Presence of melamine in human milk and the evaluation of the effects on mother

Abstract

Melamine contamination is a recent public health problem emerging globally. Present study aimed to detect the rate of melamine presence in human milk in a cohort study and to evaluate any possible differences in maternal–infant pair characteristics such as breastfeeding status, crying and sleep problems of infants, maternal postpartum depression, maternal–infant bonding, infant and maternal anthropometry, and maternal complete blood count caused by the melamine exposure. Mothers of infants born in Şanlıurfa were invited to participate in ‘Urfa Child Cohort Survey’. Overall, two breast milk samples were taken between 5 days and 15 days postpartum and between 4 weeks and 10 weeks. Randox Food’s InfiniPlex array was used to analyse the presence of melamine. Melamine was detected in 32.2% and 24.4% of the first and the second milk samples; 16.7% of mothers had two positive samples. z Scores for birth weight and z scores for height for age were detected to be significantly lower in cases with two positive samples compared to cases with negative samples. Mean maternal white blood cells counts were found to be lower in cases with repeated melamine contact. Melamine might have a detrimental effect on birth weight, infant height and maternal blood count. Further studies should be done to detect environmental contamination in different regions and countries.

Keywords

Human milk growth maternal psychopathology melamine white blood cells

Introduction

Melamine appears everywhere, namely, nitrogenous fertilizer mixtures, paints, paper coatings, plastics, textile finishes and wrinkle-free textile.¹ Cyromazine, an insecticide widely used in the cultivation of vegetables, is converted to melamine in the body of animal and human.^1
–3 Due to a high nitrogen content, melamine has also been reported to be illegally added into several foods including dairy products, candies, cakes, biscuits, chocolates, wheat gluten, pet foods, nutritional supplements and eggs in order to fortify its protein content.^2,4
–6 This is defined as melamine adulteration (MA) and MA in baby formula resulted in an outbreak of renal calculi, renal failure and death in infants in China in 2008.⁷ The effects of MA have also been evaluated in experimental animals including mice, rats, fish, chicken, cats, dogs and pig models.⁸ In rodent models, maternal exposures led to the transfer of melamine to serum, amniotic fluid and foetal tissues including brain, heart, kidney, liver, lungs as well as into the milk.⁹ Also, when cows and goats were fed the melamine-added feeds, melamine was detected in their milk.^6,10,11 Therefore, besides the intake of contaminated food, infants may come into contact with melamine antenatally through a placental barrier or postnatally through breastfeeding. To date, limited data are available on assessing the transfer of melamine into human milk.^12
–14 Although the exposure level is very little compared to self-consumption of contaminated food, melamine exposure to pregnant and lactating women can be a potential risk to the offspring during critical periods of development.¹⁴ Since foetuses and infants are more vulnerable to the effects of environmental pollution, any possible route of melamine transfer should be prevented.^1,3,6

There are several studies about MA and renal disorders, however, its extra-renal effects are not well researched.^3,7,8,15 To best of our knowledge, there is no published report available on the effects of MA on haematopoietic parameters in human. Being at a global risk for MA, the research into mother–infant pair’s common problems with unknown pathogenesis such as postpartum depression, breastfeeding problems, and sleep and crying characteristics can also contribute in melamine risk assessment. So far, data regarding the effects of melamine on infant’s anthropometric evaluation are very scarce.¹⁵

Therefore, this study was aimed to assess the presence of melamine in breast milk and evaluate any possible differences in maternal–infant pair characteristics such as breastfeeding status, crying and sleep problems of infants, maternal postpartum depression, maternal–infant bonding, infant and maternal anthropometry, and maternal complete blood count in cases with melamine exposure. We hypothesized that the maternal exposure of MA could be detected easily with breast milk and would influence the well-being of maternal–infant pairs in the early postpartum period.

Materials and methods

Study design and participants

Subjects of the study were recruited from the Özel Şan Med Hospital in Şanlıurfa, Turkey. Mothers giving birth between May 2017 and April 2018 were invited to participate in ‘Urfa Child Cohort Survey’. Of those invited in the survey, 90 mother–infant pairs with complete follow-up schedule and two breast milk samples at different visits were selected for this study.

Visit 1 (V1): All voluntary mother–infant pairs were evaluated between 5 days and 15 days post-delivery, and they answered the questionnaire concerning mother–infant health. Items on maternal sociodemographic and health properties included maternal age, educational level, working status and health status (hypertension, diabetes, anaemia, drug use, smoking exposure status, urolithiasis, weight gain during pregnancy and weight at birth). The criterion for smoking exposure status was (1) maternal smoking, (2) environmental tobacco smoke exposure without maternal smoking and (3) no known smoke exposure for the last 3 months. In the questionnaire, items on infant characteristics included the presence of assisted reproduction, gestational length, type of delivery, birth order, gender, birth weight, head circumference at birth, any health problem including hyperbilirubinemia and urolithiasis, presence of any breastfeeding-related problem and feeding type.

Visit 2 (V2): When infants were 4–10 weeks of age, mother–infant pairs were re-evaluated by a second questionnaire, which covered infant’s crying properties, feeding and sleep-related problems, and maternal smoking status for the last months.

