Dioxins and Polychlorinated Biphenyls in Human Breast Milk: Pilot Biomonitoring Data from Greater Poland Province

Abstract

Persistent organic pollutants such as polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polychlorinated biphenyls (PCBs) remain a public health concern due to their persistence, bioaccumulation, and potential health effects. Human breast milk is an important biomonitoring matrix for assessing maternal and infant exposure to lipophilic contaminants. This pilot study aimed to determine concentrations of PCDD/Fs, dioxin-like PCBs (dl-PCBs), and non-dioxin-like PCBs (ndl-PCBs) in breast milk samples collected from five lactating women residing in the Greater Poland Province and to explore potential determinants of exposure. Following participant recruitment, sample collection, and questionnaire-based assessment performed by the authors, breast milk samples were analyzed at the accredited Laboratory of Trace Analysis (Cracow University of Technology, Poland) using isotope dilution gas chromatography coupled with tandem mass spectrometry. Toxic equivalency values (TEQ) were calculated using World Health Organization 2005 toxic equivalency factors (WHO-TEFs). WHO-PCDD/F-TEQ ranged from 0.096 to 0.22 pg/g fresh weight. Median lipid-normalized WHO-PCDD/F-TEQ and total WHO-PCDD/F-PCB-TEQ concentrations were 3.5 and 4.7 pg/g lipid, respectively, remaining below the European Food Safety Authority (EFSA) reference level of 5.9 pg/g lipid; only one sample exceeded this threshold (6.2 pg/g lipid). Lipid-normalized WHO-PCB-TEQ correlated positively with maternal age (ρ = 0.949, p = 0.0389). The observed values were within the lower range reported in recent European studies. The congener patterns suggest a combination of chronic exposure to combustion by-products and long-term bioaccumulation of historical industrial pollutants. Although limited by the small sample size, this exploratory study provides preliminary regional biomonitoring data supporting future environmental exposure research.

1. Introduction

Persistent organic pollutants (POPs) include highly toxic and lipophilic compounds such as polychlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated biphenyls (PCBs). These substances are characterized by their remarkable environmental persistence and bioaccumulation along the food chain [1,2,3]. Of particular concern are dioxins (PCDD/Fs) and dioxin-like PCBs (dl-PCBs), due to their potent biological activity. This activity is primarily mediated through activation of the aryl hydrocarbon receptor (AhR), which may result in endocrine disruption, metabolic disturbances, carcinogenesis, and interference with normal developmental processes [1,4,5]. The most extensively studied and toxic congener, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), serves as the reference compound for toxicity assessment using toxic equivalency factors (TEFs) [2,4].

Dioxins have no intentional commercial applications; instead, they are generated as unintentional by-products of various industrial and thermal processes, including waste incineration, metal processing, and chlorine-based bleaching of pulp and paper. In addition, dioxins are formed during natural processes such as volcanic eruptions and forest fires [6,7]. They are also present in the combustion products of tobacco smoke [8].

In addition to dioxin-like compounds, non-dioxin-like PCBs (ndl-PCBs) constitute an important contributor of human exposure to POPs. Although ndl-PCBs do not exert their effects through AhR activation and are therefore not included in TEQ calculations, they are recognized for distinct toxicological properties, including neurotoxicity, endocrine disruption, and immunomodulatory effects [9,10,11]. Historically, PCBs were widely used in industry, particularly as dielectric fluids in transformers and capacitors, as well as in hydraulic systems and materials such as plasticizers and sealants. Despite the global ban on PCB production, these compounds persist in the environment due to their resistance to degradation, and they continue to be released from contaminated soils, sediments, old electrical equipment, and building materials [7,11].

Dietary intake, particularly the consumption of high-fat foods of animal origin such as meat, dairy products, and fish, remains the primary source of human exposure to dioxins and PCBs [5,11]. Due to a strong affinity for lipids, they bioaccumulate in human adipose tissue and may be transferred to the developing fetus via placental transfer and postnatally through breastfeeding. Consequently, human breast milk is widely recognized as one of the most informative and reliable biological matrices for assessing population exposure to lipophilic environmental contaminants [12,13]. Its non-invasive collection, and direct relevance to early-life exposure have made breast milk a cornerstone of international biomonitoring programs coordinated by the World Health Organization [13]. Importantly, breast milk reflects not only recent exposure but also the mobilization of POPs accumulated in the maternal body over a lifetime, driven by the breakdown of maternal fat stores during lactation [12,13].

Breastfed infants represent the group most highly exposed to dioxins and PCBs on a body-weight basis during a critical period of postnatal development, when detoxification systems and regulatory mechanisms are still immature [12,14,15]. Although breastfeeding is strongly recommended due to its well-established health benefits, the presence of persistent organic pollutants (POPs) in human milk remains a concern with regard to potential subtle and long-term health effects [16]. Over the past decades, increasing attention has been directed toward the potential effects of early-life exposure to dioxins and PCBs on developmental processes, for example enamel mineralization of first permanent teeth, which overlaps with periods of placental and breastfeeding-related exposure [17,18,19].

Despite extensive international literature on dioxin and PCB contamination of human breast milk, recent data from Poland remain limited. Available studies originate from different regions and sampling periods [20,21,22,23,24]. In the context of ongoing changes in environmental emissions, air quality, and dietary patterns, updated regional evidence is needed to better characterize current exposure levels among breastfeeding women.

The present study aimed to provide preliminary data on concentrations of PCDD/Fs, dl-PCBs, and ndl-PCBs in breast milk samples collected from women residing in the Greater Poland Province. To our knowledge, this pilot study provides the first contemporary data on dioxin and PCB concentrations in human breast milk from the Greater Poland region. Although the pilot nature of the study precludes causal inference, the findings contribute to the limited body of regional biomonitoring data and may serve as a basis for future large-scale studies on environmental exposure to POPs in early life.

