Secondary Logo

Journal Logo

Original Article

Influence of seafood and vitamin supplementation on maternal and umbilical cord blood mercury concentration

Huang, Shih-Huia,b; Weng, Ken-Penc,d,e; Ger, Luo-Pingf; Liou, Huei Hanf; Lin, Ching-Chiangg,h; Wang, Chung-Chengi; Lee, Charles Tzu-Chia; Wu, Ming-Tsanga,j,k,*

Author Information
Journal of the Chinese Medical Association: May 2017 - Volume 80 - Issue 5 - p 307-312
doi: 10.1016/j.jcma.2016.11.005


    1. Introduction

    Mercury (Hg) is a toxic contaminant that can cross the placenta, and affect fetal growth and neurodevelopment.1–3 Hg toxicity to the fetus became well known after high maternal exposure incidents occurred in Japan and Iraq.2,3 Seafood consumption is the major source of Hg exposure for most people.4 However, Hg exposure associated with seafood consumption may negatively affect neurodevelopment, although seafood rich in omega-3 fatty acids is associated with enhanced neurodevelopment.5,6 This makes the potential harm of Hg exposure and the benefit of seafood consumption on health become important issues.

    According to the cord blood levels reported in Budtz-Jorgensen et al's study,7 the Environmental Protection Agency (EPA; USA) established a reference dose of 5.8 μg/L methyl Hg (MeHg) in blood for limiting exposure to MeHg in young children and women who are pregnant, are breastfeeding, or might be pregnant.8 American pregnant women are advised to avoid fish with high Hg levels, but to eat 227–340 g/wk of a variety of fish low in Hg to support fetal growth and development.9 This advice is complicated by the benefit of seafood rich in omega-3 fatty acids,5,6 and Soon et al10 reported that a significant portion of pregnant women in Hawaii consumed more than the recommended amount of seafood. Seafood is an important part of the Taiwanese diet and therefore a concern; however, seafood is only one part of the larger Taiwanese diet and should be evaluated with other nutrients in terms of mercuric intoxication. The protective effects of selenium, zinc, vitamin E, and vitamin B complex have been implicated in the alteration of Hg metabolism.10,11 Most data about the protective effects of these nutrients are derived from animal research,11 and therefore, implications of these data for human populations require further study.

    Asian-American people and women of child-bearing age, who consume seafood more frequently than other race/ethic groups, have higher mean MeHg levels.12,13 Taiwan is a large Asian island where people frequently consume seafood. Hsu et al’s14 study showed that 89% of pregnant women contain blood Hg exceeding the recommended value of 5.8 μg/L. Chien et al15 demonstrated that the mean Hg concentration in hair was 1.73±2.12 μg/g in Taiwanese women of child-bearing age, exceeding the EPA reference dose of 1 μg/g. These two studies14,15 are consistent with previous studies reporting the risk of elevated Hg in Asians.12,13 However, the data on the accumulation of Hg in pregnant women in Taiwan remain limited.14,15 Therefore, in this study, we measured total Hg in paired maternal/neonatal blood samples, and further investigated the relationship between Hg levels and seafood as well as vitamin intake during pregnancy in Taiwan.

