Microplastics in the brain The BfRshort forGerman Federal Institute for Risk Assessment assesses a controversial study on microplastics and nanoplastics in human organs

This study deals with the detection of microplastic particles in samples of organs from deceased individuals from various hospitals in the USA at two different sampling times (2016 and 2024). It is therefore a multicentre observational study in which the groups were also divided according to diagnosed dementia, among other factors. Organs such as the liver, kidneys and brain were examined for the presence of microplastics of various polymer types using analytical methods, in particular pyrolysis gas chromatography-mass spectrometry (Py-GC/MS). Several types of plastic were detected in the organ samples, with polyethylene (PE) consistently accounting for the highest proportion. In addition, higher plastic content was measured in the brain samples of dementia patients. Plastic particles were detected in particular in immune cells in the walls of the cerebral blood vessels (cerebrovascular walls). No other health parameters were examined. Beyond the correlation, the study does not make any claims about causal links between plastic exposure and health effects. However, the study notes that the plastic levels measured in the newer samples were higher than those in the older samples, which is attributed to increased environmental exposure. The study emphasises the need to investigate microplastic exposure, intake and distribution pathways, and the resulting potential health effects in more detail.

Fundamentals

The study by Nihart et al.short foret alii (lat. "and others") appears to have been conducted with great care and uses advanced, state-of-the-art methods for the detection of microplastics in human organ samples. The focus is on the brain, although other organs, such as the liver and kidneys, were also examined. 

The key findings of the study are that:

• the human brain contains measurable amounts of microplastics, especially polyethylene. The particles ranged in size from 0.2 to 1 micrometre (µmshort formicrometre). It is unclear how these particles got there.
• The amounts of microplastics have increased in recent years. Regardless of age, gender and other socio-economic factors, an increase was observed in deceased individuals in 2024 compared to 2016.
• Higher amounts were found in brain samples from deceased individuals with detected dementia than in the other groups, regardless of age.

These findings are highly significant in terms of their timeliness and novelty. This is the first time that such data has been demonstrated in humans. The results appear to be fundamentally rational, but must be verified by further research. In recent years, comparable studies have been published that have detected microplastics in other human organs, such as External Link:arterial plaques, External Link:olfactory bulbs, External Link:blood and External Link:placental tissue. These studies reinforce the hypothesis that microplastics could spread throughout the human body and accumulate in organs. However, it is also known that the body, and the brain in particular, has efficient barriers and protective mechanisms against the uptake and accumulation of undesirable substances. In addition, the intake of particles into the body is highly dependent on the route of exposure and particle properties, such as particle size.

A major uncertainty factor here is the measurement analysis, which is still in its infancy in terms of detecting plastic particles in complex matrices. For this reason, every newly published study is currently being critically discussed among experts. Some experts have positively assessed the significance of the work, while others have criticised the analytical approach. Comments known to the BfRshort forGerman Federal Institute for Risk Assessment and published include, for example: 

• Kozlov, M. Your brain is full of microplastics: are they harming you? Nature 2025, 638 (8050), 311-313. DOI: 10.1038/d41586-025-00405-8, From NLM.
• McMurray, C. ‘Spoonful of plastics in your brain’ paper has duplicated images. The Transmitter 2025. DOI: 10.53053/GTDK5632.
• Galloway, T.; Henry, T. B.; Myridakis, A. Expert reaction to a study investigating the accumulation of microplastics in human organsExternal Link: https://www.sciencemediacentre.org/expert-reaction-to-a-study-investigating-the-accumulation-of-microplastics-in-human-organs/ 2025.
• Hagelskjaer, Oshort foroxygen. A spoonful of nanoplastics in our brains? - A number based entirely on false positives?External Link: https://www.microplasticsolution.com/post/a-spoonful-of-nanoplastics-in-our-brains-a-number-based-entirely-on-false-positives.
• Xu, J. L.; Wright, S.; Rauert, C.; Thomas, K. V. Are microplastics bad for your health? More rigorous science is needed. Nature 2025, 639 (8054), 300-302. DOI: 10.1038/d41586-025-00702-2 From NLM.
• Golovin, A. Fact-Checking: Bioaccumulation of microplastics in decedent human brainsExternal Link: https://medium.com/@axegggl/fact-checking-bioaccumulation-of-microplastics-in-decedent-human-brains-nature-medicine-2025-1650e349dc10.