Mothers and infants were weighed, and height and head circumference of infants were measured at V1 and V2.

Outcomes of the study

The outcomes examined in this study were breastfeeding status, colicky crying and sleep problems of infants, postpartum depression, maternal–infant bonding, infant and maternal anthropometry and maternal complete blood count.

Breastfeeding status were evaluated according to ‘early initiation of breastfeeding within 1 h of delivery’, ‘delayed onset of lactogenesis which was defined as maternal perception of onset of noticeable fullness after 72 h postpartum’, ‘exclusive breastfeeding that is the infant only receives breast milk with vitamin–mineral supplement, without any additional food or drink, not even water’ and ‘breastfeeding problems were accepted as presence of anyone of cracked nipples and insufficient milk during follow-up visits’.

Length of crying episodes in a day and weekly episodes were asked. Inconsolable infant crying lasting for >2 h per day and >2 days/week was defined as ‘colicky crying’ in the protocol.

Sleep-related problems were defined as ‘absence or presence’ according to maternal perception.

To assess mother–infant bonding, Postpartum Bonding Questionnaire (PBQ) with self-rating 25-item questionnaire scored on a 6-point Likert scale was used.^16,17 A high score indicates worse bonding.

Edinburgh Postnatal Depression Scale (EPDS) composed of self-rating 10-item questionnaire with a 4-point Likert scale was used to identify postpartum depression.¹⁸

Maternal body mass index was calculated using weight and height at each visit. For each visit, z scores of head circumference for age, weight for age, length for age and body mass index for age (HCZ, WAZ, LAZ and BAZ, respectively) of infants were calculated with World Health Organization Multicentre Growth Reference Study.¹⁹

Maternal data on complete blood count (haemoglobin value, haematocrit value, mean corpuscular volume, white blood cells and thrombocyte count) which were evaluated at 1 month postpartum were taken from hospital file.

Samples collection and analysis

At both visits (V1 and V2), breast milk samples were obtained from the participants via manual pumping. Samples were taken into polyethylene tubes, and immediately after collection, they were kept frozen (−20°C) until analysed. InfiniPlex kit for Milk array (EV4076, Lot number: 9034, Randox Food Diagnostics, Crumlin, UK) was used to analyse the presence of melamine in the milk according to the manufacturer’s instructions.²⁰ The detection limits for melamine and cyromazine contamination are 0.200 and 0.027 mg/kg milk.

Statistical analyses

Data were analysed using SPSS-Windows 23.0 (SPSS Inc, Chicago, Illinois, USA).

According to the status of melamine detection in visits, three ‘melamine-exposed groups’ were designed: not detected at all (M0), in one visit (M1) and in both visits (M2).

Shapiro–Wilk test was used to evaluate the normality of continuous variables in melamine exposure groups. Data with skewed distribution were shown as median and interquartile values and Kruskal–Wallis H test was applied to compare groups.

Normally distributed data were given as mean, standard deviation (SD) and 95% confidence interval (CI) and one-way analysis of variance was used to compare groups. Anthropometric measurements were compared with ‘General Linear Models, Univariate Analysis of Variance’ after controlling for confounding factors at relevant visit data. ‘Birth WAZ and HCZ data’ was compared in melamine-exposed groups with adjustment of gestational length (weeks) and maternal smoking exposure. ‘Infant’s anthropometry on admission (V1)’ with adjustment of adjusted for gestational length (weeks), maternal smoking exposure, feeding type of infant (exclusive breastfed vs. mixed feeding), infant’s age and gender. ‘Infant’s anthropometry on V2’ adjusted for gestational length (weeks), feeding type (exclusive breastfed vs. mixed feeding), smoking exposure (no known exposure, environmental exposure vs. maternal smoking), infant’s age (days) at V2 and gender.

The percentages in melamine-exposed groups were calculated and the χ² test was used to compare proportions.

Only parameters having the p value less than 0.1 were included in further analysis. Multiple logistic regression with backward stepwise condition was performed to detect the effect of melamine exposure (M1 vs. M0 and M2 vs. M0), gestational length (weeks), birth type (caesarean delivery vs. normal delivery), early initiation of breastfeeding (<1 h vs. other) and birth order (1 vs. ≥1 child) on delayed onset of lactogenesis.

Statistical significance level was selected as p < 0.05.

Results

General characteristics of mother–infant pairs

In the study, 36.7% of the participants were primiparous and mean (±SD) maternal age calculated was 27.9 (±5.3) years. Mean (±SD) gestational length recorded was 38.3 (±1.5) weeks (Table 1). Of all, 18.9% of mothers stated they had some health problems (number) during gestational period, [gestational diabetes (3), hypothyroidism (2), hypertension (1), iron deficiency anaemia (5), oedema on their legs (2), urinary tract infection (3) and upper respiratory infections (2)].

Table 1.

Background characteristics of mother–infant pairs, n = 90.

