My Healthcare News & Research — August 11, 2026 · Microplastics Were Found in 77% of Lungs From 2024 Autopsies, Against 19% From 1991

On Tuesday, August 11, 2026, The Lancet Regional Health – Europe published online a study that looked for plastic particles in human lungs taken at autopsy 33 years apart. Italian researchers went back to stored lung tissue from 42 people whose autopsies were done in 1991 and compared it with tissue from 57 people autopsied in 2024, all at the same forensic medicine institute of the University Hospital of Campania “Luigi Vanvitelli” and, the authors write, “using identical procedures.” Using an infrared imaging method that identifies each particle one by one, they found microplastics in 8 of 42 lungs (19%) from 1991 and in 44 of 57 lungs (77%) from 2024. The particles found in 2024 were smaller, and they came from a wider mix of plastics.

This is a comparative post-mortem study of 99 people: observational, cross-sectional and retrospective, with one hospital, two calendar years and stored tissue. It can show that the two groups of lungs differed. It cannot show why they differed, whether the plastic did any harm, or whether the same pattern holds elsewhere. The authors call their own observations “hypothesis-generating.” They also report that lungs containing microplastics showed more signs of scarring and inflammation under the microscope, and they state that this link “is correlational and does not establish causality.”

This report goes back to the full open-access paper (PubMed PMID 42620939) and gives its numbers as the paper states them. That includes the detail that the headline particle counts describe only the lungs in which plastic was found, plus the limits the authors list themselves. It also places the study among earlier lung, brain, placenta and artery-plaque findings, and the open argument over how plastics in human tissue are measured.

Table of Contents

  1. At a Glance
  2. What Was Done
  3. What Was Found, Number by Number
  4. How Plastic Reaches the Lungs
  5. How It Fits With Earlier Research
  6. The Measurement Debate
  7. What It Cannot Show
  8. What the Authors and Later Coverage Said
  9. What Remains Unknown
  10. Sources and Primary Documents
  11. Key Research Papers
  12. Connections

At a Glance

Back to Table of Contents

What Was Done

The two groups. The first group was made up of autopsies done from January to December 1991; the second, of autopsies done from January to September 2024. Both came from the Institute of Legal/Forensic Medicine of the University Hospital of Campania “Luigi Vanvitelli.” To be included, the autopsy had to have been done within 48 hours of death, with no signs of decomposition. The person could not have died of a lung cause, and the lung tissue had to be well preserved. Age and sex had to be on record. Work histories came from archived records or, where these were missing, from phone interviews with surviving relatives. Each person was then classed as having no, low or high risk of exposure to plastic particles at work.

How alike the groups were. The mean age was 61 ± 11 years in the 1991 group and 64.9 ± 9.4 years in the 2024 group. Men made up 69.1% and 59.6% respectively. Smokers numbered 31 (73.8%) and 35 (61.4%). The paper reports no differences between the two groups in age, sex, smoking, occupational exposure risk, cause of death or season of death. Most people in both groups were Italian.

The tissue. Only paraffin-embedded tissue from the same part of the lung, the posterior basal segment of the lower lobes, was used, to reduce variation between regions of the lung. For each person, about 1.0 gram of tissue was freed from its paraffin and dissolved chemically, with potassium hydroxide and then hydrogen peroxide. The liquid was passed through a gold-coated membrane filter with 0.8 µm pores, and whatever stayed on the filter was scanned.

The main measurement. The laser-direct infrared (LDIR) instrument finds each particle on the filter and records its infrared spectrum. It then matches that spectrum against a library of plastics. A particle counted as a microplastic only if the match scored at least 0.80 on the instrument’s hit-quality index. The validated size range was 10 to 500 µm. The primary end point was the number of microplastic particles per gram of lung tissue. Secondary end points were whether any plastic was present, particle size and the type of plastic.

Cross-checks. Ten samples from each year were picked at random and also tested by Py-GC/MS, which heats the sample until plastics break down into chemical fragments typical of each polymer. That method measures the mass of 11 different plastics. The same 20 samples were examined by scanning electron microscopy with X-ray element mapping. Finally, all 99 samples were stained for collagen I and for CD68. A pathologist who did not know the plastic results scored the stains, and image-analysis software measured them as well.

