Are Microplastics Making Us Sick? What the Human Evidence Actually Shows

The question

Do microplastics (roughly 1 micrometer to 5 millimeters) and nanoplastics (below 1 micrometer) that accumulate inside the human body cause disease — and how strong is the evidence that they do, compared with what detection, toxicology, and epidemiology actually establish?

Two claims are routinely blended in coverage of this topic, and they have different answers:

1. Presence: plastic particles enter the body through food, water, air, and medical procedures, and reach blood and solid tissues. This is an empirical detection question.
2. Harm: those particles, at the doses and sizes humans actually carry, raise the risk of clinical disease. This is a causal question requiring designs that separate exposure from confounding, contamination, and reverse causation.

This brief keeps the two apart, treats cardiovascular disease as the leading test case — it is where the strongest human clinical signal exists — and asks what each class of design actually identifies.

Why it matters

Global plastic production exceeds 400 million tonnes a year and is projected to keep rising; particles shed from plastics, packaging, synthetic textiles, tires, and open burning are now detected in air, food, drinking water, and essentially every human tissue examined. The question sits directly in the policy arena: negotiations on a global plastics treaty have run through the mid-2020s, with health arguments carrying much of the political weight. Regulators, clinicians, and households are each implicitly answering the causal question — tightening rules, buying filters, or dismissing the concern — usually with less evidence than they assume.

The stakes are asymmetric in a specific way. If particles themselves are the hazard, the interventions are filtration, exposure control, and material redesign. If the hazard is mostly the chemicals that plastics carry or leach — phthalates, bisphenols, flame retardants — the interventions are chemical regulation. If particles are merely vectors that deliver those chemicals into tissues, both strategies matter. These are different mechanisms, and the current evidence supports them to very different degrees; conflating them inflates whichever number sounds scarier.

What each design actually identifies

Tissue-detection studies establish presence, not harm. The first quantification of plastic polymers in human blood (22 healthy donors, pyrolysis–gas chromatography–mass spectrometry, mean of the quantifiable sum about 1.6 µg/mL) showed that particles are bioavailable for uptake into the bloodstream. Electron-microscopy-guided infrared spectroscopy found microplastic fragments in 11 of 13 human lung samples at levels significantly above procedural and laboratory blanks — one of the very few studies in the field with contamination controls strong enough to trust the positive. Raman microspectroscopy detected fragments in 4 of 6 placentas. None of these designs distinguishes a person whose particles matter from a person whose particles are inert cargo; they establish internal exposure and nothing about its consequences.

The flagship clinical study is a prognostic cohort, not an exposure experiment. The most cited human result followed 304 patients undergoing carotid endarterectomy for asymptomatic stenosis; 257 completed roughly 34 months of follow-up. Polyethylene was detected in 58.4% of plaques and polyvinyl chloride in 12.1%. Patients with detectable particles in their excised plaque had a hazard ratio of 4.53 (95% CI 2.00–10.27) for a composite of myocardial infarction, stroke, or death. The design identifies a prognostic association within people who already have severe carotid disease severe enough to warrant surgery. It does not identify what would happen if exposure were reduced, in patients or in the population, and its confidence interval spans anything from a doubling to a tenfold effect. A second, similar cohort — 110 myocardial-infarction patients followed about 31 months — reported that polyvinyl chloride in the coronary circulation independently predicted major adverse cardiac events, with thrombus burden correlating with inflammatory markers. Two studies in the same design family do not constitute replication under independent measurement conditions; both used the same analytical platform.

The systematic reviews say what the primary papers cannot. A 2026 systematic review of every in-vivo human study — 25 papers from 5,522 records — found the entire cardiovascular MNP-outcome literature to be 5 studies totaling 454 participants, and graded the human evidence at moderate-to-high risk of confounding and exposure-measurement bias, with limited consistency across detection modalities. Cardiology's own review journal concluded in 2026 that causal associations remain unproven in humans, citing inconsistent detection methods, sparse epidemiology, and the poor fit between experimental models (usually pristine polystyrene spheres at high doses) and human internal dosimetry. A December 2025 planetary-health review reached the same verdict: direct evidence linking MNP exposure to human disease remains scarce.

The measurement method is itself contested. A 2025 laboratory study tested pyrolysis–gas chromatography–mass spectrometry — the platform behind the blood-detection paper and both cardiovascular cohorts — against the task of quantifying polyethylene and polyvinyl chloride in human blood. It found matrix interferences producing false-positive polyethylene signals, realistic detection limits up to 20 times worse than nominal ones, and concluded the method is currently unsuitable for these polymers in biological matrices; in a pilot population sample, no polymer except interference-laden polyethylene cleared detection. A 2026 follow-up by the same group proposed confidence-communication standards for MNP identification in health studies. This matters directly: the exposure classifier in the flagship studies — plaque “positive” versus “negative” — depends on exactly the quantification properties now in doubt, in both directions. Contamination or interference can create false positives; detection limits can misclassify low-burden patients as negative.