Background characteristics
Maternal age, years	Mean (95% CI)	27.9 (26.8, 29.1)
Maternal education ≥12 years	n (%)	31 (34.4)
Working mothers	n (%)	15 (16.7)
Assisted reproduction for this pregnancy	n (%)	7 (7.8)
Any maternal health problem during the gestational period	n (%)	17 (18.9)
Maternal weight gain during pregnancy, kg	Mean (95% CI)	10.4 (8.8, 12.0)
Maternal body mass index at birth	Mean (95% CI)	30.0 (29.0, 31.0)
Gestational length, week	Mean (95% CI)	38.3 (38.0, 38.6)
Gestational length <38 weeks	n (%)	20 (22.2)
Birth order, first child	n (%)	33 (36.7)
Birth type, caesarean section	n (%)	66 (73.3)
Gender, male	n (%)	39 (43.3)
Infant’s HCZ-0, at birth	Mean (95% CI)	−0.07 (−0.23, 0.09)
Infant’s WAZ-0, at birth	Mean (95% CI)	−0.36 (−0.55, −0.17)
Low birth weight < 2.5 kg	n (%)	6 (6.7)
Any infant’s health problem seen following delivery	n (%)	6 (6.7)
Early initiation of breastfeeding ≤ 1 h	n (%)	48 (53.3)
Delayed onset of lactogenesis	n (%)	18 (20.0)

CI: confidence interval; HCZ-0: z scores of head circumference for age taken at birth; WAZ-0: z scores of weight for age taken at birth.

Mean (±SD) birth weight of infants calculated was 3132 (±412) g. Six infants had a problem at birth: four with respiratory distress syndrome, two with neonatal transient tachypnoea. Meconium aspiration was also present in one of two tachypnea cases. Physical examination of the infants at V1 showed normal health status, except four cases of hyperbilirubinemia (with no need of phototherapy), two cases of mild omphalitis. At V2, one case of developmental dysplasia of hip, four cases of upper respiratory infections and one case of hyperphenylalaninemia were detected.

Early initiation of breastfeeding was performed in 53.3% of mother–infant pairs. One in five reported delayed lactogenesis after delivery (Table 1). One-third had breastfeeding problems at V1. The proportions of exclusively breastfed infants were 62.2% at V1 and 45.6% at V2 (Table 2).

Table 2.

Characteristics of infants (age, anthropometric parameters, health problems, crying and sleep problems), mothers (smoking, scores of EPDS and PBQ, BMI and blood count) and breastfeeding (problems and melamine content) at visit 1 and visit 2, n = 90.

		Visit 1	Visit 2
Infant age, days	Mean (95% CI)	7.1 (6.6, 7.5)	46.8 (44.3, 49.2)
Infant health problems	n (%)	6 (6.7)	6 (6.7)
Exclusive breastfeeding	n (%)	56 (62.2)	41 (45.6)
Breastfeeding problems, including cracked nipples and insufficient milk	n (%)	33 (36.7)	54 (60.0)
Maternal smoking exposure during the last 2 months
No known smoke exposure	n (%)	49 (54.4)	46 (51.1)
Environmental tobacco smoke exposure	n (%)	31 (34.4)	27 (30.0)
Maternal smoking	n (%)	10 (11.1)	17 (18.9)
Maternal BMI	Mean (95% CI)	28.1 (27.2, 29.0)	27.1 (26.3, 28.0)
Daily maternal body mass index change from birth to visit date	Mean (95% CI)	−0.29 (−0.36, −0.23)	−0.07 (−0.08, −0.06)
Infant’s WAZ	Mean (95% CI)	−0.40 (−0.62, −0.18)	−0.13 (−0.34, 0.09)
Infant’s HCZ	Mean (95% CI)	−0.49 (−0.67, −0.31)	−0.48 (−0.66, −0.31)
Infant’s LAZ	Mean (95% CI)	−0.21 (−0.43, 0.01)	−0.28 (−0.53, −0.03)
Infant’s BAZ	Mean (95% CI)	−0.44 (−0.69, −0.18)	0.04 (−0.18, 0.26)
Colicky crying	n (%)		27 (30.0)
Sleep problems according to maternal perception	n (%)		10 (11.1)
EPDS, score	Median (25–75 p)		7.0 (3.0–12.0)
EPDS, score ≥ 13	n (%)		20 (22.2)
PBQ score, total (n = 65)	Median (25–75 p)		7.0 (3.0–14.0)
Breast milk melamine or cyromazine levels ≥ DL^a	n (%)	29 (32.2)	22 (24.4)
Maternal blood count at visit 2
Haemoglobin value, g/dL	Mean (95% CI)		12.4 (12.2, 12.6)
Haematocrit value, %	Mean (95% CI)		38.3 (37.7, 38.9)
Mean corpuscular volume, fl	Mean (95% CI)		85.5 (83.9, 87.1)
White blood cells, ×10³/µL	Mean (95% CI)		7.2 (6.8, 7.5)
Thrombocytes, ×10³/µL	Mean (95% CI)		284 (269, 300)

EPDS: Edinburgh Postnatal Depression Scale; PBQ: Postpartum Bonding Questionnaire; BMI: body mass index; visit 1: admission visit; visit 2; follow-up visit; CI: confidence interval; WAZ: z scores of weight for age; HCZ: z scores of head circumference for age; LAZ: z scores of length for age; BAZ: z scores of BMI for age; EPDS: Edinburgh Postnatal Depression Scale; PBQ: Postpartum Bonding Questionnaire.

