Erythropoietin Improved Neurologic Outcomes in Newborns With Hypoxic-Ischemic Encephalopathy
OBJECTIVE: The purpose of this study was to evaluate the efficacy and safety of erythropoietin in neonatal hypoxic-ischemic encephalopathy (HIE), by using a randomized, prospective study design.
METHODS: A total of 167 term infants with moderate/severe HIE were assigned randomly to receive either erythropoietin (N = 83) or conventional treatment (N = 84). Recombinant human erythropoietin, at either 300 U/kg (N = 52) or 500 U/kg (N = 31), was administered every other day for 2 weeks, starting <48 hours after birth. The primary outcome was death or disability. Neurodevelopmental outcomes were assessed at 18 months of age.
RESULTS: Complete outcome data were available for 153 infants. Nine patients dropped out during treatment, and 5 patients were lost to follow-up monitoring. Death or moderate/severe disability occurred for 35 (43.8%) of 80 infants in the control group and 18 (24.6%) of 73 infants in the erythropoietin group (P = .017) at 18 months. The primary outcomes were not different between the 2 erythropoietin doses. Subgroup analyses indicated that erythropoietin improved long-term outcomes only for infants with moderate HIE (P = .001) and not those with severe HIE (P = .227). No negative hematopoietic side effects were observed.
CONCLUSION: Repeated, low-dose, recombinant human erythropoietin treatment reduced the risk of disability for infants with moderate HIE, without apparent side effects.
Perinatal asphyxia-induced brain injury is one of the most common causes of death and long-term neurologic impairments (cerebral palsy, mental retardation, learning disability, and epilepsy) in term and preterm neonates.1 The current treatment for hypoxic-ischemic encephalopathy (HIE) is predominantly supportive, to maintain physiologic parameters. Multicenter trials assessing the effects of hypothermia demonstrated improved outcomes for term neonates with moderate HIE if the infants underwent cooling within 6 hours.2,3 Hypothermia is not widely used in China, because efficacy is uncertain if treatment is delayed >6 hours.4 The safety and efficacy of hypothermia for preterm infants need to be proven and such treatment may not be suitable for this group,5 leaving a substantial proportion of patients with asphyxia without effective treatment. There is a need for effective treatments that can be implemented even >6 hours after the injurious event.
Erythropoietin was originally identified on the basis of its role in erythropoiesis. Clinical trials demonstrated the safety and efficacy of recombinant human erythropoietin in the prevention and treatment of anemia of prematurity.6,7 In addition, erythropoietin may mediate an adaptive tissue response to stress and may mediate tissue protection.8 Erythropoietin and its receptors were upregulated after brain injury, and the levels of erythropoietin in cerebrospinal fluid (CSF) were correlated positively with outcomes.9 Systemically administered erythropoietin was neuroprotective in neonatal brain injury models.10–13 To date, there are no reports evaluating possible effects of erythropoietin on neonatal HIE. A clinical study using recombinant human erythropoietin to prevent anemia of prematurity found that elevated serum erythropoietin concentrations were correlated with higher Mental Developmental Index (MDI) scores.14 Furthermore, a placebo-controlled trial in adult patients after stroke, using a dose of 33000 U per injection (450 U/kg of body weight), showed beneficial results.15 However, erythropoietin doses used in animal neuroprotection studies (1000–30000 U/kg) are invariably much higher than the doses for anemia treatment (250–500 U/kg) and are not considered suitable for human neonates because of potential side effects.16 Because of the lack of reports on the safety of higher doses of erythropoietin in neonates, a low-dose regimen was chosen. The aim of this study was to investigate whether systemic administration of low doses (300–500 U/kg) of recombinant human erythropoietin could improve neurodevelopmental outcomes at 18 months for infants with moderate or severe HIE.
