Research Paper

Mediators of Inflammation, 7, 309–312 (1998)

SERUM levels of IL-1b , IL-6 and TNF-a were m easured in 48 healthy, term ed neonates on the 1st (N1), 5th (N5) Inflammatory cytokines in newborn and 40th (N40) day after birth, com pared with th ose infants in m aternal serum (MS), um bilical cord (UC) and adult controls. Cytokine values in N1 and N5 were significantly elevated, th an those in UC and in CA controls (P<0.0001). IL-1b and IL-6 declined sig- A. Sarandakou, G. Giannaki, nificantly from N1 to N40 (P<0.0001), while TNF-a A. Malamitsi-Puchner, D. Rizos, E. Hourdaki, in creased significantly from N1 to N5 and declined E. Protonotariou and I. Phocas thereafter. MS` IL-1b and IL-6, but not MS` TNF-a , w ere significantly higher than those of controls (P<0.0001). IL-1b values depended on the m ode of Hormonological Laboratory and Neonatal Unit, 2nd delivery. In conclusion, th e in creased concentrations of IL-1b , IL-6 and TNF-a durin g th e perin atal period Department of Obstetrics and Gynaecology, m ight suggest th eir involvem ent in an in flam m ation- ‘Areteion’ University Hospital, Athens, Greece like process durin g norm al parturition, and reflect also a newborn im m une response to th e stress of delivery and environm ental changes. CACorresponding Author Tel/Fax: (+30) 017 286 229 Key words: IL-1b , IL-6, TNF-a , Neonatal serum, Maternal serum, Umbilical cord, Labour, Neonatal immune response, Perinatal period, Early neonatal life, Normal values of serum inflammatory cytokines in neonates