^a Detection limits were 0.200 for melamine and 0.027 mg/kg-milk for cyromazine.

Melamine detection and mother–infant characteristics

Melamine was detected in 32.2% and 24.4% of the first and the second samples, respectively (Table 2). Of the mothers enrolled in the study, 60% had no melamine present in either of the samples while 16.7% of mothers had two positive samples.

Maternal age, level of education, working status, health problems, presence of assisted reproduction for pregnancy, birth order, gestational length, mode of delivery and infant’s gender did not differ among melamine-exposed groups (Table 3). Maternal anthropometric evaluations were also similar for all groups (Tables 4 and 5).

Table 3.

Background characteristics of mother–infant pairs according to melamine-detection groups.

Background characteristics		M0	M1	M2	p
Overall	n	54	21	15
Maternal age, years	Mean (95% CI)	28.3 (26.8, 29.9)	26.9 (25.0, 28.8)	28.0 (25.1, 30.9)	0.585
Maternal education ≥12 years	n (%)	16 (29.6)	8 (38.1)	7 (46.7)	0.434
Working mothers	n (%)	6 (11.1)	4 (19.0)	5 (33.3)	0.117
Assisted reproduction	n (%)	3 (5.6)	2 (9.5)	2 (13.3)	0.575
Any maternal health problem	n (%)	10 (18.5)	5 (23.8)	2 (13.3)	0.727
Gestational length, week		38.4 (38.0, 38.8)	38.5 (37.9, 39.1)	37.5 (36.8, 38.3)	0.085
Gestational age <38 weeks	n (%)	11 (20.4)	3 (14.3)	6 (40.0)	0.164
Birth order, first child	n (%)	18 (33.3)	8 (38.1)	7 (46.7)	0.630
Birth type, caesarean section	n (%)	44 (81.5)	12 (57.1)	10 (66.7)	0.083
Gender, male	n (%)	22 (40.7)	10 (47.6)	7 (46.7)	0.830
Maternal weight gain during pregnancy period, kg	Mean (95% CI)	11.3 (8.8, 13.7)	9.8 (7.0, 12.6)	8.3 (5.6, 10.9)	0.379
Maternal BMI-b	Mean (95% CI)	30.6 (29.3, 31.9)	29.9 (27.9, 32.0)	28.1 (25.8, 30.5)	0.203
Low birth weight < 2.5kg	n (%)	1 (1.9)	1 (4.8)	4 (26.7)	0.003
Infant WAZ-b	Mean (95% CI)	−0.17 (−0.40, 0.07)^a	−0.45 (−0.80, −0.09)^ab	−0.93 (−1.51, −0.35)^b	0.013
Infant WAZ-b^d	Mean (95% CI)	−0.18 (−0.40, 0.03)^a	−0.50 (−0.85, −0.15)^ab	−0.79 (−1.21, −0.37)^b	0.032
Infant HCZ-b	Mean (95% CI)	0.00 (−0.23, 0.23)	−0.09 (−0.40, 0.23)	−0.31 (−0.61, −0.01)	0.380
Early initiation of breastfeeding, within first hour	n (%)	27 (50.0)	12 (57.1)	9 (60.0)	0.729
Delayed onset of lactogenesis	n (%)	8 (14.8)	4 (19.0)	6 (40.0)	0.097

CI: confidence interval; M0: absence of melamine in the samples; M1: one positive sample; M2: two positive samples; BMI-b: body mass index taken at delivery; HCZ-b: z scores of head circumference for age taken at birth; WAZ-b: z scores of weight for age taken at birth.

^a,b,cValues with different letter in the same row are different (p < 0.05).

^d Adjusted for gestational length, maternal smoking exposure.

Table 4.

Characteristics of infants (age, anthropometric parameters), mothers (smoking and BMI) and breastfeeding (problems) at visit 1 (admission visit) according to melamine-detection groups.

Visit 1		M0	M1	M2	p
Infant age at visit 1, days	Mean (95% CI)	7.4 (6.9, 8.0)	6.4 (5.5, 7.4)	6.8 (5.6, 8.0)	0.157
Exclusive breastfeeding	n (%)	33 (61.1)	14 (66.7)	9 (60.0)	0.889
Any breastfeeding problem	n (%)	18 (33.3)	6 (28.6)	9 (60.0)	0.113
Maternal smoking exposure, last 3 months					0.829
No known smoke exposure	n (%)	29 (53.7)	9 (42.9)	8 (53.3)
Environmental tobacco smoke exposure	n (%)	11 (20.4)	4 (19.0)	2 (13.3)
Maternal smoking	n (%)	14 (25.9)	8 (38.1)	5 (33.3)
Maternal BMI-1	Mean (95% CI)	28.4 (27.1, 29.7)	28.6 (26.9, 30.3)	26.4 (24.3, 28.5)	0.254
Maternal BMI change daily since delivery	Mean (95% CI)	−0.32 (−0.36, −0.27)	−0.25 (−0.48, −0.01)	−0.28 (−0.41, −0.16)	0.671
Infant HCZ-1	Mean (95% CI)	−0.43 (−0.69, −0.18)	−0.41 (−0.76, −0.07)	−0.77 (−1.11, −0.44)	0.367
Infant WAZ-1	Mean (95% CI)	−0.29 (−0.56, −0.01)	−0.39 (−0.83, 0.05)	−0.82 (−1.47, −0.16)	0.215
Infant WAZ-1^d	Mean (95% CI)	−0.26 (−0.51, −0.02)	−0.55 (−0.95, −0.16)	−0.67 (−1.14, −0.20)	0.224
Infant LAZ-1	Mean (95% CI)	0.01 (−0.22, 0.24)^a	−0.09 (−0.41, 0.22)^a	−1.16 (−2.08, −0.24)^b	<0.001
Infant LAZ-1^d	Mean (95% CI)	0.03 (−0.22, 0.28)^a	−0.23 (−0.62, 0.17)^a	−1.06 (−1.53, −0.58)^b	0.001
Infant BAZ-1	Mean (95% CI)	−0.44 (−0.75, −0.12)	−0.53 (−1.07, 0.01)	−0.30 (−1.15, 0.55)	0.855

