Literature search strategy

The search was conducted across Web of Science Core Collection, Scopus, PubMed and Google Scholar databases. Key search terms included microplastics, nanoplastics, submicron plastics, PFAS, food web, trophic transfer, bioaccumulation, human exposure, analytical methods such as liquid chromatography–mass spectrometry (LC–MS/MS), Raman spectroscopy, Fourier-transform infrared (FTIR) spectroscopy, thermal analysis, risk assessment and food safety. Relevant articles were published between 2000 and June 2026. Both recent and older sources were included if considered valuable for the subject of research. The peer-reviewed original papers, reviews and meta-analyses, as well as regulatory guidelines and recommendations from the World Health Organization (WHO), European Food Safety Authority (EFSA), Food and Agriculture Organization of the United Nations (FAO), Organization for Economic Cooperation and Development (OECD), United States Environmental Protection Agency (USEPA) and other international organisations were prioritised. Attention was paid to the articles that demonstrated the occurrence of micro/nanoplastics and PFAS in food webs, analytical methods for their identification and quantification, environmental fate, mechanisms of biological transfer, toxicological effects and regulatory aspects. The articles with conflicting conclusions were also included and critically assessed.

Physical particulate effects

One of the MNPsʼ attributed impacts is that recalcitrant MPs and NPs may trigger local inflammatory processes, oxidative stress due to the enhanced production of ROS and barrier dysfunction. The persistence of insoluble and enzymatically resistant particles causes prolonged stimulation of macrophages, leading to MPs accumulation in tissues (Kadac-Czapska et al., Reference Kadac-Czapska, Ośko, Knez and Grembecka2024).

Biological interactions beyond local toxicity

Experimental studies have shown that the exposure to MNPs may alter gut microbial community composition, particularly the abundance of short-chain fatty acids producing bacteria. The effect depends on the polymer type, particle size, surface modification, exposure duration and study models. Moreover, there is evidence of the association between MNP exposure, dysbiosis and intestinal inflammation. However, it is unclear how the observations relate to human exposure since dysbiosis was not reported for all cases and may be triggered by other factors such as diet or inflammation (Boctor et al., Reference Boctor, Oweda and El-Hadidi2023; Bora et al., Reference Bora, Gogoi, Sharma, Borah, Deka, Bora, Naorem, Das and Teli2024; Ebaid et al., Reference Ebaid, Boctor, Husien, Abdelaal, Gazar, Badr and Farag2026).

A significant challenge associated with the interpretation of MNP toxicology is the discrepancy between the exposure conditions used in experiments and the physicochemical properties of MNPs observed in the environment. Most often experiments are conducted using engineered MNPs which are present at concentrations far higher than those observed in the environment in order to observe an effect. Nonetheless, these types of experiments are essential in elucidating potential mechanisms of action, as well as environmental risk assessment provided that the properties of the tested MNPs and exposure paradigms correspond to the ones present in nature such as particle size, surface structure, ageing and exposure regimes. By contrast, chronically exposed organisms are frequently subjected to a mixture of MNPs typically found within their ambient environment at concentrations below those typically tested in laboratories. However, these lower concentrations cannot be easily approximated experimentally because of the complexities of realistically recreating environmental exposure scenarios with sufficient spatial and temporal resolution and analytical challenges in characterising abundance and sizes of MNPs in the environment (Shruti et al., Reference Shruti, Pérez-Guevara, Elizalde-Martínez and Kutralam-Muniasamy2020; Hasegawa et al., Reference Hasegawa, Mizukawa, Yeo, Sekioka, Takada and Nakaoka2022; Monikh et al., Reference Monikh, Materić, Valsami-Jones, Grossart, Altmann, Holzinger, Lynch, Stubenrauch and Peijnenburg2025).