2. Materials and Methods

2.1. Study Design and Biological Material

The study protocol was approved by the Bioethics Committee of the Karol Marcinkowski University of Medical Sciences in Poznan (decision no. 453/25, issued on 12 June 2025). Five lactating women residing in the Greater Poland Province were enrolled in the study. All participants were active donors of a human milk bank, namely The Regional Human Milk Bank at the Gynecology and Obstetrics Clinical Hospital of the Poznan University of Medical Sciences.

Participants were eligible for inclusion in the study if they met all of the following criteria:

-

currently lactating, actively breastfeeding their own infants;

-

provided written informed consent to participate in the study and to donate breast milk samples for research purposes;

-

permanent residence in the Greater Poland Province, Poland;

-

fulfillment of human milk bank donor requirements, including good general health, absence of acute infections, and no medical contraindications to milk donation;

-

negative results of routine serological screening tests required by Polish milk bank protocols (HIV, hepatitis B and C, cytomegalovirus, and syphilis);

-

ability and willingness to provide three serial milk samples collected at one-month intervals.

Participants were excluded from the study if any of the following conditions were present:

-

use of psychoactive drugs, tobacco smoking, or alcohol consumption;

-

excessive caffeine intake, defined as consumption of more than three cups of coffee per day or equivalent amounts of caffeine-containing beverages;

-

inability to comply with the sample collection schedule.

The study group comprised women aged 22 to 36 years, with a mean age of 31 years (median 34). All participants were enrolled based on voluntary consent. Each participant provided three breast milk samples collected at one-month intervals. The volume of a single sample was approximately 100 mL, resulting in a total volume of the pooled sample of about 300 mL of breast milk per donor.

The heterogeneity of sampling periods and participant characteristics was intentional and aimed to capture a broader range of potential exposure scenarios in this exploratory pilot study.

Milk samples were collected in accordance with standard human milk bank procedures, with strict hygienic procedures and measures preventing secondary contamination. After collection, samples were properly labeled, secured, and stored at a controlled temperature of −22 °C until laboratory analysis. Storage and transport conditions were designed to ensure the stability of the analyzed lipophilic compounds and to minimize the risk of sample degradation or contamination.

Breast milk samples were analyzed for concentrations of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs), dioxin-like polychlorinated biphenyls (dl-PCBs), and non-dioxin-like polychlorinated biphenyls (ndl-PCBs). The analyses allowed for the determination of individual congener profiles and calculation of TEQ, enabling assessment of potential infant exposure to the investigated compounds.

A structured questionnaire was used to collect information on maternal and infant characteristics, as well as lifestyle, environmental, and dietary factors potentially influencing exposure to dioxins and polychlorinated biphenyls. The questionnaire included data on maternal age, current body mass index (BMI), parity, breastfeeding history, and perinatal characteristics, including gestational age and birth weight. Information on environmental exposure was obtained through questions on place of residence, household heating sources, passive exposure to tobacco smoke, and potential contact with waste burning emissions in the residential area. In addition, participants were asked about selected behavioral factors, such as checking air quality before outdoor walks and the use of air purifiers. The question regarding air quality checking was intended to reflect general environmental awareness and did not assess specific behavioral responses or the exact sources of air quality information. Dietary habits were assessed with a focus on major sources of dioxin and PCB exposure, including the frequency of consumption of meat, fish, and dairy products both high-fat (e.g., cheese, cream, butter, whole milk) and low-fat (e.g., semi-skimmed milk, low-fat yogurt, and curd cheese). The questionnaire was developed for the purposes of this study and was not formally validated.

Ambient PM10 data (particulate matter with an aerodynamic diameter ≤ 10 µm) were retrieved from the Polish National Air Quality Monitoring System (https://powietrze.gios.gov.pl/pjp/archives, accessed on 13 April 2026). PM10 is commonly used as a proxy indicator of urban air pollution associated with traffic, residential heating, and other anthropogenic emission sources. Mean concentrations were calculated for each participant based on data from the nearest monitoring stations in the area of residence. To reflect recent environmental exposure, PM10 values were averaged over the 7 days preceding breast milk collection. In addition, annual mean PM10 concentrations for 2025 were obtained to characterize background environmental exposure at each study location. The calculated values are presented in

Table 1. Selected questionnaire-derived characteristics of breast milk donors included in the case-by-case analysis.

Variable	Case 1	Case 2	Case 3	Case 4	Case 5
Maternal and infant characteristics
Maternal age (years)	22	27	36	36	34
BMI (kg/m2)	23.8	23.2	27.1	26.2	25.0
Breastfeeding child order	First child	Second child	First child	Third child	Second child
Months postpartum	1, 2, 3	1, 2, 3	1, 2, 3	6, 7, 8	1, 2, 3
Sampling period	August–October	September–November	August–October	August–October	October–December
Gestational age at birth (weeks)	26	39	27	41	25
Birth weight (g)	685	3350	815	4190	700
Lifestyle and environmental factors
Place of residence	Small town in an urban–rural municipality (~2000 inhabitants)	Large city (>500,000 inhabitants)	Medium-sized city (~70,000 inhabitants)	Large city (>500,000 inhabitants)	Medium-sized city (~100,000 inhabitants)
PM10 (7-day mean, µg/m3)	20.4	25.3	19.8	19.2	33.4
PM10 (annual mean, µg/m3)	26.1	24.9	20.1	24.9	25.8
Primary household heating	Wood-based	District	Gas	District	District
Passive smoking exposure	Yes (outdoors only)	No	No	No	No
Self-reported exposure to waste burning	Unknown	No	No	No	Unknown
Checking air quality	Never	Yes	Occasionally	Occasionally	Occasionally
Air purifier use	No	No	No	No	No
Dietary habits
Meat consumption	Daily	≤1 time per week	Several times per week	Daily	Daily
High-fat dairy consumption	Daily	Several times per week	Several times per week	Daily	Daily
Low-fat dairy consumption	≤1 time per week	Several times per day	≤1 time per week	Several times per week	Several times per day
Fish consumption	No	Occasionally	Occasionally	Once per week	Once per week

.

