of colonies. Br J Haematol 1996; 92: 23–28. prematurity. Semin Perinatol 2009; 33: 29–34. 22 Harker LA. Kinetics of thrombopoiesis. J Clin Invest 1968; 47: 458–465. 2 Roberts I, Stanworth S, Murray NA. Thrombocytopenia in the neonate. Blood Rev 23 Harker LA. Thrombokinetics in idiopathic thrombocytopenic purpura. Br J Haematol 2008; 22: 173–186. 1970; 19: 95–104. 3 Christensen RD, Calhoun DA. Congenital neutropenia. Clin Perinatol 2004; 31: 24 Sola-Visner MC, Christensen RD, Hutson AD, Rimsza LM. Megakaryocyte size and 29–38. concentration in the of thrombocytopenic and nonthrombocytopenic 4 Ohls RK, Ehrenkranz RA, Wright LL, Lemons JA, Korones SB, Stoll BJ et al. Effects of neonates. Pediatr Res 2007; 61: 479–484. early erythropoietin therapy on the transfusion requirements of preterm infants below 25 Hu Z, Slayton WB, Rimsza LM, Bailey M, Sallmon H, Sola-Visner MC. Differences 1250 grams birth weight: a multicenter, randomized, controlled trial. Pediatrics 2001; between newborn and adult mice in their response to immune thrombocytopenia. 108: 934–942. Neonatology 2010; 98: 100–108. 5 Bernstein HM, Calhoun DA, Christensen RD. Use of myeloid colony-stimulating factors 26 Sola MC, Calhoun DA, Hutson AD, Christensen RD. Plasma thrombopoietin in neonates with septicemia. Curr Opin Pediatr 2002; 14: 91–94. concentrations in thrombocytopenic and non-thrombocytopenic patients in a neonatal 6 Basser RL, O’Flaherty E, Green M, Edmonds M, Nichol J, Menchaca DM et al. intensive care unit. Br J Haematol 1999; 104: 90–92. Development of pancytopenia with neutralizing antibodies to thrombopoietin after 27 Sola MC, Dame C, Christensen RD. Toward a rational use of recombinant multicycle chemotherapy supported by megakaryocyte growth and development factor. thrombopoietin in the neonatal intensive care unit. J Pediatr Hematol Oncol 2001; Blood 2002; 99: 2599–2602. 23: 179–184. 7 Li J, Yang C, Xia Y, Bertino A, Glaspy J, Roberts M et al. Thrombocytopenia 28 Murray NA, Howarth LJ, McCloy MP, Letsky EA, Roberts IA. Platelet transfusion in the caused by the development of antibodies to thrombopoietin. Blood 2001; 98: management of severe thrombocytopenia in neonatal intensive care unit patients. 3241–3248. Transfus Med 2002; 12: 35–41. 8 Nurden AT, Viallard JF, Nurden P. New-generation drugs that stimulate platelet 29 Dame C. Developmental biology of thrombopoietin in the human fetus and neonate. production in chronic immune thrombocytopenic purpura. Lancet 2009; 373: Acta Paediatr Suppl 2002; 91: 54–65. 1562–1569. 30 Garcia MG, Duenas E, Sola MC, Hutson AD, Theriaque D, Christensen RD. 9 George JN, Terrell DR. Novel thrombopoietic agents: a new era for management of Epidemiologic and outcome studies of patients who received platelet transfusions in the patients with thrombocytopenia. Haematologica 2008; 93: 1445–1449. neonatal intensive care unit. J Perinatol 2001; 21: 415–420. 10 Bussel JB, Kuter DJ, George JN, McMillan R, Aledort LM, Conklin GT et al. AMG 31 Dohner ML, Wiedmeier SE, Stoddard RA, Null Jr D, Lambert DK, Burnett J et al. Very 531, a thrombopoiesis-stimulating protein, for chronic ITP. N Engl J Med 2006; 355: high users of platelet transfusions in the neonatal intensive care unit. Transfusion 1672–1681. 2009; 49: 869–872.

Journal of Perinatology Novel thrombopoietic agents H Sallmon et al 769

32 Brown RE, Rimsza LM, Pastos K, Young L, Saxonhouse MA, Bailey M et al. Effects of 39 Pastos KM, Slayton WB, Rimsza LM, Young L, Sola-Visner MC. Differential effects of sepsis on neonatal thrombopoiesis. Pediatr Res 2008; 64: 399–404. recombinant thrombopoietin and bone marrow stromal-conditioned media on 33 Fleming GM, Gurney JG, Donohue JE, Remenapp RT, Annich GM. Mechanical neonatal versus adult megakaryocytes. Blood 2006; 108: 3360–3362. component failures in 28,171 neonatal and pediatric extracorporeal membrane 40 Jelkmann W. Erythropoietin after a century of research: younger than ever. Eur J oxygenation courses from 1987 to 2006. Pediatr Crit Care Med 2009; 10: 439–444. Haematol 2007; 78: 183–205. 34 Cremer M, Paetzold J, Schmalisch G, Hammer H, Loui A, Dame C et al. Immature 41 Ohlsson A, Aher SM. Early erythropoietin for preventing red blood cell transfusion in platelet fraction as novel laboratory parameter predicting the course of neonatal preterm and/or low birth weight infants. Cochrane Database Syst Rev 2006; 3: thrombocytopenia. Br J Haematol 2009; 144: 619–621. CD004863. 35 Erhardt JA, Erickson-Miller CL, Aivado M, Abboud M, Pillarisetti K, Toomey JR. 42 Dame C, Wolber EM, Freitag P, Hofmann D, Bartmann P, Fandrey J. Thrombopoietin Comparative analyses of the small molecule agonist gene expression in the developing human central nervous system. Brain Res Dev eltrombopag and thrombopoietin on in vitro platelet function. Exp Hematol 2009; 37: Brain Res 2003; 143: 217–223. 1030–1037. 43 Ehrenreich H, Hasselblatt M, Knerlich F, von Ahsen N, Jacob S, Sperling S et al. A 36 Geddis AE. Congenital amegakaryocytic thrombocytopenia and thrombocytopenia hematopoietic growth factor, thrombopoietin, has a proapoptotic role in the brain. Proc with absent radii. Hematol Oncol Clin North Am 2009; 23: 321–331. Natl Acad Sci USA 2005; 102: 862–867. 37 Sola MC, Du Y, Hutson AD, Christensen RD. Dose-response relationship of megakaryocyte 44 Samoylenko A, Byts N, Rajalingam K, von Ahsen N, Rapp UR, Ehrenreich H et al. progenitors from the bone marrow of thrombocytopenic and non-thrombocytopenic Thrombopoietin inhibits nerve growth factor-induced neuronal differentiation and ERK neonates to recombinant thrombopoietin. Br J Haematol 2000; 110: 449–453. signalling. Cell Signal 2008; 20: 154–162. 38 Sola MC, Christensen RD, Hutson AD, Tarantal AF. , pharmacody- 45 Widness JA, Veng-Pedersen P, Modi NB, Schmidt RL, Chestnut DH. Developmental namics, and safety of administering pegylated recombinant megakaryocyte growth and differences in erythropoietin pharmacokinetics: increased clearance and distribution in development factor to newborn rhesus monkeys. Pediatr Res 2000; 47: 208–214. fetal and neonatal sheep. J Pharmacol Exp Ther 1992; 261: 977–984.

Journal of Perinatology