Micro and nanoplastics as dynamic chemical carriers

Apart from their direct particulate effects, MNPs represent dynamic chemical carriers whose relevance depends on the interaction between environmental ageing, contaminant chemistry and gastrointestinal conditions (Saeed et al., Reference Saeed, Afzal, Nimra, Ahmad, Rehman, Maqsood, Jan, Ahmad, Karim, Jeon, Mustafa, Zaman and Khawar2026). The affinity of externally adsorbed chemicals such as PFAS is also more dynamic than fixed. Natural weathering causes oxygen-containing groups, such as carbonyl and carboxyl, to be generated on polymer surfaces, thus modifying the surface polarity and increasing the specific surface area. The relatively hydrophobic surface has an innate tendency to accumulate hydrophobic organic contaminants. At the same time, a weathered surface becomes increasingly polar and, accordingly, the hydrophobic content decreases for strongly hydrophobic compounds, with the ability to preferentially absorb short-chain ionisable PFAS molecules (Lenka et al., Reference Lenka, Kah and Padhye2023).

Gastrointestinal desorption kinetics and plasticisers

Upon ingestion, the physiological gradients of the gut promote rapid desorption of these chemicals. Core body temperature close to 37 °C and the pH range, along with the presence of digestive enzymes and bile salts, may contribute to the accelerated desorption of intrinsic plasticizers such as phthalates and bisphenols. This process occurs at a significantly higher level than in seawater under simulated gut conditions (Bakir et al., Reference Bakir, Rowland and Thompson2014; Chiu et al., Reference Chiu, Warner, Nowak, Flaws and Mei2020). Although MNPs pose a problem by carrying various types of absorbed organic pollutants, there are disputes about the toxicological importance of their translocation. From a mass balance perspective, the amount of PFAS, BPA and other additives ingested with microplastics is expected to be lower than the intake from food, water and commercial products. Moreover, under certain physiological conditions, MNPs can behave as a passive sorbent medium. Therefore, the actual challenge presented by microplastics may be related to their ability to persist in the body and their impact on biodistribution rather than the mass of co-ingested toxins.

In April 2023, the European Food Safety Authority (EFSA) drastically reduced the tolerable daily intake (TDI) for bisphenol A (BPA) from 4 μg/kg bw per day down to 0.2 ng/kg bw per day (Prueitt and Goodman, Reference Prueitt and Goodman2024). This unprecedented revision reduced the TDI by approximately 20,000-fold and was anchored by ‘non-guideline’ academic data demonstrating an increase in the percentage of Th17 immune cells in the spleen; an intermediate biomarker linked to allergic lung inflammation and autoimmune disorders. Because EFSA estimated that mean and high dietary exposures across all age groups exceed this new threshold by two to three orders of magnitude, the derivation highlights the potential for biological effects at very low exposure levels of the chemicals associated with MNPs.

However, the attempt to relate these intermediate biomarkers to broad hazard statements has caused the EFSA to differ from both the EMA and the German Federal Institute for Risk Assessment (BfR) in its conclusions regarding these compounds’ risk to humans. The EFSA’s reliance on a T-cell assay, which focuses on the mechanism of action rather than apical disease endpoints, is at odds with the EMA and BfR’s established procedures. The BfR dismissed the EFSA’s proposed 0.2 ng/kg derivation as unwarranted due to a lack of biological plausibility and adequate control of background contamination in the related studies, instead calculating a TDI of 200 ng/kg bw/day (Zoeller et al., Reference Zoeller, Birnbaum, Collins, Heindel, Hunt, Iguchi, Kortenkamp, Myers, Vom Saal, Sonnenschein and Soto2023; Juric et al., Reference Juric, Franz and Welle2024). While this still represents a 20-fold reduction from the 2015 baseline, it remains 1,000 times higher than EFSA’s threshold. In addition, international bodies, including the U.S. Food and Drug Administration (FDA) and Food Standards Australia New Zealand (FSANZ), have thus far maintained their current safety ratings, arguing that the incorporation of non-guideline immunological endpoints overstates the risk at typical consumer exposure levels (Prueitt and Goodman, Reference Prueitt and Goodman2024; Vom Saal et al., Reference Vom Saal, Antoniou, Belcher, Bergman, Bhandari, Birnbaum, Cohen, Collins, Demeneix, Fine, Flaws, Gayrard, Goodson, Gore, Heindel, Hunt, Iguchi, Kassotis and Kortenkamp2024; Wen et al., Reference Wen, Wang, Tian and Wang2024).

