Umbilical vein interleukin-6 levels in very low birth weight infants developing intraventricular hemorrhage

Brain & Development 27 (2005) 483–487 www.elsevier.com/locate/braindev

Original article

Umbilical vein interleukin-6 levels in very low birth weight infants developing intraventricular hemorrhage Rajeev Kassala, Mujahid Anwarb, Fawaz Kashlana, John Smulianc, Mark Hiattb, Thomas Hegyia,* a

Division of Neonatology, Department of Pediatrics, University of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, One Robert Wood Johnson Place, MEB 312C, New Brunswick, NJ 08903-0019, USA b Division of Neonatology, Department of Pediatrics, St Peter’s University Hospital, New Brunswick, NJ, USA c Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology and Reproductive Sciences, University of Medicine and Dentistry-Robert Wood Johnson Medical School, New Brunswick, NJ, USA Received 17 August 2004; received in revised form 17 December 2004; accepted 20 December 2004

Abstract To assess the relationship between perinatal infection/inflammation as reflected by umbilical vein interleukin-6 (IL-6) levels and the development of periventricular–intraventricular hemorrhage (IVH) in very low birth weight (VLBW) infants, we tested the hypothesis that VLBW infants who develop IVH have higher concentrations of IL-6 in an umbilical vein sample compared to infants without IVH. An inception cohort of 69 VLBW infants was followed from birth until discharge or death to determine the development of IVH by serial neuroultrasounds. Umbilical vein IL-6 levels were measured using commercially available ELISA kit (Endogen Laboratories, Woburn, MA) and compared in IVH and control cohorts. Twenty-two (32%) infants developed IVH, including 18 (82%) with grade I or II and 4 (18%) with grade III or IV. One of these infants also developed periventricular leukomalacia. The umbilical vein IL-6 levels were significantly elevated in infants with IVH with median value of 87 pg/ml (25th percentile value 30 pg/ml and 75th percentile value 310 pg/ml) compared with infants without IVH, with a median value of 0 pg/ml (25th percentile value 0 pm/ml and 75th percentile value 4 pg/ml) (PZ0.007). Umbilical vein IL-6 levels are elevated in neonates who subsequently develop IVH. q 2005 Elsevier B.V. All rights reserved. Keywords: Cytokines; Interleukin-6; Intraventricular hemorrhage; Prematurity; Very low birth weight infant

1. Introduction An inverse relationship exists between gestational age and intraventricular hemorrhage (IVH), with the most immature infants having the highest frequency [13,14]. Infants younger than 28 weeks of gestation have Abbreviations: IL-6, interleukin-6; IVH, periventricular–intraventricular hemorrhage; nl, natural logarithm; pg, picograms; GBS, Group B Streptococcus; RDS, respiratory distress syndrome; PIH, pregnancy induced hypertension/preeclampsia; PDA, patent ductus arteriosus; IUGR, intrauterine growth retardation; NUS, neuroultrasound; IMV, intermittent mandatory ventilation; PPV, positive pressure ventilation; BBB, blood–brain barrier; CSF, cerebrospinal fluid; VLBW, very low birth weight infants (!1501 g birth weight). * Corresponding author. Tel.: C1 732 235 7354; fax: C1 732 235 6609. E-mail address: [email protected] (T. Hegyi). 0387-7604/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2004.12.003

approximately a threefold greater risk of hemorrhage than infants born between 28 and 31 weeks of gestation. Most hemorrhages occur within the first two postnatal days, and virtually all occur within 1 week of birth [10–12]. Although many perinatal and neonatal events have been linked to IVH, the direct cause and pathogenesis of IVH remain elusive. Prematurity has the strongest association with this disorder [1,2,16]. Neonatal hypotension and acidosis are also associated with IVH [18]; however, they may reflect the severity of the infants’ underlying disease, rather than a causal relationship. Recently, cytokines have been implicated in the casual pathway for IVH, with IL-6 the most widely studied. IL-6 levels are elevated in the serum and cerebrospinal fluid of humans with brain injury [7], stroke [17], hypoxic-ischemic encephalopathy [8], meningitis [4] and periventricular

484

R. Kassal et al. / Brain & Development 27 (2005) 483–487

leukomalacia [23]. In recent studies, Yoon and coworkers reported that elevated levels of amniotic fluid IL-6 may be predictive of IVH [22] and Gomez et al. [3] showed that an elevated fetal plasma IL-6 level was an independent risk factor for severe neonatal morbidity, including IVH. Based on this data, we hypothesized that VLBW infants with IVH will have higher levels of umbilical vein interleukin-6 (IL-6) when compared to VLBW infants without IVH.