    2. Methods

    2.1. Participants and questionnaires

    For this cross-sectional study, 145 healthy pregnant women with a mean age of 28.1 years were recruited from September 30, 2010 to May 25, 2011 in the Department of Obstetrics and Gynecology of Fooyin University Hospital in Dong Gang, Taiwan. The city of Dong Gang is a major seaside fishing area in southern Taiwan. The participants received a detailed explanation of the study procedures before consenting to participate. The research protocol was approved by the Institutional Review Board of Fooyin University (FYH-IRB-099-04-02-A). Written informed consent was obtained from all participants. Data on gestational age, prenatal examination results, and characteristics of the newborn were obtained from medical records. We used interviewer-administered questionnaires to collect information on demographic characteristics, smoking habit, alcohol drinking habit, betel-nut-chewing habit, use of Chinese medicine, seafood consumption, vitamin supplement, degree of education, etc. The semiquantitative food frequency questionnaire was derived and modified from previously validated studies.16,17 Content validity of the food frequency questionnaire was assessed by an expert panel consisting of six nutrition experts. After gathering opinions from the experts, questions without a precise content were excluded. The degree of seafood consumption was defined as the sum scores of seafood consumption items queried in the questionnaire. The score of seafood consumption was based on the answers to the following three questions. The first question was the following: “Approximately how many grams of fresh fish do you eat per day on average?” The following were the different response options to that question: (1) small amount, 0–30 g/d (1 point); (2) moderate amount, 31–60 g/d (2 points); (3) large amount, 61–90 g/d (3 points); and (4) large amount exceeding more than 90 g/d (4 points). The second question was as follows: “Approximately how many grams of shellfish do you eat per day on average?” The different response options to that question were the following: (1) small amount, 0–10 g/d (1 point); (2) moderate amount, 11–14 g/d (2 points); and (3) large amount, ≥15 g/d (3 points). The third question was as follows: “Approximately how many grams of canned seafood (fish or shellfish) do you eat per day?” The different response options to that question were the following: (1) small amount, 0–10 g/d (1 point); (2) moderate amount, 11–14 g/d (2 points); and (3) large amount, ≥15 g/d (3 points). The higher the sum sores, the greater the consumption of seafood. In our analysis, participants with 3 points were categorized into those people whose consumption approximated the EPA recommended ranges (227–340 g/wk). Those with scores over 3 were categorized into the high-seafood-consumption group whose consumption exceeded those recommendations. Those with ≤3 points were categorized into the less-seafood-consumption group. Anthropometric measurements of newborns were made by delivery room staff following standard procedures.

    2.2. Blood sampling and sample analysis

    Umbilical cord and maternal venous whole blood samples were separately collected into 9 mL standard laboratory-issued EDTA tubes. All samples were processed within 2 hours of delivery and stored at –80°C. Samples were sent to the Department of Biomedical Engineering and Environmental Sciences, Ultra Trace Micro-Analysis Laboratory at National Tsing Hua University, and were analyzed by inductively coupled plasma-mass spectrometry (7500ce; Agilent, Tokyo, Japan). The limit of detection for Hg was 0.353 μg/L (ppb). The intraday precision was tested with seven repeats of three samples of different concentrations. The average relative standard deviations were 0.49–0.71%. The interday precision was determined by analyzing the three samples three times every day, for 5 consecutive days. The average relative standard deviations were 1.01–2.76%, and the recovery was 98–102% of certified values.

    2.3. Statistical analysis

    Sociodemographic characteristics were analyzed descriptively. Hg concentrations in maternal blood (mHg) and that in umbilical cord blood (fHg) were described as mean±standard deviation and categorized into 5%, 25%, 50%, 75%, and 95% percentiles. Pearson's correlation was used to analyze the association between Hg concentration in maternal and fetal blood. Receiver operating characteristic curve analysis was used for dichotomization of the frequencies of vitamin supplements. Multivariate logistic regression was performed to further analyze the association after controlling for race/ethnicity, education, cigarette smoking, alcohol drinking, betel quid chewing, Chinese medicine use, vitamin supplementation, and seafood consumption. Unadjusted odds ratios and fully adjusted odds ratios were reported. A p value of 0.05 was considered significant. All statistical analyses were performed using SAS (version 9.3; SAS Institute Inc., Cary, NC, USA).

    3. Results

    Between September 2010 and May 2011, 345 women visited the Department of Obstetrics and Gynecology of Fooyin University Hospital to receive third-trimester prenatal checkups (after 24 weeks). A total of 150 women were excluded because of chronic diseases (diabetes mellitus and hypertension), infectious diseases, twins, and stillbirth, leaving 195 eligible candidates. Fifty of these women did not want to participate in the study, leaving us with 145 participants, all of whom signed written consent forms. None of the mothers had complications during the study period, and only singleton infants were recruited into our study. These participants agreed to provide both samples of their own blood before delivery and cord blood samples at delivery, gave complete postpartum interviews, and allowed access to medical record data.