Between August 2003 and January 2007, infants who were born at >37 weeks of gestation, with body weights of >2500 g, and were admitted to the NICU with clinical evidence of perinatal HIE were screened for eligibility (Fig 1). The sample size was determined on the basis of the following assumptions: 60% of the patients would suffer from moderate asphyxia and 40% from severe asphyxia, and 60% of the patients with moderate asphyxia and 90% with severe asphyxia in the control group would be dead or disabled after 18 months. If the relative risk would be reduced 40% in the moderate asphyxia group and 10% in the severe asphyxia group with erythropoietin treatment, then 65 patients would need to be recruited in each group for a significance level of 5%, with 80% power and 20% lost to follow-up monitoring. The inclusion criteria were Apgar scores of ≤5 at 5 minutes after birth or continued need for resuscitation (including endotracheal or mask ventilation) at 10 minutes after birth. The severity of encephalopathy (moderate or severe) was assessed by certified examiners according to the criteria described by Sarnat and Sarnat,17 consisting of altered state of consciousness (lethargy, stupor, or coma) and ≥1 of hypotonia, abnormal reflexes (including oculomotor or pupillary abnormalities), absent or weak sucking, or clinical seizures. Exclusion criteria were major congenital abnormalities, head trauma or skull fracture causing intracranial hemorrhage, body temperature of <34°C, financial problems of the parents, lack of permanent address, and postnatal age of >48 hours. Written informed consent was obtained from ≥1 of the parents for all patients assigned randomly to erythropoietin treatment. This study was conducted in 2 neonatal centers in Zhengzhou, China, and was approved by the local ethics committee in accordance with the Helsinki Declaration.
Study Design and Treatment Procedures
For patients fulfilling the inclusion criteria, complete obstetric history was obtained. The severity of HIE was determined as described above, and then patients were assigned randomly to either erythropoietin or conventional treatment. Randomization was performed separately in the 2 hospitals, according to the severity of HIE. In the first phase, we used 300 U/kg. Because we did not observe any obvious side effects, we increased the dose to 500 U/kg for the second phase, speculating that a higher dose would be more effective. In the erythropoietin treatment group, patients were given recombinant human erythropoietin at either 300 U/kg or 500 U/kg, administered subcutaneously the first time and then intravenously every other day for 2 weeks. The time to the first erythropoietin injection was 1 to 48 hours after birth (median: 20 hours for 300 U/kg and 24 hours for 500 U/kg). Each intravenous injection was completed in <2 minutes. The supportive care and other treatments were identical between the groups. Additional clinical variables assessed before and 2 weeks after treatment included liver and renal function, blood hemoglobin level, hematocrit level, platelet count, and reticular red blood cell count.
Neurologic signs were assessed and scored, according to the method described by Thompson et al,18 every day for 7 days, because the evaluation on day 7 is a useful time point with regard to prognosis.19 The scoring consists of clinical assessment of 9 neurologic signs, with scores from 0 to 3 and a maximal score of 22. The higher the score is, the more the affected infant is at risk of neuropsychomotor disabilities.
The 20-item neonatal behavioral neurologic assessment was performed at postnatal days 14 and 28.20 Each item is scored as 0, 1 or 2, and the maximal score is 40 for normal term infants. The investigators performing the neonatal behavioral neurologic assessment underwent the same training program, to ensure strong interobserver agreement.
All surviving infants were evaluated regularly every 6 months through gross neurologic assessment and MDI testing, with a final examination at 18 months. Assessment of neuromotor disability was based on the presence of cerebral palsy and functional disability, graded according to the 5-level classification described by Palisano et al.21 Psychological development was evaluated by using the Bayley Scales of Infant Development, Second Edition.22 Criteria for moderate/severe disability included cerebral palsy, severe hearing loss, blindness, gross motor function classification levels 3 through 5, and MDI of <70.
In this study, the doctors and nurses responsible for treatment were not blinded but the investigators performing the short- and long-term outcome assessments were blinded to the patients' group allocation. The final evaluation at 18 months of age was performed by doctors from another department, who were blinded to the treatment protocol and were not allowed to inquire about the treatment history from the parents.
