Introduction

Human milk has been recognized as the main source of nutrition for infants (Agostoni et al. 2010; Tudehope 2013; de Halleux et al. 2017) because of its abundant nutritional content: energetic macronutrients and fatty acids (Grote et al. 2016), bioactive content, important enzymes (Khaldi et al. 2014), antimicrobial and anti-inflammatory properties (Laiho et al. 2003; Lepage and Perre 2012), and docosahexaenoic (omega-3 fatty) acid (Affinito et al. 1996). The antibodies (Lepage and Perre 2012; Rogier et al. 2014; Sherry et al. 2015; Marangoni et al. 2016) in human milk reduce the risk of diseases, such as necrotizing enterocolitis (Zhou et al. 2015; Buckle and Taylor 2017; Pammi and Suresh 2017), nosocomial infections, sepsis (Cortez et al. 2018), lung disease (Villamor-Martinez et al. 2018), severe diarrhea, otitis media, and respiratory infections (Comite de nutrition de la Societe francaise de et al. 2013). In addition, human milk has been linked with long-term health conditions, such as decreased susceptibility to allergies and obesity (Admyre et al. 2007; Newburg et al. 2010; Crume et al. 2011). Human milk consumption is associated with improved neurodevelopmental outcomes in vulnerable infants (Lucas et al. 1992; Stefanescu et al. 2016; Cortez et al. 2018).

For Taiwanese infants, breastfeeding has been promoted by the Health Promotion Administration (HPA), Ministry of Health and Welfare since 2001 (Taiwan-Health-Promotion-Administration 2015). The HPA reported that the exclusive breastfeeding rate 1 month after birth rose dramatically from 5.4% in 1989 to 61.8% in 2011, when 24.2% of 6-month postpartum women reported exclusive breastfeeding (Taiwan-Health-Promotion-Administration 2012). As for exclusive breastfeeding rate, 40.1% of Taiwanese mothers reported exclusive breastfeeding at 1 month and 29.3% at 2 months postpartum (Chang et al. 2019). According to a 2016 telephone interview in Taiwan by the HPA, the breast milk feeding rate for infants under 4 months was more than 50% and 44.8% for infants under 6 months, which is comparable to the 41% recorded in the global database (MOHW-HPA 2019).

Although the advantages associated with breastfeeding are well known, heavy metal toxicity in human milk has received increased research attention (Chao et al. 2014; Jeong et al. 2017; Dix-Cooper and Kosatsky 2018; Vahidinia et al. 2019). Mining, battery manufacturing, drinking water tubes, paint, makeup, and other consumer products are sources of Pb exposure (Nriagu et al. 1996; Piomelli 2002; Cardoso et al. 2014); heavy metals can also be found in the soil, water, air, and the food chain. The concentrations of toxic and trace metals are higher in pig and cattle verscera than in the muscle (Niemi et al. 1991; Sedki et al. 2003; Akan et al. 2010). The postpartum confinement is popular in many cultures, especially in traditional Chinese culture worldwide (Dennis et al. 2007). Including Taiwan, dietary pattern of Chinese-style postpartum confinement provides pig liver, pig kidney, fish, sea food, and chicken heart etc. to postpartum mothers during the confinement month after childbirth. As mentioned, those foods make women exposed to heavy metals. If Pb and Hg, which are both regarded as neurotoxicants, accumulate in human breast milk (Braun et al. 2008; Díez 2009; Caserta et al. 2013; Kvestad et al. 2018), they will be passed down to infants (Sowers et al. 2002; Park et al. 2018).

Attention should be given to the metal effects on children for several reasons. First, their tolerance for toxins is lower than that of adults (Needleman 2004). Second, maternal prenatal nutrition and a child's nutrition during the first 2 years are crucial for healthy neurodevelopment and lifelong mental health (Al-Saleh et al. 2016; Schwarzenberg and Georgieff 2018). Therefore, low levels of Pb and methylmercury (MeHg) exposure from breastfeeding can pass through the blood–brain barrier and cause neurological and developmental disorders (Chien et al. 2006; Díez 2009; Kern et al. 2020).