2.2. Analytical Methods (PCDD/Fs, dl-PCBs, ndl-PCBs)

Analyses were performed using an isotope dilution reference method based on gas chromatography coupled with tandem mass spectrometry (TSQ 9000 Triple Quadrupole GC–MS/MS System, Thermo Scientific, Waltham, MA, USA), in accordance with Commission Regulation (EU) No 709/2014 [25]. The analytical procedure involved the use of 13C-labeled internal standards, solvent extraction of lipids, multistep clean-up using chromatographic columns, and fractionation prior to final determination. The analyses were conducted in an accredited laboratory specializing in the trace analysis of persistent organic pollutants—the Laboratory of Trace Analysis named after Professor Adam Grochowalski, Faculty of Chemical Engineering and Technology, Cracow University of Technology, Poland.

TEQ were calculated using the World Health Organization 2005 TEFs (WHO-TEFs), with 2,3,7,8-TCDD as the reference compound, in accordance with Commission Regulation (EU) 2023/915 [26]. Results were expressed as WHO-PCDD/F-TEQ, WHO-PCB-TEQ, and combined WHO-PCDD/F-PCB-TEQ. Non-dioxin-like PCBs were reported separately as the sum of six indicator congeners (ICES-6: PCB-28, 52, 101, 138, 153, and 180).

All results were reported using the upper-bound approach, whereby concentrations below the limit of quantification (LOQ) were assigned values equal to the LOQ, providing a conservative estimate of exposure. This approach assumes a worst-case scenario for non-quantified congeners and minimizes the risk of underestimating cumulative exposure. Concentrations were expressed both per gram of fresh weight (fw), in accordance with European regulatory limits for food intended for infants and young children [26], and per gram of lipid, allowing comparison with international human milk biomonitoring studies.

2.3. Quality Assurance and Quality Control

All analyses were conducted using validated analytical procedures under comprehensive quality assurance and quality control conditions. Measurement uncertainty was expressed as expanded uncertainty with a coverage factor of k = 2, corresponding to a confidence level of approximately 95%. LOQs were determined empirically based on actual measurement data for each analytical series. The estimated uncertainty of congener determination amounted to approximately 23% for PCDD/Fs, 24% for dl-PCBs, and 21% for ndl-PCBs.

2.4. Regulatory Reference Values

For contextual interpretation of results, concentrations were compared with maximum levels established by European Union legislation for food intended for infants and young children. According to Commission Regulation (EU) 2023/915, the maximum permitted levels are 0.10 pg/g fresh weight for WHO-PCDD/F-TEQ, 0.20 pg/g fresh weight for WHO-PCDD/F-PCB-TEQ, and 1.0 ng/g fresh weight for the sum of indicator ndl-PCBs (ICES-6) [26].

2.5. Data Analysis

The obtained data were analyzed using descriptive methods. Due to the pilot nature of the study and the small sample size (n = 5), statistical analyses were limited. Exploratory correlation analysis was restricted to maternal age and conducted using Kendall’s tau coefficient, which is suitable for small samples. A significance level of p < 0.05 was adopted. In addition, a case-by-case analysis was conducted by relating the analytical results of individual breast milk samples to selected demographic, environmental, and dietary information obtained from donor questionnaires.

3. Results

3.1. Characteristics of the Study Participants and Breast Milk Samples

Selected demographic, environmental, and dietary characteristics of the donors derived from the questionnaire are summarized in Table 1.

Concentrations of PCDD/Fs, dl-PCBs, and ndl-PCBs are presented in

Table 2. Concentrations of PCDD/Fs and PCBs in human breast milk samples expressed as WHO-TEQ and non-dioxin-like PCBs (ICES-6), reported per gram of fresh weight (fw) and lipid.

Case	Lipid Content	WHO-PCDD/F-TEQ		WHO-PCB-TEQ		WHO-PCDD/F-PCB-TEQ		ndl-PCBs (ICES-6)
	%	(pg/g fw)	(pg/g Lipid)	(pg/g fw)	(pg/g Lipid)	(pg/g fw)	(pg/g Lipid)	(ng/g fw)	(ng/g Lipid)
1	4.37	0.096	2	0.020	0.44	0.116	2.44	0.35	7.8
2	5.09	0.22 *	4.3	0.058	1.1	0.28 *	5.4	1.4 *	28
3	2.96	0.13 *	4.4	0.054	1.8	0.18	6.2	0.63	21
4	4.32	0.12 *	2.9	0.069	1.6	0.19	4.5	0.78	15
5	4.26	0.14 *	3.5	0.049	1.2	0.19	4.7	0.32	7.6
Median	4.32	0.13	3.5	0.054	1.2	0.19	4.7	0.63	15

* Values exceeding EU maximum levels for food intended for infants and young children [26]: WHO-PCDD/F-TEQ: 0.10 pg/g fw, WHO-PCDD/F-PCB-TEQ: 0.20 pg/g fw, ndl-PCBs (ICES-6): 1.0 ng/g fw. Results are expressed as upper-bound concentrations. Measurement uncertainty is reported as expanded uncertainty (k = 2, 95% confidence level): ~23% for PCDD/Fs, ~24% for dl-PCBs, and ~21% for ndl-PCBs.

, expressed both on a fresh weight basis and normalized to lipid content.