Contamination controls. Plastic is everywhere in a laboratory, so the authors describe steps to keep stray particles out of the results. All work used non-plastic equipment. Every batch included two blanks: a procedural blank of paraffin-only sections with no tissue, run through the whole process, and a laboratory blank of pre-cleaned filters left open on the bench. The paper states that both kinds of blank “were devoid of microplastic particles above the spectral confidence threshold.” For the Py-GC/MS work, samples were handled under a laminar-flow hood, glassware was rinsed with solvent, and instrument, reagent and room-air blanks were run.

Statistics. Because most 1991 lungs had no detectable plastic at all, the team added a “two-part hurdle model.” It looks separately at the chance that any plastic is found and at how much is found when it is. The uncertainty of the combined estimate was measured by resampling the data 10,000 times.

The study was funded by the Italian Ministry of Health, through its Ricerca Corrente programme to a research hospital, and by a university research fund. The paper states that the funders had no role in the study. One author reported study and travel support from a philanthropic foundation and consulting fees from a research centre.

Back to Table of Contents

What Was Found, Number by Number

How many lungs held plastic

How much plastic, in the lungs that had some

The paper’s abstract gives the median burden as rising “from 12.9 particles (11.3–13.7) to 19.9 (15.5–28.2)” per gram of tissue. The results section adds a qualifier that the abstract leaves out: these medians are “among those with evidence of MPs.” In other words, they describe only the 8 positive lungs from 1991 and the 44 positive lungs from 2024, not all 99. The numbers in brackets are the middle half of the values (first to third quartile).

The hurdle model

The authors add two cautions about this model in their own discussion: it “showed a low deviance,” and “much of the apparent difference between groups observed in this analysis was driven by prevalence rather than by differences among positive samples.” Put simply, the biggest change was in how many lungs held any plastic at all, more than in how much each positive lung held.

Particle size

Which plastics

For each person with plastic, the team worked out what share of their particles came from each polymer. They then averaged those shares across the group (mean ± standard deviation):

Polyethylene is the plastic of films and bags. PET is used in drink bottles and polyester fibres, PVC in pipes, flooring and cables, often softened with phthalates, and polystyrene in foams and rigid containers. The paper itself does not trace the particles to any source. As for shape, “Fragments were the predominant morphotype in both cohorts,” and fibres, pellets and spheres made up smaller shares.

The cross-check: Py-GC/MS on 20 samples

By Py-GC/MS, plastic was found in 2 of 10 samples from 1991 and 8 of 10 from 2024 (p < 0.05). The paper says this method “corroborated” the higher prevalence and the more varied mix of polymers in 2024. It could not do more than that. With only two positive 1991 samples, the authors write, it was not possible to compare the average total mass of plastic between the years. They therefore describe the Py-GC/MS findings as “supportive rather than confirmatory.” They also raise a point that cuts the other way: “Given that particles in 2024 were smaller, it may be possible that overall mass according to Py-GC/MS is not increased when considering only positive samples.” More particles do not necessarily mean more plastic by weight.

Electron microscopy

Under the scanning electron microscope with X-ray element mapping, some particles showed less carbon and more sulfur, and such particles were “more pervasive” in the 2024 samples. The authors are explicit that this is not evidence of plastic: “the detection of sulfur-containing particles cannot be considered specific evidence of MPs,” because sulfur is a common part of air pollution from many sources. They class these as “ancillary findings rather than corroborative evidence.”

Scarring and inflammation markers

Across both years, 52 samples had microplastics and 47 did not, and the two sets of people did not differ in their recorded characteristics. The plastic-positive samples showed more collagen I, the main fibre of scar tissue, and more CD68, a marker of macrophages, the immune cells that engulf foreign material (p < 0.01 for both). This held “independently of the case series,” that is, within the 1991 and 2024 groups alike.

The authors set out why this cannot be read as plastic causing fibrosis. The relationship “is correlational and does not establish causality.” Hidden lung disease may have been present even though lung deaths were excluded. The order may also run the other way: inflammation and scarring that came first could slow the lung’s clearance of particles, so that particles build up where the tissue is already damaged. They call these results “exploratory and hypothesis-generating rather than mechanistic evidence” of harm from microplastics.