Considerable research reported heavy metal inspections in foodstuffs. In Algeria, the levels of Pb in 15 fruits and vegetables measured 12.33–39.33 mg/kg dry weight, and several values exceeded the international threshold. The estimated daily intake (EDI) and total hazard quotient (HQ) were also all higher than the thresholds set by the Food and Agriculture Organization and World Health Organization (FAO/WHO) (Cherfi et al. 2014). In Uganda, measurements of Pb levels in beef, mutton, chicken, and their viscera showed levels ranging from 0.04 to 1.11 mg/kg, which were higher than the suggested level (0.5 mg/kg wet weight) set by the European Food Safety Authority. Moreover, the risk assessment showed that the HQ was above 1 for sheep intestines and beef liver, kidney, and lung, whereas the HQ of 0.99 was detected for chicken liver (Ogwok et al. 2020). Burger et al. measured arsenic (As), cadmium (Cd), chromium (Cr), Pb, manganese (Mn), Hg, and selenium (Se) levels, and although the average Hg level in all fish did not exceed the 1.0 ppm limit set by the Taiwan Food and Drug Administration (FDA), the 18% Hg level in tuna did. Furthermore, the Pb levels in shellfish and shrimp exceeded the Codex Alimentarius Commission suggestion of 0.2 ppm (Burger and Gochfeld 2004, 2005). In Taiwan, regarding the measurement of Cd, Pb and Hg in 1939 foodstuffs, only Pb levels in one chickpea exceeded 0.2 µg/g, whereas the Cd level in one shellfish exceeded the 2 µg/g set by the FDA (Lee et al. 2018). Although numerous studies have reported metals levels in foodstuff, the measurements of toxic metals in breast milk are still limited (Iwai-Shimada et al. 2015; Al-Saleh et al. 2016; Vollset et al. 2019), especially for longitudinal measurement on the breast milk donated to the human milk bank.

The study sought to determine whether the demographic characteristics, life habits and dietary pattern of breastfeeding mothers is related to the accumulation of Pb, Hg, and MeHg in their breast milk, and to assess the health risk of Pb and Hg exposure of the offspring and recipients through breast milk. Finally, further to evaluate whether these metals should be measured before the milk is donated.

Materials and Methods

Subject Enrollment

This study was approved by the Ethics Committee of the National Cheng Kung Hospital (Tainan, Taiwan, encoded: A-ER-108-193). The study was conducted at the Taiwan Southern Human Milk Bank (TSHMB) of National Cheng Kung University Hospital, the second human milk bank in Taiwan. TSHMB is a non-profit unit supported by HPA Taiwan. The operation of the TSHMB is similar to that of the first bank in Taiwan (Chang et al. 2013) and other banks worldwide (Moro et al. 2019). One local rule of the TSHMB is that the donor must make contact within 3 months postpartum and be willing to cooperate with a long-term donation. Single-batch donations are not allowed. All participants must prove to have a healthy medical history and have their blood tested for infectious diseases, including hepatitis A, B, and C virus, human immunodeficiency virus, Human T-lymphotropic virus, and cytomegalovirus. The cost of tests is supported by TSHMB. All participants are volunteers without incentive payments and commit to donate long-term. After receiving certificated of blood test report, donors are supplied with sterile glass bottles. Expressed or pumped milk was collected and refrigerated by donors immediately. A staff from the TSHMB visits the donors once a week to bring the frozen milk back for processing and pasteurization. Under a friendly support system, qualified donors can donate breast milk by their will without time limitation but maximally up to 9 months postpartum (Chen et al. 2022).

When donor mother contacted the TSHMB by themselves through the internet to donate breast milk, mother-infants-dyads were prospectively invited to this study, and consent was obtained before the first donation. The exclusion criteria were that the delivery could not be a multiple-gestation and the participants did not live adjacent to, or have an occupation related to, a smelter or metal recycling plant (radius < 1 km).