M0: absence of melamine in the samples; M1: one positive sample; M2: two positive samples; CI: confidence interval; anthropometric measurement taken at the first visit (admission visit), BMI-1: body mass index; HCZ-1: z scores of head circumference for age; WAZ-1: z scores of weight for age; LAZ-1: z scores of length for age; BAZ-1: z scores of body mass index for age.

^a,b,cValues with different letter in the same row are different (p < 0.05).

^d Adjusted for gestational length, maternal smoking exposure, feeding type of infant, infant’s age and gender.

Table 5.

Characteristics of infants (age, anthropometric parameters, crying and sleep problems), mothers (smoking, scores of EPDS and PBQ, BMI and blood count) and breastfeeding (problems and melamine content) at visit 2 (V2; follow-up visit) by melamine-detection groups.

Visit 2		M0	M1	M2	p
Infant age at V2, days	Mean (95% CI)	47.7 (44.1, 51.4)	46.0 (41.4, 50.6)	44.4 (40.9, 47.9)	0.586
Maternal BMI-2	Mean (95% CI)	27.6 (26.4, 28.8)	27.1 (25.4, 28.7)	25.4 (23.3, 27.6)	0.197
BMI change from delivery to V2, daily	Mean (95% CI)	−0.07 (−0.08, −0.06)	−0.07 (−0.11, −0.03)	−0.06 (−0.08, −0.05)	0.871
Exclusive breastfed	n (%)	21 (38.9)	12 (57.1)	8 (53.3)	0.291
Any breastfeeding problem at V2	n (%)	35 (64.8)	9 (42.9)	10 (66.7)	0.185
Smoking exposure, last month					0.783
No known smoke exposure	n (%)	26 (48.1)	8 (38.1)	5 (33.3)
Environmental tobacco smoke exposure	n (%)	15 (27.8)	8 (38.1)	5 (33.3)
Maternal smoking	n (%)	13 (24.1)	5 (23.8)	5 (33.3)
Maternal blood count
Haemoglobin, g/dL	Mean (95% CI)	12.3 (12.0, 12.7)	12.2 (11.7, 12.7)	12.9 (12.4, 13.4)	0.177
Haematocrit, %	Mean (95% CI)	38.2 (37.4, 39.0)	37.8 (36.4, 39.2)	39.2 (37.9, 40.4)	0.360
Mean corpuscular volume, fl	Mean (95% CI)	84.5 (82.4, 86.6)	86.5 (82.3, 90.7)	87.8 (84.7, 90.8)	0.293
White blood cells, ×10³/µL	Mean (95% CI)	7.5 (7.0, 8.1)^a	6.9 (6.4, 7.5)^ab	6.3 (5.6, 6.9)^b	0.044
Thrombocytes, ×10³/µL	Mean (95% CI)	292 (270, 313)	268 (242, 294)	280 (238, 322)	0.446
Infant characteristics
Colicky crying	n (%)	19 (35.2)	10 (47.6)	8 (53.3)	0.354
Sleep problems according to maternal perception	n (%)	5 (9.3)	3 (14.3)	2 (13.3)	0.788
Infant WAZ-2	Mean (95% CI)	−0.05 (−0.31, 0.21)	−0.07 (−0.56, 0.43)	−0.48 (−1.13, 0.17)	0.347
Infant HCZ-2	Mean (95% CI)	−0.38 (−0.62, −0.15)	−0.50 (−0.90, −0.11)	−0.81 (−1.25, −0.37)	0.223
Infant LAZ-2	Mean (95% CI)	−0.02 (−0.34, 0.30)^a	−0.47 (−0.91, −0.03)^ab	−0.97 (−1.68, −0.26)^b	0.016
Infant LAZ-2^d	Mean (95% CI)	−0.01 (−0.31, 0.29)^a	−0.50 (−0.98, −0.03)^a	−0.96 (−1.53, −0.38)^b	0.012
Infant BAZ-2	Mean (95% CI)	−0.06 (−0.33, 0.22)	0.26 (−0.28, 0.80)	0.07 (−0.52, 0.65)	0.505
EPDS, score	Mean (95% CI)	5.5 (3.0, 12.0)	9.0 (3.0, 14.0)	9.0 (6.0, 12.0)	0.316
EPDS, score ≥13	n (%)	11 (20.4)	6 (28.6)	3 (20.0)	0.726
PBQ, total	Median (25–75 p)	5 (2–14) N = 34	10 (4–18) N = 16	9.5 (5.5–18.3) N = 15	0.582