When expressed on a fresh weight basis, WHO-PCDD/F-TEQ ranged from 0.096 to 0.22 pg/g fw. Exceedance of the EU maximum level for PCDD/Fs (0.10 pg/g fw) was observed in four out of five samples. Total WHO-PCDD/F-PCB-TEQ concentrations ranged from 0.116 to 0.28 pg/g fw. The regulatory limit for total TEQ (0.20 pg/g fw) was exceeded in one sample. Concentrations of ndl-PCBs ranged from 0.32 to 1.4 ng/g fw. The EU maximum level for ndl-PCBs (1.0 ng/g fw) was exceeded in one sample. When expressed on a lipid weight basis, WHO-PCDD/F-TEQ concentrations ranged from 2.0 to 4.4 pg/g lipid, WHO-PCB-TEQ from 0.44 to 1.8 pg/g lipid, total WHO-PCDD/F-PCB-TEQ from 2.44 to 6.2 pg/g lipid, and ndl-PCBs (ICES-6) from 7.6 to 28 ng/g lipid.

A strong positive correlation was observed between maternal age and WHO-PCB-TEQ expressed per lipid weight (ρ = 0.949, p = 0.0389). No significant associations were found between maternal age and lipid-normalized PCDD/F or ndl-PCB levels (ρ = 0.316, p = 0.6056 and ρ = 0.105, p = 1.000).

Table 3. Mass concentrations and WHO-TEQ contributions of PCDD/F and dioxin-like PCB congeners and concentrations of non-dioxin-like PCBs in human milk samples.

PCDD/Fs Congeners (pg/g Lipid)	Case 1		Case 2		Case 3		Case 4		Case 5
	Mass	TEQ	Mass	TEQ	Mass	TEQ	Mass	TEQ	Mass	TEQ
2,3,7,8-TCDD	0.15	0.15	0.34	0.34	0.4	0.4	0.27	0.27	0.34	0.34
1,2,3,7,8-PeCDD	0.76	0.76	1.59	1.59	1.6	1.6	1	1	1.2	1.2
1,2,3,4,7,8-HxCDD	0.29	0.029	0.65	0.065	0.61	0.061	0.36	0.036	0.53	0.053
1,2,3,6,7,8-HxCDD	1.2	0.12	2.5	0.25	3	0.3	2.4	0.24	1.5	0.15
1,2,3,7,8,9-HxCDD	0.41	0.041	0.65	0.065	0.67	0.067	0.51	0.051	0.58	0.058
1,2,3,4,6,7,8-HpCDD	2.4	0.024	2.5	0.025	2.8	0.028	2.2	0.022	3.4	0.034
OCDD	10	0.003	9.4	0.0028	13	0.0039	15	0.0045	10	0.0030
2,3,7,8-TCDF	0.21	0.021	0.32	0.032	0.38	0.038	0.3	0.03	0.89	0.089
1,2,3,7,8-PeCDF	0.15	0.0045	0.37	0.011	0.25	0.0075	0.14	0.0042	0.77	0.023
2,3,4,7,8-PeCDF	2.1	0.63	5.1	1.5	4.9	1.5	3	0.90	3.6	1.1
1,2,3,4,7,8-HxCDF	0.89	0.089	1.5	0.15	2.1	0.21	1.3	0.13	1.8	0.18
1,2,3,6,7,8-HxCDF	0.64	0.064	1.5	0.15	1.4	0.14	1.1	0.11	1.6	0.16
1,2,3,7,8,9-HxCDF	n.d.	0.0086	n.d.	0.0089	n.d.	0.011	n.d.	0.017	n.d.	0.013
2,3,4,6,7,8-HxCDF	0.3	0.030	0.63	0.063	0.73	0.073	0.46	0.046	1	0.10
1,2,3,4,6,7,8-HpCDF	0.55	0.0055	0.91	0.0091	0.55	0.0055	n.d.	0.0091	1.2	0.012
1,2,3,4,7,8,9-HpCDF	n.d.	0.0018	n.d.	0.0022	n.d.	0.0026	n.d.	0.0041	n.d.	0.0025
OCDF	0.44	0.00013	0.17	0.000051	n.d.	0.000051	n.d.	0.000075	0.28	0.000084
dl-PCB Congeners (pg/g Lipid)	Case 1		Case 2		Case 3		Case 4		Case 5
	Mass	TEQ	Mass	TEQ	Mass	TEQ	Mass	TEQ	Mass	TEQ
PCB 77	7.94	0.00079	3.6	0.00036	7.7	0.00077	3.2	0.00032	5.1	0.00051
PCB 126	3.33	0.33	7.1	0.71	14	1.4	13	1.3	9.9	1.0
PCB 169	2.53	0.076	11	0.33	9.3	0.28	5.7	0.17	3.9	0.12
PCB 81	0.44	0.00013	0.57	0.00017	0.76	0.00023	0.6	0.00018	0.95	0.00029
PCB 105	132	0.0040	172	0.0052	403	0.012	571	0.017	220	0.0066
PCB 114	20.9	0.00063	91	0.0027	82	0.0025	80	0.0024	33	0.0010
PCB 118	628	0.019	911	0.027	1811	0.054	2105	0.063	875	0.026
PCB 123	5.42	0.00016	5	0.00015	11	0.00033	17	0.00051	9.8	0.00029
PCB 156	229	0.0069	1331	0.040	831	0.025	432	0.013	219	0.0066
PCB 157	40.3	0.0012	226	0.0068	158	0.0047	95	0.0029	54	0.0016
PCB 167	80.5	0.0024	278	0.0083	304	0.0091	191	0.0057	90	0.0027
PCB 189	25.6	0.00077	123	0.0037	82	0.0025	46	0.0014	28	0.00084
ndl-PCB Congeners (ng/g Lipid)	Case 1		Case 2		Case 3		Case 4		Case 5
	Mass		Mass		Mass		Mass		Mass
PCB 28	0.29		0.42		0.47		0.47		0.37
PCB 52	0.17		0.11		0.23		0.12		0.14
PCB 101	0.59		0.13		0.27		0.16		0.20
PCB 138	2.07		5.03		5.85		3.86		1.7
PCB 153	3.34		13.1		8.98		6.92		3.12
PCB 180	1.33		8.96		5.39		3.10		2.0

For PCDD/Fs and dl-PCBs, mass concentrations are expressed as pg/g lipid and TEQ as pg WHO-TEQ/g lipid. n.d.—not detected (below the limit of quantification).

presents the congener profile of PCDD/Fs, dioxin-like PCBs (dl-PCBs), and non-dioxin-like PCBs (ndl-PCBs) detected in the analyzed milk samples, including their mass concentrations and contributions to WHO-TEQ.