Samples of blood and lumbar CSF were collected before (n = 10) and 3 hours (n = 10), 8 hours (n = 8), or 24 hours (n = 7) after subcutaneous administration of a single dose of 500 U/kg recombinant human erythropoietin, from different patients. Lumbar CSF samples from untreated healthy newborns were not available, for ethical reasons. The samples were centrifuged and the supernatants were frozen at −20°C. CSF samples with blood contamination (visible or microscopic) were excluded. Erythropoietin was detected by using a radioimmunoassay, according to the manufacturer's protocol (R-20; East Asia Immunotechnology Institute, Beijing, China).
The outcome differences were analyzed with Fisher's exact tests for categorical variables, and P values are 2-sided. Analysis of variance with Bonferroni correction was used when >2 groups were compared. Data were expressed as mean ± SD or median and range, and P values of <.05 were considered statistically significant.
Erythropoietin Levels in Serum and CSF
Endogenous erythropoietin was detected in both serum and CSF from the patients with HIE. The concentration in serum was 4.6 times higher than that in CSF (Fig 2). After subcutaneous administration, the erythropoietin level in serum increased threefold (P = .0046), reached a peak 3 hours after administration, and then decreased gradually 8 and 24 hours after administration (Fig 2A). The changes in CSF were similar to those in serum. The peak CSF level was 23.75 mU/mL (range: 13.75–43.75 mU/mL) at 3 hours after injection, which was significantly higher than the level before erythropoietin injection (Fig 2B).
Baseline Characteristics and Follow-up Data
A total of 167 patients were enrolled in this study; 9 of them dropped out during treatment, at the parents' request. Of the remaining 158 patients, 82 were assigned to the control group and 76 to the erythropoietin group, 47 of whom were allocated to the 300 U/kg erythropoietin subgroup and 29 to the 500 U/kg erythropoietin subgroup. The baseline characteristics of the patients in the control and recombinant human erythropoietin groups were not significantly different (Table 1). Neurologic improvement attributable to erythropoietin was observed at postnatal day 7, with erythropoietin-treated patients having significantly lower Thompson scores, compared with control subjects (Table 2). Persistent beneficial effects of erythropoietin treatment were demonstrated by higher neonatal behavioral neurologic assessment scores at 14 and 28 days of age (Table 3).
Outcomes at 18 Months
Data on outcomes at 18 months were available for 153 (96.8%) of the 158 patients. Two patients in the control group and 3 in the erythropoietin group (2 in the 300 U/kg erythropoietin subgroup and 1 in the 500 U/kg erythropoietin subgroup) were lost to follow-up monitoring. Four patients in the control group died, at 4, 18, and 25 days and 2 months of age. Two of those patients had severe HIE and 2 moderate HIE. Three patients in the erythropoietin group died, at 7 days, 14 days, and 3 months of age, and all of those patients had severe HIE (Table 4). Death was related directly to either severe brain injury or infection. Death or moderate/severe disability was present for 35 (43.8%) of 80 infants in the control group and 18 (24.6%) of 73 infants in the erythropoietin group (P = .017). The rates of disability were 40.8% in the control group and 21.4% in the erythropoietin group (P = .013). The proportions of subjects with MDI scores of <70 were 17 (22.4%) of 76 infants in the control group and 7 (10%) of 70 infants in the erythropoietin group (P = .048). The proportions with cerebral palsy were 14 (18.4%) of 76 infants in the control group and 5 (6.8%) of 70 infants in the erythropoietin group (P = .051). Benefit from erythropoietin treatment, as judged on the basis of death and disability, was seen with moderate HIE (P = .001) but not severe HIE (P = .227) (Table 4). The final analysis also revealed that overall disability was significantly reduced after erythropoietin treatment for girls (P = .029) but not for boys (P = .118). The primary outcomes at 18 months of age did not show any significant differences between the 2 erythropoietin dose groups (Table 5).
Erythropoietin was well tolerated, and neither allergic reactions nor venous thromboses were observed with this treatment protocol. Liver and renal functions, as well as electrolyte levels, were not different between the erythropoietin and control groups. Hemoglobin and reticulocyte levels decreased significantly at 2 weeks after birth in the control group, but this decrease was prevented in the erythropoietin groups (Table 6).