EPDS: Edinburgh Postnatal Depression Scale; PBQ: Postpartum Bonding Questionnaire; BMI: body mass index; M0: absence of melamine in the samples; M1: one positive sample; M2: two positive samples; CI: confidence interval; anthropometric measurement taken at the second visit (V2; follow-up visit, BMI-2: body mass index; HCZ-2: z scores of head circumference for age; WAZ-2: z scores of weight for age; LAZ-2: z scores of length for age; BAZ-2: z scores of body mass index for age.

^a,b,cValues with different letter in the same row are different (p < 0.05).

^d Adjusted for infant’s age (days) at V2, gender, gestational length (weeks), feeding type (exclusive breastfed or mixed feeding) and smoking exposure (no, environmental exposure, maternal smoking) at V2.

When maternal complete blood count values were compared, leucocytes count of M2 cases was found to be significantly lower than other groups (p = 0.044, Table 5).

The detection rate of delayed onset of lactogenesis in M2 was slightly higher compared to the counterparts (p = 0.097, Table 3). However, multiple logistic regression with stepwise method revealed only birth order (1 vs. ≥1 child) affected delayed lactation (odds ratio = 0.15, %95 CI = 0.05–0.47) among factors including melamine-exposed groups, gestational length, birth type and early initiation of breastfeeding. There were no differences present in the frequency of being exclusively breastfed among groups at both visits.

WAZ scores of infants at birth (WAZ-b) were detected to be significantly lower in M2 compared to M0 (Table 3). Similarly, low WAZ-b was observed in M2 after controlling gestational length, presence of maternal health problems and maternal smoking exposure. However, WAZ-1 and WAZ-2 scores were found to be similar in melamine-detection groups (Tables 4 and 5). On the other hand, low HAZ-1 and HAZ-2 scores of infants were recorded in the M2 group compared to M0 (Tables 4 and 5). Adjusting gestational length, mode of delivery (vaginal vs. caesarean delivery), infant gender (male vs. female), age (days), maternal and neonatal health problems (presence vs. absence), maternal smoke exposure (environmental and maternal smoke vs. absence) and feeding type of infant (exclusive breastfed vs. mixed) at both visits had no impact on the results. Melamine contamination did not affect HCZ and BAZ scores of infants during the follow-up period (Tables 4 and 5).

Proportions of colicky crying, sleep problems according to maternal perception were similar in all groups. Melamine presence did not alter maternal EPDS and PBQ scores (Table 5).

Discussion

Melamine transfer to human milk

This study showed that 40.0% of breast milk from mothers in Şanlıurfa, Turkey, had melamine and/or cyromazine contamination at the level of ≥0.200 mg/kg and/or ≥0.027 mg/kg, respectively. Exposure to melamine might change depending on region and country.^13,14 One study with the same methodology reported melamine contamination in 3.6% (3/83) of breast milk samples in Eskişehir, Turkey.¹³ So far, the previous study in Ankara, Turkey showed that 20.8% of 77 milk samples were positive for melamine presence (range = 10.6–76.4 ng/L).¹² Compared to studies in Turkey, a recent study in the United States reported the sum of melamine and its three derivatives in breast milk at a lower concentration (0.176–10.0 ng/mL).¹⁴ However, melamine detection in breast milk has raised concerns about the environment surrounding of mother–infant pairs. Interestingly, melamine and its derivatives were detected in 341 samples of indoor dust collected from 12 countries; highest in the United States and lowest in India.²¹ Previously, 300 samples of milk and dairy products from Ankara, capital of Turkey, were analysed for the presence of melamine and detected in 44%, 8% and 2% of yogurt, milk powder and cheese samples, respectively.²²

Melamine is not metabolized or bio-transformed in the body. It is eliminated by the kidney and has been detected in urine.²³ The pharmacokinetics of melamine revealed that half-life and clearance time were depended on the species; it was as short as 90–120 h in animals like pigs, chickens and rats.^6,10,11 In our study, 16.7% of mothers had two positive samples showing continuous or repeated exposure of melamine. During the lactation period, maternal exposure to melamine, either through contaminated foods or use of dishes, can result in melamine transfer to the infants via milk.¹⁴

Health concerns and melamine

Previous studies revealed that prenatal and postnatal exposures to melamine caused altered birth weight and crown-to-rump length in rat models.^9,24 One study reported high percentages of delayed growth (under height and underweight) in melamine-related urinary stones group than the control group.¹⁵ In the present study, cases of melamine contact in two periods (M2 group) had lower z scores of birth weight than others and showed disruptions in height for age during the follow-up period. However, none of the mother–infant pairs had a history of urolithiasis. On the other hand, the previous study in the United States reported no differences among cases in the melamine compound’s according to birth weight.¹⁴ This might be due to low dose exposure of mother–infant pairs in the United States. There is no published study to evaluate growth in the exclusively breastfed infant. Further studies in a large sample size are necessary to detect the effect of melamine in infant growth.