Introduction Patients and methods

The inflammatory cytokines, interleukin-1b (IL-1b ), This study w as approved by the Ethics Committee of interleukin-6 (IL-6) and tumour necrosis factor-a the Areteion Teaching Hospital. Follow ing informed (TNF-a ) are important mediators of host response to maternal consent, peripheral blood samples w ere stress and infection. All three cytokines have already collected from 48 healthy, termed neonates (clinical been found in cells, tissues and fluids associated w ith characteristics in the Table 1), during the 1st (N1), the pregnanc y.1–5 It has been hypothesized that IL-1b may 5th (N5) and the 40th (N40) days after birth. In 25 of play a significant role in the embryogenesis, develop- these cases, maternal serum (MS) during the first ment and differentiation of the fetus by regulating the stage of labour and umbilical cord (UC) serum in the grow th and differentiation of tissues and/or organs.2 second stage of labour w ere also obtained. Serum Moreover, IL- 1b , stimulating prostaglandin biosyn- samples from 25 healthy adults w ere also analysed thesis by intrauterine tissues, may signal the onset of simultaneously, as controls. labour and also strongly upregulate IL-6 production. Clinical examination of neonates, their mothers and In term pregnancy IL-6 may orchestrate biochemical, controls, as w ell as serum reactive protein (CRP) immunological and physiological changes that con- determinations w ere performed and all subjects were tribute to maternal and fetal survival.3 In addition, judged infection-free. TNF-a , detected in the placenta, amniotic fluid and Blood samples, collected in pyrogen-free tubes, decidua, may contribute to the regulation of cell w ere centrifuged at once after clotting and stored proliferation at the uteroplacental unit.6,7 aliquotted in –30°C until assay. All three cytokines Elevated values of IL-1b , IL-6 and TNF-a during both term and preterm labour have already been reported 2,3 ,8 –11 in amniotic fluid samples, in maternal Table 1. Clinical characteristics of the studied neonates serum7,1 2 –14 and in cervicovaginal secretions;15 –16 but, little is know n about the presence of these Sex: Females n = 28 inflammatory cytokines in early neonatal life. For this Males n = 20 reason, our aim w as to study the concentrations of IL- Birthweight (mean, range) 3322g (2300–4380g) Weeks of pregnancy (mean, range) 40 (37–42) 1b , IL-6 and TNF-a in new born infant serum on the Mode of delivery first days after birth, compared w ith those in maternal elective caesarean section n = 13 and umbilical cord serum during parturition , in order urgent caesarean section n = 6 vaginal delivery n = 29 to investigate the changes of all three inflammatory Apgar scores 9–10 cytokines during the perinatal period. 0962-9351/98/050309-04 $9.00 © 1998 Carfax Publishing Ltd 309 A. Sara nda ko u et al. w ere determined using solid phase sandw ich enzyme (P<0.055), the low est being found in the cases immunoassays. Sensitivity, interassay CV% and trade w ith elective caesarean section and the highest in names were, for IL-1b : 4.3 pg/ml, 8.2, Biokine® IL-1b ; those w ith an urgent one. for IL- 6: 0.7 pg/ml, 4.4, Biotrak IL-6 Amersham Life d A significant correlation w as observed betw een IL- Science; and for TNF-a : 3.0 pg/ml, 8.2, TNF-a EAS- 1b values in MS and UC (r=0.77; P<0.001), in MS IATM, Medgenix Diagnostics. and N1 (r=0.6; P<0.009), in MS and N5 (r=0.5; Data w ere analysed using non parametric methods, P<0.03) and especially in N1 and N5 (r=0.9; as they did not show a normal distribution (Kolmo- P<0.0001). No correlation w as found betw een gorov-Smyrnov test). A Wilcox on paired or non paired either IL-6 or TNF-a values in the same samples. test, as appropriate, was used in comparing the d Serum values of IL-1b and TNF-a in N1 were distribution of cytokines in the different serum strongly interrelated (r=0.7; P<0.01). samples, w hile Kruskal-Wallis ANOVA w as used to compare the cytokine values among the three groups Discussion according to the mode of delivery. Correlations were evaluated w ith Spearman rank correlation coefficients. The patterns of inflammatory cytokines IL-1b , IL-6 and TNF-a in new born infants during the perinatal Results period demonstrated significant changes in cytokine values, related to the time after birth. TNF-a presente d Cytokine concentrations, as median values and ranges, significantly elevated umbilical cord values, compared in the different serum samples are given in Table 2. w ith those in controls and in maternal serum, w ith a significant increase in neonatal serum on the 1st and d All three cytokines w ere significantly elevated in further on the 5th day after birth, declining slow ly the neonatal samples N1 and N5, compared w ith thereafter. The concentrations of IL-1b and IL-6 in those found in UC and in adult controls umbilical cord w ere also higher, marginally though, (P<0.0001). than those in controls, increasing markedly further on d Neonatal serum values of IL-1b and IL-6 declined the 1st day after birth and declining thereafter, until significantly from N1 to N40 (P<0.0001), w hile normalization 40 days later. those of TNF-a increased significantly from N1 to These data, concerning elevated values of all three N5 (P<0.0001) and declined thereafter. inflammatory cytokines in healthy new born infants, d Umbilical cord values of TNF-a were significantly are the new findings of this study and may possibly higher than those in controls (P<0.0001), but reflect an immune response of neonates to the stress low er than in N40. The UC values of IL-1b and IL-6 of delivery and environmental influences during the w ere higher, not significantly though, than those in early neonatal life. This hypothesis is also supported adult controls and in neonatal samples N40. by the dependence of IL-1b values in UC and N1 on d Maternal serum IL-1b and IL-6, but not TNF-a value s the mode of delivery and the elevation of cytokine w ere significantly higher than those in controls values in UC, mainly of TNF-a , compared w ith those (P<0.0001). in controls. d Maternal serum IL-1b values were also significantly To the best of our know ledge, serum levels of these higher than those in N1 (P<0.0001), w hile those of inflammatory cytokines in healthy, non-infected neo- IL-6 and TNF-a w ere significantly low er nates during the early neonatal period are reported (P<0.0001). here for the first time. Since all these pregnancies d Serum IL-1b values w ere strongly dependent on the w ere uneventful and led to the birth of healthy, mode of delivery in UC (P<0.03), in N1 (P<0.001), termed new borns, these levels may be considered to in MS (P<0.02), and marginally in N5 samples be a good reflection of the normal range.

