Objectives. To determine the initial inflammatory cytokine response in term infants born to mothers with clinical chorioamnionitis and to assess whether the cytokine response is associated with birth depression, abnormal neurologic examination, and hypoxic-ischemic encephalopathy (HIE).
Methods. Infants who were exposed to chorioamnionitis and admitted to the neonatal intensive care unit (n = 61) were studied prospectively. Cytokine concentrations were measured from umbilical cord blood and at 6 and 30 hours after birth. Control values (n = 50) were determined from cord blood of healthy term infants. Enzyme-linked immunosorbent assays were performed for interleukin (IL)-1β; IL-6; IL-8; regulated on activation, normal T-cell expressed and secreted (RANTES); macrophage inflammatory protein-1α; and tumor necrosis factor-α. Serial blinded neurologic examinations using a modified Dubowitz score were performed simultaneously at 6 and 30 hours.
Results. Cord IL-6 (1071 ± 1517 vs 65 ± 46 pg/mL), IL-8 (2580 ± 9834 vs 66 ± 57 pg/mL), and RANTES (95 917 ± 16 518 vs 54 000 ± 14 306 pg/mL) concentrations only were higher in infants with chorioamnionitis versus control infants. IL-6 increased at 6 hours to 1451 ± 214 pg/mL, followed by a 5-fold decline at 30 hours in contrast to progressive decreases over time in IL-8 and RANTES. There was no relationship between cytokines and birth depression. Modified Dubowitz score correlated with IL-6 at 6 hours (r = 0.5). Infants with HIE/seizures (n = 5) had significantly higher cytokine concentrations at 6 hours versus infants without either (n = 56): IL-6 (3130 vs 1219 pg/mL), IL-8 (5433 vs 780 pg/mL), and RANTES (97 396 vs 46 914 pg/mL).
Conclusions. There was a significant association between abnormalities in the neurologic examination and cytokine concentrations, with the highest cytokines concentrations observed in infants who developed HIE/seizures.
There is increasing evidence supporting an association between placental infection/inflammation in term infants and the development of cerebral palsy in early childhood. This association was first noted in the 1950s by Eastman et al,1 who reported that intrapartum fever was 7 times more common in mothers of infants with cerebral palsy compared with control infants. Interest in this area has been rekindled by the work of Nelson et al,2 who analyzed characteristics of a population registry of infants with cerebral palsy and reported that a clinical or histologic diagnosis of chorioamnionitis was associated with an 8-fold increased risk of cerebral palsy. In addition, studies of stored neonatal blood samples of children with cerebral palsy revealed that concentrations of inflammatory cytokines were increased, suggesting that perinatal inflammatory processes may be important in the cause of cerebral palsy.3 Understanding the association between perinatal infection and cerebral palsy has been limited by retrospective data collection, which precludes assessment of the early neurologic manifestations of neonates exposed to perinatal infection. Although cytokines may represent an important link between perinatal infection and brain injury, evaluation of the neonatal neurologic examination could possibly distinguish this association from other pathways that lead to brain injury. The objectives of this study were 1) to determine the early postnatal changes in inflammatory cytokine concentrations in symptomatic term newborns exposed to clinical chorioamnionitis and 2) to determine whether any of the cytokines studied were correlated with short-term neonatal neurologic outcome, including depression at birth, abnormalities in the neurologic examination, and hypoxic-ischemic encephalopathy (HIE).
This was a prospective cohort study of 61 term infants who had an estimated gestational age of ≥36 weeks and were admitted to the neonatal intensive care unit (NICU) at Parkland Memorial Hospital (Dallas, TX) between July 1999 and December 2001. Of the 23 231 nursery admissions (NICU and newborn nursery) during the study interval, 1660 (7%) term infants had a maternal history of clinical chorioamnionitis. Clinical chorioamnionitis was defined according to obstetrical criteria of maternal fever associated with fetal or maternal tachycardia, uterine tenderness, or foul-smelling amniotic fluid.4 Of the infants born to mothers with clinical chorioamnionitis, 1571 were asymptomatic and triaged to the regular newborn nursery, whereas 89 (5%) were admitted to the NICU and constituted the target population. Of these 89 infants, 18 were triaged to the newborn nursery within 4 hours and consent could not be obtained in 10 infants, resulting in a study population of 61 infants. The study was approved by the Institutional Review Board at the University of Texas Southwestern Medical Center at Dallas. Informed consent was obtained from the parents of eligible infants before enrollment.