The same low-dose logic extends to the legacy POPs, in 2018, EFSA lowered the tolerable weekly intake for dioxins and dioxin-like PCBs to 2 pg WHO-TEQ per kilogram body weight, a value that average European dietary exposure already exceeds according to many reports (Knutsen et al., Reference Knutsen, Alexander, Barregård, Bignami, Brüschweiler, Ceccatelli, Cottrill, Dinovi, Edler, Grasl‐Kraupp, Hogstrand, Nebbia, Oswald, Petersen, Rose, Roudot, Schwerdtle, Vleminckx and Vollmer2018; Kannan and Vimalkumar, Reference Kannan and Vimalkumar2021). The repeated retroactive reductions of regulatory exposure limits extend beyond bisphenols and dioxins and reflects a broader evolution in toxicological risk assessment. As shown in Table 1, the largest reductions, often spanning several orders of magnitude, have occurred following advances in analytical sensitivity and the incorporation of more subtle biological endpoints, including immunotoxicity, endocrine disruption and neurodevelopmental effects. Thus, the main sources of uncertainty associated with the human health effects of MNPs include dose metrics, exposure assessment, dose–response and mechanistic domains.

Table 1.

Regulatory threshold revisions for persistent environmental and food-web contaminants

Table 1. Long description

The table contains 11 rows of contaminants with their respective regulatory changes:

* B P A: E F S A T D I metric. Previous threshold 4,000 n g / k g b w / day (2015) revised to 0.2 n g / k g b w / day (2023), a 20,000 times reduction based on immunotoxicity.

* P F O A: U S E P A R f D metric. Previous threshold 20 n g / k g b w / day (2016) revised to 0.0015 n g / k g b w / day (2024), a 13,333 times reduction based on immune suppression.

* P F O S: U S E P A R f D metric. Previous threshold 20 n g / k g b w / day (2016) revised to 0.0079 n g / k g b w / day (2024), a 2,531 times reduction based on immune suppression.

* P F B S: U S E P A R f D metric. Previous threshold 20,000 n g / k g b w / day (2014) revised to 300 n g / k g b w / day (2021), a 66.7 times reduction based on thyroid toxicity.

* GenX: U S E P A R f D metric. Previous threshold 80 n g / k g b w / day (2018) revised to 3.0 n g / k g b w / day (2021), a 26.7 times reduction based on hepatic lesions.

* Melamine: U S F D A T D I metric. Previous threshold 0.63 m g / k g b w / day (2007) revised to 0.063 m g / k g b w / day (2008), a 10.0 times reduction based on renal pathology.

* Dioxins and D L - P C B s: E F S A T W I metric. Previous threshold 14 p g T E Q / k g b w / week (2001) revised to 2.0 p g T E Q / k g b w / week (2018), a 7.0 times reduction based on reproductive toxicity.

* D B P: E F S A T D I metric. Previous threshold 0.05 m g / k g b w / day (1995) revised to 0.01 m g / k g b w / day (2005), a 5.0 times reduction based on embryotoxicity.

* Aluminium: J E C F A P T W I metric. Previous threshold 7 m g / k g b w / week (1989) revised to 2.0 m g / k g b w / week (2011), a 3.5 times reduction based on developmental toxicity.

* Cadmium: E F S A T W I metric. Previous threshold 7 micro-g / k g b w / week (pre-2009) revised to 2.5 micro-g / k g b w / week (2009), a 2.8 times reduction based on renal tubular toxicity.

* Methyl mercury: J E C F A P T W I metric. Previous threshold 3.3 micro-g / k g b w / week (1988) revised to 1.6 micro-g / k g b w / week (2003), a 2.1 times reduction based on foetal neurodevelopment.

Note: The dataset illustrates how regulatory thresholds for established chemical contaminants have been revised as toxicological evidence, analytical sensitivity and risk-assessment frameworks have evolved. These revisions reflect changes in hazard identification, exposure assessment and regulatory interpretation for specific chemical contaminants and should not be interpreted as predictive benchmarks for microplastic regulation.

