Fax +41 61 306 12 34E-Mail karger@karger.chwww.karger.com

Original Paper

Neonatology 2007;92:111115 DOI: 10.1159/000100965

Different Degrees of Fetal Oxidative Stress in Elective and EmergentCesarean Section

Samuel Lurie a, d Zippora Matas b Mona Boaz c, d Asora Fux c

Abraham Golan a, d Oscar Sadan a, d

Departments of a Obstetrics and Gynecology, b Biochemistry, and c Epidemiology and Statistics Unit,Edith Wolfson Medical Center, Holon , and d Sackler School of Medicine, Tel Aviv University, Tel Aviv , Israel

vs. 0.7 8 0.3 nmol/l, p ! 0.05) and in umbilical cord erythro-cytes (159.6 8 48.6 nmol/g Hb vs. 85.8 8 5.2 nmol/g Hb,p ! 0.05) in women delivering by emergent cesarean com-pared to those delivering by elective cesarean. GPX activity was enhanced in amniotic fluid (12.4 8 2.2 U/l vs. 5.1 8 0.6 U/l, p ! 0.05) and GPX activity/hemoglobin ratio was higher in cord blood (22.0 8 0.8 U/g Hb vs. 18.7 8 0.9 U/g Hb, p ! 0.05) in women delivering by emergent cesarean compared to those delivering by elective cesarean. Conclusion: Dis-tressed fetuses delivered by emergency cesarean exhibited increased malondialdehyde concentrations, an indicative parameter for oxidative damage, and enhanced GPX activity an antioxidant enzyme, in amniotic fluid and umbilical cord blood compared to non-distressed fetuses delivered by elective cesarean section. This is probably an indication of higher fetal oxidative stress. Copyright 2007 S. Karger AG, Basel

Introduction

Oxidative stress has been implicated in several new-born conditions such as bronchopulmonary dysplasia, necrotizing enterocolitis, or periventricular leukomala-cia [1] . Cell damage caused by oxidative stress may result from an increase in oxidant generation, or a decrease in antioxidant protection, or a failure to repair oxidative

Key Words Oxidative stress Cesarean section Non-reassuring fetal heart pattern

Abstract Background: Several studies have addressed the influence of labor and mode of delivery on oxidative stress. Still it is unclear whether oxidative stress is related to delivery itself or whether it reflects a pre-existing fetal oxidative status. Objective: To investigate whether the degree of fetal oxida-tive stress is different between distressed fetuses that were delivered by emergent cesarean section and non-distressed fetuses that were delivered by elective cesarean section. Methods: The protocol of this prospective study was ap-proved by the Institutional Review Board Committee. Amni-otic fluid and umbilical artery blood were prospectively col-lected from 21 parturients who were delivered by an emergent cesarean section for non-reassuring fetal heart rate pattern and from 21 parturients who were delivered by an elective cesarean section in a tertiary care center. Oxida-tive stress was evaluated in amniotic fluid, umbilical cord plasma and erythrocytes by determining malondialdehyde concentration and glutathione peroxidase (GPX) activity. Results: Malondialdehyde concentration was higher in am-niotic fluid (mean 8 SEM) (2.2 8 0.7 nmol/l vs. 0.6 8 0.02 nmol/l, p ! 0.05), in umbilical cord plasma (1.2 8 0.2 nmol/l

Received: August 25, 2006 Accepted after revision: November 15, 2006 Published online: March 21, 2007

formerly Biology of the Neonate

Samuel Lurie, MD Department of Obstetrics and Gynecology, Edith Wolfson Medical Center IL56100 Holon (Israel) Tel. +972 3 502 8329, Fax +972 8 936 6259 E-Mail lurie@nashim.net or drslurie@hotmail.com

2007 S. Karger AG, Basel 16617800/07/09220111$23.50/0

Accessible online at: www.karger.com/neo

Lurie /Matas /Boaz /Fux /Golan /Sadan

Neonatology 2007;92:111115 112

damage [2, 3] . With severe oxidative stress, lipid peroxi-dation and protein damage may occur [4, 5] . Malondial-dehyde (MDA) is a marker of lipid peroxidation induced by oxygen-free radicals [6] and reflects the degree of oxi-dative stress [7] . Glutathione peroxidase represents one of several antioxidant defense system enzymes [8] .

