Lead, cadmium and mercury levels in pregnancy: the ...

Journal of Developmental Origins of Health and Disease (2014), 5(1), 16–30. © Cambridge University Press and the International Society for Developmental Origins of Health and Disease 2013 doi:10.1017/S2040174413000500

R EV I EW

Lead, cadmium and mercury levels in pregnancy: the need for international consensus on levels of concern C. M. Taylor*, J. Golding and A. M. Emond Centre for Child and Adolescent Health, School of Social and Community Medicine, University of Bristol, UK

For heavy metals that have any degree of transfer though the placenta to the fetus, it is unlikely that there are safe limits for maternal blood levels. The only means of reducing fetal exposure is to minimise maternal exposure. There are few recommendations for levels of concern. With the exception of US recommendations for maternal Pb levels, but there are no international levels of concern or cut-off levels specifically for pregnancy for heavy metals, so that comparisons can generally only be made with national reference values relating to similar physiological statuses or age groups. These include recommendations for Cd levels by Germany (reference value for non-smoking adults aged 18–69 years, 1 µg/l) and for Hg by Germany (reference value for adults age 18–60 years with fish intake ⩽3 times per month, 2.0 µg/l) and the USA (cut-off level for women, 5.8 µg/dl). To illustrate the lack of cohesion, we present data on blood Pb, Cd and Hg levels from pregnant women enroled in the UK Avon Longitudinal Study of Parents and Children study and compare the values with present levels of concern and recommended cut-off values. We also compare the levels with those found in other groups of pregnant women worldwide to strengthen the database for the development of levels of concern in pregnancy. The need for clarity of terminology in describing levels of concern is discussed. There is a pressing need for international consensus on levels of concern for all age groups and physiological statuses, particularly for pregnancy. Received 3 April 2013; Revised 18 October 2013; Accepted 25 October 2013; First published online 28 November 2013 Key words: ALSPAC, cadmium, lead, levels of concern, mercury, pregnancy

Introduction Industrial pollution, as well as natural processes, have increased the exposure to heavy metals in recent years. The effect of heavy metals on public health has been of increasing concern, and in particular the effects on growth and development. During pregnancy, heavy metals may contribute to adverse outcomes including pre-eclampsia, preterm delivery and small for gestational age babies.1,2 Pb and Hg are known to be transferred freely though the placenta, although the placenta is thought to be at least a partial barrier for Cd transfer.3 The fetus is particularly vulnerable to the effects of heavy metals because of the high rate of cell division and differentiation. Thus, relatively low levels of exposure that do not harm the mother may have a profound effect on the development of the fetus and its growth and development during childhood. There are no known safe thresholds for levels of heavy metals in pregnancy, and environmental exposure should be kept to a minimum. Low-level lead exposure in children does not cause overt clinical symptoms, but has permanent effects on cognition, behaviour and school performance.4,5 Other adverse outcomes may include preterm delivery and low birth weight and effects on growth, IQ and neurodevelopment.6,7 Accumulation of Cd has been associated with effects on motor and *Address for correspondence: C. M. Taylor, Centre for Child and Adolescent Health, School of Social and Community Medicine, University of Bristol, Oakfield House, Oakfield Grove, Bristol BS8 2BN, UK. (Email: [email protected])

perceptual behaviour in childhood, and may have effects at an older age that are as yet unknown.8 Hg has adverse neurodevelopmental effects, including associations with learning and memory deficits, delayed cognitive performance, preterm birth and birth defects.9 For heavy metals that have any degree of transfer though the placenta to the fetus, there are no identified safe limits for maternal levels. The only means of reducing fetal exposure is to minimise maternal exposure. It is therefore important to have reliable data on blood levels of heavy metals in pregnancy from a range of countries and types of populations, which have a variety of environmental exposures, as well as data on the health effects in the resulting children. There are very few recommendations for levels of concern for Pb, Cd and Hg, and only three recommendations, for Pb levels, relate specifically to pregnancy.10–12 It is therefore difficult to compare levels of Pb, Cd and Hg found in pregnancy worldwide with existing levels of concern as they do not reflect pregnancy, as an especially critical time for the effects of exposure. To illustrate the lack of cohesion, we present data on blood Pd, Cd and Hg levels from pregnant women enroled in the UK Avon Longitudinal Study of Parents and Children (ALSPAC) study in 1991–1992 and compare the values with these levels of concern and recommended cut-off values, and with levels found in other groups of pregnant women worldwide. We discuss the need for clarity of terminology in describing levels of concern and the need for international consensus on levels of concern particularly for children and pregnant women.