Abbreviations: BPA, bisphenol A; bw, body weight; DBP, di-butyl phthalate; DL-PCBs, dioxin-like polychlorinated biphenyls; EFSA, European Food Safety Authority; GenX, hexafluoropropylene oxide dimer acid (HFPO-DA); JECFA, Joint FAO/WHO Expert Committee on Food Additives; PFBS, perfluorobutane sulphonic acid; PFOA, perfluorooctanoic acid; PFOS, perfluorooctane sulphonate; PTWI, provisional tolerable weekly intake; RfD, reference dose; TEQ, toxic equivalency; TDI, tolerable daily intake; TWI, tolerable weekly intake; US EPA, United States Environmental Protection Agency; US FDA, United States Food and Drug Administration.

As detailed in Table 1, regulatory thresholds for several established chemical contaminants have been retroactively changed over the years, following advances in analytical capability, exposure assessment and toxicological understanding. These examples demonstrate that regulatory values are dynamic and may change when new evidence identifies previously unrecognised hazards or improves understanding of sensitive biological endpoints. However, they are not directly applicable to microplastics and nanoplastics, because the latter lack an exposure health-based model, and have no unified analytical methods for determining their concentrations in the human body. Consequently, the lack of health-based guidance values for MNPs does not necessarily indicate either their safety or the likelihood of imposing more stringent requirements in the future. The assessment of whether the critical values were exceeded or are close to being exceeded requires an initial determination of the amount of microplastics ingested. Nonetheless, the number varies significantly due to the use of different methods, each of which has its own constraints.

Quantifying human dietary exposure

Dietary intake of MNPs is widely reported, but the magnitude of exposure is subject to considerable uncertainty stemming from differences in analytical sensitivity, particle size resolution, reporting units and differing assumptions about the relationship between particle number and mass. Mechanistically, this uncertainty reflects the fact that exposure is dictated not only by particle number but also by physicochemical factors such as particle size, gastrointestinal retention, polymer density and fragmentation state. As a result, a particle count is not necessarily indicative of the total mass of plastic ingested, since the relative contribution of small particles (rich in surface area and thus in surface-active compounds) can be disproportionately large. Estimates of human exposure to plastics span a range of six orders of magnitude, largely because they rely on different particle-size distributions, concentration datasets and particle-to-mass conversion assumptions. In their study, Senathirajah et al. (Reference Senathirajah, Attwood, Bhagwat, Carbery, Wilson and Palanisami2021) address the challenge by performing a comprehensive literature review that provides a first-order estimate of the number of polymer particles ingested, as well as the corresponding mass of these particles. The study synthesised data from 59 publications to estimate human ingestion of microplastics by combining reported particle counts with particle-size and density assumptions. The resulting estimates were highly variable, with reported intake spanning approximately 0.1–5 g plastic per person per week, illustrating the uncertainty introduced when particle numbers are converted to mass. Within that estimate, shellfish accounted for approximately 26.4 g per capita per year, followed by salt at 7.4 g, beer at 0.5 g and drinking water at 0.08 g. The upper bound was subsequently popularised, through a 2019 WWF report, as the claim that a person consumes a credit card of plastic each week, a framing that replaced the 0.1 to 5 g interval with a flat 5 g figure representing the ceiling rather than the central tendency, a restatement later criticised on methodological grounds (WWF, 2019; Pletz, Reference Pletz2022).

Count-based assessments place the burden lower but still substantial. Cox et al. (Reference Cox, Covernton, Davies, Dower, Juanes and Dudas2019), using 402 data points from 26 studies and more than 3,600 processed samples, estimated annual dietary intake at 39,000 to 52,000 particles, rising to 74,000 to 121,000 once inhalation is included and noted that conservative assumptions render these values a probable underestimate. Drinking water source alone re-shaped the total, with bottled water averaging 94 particles per litre against 4.2 for tap, a 22-fold difference equivalent to an additional 90,000 particles annually for exclusively bottled consumers versus roughly 4,000 for tap. The mass-based and count-based estimates do not reconcile. A bottom-up model by Mohamed Nor et al. (Reference Mohamed Nor, Kooi, Diepens and Koelmans2021) returns a median adult intake near 4.1 μg per week, approximately six orders of magnitude below the upper Senathirajah bound, while a re-analysis of the original calculation reproduced 0.1 to 5.5 g per week across plausible particle-size scenarios and judged the 5 g ceiling physically unreasonable (Pletz, Reference Pletz2022). This divergence reflects fundamental differences in particle-size assumptions, detection limits and conversion approaches.