The results of our investigation demonstrated that administration of recombinant human erythropoietin at a dose of 300 or 500 U/kg every other day for 2 weeks was safe and resulted in improved neurologic outcomes, as assessed at 18 months of age, for patients with moderate but not severe HIE. This study raises the following questions. (1) Can the patient population in this study be compared readily with those in the hypothermia trials? (2) What are the pharmacokinetic characteristics of erythropoietin, including issues of dosage and timing, and does administered erythropoietin cross the blood-brain barrier (BBB)? (3) How do the effectiveness, side effects, and potential of erythropoietin therapy compare with those of induced hypothermia?
Death or moderate/severe disability was present for 35 (43.8%) of 80 infants in the control group and 18 (24.6%) of 73 infants in the erythropoietin group (P = .017). The death and disability rate for the control group was lower than that in the hypothermia study (66%),2 probably because of the small proportion of severe HIE in our study (26.25%, compared with 37.7%).2 Consequently, these results demonstrated a distinctly beneficial effect on neurodevelopmental outcomes even when erythropoietin was administered up to 48 hours after birth, at a dosage similar to those used for treatment for anemia. Erythropoietin had no effect in reducing the mortality rate, but the rate of disability was reduced from 40.8% to 21.4% (P = .013). The overall effect of erythropoietin was to halve the incidence of death or disability. However, a statistically significant effect was demonstrated only for infants with moderate HIE. This is analogous to the results noted after head cooling, which reduced significantly the rate of disability in infants with less-severe amplitude-integrated electroencephalographic changes.3 The rate of disability after treatment with erythropoietin was similar to that seen after induced hypothermia,2,3,23 with reduction in the rate of disability among survivors from 46% to 40% in the control group compared to 31% to 27% in the hypothermia group. We report a similar or greater reduction in the rate of disability, from 40.8% to 21.4%, for infants treated with erythropoietin (Table 4). In reducing the proportion of infants with MDI scores of <70, erythropoietin had an effect similar to that of induced hypothermia, reducing the proportion from 22.4% in the control group to 10%.
Compared with induced hypothermia, erythropoietin treatment is technically much simpler to perform and has fewer side effects. Another major advantage of the erythropoietin treatment is that it seems to be effective even when started >6 hours but <48 hours after birth, which indicates, at least in part, different mechanisms of action. Erythropoietin treatment, as proposed in our study, should be considered an alternative to or a treatment to be combined with hypothermia. Important questions concern dosage and timing, erythropoietin transfer across the BBB, and the mechanisms of action. Reported neuroprotective doses of recombinant human erythropoietin in animal models range from 1000 to 30 000 U/kg.10,12,24–26 In adult stroke patients, a dose of 33 000 U/day (∼470 U/kg) for 3 consecutive days resulted in improved outcomes at 30 days.15 However, a phase II study administering 40 000 U/day intravenously for 3 days to adult stroke patients was stopped because of an increased mortality rate, which was attributed at least in part to intracranial hemorrhage.27 In the current study, we chose a dose that was in the range (per kilogram of body weight) of the approved doses for anemia treatment and close to that used for treating adult stroke patients. Repeated administration of erythropoietin, at both doses, every other day for 2 weeks resulted in improved neurologic outcomes for newborn infants with HIE. The 2 doses tested (300 and 500 U/kg) had indistinguishable effects and were in the same range as those used for anemia treatment (250–500 U/kg). However, further investigations are required to elucidate the dose-response relationship.