2. Materials and methods Infants born at St Peter’s University Hospital from July 1998 to April 1999 and weighing below 1501 g at birth were eligible for enrollment into this prospective study, which was approved by the institutional review board of the hospital. After informed consent was obtained, we collected 1.5–2.0 cm3 of blood from the umbilical vein. The blood was centrifuged at 4000 rpm for 5 min and the serum was separated and stored in polypropylene tubes at K70 8C The cord plasma values were not compared to samples from the infants following delivery due to limitations set by the Institutional Review Board. Specimens were thawed and assayed in batches for IL-6 levels using a commercially available ELISA kit (Endogen laboratories, Woburn, MA). The sensitivity of the assay was less than 1 pg/ml of human IL-6 and the intra-assay and inter-assay coefficients of variation were less than 10% each. Samples were assayed in duplicates. Maternal clinical data recorded included demographics, duration of rupture of membranes, GBS status, maternal fever (O100.4 F), presence of labor, use of antenatal steroids, intrapartum antibiotics, histological evidence of chorioamnionitis and mode of delivery with indication. Selected variables that could affect the incidence of IVH in the infants were also studied including birth weight, gestational age, pregnancy induced hypertension, intrauterine growth retardation, respiratory distress syndrome, mechanical ventilation and patent ductus arteriosus with neonatal indomethacin exposure. All infants in the study were monitored for evidence of IVH by neurosonograms using a portable real-time sector scanner interfaced with a 7.5 MHz transducer (Advanced Technology Laboratories, Bothell, Washington). Neurosonograms were performed at 7–10 days and at 3–4 weeks of age on all infants and repeated in infants with IVH as indicated. Screening at 7–10 days is standard in the NICU with the goal of identifying all IVH, most of which are known to occur within the first several days of life. A transfontanelle approach was used and both coronal and sagittal views were obtained. A pediatric radiologist who was unaware of the infant’s clinical course reviewed the ultrasound studies. The grading system for the hemorrhage was adapted from the scheme of Papile et al. [10]: Grade 0, no IVH; Grade I, blood in the periventricular germinal matrix regions; Grade II,

blood within the lateral ventricular system without ventricular dilatation; Grade III, blood within and distending the lateral ventricles; and Grade IV, blood within the ventricular system accompanied by parenchymal extension.

3. Statistical analysis At the time of the analysis, the infants were divided into two groups based on the presence of IVH. Since IL-6 values were not normally distributed, they were analyzed using non-parametric tests and expressed as median and 25th and 75th percentiles. Continuous variables in the two groups were compared using the Mann–Whitney U-test and the c2test was used for comparison of nominal variables. Multivariable logistic regression analysis was used to evaluate which variables were most predictive of IVH. Candidate variables for the regression analysis included those antenatal, intrapartum and immediate peripartum risk factors with a statistical association with IVH of P close to or less than 0.05. A P-value !0.05 was considered statistically significant.

4. Results During the period of the investigation 210 infants weighing below 1500 g were delivered and 69 were successfully enrolled into the study. The reason that only one-third of the eligible subjects were recruited was the lack of availability of the study investigators round the clock inhouse to obtain consent, as according to our protocol consent was obtained prenatally. None of the parents who were approached for consent declined to participate in the study. Of these 69 preterm deliveries, 38 (55%) resulted from preterm labor, 15 (22%) were associated with PIH, 11 (16%) resulted from placental abruption and 9 (13%) had IUGR. During the study period our incidence of IVH in all babies with a birth weight !1501 g who survived more than 7 days of age was 25%. Among the study infants 22 (32%) developed IVH, of which 18 (82%) had grade I–II IVH and 4 (18%) had grade III–IV IVH. There were no significant differences in demographic or birth characteristics between the IVH and control groups (Table 1). However, there was a trend towards lower gestational age in the IVH group (PZ 0.071) and a need for positive pressure ventilation in the delivery room (PZ0.052). Other clinical variables that may affect the incidence of IVH are reported in Table 2. Of note, the general levels of illness was similar in the two groups, with no significant differences found in any of noted determining variables, except for the need for mechanical ventilation (PZ0.022). All infants were initially treated with antibiotics and there were no infants identified to be septic. Umbilical vein IL-6 levels did not correlate with gestational age or birth weight. These levels in relation to

R. Kassal et al. / Brain & Development 27 (2005) 483–487 Table 1 Clinical characteristics of infants with and without intraventricular hemorrhage IVH (nZ22) Birth weight (grams) Gestational age (weeks) Males (%) One minute Apgar Five minute Apgar Caesarean delivery (%) Cord blood pH Cord blood base deficit PPV in delivery room (%)