Agricultural produce: The crack-entry pathway

Contrary to the early assumption that root membranes act as absolute barriers, experimental studies demonstrate uptake of sub-micrometric plastics in selected crop species, indicating that plant uptake is biologically plausible under some environmental conditions. Mechanistically, uptake depends on particle size, surface chemistry and plant anatomical features. In some crops, sub-micrometre particles have been shown to enter through transient discontinuities associated with lateral root emergence, followed by movement through the apoplast and xylem.

However, this pathway is strongly constrained by particle dimensions, aggregation state, surface charge and interactions with soil organic matter, which determine mobility and bioavailability under field conditions. Li et al. (Reference Li, Shi, Yang, Xiao, Kehoe, Gun’ko, Boland and Wang2020) demonstrated that sub-micrometre polystyrene enters crops such as wheat and lettuce via a crack-entry mode, bypassing the Casparian strip at transient fissures where lateral roots emerge and then translocates through the apoplast into the xylem. Field surveys of produce report high baseline contamination: Oliveri Conti et al. (Reference Oliveri Conti, Ferrante, Banni, Favara, Nicolosi, Cristaldi, Fiore and Zuccarello2020) recorded between 52,050 and 233,000 particles per gram across six commonly consumed fruits and vegetables, with apples and carrots the most contaminated, and an estimated daily intake from apples reaching roughly 4.6 × 10⁵ particles for adults and 1.4 × 10⁶ particles for children. Such crops can additionally act as sinks for PFAS introduced through contaminated biosolids and mulching films (Boctor et al., Reference Boctor, Hoyle, Farag, Ebaid, Walsh, Whiteley and Murphy2025). Importantly, detection of particle uptake does not equate to efficient dietary transfer. Soil-to-plant movement represents a series of selective barriers, including rhizosphere retention, aggregation, root membrane interactions and dilution during plant growth. Consequently, laboratory demonstrations of uptake provide mechanistic evidence for possibility rather than quantitative estimates of human dietary exposure.

Trophic transfer of flame retardants in animal proteins

Minimally processed terrestrial proteins generally carry lower baseline concentrations, whereas marine food webs favour biomagnification (Gray, Reference Gray2002). Microplastics can serve as a dietary vector for brominated flame retardants. In a controlled prey–predator experiment, Hasegawa et al. (Reference Hasegawa, Mizukawa, Yeo, Sekioka, Takada and Nakaoka2022) exposed the benthic fish Myoxocephalus brandtii either to polyethylene microplastics compounded with the flame retardants BDE-209 and DBDPE suspended in water, or to mysid prey (Neomysis spp.) pre-exposed to the same particles; prey-mediated ingestion produced significantly higher tissue concentrations of both flame retardants than waterborne exposure, while three co-loaded ultraviolet stabilisers accumulated similarly by either route, a divergence the authors attributed to the greater hydrophobicity of the flame retardants. These observations suggest an important distinction between the trophic behaviour of the polymer and its associated chemicals. Trophic magnification factors above one have been reported for legacy POPs and several brominated flame retardants, whereas particle concentrations may decline at higher trophic levels. Thus, trophic transfer may amplify the chemical component of plastic-associated exposure more consistently than the particle burden itself.

Beverages and liquid matrices

Liquid matrices are uniquely susceptible to processing shear and to migration from packaging (Table 2) (Kumar et al., Reference Kumar, Kumar, Wei, Chopra, Rudra and Rabbani2025). Bottled water is an especially intense exposure route: using stimulated Raman scattering microscopy, Qian et al. (Reference Qian, Gao, Lang, Deng, Bratu, Chen, Stapleton, Yan and Min2024) estimated an average of about 2.4 × 10⁵ particles per litre, roughly 90% of them nanoplastics and around two to three orders of magnitude above earlier microplastic-only counts, with unbottled tap water markedly lower. These absolute values are contested, as a subsequent analysis argued that the procedural blanks were themselves contaminated and that polyamide being identified as dominant in polyethylene terephthalate bottles points to contamination or misidentification (Materić, Reference Materić2024). Commercial milk has likewise been reported to carry MNPs consistent with abrasion of polymer milking and processing tubing, though published particle counts remain inconsistent and await harmonised confirmation.