Each infant underwent serial neurologic examinations that were performed by the same investigator (L.S.) in the first 12 hours after birth (mean age: 6 ± 5 hours) and a second examination at 24 hours after the initial assessment. Infants with abnormalities in the second neurologic examination had daily examinations performed until normalization or hospital discharge.
Modified Dubowitz Score
A modified Dubowitz score (MDS) was used to quantify the extent of hypotonia,5 and has previously been used by our group.6 The MDS consisted of an assessment of 7 items (posture, arm traction, arm recoil, leg recoil, popliteal angle, head lag, and abnormal movements) and was performed without knowledge of the cytokine concentration. Each item was given a score from 1 to 5, with a higher score indicating hypotonia; a total composite score ranging from 7 to 35 was possible. The test was standardized on healthy term newborns (n = 20) within the first 12 hours after birth, and a value of 15 ± 2 (mean ± standard deviation [SD]) was obtained.
The Sarnat staging7 was used to characterize the degree of the encephalopathy and was simplified as follows: Sarnat 1 (mild encephalopathy) characterized by hyperalertness, irritability, and exaggerated reflexes; Sarnat 2 (moderate encephalopathy) characterized by lethargy, hypotonia, and seizures; and Sarnat 3 (severe encephalopathy) characterized by stupor, flaccidity, and suppression of brainstem and autonomic function. Infants with persistent abnormalities in their neurologic examinations beyond 1 day were further evaluated by neuroimaging studies, as part of clinical care.
Adverse Short-Term Neurologic Outcome
Adverse short-term neurologic outcome was predefined as follows:
Depression at birth: included infants who required bag and mask ventilation for >2 minutes or intubation in the delivery room and/or were given an Apgar score ≤5 at 5 minutes.
Abnormal MDS: an abnormal MDS was defined as 2 SDs above a normal score of 15 (ie, an MDS >19); a higher score was indicative of hypotonia.
HIE and/or seizures: a diagnosis of HIE was defined in accordance with American Academy of Pediatrics/American College of Obstetricians and Gynecologists recommendations8 as the presence of a sentinel perinatal event with all of the following: Apgar score ≤3 at 5 minutes, a cord pH ≤7.00, moderate or severe encephalopathy (Sarnat 2–3), and evidence of non-central nervous system dysfunction manifested by oliguria, hypotension, pulmonary hypertension, or an elevated blood urea nitrogen/creatinine. Infants with seizures, in the absence of HIE, were also included in this category.
Cytokine Immunoassays and Evaluation of Infection
In all infants who were born to mothers with chorioamnionitis, 1 mL of blood was obtained from the umbilical cord of the infant and at 6 and 30 hours of age to coincide with the neurologic examination. Arterial, venous, or capillary blood, whichever was available, was used for postnatal sampling. Samples were centrifuged at 2700 rpm for 10 minutes, and the plasma was stored at −70°C for multiple cytokine immunoassays.
Control values for cytokine assays were obtained from the umbilical cord blood of 50 healthy term newborns delivered by repeat elective cesarean section. These infants were born on the same or consecutive day after enrollment of a patient with chorioamnionitis, did not have evidence of infection, and had Apgar scores ≥7 at 1 and 5 minutes.
Plasma levels of cytokines interleukin (IL)-1β; IL-8; tumor necrosis factor (TNF)-α; regulated on activation, normal T-cell expressed and secreted (RANTES); macrophage inflammatory protein (MIP)-1α; and IL-6 were measured by double sandwich ELISA immunoassay technique using commercial kits specific for human cytokines (R & D Systems, Inc, Minneapolis, MN, for all except IL-6, which was measured with Immunotech, Coulter Co, Paris, France). Detection limits of the assays as indicated by the manufacturers were 4.4 pg/mL for TNF-α, 3.9 pg/mL for IL-1β, 0.35 pg/mL for IL-6, 4.7 pg/mL for IL-8, and <60 pg/mL for RANTES and MIP-1α. The volume of plasma needed for each assay was 50 to 200 μL. The coefficient of variation for both intra- and interassay precision was <10%.