The comments mainly criticise the measurement methods used, but also partly the implementation of the manuscript (i.e. the presentation of the data: for example, figures were incorrectly labelled or mixed up), the interpretation of the data and the publication as a whole. At the same time, the importance of robust science for this important field of research is emphasised. 

The aim of this opinion is to evaluate both the study by Nihart et al.short foret alii (lat. "and others") and the published comments from a technical perspective and in line with the latest scientific findings, to critically review the results presented and to place them in the known scientific context. It also aims to highlight the need for further research and establishment of standards based on these data. In addition, suggestions are made for the design of future comparable studies, which should reduce future uncertainties and thus enable a plausible, clear and scientifically comprehensible interpretation of the data.

Analysis and particle quantification

The focus of criticism is on analytical detection technology. Several analytical methods are used in the study to determine the concentrations of polymers in the tissues examined and their particulate form: pyrolysis gas chromatography-mass spectrometry (Py-GC/MS), attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR), electron microscopy (as a combination of scanning electron microscopy and energy-dispersive X-ray spectroscopy (SEM-EDS) and as transmission electron microscopy (TEM)) and polarised light microscopy. The detection methods used are modern, well-established and fundamentally suitable for this purpose, but like any technology, they require establishment work when applied to new questions. This applies both to the optimisation of sample preparation and to the development of a measurement methodology adapted to the respective sample matrix and target substances, which enables the target substances in the respective samples to be determined sensitively, accurately and precisely. In this study, calibrations and quality control steps were carried out, but in many cases there is a lack of suitable process controls to adequately assess background contamination or method-related contamination.

The amount of different polymer types, such as polyethylene (PE), polypropylene (PP), polyamide 6 and 66 (PA-6 and PA-66, also known as nylon-6 and nylon-66), polyvinyl chloride (PVC) and polycarbonate (PC), in the tissue samples was determined using Py-GC/MS. To this end, tissue samples were first broken down by basic digestion with aqueous potassium hydroxide solution (KOH) at 40 °Cshort fordegrees Celsius for several days. The residue insoluble after this digestion was separated from the supernatant by ultracentrifugation and examined for its polymer content using Py-GC/MS. It is particularly important to note that all tissues removed during the autopsy came into contact with plastic. For example, the removed brains were temporarily stored in plastic buckets containing formalin solution. The ultracentrifuge tubes used to separate insoluble components were made of PC or high-density PE (HDPE). In addition, all organs were cut with scalpels on plastic boards made of PE for storage until examination and then stored in plastic containers filled with formalin. The extent to which components of the cutting board lead to contamination of the samples during the cutting process and how relevant such a possible entry into the samples would be was not investigated. However, PE was identified as the main component of the plastic fraction in all tissues examined. Furthermore, the study does not mention whether the fixation and digestion solutions used (formalin, aqueous KOH) were taken from commercially available plastic containers and could therefore represent a source of contamination. Although an examination of formalin and KOH solutions for plastic components did not reveal any evidence of this, the sample preparation and data obtained from these controls are only described in outline and are therefore of little significance. Extensive process controls would have been appropriate here to underpin the significance of the study. 

Another uncertainty arises from the quantification of individual polymers using Py-GC/MS. Here, several pyrolysis fragments are normally used for each polymer. These fragments break down again in the mass spectrometer into several fragment ions, the frequency (intensity) of which can be used to determine the concentration of the polymer. In order to rule out the possibility that signals of other, non-polymeric compounds (in the case of PE, these are, for example, the alkyl chains of fatty acid residues) overlap with these signals, they should occur in certain ratios. For example, Rauert et al.short foret alii (lat. "and others") report that signal ratios of the C10-alkenes derived from PE toC12and C14 -alkenes and C21 -alkadienes indicate the presence of such interference (Environ. Sci. Technol. 2025, 59, 1984 − 1994;External Link: https://doi.org/10.1021/acs.est.4c12599). Although pyrolysis fragments and ions are named in the publication for the identification and quantification of the respective polymers, it is not clear which specific ions were used for the important quantification, nor what the ratio of the signals to each other is. Thus, interference from non-polymeric components and a significant overestimation of the polymer content cannot be ruled out.