Table 2.

Representative polymer types and associated chemical classes are reported in common food matrices

Table 2. Long description

The table consists of three columns and ten rows of data.

* Row 1: Bottled water (Qian et al., 2024). Polymers: P E T, polyamide (P A). Chemicals: Phthalates, bisphenol A (B P A), P F A S.

* Row 2: Apples (fruit) (Oliveri Conti et al., 2020). Polymers: Polyethylene (P E). Chemicals: Agricultural stabilisers, P F A S.

* Row 3: Salad crops (carrots) (Li et al., 2020; Oliveri Conti et al., 2020). Polymers: Polypropylene (P P), P E. Chemicals: Mulching-film plasticisers, P F A S.

* Row 4: Milk (commercial) (Kutralam-Muniasamy et al., 2020). Polymers: P E, P P. Chemicals: Tubing plasticisers, endocrine disruptors.

* Row 5: Soft drinks (Shruti et al., 2020). Polymers: P E T, epoxy can-lining resins. Chemicals: B P A and B P S from linings, phthalates.

* Row 6: Eggs (Liu et al., 2022). Polymers: Polyethylene (P E). Chemicals: Lipophilic endocrine disruptors partitioning into yolk.

* Row 7: Fish (seafood) (Hasegawa et al., 2022). Polymers: P E, polystyrene (P S). Chemicals: P C B s, brominated flame retardants, P F A S.

* Row 8: Poultry (chicken) (Shifa and Mostafa, 2026). Polymers: P E, P P. Chemicals: Feed-derived P F A S, stabilisers.

* Row 9: Beef (minimally processed) (Hantoro et al., 2026). Polymers: Polyethylene (P E). Chemicals: Environmental contaminants, P F A S.

* Row 10: Tap water (Cox et al., 2019). Polymers: P E, P P, P V C. Chemicals: P F A S from environmental runoff.

Note: Entries summarise the dominant polymers and the additive or adsorbed chemical classes described in the cited literature; they are indicative rather than exhaustive, and reported concentrations vary with method and region.

Abbreviations: PA, polyamide; PE, polyethylene; PET, polyethylene terephthalate; PP, polypropylene; PS, polystyrene; PVC, polyvinyl chloride.

Dietary exposure to forever chemicals (PFAS)

While MNPs vector adsorbed contaminants, PFAS are pervasive dietary contaminants. Human exposure occurs predominantly through the consumption of contaminated drinking water, fish and eggs (Rekibi et al., Reference Rekibi, Duflos, Grard and Dehaut2025). In aquatic food webs, perfluorooctane sulphonate (PFOS) biomagnifies, rendering predatory fish a primary vector for human intake. In terrestrial systems, agricultural soils amended with contaminated biosolids transfer PFAS into crop tissues and livestock feed, establishing a direct pathway into dairy and meat products. This direct molecular exposure parallels and compounds the particle-mediated delivery of adsorbed forever chemicals (Rekibi et al., Reference Rekibi, Duflos, Grard and Dehaut2025).

Vulnerable windows: Early-life exposure

Exposure is not uniform across the population. Polypropylene infant feeding bottles, which dominate the global market, release roughly 1.3 to 16.2 million microplastic particles per litre during standard formula preparation, with sterilisation and hot water markedly increasing release (Li et al., Reference Li, Shi, Yang, Xiao, Kehoe, Gun’ko, Boland and Wang2020). Because intake scales inversely with body mass, the estimated daily intake per kilogram is roughly threefold higher in children than adults across matrices such as apples and bottled water (Oliveri Conti et al., Reference Oliveri Conti, Ferrante, Banni, Favara, Nicolosi, Cristaldi, Fiore and Zuccarello2020), which, with the placental transfer described in section ‘Biological fate of MNPs: From barrier crossing to human tissue evidence,’ argues for age-stratified rather than population-average risk modelling. Collectively, dietary studies not only demonstrate widespread exposure but also reveal the central limitation of current risk assessment: exposure estimates remain strongly dependent on analytical capability. Therefore, understanding the biological significance of MNP exposure requires first addressing whether current methodologies can accurately resolve particle abundance, size distribution and polymer identity.