Back to Table of Contents

How Plastic Reaches the Lungs

Plastic objects shed fragments as they wear, weather and break down. Fragments smaller than 5 millimetres are called microplastics, and those smaller than 1 micrometre are called nanoplastics. The paper uses these definitions. Particles light enough to float are part of the indoor and outdoor air. The lung, the authors write, is “continuously exposed to inhaled particles” and is a potential reservoir for airborne contaminants, which is why they chose it.

What happens to a breathed-in particle depends on its size. The authors note that particles under 10 µm are “those more likely reaching the alveoli and the blood.” The alveoli are the tiny air sacs where oxygen enters the bloodstream, and ordinary air pollution is measured in the same size bands (PM10 and the fine particles of PM2.5). Their method could only see particles of 10 µm and larger, so the size range most likely to travel deepest was outside its view.

The lung is not a sealed store. The authors describe it as “a transient environment, with established trafficking and clearance mechanisms.” Mucus and tiny beating hairs (cilia) move trapped material up and out of the airways, and macrophages engulf particles in the air sacs. CD68, which the study measured, marks these macrophages. As the authors note, clearance depends on how well the lung is working, so what a lung holds at death reflects clearance as well as what was breathed in. In their account, a lung already inflamed or scarred may clear particles less well.

Concern about plastics in the lung did not start with this study. The authors point to occupational studies that report lung diseases in workers exposed to different plastics. They state that whether the everyday exposure of the general population causes disease “is now being actively explored.” A 2024 review in the European Respiratory Review reached the same open verdict: plastic particles can affect lung tissue, but “it remains unclear to which extent this occurs upon exposure to ambient levels.”

Back to Table of Contents

How It Fits With Earlier Research

Plastic in lungs: the first reports

In 2021 a team reported in the Journal of Hazardous Materials on lung tissue obtained at autopsy. They found 33 polymeric particles and 4 fibres in 13 of 20 samples. Every particle was smaller than 5.5 µm and the fibres measured 8.12 to 16.8 µm. The most frequent polymers were polyethylene and polypropylene. Every particle in that study was below the 10 µm lower limit of the 2026 study’s main method.

In 2022 another team used a different infrared technique (µFTIR), able to see particles down to 3 µm, on 13 lung tissue samples (Science of the Total Environment). They identified 39 microplastics in 11 of 13 samples, an average of 1.42 ± 1.50 per gram. After subtracting what turned up in their blank controls, the figure was 0.69 ± 0.84 per gram. Polypropylene (23%), PET (18%) and resin (15%) were the most common types. The lower lung held the most, 3.12 ± 1.30 per gram unadjusted, against 0.80 in the upper and 0.41 in the middle regions. The 2026 study took all its samples from the lower lobes.

These counts are not directly comparable with the 2026 figures. The studies used different instruments and different smallest-visible sizes, and their blank-correction methods differed as well. The 2026 authors themselves cite a 2021-collected autopsy lung study, run with a different method, whose results were “similar to those presented here for the 2024 series” in how often plastic was found and which polymers appeared.

Changes over time: brain, liver and placenta

In their search of earlier research, the authors found only two studies that tracked human tissue over time. A study in Nature Medicine (published February 2025) compared autopsies from 2016 and 2024. It reported that time of death was a significant factor, with higher concentrations in 2024 in liver and brain samples (P = 0.01) but, the 2026 paper notes, not in kidney. Concentrations were not influenced by age, sex, race/ethnicity or cause of death. Its concentration figures came from Py-GC/MS, with other methods used to confirm the presence of particles. A 2023 study from Hawai’i (Environment International) used Raman spectroscopy on discarded placentas collected in 2006, 2013 and 2021. It reported a significant increase in microplastics over the 15 years, with changes in the size and chemical make-up of the polymers. The lung study covers the longest span of the three.

