Main

PCB and HCB are ubiquitous compounds that have carcinogenic and teratogenic properties (1, 2). They are chemically very stable and readily fat-soluble substances, so that they accumulate in the food chain. Therefore, the most important source of PCB and HCB uptake in man is via diet (3). Although the amounts of PCB and HCB ingested by this route are suggested to have no adverse effects on health in the general population (4), the prenatal uptake of PCB and HCB is a considerable problem in infants and young children (5), possibly leading to long-lasting adverse health effects (6, 7).

Parental smoking considerably contributes to childhood morbidity (8). Maternal smoking during pregnancy has severe adverse health effects on the fetus and newborn child (9). Spontaneous abortion, abruptio placentae, and intrauterine growth retardation are the most common clinical problems (10–12), whereas the correlation between maternal smoking and childhood cancer is still under discussion. We were recently able to demonstrate the presence of a tobacco-specific transplacental carcinogen, a derivative of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, in the urine of neonates born to smoking mothers (13). Finette et al. (14) found characteristic mutations in the HPRT gene in cord blood T lymphocytes associated with passive maternal exposure to tobacco smoke. Both studies give a possible biologic link between intrauterine exposure to environmental tobacco smoke and the later occurrence of childhood cancer (15) as suggested by some recent investigations (16, 17).

PCB and HCB have tumor-promoting properties if they are applied together with transplacental carcinogens, e.g. 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, in neonatal mice (18). Despite this obvious cocarcinogenic effect of PCB and tobacco-specific nitrosamines in offspring of smoking parents, to date there are no studies about the influence of parental smoking on the PCB and HCB concentrations in newborns. It was the purpose of the present study to investigate whether parental smoking by itself would increase the prenatal uptake of such organochloric compounds.

METHODS

The present study was approved by the Committee on Ethics in Medical Research of the Philipps-University Marburg, Germany. With the informed written consent of the parents, blood samples were taken from 80 full-term healthy neonates. Thereby, only those newborns whose parents had always lived in Fulda (a small German town with approximately 100 000 inhabitants and a large rural catchment area) and environs were included in the study. This restriction was made to exclude differences in the maternal burden with PCB and HCB caused by previous environmental influences. In addition, no mother had ever worked in PCB and HCB processing industries.

Blood specimens were obtained from an unblocked peripheral vein within the first 12 h of neonatal life, in every case before the first oral feeding. Venous blood was sampled by directly dropping into minicaps without addition of an anticoagulant and immediately centrifuged (750 ×g; 5 min.). The supernatant liquid (serum) was transferred to glass vessels that had been rinsed with acetone and petroleum ether. They were then kept deep-frozen at −20°C until analysis.

Analytic methods.

The serum was homogenized by shaking it with formic acid (19). Then the PCB together with the persistent halogenated pesticides and 2,2′,6,6′-tetrachlorobiphenyl and Mirex, the two substances that served as internal standards, were extracted by shaking the sample with heptane. Lipids, which were also extracted, were removed by filtering the extract through a silica gel column (20). The cleansed organic phase was concentrated and analyzed by means of capillary gas chromatography, and an electron capture detector served as a detector. Because of the numerous PCB congeners, quantitative determination of all the components is too time-consuming in practice (21). Six PCB congeners [International Union for Pure and Applied Chemistry (IUPAC) no. 28, 52, 101, 138, 153, and 180] were analyzed. Their concentrations allow estimation of the whole body burden with PCB (22). A solution containing these PCB in n-heptane that also included the two internal standards was used for calibration. The results were calculated by dividing the resulting peak areas by the area of the corresponding internal standard and corrected with the response factor specific for the substance (23).

HCB present in its free form in the native blood samples was extracted directly from serum by using isooctane (2,2,4-trimethylpentane). The substance was separated in a quartz capillary with a chemically bonded silicon phase (OV 1701). An electron capture detector served as a detector. The calibration standards were prepared using serum samples of persons who were not previously exposed to HCB. These samples were spiked with known quantities of HCB, processed in the same manner as the assay samples, and analyzed. Evaluation was carried out with the help of calibration curves in which the peak areas were plotted as a function of the concentrations used (24, 25).