The toxicokinetic bridge and the asymmetry of evidence

The health endpoints of the chemical payload are far better characterised than any particulate effect, and their persistence is physiological as well as environmental. Human serum elimination half-lives are measured in years, with arithmetic means near 5.4 years for PFOS, 8.5 years for PFHxS and 3.8 years for PFOA (Olsen et al., Reference Olsen, Burris, Ehresman, Froelich, Seacat, Butenhoff and Zobel2007), so chronic low-dose intake combined with slow clearance makes tissue accumulation expected rather than anomalous. The hazard is also formally graded: in November 2023, the International Agency for Research on Cancer re-classified PFOA as a Group 1 human carcinogen, citing limited human evidence for renal cell and testicular cancer with strong mechanistic evidence, and classified PFOS as Group 2B (Zahm et al., Reference Zahm, Bonde, Chiu, Hoppin, Kanno, Abdallah, Blystone, Calkins, Dong, Dorman, Fry, Guo, Haug, Hofmann, Iwasaki, Machala, Mancini, Maria-Engler and Møller2024). Comparable levels of evidentiary certainty have not yet been established for particle-specific microplastic effects, and that contrast frames the risk characterisation in section ‘Weighing the evidence: A confidence-based risk assessment framework.’

Crossing the blood–brain barrier

On entering the circulation, bare polymer surfaces are rapidly coated by host proteins and lipids to form a biomolecular corona that governs the particle’s biological fate. Molecular-dynamics modelling indicates that cholesterol within this corona enhances the uptake of nanoplastics across the lipid bilayer of the blood–brain barrier (BBB), supporting a mechanistic basis for nanoplastic interaction and translocation across BBB lipid membranes (Kopatz et al., Reference Kopatz, Wen, Kovács, Keimowitz, Pichler, Widder, Vethaak, Hollóczki and Kenner2023).

(A) Following oral exposure, polymers were detected in the frontal cortex, liver and kidney, indicating translocation beyond the gastrointestinal tract and accumulation in distant organs (adapted from Nihart et al., Reference Nihart, Garcia, El Hayek, Liu, Olewine, Kingston, Castillo, Gullapalli, Howard, Bleske, Scott, Gonzalez-Estrella, Gross, Spilde, Adolphi, Gallego, Jarrell, Dvorscak and Zuluaga-Ruiz2025). (B) The EFSA 2023 reassessment of BPA lowered the tolerable daily intake by approximately 20,000-fold, highlighting growing recognition of potential health risks associated with plastic-derived contaminants (adapted from (Lambré et al., Reference Lambré, Barat Baviera, Bolognesi, Chesson, Cocconcelli, Crebelli, Gott, Grob, Lampi, Mengelers, Mortensen, Rivière, Silano, Steffensen, Tlustos, Vernis, Zorn, Batke and Bignami2023).

Neural tissue has emerged as a potential site of particle accumulation, though both the magnitude and biological significance of reported burdens remain actively debated. Using Py-GC/MS, Nihart et al. (Reference Nihart, Garcia, El Hayek, Liu, Olewine, Kingston, Castillo, Gullapalli, Howard, Bleske, Scott, Gonzalez-Estrella, Gross, Spilde, Adolphi, Gallego, Jarrell, Dvorscak and Zuluaga-Ruiz2025) reported high concentrations of micro- and nanoplastics in the human frontal cortex, with a roughly 50% rise in median concentration from 3,345 μg/g in 2016 decedents to 4,917 μg/g in 2024 decedents, and polyethylene as the dominant polymer (Figure 2).

Figure 2.

Evidence suggesting systemic polymer translocation. Reported tissue burdens remain under active investigation because quantitative estimates remain highly dependent on analytical methodology and contamination control.

Figure 2. Long description

Panel A, titled Systemic Translocation Nihart et al., 2025, is a bar graph. The y-axis is Polymer Mass in micrograms per gram tissue, ranging from 0 to 5000. The x-axis lists three tissue types. Frontal Cortex shows a tall dark gray bar at approximately 4100 with a large error bar. Hepatic Tissue and Renal Tissue both show much shorter dark gray bars at approximately 400 with smaller error bars.