986C294

No IVH (nZ 47)

P-value

1047C262

0.388

27.0C2.4

28.2C2.7

0.071

12 (54%) 6.5 (3,10)

19 (40%) 6.0 (1,9)

0.305 0.135

8.5 (6,10)

8 (2,10)

0.784

15 (68%)

32 (68%)

0.622

7.31G0.07 K2.7G2.8

7.28G0.11 K2.7G4.8

0.469 0.990

18 (82%)

26 (55%)

0.052

IVH and maternal clinical variables are reported in Table 3 and in the Fig. 1. The levels of IL-6 were significantly higher in the IVH group [87 pg/ml (30,310)] as compared to controls [0 pg/ml (0,4)] PZ0.007. There was a trend towards higher IL-6 values in infants born to mothers with fever and histological chorioamnionitis; however, these differences were not statistically significant. One infant with IVH also suffered from PVL and had an IL-6 level of 365 pg/ml. Using multivariable logistic regression analysis, we determined the best model for predicting IVH with the following variables previously assessed to be potentially relevant on univariable analysis: gestational age (PZ 0.071), positive pressure ventilation in the delivery room Table 2 Selected variables which may affect the incidence of intraventricular hemorrhage

a

IVH (nZ22)

No IV (nZ47)

P-value

4 (18%)

11 (24%)

0.594

2 (9%)

7 (15%)

0.485

17 (77%)

33 (72%)

0.628

18 (82%)

34 (74%)

0.472

20 (91%)

29 (64%)

0.022

14 (64%)

23 (50%)

0.291

10 (45%)

13 (28%)

0.161

Indomethacin treatment of patent ductus arteriosus.

Table 3 Cord blood interleukin-6 levels (Pg/Ml) in relation to intraventricular hemorrhage and other clinical variables related to infection and inflammation Condition

Present

Absent

Pvalue

Intraventricular hemorrhage Rupture of membranes O12 h Maternal GBS colonizationb Maternal fever Preterm labor Histological chorioamnionitis Maternal steroids

87 (30,310) (nZ22)

0a(0,4) (nZ47)

0.007

0 (0,17) (nZ13)

4 (0,37) (nZ56)

0.289

0 (0,32) (nZ6)

0.161

638 (15,1261) (nZ2) 1 (0,98) (nZ38) 30 (0,236) (nZ17)

18 (0,192) (nZ28) 1 (0,37) (nZ67) 3 (0,32) (nZ31) 1 (0,32) (nZ52)

0.163 0.942 0.190

4 (0,37) (nZ51)

1 (0,30) (nZ18)

0.697

Data is expressed as median (25th percentile, 75th percentile). a 0 pg/ml indicates !1 pg/ml, the limit of detection. b Group B Streptococcus (nZtested).

Continuous variables are expressed as meanGSD, ordinal variables (Apgar score) as median (range).

Pregnancy induced hypertension (%) Intrauterine growth retardation (%) Maternal steroids (%) Respiratory distress syndrome (%) Mechanical ventilation (%) Patent ductus arteriosus (%) Indomethacina (%)

485

(PZ0.052), mechanical ventilation in the immediate neonatal period (PZ0.022) and umbilical vein IL-6 levels (PZ0.007). In the multivariable analysis, only umbilical vein IL-6 levels were found to be significantly associated with IVH (P!0.001) [odds ratio 5.0 with 95% confidence interval of 2.0 and 13.0 for each unit change in the value of natural log of IL-6]. In the multivariable analysis, the P-value for gestational age changed to 0.564, for positive pressure ventilation in delivery room to 0.173 and for mechanical ventilation to 0.712. Utilizing a log IL-6 level of two to determine subsequent IVH, we found a sensitivity of 91%, specificity of 83%, a positive predictive value of 71% and a negative predictive value of 95%.