Plastic in artery plaque

The same lead author and several of the same co-authors published a study in the New England Journal of Medicine in March 2024. It analysed plaque removed from the neck (carotid) arteries of people having surgery for carotid artery disease that had not caused symptoms, using Py-GC/MS. Of 257 people followed for a mean of 33.7 months, polyethylene was found in the plaque of 150 (58.4%) and PVC in 31 (12.1%). People whose plaque contained micro- and nanoplastics had a higher risk of a combined outcome of heart attack, stroke or death from any cause (hazard ratio 4.53, 95% CI 2.00 to 10.27). That study was also observational. It showed an association, not proof that the plastic caused the events (see atherosclerosis).

The bigger picture on plastic production

The 2026 authors write that plastic production “has increased almost six-fold since the early 1990s, reaching an annual rate of nearly 600 million metric tons at present.” The 2025 Lancet Countdown on health and plastics gives its own production figures: from 2 million tonnes in 1950 to 475 million tonnes in 2022, with a projection of 1,200 million tonnes by 2060, and less than 10% of plastic recycled. The site’s January 2026 report on the health cost of plastics’ life cycle covers related research.

Back to Table of Contents

The Measurement Debate

Measuring plastic inside human tissue is hard, and scientists disagree about how far some published numbers can be trusted. The 2026 paper addresses this head-on. Its introduction says that earlier results showing more plastic in recent tissue “are debated mainly due to possible methodological challenges in assessing MPs in human brain samples with pyrolysis–gas chromatography/mass spectrometry (Py-GC/MS).”

The two kinds of method. Spectroscopy methods such as LDIR, µFTIR and Raman look at individual particles. They give a count, a size, a shape and a polymer identity for each particle, but only above a minimum size. Py-GC/MS burns the whole sample and reads the chemical fragments. It gives a total mass for each polymer and can pick up much smaller pieces, but it cannot count or size particles. Some fragments can also come from other material left in the sample. The 2026 authors acknowledge this for their own Py-GC/MS work: “potential matrix interference (e.g., from incompletely removed lipids) cannot be entirely excluded.”

The published critique of the brain study. In November 2025, Nature Medicine published a “Matters Arising” letter by a group of scientists about the 2016-vs-2024 brain study. The letter argued that the study “appear[s] to face methodological challenges, such as limited contamination controls and lack of validation steps,” which bears on how reliable its concentrations are. The lung paper cites this letter. Its authors argue that adding a second, particle-based technology (LDIR) in a different organ “might mitigate the concerns” raised about Py-GC/MS in brain samples.

A proposed standard. In 2026 a group of researchers in the field published a framework in Environment & Health for “communicating confidence” in plastic identification in human studies. It describes the analytical challenge as compounded by the ubiquity of plastic particles and the resulting risk of sample contamination. The framework proposes combining “orthogonal” methods, meaning independent techniques that check one another, sets minimum data for each confidence level, and calls for harmonised reporting criteria. The lung paper cites this framework and describes its LDIR plus Py-GC/MS plus electron microscopy design as consistent with it. In that design the cross-check ran on 20 samples and, by the authors’ account, supported the prevalence and polymer-mix findings but could not compare plastic mass between the years.

Back to Table of Contents

What It Cannot Show

The authors list a long set of limits. In their words or close to them:

Two further cautions follow from the paper’s own numbers. The headline medians of 12.9 and 19.9 particles per gram describe only plastic-positive lungs. And the paper offers no direct mass comparison between the years, so it does not show whether the total weight of plastic in lungs rose.

Back to Table of Contents

What the Authors and Later Coverage Said

The authors’ own conclusion. The paper concludes that the results “suggest that MPs contamination of the human lung is more prevalent in samples collected recently compared with those obtained three decades ago.” Given the observational, retrospective design, it calls the findings “hypothesis-generating,” needing “confirmation in future studies with larger sample sizes and multiple time points.” The authors add that the observations “might suggest the need for systematic biomonitoring of MPs also in living individuals.” For future work they point to optical PTIR, an infrared technology that shows particles inside the tissue itself. Combined with staining of neighbouring sections, they write, it could firmly link particles to inflammation and scarring.

Agencies. No statement by a health agency about this study was found.