Panel B, titled Regulatory Shift: EFSA B P A Threshold, is a logarithmic bar graph. The y-axis is Tolerable Daily Intake in nanograms per kilogram body weight per day, ranging from 10 super minus 4 to 10 super 3. The x-axis has two categories. Pre-2023 T D I is represented by a tall dark red bar reaching the 10 super 3 level. Revised 2023 T D I is represented by a very short green bar at the 10 super minus 4 level. A curved arrow points from the top of the red bar down to the green bar, labeled 20,000-fold Reduction.

Brains from individuals with documented dementia carried a markedly higher median of 26,076 μg/g; the authors caution against causal interpretation, noting that brain atrophy, impaired blood–brain barrier integrity and reduced clearance are features of dementia that would themselves be expected to raise tissue concentrations. The absolute concentrations have been questioned: an author correction resolved duplicated supplementary images, but a separate published critique in the same journal flags limited contamination controls and absent validation steps that bear on the reliability of the reported values (Monikh et al., Reference Monikh, Materić, Valsami-Jones, Grossart, Altmann, Holzinger, Lynch, Stubenrauch and Peijnenburg2025). More broadly, absolute tissue burden estimates remain highly sensitive to sampling strategy, contamination control, analytical workflow and polymer identification methodology, making quantitative comparisons between studies difficult. Independent replication using rigorously contamination-controlled workflows, harmonised analytical protocols and inter-laboratory validation is therefore required before the reported tissue burdens can be considered robust. At present, these findings demonstrate biological plausibility rather than definitive population-level exposure.

Cardiovascular pathology and plaque vulnerability

The systemic presence of these particles is linked to cardiovascular morbidity. In a 2024 prospective, multicentre observational study, Marfella et al. (Reference Marfella, Prattichizzo, Sardu, Fulgenzi, Graciotti, Spadoni, D’Onofrio, Scisciola, La Grotta, Frigé, Pellegrini, Municinò, Siniscalchi, Spinetti, Vigliotti, Vecchione, Carrizzo, Accarino and Squillante2024) evaluated 304 patients (257 completing follow-up) undergoing carotid endarterectomy; polyethylene was detected within the atheroma of 58.4% of the cohort at a mean of 21.7 μg per milligram of plaque and polyvinyl chloride in 12.1% at a mean of 5.2 μg per milligram. Patients with detectable plastics in their plaque had a higher risk of the primary endpoint of myocardial infarction, stroke or death from any cause, with a hazard ratio of 4.53 (95% CI, 2.00–10.27; P < 0.001). Because the study is observational and single-cohort, and though concerns have been raised regarding pre-analytical contamination control and the potential for intraoperative contamination, limitations that warrant independent replication, the association is best read as strongly suggestive but insufficient to establish causality, particularly in the absence of independent cohort replication and standardised contamination controls.

Maternal–foetal transfer and reproductive sinks

Because MNPs and their associated endocrine-disrupting payloads have been reported to cross biological barriers, reproductive tissues may act as potential sites of accumulation. Ragusa et al. (Reference Ragusa, Svelato, Santacroce, Catalano, Notarstefano, Carnevali, Papa, Rongioletti, Baiocco, Draghi, D’Amore, Rinaldo, Matta and Giorgini2021) provided the first evidence of microplastics in human placentas, showing that maternal dietary or environmental exposures may result in maternal transfer of these pollutants to the foetus. Although detection has now been reported by multiple independent studies, reported concentrations remain method dependent, and continued improvements in contamination control and analytical standardisation are required before absolute tissue burdens can be confidently compared across studies.

Subsequent Py-GC/MS analysis detected microplastics in all 47 canine and 23 human testes examined, averaging 329 μg/g in human testis, with polyethylene the dominant polymer and polyvinyl chloride correlating with reduced sperm counts in the canine samples (Hu et al., Reference Hu, Garcia, Nihart, Liu, Yin, Adolphi, Gallego, Kang, Campen and Yu2024). The detection of particles in reproductive tissues raises concern because plastic-associated additives can possess endocrine activity; however, direct in vivo linkage between tissue-resident particles, additive release and reproductive outcomes remains unresolved.