Toxicology supplies plausibility, not magnitude. Laboratory studies converge on credible mechanisms: oxidative stress, NF-κB-mediated inflammation, endothelial senescence and dysfunction, macrophage foam-cell formation, pro-thrombotic effects on fibrinogen, and mitochondrial damage in cardiomyocytes. That is a real biological plausibility case. But the doses are often orders of magnitude above measured human internal burdens, the polymers are usually not the polyethylene and PVC that human plaques actually contain, and the particles are typically pristine spheres rather than weathered, irregularly shaped environmental fragments. The World Health Organization's assessments (2019 for drinking water, extended in 2022 to diet and inhalation) found no reliable evidence of concern for microplastic particles at current exposures, judged particles above 150 µm unlikely to be absorbed at all, and flagged the evidence on nanoparticle uptake as extremely limited and derived from unrealistically high animal doses.

The best current answer

Presence in human tissues is established; causation of disease is not. The honest summary of the field as of late 2026:

• Exposure: humans absorb and distribute micro- and nanoplastics to blood, lung, placenta, and vascular tissue at low but measurable levels. Confidence ~90%, based on converging detection studies, including some with strong contamination controls.
• Hazard signal: a small but internally consistent set of clinical studies associates higher tissue MNP burdens with inflammatory markers and adverse cardiovascular events. Confidence ~65% that some real pro-inflammatory, pro-thrombotic effect of particles exists in humans at realistic doses — the toxicology is coherent and two cohorts point the same way.
• Causal attribution of the observed clinical associations: weak. Confidence ~35% that the carotid-plaque hazard ratio is primarily causal rather than substantially driven by confounding (systemic inflammation, plaque vulnerability, cumulative vascular damage trapping debris), reverse causation, or exposure misclassification.
• Population-level importance: unsupported. Confidence ~15% that MNPs are currently a material independent driver of cardiovascular disease in the way that smoking, blood pressure, or LDL cholesterol are. No study has estimated a population-attributable fraction that survives the bias structure of the existing designs.
• Clinical actionability: ~5% that current evidence justifies routine testing, risk-score changes, or MNP-specific treatment. Major reviews explicitly say otherwise.

The gap between the first and last numbers is where headlines live.

A separate and easily conflated hazard: plastic chemicals. The epidemiology of phthalates — chemical additives, not particles — is more mature and carries larger claims: a nationally representative United States cohort linked urinary DEHP metabolites to all-cause and cardiovascular mortality (third-tertile hazard ratios around 1.4–1.5), and extrapolations from those associations have produced estimates of roughly 90,000–107,000 attributable deaths a year among Americans aged 55–64, later extended globally to about 356,000 cardiovascular deaths in 2018. These are extrapolations on top of observational hazard ratios, and they have already needed public correction: the Minderoo–Monaco Commission's headline figure was found to sum two years of deaths into an annual estimate — a factor-of-two error acknowledged in a 2023 correction. The particle literature and the chemical literature support different interventions and different regulatory claims; strength of one is not evidence for the other.

Mechanisms: three hazards wearing one label

1. The particle itself. Translocation across gut or lung epithelium, distribution in blood, deposition in vascular tissue, then inflammatory and thrombotic signaling. This is the mechanism the cardiovascular cohorts are implicitly testing, and it is the one the human data are (weakly) about.
2. Leached and carried chemicals. Phthalates, bisphenols, heavy metals, and adsorbed organic pollutants hitchhiking on particles or entering through the same products. This mechanism's epidemiology exists (urinary metabolite studies) but says nothing about particle counts.
3. Biofilms and pathogens. Microbial communities on particle surfaces. Institutional assessments judge this secondary to ordinary microbial risk in water and food.

Different identification strategies are required for each: measuring internal particle dose for mechanism 1, measuring metabolite exposure and intervening on chemicals for mechanism 2. Adding coefficients across the two literatures — as popular burden estimates sometimes do — double-counts the same plastics system and silently assumes independence between correlated exposures.