    3.1. Characteristics of the study population

    As seen in Table 1, the participants had a mean age of 28.1±5.2 years, and prepregnancy and perinatal body mass indexes of 22.0±3.3 kg/m2 and 27.8±3.9 kg/m2, respectively. Almost 90% of the participants (84.8%) were of Han Chinese origin, with the remaining participants from various Southeast Asian countries. Almost half of the participants (49.0%) had graduated from high school. Of the participants, 9.0%, 7.6%, and 10.0% smoked cigarettes, consumed alcohol, and chewed betel quid during pregnancy, respectively. A minority (10.0%) of the participants said that they used Chinese herbal medicine within a month before birth. For logistic regression analysis of fHg, frequencies of vitamin supplements were dichotomized as low and high levels with the cutoff value (>2.5 times per week) based on the receiver operating characteristic curve analysis. Therefore, participants with vitamin supplements were categorized into two groups: those with vitamin intake >3 times/wk and those with vitamin intake ≤3 times/wk. Of all the women taking vitamins, 42.1% reported taking vitamin supplements >3 times/wk. A total of 126 women reported consuming seafood, with 61.9% being categorized into the high-seafood-consumption group. In total, these women gave birth to 71 female and 74 male neonates. Mean birth weight, birth length, head circumference, and chest circumference were 3081.6±387.3 g, 49.6±2.5 cm, 34.3±1.4 cm, and 32.6±1.6 cm, respectively. Most of the neonates (74.5%) were full term, and only 10 (6.9%) had low birth weight (<2500 g).

    Table 1
    Table 1:
    Maternal and neonatal characteristics (N = 145).

    3.2. Relationship between maternal and cord blood Hg concentration

    Table 2 presents the distribution of Hg concentrations in maternal and umbilical cord blood samples. Mean mHg concentrations and fHg were 3.6±3.9 μg/L and 5.0±7.3 μg/L, respectively. Geometric mean mHg and fHg were 2.1 μg/L and 2.5 μg/L, respectively. Hg concentrations were about 1.4 times higher in umbilical cord blood than in maternal blood; mHg were strongly correlated with fHg (r=0.76, p<0.0001). The linear regression equation for the relationship between mHg and fHg was as follows: fHg=1.405 × mHg – 0.068 (Fig. 1).

    Table 2
    Table 2:
    Data on Hg levels in maternal blood and umbilical cord blood (n = 145).
    Fig. 1
    Fig. 1:
    Association between mercury levels in maternal blood and cord blood; r = 0.76, n = 145, p < 0.0001.

    3.3. Effect of seafood and vitamin use on Hg level

    As can be seen in Table 3, 29 mothers (20%) exceeded the US EPA reference dose (maximum acceptable dose of a toxic substance) of Hg in pregnancy (Hg>5.8 μg/L). Han Chinese women did not have a significantly higher incidence of Hg>5.8 μg/L than southeast Asian women [adjusted odds ratio (AOR) 1.48, 95% confidence interval (CI): 0.42–5.16, p=0.541]. There was also no significant difference between subgroups with Hg>5.8 μg/L or Hg≤5.8 μg/L in terms of education level, cigarette smoking, alcohol drinking, betel quid chewing, and use of Chinese herbal medicine. Altogether, 97.5% of women taking vitamins >3 times/wk had Hg concentrations ≤5.8 μg/L, while 25.6% of those with high seafood consumption had maternal Hg levels of >5.8 μg/L. Women with vitamin intake >3 times/wk had a significantly lower incidence of Hg>5.8 μg/L than those with vitamin intake ≤3 times/wk (AOR 0.06, 95% CI: 0.01–0.49, p=0.008). On the contrary, women with high seafood consumption had a significantly higher incidence of Hg>5.8 μg/L than those with low seafood consumption (AOR 2.91, 95% CI: 1.04–8.15, p=0.042).