Of 41 lactating women, 39 mothers aged 20–40 years old were recruited. Two dyads were excluded for the non-qualified of breast milk or temporary donation. The mothers filled out a background information questionnaire in their demographic, age, occupation, education, life habits (smoking, exercising, and cosmetic or perfume usage), residential environment, outdoor activities, type of drinking water, and eating frequency and quantity. For dietary intake, oil, viscera, milk, fish, and seafood intake were all recorded separately categories based on Taiwan National Food Consumption Database. For each dyads, the monthly body weights and estimated daily milk volume of infants were obtained from the mothers by the telephone visits. Using the information from mother-infant-dyads, exposures for vulnerable recipients of donor milk would be further estimated.

Samples Collection and Metal Measurement

One additional tube of a participant’s blood was taken during the required TSHMB blood tests when donors joined the donation. One milliliter (mL) of venous blood from each participant was stored at − 20 °C for metals analysis.

Participants’ breast milk was collected every 1–2 weeks from their first donation and maximally up to 9 months. The staff of the TSHMB helped to obtain samples for further analysis during the handling unpasteurized donor milk. Fifteen mL of breast milk were collected in glass bottles and stored at − 20 °C before metals analysis. The body weight and daily breast milk consumption of infants were also reported from their mother along with each breast milk sampling for calculating the health risk for those infants from breast milk consumption. As well, the changes in food intake of lactating women and breast milk feeding of infants were also recorded at each sampling time.

During the analysis, the breast milk was thawed at 4 °C, heated to 40 °C, and shaken for homogeneity. Then, 1.5 mL 67% HNO3 was added to 500 μL of breast milk, and 2.5 mL 67% HNO3 was added to 250 μL of blood, which were adopted from the analysis method for metals in beverages and dairy food, reported from Taiwan FDA (MOHWH0023.00) (TFDA 2020).

The Pb levels in the milk and blood samples were measured by inductively coupled plasma mass spectrometry (ICP–MS, Nexion 2000, PerkinElmer) (MOHWH0023.00) (TFDA 2020).

The Hg concentrations were analyzed by an automatic mercury analyzer (MA-2000), and the method was modified from Hg in solids and solutions by thermal decomposition, amalgamation, and atomic absorption spectrophotometry (USEPA 7473), respectively. The breast milk was de-freeze at 40 °C water bath for reducing the delamination of fat from liquid. Then 100 µL was obtained for Hg analysis by automatic mercury analyzer.

The MeHg levels were measured by liquid chromatograph/inductively coupled plasma mass spectrometry (LC/ICP–MS, Nexion 2000, PerkinElmer), which were adopted from the analysis method for MeHg in foodstuff, reported from Taiwan FDA (MOHWH0018.00). The accuracy of the analysis and instrumentation were validated using certified reference materials from the PerkinElmer Pure Plus NexION Setup solution (Lot#32-13GSX1).

The recovery efficiency tests for Pb, Hg, and MeHg were conducted using the same sample analysis procedure but with the addition of a standard solution. In the current study, recovery rates were respectively 103%, 109%, and 89% for Pb, Hg, and MeHg in breast milk and 108% and 93% for Pb and Hg in blood. The method detection limit (MDL) was shown using a concentration slightly lower than the lowest concentration of the calibration curve. Measurements at this concentration were repeated seven times to estimate the standard deviation, and the MDL was set at a 3-time standard deviation. The MDLs for Pb, Hg, and MeHg in breast milk were 0.047, 0.132, and 0.082 μg/L, respectively, and for Pb, Hg in the blood it was 0.230 and 0.132 μg/L.