Lead, cadmium and mercury levels in pregnancy Terminology in national recommendations There is a variety of terminology used in national recommendations. Although ‘reference value’ is used by several bodies to indicate the upper margin of the current background exposure of the general population to a given environmental toxin at a given time (e.g., Germany13) it implies a degree of normality and an associated physiological requirement – but this is not applicable to Pb, Cd or Hg levels. Other bodies refer to a ‘benchmark dose’ level from which a ‘reference dose’ for daily oral exposure can be derived (e.g., the USA14). The NoObserved-Adverse-Effect Level (NOAEL; the highest exposure level at which there are no biologically significant increases in the frequency or severity of adverse effects between the exposed population and a control population) and the LowestObserved-Adverse-Effect Level (LOAEL; the lowest exposure levels at which there are biologically significant increases in the frequency or severity of adverse effects between the exposed population and a control population) would appear to be more helpful in this context.

17

of blood Hg in pregnancy in the UK, and only the second report of Cd in pregnancy in the UK, to our knowledge. Methods For details of the methods of sample collection and analysis, see Supplementary Text S1. In brief, the study sample was derived from the ALSPAC study, a population-based study investigating environmental and genetic influences on the health, behaviour and development of children. Ethics approval for the study was obtained from the ALSPAC Ethics and Law Committee and Local Research Ethics Committees. Whole blood samples were collected from pregnant women in acid-washed vacutainers and analysed by inductively coupled plasma mass spectrometry in standard mode [R. Jones, Centers for Disease Control and Prevention (CDC), Bethesda, MD, USA]. The analyses were completed on 4285 women for Pb, 4286 women for Cd and 4134 women for Hg. Results and comparison with national levels of concern

National recommendations for levels of concern for Pb, Cd and Hg There are few recommendations available nationally for Pb, Cd and Hg levels, and none that relate specifically to pregnant women, with the exception of recommendations for Pb levels in pregnant women by the US Centers for Disease Control and Prevention,10 the Association of Occupational and Environmental Clinics11 and the American College of Obstetricians and Gynecologists.12 Some recommendations focus specifically on health and safety in the workplace and relate to monitoring of individuals rather than populations (e.g., the US Department of Labour Occupational Safety and Health Administration, and the Health and Safety Executive in the UK). A summary of recommendations for Pb, Cd and Hg for children or adults, or women were given specifically, are shown in Table 1. In Germany there is regular monitoring of heavy metal levels in the population,18 as there is in the USA though the National Health and Nutrition Examination Survey (NHANES) programme. There are only three countries that have published levels of concern for heavy metals to our knowledge: Pb (USA, Australia and Germany13,15,17), Cd (Germany only13) and Hg (both USA and Germany9,13,14,16). There is clearly a need for an international consensus on levels of concern for values for heavy metals, particularly for pregnant women, which should be founded on regular monitoring in this group. Comparison of data from the ALSPAC study with national levels of concern The ALSPAC study The data from the ALSPAC study represent the largest cohort on which blood levels of Pb, Cd and Hg in pregnancy have been reported. In addition, they provide the first reported level