    Table 3
    Table 3:
    Logistic regression analysis of Hg levels in maternal blood (N = 145).

    4. Discussion

    In this study, we attempted to determine the effect of dietary sources on Hg levels in pregnant women. Our results showed that there was a strong correlation between mHg and fHg, and that prenatal vitamin use decreased the mHg. The effect of prenatal vitamin use on Hg levels in women during pregnancy is a very important finding.

    The geometric mean of the mHg in this series was 2.1 mg/L, a value much lower than that reported in 2007 in a study of pregnant Taiwanese woman (8.6 mg/L).14 Hsu et al14 reported 89% of the mHg exceeded 5.8 μg/L. However, in the current study, 20% of mother's blood samples exceeded 5.8 μg/L. Two reasons may explain the discrepancy of Hg levels between Hsu et al's14 and our studies. First, the study period used in two investigations was quite different. Additionally, their study14 was performed 7 years prior to ours. In recent years, pregnant women in Taiwan have been educated about the potential risk of Hg toxicity associated with seafood consumption. Second, our sample size (n=145) was larger than that in Hsu et al's14 study (n=65). Small sample size might be associated with an underpowered analysis.

    Compared with the geometric means reported for other countries, our participants had higher blood concentrations of Hg at the time of delivery than pregnant women in Canada (0.48 mg/L),18 Poland (0.83 mg/L),19 and the USA (1.6 mg/L).20 By analyzing data from the 2011–2012 US National Health and Nutrition Examination Survey, Mortensen et al's12 study showed that Asians had higher MeHg concentrations. Buchanan et al13 further reported that Asian women of child-bearing age, who consumed seafood more frequently than other race/ethic groups, had higher mean MeHg levels. In Taiwan, fish consumption during pregnancy is generally higher than that found in other countries because traditionally it is thought that eating fish during pregnancy makes for better nutrition for the neonate.14 Taiwanese studies,14,15 including our investigation, are in agreement with the previous studies reporting the risk of elevated Hg in Asians.12,13

    Our result showed a strong correlation of Hg in maternal and umbilical cord blood of paired maternal/neonatal samples. The component transport between umbilical cord and maternal serum is very complex.21 Usually, fHg is higher than mHg due to its high affinity for fetal hemoglobin.22 Previous studies,3,14 including our investigation, have also documented higher Hg levels in cord blood than in maternal blood. Although there was no significant association of Hg level and birth outcome in this study, Hg has been reported to be associated with developmental delay in children whose mothers were exposed to it during pregnancy.23,24 This raises the question as to how much seafood a pregnant woman can eat without the risk of Hg toxicity. A Taiwanese study by Chien et al25 showed that 21.6–24.3% and 45.6–57.4% of the daily Hg dose estimates exceeded the reference dose for typical and high seafood consumers, respectively.25 Their analysis suggested that the acceptable ingestion rate of fish for women during childbearing is 90.8±15.7 g/d.25 As advised by the Food and Drug Administration and EPA in 2014, pregnant woman should eat 227–340 g of a variety of fish each week from choices that are lower in Hg content.9 Further studies are required to elucidate the optimal dose of fish during pregnancy in the Asian race.