Exposure Measurement and Risk Characteristics

Due to the breast milk will be donated to TSHMB for vulnerable recipients, the exposure and risk assessment was processed based on the donor mothers and their infant dyads for assessing their consumption safety of breast milk. Estimated breast milk consumption and body weight (BW) of infants in each dyads were provided by mother during monthly interviews. This data was provided for calculating the estimated daily intake (EDI) of dietary exposure to toxic metals from breast milk consumption for infants in each dyads. The hazard quotient (HQ) of Pb was calculated based on the BMDL01 (0.5 µg/kg·BW/day), set by EFSA in 2010. The HQ for Hg was calculated based on the PTWI (4 µg/kg·BW/week), set by the WHO in 2011. The EDI, HQ and hazard index (HI) risks are expressed by the following equations:

$${\text{EDI}}\left( {{{{\mu g}} \mathord{\left/ {\vphantom {{{\mu g}} {{\text{kg}}}}} \right. \kern-\nulldelimiterspace} {{\text{kg}}}} \cdot { }{{{\text{bw}}} \mathord{\left/ {\vphantom {{{\text{bw}}} {{\text{day}}}}} \right. \kern-\nulldelimiterspace} {{\text{day}}}}} \right) = \frac{{{\text{Metal concentration}}\left( {{{{\mu g}} \mathord{\left/ {\vphantom {{{\mu g}} {\text{L}}}} \right. \kern-\nulldelimiterspace} {\text{L}}}} \right) \times {\text{Daily milk consumption}}\left( {\text{L}} \right)}}{{{\text{Body weight }}\left( {{\text{kg}}} \right)}}$$
(1)
$$\mathrm{HQ}=\frac{\mathrm{EDI}(\mathrm{\mu g}/\mathrm{kg bw}/\mathrm{day})}{\mathrm{TDI}(\mathrm{\mu g}/\mathrm{kg bw}/\mathrm{day})}$$
(2)
$$\mathrm{HI}=\sum_{i}{\mathrm{HQ}}_{i}$$
(3)

In addition, the HI can be calculated as the sum of the individual HQ of metals based on their reproductive effects. An HQ or HI < 1 indicated that the daily exposure dose may not cause adverse health effects and vice versa.

The Monte Carlo method is a probabilistic distribution function that characterizes the parameters used in estimating daily doses and risks. Sensitivity analysis was used to illustrate and rank variation in input variables based on their relative contributions to model output variability and uncertainty (US EPA 2001). The correlation coefficient (r) measures the strength and direction of the linear association between the values of two quantitative variables. If the model output variable (e.g., HI) and input variable are highly correlated, it means that the output is sensitive to that input variable. By squaring the coefficient, the result can be expressed as a percentage contribution to the variance of the output. The Monte Carlo simulation used @Risk 7.5.1 (Palisade Corp., Ithaca, NY, USA).

Results

The average age of the 39 lactating women in the study was 33.1 years, and no smokers or nor drinkers (Table 1). Twenty-one lived near the potential sources of air pollution, including traffic, industrial pollution, a night market, or a temple. Thirteen (33%) subjects reported using incense, and 18 (46%) used cosmetics frequently.

Table 1 Demographic data of the nursing women
Full size table

Figure S1 shows the numbers of samples in each postpartum month. A total of 228 breast milk samples were obtained from the donors, and the distribution of metal concentration is shown in Table 2. The average Pb level of breast milk was 6.49 µg/L (standard deviation: 5.23 µg/L), and the Hg level was 0.76 µg/L (0.98 µg/L).

Table 2 Pb and Hg levels in breast milk and maternal blood
Full size table

A mother who lived within 1 km of traffic, industrial pollution, night markets or temple incense had significantly higher Pb levels than the other mothers (mean Pb level 10.30 µg/L vs. 5.49 µg/L; p = 0.025) (Table 3). A significantly higher average of Pb level was also found in mothers exposed to house dust without wearing a mask, and who used cosmetics and lipstick (9.43, 9.23, and 10.90 µg/L, respectively) than those who did not (5.06, 4.82, and 4.85 µg/L).

Table 3 The difference in average concentrations of Pb and Hg in breast milk among different daily life habits
Full size table

A significant difference in the Pb levels of breast milk was found among lactating mothers who did or did not consume viscera (10.20 vs. 4.54 µg/L, respectively, p < 0.001). A significantly higher Pb level was found in those who ate the eggs (9.08 vs. 5.46 µg/L), shellfish (7.39 vs. 6.59 µg/L), and canned food (8.01 vs. 5.30 µg/L) than those who did not (Table 4). Similarly, for Hg levels, a significant difference was found among those who ate the viscera (0.84 vs. 0.57 µg/L, p = 0.043), shellfish (0.77 vs. 0.59 µg/L), large sea fish (1.11 vs. 0.62 µg/L), and cephalopods (0.79 vs. 0.58 µg/L) than those who did not consume (p < 0.05 for all).

