Pieter Sauer

Functional aspects of developmental toxicity of polyhalogenated aromatic hydrocarbons in experimental animals and human infants

A scientific evaluation was made of functional aspects of developmental toxicity of polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) in experimental animals and in human infants. Persistent neurobehavioral, reproductive and endocrine alterations were observed in experimental animals, following in utero and lactational exposure to PCBs, PCDDs and PCDFs. The lowest observable adverse effect levels (LOAELs) for developmental neurobehavioral and reproduction endpoints, based on body burden of TCDD-toxic equivalents (TEQs) in animals, are within the range of current background human body burdens. Relatively subtle adverse effects on neurobehavioral development and thyroid hormone alterations have also been observed in infants and children exposed to background levels. Exclusive use of the toxic equivalency factor (TEF) approach may underestimate the risk of neurodevelopmental effects, because both Ah receptor dependent and independent mechanisms may be involved in these effects. The use of marker congeners and/or bioassays based on Ah receptor mediated mechanisms are rapid, low cost pre-screening alternatives for expensive and time consuming gas chromatographic-mass spectrometric analysis.

Effects of Environmental Exposure to Polychlorinated Biphenyls and Dioxins on Birth Size and Growth in Dutch Children

Lower birth weight and growth retardation has been found in studies with laboratory animals, in children born of mothers exposed to accidental high levels of polychlorinated biphenyls (PCBs) and related compounds, and in children born of mothers who consumed PCB-contaminated fish. The effect of background exposure to PCBs and dioxins on birth size and growth in human newborns, however, is still unknown. This study examined birth size and postnatal growth of term newborns in relation to their background PCB and dioxin exposure. Birth weight and weight, length, and head circumference were measured at 10 d and 3, 7, 18, and 42 mo of age in 207 children, of whom 105 were breast-fed and 102 were formula-fed during infancy. The effect of in utero exposure to PCBs on birth size, assessed by cord and maternal plasma PCB levels, was investigated in the whole group. The effect of prenatal PCB exposure on postnatal growth was studied in the formula-fed group, whereas the effect of prenatal as well as lactational exposure to PCBs and dioxins on postnatal growth was studied in the breast-fed group. After adjustment for covariates, cord and maternal plasma PCB levels where both negatively associated with birth weight. Infants with high cord plasma PCB levels (P90 = 0.80 µg/L) weighed 165 g less compared with infants with low cord plasma PCB levels (P10 = 0.20 µg/L). Cord and maternal plasma PCB levels where both significantly associated with lower growth rate, defined as change in SD score (SDS) of weight, length, and head circumference from birth to 3 mo in the formula-fed group (all p values <0.05). No negative effects of prenatal PCB exposure on growth rate were found from 3 to 42 months of age. Postnatal PCB and dioxin exposure was not negatively associated with growth rate in the breast-fed group. In utero exposure to environmental levels of PCBs is negatively associated with birth weight and postnatal growth until 3 mo of age. Although this growth delay was described in healthy term born infants, intrauterine and postnatal growth retardation are potentially harmful to the developing human and should be avoided by reducing maternal PCB and dioxin body burden, and consequently fetal exposure to these pollutants.

Adrenal function in sick very preterm infants

Some very preterm neonates admitted to the neonatal intensive care unit show circulatory and respiratory problems that improve after administration of steroids. It is unclear whether these symptoms could be caused by adrenal insufficiency. The objective of our study was to investigate the cortisol levels and the cortisol release from the adrenals after ACTH in very preterm infants with and without severe illness and to find whether a relation exists between adrenal function and outcome. An ACTH test (0.5 μg) was performed on d 4 in 21 very preterm infants (gestational age, 25.6–29.6 wk; birth weight, 485–1265 g). Baseline cortisol and 17-hydroxyprogesterone (17OHP) levels and the cortisol levels 30, 60, and 120 min after ACTH administration were measured. The Score for Neonatal Acute Physiology was used to measure illness severity. All infants showed an increase in cortisol levels after ACTH, but the cortisol levels were significantly lower in the ventilated more severely ill infants. After adjusting for birth weight and gestational age, the mean baseline cortisol levels and cortisol/17OHP ratios were significantly lower and the 17OHP levels significantly higher in the ventilated infants compared with the nonventilated infants. Patients with an adverse outcome had significantly lower baseline cortisol/17OHP ratios and 60-min cortisol levels during ACTH testing (p = 0.002 and p = 0.03, respectively). These data suggest an insufficient adrenal response to stress in sick ventilated very preterm infants with gestational ages younger than 30 wk compared with nonventilated less sick preterm infants. Further studies are required to investigate whether supplementation with physiologic doses of hydrocortisone may benefit the outcome.