Evaluation for Neonatal Sepsis
All infants who were born to mothers with chorioamnionitis had 2 sets of blood cultures obtained at birth and complete blood counts at 0, 12, and 24 hours as part of the evaluation for infection. The total white cell count was measured with a Coulter S Counter. An immature to total neutrophil proportion was calculated. A ratio >0.17 was considered abnormal.9 Chest radiographs and lumbar puncture were obtained as clinically indicated. Infants with a negative sepsis workup and a hospital course not consistent with sepsis had antibiotics discontinued at 48 hours, whereas infants with a clinical diagnosis of sepsis or pneumonia were treated for at least 7 days.
Placental specimens were available in 26 (43%) of the enrolled infants. Placentas were evaluated within 48 hours for inflammation, using the histologic grading system of Salafia et al.10 Four grades of inflammation were used to assess the amnion, chorion-decidua, umbilical cord, and chorionic plate. Grade 2 inflammation, characterized by multiple foci of 5 or more polymorphonuclear leukocytes or a larger focus in the subchorionic fibrin, was used as the cutoff for clinically important placental inflammation because this grade has been shown to be a sensitive indicator of culture-proven amniotic infection.11
The medical charts of both mother and infant were reviewed, and the following data were collected: complications during pregnancy and during labor, the maternal temperature closest to the time of delivery, mode of delivery, delivery room resuscitation, Apgar scores, initial neonatal temperature, use of cardiovascular medications in the neonate, postnatal complications, days in hospital, and discharge diagnosis.
Mann-Whitney rank sum test was used to compare 2 variables for data that were not normally distributed. Analysis of variance for ranks was used when >2 nonnormally distributed variables were compared. Repeated measures analysis of variance for ranks was used to compare >2 related samples. Correlations were done using the Spearman rank test. All results are presented as mean ± SD and/or median with 25% and 75% quartiles; significance was denoted at P < .05.
The enrolled 61 infants had a birth weight of 3496 ± 88 g and a gestational age of 40 ± 0.2 weeks. The mode of delivery was vaginal for 31 infants (51%) and cesarean section for 30 infants (49%), including 17% that were emergent deliveries. Labor was complicated by meconium staining with fetal heart rate abnormalities in 14 infants (23%). At delivery, 39 infants (64%) required bag and mask ventilation of any duration, 18 (29%) required intubation, and 2 (3%) required chest compressions and epinephrine. An Apgar score ≤5 at 5 minutes was present in 7 infants (11%), and a cord umbilical pH <7.00 was noted in 13 infants (20%). The indications for admission to the NICU were respiratory distress in 48 infants (79%) and birth depression in 13 infants (21%). Mechanical ventilation beyond 24 hours was needed in 7 infants (11%), and 10 infants (16%) required nasal continuous positive airway pressure. No infant had evidence of systemic hypotension, and none required volume replacement therapy in the NICU. Mean maternal temperature was 38.3 ± 0.3°C, and the initial mean neonatal rectal temperature on admission was 38.2 ± 0.4°C (mean ± SD). Abnormal white blood cell count indices were present in 40 infants (66%), and 28 infants (47%) had serial white blood cell abnormalities. No infant had positive blood cultures, but all mothers were treated with antibiotics before delivery. The average duration of hospital stay was 7 days (25%–75% percentile: 3–8 days). The diagnosis of clinical chorioamnionitis was confirmed by placental pathology in 19 of the 26 available placentas (73%).
Umbilical Cord Concentrations
Cytokine concentrations were higher in the cord blood of infants who were born to mothers with chorioamnionitis compared with control infants for IL-6 (1071 ± 1517 vs 65 ± 46 pg/mL), IL-8 (2580 ± 9834 vs 66 ± 57 pg/mL), and RANTES (95 917 ± 16 518 vs 54 090 ± 14 306 pg/mL; Fig 1). No differences were noted for IL-1β, TNF-α, and MIP-1α (Table 1).