The congener profile of PCDD/Fs in the analyzed human milk samples was dominated by highly chlorinated congeners. Among PCDDs, the highest mass concentrations were observed for OCDD, whereas among PCDFs the predominant congener was 2,3,4,7,8-PeCDF. OCDD reached values of approximately 9–15 pg/g lipid across the analyzed samples. Other abundant congeners included 1,2,3,6,7,8-HxCDD and 1,2,3,4,6,7,8-HpCDD. The main contributors to the total PCDD/F-TEQ were 2,3,4,7,8-PeCDF and 1,2,3,7,8-PeCDD, which exhibited the highest TEQ contributions among the analyzed congeners. Among dioxin-like PCBs, PCB 118, PCB 105, and PCB 156 showed the highest mass concentrations. However, the largest contribution to the total dl-PCB TEQ was associated with PCB126, reflecting its high toxic equivalency factor. The ndl-PCB fraction was dominated by the indicator congeners PCB 153, PCB 138, and PCB 180.

3.2. Case-by-Case Analysis

Sample no. 1 was obtained from the youngest participant (22 years old), residing in an urban–rural area. This sample exhibited the lowest WHO-PCDD/F-TEQ, WHO-PCB-TEQ, and WHO-PCDD/F-PCB-TEQ among all analyzed samples. The total ndl-PCB concentration (ICES-6) was also among the lowest. The questionnaire indicated daily consumption of meat and high-fat dairy products and the use of wood-based heating. Occasional passive exposure to tobacco smoke was reported, although smoking by household members occurred exclusively outdoors. Notably, this participant was the only one who reported no consumption of fish.

Sample no. 2 originated from a 27-year-old participant residing in a large urban area. It showed relatively high WHO-PCDD/F-TEQ and WHO-PCDD/F-PCB-TEQ concentrations and the highest ndl-PCB concentration (ICES-6) among all analyzed samples. All measured parameters exceeded the corresponding EU maximum levels for food intended for infants and young children. Based on the questionnaire data, the participant reported frequent intake of low-fat dairy products, rare consumption of meat and fish, and consumption of high-fat dairy products several times per week. The household used district heating, and no passive smoking exposure or waste burning in the residential area was reported. The participant declared regular checking of air quality before outdoor activities. Despite the absence of self-reported high-risk environmental exposures, this sample exhibited elevated concentrations of both dioxins and PCBs, particularly ndl-PCBs, indicating that factors not captured by the questionnaire, such as past long-term exposure to historical industrial pollutants (e.g., from old electrical equipment or building materials) common in large metropolitan areas, might have contributed to the observed contamination levels.

Sample no. 3 was obtained from a 36-year-old participant with the highest BMI, breastfeeding her first child, living in a medium-sized city. This sample showed elevated WHO-PCDD/F-TEQ concentrations, exceeding the applicable EU maximum levels. The ndl-PCB concentration was within the regulatory limit. It also exhibited the highest total toxicity in the study, with a value of 6.2 pg WHO-TEQ/g lipid. The questionnaire data indicated meat and high-fat dairy consumption several times per week. The primary source of household heating was gas. No passive smoking exposure or waste burning in the residential area was reported, and air quality was checked occasionally before outdoor walks.

Sample no. 4 originated from a 36-year-old participant residing in a large city and was collected during a later stage of lactation (6–8 months postpartum). This sample showed WHO-PCDD/F-TEQ concentrations exceeding the EU maximum level, while WHO-PCDD/F-PCB-TEQ and ndl-PCB concentrations remained below the respective regulatory thresholds. The participant reported daily consumption of meat and high-fat dairy products, with fish consumption once per week. The household used district heating, and no passive smoking exposure or waste burning in the residential area was declared. Air quality was checked occasionally before outdoor activities. Case no. 4 differed from the others due to the later timing of milk sampling and the fact that the participant was breastfeeding her third child.

Sample no. 5 was obtained from a 34-year-old participant living in a medium-sized city. This sample showed elevated WHO-PCDD/F-TEQ concentrations, exceeding the corresponding EU maximum levels. However, the ndl-PCB concentration was the lowest in the study group. The questionnaire revealed daily consumption of meat and high-fat dairy products, with fish consumption reported once per week. The household used district heating. No passive smoking exposure was reported, while the participant was not aware of any exposure to waste burning in the residential area. Air quality was checked occasionally before outdoor walks. The sample collection period (October–December) coincided with the heating season and represented the only sample fully collected during this period.

4. Discussion

4.1. Current Levels of Dioxins and PCBs in Human Milk

The WHO-PCDD/F-PCB-TEQ concentrations measured in the present study fall within the range reported in international human biomonitoring programs, including the global WHO/UNEP surveys on PCDD/Fs and dioxin-like PCBs in human milk. Such long-term monitoring initiatives have consistently demonstrated substantial regional variability in concentrations, reflecting differences in industrial history, environmental contamination, and dietary patterns [12,13].

Over the past decades, many countries have reported a pronounced decline in PCDD/F and PCB levels following regulatory restrictions on production and emissions. However, more recent data suggest that the decreasing trend has slowed or stabilized in several regions, indicating that background exposure persists despite earlier reductions [27]. In Central Europe, emissions from energy production and domestic heating have been identified as particularly important contributors to environmental contamination, reflecting the continued reliance on solid fuels across both industrial and residential sectors. In addition, uncontrolled waste burning, including both use in domestic heating systems and open burning in outdoor environments, may further contribute to the formation and release of dioxin-like compounds [28].