Statistical analysis.

Information about parental smoking behavior, the geographic origin of the families, and their actual and previous working places was recorded by history taking, and the results were documented in a standardized questionnaire. Because the range of concentrations of all measurements was not distributed normally, statistical analysis was done by nonparametric testing only. Mean differences in group comparisons were calculated using the Wilcoxon test for independent samples. A p value below 0.01 was regarded as statistically significant.

Two newborns with birth weights below 2000 g were included in the study (see Table 1). Although we are concerned about these very small birth weights in two apparently healthy and full-term neonates, we have no explanation other than a mistake in the determination of the calculated birth dates of both patients.

Table 1 Personal data (median, range) of newborns and their parents in the study groups of active smoking mothers, passive smoking mothers, and nonsmoking families * No. of cigarettes smoked per day by the fathers.
Full size table

RESULTS

The study group consisted of 48 male and 32 female newborns. Mean gestational age was 40 wk (range, 38 to 42 wk), and mean birth weight was 3335 g (1670 to 5450 g). The mean maternal and paternal ages were 28 (19 to 40 y) and 30 y (18 to 52 y), respectively. Twelve mothers (15%) and 41 fathers (51.3%) had smoked during the pregnancy. In eight families, both parents were smokers. These families were assigned to the study group of active smoking mothers (see below). Except for the parents, no other smokers were present in any family. All parents who denied smoking during the pregnancy also denied intense exposure to environmental tobacco smoke, e.g. in their workplace. Therefore, we composed three study groups for statistical analysis:1) active smoking mothers (n = 12);2) passive smoking mothers, i.e. smoking fathers (n = 33); and 3) nonsmoking families (n = 35). Personal data for each group are given in Table 1.

According to the analytic method used in the present study, the limit of detection was 0.2 μg/L for the PCB congeners 28, 52, and 101 and 0.1 μg/L for the PCB congeners 138, 153, and 180 as well as HCB, respectively. The samples containing nondetectable levels of PCB and HCB were regarded as zero (substance not present) in statistical analysis. The PCB and HCB concentrations in the whole study population were as follows (median, range): PCB congeners 28, 52, and 101 were below the detection limit in every sample; PCB 138, 0.34 (<0.10–1.01); PCB 153, 0.42 (<0.10–1.42); PCB 180, 0.17 (<0.10–0.78); total PCB (sum of PCB congeners 138, 153, and 180; ΣPCB), 0.96 (<0.10–3.14); and HCB, 0.61 (<0.10–4.38). PCB congeners 138 and 153 as well as HCB were detectable in 96% and PCB 180 in 68% of all serum samples. The results of the PCB and HCB measurements in each subgroup are summarized in Table 2. PCB congener 153 regularly showed the highest serum concentrations, followed by PCB 138 and PCB 180.

Table 2 Serum concentrations of the PCB congeners 138, 153, and 180, total PCB concentration, and HCB concentration (median, range) in the study groups of active smoking mothers, passive smoking mothers, and nonsmoking families Data in underlined italics indicate a statistically significant difference (p < 0.01) in the Wilcoxon test for independent samples.
Full size table

Neonates born to active smoking mothers had the highest PCB and HCB concentrations compared with children of passive and nonsmoking mothers. These differences were statistically significant in the cases of PCB 138, ΣPCB, and HCB (Fig. 1), although a trend toward higher values also was present regarding PCB 153 and 180. Newborns of passive smoking mothers had higher PCB and HCB concentrations than children of nonsmoking families but lower values than those of active smoking mothers. These differences were statistically significant for all compounds with the exception of PCB 180 (Table 2).

Figure 1
figure 1

Neonatal serum concentrations (μg/L; median, 25th and 75th percentile by rank) of PCB 138, total PCB (sum of PCB congeners 138, 153, and 180), and HCB in children of active smoking women (n = 12), passive smoking women (n = 33), and nonsmoking families (n = 35). The differences between each study group are statistically significant (p < 0.01) for all compounds in the Wilcoxon test for independent samples.