The elevated cytokines IL-6, IL-8, and RANTES in infants with chorioamnionitis showed 2 distinct patterns of change over the 3 postnatal sampling times (Fig 2). First, IL-6 concentration increased at 6 hours from 1071 ± 1517 to 1451 ± 214 pg/mL, followed by a 5-fold decline at 30 hours to 280 ± 76 pg/mL. Each value differed from the other time points (P < .01). Second, IL-8 and RANTES concentrations progressively declined over time. Thus, IL-8 decreased from 2580 ± 9834 to 1219 ± 750 pg/mL at 6 hours and to 742 ± 450 pg/mL at 30 hours. Both values differed from the umbilical cord value (P < .01). Similarly, RANTES declined at 6 hours from 95 917 ± 16 518 to 50 000 ± 4000 pg/mL and to 41 900 ± 3000 pg/mL at 30 hours. Only the latter concentration differed from the umbilical cord value (P < .01). Postnatal values for IL-1β, TNF-α, and MIP-1α did not differ from the umbilical cord values. Postnatal concentrations of IL-6, IL-8, and RANTES were not correlated with Apgar scores, neonatal temperature, immature to total neutrophil proportion, and days in the hospital. A weak correlation between cord pH and IL-6 at 6 hours was noted (r = 0.3; P < .05). However, after exclusion of the 4 infants with HIE, that correlation became nonsignificant (r = 0.1; P = .6). In the 26 infants with available placental pathology, there were no differences in cord or postnatal concentrations for any of the cytokines studied between the 19 infants with and the 7 infants without histologic chorioamnionitis.
Short-Term Neurologic Outcomes
Birth depression was present in 23 infants (38%), and an MDS >19 was present in 8 infants (13%) at 6 hours and persisted at 30 hours in 4 infants (7%). All infants with an abnormal MDS demonstrated similar findings characterized by a decrease in spontaneous movements, proximal weakness, head lag, reduced arm recoil and traction, and incomplete hip flexion. These abnormalities normalized after 24 hours, except in the 4 infants with a diagnosis of HIE. In these infants, the encephalopathy was characterized as severe in 2 infants and moderate in the remaining 2 infants. The 2 infants with severe encephalopathy both had generalized tonic clonic seizures associated with burst suppression on the electroencephalogram recording and magnetic resonance imaging (MRI) evidence of marked edema with diffuse bilateral infarctions. Both infants died after intensive care was withdrawn. Placental pathologies in both cases revealed acute funisitis and chorioamnionitis. For the infants with moderate encephalopathy, the hospital course was complicated by severe persistent pulmonary hypertension in 1 infant, requiring prolonged sedation and mechanical ventilation. These factors prevented a determination of the exact length of the encephalopathy, but generalized seizures were evident on day 1 and there was subsequent MRI evidence of diffuse infarction. The second infant did not exhibit clinical seizures, and the MRI was normal. This infant improved by the second day, and the neurologic examination was normal on discharge. The 1 infant with seizures in the absence of HIE had a generalized seizure at 1 hour of age that lasted 15 minutes and was accompanied by mouthing, sucking, and posturing and was associated with lateral eye deviation and desaturation. The cause of the seizure was not determined. Thus, the Apgar scores were 8 and 9 at 1 and 5 minutes, respectively; cord umbilical arterial pH was 7.2; and blood sugar, serum electrolytes, cerebrospinal fluid (CSF) studies, and MRI all were normal. This infant exhibited initial diffuse hypotonia that improved rapidly and was considered normal by the second postnatal day.
