Oxidant and antioxidant status in mothers and their newborns according to birthweight

1. Introduction 

Pregnancy, mostly because of the increased oxygen requirement and mitochondria-rich placenta, is a condition exhibiting increased susceptibility to oxidative stress. Evidence for this concept includes studies demonstrating elevated levels of oxidative stress markers in normal pregnancy [1]. Plasma and erythrocyte malondialdehyde (MDA) levels were significantly higher while erythrocyte glutathione (GSH) levels and superoxide dismutase (SOD) activity were significantly lower in pregnant women in the third trimester than in nonpregnant women [2], [3]. The perinatal period and the delivery in particular is a critical time for maintaining a balance between the production of free radicals and the incompletely developed antioxidative protection of the fetus and newborn. Delivery represents a significant oxidative stress for a fetus, which passes from a hypoxic intrauterine space to a normooxic environment [4]. Lipid peroxidation and antioxidant status are changed during delivery, and these changes affect the fetus by creating oxidative stress [2], [5]. However, in uncomplicated term pregnancy, appropriate for gestational age (AGA) newborns possess antioxidant defense capable of resisting the physiological oxidative stress at birth [4]. Several studies have addressed the influence of labor and mode of delivery on oxidative stress [5], [6], [7]. Oxidative stress may be related to delivery or to a pre-existing fetal oxidative status [7]. Indeed, high oxidative stress was found in the fetal circulation, regardless to the mode of delivery [6].

Oxygen free radicals have been implicated in the etiology of premature delivery, fetal growth restriction, eclampsia, maternal infections and maternal malnutrition [4], [8], [9], [10]. Risk may, however, depend on the mother's antioxidant status which potentially protects the maternal–fetal unit, thus increasing intrauterine growth and infant weight at birth [2], [4], [11]. In fact, the steady-state formation of ROS produced during cell metabolism is normally balanced by a similar rate of consumption by antioxidants. Under normal conditions, protective intracellular enzymes mainly catalase and superoxide dismutase, and non-enzymatic antioxidants such as glutathione and vitamins A, C and E prevent ROS accumulation. Oxidative stress may result from imbalance in this pro-oxidant–antioxidant equilibrium.

A significant correlation was found between some maternal and cord blood oxidative stress markers [12]. However, few data provided a global estimation of oxidative stress in mothers and their newborns, according to birth weight. During pregnancy complicated by intrauterine growth restriction, the MDA concentration in amniotic fluid was higher and total antioxidant capacity of the serum was lower than in normal pregnancy [13], [14]. Oxidative stress was induced both in small for gestational age (SGA) newborns and their mothers which is manifested as increased lipid peroxidation and protein oxidant damage [15]. The implication of oxidative status in large for gestational age (LGA) newborns of healthy mothers is not well known.

Epidemiological studies suggest that there is a relationship between SGA and LGA newborns and developing metabolic syndrome in adulthood [16]. Oxidative stress has been implicated in adult metabolic syndrome [16]. The question of the possible link between fetal oxidant and antioxidant status and the development of long term metabolic abnormalities is of a great interest.

The aim of the present study is to investigate oxidative stress status in AGA, SGA and LGA newborns and in their respective mothers. This status was evaluated by assaying both plasma total antioxidant capacity (ORAC), markers of lipid and protein oxidation and blood antioxidant defenses, namely erythrocyte superoxide dismutase, catalase, glutathione peroxidase and reductase activities, plasma vitamin A, C and E levels.

2. Materials and methods 

2.1. Patients 

The study population included 139 women giving birth at the obstetrics and genecology department of Tlemcen Hospital, Tlemcen, Algeria. They were recruited successively among the admitted women at the hospital. A written consent was required from all the subjects, and the study was approved by the Tlemcen Hospital Committee for Research on Human Subjects.

The women claimed to have no history of chronic diseases, eclampsia, infections or fetal anomalies. All were tested for gestational diabetes according to the World Health Organization criteria, and all had normal glucose tolerance test during the third trimester and within 48h of birth. Care was taken to ensure that all the subjects were of similar age, weight, height, gestational age and parity. All these women had uncomplicated singleton pregnancies. None showed any abnormalities during labor and delivered vaginally at term. Gestational age was estimated by the last menstrual period and confirmed by a first-trimester ultrasound scan. Newborn weight was recorded immediately after delivery. Appropriate growth was defined by the presence of ultrasonographic signs (when biparietal diameter and abdominal circumference were between the 10th and 90th percentiles) according to the normograms of Campbell and Thoms [17] and by postnatal confirmation of a birth weight between the 10th and 90th percentiles (between 2600 and 3900g) according to our population standard curves (unpublished data). Small for gestational age newborns with birth weight less than 10th percentile or less than 2500g at term, and large for gestational age newborns with birth weight over 4000g at term (>90th percentile) were also identified.