Table 4 The difference of average concentration of Pb and Hg in breast milk (μg/L) among different dietary intake of the study population
Full size table

A significant difference in Pb levels was also observed among lactating women who drank only tap water (18.3 µg/L), tap water plus reverse osmosis water or bottled water (6.95 µg/L), or water without any tap water (5.30 µg/L).

The average infant EDIs of Pb were 0.628 and 0.069 ng/kg·body weight (BW)/day of Hg (Table 5). The exact values for 50% HQ for Pb and Hg exposure were 0.989 and 0.097, respectively, and the total HI for Pb and Hg was more than 1. Meanwhile, 95% of the upper limit of HQ was 3.782 for Pb and 0.326 for Hg exposure. According to the monthly measurements of Pb and Hg levels in breast milk, significant differences in EDI and HQ for Pb and Hg exposure were found each month (p < 0.05, separately) (Table S1).

Table 5 Estimated daily intake (EDI) and hazard quotient (HQ) of Pb and Hg of infants consuming the breast milk
Full size table

The decreasing trend of EDI and HQ of Hg and Pb exposure was found by their delivery month, and such finding may be due to the increase in infant BW. However, the HQs of Pb (Fig. 1a) in M1–M9 were higher than 1, and that of Hg (Fig. 1b) in M1–M4 was higher than 0.5. If the total HQs of Pb and Hg were summarized at 4 months after delivery, the HIs before month 4 were all higher than 1, and the most predominant risk contributor was Pb (Figs. S2 and S3).

Fig. 1
figure 1

Hazard quotient (HQ) of Pb (panel a) and Hg (panel b) of infants consuming the breast milk in each month

Full size image

The results of sensitivity analysis showed that the most influential variables in HI were the concentration of Pb in breast milk (52.8%) and the amount ingested by infants (34.2%) (Fig. S2). In addition, the literature cites environmental pollution sources, mask wearing while performing household cleaning, amalgam fillings, cosmetic usage, drinking water, and consumption of viscera and canned food as important factors.

Discussion

Pb and Hg Concentrations in Breast Milk

When comparing the Pb and Hg levels with other studies (Table 6), the results of the current study were similar to those found in Korea (8.79 µg/L vs. 0.94 µg/L) (Park et al. 2018). The Pb level was higher than those found in Austria (1.63 µg/L), Greece (0.48 µg/L), and Norway (0.2 µg/L) (Gundacker et al. 2002; Leotsinidis et al. 2005; Vollset et al. 2019), but lower than those in Spain (15.56 µg/L) (García-Esquinas et al. 2011), Lebanon (18.17 µg/L), and Iran (median: 41.9 µg/L) (Bassil et al. 2018; Samiee et al. 2019). The Hg level in the current study was comparable to those of other research (García-Esquinas et al. 2011; Iwai-Shimada et al. 2015; Al-Saleh et al. 2016; Vollset et al. 2019), but considerably lower than that in Austria (Gundacker et al. 2002). The MeHg level was not detected in all breastmilk samples.

Table 6 Reviews of metal concentration in breast milk
Full size table

Habit Factors Influence Metal Concentration in Breast Milk

In this study, we identified several life habit factors, including living place, cosmetic usage, and lipstick usage, which may be related to the increased levels of the studied metals (Table 3). Given that particulate matter inside a house may contain Pb from paint or furniture covering, lactating mothers may inhale them while doing housework if they do not wear a mask (Khanjani et al. 2018). Meanwhile, they still need to go out to shop or work, but such variable was not considered in this study. Feizi et al. (2019) indicated that high Pb concentrations were found in lipstick and eye pencil; thus, childhood Pb poisoning from breast milk poses a concern (Hon et al. 2017). Moreover, a positive association between dental amalgam fillings and the risk of Hg exposure has been proven in other research (Norouzi et al. 2012; Grzunov LetiniĿ et al. 2016; Vollset et al. 2019), but clinically strong evidence is still limited (FDA 2017). Although the Hg level was high in lactating women who had amalgam fillings, it was not statistically significant.