The characteristics of the ALSPAC study population with blood Pb levels have been presented elsewhere.19 In brief, the subsample was slightly older and better educated that the rest of the ALSPAC sample, but were similar in terms of social class, housing status, rate of cigarette smoking and parity. The mean blood levels of Pb, Hg and Cd in the ALSPAC study are shown in Table 2. As well as showing a summary of the recommendations for levels of concern for blood Pb, Cd and Hg, Table 1 also shows the percentage of pregnant women in the present sample that had blood levels of Pb, Cd or Hg at or above each recommended value. Several values were markedly greater than the recommended values: for example, 14.4% of blood Pb levels were above the value of 5 µg/dl suggested as an upper limit by the US Association of Occupational and Environmental Clinics, the Centers for Disease Control and Prevention and the American College of Obstetricians and Gynecologists,10–12 and 39.0% of blood Hg values were above the reference level of 2.0 µg/l suggested by the German Federal Environment Agency13 for adults aged 18–60 years old with fish intake ⩽3 times per month. Comparison of blood levels of Pb, Cd and Hg in pregnancy in ALSPAC with previous studies worldwide Comparisons of blood levels of Pb, Cd and Hg in pregnancy from the ALSPAC study with previous studies worldwide are shown in Tables 3–5, respectively). In summary, the levels of Pb, Cd and Hg were higher than in similar developed countries. As has been discussed in detail for Pb,19 there are several factors that should be taken in account when comparing maternal blood levels between studies. (1) The sample sizes in published studies range widely (see Tables 3–5). (2) Several analytical techniques for analysis have been used, with varying levels of accuracy, including commercial kits, atomic

18

Pb

Country

Recommending body

USA

Centers for Disease Control and Prevention15 American College of Obstetricians and Gynecologists12 Centers for Disease Control and Prevention10 US Association of Occupational and Environmental Clinics11 Federal Environmental Agency13

USA USA USA Germany Australia Cd

Germany

Australian Government, National Health and Medical Research Council17 Federal Environmental Agency13

Hg

Germany

Federal Environmental Agency13

USA

Committee on the Toxological Effects of Methylmercury, Board on Environmental Studies and Toxicology, National Research Council9 Mahaffey et al.16 US Environmental Protection Agency14

– USA

Description of group that recommendation applies toa

Value (whole blood level)

Percentage of pregnant women with blood level greater than or equal to ‘reference value’ in the ALSPAC study

Reference value for childrenb

5 µg/dl

14.4 (619/4285)

Pregnant women

Maximum 5 µg/dl

14.4 (619/4285)

Pregnant women

Maximum 5 µg/dl

14.4 (619/4285)

Pregnant women

Maximum 5 µg/dl

14.4 (619/4285)

Reference value for women aged 18–69 years All population

7 µg/dl

3.0 (130/4285)

10 µg/dl

0.4 (15/4285)

1 µg/l

Reference value for non-smoking adults aged 18–69 years Reference value for adults aged 18–60 years with fish intake ⩽3 times per month

2.0 µg/l

BMDL in cord blood

5.8 µg/l

18.9 (812/4285) (smokers plus non-smokers) 3.6 (107/2987) (non-smokers only) 44.0 (1821/4134) (fish-eaters and non-fish eaters) 39.0 (805/2065) (eat white fish never/rarely/once per 2 weeks) 40.4 (1072/2652) (eat oily fish never/rarely/once per 2 weeks) 45.5 (1535/3374) (eat shellfish never/rarely/once per 2 weeks) 0.9 (38/4134)

Reference dose level for womenc Blood level equivalent to the EPA’s recommendation for maximum oral reference dose (0.1 µg/kg per day)

3.5 µg/l 5.8 µg/l

8.6 (357/4134) 0.9 (38/4134)

ALSPAC, Avon Longitudinal Study of Parents and Children; BMDL, Benchmark dose level; EPA, Environmental Protection Agency. Terminology is that used by the recommending body. b Level at which monitoring in children will be carried out (based on population of children in the USA in the top 2.5% of blood Pb levels). c Suggested as a revision of US National Research Council recommendation. a

C. M. Taylor et al.

Table 1. Summary of recommendations for levels of concern for blood Pb, Cd and Hg

Lead, cadmium and mercury levels in pregnancy Table 2. Maternal blood levels of Pb, Cd and Hg in the ALSPAC study

n Mean ± S.D. Median Geometric mean Range

Pb (µg/dl)a

Cd (µg/l)

Hg (µg/l)