Several studies have addressed the influence of labor and mode of delivery on oxidative stress [4, 5, 9, 10] . Some investigators observed that delivery by cesarean section is associated with an increased oxidative stress [4, 5, 9] while others have failed to detect an association between oxidative stress and mode of delivery [10] . Furthermore, it is unclear whether oxidative stress is related to delivery itself or reflects a pre-existing fetal oxidative status. For example, amniotic fluid MDA concentrations were found to be higher in pregnancy with intra-uterine growth re-striction [11] . Therefore, the objective of this study was to investigate whether the degree of fetal oxidative stress differs between fetuses delivered by emergent cesarean section for pre-existing fetal compromise and non-dis-tressed fetuses delivered by elective cesarean section without pre-existing fetal compromise.

Materials and Methods

Study Design The protocol of this prospective study was approved by the

Institutional Review Board Committee (protocol number 92003-0889). Subjects for the study were recruited among women who delivered by a cesarean section at the Wolfson Medical Center delivery ward, a tertiary health care facility during 2004. During the study period, we had 253 cesarean section of which 139 were elective and 114 were emergent. The clinicians approached the women for consent for participation in the study before taking specimens for analysis. Umbilical artery blood samples (after cord separation) and amniotic fluid samples (before fetal deliv-ery) were collected during cesarean section. The samples were placed into vacuum tubes containing EDTA as anticoagulant. The samples were immediately centrifuged at 3,000 g at 4 C for 10 min and plasma and amniotic fluid supernatant serum were removed and frozen at 70 C. Packed erythrocytes obtained from umbilical artery blood were washed three times with cold saline solution, lysed with distilled water and three freeze-thaw cycles. The supernatant solution obtained by centrifugation of cell lysate for 20 min at 10,000 g was used for biochemical assays. Aliquots of the same umbilical artery blood were obtained in heparin washed syringes and tested for pH. The oxidative stress was de-termined by assessing levels of thiobarbituric acid reactive sub-stances which quantifies MDA and by the activity of glutathione peroxidase, an antioxidant enzyme. Due to laboratory restriction, the samples were collected during mornings of non-weekend days. All the samples were measured in the same run. Sample size was calculated so that with a sample size of 42 subjects (n = 21 in each group), the present study was designed to have 80% power

to detect a true, by-method difference in umbilical cord plasma MDA of 0.6 8 0.09 nmol/l using the t test for independent sam-ples, assuming a two-sided of 0.05. This difference (0.6 nmol/l) was based on a previously published data by Rogers et al. [12] while allowing a 15% (0.09 nmol/l) deviation from this differ-ence.

Study Participants All women were healthy, taking no medication except multi-

vitamin supplementation and did not have any pregnancy com-plication. None of them had clinical evidence of chorioamionitis such as fever, leukocytosis, uterine tenderness or fetal tachycar-dia. Two study groups were formed consisting of 21 pregnant women, each. The first group included women undergoing an elective cesarean section, while the second group included wom-en undergoing cesarean section for non-reassuring fetal heart rate pattern (presumed fetal distress). The indications for elective ce-sarean section were suspected macrosomia (38.1%), previous ce-sarean section (33.4%), breech presentation (23.8%), and patient request (4.7%). The diagnosis of non-reassuring fetal heart pat-tern were made in accordance with the guidelines of the Ameri-can College of Obstetricians and Gynecologists (ACOG) and in-cluded either prolonged fetal bradycardia, or late decelerations, or severe repeated variable decelerations. The demographic charac-teristics of the women who participated in the study are summa-rized in table 1 . The two groups were similar with respect to ma-ternal age, gestational age, fetal gender, types of anesthesia used and fetal birth weight.

Laboratory Tests Amniotic fluid, plasma and erythrocytes hemolysates MDA

concentrations were assayed using 2-thiobarbituric acid accord-ing to Ohkawa et al. [13] using 1,1,33 tetramethoxymagane as a

Table 1. Demographic characteristics of all women with elective and emergent cesarean sections

Electivecesareans (n = 21)

Emergentcesareans (n = 21)

Age, years 33.586.4 31.886.3Parity 1.880.9 1.081.1*Gestational age, weeks 38.881.7 39.982.1 Birth weight, g 3,396.18689.6 3,112.68644.6Fetal gender

Male 12 (57.2) 15 (71.4)Female 9 (43.8) 6 (28.6)

Type of anesthesiaRegional 20 (95.2) 15 (71.4)General 1 (5.8) 6 (28.6)

pH 7.3280.06 7.2380.07*Apgar score at 5 min 9.980.4 9.880.5

The data is expressed as mean 8 standard deviation or as number (%).

