Christine A. Gleason

A Phase I/II Trial of High-Dose Erythropoietin in Extremely Low Birth Weight Infants: Pharmacokinetics and Safety

OBJECTIVES. High-dose recombinant erythropoietin is neuroprotective in animal models of neonatal brain injury. Extremely low birth weight infants are at high risk for brain injury and neurodevelopmental problems and might benefit from recombinant erythropoietin. We designed a phase I/II trial to test the safety and determine the pharmacokinetics of high-dose recombinant erythropoietin in extremely low birth weight infants. METHODS. In a prospective, dose-escalation, open-label trial, we compared 30 infants who were treated with high-dose recombinant erythropoietin with 30 concurrent control subjects. Eligible infants were <24 hours old, ≤1000 g birth weight, and ≤28 weeks of gestation and had an umbilical artery catheter in place. Each infant received 3 intravenous doses of 500, 1000, or 2500 U/kg at 24-hour intervals beginning on day 1 of age. Blood samples were collected at scheduled intervals to determine recombinant erythropoietin pharmacokinetics. Safety parameters were also evaluated. In the concurrent control group, only clinical data were collected. RESULTS. Mean erythropoietin concentrations 30 minutes after recombinant erythropoietin infusion were 5973 ± 266, 12291 ± 403, and 34197 ± 1641 mU/mL after 500, 1000, or 2500 U/kg, respectively. High-dose recombinant erythropoietin followed nonlinear pharmacokinetics as a result of decreasing clearance from the lowest dosage (17.3 mL/hour per kg for 500 U/kg) to the highest dosage (8.2 mL/hour per kg for 2500 U/kg). Steady state was achieved within 24 to 48 hours. Both 1000 and 2500 U/kg recombinant erythropoietin produced peak serum erythropoietin concentrations that were comparable to neuroprotective concentrations that previously were seen in experimental animals. No excess adverse events occurred in the recombinant erythropoietin–treated infants compared with control infants. CONCLUSIONS. Early high-dose recombinant erythropoietin is well tolerated by extremely low birth weight infants, causing no excess morbidity or mortality. Recombinant erythropoietin dosages of 1000 and 2500 U/kg achieved neuroprotective serum levels.

Erythropoietin Concentrations in Cerebrospinal Fluid of Nonhuman Primates and Fetal Sheep following High-Dose Recombinant Erythropoietin

Erythropoietin (Epo) decreases neuronal injury and cell death in vitro and in vivo. To lay the groundwork for use of Epo as a potential therapy for brain injury, we tested the hypothesis that systemic dosing of high-dose recombinant Epo (rEpo) would result in neuroprotective rEpo concentrations in the spinal fluid of adult and developing animals. This report characterizes the pharmacokinetics of high-dose rEpo in the blood and spinal fluid of juvenile and adult nonhuman primates (n = 7) and fetal sheep (n = 37) following a single injection. Timed blood and spinal fluid samples were collected following rEpo injection. Epo accumulation in spinal fluid was dependent on peak serum concentration and time following injection. We demonstrate that high-dose rEpo was well tolerated and results in neuroprotective concentrations in spinal fluid of adult and developing animal models by 2–2.5 h after injection.

A new model of neonatal stress which produces lasting neurobehavioral effects in adult rats.

Background: During critical care in neonatal intensive care units (NICU), infants experience stressors and treatments that may produce lasting effects on adult health. An animal model simulating the NICU experience is needed to understand the impact of specific neonatal stressors. Objective: We combined approaches to develop a neonatal rat model simulating NICU stressors in order to examine the hypothesis that early stress and morphine sulfate (MS) exposure would affect development and alter adult behavior. Methods: Rat pups were exposed to stressors and given twice daily MS injections (2 mg/kg s.c.) for 5 days (postnatal days 3–7). Stress included daily maternal separation (from 08.00 to 16.00 h), hand feedings, a daily hypoxia/hyperoxia episode (100% N2 for 8 min, then 100% O2 for 4 min), and cold exposure (4°C for 20 min/day). Five treatment groups were formed: (1) ‘control control’ (dam reared and untreated); (2) control vehicle; (3) stress vehicle; (4) control morphine, and (5) stress morphine. Early growth and developmental indices were measured. Adult neurobehavioral tests were paw flick, passive avoidance, and forced swimming. Neonatal MS pharmacokinetics, neonatal and adult corticosterone levels, and adult hematocrit and blood pressure values were measured. Results: Neonatal stress significantly increased the mortality. Neonatal stress and MS treatment slowed early growth. Neonatal MS impaired passive avoidance learning and increased frequency, duration, and distance of forced swimming. There were no differences in corticosterone, hematocrit, or blood pressure values. Conclusions: This model simulates NICU stressors and enables measurement of acute physiological and long-term neurobehavioral indices. Neonatal MS treatment impaired the adult cognitive functioning.