4285 3.67 ± 1.47 3.41 3.42 0.29–19.14

4286 0.58 ± 0.63 0.29 0.37 0.14–6.30

4134 2.07 ± 1.10 1.86 1.83 0.17–12.76

ALSPAC, Avon Longitudinal Study of Parents and Children. Values reported in Taylor et al.19

a

absorption spectroscopy and inductively coupled plasma mass spectroscopy (e.g., for Pb, Al-Jawari et al.,43 Al-Saleh et al.28 and Osman et al.,3 respectively). (3) There is a range of gestational ages at which the blood samples have been taken (for Pb for example, at 12 weeks of gestation,57 at birth,61 or shortly after birth.46 (4) The date that the study was performed is not always reported (see Tables 3–5). Lead The blood Pb levels reported here from the ALSPAC study (3.67 ± 1.47 µg/dl) are higher than those in similar developed countries at about the same time (e.g., Sweden3 (median 1.14 µg/dl) and the USA15 (1.9 ± 1.68 µg/dl); see Table 3).19 Assuming a fetal:maternal ratio of blood lead of about 0.8,27,41,83 this would result in a mean neonatal blood lead level of about 2.9 µg/dl. The level is, nevertheless, in alignment with a trend for a gradual reduction in values in pregnant women with time in the UK (Fig. 1), in accordance with the eventual complete bans on lead in petrol (in 2000 following gradual withdrawal during the 1990s) and paint (in 1992), replacement of lead water pipes, and a reduction in cigarette smoking levels in the UK. Although these data are the most recent reported levels in pregnancy in the UK, Primatesta et al.84 reported levels of 2.0 ± 0.05 µg/dl (geometric mean) in women of childbearing age in 1995 in the UK, which is also in accordance with this downward trend. Cadmium To our knowledge there are no studies on blood Cd levels in pregnancy in similar developed countries at about the same time as our study, but the present results for blood Cd in the ALSPAC study (0.58 ± 0.63 µg/l) are slightly higher than more recent studies (e.g., USA20 (geometric mean 0.18 µg/l) and Sweden23 (0.30 µg/l); see Table 4). As cigarette smoking is one of the main predictors of blood Cd levels,22,85 these disparities may reflect a decline in cigarette smoking in developed countries. The mean level of blood Cd is slightly lower than that reported by Kovar et al.86 in the UK (0.69 ± 0.65 µg/l), although their sample size was very small (n = 28).

19

Mercury Unlike Pb and Cd, there is no reason to believe that Hg levels are falling with time in pregnant women. Hg bioaccumulates in fish and shellfish, and so in countries where it would be expected that fish consumption would be high, blood Hg levels in pregnancy tend to be higher, for example, 5.18 µg/l (geometric mean) in Japan73 and 12.8 ± 16.8 µg/l in Greenland,62 although conversely levels in northern Norway22 were reported as being only 1.5 ± 1.1 µg/l (this level was not remarked on as being unexpectedly low by the authors). In developed countries with an assumption of moderate fish intake, blood Hg levels in pregnancy tend to be lower, for example, Austria24 (median 0.7 µg/l), Slovakia72 (median 0.63 µg/l) and the USA75 (geometric mean 0.97 µg/l) (see Table 5). As discussed by Golding et al.,87 the median value for Hg in the present study was more than twice that found in the USA in women of child-bearing age based on NHANES data (median values 1.86 v. 0.89 μg/l),75 as were the 10th (0.81 v. 0.15 μg/l) and 25th (0.99 v. 0.38 μg/l) centiles, respectively. The 90th and 95th centiles, however, were higher in the USA study than in the present study (3.33 v. 4.83 and 4.02 v. 5.98 µg/l, respectively). Golding et al.87 stated that this difference is unlikely to be due to a difference in consumption of seafood as consumption is less in the USA than in the UK, and Hg levels in seafood in the UK are generally higher than those in the USA.88 Similar comparisons can be made with NHANES data from 1999 to 2006.89 The mean level reported in the present study (2.07 ± 1.10 µg/l) is therefore slightly higher than for similar developed countries, although even higher levels have been found in pregnant women following IVF in the USA for women with a clinical (v. a biochemical) diagnosis of pregnancy (2.84 ± 1.86 µg/l)21 and in pregnant women living or working near the site of the World Trade Center in the USA (levels were not greater than in a control group; overall mean 2.29 ± 2.33 µg/l.71 There are no other published reports of blood Hg levels in pregnancy in the UK to our knowledge. Summary The relatively high levels of blood Pb, Cd and Hg found here may reflect the rich industrial history of the UK. With regard to Pb in particular, the final ban on leaded petrol came in force in 2000 in the UK, which is somewhat later than in other developed countries (e.g., Germany 1988, Sweden 1992, Canada 1993, USA 1995). In addition, the Avon area, in which the city of Bristol is located, has relatively high concentrations of Pb and Cd in soil and rock,90 and there has been Pb mining around the city since Roman times. Industries that could have contributed to local levels of environmental contamination with heavy metals in the Avon area include a tobaccoprocessing factory, brass works, a hat-making factory and a lead-sheet rolling mill. Indeed, a lead-shot chimney in the centre of the city and a smelting works on the north-west fringe of the city both operated until the early 1990s. Levels of Pb in air samples near the smelting works site in the early 1990s were