Previous studies showed that prenatal melamine exposure resulted in neurophysiological changes and neurotoxicity with the underlying mechanisms including synaptic plasticity, oxidative stress and the cholinergic system.²⁵ Considering the possibility of neurobehavioural effect, postpartum maternal depression and maternal–infant bonding, infants’ sleep and crying characteristics were evaluated, however, no interaction was detected with melamine exposure. However, it does not mean that there are no interactions and it is possible to observe such interactions in high-level exposure models. This hypothesis can be analysed in high exposure areas. On the other hand, unexplained crying especially when urinating had been observed in infants affected by the melamine-contaminated infant formula in China.²⁶ Renal damage might have a role in these crying.

As expected, some environmental chemicals might affect breastfeeding success. Our previous study showed that ‘delayed onset of lactogenesis’ was detected 3.33 times higher in mothers who had the highest quarter level of zearalenone in their milks than counterparts.²⁷ In the present study, delayed lactogenesis was slightly higher in mothers who had repeated melamine exposure, which is not found to be statistically significant. Further studies with large sample size are necessary to evaluate the effect of melamine exposure on breastfeeding problems.

Studies also showed changes in white blood cells, haemoglobin and haematocrit value of tilapia caused by melamine in feed²⁸ and altered white and red blood cell counts in mice with melamine exposure.²⁹ The present study also showed decreased white blood cells counts in the group with repeated melamine contact. Further research is needed to evaluate the effect of different levels of melamine contamination on haematologic parameters.

Strengths and limitations

The present study, for the first time, evaluated the impact of melamine exposure (in two periods) on maternal psychopathologic, haematologic and anthropometric parameters as well as infants’ anthropometric and sleep–crying characteristics.

As a limitation, the contamination level of melamine in breast milk was not evaluated. However, the primary aim of our study was the screening of melamine contamination of human milk. We used Randox’s patented InfiniPlex Biochip Array Technology, a standard competitive immunoassay format, specifically developed to ensure a 98% compliance with all currently monitored residues by the European Union.²⁰ Screening immunoassay methods typically provide results in less time and is cost-effective than other selective techniques, but they have no power of the methods based on high-performance liquid chromatography and mass spectrometry to unambiguously identify melamine and related compounds.^30,31 Present study showed that population in Urfa, Southeast Anatolia, had a high risk for melamine biohazards and there is a need to further investigate this area with detailed pharmacokinetic and pharmacodynamic studies in long-term follow-up cohort studies. Another limitation is the small sample size, 90 mothers with 180 breast milk samples. However, this study is a descriptive study with a convenience sample. As a strength of the study, two samples from each case were also taken with a 1-month interval, showing repeated exposure.

Conclusions

Women might have an exposure of melamine throughout their life including lactation period and melamine can pass through milk. Incidents of melamine contamination have become an international problem, risking the public health all over the world. This study with a convenient sample did not detect any effect on maternal psychopathologies, maternal bonding and infantile crying and sleeping problems. On the other hand, maternal repeated melamine exposure and transfer through breast milk lead to a negative impact on the height of her infant. Similarly, lower infant birth weight of mothers with repeated exposure could support this hypothesis. In addition, the mean value of maternal white blood cells in mothers with repeated exposure was lower than cases with no exposure. Therefore, melamine exposure of lactating women should be prevented. Breast milk screening can easily evaluate the environmental risks for mother–infant pairs. Further studies including environmental history and food sampling will be helpful in revealing the source of melamine contamination.

Footnotes

Acknowledgement

The authors are grateful to lactating mothers supplying breast milk.

Data accessiblity

Research data can be requested from the corresponding author.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The funding organization(s) played no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report or in the decision to submit the report for publication.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Hacettepe University, School of Medicine Institutional Review Board (project number: GO 17/687) approved the protocol.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported by Hacettepe University Scientific Research Projects Coordination Unit [Project No. THD-2017-15742].

Informed consent

Informed consent was obtained from parents included in the study.

ORCID iD

SS Yalçin

References

WHO. Toxicological and health aspects of melamine and cyanuric acid: report of a WHO expert meeting in collaboration with FAO, supported by Health Canada, Ottawa, Canada, 1–4 December 2008. Geneva: World Health Organization, 2009.

Alemanno

. European food import safety regime under a stress test: the melamine contamination of the global food supply chain. Erasmus L Rev 2010; 3(4): 203–215.

Chu

Wang

. Toxicity of melamine: the public health concern. J Environ Sci Health C Environ Carcinog Ecotoxicol Rev 2013; 31(4): 342–386.

EFSA. EFSA panel on contaminants in the food chain EFSA panel on food contact materials, enzymes, flavourings processing aids scientific opinion on melamine in food and feed. EFSA J 2010; 8(4): 1573.

Gabriels

Lambert

Smith

, et al.

Melamine contamination in nutritional supplements – is it an alarm bell for the general consumer, athletes, and ‘weekend warriors’?

Nutr J 2015; 14: 69.