Extraction of the lipids remaining in the tissue samples after basic digestion with KOH could enable further depletion of interfering fat components, thereby minimising false-positive findings of PE. A comparison with the pyrolysis products of different fatty acid standards could also facilitate the assignment of the signals detected by Py-GC/MS. Complementary to Py-GC/MS, thermogravimetric analysis (TGA) and dynamic differential calorimetry (DSC) would also be useful analytical methods for distinguishing between PE and fatty acid residues. However, considering the enormous difference in plastic concentration between normal and dementia-affected brains, the many times higher measured concentration of PE in dementia-affected brains cannot be attributed to an equally higher concentration of fats. Ultimately, it is unclear to what extent misinterpretations have occurred in the assignment of spectroscopic signals and what influence this has on the reported polymer concentrations.

Another point of criticism relates to the lack of measured FTIR spectra, which would be necessary for the detection of the identity of the signals. In addition, TEM was used for the detection of different morphologies of the particles in the tissue samples (see Figs. S9 and S16). If it is postulated that the vast majority of the polymer particles found consist of PE, a more similar morphology would be expected. 

The publication does not comment on possible sources of plastic and routes of absorption through which the microplastic particles may have entered the brain. Furthermore, no distinction is made as to whether the particles actually crossed the blood-brain barrier. An accumulation in the blood vessels is visible in Fig. 2f. However, the images taken with polarisation microscopy do not allow a clear conclusion as to whether the detected particles have crossed the blood-brain barrier or are located exclusively within the blood vessels of the brain (see Fig. 2e). 

These limitations underscore that the finding – significant amounts of microplastics in the brain – warrants further investigation, which should include verification and reproduction of the result. In doing so, potential process-related contamination of the samples should be avoided at all costs by refraining from using plastic as much as possible during sample processing, and stringent process controls should be implemented. In addition, alternative analysis techniques to confirm polymer identity and record particle size distribution, as well as larger sample sets examined in independent laboratories, could significantly improve the scientific significance of these important data.

Interpretation of results and correlation with health parameters (toxicology)

Although the study focuses primarily on exposure to and detection techniques for microplastics and nanoplastics (MNP), it also expresses concern about the impact on health, with a rationale for the ever-increasing level of exposure to MNP in the environment. The authors describe that the actual extent to which MNP cause impairment in human health is still unclear, but refer to studies linking inflammation and cardiovascular disease to MNP. Reference is also made to the study by Marfella et al.short foret alii (lat. "and others") (N. Engl. J. Med. 2024, 390, 900−910;External Link: https://doi.org/10.1056/NEJMoa2309822) on microplastics in samples of human atheroma plaques. The BfRshort forGerman Federal Institute for Risk Assessment has examined this study in detail (External Link:https://www.bfr.bund.de/mitteilung/erhoehen-mikroplastikpartikel-das-risiko-fuer-einen-schlaganfall/). In fact, there is currently no evidence of a causal link between current MNP exposure and health impairments. The authors mention toxicological effects from in vitro and animal studies, but at the same time make it clear that these data were often generated with extremely high doses of MNP that are not realistically achievable in humans. However, they argue that the ever-increasing environmental pollution is a reason to take this into account nonetheless. In addition, the actual intake into the organism as well as the bioavailability and distribution within the body are not yet fully understood.