Erythropoietin is larger than the molecular weight threshold for lipid-mediated transport across the BBB. Whether it is able to cross the BBB is an important issue for any systemically administered neuroprotective agent. Animal and human studies with high-dose erythropoietin treatment suggested that recombinant human erythropoietin can enter the brain either through a receptor-facilitated process8 or in a manner dependent on time, dose, and peak serum concentration.28,29 Endogenous erythropoietin levels are known to increase after hypoxic-ischemic insults,30 and the increased erythropoietin concentrations measured after HIE reflect the combined contributions of endogenous and exogenous erythropoietin. However, judging from the time course and from published data, we can safely infer that the increases are at least partly related to erythropoietin administration. First, the time course of erythropoietin concentrations in CSF was mirrored by that in serum. Second, erythropoietin concentrations in serum and CSF in one study were reported to be increased after HIE, compared with control values, but levels peaked ∼3 days after asphyxia, with the erythropoietin concentrations in serum being twice as high as those in CSF.31 In contrast, in our investigation the erythropoietin concentration peaked 3 hours after erythropoietin administration and reached 5- to 10-fold higher levels than before treatment. The erythropoietin concentration in blood was lower than that in a report on premature newborns, showing a peak concentration of 740 mU/mL 4 to 12 hours after administration.32 Because we did not assess erythropoietin concentrations between 3 and 8 hours after erythropoietin administration, the peak concentration might have occurred between those time points. Investigations with labeled erythropoietin would be necessary to elucidate the pharmacokinetic features and relative contributions of exogenous and endogenous erythropoietin. In this study, we noted peak erythropoietin concentrations in CSF ranging from 13.75 to 43.75 mU/mL, 3 hours after subcutaneous administration, similar to or higher than the concentration of 15 mU/mL in a human stroke trial.15 The peak concentrations of erythropoietin in CSF were reported to occur 8 to 24 hours after administration for adults and 4 to 12 hours after administration for preterm infants.29,32,33 These differences may be attributable to differences in brain developmental stages,34 the integrity of the BBB, and/or the administration route and dose.32,35 However, optimal dosage regimens remain to be determined for newborn infants treated for HIE.
Erythropoietin at a dose of 5000 U/kg has been shown to prevent neuronal apoptosis,36 to promote neurogenesis,26 to increase cell proliferation,37 and to reduce the inflammatory response in the brain.38 Apoptotic cell death contributes relatively more to ischemic injury in the immature brain than in the adult brain,39 persists for several weeks after hypoxia-ischemia, and is correlated with the prognosis after brain injury.40–42 Neurogenesis is a slow process and may contribute to functional recovery after brain injury.43,44 These delayed processes of degeneration and regeneration after an insult indicate that neuroprotective therapy for HIE should continue during the recovery phase. Consequently, we think that it was rational to continue treatment for 2 weeks and that this prolonged treatment contributed to the neuroprotective effects observed. We speculate that shorter or longer treatment periods would have resulted in different outcomes.
Erythropoietin has been widely used to prevent anemia of prematurity, without toxic effects even after months of administration.6,7,45 We did not find any obvious hematopoietic effect after 2 weeks of erythropoietin treatment, which may be attributable in part to frequent blood sampling or HIE inducing an intrinsic response with increased endogenous erythropoietin levels, making the contribution of exogenous erythropoietin seem smaller. However, decreased hemoglobin and reticulocyte levels in the control group, compared with unchanged parameters in the erythropoietin group, indicated strongly that administered erythropoietin did induce a hematopoietic response. In adult patients, erythropoietin increased the risk of hypertension and thrombosis through enhanced platelet reactivity.16 In this investigation, we did not find any obvious erythropoietin-related toxicity, and the doses used (300–500 U/kg), with administration for a period of 2 weeks, for term infants seemed to inflict no short-term hazards or risks. Because erythropoietin has been administered to a large number of premature infants for decades, negative side effects outweighing the positive effects observed here seems an unlikely scenario.
We found that systemic administration of recombinant human erythropoietin at a dose of 300 or 500 U/kg every other day for 2 weeks was safe and resulted in improved neurologic outcomes for patients with moderate HIE, as assessed at 18 months of age. We speculate that erythropoietin treatment could be an important adjunct or alternative to therapeutically induced hypothermia.
This work was supported by the Medical Science Academy of Henan, a grant from the Department of Education of Henan Province, the National Natural Science Foundation of China, the Ministry of Education of China (211 Project), and the Swedish Research Links Program through the Swedish International Development Cooperation Agency.
- Accepted March 17, 2009.