* p < 0.05 when compared to elective cesarean group.

Fetal Oxidative Stress during Cesarean Neonatology 2007;92:111115 113

standard. The results were expressed for plasma and amniotic flu-id as nanomoles per milliliter or nanomoles per gram of protein and for erythrocytes as nanomoles per gram of hemoglobin. Pro-tein concentrations were measured on Olympus AU 2700 ana-lyzer with a commercial kit. Hemoglobin levels were measured according to the method of Drabkin [14] .

Glutathione peroxidase (GPX) activity was assayed by the method described by Konukoglu et al. [15] . One unit of enzyme activity was defined as micromoles of nicotianamide adenosine dinucleotide phosphate oxidized per minute assuming 6.22 ! 10 3 as micromolar absorbance of nicotianamide adenosine dinucleo-tide phosphate at 340 nm. The activity was expressed as units per liter or units per gram of protein for plasma and amniotic fluid and as units per gram of hemoglobin for erythrocytes.

The coefficient of variation of the intra- and inter-assay preci-sion for MDA was 3.5 and 5.2%, respectively, and for GPX 6.1 and 8.2%, respectively. The minimal detectable level of MDA was 0.12 nmol/l and of GPX activity 2.7 U/l.

Statistical Analysis Analysis of data was carried out using SPSS statistical analysis

software (SPSS Inc., Chicago, Ill., USA, 1999). For continuous variables, such as age and parameters of oxidative stress, descrip-tive statistics were calculated and reported as mean 8 standard deviation. Normalcy of distribution of continuous variables was assessed using the Kolmogorov-Smirnov test. Normally distrib-uted continuous variables were compared by group using the t test for independent samples. Continuous variables with distribu-tions deviating significantly from normal were compared by group using the Mann-Whitney U test. Categoricals were de-scribed using frequency distributions and compared by group us-ing the chi-square ( 2 ). Because pH and parity differed between the two groups, variables identified as differing by group were re-analyzed using univariate general linear modeling including group as the fixed factor and pH, parity and other relevant con-tinuous variables associated with the outcome variable in ques-tion as covariates. Spearmans rho ( ) was used to calculate cor-relation coefficients between continuous variables. All tests were two-sided and considered significant at p ! 0.05.

Results

MDA concentration and glutathione peroxidase activ-ity in amniotic fluid and umbilical cord arterial blood are presented in table 2 . In general, MDA concentration and glutathione peroxidase activity were significantly higher in amniotic fluid and umbilical cord arterial blood in the emergent cesarean section group (non-reassuring fetal heart rate pattern). Note that the umbilical artery pH was indeed significantly lower in this group ( table 1 ).

Lower umbilical artery pH was found to have inverse correlation with glutathione peroxidase activity per gram protein in the amniotic fluid (r = 0.44, p ! 0.05), with glu-tathione peroxidase activity in umbilical artery cord erythrocytes (r = 0.34, p ! 0.05), with MDA concentration per gram protein in amniotic fluid (r = 0.47, p ! 0.05), with MDA concentration in umbilical artery cord plasma (r = 0.43, p ! 0.05) and with MDA concentration per hemoglo-bin in umbilical artery cord plasma (r = 0.33, p ! 0.05). Umbilical artery pH did not correlate with glutathione peroxidase activity in amniotic fluid or MDA concentra-tion in amniotic fluid. Lower Apgar score at 5 min was not found to correlate significantly with any of the above-mentioned parameters except for MDA concentration in umbilical artery cord erythrocytes (r = 0.32, p ! 0.05).