Fetal Exposure to PCBs and Their Hydroxylated Metabolites in a Dutch Cohort

Polychlorinated biphenyls (PCBs) are still the most abundant pollutants in wildlife and humans. Hydroxylated PCB metabolites (OH-PCBs) are known to be formed in humans and wildlife. Studies in animals show that these metabolites cause endocrine-related toxicity. The health effects in humans have not yet been evaluated, especially the effect on the fetus and newborn. The aim of this study is to measure the levels of PCBs and OH-PCBs in maternal and cord blood samples in a population with background levels of PCBs. We analyzed 51 maternal and corresponding cord blood samples in the northern part of the Netherlands. The PCB concentrations in maternal plasma ranged from 2 to 293 ng/g lipid, and OH-PCB concentrations from nondetectable (ND) to 0.62 ng/g fresh weight. In cord plasma, PCB concentrations were 1–277 ng/g lipid, and OH-PCB concentrations, ND to 0.47 ng/g fresh weight. The cord versus maternal blood calculated ratio was 1.28 ± 0.56 for PCBs and 2.11 ± 1.33 for OH-PCBs, expressed per gram of lipid. When expressed per gram fresh weight, the ratios are 0.32 ± 0.15 and 0.53 ± 0.23 for PCBs and OH-PCBs, respectively. A significant correlation between the respective maternal and cord levels for both PCBs and OH-PCBs was found. Our results indicate that OH-PCBs and PCBs are transferred across the placenta to the fetus in concentrations resulting in levels of approximately 50 and 30%, respectively, of those in maternal plasma. More research in humans is needed to evaluate potential negative effects of these endocrine disruptors on the fetus.

PERINATAL EXPOSURE TO POLYCHLORINATED BIPHENYLS AND DIOXINS THROUGH DIETARY INTAKE.

Polychlorinated biphenyls (PCBs) and dioxins (polychlorinated dibenzo-p-dioxins and dibenzofurans) are potentially hazardous compounds. Since food is the major source (> 90%) for the accumulation of PCBs and dioxins in the human body, food habits in women determine the degree of fetal exposure and levels in human milk. In order to investigate an association between dietary intake and PCB and dioxin levels in human milk and PCB levels in maternal and cord plasma, the food intake of 418 Dutch women during pregnancy was recorded using semi-quantitative food frequency questionnaires. After adjusting for covariates, a weak association was found between the estimated dietary intake of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD), dioxins, and planar PCBs and their corresponding levels in breast milk. The estimated dietary intake of 2,3,7,8-TCDD, dioxins, and planar PCBs was also related to the PCB levels in maternal and cord plasma. Dairy products accounted for about half and industrial oils for about a quarter of the estimated 2,3,7,8-TCDD, dioxin, and the planar PCB intake. It is concluded that the contribution of a pregnancy related diet to PCB and dioxin levels in human milk and to PCB levels in maternal and cord plasma is relatively low. Decrease of exposure to PCBs and dioxins of the fetus and the neonate probably requires long-term reduction of the intake of these pollutants. Substitution of normal cheese by low-fat cheese and the use of vegetable oils instead of fish oils in the preparation of foodstuffs by the food industry could contribute to a reduced intake of PCBs and dioxins.

BEHAVIOUR PROBLEMS OF VERY LOWBIRTHWEIGHT CHILDREN

Parent and clinician reports of behaviour problems among very low‐birthweight (VLBW) children at 3½ years of age were studied in relation to indicators of neonatal cerebral damage, cognition and social factors. VLBW children had more depressed behaviour and more internalizing problems by parent report, and also scored significantly more often within the clinical range on total problem scores, than children in the comparison group. Neither neonatal cerebral ultrasound nor neurological examinations were directly associated with behavioural outcome. Cerebral damage was related to cognitive development. Cognition directly influenced behaviour problems according to clinician report, while the home environment did so according to parent report. The authors suggest that depressed behaviour of preschool VLBW children might be associated with parental reactions to the birth of a VLBW child, and that their attention problems might be linked indirectly to brain damage via cognitive impairments.

Neonatal cerebral ultrasound, neonatal neurology and perinatal conditions as predictors of neurodevelopmental outcome in very low birthweight infants

To determine the assessments before discharge from the intensive care unit, that will predict outcome most accurately, a prospective longitudinal study in a cohort of 79 high risk VLBW children was conducted from birth to 3.6 years of age. Birthweight, gestational age, obstetrical and neonatal optimality, neonatal neurological examinations and neonatal cerebral ultrasound were studied in relation to outcome. The best predictor for outcome was a simple cerebral ultrasound classification according to the presence or absence of ventriculomegaly and intraparenchymal damage of any cause. Infants with normal neonatal cerebral scans or abnormal scans without ventriculomegaly almost invariably had a normal neurological outcome. In infants with cerebral lesions with ventriculomegaly the incidence of normal neurological outcome decreased to less than 50%. Intraparenchymal damage was associated with cerebral palsy as well as other (mental and sensori) handicaps in over 85% of the cases. Neonatal neurological examinations at preterm age had additional value in predicting neurological outcome especially in the group with ventriculomegaly. Neither birthweight, nor gestational age, obstetrical or neonatal optimality were independent variables in the prediction of outcome in high risk VLBW children at 3.6 years of age.