Counterevidence and unresolved weaknesses

• Contamination. A letter to the New England Journal of Medicine argued that the surgical environment and equipment are rich in polyethylene and polyvinyl chloride, that the study described no pre-analytical anti-contamination procedures beyond glass tubes, and that no operating-room blanks were reported — meaning some or all plaque detections could be exogenous. The original authors corroborated pyrolysis results with electron microscopy and stable-isotope analysis, which answers part but not all of this: microscopy shows particles inside plaque macrophages, which is stronger than surface presence, but the correspondence's core demand — field blanks from the operating room — is standard practice the study did not document.
• Analytical validity. The 2025 pyrolysis-method audit is the sharpest single piece of counterevidence in the field. Until exposure classification is reproduced with spectroscopic, single-particle, or otherwise validated methods with honest detection limits, hazard ratios built on it are provisional regardless of their p-values.
• Confounding and reverse causation. Patients with vulnerable, inflamed plaques may accumulate or retain more particles; nothing in the existing cohorts excludes that the plaque biology drives the detection as much as the reverse.
• Dose relevance. No human study has established a dose–response curve between external exposure, internal burden, and clinical risk — a prerequisite for causal claims of the kind lead, smoking, or cholesterol epidemiology eventually met.
• The adjacent literature's correction history. The factor-of-two error in a widely repeated phthalate-mortality estimate, propagated through commission reports before correction, is a cautionary precedent for how quickly model-dependent numbers outrun their data in this field.

What the evidence cannot yet tell us

• Whether reducing human MNP exposure reduces any clinical outcome. No human intervention of any kind has tested this; even short-term biomarker trials are essentially absent.
• Which particles matter. Environmental fragment populations are polydisperse, weathered, and polymer-mixed; laboratory studies overwhelmingly test polystyrene spheres. The human cardiovascular signal implicates polyethylene and PVC specifically; nobody has shown those polymers are the bioactive ones.
• Whether nanoplastics — the fraction small enough to cross barriers efficiently — dominate the hazard, and whether any current method measures them reliably in tissue.
• Whether detected burdens reflect chronic accumulation or recent exposure, which determines whether tissue measurement can ever be a usable exposure metric.
• The population-level share of disease attributable to MNPs, if any. It has not been estimated with a defensible design.

The next material advance is not another detection study or another narrative review; the field produces both in volume. It is either (a) an independent replication of the carotid-plaque association using validated spectroscopic quantification with operating-room and laboratory blanks, registered in advance, or (b) a human exposure-reduction study — dietary, occupational, or medical — with internal-dose verification and validated inflammatory or vascular endpoints. Both are feasible with current technology; neither exists.

Calibration, reversal conditions, and the practical verdict

Current calibration. Exposure ~90% established; a real vascular hazard signal ~65%; the flagship clinical associations mainly causal ~35%; MNPs a major population driver of cardiovascular disease ~15%; clinically actionable today ~5%. These are judgments about evidence states, not about the physics: nothing here says plastics are good for you, and the toxicology would be surprising if the true causal effect were exactly zero.

What would raise confidence. A preregistered, multicenter replication of the plaque–events association using validated single-particle or spectroscopic methods with documented blanks, finding a dose-response and surviving negative-control outcomes, would move the causal estimate toward ~70%. A randomized exposure-reduction trial showing reduced internal burden plus improved vascular biomarkers — or a natural experiment in which a policy-driven exposure drop tracked improved endpoints — would raise it further.

What would lower confidence. Independent replication finding plaque signals indistinguishable from blanks, or hazard ratios collapsing toward the null once detection limits and contamination are handled as the 2025 audit implies they must, would move the causal estimate toward ~10% and reframe the flagship result as a measurement artifact. Consistent failure of spectroscopic methods to confirm pyrolysis-based detections in the same samples would do the same.

The practical verdict. For individuals: the existing evidence does not support panic, clinical testing, or supplements and cleanses marketed against microplastics; it does support the ordinary hygiene of reducing unnecessary plastic contact where convenient — not because that is proven to prevent disease, but because the hazard signal is coherent and the cost of most reductions is low. For institutions and treaty negotiators: the particle evidence justifies funding standardized measurement, preregistered cohorts, and exposure-reduction trials now, and justifies neither dismissal nor large burden-of-disease numbers attributed specifically to particles. The chemical-additive case is the stronger public-health argument in the plastics portfolio today — and even it rests on observational hazard ratios whose population extrapolations have already been halved once in public. A field this young deserves exactly the amount of confidence its designs can carry: presence, yes; harm, plausible; magnitude, unknown.