To provide a broader context for interpreting the results of the present pilot study, previously published data on dioxin and PCB concentrations in human breast milk from Poland and international monitoring programs are summarized in

Table 4. Comparison of WHO-PCDD/F-TEQ, WHO-PCDD/F-PCB-TEQ, WHO-dl-PCB-TEQ and ndl-PCB concentrations in human breast milk in national and international studies.

Region	Sampling Period	WHO-PCDD/F-TEQ 1	WHO-PCDD/F-PCB-TEQ 1	WHO-dl-PCB-TEQ 2	ndl-PCBs 3	Sample Type	Method
Greater Poland Province (this study)	2025	3.5 (2–4.4)	4.7 (2.44–6.2)	0.05 (0.02–0.07)	15 (7.6–28)	Individual composite 5 (n = 5)	GC-MS/MS
Greater Poland Province [20]	2004	n.r.	n.r.	n.r.	133 (63.0–413) 4	Individual (n = 22)	GC-MS
Lesser Poland Province [21]	n.r. (study funded in 2014)	0.815 (0.31–7.0)	1.2 (0.35–8.0)	n.r.	7.55 (0.35–35.0)	Individual Composite 5 (n = 4)	GC-MS/MS
Northwestern Poland [22]	n.r. (ethical approval in 2015)	n.r.	n.r.	n.r. (0.033–5.676)	13.78 (0.343–20.31)	Individual longitudinal (n = 920 from 96 donors)	GC-MS
Central Poland [23]	2007–2011	7.42 (0.52–22.81)	9.63 (1.59–25.17)	2.59 (0.02–6.00)	n.r.	Individual (n = 110)	HRGC/HRMS
Central Poland [24]	2008–2010	6.065 (0.305–10.57)	7.71 (0.431–14.27)	1.519 (0.022–4.498)	n.r.	Individual (n = 40)	HRGC/HRMS
Europe (WHO/UNEP data compilation) [13]	2000–2019	6.50 (2.40–18.6)	12 (4.10–26.9)	n.r.	118 (14.6–1009)	Pooled (n = 92)	HRGC/HRMS
Non-European countries (WHO/UNEP data compilation) [13]	2000–2019	3.48 (1.01–41.2)	5.38 (1.29–49.0)	n.r.	16.5 (0.90–96.5)	Pooled (n = 140)	HRGC/HRMS

Data presented as median (range). ¹ pg/g lipid; ² pg/g fresh weight; ³ ng/g lipid; ⁴ Σ15 PCBs (not ICES-6); ⁵ composite samples were prepared by combining up to three milk samples collected from the same donor; n.r.—not reported. Values are reported in the units used in the original publications. Abbreviations: GC–MS/MS, gas chromatography–tandem mass spectrometry; GC–MS, gas chromatography–mass spectrometry; HRGC/HRMS, high-resolution gas chromatography/high-resolution mass spectrometry.

.

It should be noted that Poland has not participated in the WHO/UNEP coordinated global human milk surveys assessing PCDD/Fs and dioxin-like PCBs. Consequently, the available information on POP concentrations in breast milk from Poland originates from independent regional studies conducted at different times and using varying analytical approaches. According to WHO/UNEP global surveys conducted between 2000 and 2019, median WHO-PCDD/F-PCB-TEQ levels in pooled European milk samples were approximately 12 pg/g lipid, whereas non-European countries reported median values around 5 pg/g lipid [14]. The median observed in the present study (4.7 pg/g lipid; range 2.44–6.2) lies closer to the lower range of international data. The reported concentrations are markedly lower than those documented in the 1980s and 1990s, when substantially higher exposures were observed in several industrialized regions [12,13].

The observed values are in line with the long-term reduction in environmental emissions of dioxins and PCBs following regulatory restrictions, although the persistence and bioaccumulative nature of these compounds continue to result in detectable background exposure [27,28]. When interpreting ndl-PCB levels, methodological differences between studies must be considered. The 2004 study from Greater Poland reported Σ15 PCBs (median ~133 ng/g lipid), whereas the present study quantified only the ICES-6 congeners [20]. Recalculation of the earlier data by summing the median concentrations of the four ICES-6 congeners common to both datasets (PCB 101, PCB 138, PCB 153, and PCB 180) yielded an estimated median concentration of approximately 85.3 ng/g lipid, indicating that current ndl-PCB concentrations are markedly lower than those reported two decades earlier.

Data from central Poland also provide useful reference values for interpreting the present findings. A large cohort study conducted in Łódź between 2007 and 2011, including 110 lactating women, reported total WHO-PCDD/F-PCB-TEQ ranging from 1.59 to 25.17 pg/g lipid (median 9.63 pg/g lipid) [23]. A smaller study from the same region conducted in 2008–2010 reported WHO-PCDD/F-PCB-TEQ concentrations ranging from 0.431 to 14.27 pg/g lipid (median 7.71 pg/g lipid) [24]. Although direct comparisons are limited by differences in study design and sampling period, the concentrations observed in the present study were at the lower end of values reported in central Poland cohorts.

Similarly, data from Lesser Poland [21] and northwestern Poland [22] indicate variability in PCB and TEQ levels across regions, with concentrations generally overlapping international ranges. Witczak et al. demonstrated decreasing PCB concentrations with advancing lactation, supporting the role of breastfeeding as an elimination pathway for lipophilic pollutants [22]. In our study, Case 4, who was breastfeeding her third child and was in later lactation (6–8 months postpartum), did not demonstrate lower PCB concentrations compared to the other participants.