    According to our multiple logistic regression analysis, the status of prenatal vitamin use significantly decreased the mHg. It is an interesting issue about whether vitamins or dietary modifications may be used to modify Hg intoxication. Metabolized MeHg produces active oxygen species (superoxide radicals, hydroxyl radicals, singlet oxygen, and peroxides), so vitamins E and C, which have antioxidant properties, may modify MeHg toxicity.11 In a study by Al-Attar,26 administration of vitamin E was found to protect against heavy metal-induced renal and testicular oxidative stress and injuries in male mice. Another different study by Al-Attar27 reported that vitamin E protected against the heavy metal-induced liver injury in albino mice, suggesting that the attenuating effect of vitamin E might be due to its antioxidant activity. Abd El-Aziz et al28 reported that vitamin E might ameliorate some aspects of MeHg developmental toxicity in rat fetuses. Kim et al29 also found a negative association between serum folate and blood Hg concentrations in pregnant Korean women. Their findings suggest that folate is associated with the blood Hg level through its participation in the Hg detoxification process.29 These previous findings26–29 may partially explain the effect of prenatal vitamin use on decreasing maternal levels of Hg. The effect of vitamin use on heavy metals in pregnant woman still requires further investigation.

    This study had several limitations. One is that it was a single-center investigation of pregnant woman utilizing a modest sample size. Another limitation is that the contents of vitamins used by the participants in this study were not studied in detail.

    In conclusion, our results showed a positive correlation between mHg and fHg, and that high seafood consumption was an independent risk factor for a high maternal Hg level. Although an association between prenatal vitamin supplementation and reduced maternal Hg level was found, the protective effect of vitamin supplementation to modify Hg toxicity requires further detailed study.


    Part of this work was supported by grants from Kaohsiung Veterans General Hospital (VGHKS104-83, VGHUST104-G7-7-3, VGHKS105-106, and VGHUST105-G3-1-3) and Kaohsiung Medical University (MOST104-2314-B-037-012-MY2 and MOST105-2632-B-037-002).