References

1. Marfella, R., Prattichizzo, F., Sardu, C., et al. (2024). Microplastics and Nanoplastics in Atheromas and Cardiovascular Events. New England Journal of Medicine, 390, 900–910. doi:10.1056/NEJMoa2309822
2. Glazier, H. A. (2024). Microplastics and Nanoplastics in Atheromas. New England Journal of Medicine, 390, 1727–1728. doi:10.1056/NEJMc2404154
3. Leslie, H. A., van Velzen, M. J. M., Brandsma, S. H., Vethaek, A. D., Garcia-Vallejo, J. J., & Lamoree, M. H. (2022). Discovery and quantification of plastic particle pollution in human blood. Environment International, 163, 107199. doi:10.1016/j.envint.2022.107199
4. Jenner, L. C., Rotchell, J. M., Bennett, R. T., Cowen, M., Tentzeris, V., & Sadofsky, L. R. (2022). Detection of microplastics in human lung tissue using μFTIR spectroscopy. Science of the Total Environment, 831, 154907. doi:10.1016/j.scitotenv.2022.154907
5. Ragusa, A., Svelato, A., Santacroce, C., et al. (2021). Plasticenta: First evidence of microplastics in human placenta. Environment International, 146, 106274. doi:10.1016/j.envint.2020.106274
6. Tran, H. A. A., Ng, J. J., Tang, D. H., Yong, C. T., & Ho, A. F. W. (2026). Health impacts of micro- and nanoplastics in humans: systematic review of in vivo evidence. Environmental Health, 25(1), 26. doi:10.1186/s12940-026-01282-y
7. Zhang, Y., Gao, Q., Gao, Q., et al. (2025). Microplastics and nanoplastics increase major adverse cardiac events in patients with myocardial infarction. Journal of Hazardous Materials, 489, 137624. doi:10.1016/j.jhazmat.2025.137624
8. Rauert, C., Charlton, N., Bagley, A., Dunlop, S. A., Symeonides, C., & Thomas, K. V. (2025). Assessing the Efficacy of Pyrolysis–Gas Chromatography–Mass Spectrometry for Nanoplastic and Microplastic Analysis in Human Blood. Environmental Science & Technology, 59, 1984–1994. doi:10.1021/acs.est.4c12599
9. Thomas, K. V., Belz, S., Booth, A. M., et al. (2026). Communicating Confidence in the Reliability of Micro- and Nanoplastic Identification in Human Health Studies. Environment & Health. doi:10.1021/envhealth.5c00671
10. Aimo, A., Panichella, G., Tommasi, E., Revuelta-López, E., Berastegui, E., & Bayes-Genis, A. (2026). The effects of microplastics and nanoplastics on cardiovascular disease: mechanisms and perspectives. Nature Reviews Cardiology. doi:10.1038/s41569-026-01277-9
11. AzizUllah, & Zaman, W. (2026). Microplastics, Nanoplastics, and the Cardiovascular Exposome: Emerging Evidence for Vascular Deposition, Thrombotic Pathways, and Cardiovascular Risk. Comprehensive Physiology, 16(5). doi:10.1002/cph4.70259
12. Ali, N., Katsouli, J., Auyang, E., & Bernardino de la Serna, J. (2025). Microplastic and nanoplastic pollution and associated potential disease risks. The Lancet Planetary Health, 9(12), 101390. doi:10.1016/j.lanplh.2025.101390
13. Lamoree, M. H., van Boxel, J., Nardella, F., et al. (2025). Health impacts of microplastic and nanoplastic exposure. Nature Medicine, 31, 2873–2887. doi:10.1038/s41591-025-03902-5
14. World Health Organization. (2019). Microplastics in drinking-water. WHO publications
15. World Health Organization. (2022). Dietary and inhalation exposure to nano- and microplastic particles and potential implications for human health. WHO publications
16. Cox, K. D., Covernton, G. A., Davies, H. L., Dower, J. F., Juanes, F., & Dudas, S. E. (2019). Human Consumption of Microplastics. Environmental Science & Technology, 53, 7068–7074. doi:10.1021/acs.est.9b01517
17. Trasande, L., Liu, B., & Bao, W. (2022). Phthalates and attributable mortality: A population-based longitudinal cohort study and cost analysis. Environmental Pollution, 292(Pt A), 118021. doi:10.1016/j.envpol.2021.118021
18. Landrigan, P. J., Raps, H., Cropper, M., et al. (2023). The Minderoo-Monaco Commission on Plastics and Human Health. Annals of Global Health, 89(1), 23. doi:10.5334/aogh.4056
19. Correction: The Minderoo-Monaco Commission on Plastics and Human Health. (2023). Annals of Global Health. doi:10.5334/aogh.4331
20. Hyman, S., Acevedo, J., Giannarelli, C., & Trasande, L. (2025). Phthalate exposure from plastics and cardiovascular disease: global estimates of attributable mortality and years life lost. eBioMedicine, 117, 105730. doi:10.1016/j.ebiom.2025.105730
21. Landrigan, P. J., et al. (2025). The Lancet Countdown on health and plastics. The Lancet. doi:10.1016/S0140-6736(25)01447-3
22. Mahalingaiah, S., Nadeau, K. C., & Christiani, D. C. (2025). Microplastics and Human Health. JAMA, 1941–1942. doi:10.1001/jama.2025.14718