4.2. Congener Profile and Environmental Background

The congener distribution observed in the present study revealed characteristic patterns for both PCDD/Fs and PCBs detected in the analyzed human milk samples. PCDD/Fs are not intentionally produced but are formed as by-products of various thermal and industrial processes. Major sources of their environmental emissions include combustion-related activities such as waste burning, domestic heating using coal or biomass, traffic emissions, and certain industrial processes, including ferrous and non-ferrous metal production. Recent studies from Central Europe indicate that contemporary environmental contamination with PCDD/F is largely associated with diffuse combustion-related sources, particularly road transport, domestic wood burning, and coal or household waste combustion [27]. Within the PCDD/F group detected in the present study, higher chlorinated congeners predominated, with OCDD representing the largest proportion of total PCDD/F mass concentrations, followed by HpCDD and HxCDD congeners. Such profiles are generally attributed to the environmental persistence, lipophilicity, and strong bioaccumulation potential of highly chlorinated compounds [12,13,27].

In addition to PCDD/Fs, a considerable proportion of the total toxic equivalency was associated with dl-PCBs detected in the analyzed samples. Among these compounds, mono-ortho congeners, particularly PCB 118, PCB 105, and PCB 156, accounted for the largest proportion of dl-PCB mass concentrations. Similar patterns have been reported in other Polish human biomonitoring studies [20,21,23]. The non-ortho congener PCB 126 represented the major contributor to WHO-PCB-TEQ due to its high TEF. The non-dioxin-like PCB fraction was dominated by the indicator congeners PCB 153, PCB 138, and PCB 180, which are commonly reported as the most abundant PCB congeners in human milk due to their high environmental persistence and strong bioaccumulation potential [20,22]. This pattern was particularly pronounced in Case 2, a resident of a large urban area, who exhibited the highest ndl-PCB concentration in the study (28 ng/g lipid). Despite the absence of clearly identified high-risk exposures in the questionnaire, the elevated concentrations observed in Case 2 may reflect the influence of long-term urban background exposure, potentially related to historical industrial activity, legacy contamination in building materials, or other environmental sources typical of large cities.

Unlike PCDD/Fs, which are formed unintentionally during combustion and various industrial processes, PCBs were intentionally manufactured and widely used in electrical equipment, construction materials, and hydraulic systems. Due to their high environmental persistence and long biological half-lives, PCBs continue to contribute to human exposure decades after regulatory restrictions were implemented [12,27,28].

Taken together, the observed congener profile suggests that the measured concentrations are more consistent with background environmental exposure influenced by diffuse combustion-related emissions.

4.3. Interindividual Variability and Determinants of Exposure

The case-by-case analysis revealed considerable interindividual variability in contaminant concentrations among the analyzed breast milk samples. A previous Polish study demonstrated that concentrations of dl-PCBs and their TEQ were significantly associated with maternal age (r = 0.3814 and r = 0.2817, respectively), whereas no such relationship was observed for PCDD/F. Instead, PCDD/F levels in breast milk appeared to depend primarily on dietary habits (particularly fish and dairy consumption) and environmental exposure related to place of residence (e.g., waste burning and traffic-related pollution), rather than age [23]. This interpretation is consistent with recent evidence from a study conducted in Taranto, Italy [29], one of the most industrially contaminated regions in Europe. The authors reported that age-related increases were most pronounced for ndl-PCBs and dl-PCBs, whereas no significant association with age was observed for PCDDs and PCDFs. Such observations support the concept that PCBs reflect long-term bioaccumulation, while dioxins and furans are more strongly influenced by current environmental and dietary exposures. Age remains a key determinant of PCB levels, reflecting decades of accumulation of these persistent compounds, whose production has long been discontinued. In line with previous observations, a strong positive correlation between maternal age and dl-PCB-TEQ was observed in our study (ρ = 0.949, p = 0.0389).

Parity and lactation history may also contribute to variability, as the transfer of lipophilic compounds through breastfeeding is considered an important elimination pathway for persistent organic pollutants. Such mechanisms may partly explain differences observed between participants of similar age but different reproductive histories.

Dietary habits represent another important factor influencing exposure patterns. Consumption of animal-derived foods, particularly fish and high-fat dairy products, is widely recognized as a major pathway of exposure to dioxins and PCBs due to their bioaccumulation in food chains [23,30]. In the present study, the participant with the lowest contaminant levels reported no fish consumption, which may suggest a potential dietary influence, although the small sample size precludes definitive conclusions.

Interesting observations have been reported in a pilot study conducted in Lesser Poland, where higher concentrations of dioxins and ndl-PCBs were observed in a participant consuming a traditional diet and residing in a small town, whereas markedly lower levels were found in a vegan participant living in a large city [21]. The observed pattern suggests that dietary choices—particularly the avoidance of animal-derived fats—may substantially reduce exposure even in urban environments. These observations may have practical implications for nutritional counseling in lactating women.

Seasonality may also influence environmental exposure in Central Europe due to increased emissions during the heating season. However, no consistent seasonal pattern was observed in the present study. Given that PM10 is commonly used as an indicator of combustion-related emissions, which represent an important environmental source of dioxins, short-term PM10 levels were analyzed as a proxy for recent exposure. Short-term PM10 levels were more variable than annual averages, which remained similar across locations. No consistent relationship with PCDD/F concentrations was observed. Recent studies have reported a weak association between modeled ground-level PM10 exposure estimates derived using atmospheric dispersion modeling and dioxin levels in human milk, with only small increases in TEQ observed at higher PM10 concentrations [31].

Taken together, the available evidence indicates that contaminant levels in human milk are determined by a complex interplay of biological, lifestyle, and environmental factors, with the relative contribution of each varying between individuals.

4.4. Breast Milk as a Biomonitoring Matrix

When expressed on a fresh-weight basis and compared with EU regulatory limits for food intended for infants and young children, exceedances of the maximum level for WHO-PCDD/F-TEQ were observed in four out of five samples. However, these limits were established for commercially produced infant foods and are presented here only as a reference point, as human breast milk reflects cumulative maternal exposure and may therefore contain higher concentrations of lipophilic contaminants.