Because of differences in parity and umbilical artery cord pH, all variables differing significantly by group (emergent vs. elective cesareans) were re-analyzed using general linear modeling. In all models, parity and um-bilical artery cord pH were included as covariates and group (emergent vs. elective cesarean sections) was in-cluded as the fixed factor. Additionally, other variables associated with the outcome variable were included as covariates. When general linear models were re-calcu-

Electivecesareans (n = 21)

Emergentcesareans (n = 21)

Amniotic fluid GPXU/l 5.180.5 12.482.1*U/g protein 1.680.2 3.780.6*

Amniotic fluid MDAnmol/l 0.680.02 2.280.7*nmol/g protein 274.1851.9 602.38121.2*

Cord erythrocyte GPX, U/g Hb 18.780.9 22.080.8* Cord plasma MDA, nmol/l 0.780.3 1.280.2* Cord erythrocyte MDA, nmol/g Hb 85.885.1 159.6848.6*

The data is expressed as mean8 standard error of mean (SEM). Hb = Hemoglobin.* p < 0.05 when compared to elective cesarean group.

Table 2. Glutathione peroxidase activity (GPX) and malondialdehyde (MDA) in amniotic fluid and umbilical cord in elective and emergent ceareans

Lurie /Matas /Boaz /Fux /Golan /Sadan

Neonatology 2007;92:111115 114

lated including type of anesthesia or fetal gender, no dif-ference in parameters was seen, so final models were without this inclusion. Univariate analysis of variance significantly (p ! 0.05) differed by group (emergent vs. elective cesarean sections), even after controlling for par-ity and umbilical artery cord pH for MDA concentration in umbilical artery cord plasma and in amniotic fluid and for glutathione peroxidase activity in amniotic fluid.

None of the neonates in our study had bronchopulmo-nary dysplasia, necrotizing enterocolitis or periventricu-lar leukomalacia.

Discussion

We have observed that MDA concentration was signif-icantly higher in amniotic fluid and umbilical artery cord plasma and erythrocytes and that glutathione peroxidase activity was significantly higher in amniotic fluid and um-bilical artery cord blood in fetuses delivered by an emer-gent cesarean section, i.e. were in greater distress. The ef-fect was present even after controlling for maternal parity and umbilical artery cord pH. The type of anesthesia (re-gional vs. general) or fetal gender did not have any influ-ence on MDA concentration or glutathione peroxidase ac-tivity in amniotic fluid or in umbilical cord blood.

Oxidative stress takes place as a consequence of inad-equate inactivation of accumulating oxygen free radicals by antioxidant defense system [16] . The effects of this im-balance are difficult to predict especially in the second half of pregnancy. During organogenesis, oxidative stress was implicated in occurrence of fetal malformations in diabetic mothers [17] . The antioxidant capacity of rat em-bryo was reported to gradually increase with gestational age [18] . This increase might be the basis for better fetal coping with oxidative stress in the second half of preg-nancy. Term labor was shown to trigger an up-regulation of fetal antioxidant reserve that was suggested to protect against the relative hyperoxia that is experienced by the

newborn infant at birth [19] . Inadequate fetal antioxidant reserve was thought to enhance the vulnerability to free radical damage in the preterm neonate [19] .

MDA is a highly reactive 3 carbon dialdehyde produced as a byproduct of polyunsaturated fatty acid peroxidation and arachidonic acid metabolism. Increased MDA levels are considered as an indicator of lipid peroxidation in-duced by oxygen free radicals [6] and may be indicative of ischemia reperfusion injury [19, 20] . Therefore, increased MDA concentration in distressed fetuses during cesarean as compared to fetuses that were not in distress may rep-resent increased pre-existing fetal oxidative status. It may be deduced that this increased pre-existing fetal oxidative status was caused by distress resulting from the condition of the fetus as expressed by the non-reassuring fetal heart rate pattern. Fetal oxygenation is known to modulate the heart rate through the brain. Various changes of oxygen-ation of the fetus may occur during labor which may in-duce causing ischemic reperfusion periods during which there is an increased production of reactive oxygen spe-cies. Increased production of reactive oxygen species in turn may cause increased glutathione peroxidase activity. The observed increase in glutathione peroxidase activity in more distressed fetuses as observed in our study fur-thers the notion that the observed oxidative stress may reflect a pre-existing fetal oxidative status rather than be-ing a delivery related phenomenon. Hence, MDA levels and glutathione peroxidase activity could possibly serve as markers of fetal distress and could be determined in am-niotic fluid in equivocal clinical situations.