Three groups were then selected and studied: Group 1 consisted of 56 mothers and their AGA newborns (control group). Group 2 consisted of 45 mothers and their SGA newborns, and Group 3 consisted of 38 mothers and their LGA neonates. Maternal and neonate characteristics are shown in Table 1.

Table 1. Maternal and neonatal characteristics

	Characteristics	Group 1 (AGA)	Group 2 (SGA)	Group 3 (LGA)
	Number	56	45	38
	Age (years)	25±1	24±1.2	25±1.6
	BMI (kg/m2)	22.80±2.3	22.14±1.7	23.07±1.5
	Parity	2±1	2±1	2±1
	Gestational age (weeks)	38.50±0.56	38.11±0.50	39.20±0.80
	Birth weight (g)	3290±58	2295±33	4384±70
	M/F sex ratio	30/26	20/25	22/16

Values are means±S.E.M. BMI, body mass index (weight/height2); M/F, males/females.

3. Results 

3.1. Lipid parameters 

No statistically significant differences between plasma and lipoprotein—cholesterol, triglyceride or apolipoprotein (apo) levels in mothers of AGA newborns (Group 1) and mothers of LGA infants (Group 3) or in their respective newborns were confirmed (Table 2). However, mothers of SGA babies (Group 2) had significantly lower plasma triglyceride and apo B100 concentrations than those of other mothers. In their SGA newborns, plasma triglyceride and apo B100 values were significantly higher, whereas plasma, LDL- and HDL-cholesterol and apo A-I amounts were significantly lower than in AGA and LGA newborns.

Table 2. Lipid, lipoprotein and apolipoprotein levels in mothers and their newborns

		Group 1 (AGA)	Group 2 (SGA)	Group 3 (LGA)
	Mothers
	Total cholesterol (mmol/L)	6.60±0.33	6.52±0.45	6.74±0.58
	Triglycerides (mmol/L)	2.35±0.18	1.96±0.11*	2.57±0.27$
	HDL-C (mmol/L)	1.69±0.14	1.57±0.27	1.72±0.32
	LDL-C (mmol/L)	3.24±0.22	3.30±0.36	3.56±0.44
	apo A-I (g/L)	2.70±0.35	2.38±0.54	2.64±0.41
	apoB 100 (g/L)	1.43±0.18	0.89±0.14*	1.52±0.23$
	Newborns
	Total cholesterol (mmol/L)	1.50±0.13	1.17±0.14*	1.61±0.38$
	Triglycerides (mmol/L)	0.67±0.05	1.25±0.22*	0.78±0.14$
	HDL-C (mmol/L)	0.97±0.06	0.66±0.07*	0.99±0.15$
	LDL-C (mmol/L)	0.48±0.04	0.33±0.05*	0.49±0.06$
	Apo A-I (g/L)	0.66±0.05	0.38±0.06*	0.69±0.12$
	Apo B 100 (g/L)	0.32±0.03	0.51±0.04*	0.35±0.08$

Values are means±S.E.M. HDL-C, high density lipoprotein-cholesterol; LDL-C, low density lipoprotein-cholesterol; Apo, apolipoprotein. Statistical analysis was performed using ANOVA followed by the least significant difference (LSD) test.

* Significantly different (P<0.01) compared to AGA group. $ Significantly different (P<0.01) compared to SGA.

3.2. Oxidative stress markers 

No statistically significant differences in biochemical markers of oxidative stress were observed between AGA and LGA mothers (Table 3). In contrast, SGA mothers had significantly higher plasma hydroperoxide and carbonyl protein levels as compared to AGA mothers, while ORAC, vitamin C and E values, superoxide dismutase and catalase activities were found to be significantly decreased (Table 3, Table 4). In the newborn, several alterations in biochemical oxidative stress markers were noted in both SGA and LGA groups. The levels of ORAC and vitamin E were found to be statistically lower whereas those of hydroperoxides and carbonyl proteins were significantly higher in these SGA and LGA newborns as compared to the AGA newborns. In addition, LGA babies had significantly lower vitamin A and significantly higher superoxide dismutase activity in comparison with control neonates. SGA newborns showed significantly lower vitamin C, catalase and superoxide dismutase levels than those of controls. Glutathione peroxidase and reductase activities were similar in both groups (Table 4). In general, variations in biochemical markers of oxidative stress observed in SGA newborns were parallel to those seen in their mothers. However, changes observed in LGA babies were different from those seen in their mothers. Alterations of oxidant and antioxidant status were noted in LGA newborns despite normal status in their mothers.