Metal Concentration in Breast Milk Under Different Dietary Intake Patterns

We observed several dietary sources associated with metal concentration. These results are similar to those from previous studies regarding the positive relationship between Hg levels in breast milk and the consumption frequency of fish, shrimp, crab and other shellfish (Gaxiola-Robles et al. 2014; Grzunov LetiniĿ et al. 2016; Vollset et al. 2019). Moreover, other studies suggested that different canned food, especially fruit, may contain a contaminant (Al Zabadi et al. 2018; Fathabad et al. 2018).

A significant difference in Pb levels was also detected among lactating women who drank only tap water (18.3 µg/L), tap water plus reverse osmosis water or bottled water (6.95 µg/L), or water without any tap water (5.30 µg/L). The result was consistent with that of Cardoso et al. (2014), who indicated that drinking water is an important pathway for metal exposure through breast milk based on observations of similar profiles.

Estimating Exposure of Infants to Metals from Breast Milk

In Korea, the Hg and Pb levels in 207 breast milk samples were monitored. The results showed that 45% exceeded the normal range of Pb suggested by WHO, and that 71% of 15-day-old infants and 56% of 30-day-old infants were at potential risk from Pb exposure based on the Monte Carlo simulation (Park et al. 2018). In Brazil, MeHg accounted for 11.8% of total Hg, and the mean weekly intake for MeHg was 0.16 ± 0.22 μg/kg·BW (0.679 μg/kg BW/week in the present study, Table 5), which represented 10% of the provisional tolerable weekly intake (42.4% in the present study). Several studies indicated no health concern for the breastfed infants (Rebelo et al. 2017), but not for immature and vulnerable infants. In Iran, the highest levels of Pb were observed after 2 months of delivery, and up to 94% of the breast milk samples exceeded the suggested WHO limit of < 5 μg/L for Pb contamination (Samiee et al. 2019). In our study, 42.9% of breast milk samples were over 5 μg/L (Table S2). Samiee et al. (2019) suggested a 61% unacceptable non-cancerous health risk levels or HQ for Pb of the breast milk samples. These studies, including ours, all point to a potential risk of toxic metals, especially Pb, for infants via the consumption of breast milk (Park et al. 2018; Samiee et al. 2019).

Sensitivity Analysis of the EDI

Previous studies and our data all proved that infants are at risk of Pb exposure within 1 or 2 months postpartum (Park et al. 2018; Samiee et al. 2019). The breastmilk, which is donated by lactating mothers, is intended for premature delivery and infants, vulnerable to well-known neurological toxicity of Pb, at the 1st month postpartum. Utilizing mother-infant-dyads recruit from the milk bank to estimate the health risk of the infants of dyads and the vulnerable recipients of donor milk, we can confirm that the quality of breastmilk is extremely important.

This study was several research that recently conducted a follow-up of breast milk samples and questionnaires. Thus, it provides sufficient information to investigate the trend of risk during breastfeeding. However, this study also presented several limitations. First, inevitable recall and response bias existed in the questionnaire interviews. Second, except for dietary intake (Table 4), living near any polluted areas within 1 km, clean habit, and cosmetic usage are all influencing factors in metal concentrations in the breast milk. However, these mothers may need to go out to shop without any mask or to work in a polluted area while their residential place are in rural areas. These uncontrolled environmental factors will be the confounders in this study. Third, other heavy metals besides those tested can cause joint toxic reactions. Thus, the synergistic risk can be evaluated if more toxic metals were monitored simultaneously.