    1. Clarkson TW, Magos L, Myers GJ. The toxicology of mercury—current exposures and clinical manifestations. N Engl J Med. 2003;349:1731-1737.
    2. Matsumoto H, Koya G, Takeuchi T. Fetal Minamata disease. A neuropathological study of two cases of intrauterine intoxication by a methyl mercury compound. J Neuropathol Exp Neurol. 1965;24:563-574.
    3. Amin-Zaki L, Elhassani S, Majeed MA, Clarkson TW, Doherty RA, Greenwood MR, et al. Perinatal methylmercury poisoning in Iraq. Am J Dis Child. 1976;130:1070-1076.
    4. United States Environmental Protection Agency. Mercury, 2013. Available at: [Accessed 15 Jan 2016].
    5. Helland IB, Smith L, Saarem K, Saugstad OD, Drevon CA. Maternal supplementation with very-long-chain n–3 fatty acids during pregnancy and lactation augments children's IQ at 4 years of age. Pediatrics. 2003;111:e39-e44.
    6. Hibbeln JR, Davis JM, Steer C, Emmett P, Rogers I, Williams C, et al. Maternal seafood consumption in pregnancy and neurodevelopmental outcomes in childhood (ALSPAC study): an observational cohort study. Lancet. 2007;369:578-585.
    7. Budtz-Jorgensen E, Grandjean P, Keiding N, White RF, Weihe P. Benchmark dose calculations of methylmercury-associated neurobehavioural deficits. Toxicol Lett. 2000;112–13:193-199.
    8. United States Environmental Protection Agency. Methylmercury reference dose for chronic oral exposure, 2013. Available at: [Accessed 15 Jan 2016].
    9. United States Environmental Protection Agency. EPA-FDA advisory on mercury in fish and shellfish, 2014. Available at: [Accessed 15 Jan 2016].
    10. Soon R, Dye TD, Ralston NV, Berry MJ, Sauvage LM. Seafood consumption and umbilical cord blood mercury concentrations in a multiethnic maternal and child health cohort. BMC Pregnancy Childbirth. 2014;14:209.
    11. Chapman L, Chan HM. The influence of nutrition on methyl mercury intoxication. Environ Health Perspect. 2000;108(Suppl 1):29-56.
    12. Mortensen ME, Caudill SP, Caldwell KL, Ward CD, Jones RL. Total and methyl mercury in whole blood measured for the first time in the U.S. population: NHANES 2011–2012. Environ Res. 2014;134:257-264.
    13. Buchanan S, Anglen J, Turyk MS. Methyl mercury exposure in populations at risk: analysis of NHANES 2011–2012. Environ Res. 2015;140:56-64.
    14. Hsu CS, Liu PL, Chien LC, Chou SY, Han BC. Mercury concentration and fish consumption in Taiwanese pregnant women. BJOG. 2007;114:81-85.
    15. Chien LC, Gao CS, Lin HH. Hair mercury concentration and fish consumption: risk and perceptions of risk among women of childbearing age. Environ Res. 2010;110:123-129.
    16. Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi J, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol. 1985;122:51-65.
    17. Ramón R, Murcia M, Ballester F, Rebagliato M, Lacasaña M, Vioque J, et al. Prenatal exposure to mercury in a prospective mother–infant cohort study in a Mediterranean area, Valencia, Spain. Sci Total Environ. 2008;392:69-78.
    18. Morrissette J, Takser L, St-Amour G, Smargiassi A, Lafond J, Mergler D. Temporal variation of blood and hair mercury levels in pregnancy in relation to fish consumption history in a population living along the St. Lawrence River. Environ Res. 2004;95:363-374.
    19. Jedrychowski W, Perera F, Rauh V, Flak E, Mróz E, Pac A, et al. Fish intake during pregnancy and mercury level in cord and maternal blood at delivery: an environmental study in Poland. Int J Occup Med Environ Health. 2007;20:31-37.
    20. Lederman SA, Jones RL, Caldwell KL, Rauh V, Sheets SE, Tang D, et al. Relation between cord blood mercury levels and early child development in a World Trade Center cohort. Environ Health Perspect. 2008;116:1085-1091.
    21. Tong XL, Wang L, Gao TB, Qin YG, Qi YQ, Xu YP. Potential function of amniotic fluid in fetal development—novel insights by comparing the composition of human amniotic fluid with umbilical cord and maternal serum at mid and late gestation. J Chin Med Assoc. 2009;72:368-373.
    22. Iyengar GV, Rapp A. Human placenta as a ‘dual’ biomarker for monitoring fetal and maternal environment with special reference to potentially toxic trace elements. Part 3: toxic trace elements in placenta and placenta as a biomarker for these elements. Sci Total Environ. 2001;280:221-238.
    23. Bakir F, Damluji SF, Amin-Zaki L, Murtadha M, Khalidi A, al-Rawi NY, et al. Methylmercury poisoning in Iraq. Science. 1973;181:230-241.
    24. Dolbec J, Mergler D, Sousa Passos CJ, Sousa de Morais S, Lebel J. Methylmercury exposure affects motor performance of a riverine population of the Tapajos river, Brazilian Amazon. Int Arch Occup Environ Health. 2000;73:195-203.
    25. Chien LC, Yeh CY, Jiang CB, Hsu CS, Han BC. Estimation of acceptable mercury intake from fish in Taiwan. Chemosphere. 2007;67:29-35.
    26. Al-Attar AM. Antioxidant effect of vitamin E treatment on some heavy metals-induced renal and testicular injuries in male mice. Saudi J Biol Sci. 2011;18:63-72.
    27. Al-Attar AM. Vitamin E attenuates liver injury induced by exposure to lead, mercury, cadmium and copper in albino mice. Saudi J Biol Sci. 2011;18:395-401.
    28. Abd El-Aziz GS, El-Fark MM, Saleh HA. The prenatal toxic effect of methylmercury on the development of the appendicular skeleton of rat fetuses and the protective role of vitamin E. Anat Rec (Hoboken). 2012;295:939-949.
    29. Kim H, Kim KN, Hwang JY, Ha EH, Park H, Ha M, et al. Relation between serum folate status and blood mercury concentrations in pregnant women. Nutrition. 2013;29:514-518.

    mercury; pregnancy; seafood; vitamin

    © 2017 by Lippincott Williams & Wilkins, Inc.