Relationship Between Cytokine Concentrations and Short-Term Neurologic Outcomes
Cytokine concentrations were comparable in infants with or without birth depression (P = .3). For the MDS, only the IL-6 concentration at 6 hours showed a positive correlation (r = 0.5; P = .01; Fig 3). The correlation remained significant even after the 4 infants with HIE were excluded (r = 0.4; P = .01). However, there was no correlation between levels of IL-6 and MDS at 30 hours (r = 0.1; P = .6), consistent with the decline in all cytokine concentrations at 30 hours after birth. The 5 infants with HIE/seizures had significantly higher concentrations of IL-6, IL-8, and RANTES at 6 hours postnatally, as compared with the 56 infants who were born to mothers with chorioamnionitis without HIE or seizures (IL-6: 3130 ± 1218 vs 1219 ± 182 pg/mL; IL-8: 5433 ± 4860 vs 786 ± 509 pg/mL; RANTES: 97 392 ± 2607 vs 46 914 ± 3868 pg/mL; P < .05; Fig 4). However, there were no differences in cytokine concentrations in the cord blood or at 30 hours postnatally between infants with or without HIE. When the data for the 56 infants without HIE were compared with those for control infants, significant differences were again noted at all time intervals (P < .01).
To begin to understand the potential association between placental infection and/or inflammation in term infants and subsequent brain injury, we used a cohort design to study symptomatic term infants who were born to mothers with clinical chorioamnionitis. This investigation was prospective, used a standardized neurologic assessment tool at predefined time intervals, and had predetermined short-term adverse neurologic outcomes. The salient observations of this study were the following. First, concentrations of specific cytokines were elevated in the umbilical cord samples as compared with healthy control subjects. Second, the pattern of postnatal change for the elevated cytokines was not uniform. Thus, IL-6 concentration peaked at 6 hours, before declining at 30 hours, whereas IL-8 and RANTES concentrations showed a stepwise decline with time. Third, a direct relationship was present between an abnormal neurologic examination and IL-6 concentrations at 6 hours of age. Fourth, infants with the most abnormal neurologic examination, ie, those infants with HIE and/or seizures, had the highest concentration of IL-6, IL-8, and RANTES at 6 hours of age.
In this study, findings of elevated cord and postnatal IL-6 and IL-8 concentrations in infants who were exposed to chorioamnionitis are consistent with previous reports of elevated inflammatory cytokines in newborns with early-onset sepsis and positive blood cultures12–15 or in newborns who are exposed solely to chorioamnionitis.16 RANTES and MIP-1α are chemokines that activate leukocytes but have not been previously studied during perinatal infection. The values for RANTES and MIP-1α in this report for the control group are consistent with the values reported for healthy term infants.17 Infants who were exposed to chorioamnionitis in this cohort had a 2-fold increase in RANTES, with no change in MIP-1α concentrations compared with control subjects. TNF-α and IL-1β were not increased in the infants that were exposed to chorioamnionitis in the present study. This is consistent with the observations that these 2 acute-phase reactants are elevated only in the presence of systemic infections that cause cardiopulmonary impairment and/or shock.18, 19
There were different postnatal temporal patterns for IL-6 when compared with IL-8 and RANTES. This could reflect differences in time response to infection, different half-lives, and complex feedback inhibitions among these 3 cytokines.20,21 The peak concentration of IL-6 at 6 hours suggests ongoing neonatal production and not maternal transfer or feto-placental production. This is supported by studies that assessed maternal concentrations of cytokines immediately before delivery and noted a 100-fold higher cytokine concentrations in neonatal blood compared with maternal samples.15,16 The observed 2- to 5-fold subsequent decline in IL-6 and IL-8 concentrations is consistent with previous reports that showed declining levels at 24 to 48 hours even in the presence of ongoing infection.22,23
The direct correlation between IL-6 concentrations and increasing hypotonia noted at 6 hours but not at 30 hours postnatally strongly suggests that IL-6 is associated with transient abnormalities of the neurologic examination. There are a number of potential explanations for this relationship. First, elevated blood cytokine concentrations could directly affect brain function, because cytokines can cross the blood brain barrier.24–26
The proximal weakness noted in the infants corresponds topographically to the watershed distribution within the parasagittal region of the cerebral cortex and raises the possibility that reduction in cerebral blood flow may have contributed to the hypotonia. However, there is no evidence that cytokines can directly alter cerebral vascular resistance. Cytokines could, however, lower cerebral blood flow via systemic effects such as hypotension. The latter was generally not present in our patient population, with the exception of 2 infants with HIE who required a fluid bolus in the delivery room. Second, the observed hypotonia could be related to an acute interruption in placental blood flow, but the absence of a correlation between the cord pH and the MDS in the absence of HIE is not supportive of this mechanism. Third, hypotonia could also represent a nonspecific manifestation of illness, without a causal relationship to cytokine concentrations.