Future Direction to Promote the Quality of Human Milk Bank and Maternal Children Health

Currently, the operation of human-donor milk banks in Europe (Kontopodi et al. 2021) and in North America (Brownell et al. 2014) mainly focuses on the safety of donor milk in accordance with the donors’ lifestyle criteria (smoking, alcohol, drug abuse, extreme diet, etc.), maternal infection issue (human immunodeficiency virus, hepatitis viruses, cytomegalovirus, sexually transmitted disease, etc.), and the pathological bacteriology of the collected or pasteurized milk. However, the monitoring of heavy metal contamination in human milk banking has been largely overlooked.

To our best knowledge after literature review, this study is the first cohort observation research conducted on a human milk bank for monitoring of the reported heavy metals. Monitoring heavy metal in donors’ milk may improve not only the health of donors’ offspring but also that of the recipients. Our study result helped donors of the milk bank but also the non-donor lactating women to adjust their lifestyle to avoid further intake of heavy metals from polluted sources. The offspring and recipients who depend on breast milk exclusively could be also protected from intake of heavy metals.

In summary, breast feeding exhibits numerous beneficial effects for mothers and children, including reduction of ovarian and breast cancer risk (Rea 2004; Leung and Sauve 2005), growth promotion, and modulation of immune system (Makrides et al. 1996; Laiho et al. 2003; Wijga et al. 2006; Lepage and Perre 2012; Khaldi et al. 2014; Rogier et al. 2014). Beyond improving physical health, breastfeeding can help in building a close relationship between mothers and their offspring (Guxens et al. 2011; Sabel et al. 2012). The WHO recommends that women breastfeed children for the first 6 months and continue breastfeeding while giving appropriate complementary foods until 2 years of age (WHO 2009; Schwarzenberg and Georgieff 2018). Hence, healthy diets and lifestyle for breastfeeding mothers should be advised. Prioritizing adequate nutrition and healthy dietary patterns during this time is a key factor to ensure a foundation for a child’s optimal neurodevelopment (Schwarzenberg and Georgieff 2018). Based on the results of daily life habits, dietary pattern and risk assessment, the future public policy in promoting breastfeeding should be considered to instruct healthy diets and lifestyle from exposure to heavy metals since the perinatal stage.

Strength and Limitation

Enrollment of participants from a milk bank and with relatively large milk sample size is a strength. Participants from the milk bank were all longitudinal and exclusively breastfed. A total of 26 out 39 (67%) participants continued to provide milk samples and breastfeed during the study period. Not including all participants from their partum immediately was a weakness. However, the enrollment from the partum immediately will result in high dropped out rate since 24.2–29.3% of the mothers were exclusively at 2–6 months postpartum (Taiwan-Health-Promotion-Administration 2012; Chang et al. 2019). Importantly, the study unit, TSHMB, exclusively enrolled the donors within 3 months postpartum and well presented the interesting study cohort.

Conclusions

The results indicated an unacceptable non-cancerous health risk (95% HI = 1.37 > 1) for Pb and Hg, for infants through the consumption of breast milk. Meanwhile, environmental conditions, including indoor air quality, residential districts, cleaning products, and maternal dietary, e.g., drinking water, viscera (organ meat), eggs, seafood and canned food and personal habits, e.g., cosmetics usage, can affect Pb and Hg concentrations under different conditions. In addition, Pb and Hg pose a potential hazard, especially for Pb exposure via breastmilk consumption. Based on the results, heavy metal in human milk, especially for Pb, should be monthly monitored not only in donor milk human milk banks but also in postpartum women who experience traditional confinement and bear children with exclusive breastfeeding. Compared to blood sampling, breast milk is much feasible to be expressed and preserved. Second, breastfeeding mothers with high levels of toxic metals in their breast milk should be given feedback to change their life or dietary style and inhibit from contacting contaminated sources. Third, the synergistic risk of other toxics should be evaluated further to ensure high-quality milk for donors’ offspring and vulnerable recipients who exclusively depend on it. Fourth, a broader public health attention should be paid if the mothers' milk obtained from the study area, irrespective to donation, have higher level of toxic metals.