Comparison of Short Term Indirect Calorimetry and Doubly Labeled Water Method for the Assessment of Energy Expenditure in Preterm Infants

The accuracy of 8-hour indirect calorimetry (IDC) as an estimate of energy expenditure was investigated in 8 healthy preterm infants (birth weight 1,270 +/- 193 g, gestational age 32 +/- 3 weeks, mean +/- SD) in comparison with an analysis over 5 days using the doubly-labeled water (2H2(18)O) method (DLW). The infants that were fed continuously by nasogastric drip with 120 kcal/kg/day of special infant formula were measured twice under thermoneutral conditions in a closed system indirect calorimeter during 8 h with a 4-day interval; simultaneously isotope decay was measured by isotope ratio mass spectrometry in urine samples collected daily during 5 days from 6 h after an oral dose of 2H2(18)O on the first day of IDC, all during the 4th postnatal week. The mean differences between carbon dioxide production rate (rCO2) measured either by single 8-hour IDC or by duplicate 8-hour IDC and the 5-day DLW method, using the two-point analysis or the multipoint analysis were not significantly different from zero. The rCO2 calculated from the DLW method using the two-point analysis differed -1.4 +/- 1.7% from that measured by the multipoint analysis. The mean differences between the metabolic rate estimated from 8 h of IDC and from the 5-day DLW method based on a measured RQ of 0.90 was -6.7 +/- 6.2% and based on the RQ of the feeding -4.5 +/- 6.0%. These differences were not significantly different from zero. We conclude that IDC over 8 h and two-point DLW measurement over 5 days, both methods that can be applied with relative ease in practice, offer an adequate average estimate of energy expenditure in continuously fed preterm infants under thermoneutral conditions.

Early assessment and neurodevelopmental outcome in very low‐birth‐weight infants: implications for pediatric practice

To determine which assessments are useful, at what age, in order to identify handicaps in very Iow‐birth‐weight infants, neonatal cerebral ultrasound findings, neurological examinations and the mental scale of the Bayley Infant Scales of Development at 1 and 2 years of age were examined in relation to neurodevelopmental outcome at 3.6 years of age in a cohort of 79 high‐risk very low‐birth‐weight infants. At 3.6 years of age, a minor handicap was found in 9 (11%) and a major handicap in 4 (5%) children. Cerebral palsy was found in 9 (11%) children at 3.6 years of age and could only be diagnosed reliably at 2 years of age. For short‐term follow‐up, as feedback to the neonatalogist, the positive predictive value of intraparenchymal damage, as detected by neonatal cerebral ultrasound, was greater than the positive predictive value of a definitely abnormal neurological examination at 1 year of age. Visual handicaps (n= 4, 5%) and severe hearing deficits (n= 1, 1 %) were all detected in the first year of life. A mental handicap was found in 7 (9%) children. It was impossible to predict mental handicaps for the individual child. Only 35% of the children with a mental delay at 2 years of age had a mental handicap at 3.6 years of age, whereas 35% had a normal cognitive outcome. Pediatricians therefore should be cautious in the interpretation of developmental tcst results in infancy. Long‐term follow‐up is essential for the child and its parents.

Simultaneous Measurement of the Rates of Appearance of Palmitic and Linoleic Acid in Critically Ill Infants

Lipolysis has been measured in humans by means of stable isotope techniques using labeled palmitic acid (PA) or glycerol as tracers. If other fatty acids(FA) such as linoleic acid (LLA) have the same rate of appearance(Ra) as PA and therefore contribute equally to oxidative and nonoxidative metabolism is unknown. We infused albumin-bound [U-13C]PA and [U-13C]LLA in seven critically ill infants (weight 3.6 ± 1.3 kg, age 57 ± 64 d) receiving 20.9 ± 5.4 kcal· kg-1·d-1 of i.v. glucose only, and measured simultaneously the Ra of PA and LLA from the isotopic enrichment of plasma FFA by mass spectrometry. A needle biopsy of the s.c. adipose tissue was obtained for FA composition. PA in adipose tissue was higher than LLA (40 ± 6.7 versus 5.4 ± 3.2 mol%,p < 0.001). The Ra values of PA and LLA were 5.73 ± 2.79 and 1.34 ± 0.92μmol·kg-1·min-1, respectively (p = 0.005). However, the ratio of the FAs Ra to their respective mol% values in adipose tissue was lower for PA than for LLA (0.15 ± 0.06 versus 0.25 ± 0.06, p = 0.02). TheRa of LLA acid was higher than could be expected from the FA composition of adipose tissue, thus indicating a preferential release of LLA during lipolysis. In critically ill infants receiving only i.v. glucose, the contribution of LLA to the oxidative and nonoxidative metabolism may be larger than what assumed from the FA composition of plasma and adipose tissue.