Although placental pathology was not consistently available, it is of interest that the 2 infants with the most severe injury both had evidence of funisitis. Funisitis is considered to be a marker of the fetal inflammatory response and has been associated with high IL-6 levels in both preterm and term infants and echodense lesions on cranial ultrasound imaging in preterm infants.27–30
The highest concentrations of cytokines were present in infants with HIE or seizures, suggesting that cytokines may indeed play an important role in the cascade of events that lead to brain injury. The possible involvement of cytokines in the pathogenesis of brain injury supports the observations of Nelson et al3 of elevated blood cytokines during the neonatal period in infants who subsequently developed cerebral palsy. Cytokines may cause irreversible brain injury via several mechanisms. First, cytokines can have a direct toxic effect via increased production of nitric oxide synthase, cyclooxygenase, and free radicals and excitatory amino acid release.31,32 Moreover, they have been shown to cause astrogliosis with white matter encephalomyelitis when injected intraperitoneally in various animal models.33 Second, cytokines may lead to brain injury in the setting of perinatal infection through a systemic inflammatory response known as the systemic inflammatory response syndrome. This is a fulminant inflammatory reaction from release of systemic cytokines with sepsis and results in septic shock, multiple organ dysfunction, and encephalopathy.34, 35 A third mechanism may be via hypoxia ischemia, which is known to be a potent stimulus for brain production of cytokines, even in the absence of infection. Expression of IL-1β and TNF-α has been demonstrated 1 to 4 hours after hypoxia ischemia.36 This is followed by neutrophil invasion and the subsequent development of necrosis and infarction.37 Furthermore, administration of receptor antagonists or neutralizing antibodies to various proinflammatory cytokines has been shown to reduce the extent of ischemic and excitotoxic damage in neonatal animal models.38–40 Moreover, higher concentrations of IL-6, IL-8, and TNF have been noted in the CSF of human newborns with perinatal asphyxia compared with infants without asphyxia and have been correlated to the degree of encephalopathy and the severity of subsequent neurodevelopmental outcome.41–43 The peak concentrations of IL-6, IL-8, and RANTES observed at 6 hours after birth in infants who developed HIE or seizures in the present study is temporally consistent with the timing of cytokine expression in animal models of hypoxic-ischemic injury,36 assuming the presence of a perinatal stimulating event.
In the present study, a spectrum of neurologic findings that ranged from an abnormal examination to transient hypotonia and finally HIE/seizures were observed and were paralleled by a progressive amplification of blood cytokine concentrations. These observations suggest that perinatal infection and HIE may be linked in 2 ways: infection could predispose to HIE, or, alternatively, HIE and intrauterine infection may be simultaneous conditions. We speculate that cytokines may represent the final common pathway leading to brain injury that bridges these 2 conditions. The results of this study provide some estimate of the contribution of perinatal infection to the development of HIE. At our institution, the incidence of moderate and severe HIE is relatively constant and approximates 0.5 to 0.75 per 1000 term infants who are admitted to our nurseries. During the 18-month study interval, 12 infants presented with moderate to severe HIE, 4 of whom were exposed to clinical chorioamnionitis. Thus, perinatal infection may have contributed to one third of the observed cases of moderate to severe HIE at birth. However, this may be an underrepresentation because not all placentas were examined and occult chorioamnionitis may have also contributed to HIE.
The findings of this study point to an association between abnormalities in the neurologic examination and cytokine concentrations, with the highest cytokine concentrations noted in the infants with the most abnormal examinations. However, the precise mechanism that links perinatal infection and brain injury remains unclear. Future studies are planned to assess the association of blood, as well as CSF cytokine concentrations, in infants with hypoxia ischemia, with and without evidence of perinatal infection. Finally, these observations may have important therapeutic implications for neuroprotective strategies in infants with HIE, because the response to therapy might differ depending on the presence or absence of perinatal infection.