Table 3. Serum oxidative stress markers in mothers and their newborns

		Group 1 (AGA)	Group 2 (SGA)	Group 3 (LGA)
	Mothers
	ORAC (Arbitrary Units)	3.69±0.36	2.26±0.25*	3.97±0.42$
	Vitamin A (μmol/L)	13.80±2.04	11.56±1.47	14.75±2.26
	Vitamin C (μmol/L)	50±5.04	25±2.16*	54.11±4.12$
	Vitamin E (μmol/L)	29±2.03	14.33±1.08*	32.68±3.13$
	Hydroperoxides (μmol/L)	3.94±0.54	6.82±0.29*	4.21±0.66$
	Carbonyl proteins (nmol/mg protein)	1.17±0.34	2.96±0.38*	1.34±0.41$
	Newborns
	ORAC (Arbitrary Units)	1.68±0.21	1.17±0.14*	1.26±0.18$
	Vitamin A (μmol/L)	5.18±0.63	4.38±0.56	2.07±0.85*$
	Vitamin C (μmol/L)	48.15±4.16	20.62±1.42*	54.31±5.07$
	Vitamin E (μmol/L)	12.26±1.04	4.68±0.65*	6.29±1.08*
	Hydroperoxides (μmol/L)	2.59±0.43	3.78±0.38*	3.80±0.52*
	Carbonyl proteins (nmol/mg protein)	1.37±0.24	1.85±0.37*	1.66±0.07*

Values are means±S.E.M. ORAC, plasma oxygen radical absorbance capacity was determined as described in Section 2. Statistical analysis was performed using ANOVA followed by the least significant difference (LSD) test.

* Significantly different (P<0.01) compared to AGA group. $ Significantly different (P<0.01) compared to SGA.

Table 4. Erythrocyte antioxidant enzyme activities in mothers and their newborns

		Group 1 (AGA)	Group 2 (SGA)	Group 3 (LGA)
	Mothers
	Superoxide dismutase (U/g Hb)	683±37	145.20±57*	705±53.26$
	Catalase (U/g Hb)	67.80±5.14	20.42±8.80*	73.68±5.86$
	Glutathione peroxidase (U/g Hb)	89±8.04	76.82±9.16	94.08±4.12
	Glutathione reductase (U/g Hb)	56.53±7.63	61.33±8.08	63.48±8.53
	Newborns
	Superoxide dismutase (U/g Hb)	74.18±2.36	46.11±3.57*	132.44±5.42*,$
	Catalase (U/g Hb)	18.73±1.14	11.54±2.08*	17.68±1.26
	Glutathione peroxidase (U/g Hb)	8.24±1.81	7.83±0.95	8.05±1.14
	Glutathione reductase (U/g Hb)	6.62±1.03	5.17±1.08	6.89±1.13

Values are means±S.E.M. Statistical analysis was performed using ANOVA followed by the least significant difference (LSD) test.

* Significantly different (P<0.01) compared to AGA group. $ Significantly different (P<0.01) compared to SGA group.

4. Discussion 

In this study, we have attempted to demonstrate that deviations in intrauterine growth patterns were associated with abnormal oxidant and antioxidant status.

In our study, all newborns were vaginally delivered, and the mixed arteriovenous umbilical cord blood was collected and used to examine oxidative markers in fetal circulation. Fogel et al. [6] showed that lipids in the umbilical arterial serum are more susceptible to peroxidation than lipids in the umbilical venous serum, indicating high oxidative stress in the fetus, regardless to the mode of delivery. However, previous findings suggest that the way of delivery has an effect on oxidative stress in newborns [5], [7]. Neonates delivered by elective cesarean section were exposed to a higher oxidative stress compared to those who had an uncomplicated vaginal delivery [5]. In addition, oxidative markers were higher in umbilical cord blood in women delivering by emergent cesarean compared to those delivering by elective cesarean [7].

Our findings showed that intrauterine growth restriction strongly alters lipoprotein metabolism, and that fetal macrosomia is not associated with lipid and lipoprotein abnormalities, during uncomplicated pregnancies. These findings are in concordance with previous studies [24], [25]. Compared with AGA newborns, SGA newborns had increased plasma triglycerides and apolipoprotein B100, and decreased LDL and HDL contents, possibly resulting from a limited ability to metabolize VLDL particles. To explain hypertriglyceridemia in SGA neonates, several mechanisms have been suggested such as hypoxemia, hypoinsulinemia, and lipoprotein lipase impairment [24]. Similar lipoprotein changes were observed in children with marasmus, due to postnatal malnutrition [26]. Mothers of SGA neonates also showed lipid alterations, such as reduced triglyceride and apolipoprotein B100 levels, in agreement with previous findings [27]. These results could be explained by reduced hepatic VLDL synthesis and secretion or by enhanced fatty acid transfer through the placenta after LPL hydrolysis.