Table 2. Cytokine values (median, range) in maternal serum (MS), umbilical cord (UC) and newborn infant serum, in the 1st (N1), 5th (N5) and 40th (N40) days after birth and in adult controls

Samples n IL-1b Significance IL-6 Significance TNF-a Significance pg/ml P* pg/ml P* pg/ml P*

MS 25 188 (15–464) 0.0001 5.7 (0.9–68) 0.0001 5.8 (2.5–10) NS UC 25 20.6 (0.0–148) 0.055 2.1 (0.7–36.0) 0.06 14.5 (8–19) 0.0001 N1 48 58.0 (14.5–240) 0.0001 21.0 (6.0–680) 0.0001 29.2 (17–54) 0.0001 N5 48 30.0 (0.0–160) 0.0001 6.7 (2.4–10.5) 0.0001 42.0 (18–79) 0.0001 N40 13 0.8 (0.0–80) NS 1.07 (0.7–11) NS 30.0 (27–44) 0.0001 Controls 25 14.0 (0.0–76) 1.8 (0.5–2.6) 4.8 (0.0–1.3)

*Compared with controls.

310 Mediators of Inflammation · Vol 7 · 1998 IL-1b , IL-6 and TNF-a in newbo rn infa nts

It has been hypothesized that the presence of branes, maternal serum cytokine values reflect func- increased TNF-a values in pyar and breast milk may tions that take place in the gestational tissues and induce the developmental and maturational processes depend on the permeability of these tissues by these of the immune system in neonates.17,18 How ever, our proteins. Thus, our elevated maternal serum IL-6 and TNF-a values in umbilical cord showed a significant low TNF-a are in agreement w ith a previous report,27 elevation already during birth, before breast feeding, regarding the permeability of fetal membranes by strengthening our suggestion, that normal term these cytokines. In contrast, our results on increased labour itself, might constitute an inflammation-like maternal serum IL-1b values do not agree w ith the process, that promotes the activation of the new born latter study. immune system. In conclusion, the presence of increased inflamma- Similarly, w e have already reported elevated serum tory cytokines IL-1b , IL-6 and TNF-a during the early values of soluble interleukin-2 receptors (sIL-2R) in neonatal period (a) suggests their contribution to the umbilical cord and in serum samples from healthy immune system alterations that may signal or propa- term neonates on the 1st and 5th < day after birth, gate the onset of parturition and (b) reflects also a increasing progressively from the 1st to the 3rd neonatal immune response to the stress of delivery sample and strongly dependent on the mode of and environmental changes in the first days after delive ry.19 The elevation of this cytokine, a character- birth. istic marker of immune activation, w as attributed to the dynamic expansion of the neonatal immune References system, in response to environmental changes, during 1. Miller LC, Isa S, LoPreste G, Schaller JG, Dinarello CA. Neonatal and after birth in the early neonatal period. This interleukin-1 beta, interleukin-6, and tumor nec rosis fac tor: c ord blood phenomenon w as also reflected in the elevation of levels and c ellular productio. J Ped ia tr 1990; 117: 961–965. 2. Tsunoda H, Tamatani T, Oomoto Y, et al. Changes in interle ukin –1 levels other grow th factors, as w e have previously reported in human amniotic fluid w ith gestational age s and de livery. Micro bio l 20 for angiogenin levels in the perinatal period. Im m u no l 1990; 34: 377–385. 3. Romero R, Avila C, Santhanam U, Sehgal PB. Amniotic fluid interleukin 6 We have recently also found, how ever, soluble in preterm labor. Assoc iation w ith infe ction. J Clin Inves t 1990; 85: intercellular adhesion molecule-1 (sICAM-1) levels in 1392–1400. 4. Gunn L, Hardiman P, Tharmaratnam S, Low e D, Chard T. Measureme nt of neonatal serum samples on the 1st day of life to be interleukin-1 alpha and inte rleukin-6 in pregnancy-associate d tissues. low er than those in controls, though increasing Repro d Fertil De velo p 1996; 8: 1069–1073. 