- ↵Eastman NJ, De Leon M. The etiology of cerebral palsy. Am J Obstet Gynecol. 955;69 :950– 958
- ↵Cunningham FG, Mac Donald PC, Gant NF, Leveno KJ, Gilstrap LC. Williams Obstetrics. 19th ed. Norwalk, CT: Appleton & Lange; 1993:863
- ↵Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress: clinical and electrographic presentation. Arch Neurol.1976;33 :669– 705
- ↵American Academy of Pediatrics/American College of Obstetricians and Gynecologists. Guidelines for Perinatal Care. 4th ed. Elk Grove Village, IL: American Academy of Pediatrics; 1997:122
- ↵Salafia CM, Weigl C, Silberman L. The prevalence and distribution of acute placental inflammation in uncomplicated term pregnancies. Obstet Gynecol. 89;73 :383– 389
- Frantz A, Steinbach G, Kron M, et al. Reduction of unnecessary antibiotic therapy in newborn infants using interleukin 8 and C-reactive proteins as markers of bacterial infection. Pediatrics.1999;104 :457– 453
- ↵Berner R, Nyemeyer CM, Leititis JU. Plasma levels and gene expression of granulocyte colony stimulating factor, tumor necrosis factor, interleukin (IL)1β, IL 6, IL 8 and soluble intracellular adhesion molecule-1 in neonatal early onset sepsis. Pediatr Res.1998;44 :469– 477
- ↵Sullivan SE, Staba SL, Christensen RD. Circulating concentrations of leukocytes chemokines in preterm and term neonates and in cord blood. Pediatr Res.2000;47 :279
- ↵Schindler R, Mancilla J, Endres S, et al. Correlations and interactions in the production of IL-6, IL-1 and TNF in human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood.1990;75 :40– 47
- ↵Buck C, Bundshu J, Gallati H, et al. Interleukin-6: a sensitive parameter for the early diagnosis of neonatal bacterial infection. Pediatrics.1994;93 :54– 58
- ↵Hack EC, Groot ER, Berma JF, et al. Increased plasma levels of interleukin-6 in sepsis. Blood.1989;74 :1704– 1710
- ↵Banks WA, Ortiz L, Plotkin SR, et al. Human IL-1α, murine IL-1α and murine IL-1β are transported from blood to brain in the mouse, by a shared saturable system. J Pharmacol Exp Ther.1991;259 :988– 996
- Banks WA, Kastin AJ, Gutierrez EG. Penetration of interleukin-6 across the murine blood brain barrier. Neurosci Lett.1994;163 :53– 56
- Naccasha N, Hinson R, Montag A, Ismail M, Bentz L, Mittendorf R. Association between funisitis and elevated interleukin-6 in cord blood. Obstet Gynecol.2001;972 :220– 224
- Kim CJ, Yoon BH, Romero R, et al. Umbilical arteritis and phlebitis mark different stages of the fetal inflammatory response. Am J Obstet Gynecol.2001;1852 :496– 500
- ↵Hagberg H, Gilliaud E, Bona E, et al. Enhanced expression of IL-1 and IL-6 m-RNA and bioactive protein after hypoxia ischemia in neonatal rats. Pediatr Res.1999;40 :603– 609
- ↵Yamasaki Y, Shoozurhara H, Onodera H, et al. Blocking of IL-1 activity is a beneficial approach to ischemia and brain edema formation. Acta Neurochir.1995;60 :S300– S302
- ↵Oygur N, Sonmez O, Saka O, et al. Predictive values of plasma and cerebrospinal fluid TNF α and interleukin 1β concentrations on outcome of full term infants with hypoxic ischemic encephalopathy. Arch Dis Child Fetal Neonatal Ed.1998;79 :F190– F193
- ↵Martin-Ancel A, Garcia A, Pascual D, et al. Interleukin-6 in the cerebrospinal fluid after perinatal asphyxia is related to early and late neurological manifestations. Pediatrics.1997;100 :789– 794
- Copyright © 2002 by the American Academy of Pediatrics