Full size image

DISCUSSION

The present study confirms a considerable prenatal uptake of PCB and HCB in human neonates. The lower chlorinated biphenyls, e.g. the PCB congeners 28, 52, and 101, were below the detection limit of 0.2 μg/L. This corresponds to previous findings in the general German adult population (26). It is well known that the total body burden of PCB and HCB increases with rising age in adults (3). The PCB and HCB concentrations found in this study exactly fit into the age-dependent reference ranges of these substances measured in the general German population, and the distribution pattern of the PCB congeners in newborns also corresponds to that in adults (3, 26). These findings seem reasonable because the prenatal uptake of these compounds in the fetus should reflect the concentration and distribution pattern of PCB and HCB in the mothers.

To date, the influence of parental smoking during pregnancy on the neonatal PCB and HCB concentrations has not been investigated. The results of the present study clearly demonstrate that active maternal tobacco smoking during pregnancy leads to a significant increase of the neonatal burden of these compounds. Mainstream tobacco smoke contains 3500 different volatile and nonvolatile compounds, and many more have yet to be identified (27). In face of the ubiquitous distribution of PCB in the environment as well as the use of HCB as an insecticide for plants, the contamination of tobacco with these substances is within the bounds of probability and could be demonstrated in experimental investigations (28, 29). Because all families investigated in the present study came from the same geographic area in Germany (Fulda is a small town with approximately 100 000 inhabitants and a large rural catchment area) and because the study groups did not differ significantly with regard to the origin of the people and their previous working places (see above), other environmental factors influencing the neonatal PCB and HCB concentrations are very unlikely.

Exposure to environmental tobacco smoke, i.e. paternal smoking during the pregnancy, also significantly increases the neonatal PCB and HCB concentrations although to a lower extent than active maternal smoking. Beyond the well-known fact that toxic compounds released during the burning of tobacco are accumulated in significantly higher concentrations in active than in passive smokers (30, 31), two additional factors may contribute to this finding.

1) The compositions of main- and sidestream tobacco smoke differ significantly (32), so that the amount of PCB and HCB in environmental tobacco smoke may be lower than that in the inhaled fraction of mainstream smoke, independent of the dilution in indoor air.

2) Active smoking women may have accumulated higher amounts of PCB and HCB previously than the general nonsmoking or even passive smoking population, leading to an increased transfer of these substances to their unborn children. This has to be investigated in further studies, because today no data are available on the presumed differences in PCB and HCB levels between adult smokers and nonsmokers.

The results of the present study are of special importance because previous observations had confirmed that bioretained polyhalogenated hydrocarbons are capable of promoting mouse lung and liver tumors initiated perinatally with environmentally relevant nitrosamines (33–35). Transplacental exposure to the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone increased the frequency of lung tumor bearers in progeny of exposed female mice, and it could be demonstrated that sequential exposure to tobacco-specific carcinogens and PCB is effective in promoting both lung and liver tumors in mice (18). Humans are continuously exposed to a variety of xenobiotics, with an early accumulation of carcinogenic and cocarcinogenic compounds during intrauterine life as this and other studies were able to show (5, 13, 36). Further studies are urgently required to investigate the possible cocarcinogenic effects of such substances in humans, especially in the face of inconsistent epidemiologic findings regarding maternal active and passive smoking and childhood cancer (15–17). Tumor-promoting effects of environmentally abundant compounds such as PCB and HCB may be a possible clue to those unexplained observations.

Overall, the present study for the first time demonstrates a significant increase in the prenatal burden of PCB and HCB in neonates born to active and passive smoking women. PCB and HCB concentrations in the offspring of active smoking mothers were significantly higher than those of the general German neonatal population, corresponding to values found before the prohibition of PCB in Germany in 1989 (5, 37). Whether these increased levels of organochloric compounds have any negative health impacts on the affected newborns is unclear. In the face of the cocarcinogenic properties of PCB and tobacco-specific carcinogens (18), adverse effects cannot be ruled out with certainty. Further studies are required to investigate the possible risks associated with parental smoking with special consideration of the tumor-promoting properties of PCB and HCB.