20

Table 3. Comparison of blood Pb levels in pregnancy in studies published from 2000 onwards Country

Blood Pb (µg/dl)

n

Year of survey

Developed countriesa Sanders et al.20 Bloom et al.21

USA (North Carolina) USA (San Francisco)

0.890 (0.19–7.72)b 0.95 ± 0.48

211 25

2009–2011 –

0.85 ± 0.65 Hansen et al.22 Gerhardssson and Lundh23 Gundacker et al.24

Norway Sweden Austria (Vienna)

0.82 ± 0.04 (0.22–4.11) 1.1 (0.42–7.9)c 2.5 (1.04–8.4)c

Rastogi et al.25 d Harville et al.26 Schell et al.27

USA USA (Pittsburg, IL) USA (Albany, NY)

1.7%>10 1.93 (0.55–4.70) 1.9 ± 1.68 1.8 ± 1.63 1.8 ± 1.65

Smargiassi et al.28 d

Canada (Montreal)

Osman et al.3

Sweden (Stockholm)

Present studye Developing country or location with high exposure identified by the authorsa Motawei et al.29

UK (Bristol area)

Egypt (Mansoura)

Menai et al.30

France (Poitiers and Nancy)

Njoku and Orisakwe31 Ugwuja et al.32 Tiwari et al.33

Nigeria (Owerri, Imo state) Nigeria (Abakaliki) India (Lucknow)

Adekunle et al.34 Vigeh et al.35

Nigeria (Lagos) Iran (Tehran)

2.1 ± 1.7b 1.139 (0.21–4.77)c 3.67 ± 1.47

37.68 ± 9.17 14.5 ± 3.18 1.7 (IQR 0.1–14.1)c 99 ± 123 36.4 ± 18.5 (2.7–73.8) 1.84 ± 0.12 1.98 ± 0.13 2.61 ± 0.11 3.62 ± 0.17 59.5 ± 2.1 3.72 ± 2.03 4.52 ± 1.63

28 210 100 52

2007–2009 2002–2003 2005

6990 140 211

1992–1995 1992–1998

–

88

1994–1999

4284

1991–1992

99 349 50 50 50 25 214 304 44

Chemical pregnancy after IVF Biochemical pregnancy after IVF

Chart review

160

115 25 901

Comment

– 2002 – – –

Socioeconomically disadvantaged group in first, second and third trimester Low petrol Pb exposure

Exposure effect

NR No association with pregnancy rate after IVF

NR NR Negative association with birth weight No associations with birth length or head circumference NR NR Association with neonatal blood Pb level

NR Negative associations of cord blood Pb level with birth weight, head circumference and length NR

Pre-exclampsia No pre-eclampsia Mid-pregnancy

Control Mild anaemia Moderate anaemia Severe anaemia

2006–2008 Term birth Preterm birth

Association with pre-eclampsia No association with birth weight or fetal growth restriction NR NR Negative associations with changes in maternal oxidant-antioxidant balance and haematological parameters NR Association with preterm birth

C. M. Taylor et al.

Authors

Saudi Arabia (nr Riyadh) France (Nancy and Poitiers)

2.897 ± 1.851 1.84 ± 1.21 (1.2–2.0)

Mirghani38

Saudi Arabia (Aseer region)