Specifically, the study compares the detected MNP concentrations in organ samples from different cohorts of varying origins. A cohort of twelve brain samples from the University of New Mexico was also examined, which came specifically from donors who had been diagnosed with various species of dementia. These included six cases diagnosed with Alzheimer's disease, three cases of vascular dementia and three cases of other forms of dementia that were not further defined, which were obtained between 2019 and 2024. The patients had an average age of 77.1 (+/- 8.7) years and were thus significantly older than most of the other cohorts. According to the authors, only the North Carolina Duke Kathleen Price Bryan Brain Bank group (13 samples) was older, with an average age of 84, although the samples were collected much earlier, between 1997 and 2013. The samples from the group with dementia had many times higher measured MNP concentrations (Fig. 1d, Table S1). With an average of 27,215 µgshort formicrogram/g, the samples contained far more measured total plastic than the other groups (NM OMI 2016: 3,420; NM OMI 2024: 4763; East Coast MD: 1404; East Coast MA: 994.8). The East Coast NC group, which was even older, also had lower values of 1259 µgshort formicrogram/g brain mass. There is no information about possible dementia in this group. The higher values are explained by structural changes such as brain tissue atrophy, weakened blood-brain barriers, and poorer clearance of metabolites from brain tissue, which are characteristics of dementia. However, these parameters were not recorded and analysed separately in the study. The authors deliberately do not establish a causal link to the causes of dementia. It should also be mentioned that the samples from the dementia cohort were taken at different locations and that the sampling took place under conditions that could not be controlled. In addition, there were no samples from other organs in the dementia cohort, so that only detection in the brain can be compared here. The toxicological part is well described in terms of methodology, and the results are presented in a comprehensible and clear manner. Gaps in knowledge are adequately addressed and no causal relationships are established that are not supported by the data. The conclusion only describes the correlation between plastic quantities and, due to increasing levels of exposure in the environment, postulates a greater need for research, also with regard to health effects, in particular neurological diseases.

Further study evaluation

The publication by Nihart et al.short foret alii (lat. "and others") deals with the quantitative detection of a previously theoretically postulated exposure and thus provides a crucial basis for environmental epidemiology. The study finds, if correctly conducted and interpreted, that microplastics (MP) are present in human brain tissue, which may be an indication that the particles can also cross the blood-brain barrier. The study's finding that MP accumulates in the human brain confirms the existence of exposure. The study itself is a cross-sectional analysis (observational study) of deceased individuals (primarily from 2016 and 2024). The sample size is small and the selection of participating hospitals was not representative. The level of overall exposure in the population and possible sources of exposure were not investigated. The design and potential for bias were not assessed in detail. 

The study identifies the brain as a significant site of accumulation of MNPs in the human body. MNPs, particularly polyethylene (PE) but also other plastics, were found in post-mortem samples of the brain, liver and kidneys. The significant increase in MNP concentrations between 2016 and 2024 (over a period of only eight years) could indicate increasing environmental pollution and thus increasing exposure of the population, but there may also be other reasons. The internal dose to humans is therefore increasing. Detailed statistical analyses in extended tables of supplementary information show that MNP concentrations in the brain were not significantly influenced by age, gender or other factors. This reinforces the conclusion that the time of death (2016 vs. 2024) and thus the potentially increased environmental exposure is the decisive factor for the increase. The most important epidemiological observation of the study is the possible link to dementia – a correlation between higher MNP concentrations and the diagnosis of dementia was found in a small sample size. However, due to its cross-sectional and retrospective design, the study cannot prove causality (i.e. it cannot show that there is a causal link between plastic particles and dementia). It is equally plausible that a blood-brain barrier damaged by dementia allows the accumulation of MNP particles or that clearance mechanisms suffer from impairment, which could lead to secondary accumulation.

The study provides important basic information on exposure, which now needs to be supplemented by validated analytical and mechanistic studies in order to determine the actual exposure and the resulting risk to human health. In the BfRshort forGerman Federal Institute for Risk Assessment's view, further epidemiological data is needed in this field of science.

Plausibility of the results and open questions

In summary, it can be acknowledged that the study by Nihart et al.short foret alii (lat. "and others") is an important key study that was planned and conducted at a comparatively good and contemporary scientific level, even though some weaknesses were identified. Important questions remain regarding contamination-free sample preparation, blank values and process controls, the validation of measurement methods and the identities of the measurement signals. The most pressing scientific and societal questions are whether microplastics actually enter the organs examined, how this happens, and whether this can be linked to health parameters. On the one hand, the results indicate that a certain concentration of microplastics and nanoplastics is present in brains. On the other hand, the mass data on detected microplastics must be considered unrealistically high due to the reasonable assumption of false positive assignments and preparation artefacts. Particularly critical sub-steps here are sample processing, analytical measurement methods and the evaluation of the data obtained. It is not certain whether the Py-GC/MS signals can be attributed exclusively to polymers or whether there have been far-reaching measurement and attribution artefacts that would lead to an overestimation of actual exposure, especially for PE. For the reasons mentioned above, it is difficult to assess the plausibility of the results. Further verification of the measurement results and follow-up studies are necessary.