Conclusion

We have demonstrated an increased production of ox-idants and enhanced activity of the anti-oxidant defense system in amniotic fluid and umbilical cord blood of more distressed fetuses during cesarean section, proba-bly indicating an increased pre-existing oxidative stress.

References 1 Saugstad OD: Oxidative stress in the new-born a 30-year perspective. Biol Neonate 2005; 88: 228236.

2 Nicholls DG, Budd SL: Mitochondria and neuronal survival. Physiol Rev 2000; 80: 315360.

3 Fiers W, Beyaert R, Declercq W, Vandena-beele P: More than one way to die: apoptosis, necrosis and reactive oxygen damage. Onco-gene 1999; 18: 77197730.

4 Georgeson GD, Szony BJ, Streitman K, Varga IS, Kovacs A, Kovacs L, Laszlo A: Antioxi-dant enzyme activities are decreased in pre-term infants and in neonates born via caesar-ean section. Eur J Obstet Gynecol Reprod Biol 2002; 103: 136139.

5 Yaacobi N, Ohel G, Hochman A: Reactive oxygen species in the process of labor. Arch Gynecol Obstet 1999; 263: 2324.

Fetal Oxidative Stress during Cesarean Neonatology 2007;92:111115 115

11 Bazowska G, Jendryczko A: Concentration of malondialdehyde (MDA) in amniotic f lu-id and maternal and cord serum in cases of intrauterine growth retardation. Zentralbl Gynakol 1994; 116: 329330.

12 Rogers MS, Mongelli JM, Tsang KH, Wang CC, Law KP: Lipid peroxidation in cord blood at birth: the effect of labour. Br J Obstet Gynaecol 1998; 105: 739744.

13 Okhawa H, Ohishi N, Yagi K: Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 95: 351358.

14 van Kampen EJ, Zijstra WG: Determination of hemoglobin and its derivatives. Adv Clin Chem 1965; 8: 141187.

15 Konukoglu D, Akcay T, Eredem T: Suscepti-bility of erythrocyte lipids to oxidation and erythrocyte antioxidant status in myocardi-al infarction. Clin Biochem 1998; 31: 667671.

16 Rice-Evans C, Burdon R: Free radical lipid interactions and their pathological conse-quences. Prog Lipid Res 1993; 32: 71110.

17 El-Bassiouni EA, Helmy MH, Abou Rawash N, El-Zoghby SM, Kamel MA, Abou Rayah AN: Embryopathy in experimental diabetic gestation: assessment of oxidative stress and antioxidant defense. Br J Biomed Sci 2005; 62: 7176.

18 Zaken V, Kohen R, Ornoy A: The develop-ment of antioxidant defense mechanism in young rat embryos in vivo and in vitro. Ear-ly Pregnancy 2000; 4: 110123.

19 Buhimschi IA, Buhimschi CS, Pupkin M, Weiner CP: Beneficial impact of term labor: nonenzymatic antioxidant reserve in the hu-man fetus. Am J Obstet Gynecol 2003; 189: 181188.

20 Nagamatsu M, Schmelzer JD, Zollman PJ, Smithson IL, Nickander KK, Low PA: Isch-emic reperfusion causes lipid peroxidation and fiber degeneration. Muscle Nerve 1996; 19: 3747.

6 Buege JA, Aust SD: Microsomal lipid perox-idation. Methods Enzymol 1978; 52: 302310.

7 Abuja PM, Albertini R: Methods for moni-toring oxidative stress, lipid peroxidation and oxidation resistance of lipoproteins. Clin Chim Acta 2001; 306: 117.

8 Freeman BA, Crapo JD: Biology of disease: free radicals and tissue injury. Lab Invest 1982; 47: 412426.

9 Inanc F, Kilinc M, Kiran G, Guven A, Kuru-tas EB, Cikim IG, Akyol O: Relationship be-tween oxidative stress in cord blood and route of delivery. Fetal Diagn Ther 2005; 20: 450453.

10 Fogel I, Pinchuk I, Kupferminc MJ, Lichten-berg D, Fainaru O: Oxidative stress in the fe-tal circulation does not depend on mode of delivery. Am J Obstet Gynecol 2005; 193: 241246.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.