5. Discussion Interleukin-6 is an inflammatory cytokine that may be a marker for neonatal morbidity. Yoon et al. [22] have shown that elevated IL-6 levels in amniotic fluid are associated with an increased risk of neonatal morbidity in pregnancies complicated by preterm labor. Subsequently, the same group reported an association between elevated levels of IL6 in umbilical cord blood and the development of brain white matter lesions, such as, periventricular leukomalacia in the neonates [24]. Weeks et al. [20] showed significantly higher levels of cord blood interleukin-6 with neonatal sepsis, necrotizing enterocolitis, congenital pneumonia, and grades II–IV intraventricular hemorrhage. In an earlier study, our group was unable to show a statistically significant difference in umbilical vein IL-6 levels and IVH in very preterm infants, however, there were too few cases of IVH to adequately address this association [5]. Using a larger cohort of very low birth weight infants for

486

R. Kassal et al. / Brain & Development 27 (2005) 483–487

Fig. 1. Scatter plot of umbilical vein IL-6 levels in relation to IVH. Horizontal bars represent median values, which were 0 for the control group and 87 for the IVH group. For clarity only measurable values (nonzero) are shown.

this study we found significantly higher umbilical vein IL-6 levels in the infants who developed IVH. The source of elevated levels of IL-6 in infants with IVH is not apparent in our study, since only seven out of 22 infants with IVH had histological evidence of chorioamnionitis. Gomez et al. [3] have shown that amniotic fluid infection is present in 10% of the pregnancies with preterm labor and in 30% of the pregnancies after premature rupture of membranes. It is possible that a fetal inflammatory response may occur in the absence of histological chorioamnionitis leading to the production of inflammatory cytokines including IL-6. IVH originates from the blood vessels in the germinal matrix. During fetal development, these vessels grow rapidly and may be prone to a variety of damaging mechanisms, which may, in part, be mediated by cytokines. Placental villous and chorionic cells [6] are a source of proinflammatory cytokines that may gain access to the fetal and neonatal circulation. It is proposed that these cytokines may damage the germinal matrix endothelium by leading to leukocyte adhesion and transendothelial migration [15,21] as well as by intravascular coagulation [19]. It also has been shown that platelets are activated and their aggregation is enhanced by IL-6 [9]. Therefore, blood may leak out of the germinal matrix vessels into the ventricles of the preterm babies, resulting in intraventricular hemorrhage. Previous study from our institution showed that approximately one third of the hemorrhages are already present by 6 h of age [10]. Most of the other hemorrhages occur by 48–72 h of age and ultrasound study by 7–10 days of age will identify over 95% of the hemorrhages. To date, the pathogenesis of intraventricular hemorrhage remains elusive despite a number of studies that have looked at various perinatal and postnatal factors associated with intraventricular hemorrhage. The purpose of our study was to see if fetal inflammatory response, as manifested by elevated IL-6 levels in umbilical venous blood, may injure the germinal matrix vessels so they bleed easily in response to any

perinatal or postnatal stress. In this exploratory study we thus chose a standardized ultrasound protocol at 7–10 days of age to identify almost all of the hemorrhages. Based on information available in the literature, we estimated that a sample size of 65–70 subjects might be enough to see a moderately large effect of fetal inflammatory response on the development of intraventricular hemorrhage. We did not have adequate information to calculate the number of subjects more precisely before starting the study. We have therefore not used any power calculation for sample size estimation. Therefore, the results of this study can only be considered preliminary. Further prospective studies are required to assess the relationship between the development of IVH and fetal inflammatory response more thoroughly. The data from this study can be used to estimate the number of subjects required for such a large prospective study. Our study is characterized by a number of limitations. The small number of infants with severe IVH prevents an analysis of a dose–response relationship between IL-6 levels and IVH severity. The initial neurosonogram obtained at 7–10 days limits our ability to assess hemorrhages more closely, since most will occur at an earlier time period. However, in spite of these issues we conclude that high levels of cord blood IL-6 are associated with IVH in VLBW infants.

References [1] Dykes FD, Lazzara A, Ahmann P, Blumenstein B, Schwartz J, Brann AW. Intraventricular hemorrhage: a prospective evaluation of etiopathogenesis. Pediatrics 1980;66:42–9. [2] Garcia-Prats JA, Procianoy RS, Adams JM, Rudolph AJ. The hyaline membrane disease-intraventricular hemorrhage relationship in the very low birth weight infant: perinatal aspects. Acta Paediatr Scand 1982;71:79–84. [3] Gomez R, Romero R, Ghezzi F, Yoon BH, Mazor M, Berry SM. The fetal inflammatory response syndrome. Am J Obstet Gynecol 1998; 179:194–202.