The current risk assessment framework developed by the European Food Safety Authority (EFSA) is based on epidemiological evidence indicating that reduced sperm concentration is the most sensitive endpoint of toxicity associated with early-life exposure to dioxins and dioxin-like PCBs [32,33]. This effect has been linked to a serum concentration of approximately 7 pg WHO-TEQ/g lipid, identified as the no-observed-adverse-effect level (NOAEL) [32]. To prevent exceeding this threshold in children, EFSA established a tolerable weekly intake (TWI) of 2 pg WHO-TEQ/kg body weight. This value was derived using toxicokinetic modeling to link external exposure with internal concentrations, taking into account the cumulative nature of these compounds, and subsequently translated into exposure scenarios during infancy. Under typical breastfeeding and dietary conditions, this corresponds to a concentration of approximately 5.9 pg WHO-TEQ/g lipid in human milk [31]. In the present study, the median concentration of total WHO-PCDD/F-PCB-TEQ was 4.7 pg TEQ/g lipid, slightly below this modeled level, with only one sample exceeding it (6.2 pg TEQ/g lipid).

Importantly, other developmental endpoints may occur at similar or only slightly higher exposure levels. EFSA associated developmental enamel defects in children with an estimated maternal weekly intake of approximately 3 pg PCDD/F-TEQ/kg body weight per week. While this dose is measured for the mother, the adverse effect manifests in the child because enamel mineralization of early developing permanent teeth overlaps with the peak period of contaminant transfer during pregnancy and breastfeeding [33].

At the same time, EFSA has highlighted important uncertainties associated with the currently applied TEFs based on the WHO 2005 scheme. In particular, concerns have been raised regarding the contribution of certain congeners, especially PCB 126, to calculated dl-PCB-related toxicity. Recent draft EFSA assessments incorporating the WHO 2022 TEFs [34] suggest substantially lower health-based guidance values, including a proposed reduction of the TWI to approximately 0.6 pg WHO-TEQ/kg body weight and a corresponding decrease in the modeled reference level for human milk to around 2.2 pg WHO-TEQ/g lipid [34]. However, application of the updated TEFs is also expected to reduce calculated total TEQ, particularly in samples with a high contribution of dl-PCBs. Notably, recalculation of exposure metrics using the revised TEFs appears to weaken previously observed associations between total TEQ and semen quality in human studies, which has led to a greater reliance on experimental animal data in the derivation of health-based guidance values [35].

Importantly, neither the present findings nor the current EFSA framework provide any basis for questioning breastfeeding. On the contrary, existing risk assessment approaches are designed to ensure its safety by limiting maternal body burden and accounting for transfer during pregnancy and lactation. Notably, EFSA states that the TWI should not be applied to breastfed infants, as their higher exposure per body weight is already accounted for in the derivation of maternal guidance values [33,35]. Breastfeeding remains the optimal method of infant feeding, and its well-established benefits clearly outweigh potential risks associated with background environmental contaminants [36].

At the same time, human milk offers a unique opportunity to monitor environmental exposure, serving as a sensitive biomarker of population-level contamination and supporting efforts to further reduce pollutant levels in the environment [37].

Further longitudinal human studies are needed to better clarify the potential long-term effects of early-life exposure to PCDD/Fs and PCBs on child health and development, including developmental dental defects.

4.5. Study Strengths and Limitations

International datasets from European monitoring programs and global WHO/UNEP initiatives are frequently used as reference points for national exposure assessments. In Poland, several studies have previously evaluated concentrations of dioxins and PCBs in human breast milk, including large cohort investigations conducted in central regions of the country. However, many of these data originate from earlier sampling periods, and contemporary regional biomonitoring data remain limited. Notably, the only available data from the Greater Poland region originate from a study conducted nearly two decades ago, which assessed PCBs only and did not include dioxins or TEQ-based toxicity assessment [20]. The present pilot study therefore provides updated and more comprehensive information on current exposure levels in the Greater Poland Province and may serve as a reference point for future monitoring initiatives.

A significant strength of this study is its comprehensive analytical scope. The substantial volume of breast milk analyzed per donor provided a sufficient amount of milk fat for extraction, which significantly enhanced the detectability of individual congeners that might otherwise fall below the limit of quantification. Specifically, our analysis quantified the full spectrum of 29 dioxin-like compounds (17 PCDD/Fs and 12 dl-PCBs) alongside the six indicator ndl-PCBs (ICES-6: PCB 28, 52, 101, 138, 153, and 180). In addition, the use of validated analytical techniques ensures high reliability of the obtained measurements. While following the conservative upper-bound approach mandated by Commission Regulation (EU) 2023/915 [26] could theoretically lead to an overestimation of toxicity levels, the actual impact of non-detected congeners on the total TEQ results is minimal, accounting for less than 1% of the final values. This confirms that the reported TEQ are highly reliable.

Furthermore, the study group was heterogeneous, encompassing a range of maternal ages, body mass indices, parity, infant characteristics (including preterm and full-term births), and diverse residential settings across the Greater Poland region. This diversity provides a broader perspective on possible exposure scenarios. Differences in sampling season and postpartum period may also have contributed to variability in contaminant concentrations between participants. For this reason, information on sampling timing and ambient PM10 levels was included to provide additional environmental context.

The primary limitation of this study is the very small sample size, which does not allow statistical inference at the population level. The results should therefore be interpreted as exploratory, providing preliminary insight into current exposure levels.

5. Conclusions

The present study confirmed the presence of dioxins and polychlorinated biphenyls in human breast milk, indicating early-life exposure among breastfed infants. The WHO-PCDD/F-PCB-TEQ concentrations measured in this study fall within the lower range of values reported in recent European and international human milk biomonitoring studies. Considerable interindividual variability was observed, consistent with bioaccumulation of these persistent compounds and the multifactorial nature of environmental exposure. Although the study is limited by its very small sample size and its exploratory design, it provides updated regional biomonitoring data for the Greater Poland Province. Future studies involving larger, representative cohorts are needed to enable statistical analysis and to further explore associations between contaminant levels and biological, environmental, and dietary factors.