Neonatal Stress or Morphine Treatment Alters Adult Mouse Conditioned Place Preference

Background: Hospitalized preterm infants may experience pain and stress, and narcotics are often administered to lessen their suffering. However, prolonged narcotic therapy may be detrimental during neonatal brain development. Using a rat model combining neonatal stress and morphine, we found that neonatal morphine impaired adult learning. Here we describe a new mouse model examining lasting effects of neonatal stress and morphine. Objective: We tested whether repeated neonatal stress and/or morphine exposure affects early neurodevelopmental or adult behaviors. Methods: Five groups of C57/BL6 mice (1: untreated; 2: morphine (2 mg/kg s.c., b.i.d.); 3: saline, 4: stress + morphine; 5: stress + saline) were treated from postnatal day (P) 5 to P9. Stress consisted of daily maternal separation/isolation (08:00–15:00 h) with gavage feedings and twice daily exposure to brief hypoxia/hyperoxia. Developmental behaviors included righting (P5) and negative geotaxis (P9). Adult behaviors included elevated plus maze, morphine place-preference conditioning, and forced-swimming. Plasma con- centrations of morphine (P7) and corticosterone (P9 and adult) were measured. Results: Neonatal stress or neonatal morphine alone impaired adult place-preference conditioning, but the combination did not (interaction p < 0.01). Adult basal corticosterones were reduced by neonatal morphine treatment. There were no substantial differences in elevated plus maze or forced-swimming times. Conclusions: Neonatal stress and morphine treatment produced long-lasting behavioral and hormonal effects which suggest that neonatal morphine reduces adult arousal and neonatal stress exaggerates adult arousal, each to a degree sufficient to alter learning, while the combined impact of these neonatal treatments does not alter adult learning.

Long-term effects of neonatal stress on adult conditioned place preference (CPP) and hippocampal neurogenesis

Critically ill preterm infants are often exposed to stressors that may affect neurodevelopment and behavior. We reported that exposure of neonatal mice to stressors or morphine produced impairment of adult morphine-rewarded conditioned place preference (CPP) and altered hippocampal gene expression. We now further this line of inquiry by examining both short- and long-term effects of neonatal stress and morphine treatment. Neonatal C57BL/6 mice were treated twice daily from postnatal day (P) 5 to P9 using different combinations of factors. Subsets received saline or morphine injections (2mg/kgs.c.) or were exposed to our neonatal stress protocol (maternal separation 8h/d × 5d+gavage feedings ± hypoxia/hyperoxia). Short-term measures examined on P9 were neuronal fluorojade B and bromodeoxyuridine staining, along with urine corticosterone concentrations. Long-term measures examined in adult mice (>P60) included CPP learning to cocaine reward (± the kappa opioid receptor (KOR) agonist U50,488 injection), and adult hippocampal neurogenesis (PCNA immunolabeling). Neonatal stress (but not morphine) decreased the cocaine-CPP response and this effect was reversed by KOR stimulation. Both neonatal stress or morphine treatment increased hippocampal neurogenesis in adult mice. We conclude that reduced learning and increased hippocampal neurogenesis are both indicators that neonatal stress desensitized mice and reduced their arousal and stress responsiveness during adult CPP testing. Reconciled with other findings, these data collectively support the stress inoculation hypothesis whereby early life stressors prepare animals to tolerate future stress.

Effects of neonatal stress and morphine on murine hippocampal gene expression

Critically ill preterm infants experience multiple stressors while hospitalized. Morphine is commonly prescribed to ameliorate their pain and stress. We hypothesized that neonatal stress will have a dose-dependent effect on hippocampal gene expression, and these effects will be altered by morphine treatment. Male C57BL/6 mice were exposed to five treatment conditions between postnatal d 5 and 9: 1) control, 2) mild stress + saline, 3) mild stress + morphine, 4) severe stress + saline, and 5) severe stress + morphine. Hippocampal RNA was extracted and analyzed using Affymetrix Mouse Gene 1.0 ST Arrays. Single gene analysis and gene set analysis were used to compare groups with validation by qPCR. Stress resulted in enrichment of gene sets related to fear response, oxygen carrying capacity, and NMDA receptor synthesis. Morphine down-regulated gene sets related to immune function. Stress + morphine resulted in enrichment of mitochondrial electron transport gene sets and down-regulation of gene sets related to brain development and growth. We conclude that neonatal stress alone influences hippocampal gene expression, and morphine alters a subset of stress-related changes in gene expression and influences other gene sets. Stress + morphine show interaction effects not present with either stimulus alone. These changes may alter neurodevelopment.

Rising prevalence of gastroschisis in Washington State.