5. Olah KS, Neilson JP, Johnson PM. Interleukin-6 in amniotic fluid at progressively thereafter and exceeding those of con- forew ater amniotomy compared w ith hindw ater sample s in w omen in 21 trols 1 month after birth. These differences in spontaneous labour. Eu r J Obs te t Gy n eco l Repro d Bio l 1995; 60: 65–67. expression of cytokines during parturition may be 6. Jaatte la M, Kuusela P, Saksela E. Demonstration of tumor necrosis fac tor due to their differentiated roles in the cytokine in human amniotic fluids and supernatans of placental and dec idual tissues. La b Inves t 1988; 58: 48–52. cascade during inflammation and immune process. 7. Opsjon SL, Wathen NC, Tingulstad S, e t a l. Tumor necrosis factor, Moreover, in agreement w ith previous interleukin-1 and interle ukin-6 in normal human pregnanc y. Am J Obs tet 7,1 8,2 2 – 24 Gy n ecol 1993; 169: 397–404. reports, this study demonstrated very 8. Rivero-Marc otegui A, Larranaga-Azc arate C, Ceres-Ruiz R, Carc ia-Merlo S. increased values of IL-1b and IL-6, but not of TNF-a , in Polymorphonuclear elastase and inte rleukin-6 in amniotic fluid in preterm labor (letter). Clin Chem 1997; 43: 857–859. maternal circulation during the first stage of labour. 9. Coultrip IL, Lien JM, Gome z R, Kape rnick P, Khoury A, Grossman JH. The The values of IL-1b were strongly related to the mode value of amniotic fluid interleukin-6 dete rmination in patients w ith preterm labor and intact membranes in the detec tion of microbial of delivery, a finding consistent w ith the dominant invasion of the amniotic cavity. Am J Obs tet Gy n ecol 1994; 171: 1– 3 role of this cytokine in the initiation of parturition. 901–911. 10. Ghidini A, Je nkins CB, Spong CY, Pezzulo JC, Salafia CM, Eglinton GC. The elevation of maternal serum IL-1b and IL-6 values Elevated amniotic fluid interleukin-6 levels during the early se cond seems to reflect a systemic reaction in the mother to trimester are associate d w ith greate r risk of subsequent preterm delivery. Am J Reprod Im m u n ol 1997; 379: 227–231 her fetus and might be analogous to the cytokine 11. Romero R, Mazor M, Sepulveda W, Avila C, Copeland D, William J. Tumor increase observed in the serum of renal transplant nec rosis factor: in term and preterm labor. Am J Obs tet Gy ne co l 1992; 25 166: 1576–1587. recipients, before acute rejection of the graft. The 12. Lahan N, Rice GE, Bishop GJ, Hansen MB, Bendtgen K, Bre nnecke SP. final increase at the onset of labour could be a similar Elevated plasma interle ukin-6: a biochemical marker of human prete rm 7 labour. Gy n eco l Obs tet Inves t 1993; 36: 145–147. ‘sign of rejection’. Another possible explanation 13. Greig PC, Murtha AP, Jimmerson CJ, He rbe rt WNP, Roitman-Johnson B, might be the physical effort of giving birth, because Allen J. Maternal serum inte rleukin-6 during pregnanc y and during te rm and prete rm labor. Obs tet Gy ne co l 1997; 90: 465–469. prolonged exercise is associated w ith increased pro- 14. Arntze n KJ, Lien E, Austgulen R. Mate rnal serum le ve ls of inte rleukin-6 duction of IL-6 from stimulated blood mononuclear and clinic al characteristics of normal delivery at term. Ac ta Obs te t 26 Gy n ecol Sca nd 1997; 76: 55–60. cells. 15. Steinborn A, Kuhnert M, Halberstadt E. Immunomodulating cytokines in In contrast, the low values of TNF-a in maternal te rm and preterm parturition. J Perina t Med 1996; 24: 381–390. 16. Rizzo G, Capponni A, Rinaldo D, Tedeschi D, Arduini D, Romanini C. serum suggest a paracrine action for this cytokine in Interleukin-6 concentrations in cervical secretions ide ntify microbial maternal and fetal tissues, as they regulate the grow th invasion of the amniotic c avity in patients w ith prete rm labor and intac t membranes. Am J Obs tet Gy n ecol 1996; 175: 812–817. of trophoblast during pregnancy and contribute to 17. Rudlof AE 1992. Tumor nec rosis fac tor-a in human milk. Ped ia tr Re s 24 the immune processes, associated w ith labour. 1992; 31: 29–33. 18. Ellis LA, Mastro AM, Picciano MF. Do milk-born cytokine s and hormones Generally, IL-1b , IL-6 and TNF-a released by chor- influenc e neonatal immune c ell function? J Nutr 1997; 127: ion, amnion and decidua and acting on these mem- 985S–988S. Mediators of Inflammation · Vol 7 · 1998 311 A. Sara nda ko u et al.