Chan

Lau

Ting

, et al. Gestational and lactational transfer of melamine following gavage administration of a single dose to rats. Food Chem Toxicol 2011; 49: 1544–1548.

Chow

. Adverse child health impacts resulting from food adulterations in the Greater China Region. J Sci Food Agric 2017; 97: 3897–3916.

Bolden

Rochester

Kwiatkowski

. Melamine, beyond the kidney: a ubiquitous endocrine disruptor and neurotoxicant? Toxicol Lett 2017; 280: 181–189.

Chu

, et al. Melamine in prenatal and postnatal organs in rats. Reprod Toxicol 2013; 35: 40–47.

10.

Baynes

Barlow

Mason

, et al. Disposition of melamine residues in blood and milk from dairy goats exposed to an oral bolus of melamine. Food Chem Toxicol 2010; 48: 2542–2546.

11.

Sun

Wang

Shen

, et al. Residues of melamine and cyanuric acid in milk and tissues of dairy cows fed different doses of melamine. J Dairy Sci 2011; 94: 3575–3582.

12.

Yurdakok

Filazi

Ekici

, et al. Melamine in breast milk. Toxicol Res 2014; 3: 242–246.

13.

Dinleyici

Yildirim

Aydemir

, et al. Human milk antibiotic residue levels and their relationship with delivery mode, maternal antibiotic use and maternal dietary habits. Eur Rev Med Pharmacol Sci 2018; 22: 6560–6566.

14.

Zhu

Kannan

. Occurrence of melamine and its derivatives in breast milk from the United States and its implications for exposure in infants. Environ Sci Technol 2019; 53: 7859–7865.

15.

Yang

Wen

, et al. Four years follow-up of 101 children with melamine-related urinary stones. Urolithiasis 2013; 41: 265–266.

16.

Brockington

Fraser

Wilson

. The postpartum bonding questionnaire: a validation. Arch Womens Ment Health 2006; 9(5): 233–242.

17.

Yalçın

Örün

Özdemir

, et al. Türk annelerde doğum sonrası bağlanma ölçeklerinin güvenilirliği. Çocuk Sağlığı ve Hastalıkları Dergisi 2014; 57: 246–251.

18.

Engindeniz

Validation of Turkish form of Edinburgh postpartum depression scale. Masters Thesis, Ege University, Izmir, 1996.

19.

WHO. Child growth standards. WHO anthro (version 3.2. 2, January 2011) and macros, https://www.who.int/childgrowth/software/en/ (2011, accessed 22 March 2019).

20.

Randox. Food’s InfiniPlex array for milk kit. https://www.randox.com/infiniplex-milk-rfd/ (accessed 26 March 2019).

21.

Zhu

Kannan

. Distribution profiles of melamine and its derivatives in indoor dust from 12 countries and the implications for human exposure. Environ Sci Technol 2018; 52: 12801–12808.

22.

Filazi

Sireli

Ekici

, et al. Determination of melamine in milk and dairy products by high performance liquid chromatography. J Dairy Sci 2012; 95: 602–608.

23.

Kong

Choi

, et al. Hong Kong Chinese school children with elevated urine melamine levels: a prospective follow up study. BMC Public Health 2011; 11: 354.

24.

Chu

Tang

, et al. Adverse reproductive effects of maternal low-dose melamine exposure during pregnancy in rats. Environ Toxicol 2017; 32: 131–138.

25.

Sun

. A brief review of neurotoxicity induced by melamine. Neurotox Res 2017; 32: 301–309.

26.

Langman

Alon

Ingelfinger

, et al. A position statement on kidney disease from powdered infant formula-based melamine exposure in Chinese infants. Pediatr Nephrol 2009; 24: 1263–1266.

27.

Memis

Yalcin

. Human milk mycotoxin contamination: smoking exposure and breastfeeding problems. J Matern Fetal Neonatal Med 2019: 1–10. DOI: 10.1080/14767058.2019.1586879.

28.

Phromkunthong

Choochuay

Kiron

, et al. Pathophysiological changes associated with dietary melamine and cyanuric acid toxicity in red tilapia. J Fish Dis 2015; 38: 161–173.

29.

Abd-Elhakim

Mohamed

AA-R

Mohamed

. Hemato-immunologic impact of subchronic exposure to melamine and/or formaldehyde in mice. J Immunotoxicol 2016; 13(5): 713–722.

30.

Xia

Liu

, et al. A review of methods for detecting melamine in food samples. Crit Rev Anal Chem 2017; 47: 51–66.

31.

Ritota

Manzi

. Melamine detection in milk and dairy products: traditional analytical methods and recent developments. Food Anal Method 2017; 11(1): 128–147.

Presence of melamine in human milk and the evaluation of the effects on mother–infant pairs in a cohort study

Abstract

Keywords

Introduction

Materials and methods

Study design and participants

Outcomes of the study

Samples collection and analysis

Statistical analyses

Results

General characteristics of mother–infant pairs

Melamine detection and mother–infant characteristics

Discussion

Melamine transfer to human milk

Health concerns and melamine

Strengths and limitations

Conclusions

Footnotes

Acknowledgement

Data accessiblity

Declaration of conflicting interests

Ethical approval

Funding

Informed consent

ORCID iD

References