The aim of this study was to assess gastroschisis prevalence in Washington (WA) State in relation to putative risk factors. Gastroschisis prevalence was calculated from the WA State birth cohort during 1987–2006 using an administrative database with birth certificate data linked with hospital discharge records and the ICD-9 procedure code 54.71, which specifies gastroschisis repair. Poisson regression analysis was used to evaluate time trends while adjusting for risk factors. Birth year was included as a linear term. Maternal age, smoking, race, residence in urban versus rural area, geographic region (eastern versus western Washington), paternal age, and infant gender were included as categorical factors. Prevalence ratios were adjusted for birth year and all of the preceding factors. Two hundred and eighty-two infants with gastroschisis were identified. In the adjusted analysis, the prevalence ratio for gastroschisis was 1.1 per year (95% CI 1.08–1.13), indicating an average 10% increase per birth year. Teen mothers were at a higher risk compared to mothers ≥25 yr old (adjusted rate ratio [aRR] 8.02; 95% CI 5.30–12.13), as were teen fathers (aRR 2.35; 95% CI 1.48–3.74) compared to fathers ≥25 years old. Maternal smoking was associated with a higher risk compared to those who were nonsmokers (aRR 1.58; 95% CI 1.19–2.09). Black mothers had a lower risk compared with white mothers. There was no association with geographic classification of mothers residence. Gastroschisis prevalence has increased in WA, particularly in teen mothers and in smokers. This is not explained by a rise in teenage pregnancies or maternal smoking. Further investigation of factors specific to teenage lifestyle is warranted.

Skin Care Management Practices for Premature Infants

OBJECTIVE:To describe current skin care practices for preterm infants in neonatal intensive care units in the United States. We hypothesized that there would be little consensus among facilities.STUDY DESIGN:Neonatal intensive care units (n = 823) listed in the 1996 United States Neonatologists Directory (American Academy of Pediatrics, Section on Perinatal Pediatrics) were sent a 28-question survey dealing with many aspects of neonatal skin care along with descriptive data about their neonatal intensive care unit. Descriptive data analysis was performed.RESULTS:A total of 305 surveys were returned (37% return rate); of these, 241 of the respondents reported admitting infants weighing ≤1000 gm. Some neonatal skin care practices showed wide consensus (>70%) (e.g., scrub procedure for staff; use of a skin barrier under tapes/adhesives), whereas other practices showed little consensus (<30%) (e.g., routine surveillance cultures; use of Aquaphor).CONCLUSION: Consensus on skin care practices was not found among neonatal intensive care units. Data from this survey can be used to develop studies to examine whether certain skin care management practices can improve neonatal outcomes.

Effects of Neonatal Stress and Morphine on Kappa Opioid Receptor Signaling

Background: Critically ill neonates experience multiple stressors during hospitalization. Opioids are commonly prescribed to ameliorate their pain and stress. However, the enduring effects of stress and opioids are not understood. The kappa opioid system is important in the mediation of stress in adults, but little is known about its function in neonates. Objectives: To characterize kappa opioid receptor (KOR) distribution in the neonatal mouse brain and test whether neonatal exposure to morphine, stress, or both, change KOR signaling. Methods: Five groups of wild-type C57BL/6 or prodynorphin (Pdyn) knockout mice were tested: (1) untreated control (dam-reared, no handling), (2) saline-injected control, (3) morphine-injected control, (4) stressed with saline injections and (5) stressed with morphine injections. Mice were treated from postnatal day 5 to postnatal day 9, after which their brains were immunolabeled with a phospho-specific KOR antibody (KOR-P), glial fibrillary acidic protein or glutamic acid decarboxylase. Results: There were no effects of saline or morphine injection on KOR-P immunoreactivity. Neonatal stress increased KOR-P labeling in wild-type brains (p < 0.05), but not in Pdyn–/– animals. Mice exposed to stress and morphine showed region-specific increases in KOR-P immunoreactivity from 38 to 500% (p < 0.05 to p < 0.001), with marked gliosis. In stressed morphine-treated Pdyn–/– animals, KOR-P immunoreactivity was absent, but gliosis increased compared to wild-type animals. Conclusions: Neonatal stress increases KOR activation via the dynorphin system. Neonatal stress plus morphine treatment further increased this response and also resulted in hippocampal gliosis. Enhanced gliosis noted in Pdyn–/– animals suggests that the endogenous dynorphin may play a role in downregulating this inflammatory response.

Cerebrovascular effects of rapid volume expansion in preterm fetal sheep.

Preterm human infants are often treated with volume expansion during their initial stabilization. There are limited data regarding the cerebral vascular effects of this therapeutic approach. The effects of blood volume expansion on cerebral vascular reactivity and oxygen metabolism in very immature animals have not been determined. We examined the effects of volume expansion, with and without hypoxia, on cerebral blood flow and metabolism in unanesthetized, chronically catheterized, preterm fetal sheep. Rapid volume expansion with i.v. dextran increased circulating blood volume. Arterial blood pressure did not increase, nor did cerebral blood flow. However, volume expansion resulted in lower arterial Hb concentration and, consequently, oxygen content without a compensatory increase in cerebral blood flow. Cerebral oxygen delivery fell significantly. Induction of severe hypoxia after volume expansion resulted in an increase in cerebral blood flow, as expected, but the increase in flow was not enough to maintain cerebral oxygen delivery. Rapid volume expansion in normovolemic preterm fetal sheep is associated with decreased cerebral oxygen delivery, and this is further compromised when oxygen content is decreased.