19. Sarandakou A, Phoc as I, Giannaki G, Rizos D, Davri Y, Zourlas PA. 24. Laham N, Brennecke SP, Bendtze n K, Ric e GE. Tumor ne crosis fac tor- Immunomarkers in new born infants. In: Cosmi EV, Di Renzo GC, e ds. alpha during human pregnancy and labor: maternal plasma and amniotic Pro ceeding s o f 2n d Wo rld Co n gre s s o f Perin a tal Medic in e. Rome: fluid c oncentrations and release of intrauterine tissue s. Eu r J End ocrin o l Monduzzi Editore , 1993; 307–310. 1994; 131: 607–614. 20. Malamitsi-Puchne r A, Sarandakou A, Giannaki G, Rizos D, Phocas I. 25. van Oers MHJ, van der He yde n AAPAM, Aarde n LA. Inte rleukin 6 (IL-6) in Change s of angiogenin se rum conc entration in the perinatal pe riod. se rum and urine of renal transplant recipients. Clin Exp Im m un o l 1988; Ped ia tr Res 1997; 41: 909–911. 71: 314–319. 21. Phocas I, Sarandakou A, Giannaki G, Malamitsi-Puchner A, Rizos D, 26. Haar PM, Petersen BK, Fomsgaard A. Effec t of physical ex ercise on in Zourlas PA. Soluble intercellular adhe sion molec ule-1 in new born vitro production of interleukin-1, inte rleukin-6, tumour ne crosis factor-a , infants. Eu r J Pedia tr 1997; 157: 153 –156. interleukin2 and interferon-g . In t Spo rts Med 1991; 12: 223–227. 22. Austgulen R, Lien E, Liabakk NB, Jacobsen G, Arntzen KJ. Inc reased leve ls 27. Kent AS, Sullivan MH, Elder MG. Transfe r of cytokines through human of cytokines and c ytokine activity modifiers in normal pregnanc y. Eu r J fe tal membranes. J Repro d Fertil 1994; 100: 81–84. Obs tet Gy n eco l Repro d Bio l 1994; 57: 149–155. 23. Buonocore G, De Filippo M, Gioia D, Pic ciolini E, et a l. Maternal and neonatal plasma c ytokine le ve ls in relation to mode of de livery. Biolo g y Received 22 June 1998; o f the Ne on a te 1995; 68: 104–110. accepted in revised form 20 July 1998

312 Mediators of Inflammation · Vol 7 · 1998