- Address correspondence to Changlian Zhu, MD, PhD, Department of Pediatrics, Third Affiliated Hospital of Zhengzhou University, Kangfuqain St 7, Zhengzhou 450052, China. E-mail:
This trial has been registered at www.clinicaltrials.gov (identifier NCT008080704).
Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject:
Erythropoietin has been shown to be neuroprotective after hypoxic-ischemic brain injury in animal models, but there is no report on newborn infants with HIE.
What This Study Adds:
This prospective study found that systemic administration of recombinant human erythropoietin resulted in improved neurologic outcomes after moderate HIE, as evaluated at up to 18 months of age.
- ↵Johnston MV, Nakajima W, Hagberg H. Mechanisms of hypoxic neurodegeneration in the developing brain. Neuroscientist.2002;8 (3):212– 220
- ↵Gunn AJ, Gunn TR, Gunning MI, Williams CE, Gluckman PD. Neuroprotection with prolonged head cooling started before postischemic seizures in fetal sheep. Pediatrics.1998;102 (5):1098– 1106
- ↵Silverman WA, Fertig JW, Berger AP. The influence of the thermal environment upon the survival of newly born premature infants. Pediatrics.1958;22 (5):876– 886
- ↵Ohls RK, Ehrenkranz RA, Das A, et al. Neurodevelopmental outcome and growth at 18 to 22 months' corrected age in extremely low birth weight infants treated with early erythropoietin and iron. Pediatrics.2004;114 (5):1287– 1291
- ↵Brines ML, Ghezzi P, Keenan S, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A.2000;97 (19):10526– 10531
- ↵Casals-Pascual C, Idro R, Gicheru N, et al. High levels of erythropoietin are associated with protection against neurological sequelae in African children with cerebral malaria. Proc Natl Acad Sci U S A.2008;105 (7):2634– 2639
- ↵Bierer R, Peceny MC, Hartenberger CH, Ohls RK. Erythropoietin concentrations and neurodevelopmental outcome in preterm infants. Pediatrics.2006;118 (3). Available at: www.pediatrics.org/cgi/content/full/118/3/e635
- ↵Stohlawetz PJ, Dzirlo L, Hergovich N, et al. Effects of erythropoietin on platelet reactivity and thrombopoiesis in humans. Blood.2000;95 (9):2983– 2989
- ↵Bayley N. Bayley Scales of Infant Development, Second Edition. San Antonio, TX: The Psychological Corporation; 1993
- ↵US Food and Drug Administration. Early communication about an ongoing safety review: epoetin alfa. Available at: www.fda.gov/cder/drug/early_comm/epoetin_alfa.htm. Accessed February 20, 2009
- ↵Chen N, Mao J, Du Y. Erythropoietin levels in serum and cerebrospinal fluid of neonates with hypoxic-ischemic encephalopathy. Chin J Contemp Pediatr.2005;7 (2):107– 111
- ↵Chapel S, Veng-Pedersen P, Hohl RJ, Schmidt RL, McGuire EM, Widness JA. Changes in erythropoietin pharmacokinetics following busulfan-induced bone marrow ablation in sheep: evidence for bone marrow as a major erythropoietin elimination pathway. J Pharmacol Exp Ther.2001;298 (2):820– 824
- ↵Sirén AL, Fratelli M, Brines M, et al. Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci U S A.2001;98 (7):4044– 4049
- ↵Villa P, Bigini P, Mennini T, et al. Erythropoietin selectively attenuates cytokine production and inflammation in cerebral ischemia by targeting neuronal apoptosis. J Exp Med.2003;198 (6):971– 975
- ↵Naylor AS, Bull C, Nilsson MK, et al. Voluntary running rescues adult hippocampal neurogenesis after irradiation of the young mouse brain. Proc Natl Acad Sci U S A.2008;105 (38):14632– 14637
- ↵Juul SE, McPherson RJ, Bauer LA, Ledbetter KJ, Gleason CA, Mayock DE. A phase I/II trial of high-dose erythropoietin in extremely low birth weight infants: pharmacokinetics and safety. Pediatrics.2008;122 (2):383– 391
- Copyright © 2009 by the American Academy of Pediatrics