On the other hand, SGA newborns as well as their mothers showed altered oxidant and antioxidant status. Our data revealed that the total antioxidant activity (ORAC) was decreased reflecting oxidative stress in these newborns and in their mothers. In addition, plasma hydroperoxide and carbonyl protein levels were increased in SGA groups. Hydroperoxides and carbonyl proteins are commonly used as indicators of lipid peroxidation and protein oxidation. Our results were in accordance with other studies which reported high oxidative damage in IUGR [14], [15]. It was found that during pregnancy complicated by IUGR, the malondialdehyde (one of the many products of the oxidized lipids breakdown) concentration in amniotic fluid was significantly higher than in normal pregnancy and determination of MDA can be used as a biochemical test in parental diagnosis of IUGR [13].

Concerning vitamins, we found no alterations in plasma vitamin A concentrations, whereas vitamin C and E levels were lower in SGA newborns and in their mothers than in AGA group. There are few studies with conflicting reports regarding the vitamin status in maternal and cord blood [8], [11], [12]. Low levels of vitamins E and C were reported in mothers with low birth weight newborns [11], [12]. Low levels of these vitamins could reflect their high utilization rate, suggesting that they might be used to reduce oxidative stress in mothers of SGA newborns. Alternatively, it is also possible that reduced vitamin C and E concentrations reflect low intake, which resulted in decreased antioxidant defense system in these mothers and thereafter in their SGA newborns.

In our study, SGA newborns and their mothers showed altered erythrocyte antioxidant enzyme activities. SOD and catalase activities were reduced in SGA newborns and in their mothers while glutathione peroxidase and reductase activities were not modified. A reduction in SOD the primary enzyme that inactivates the superoxide radical, and in catalase activity involved in the detoxification of H2O2 would lead to increased numbers of free radicals, and could thereafter be responsible for the increased levels of hydroperoxide and carbonyl proteins in SGA newborns and in their mothers. On the other hand, free radicals and lipid peroxides might be involved in the regulation of antioxidant enzyme activity and gene expression; SOD and catalase can be inactivated by high levels of these reactive radicals [28]. Similar findings have been reported in pregnancies with preeclampsia, related to a response to the placental ischemia and fetal hypoxia [10]. Erythrocyte antioxidants were lower in cord blood of SGA infants born to undernourished mothers compared to term AGA infants born to healthy mothers [9].

Altogether, our data suggested that intrauterine restriction is associated with significant oxidative stress in SGA newborns and in their mothers.

In LGA newborns, the situation was different. These newborns showed altered oxidant and antioxidant status despite normal status in their mothers. We found that hydroperoxide and carbonyl protein levels were enhanced while plasma vitamin A and E concentrations were reduced in LGA compared to controls. Erythrocyte SOD activity was significantly increased in these LGA newborns. These data confirm the existence of an oxidative stress in fetal macrosomia. Reduced ORAC values in LGA newborns compared to controls was also in favor of an oxidative stress. Similar findings have been reported in obese children [29]. The enhanced SOD activity could involve, at least partly, an induction of this enzyme by oxidative stress. It has been shown that the increased SOD activity may be a compensatory nature responding to the increased peroxide levels in obesity [29]. Increased expenditure of antioxidant vitamins A and E or/and enhanced entrapment of these vitamins in the adipose tissue have been reported in obesity [30].

Altogether, our data supported an imbalance of the oxidant and antioxidant systems in favor of an oxidative stress in IUGR as well as in fetal macrosomia.

In SGA newborns, oxidant and antioxidant status was related to that found in mothers. Oxidant stress in IUGR may be reduced by controlling maternal antioxidant status. Supplementary therapy with antioxidants including vitamins and trace elements might help to shift this oxidant and antioxidant balance.

We provided novel evidence that oxidant and antioxidant status in LGA newborns was not related to maternal status. In our study, alterations of oxidant and antioxidant markers in LGA newborns might be explained by excessive nutrients availability, fetal hyperinsulinemia or other factors that could enhance free radical production and depress natural antioxidant defenses. However, our study has a limitation because maternal blood was not always taken during delivery, but within 48h of birth. It is plausible that oxidant and antioxidant status in LGA mothers might also be impaired at delivery.

On the other hand, persisting abnormalities in oxidant and antioxidant balance in SGA and LGA newborns could be one of the processes that link fetal growth deviations with adult metabolic diseases, especially metabolic syndrome [16].

We therefore suggest follow up studies on oxidant and antioxidant status of SGA and LGA infants to investigate the long term consequences of oxidative markers impairment at birth. Nevertheless, oxidant and antioxidant status in SGA as well as in LGA neonates should be carefully considered, and appropriate management should be organized during the early postnatal period, including antioxidant supplementation.