R. Kassal et al. / Brain & Development 27 (2005) 483–487 [4] Hashim IA, Walsh A, Hart CA, Shenkin A. Cerebrospinal fluid interleukin-6 and its diagnostic value in the investigation of meningitis. Ann Clin Biochem 1995;32:289–96. [5] Kashlan F, Smulian J, Shen-Schwarz S, Anwar M, Hiatt M, Hegyi T. Umbilical vein interleukin-6 and tumor necrosis factor alpha plasma concentrations in the very preterm infant. Pediatr Infect Dis 2000;19: 238–43. [6] Kauma SW, Herman K, Wang Y, Walsh SW. Differential mRNA expression and production of interleukin-6 in placental trophoblast and villous core compartments. Am J Reprod Immunol 1993;30: 131–5. [7] Kossmann T, Hans V, Imhof HG, Trentz O, Morganti-Kossmann MC. Interleukin-6 released in human cerebrospinal fluid following traumatic brain injury may trigger nerve growth factor production in astrocytes. Brain Res 1996;713:143–52. [8] Martin-Ancel A, Garcia-Alix A, Pascual-Salcedo D, Cabanas F, Valcarce M, Quero J. Interleukin-6 in the cerebrospinal fluid after perinatal asphyxia is related to early and late neurological manifestations. Pediatrics 1997;100:789–94. [9] Oleksowicz L, Mrowiec Z, Zuckerman D, Isaacs R, Dutcher J, Puszkin E. Platelet activation induced by interleukin-6: evidence for a mechanism involving arachidonic acid metabolism. Thromb Haemost 1994;72:302–8. [10] Paneth N, Pinto-Martin J, Gardiner J, Wallenstein S, Katsikiotis V, Hegyi T, et al. Incidence and timing of germinal matrix/intraventricular hemorrhage in low birth weight infants. Am J Epidemiol 1993;137:1167–76. [11] Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weight less than 1500 g. J Pediatr 1978;(92): 529–34. [12] Partridge JC, Babcock DS, Steichen JJ, Han BK. Optimal timing for diagnostic cranial ultrasound in low-birth-weight infants: detection of intracranial hemorrhage and ventricular dilation. J Pediatr 1983;102: 281–7. [13] Perlman JM, Volpe JJ. Intraventricular hemorrhage in extremely small premature infants. Am J Dis Child 1986;140:1122–4. [14] Shankaran S, Bauer CR, Bain R, Wright LL, Zachary J. Relationship between antenatal steroid administration and grades

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

487

III and IV intracranial hemorrhage in low birth weight infants. The NICHD neonatal research network. Am J Obstet Gynecol 1995;173: 305–12. Slungaard A, Vercellotti GM, Walker G, Nelson RD, Jacob HS. Tumor necrosis factor-alpha/cachectin stimulates eosinophil oxidant production and toxicity towards human endothelium. J Exp Med 1990;171:2025–41. Szymonowicz W, Yu VY, Wilson FE. Antecedents of periventricular haemorrhage in infants weighing 1250 g or less at birth. Arch Dis Child 1984;59:13–17. Tarkowski E, Rosengren L, Blomstrand C, Wikkelso C, Jensen C, Ekholm S, et al. Early intrathecal production of interleukin-6 predicts the size of brain lesion in stroke. Stroke 1995;26:1393–8. Volpe JJ, Herscovitch P, Perlman JM, Raichle ME. Positron emission tomography in the newborn: extensive impairment of regional cerebral blood flow with intraventricular hemorrhage and hemorrhagic intracerebral involvement. Pediatrics 1983;72:589–601. Wada H, Tanigawa M, Wakita Y, Nakase T, Minamikawa K, Kaneko T, et al. Increased plasma level of Interleukin-6 in disseminated intravascular coagulation. Blood Coagul Fibrinolysis 1993;4:583–90. Weeks JW, Reynolds L, Taylor D, Lewis J, Wan T, Gall SA. Umbilical cord blood interleukin-6 levels and neonatal morbidity. Obstet Gynecol 1997;90:815–8. Westlin WF, Gimbrone Jr. MA. Neutrophil-mediated damage to human vascular endothelium. Role of cytokine activation. Am J Pathol 1993;142:117–28. Yoon BH, Romero R, Kim CJ, Jun JK, Gomez R, Choi JH, et al. Amniotic fluid Interleukin-6: a sensitive test for antenatal diagnosis of acute inflammatory lesions of preterm placenta and prediction of perinatal morbidity. Am J Obstet Gynecol 1995;172:960–70. Yoon BH, Romero R, Kim CJ, Koo JN, Choe G, Syn HC, et al. High expression of tumor necrosis factor-alpha and interleukin-6 in periventricular leukomalacia. Am J Obstet Gynecol 1997;177: 406–11. Yoon BH, Romero R, Yang SH, Jun JK, Kim IO, Choi JH, et al. Interleukin-6 concentrations in umbilical cord plasma are elevated in neonates with white matter lesions associated with periventricular leukomalacia. Am J Obstet Gynecol 1996;174:1433–40.