Since then (September 17, 2026). A chromatography trade publication covered the study under a headline crediting Py-GC/MS. Its main figures (19% vs 77%; medians of about 13 and about 20 particles per gram) match the paper. Some details do not. The coverage attributed the sulfur-containing particles to Py-GC/MS, while the paper found them by electron microscopy and says they are not specific evidence of plastic. The paper’s main measurement was LDIR imaging, and its Py-GC/MS results are described by the authors as supportive. No other news coverage, press release or expert-reaction page about the lung study was found for this report.

Back to Table of Contents

What Remains Unknown

Back to Table of Contents

Sources and Primary Documents

Every figure attributed to the 2026 study above was taken from the paper’s open-access full text in PubMed Central (PMC13487393), not from news coverage. The publication date (online August 11, 2026; received March 24, 2026; accepted July 22, 2026) is from the PubMed record.

  1. Marfella R, Prattichizzo F, La Grotta R, Pellegrini V, Metallo A, Graciotti L, et al. (2026). Microplastics in human lung tissue from autopsy samples collected in 1991 and 2024: a comparative post-mortem study. The Lancet Regional Health – Europe 69:101804. — doi:10.1016/j.lanepe.2026.101804 · PubMed PMID: 42620939
  2. Full text of the same paper in PubMed Central, including the methods, contamination controls, results and limitations sections. — PMC13487393
  3. A Monikh F, Materić D, Valsami-Jones E, Grossart HP, Altmann K, Holzinger R, et al. (2025). Challenges in studying microplastics in human brain. Nature Medicine 31(12):4034-4035. — doi:10.1038/s41591-025-04045-3 · PubMed PMID: 41233597

Later coverage: LCGC / Chromatography Online, September 17, 2026 (trade-press report).

Back to Table of Contents

Key Research Papers

Background papers on plastic in human lungs and other tissues, on trends over time, and on how such plastic is measured. Each entry was checked against PubMed before it was written down.

  1. Amato-Lourenço LF, Carvalho-Oliveira R, Júnior GR, Dos Santos Galvão L, Ando RA, Mauad T (2021). Presence of airborne microplastics in human lung tissue. Journal of Hazardous Materials 416:126124. — PubMed PMID: 34492918
  2. Jenner LC, Rotchell JM, Bennett RT, Cowen M, Tentzeris V, Sadofsky LR (2022). Detection of microplastics in human lung tissue using μFTIR spectroscopy. Science of the Total Environment 831:154907. — PubMed PMID: 35364151
  3. Marfella R, Prattichizzo F, Sardu C, Fulgenzi G, Graciotti L, Spadoni T, et al. (2024). Microplastics and Nanoplastics in Atheromas and Cardiovascular Events. New England Journal of Medicine 390(10):900-910. — PubMed PMID: 38446676
  4. Nihart AJ, Garcia MA, El Hayek E, Liu R, Olewine M, Kingston JD, et al. (2025). Bioaccumulation of microplastics in decedent human brains. Nature Medicine 31(4):1114-1119. — PubMed PMID: 39901044
  5. Weingrill RB, Lee MJ, Benny P, Riel J, Saiki K, Garcia J, et al. (2023). Temporal trends in microplastic accumulation in placentas from pregnancies in Hawai’i. Environment International 180:108220. — PubMed PMID: 37741006
  6. Thomas KV, Belz S, Booth AM, Clift MJD, Cross RK, Davies G, et al. (2026). Communicating Confidence in the Reliability of Micro- and Nanoplastic Identification in Human Health Studies. Environment & Health 4(5):804-811. — PubMed PMID: 42164877
  7. Vasse GF, Melgert BN (2024). Microplastic and plastic pollution: impact on respiratory disease and health. European Respiratory Review 33(172). — PubMed PMID: 39009408
  8. Landrigan PJ, Dunlop S, Treskova M, Raps H, Symeonides C, Muncke J, et al. (2025). The Lancet Countdown on health and plastics. Lancet 406(10507):1044-1062. — PubMed PMID: 40769171

PubMed Topic Searches

  1. PubMed: Microplastics in human lung tissue
  2. PubMed: Microplastics in autopsy tissue over time
  3. PubMed: Laser-direct infrared imaging of microplastics in human samples
  4. PubMed: Py-GC/MS, contamination and plastics in human tissue
  5. PubMed: Inhaled microplastics, lung inflammation and fibrosis

Back to Table of Contents

Connections

Back to Table of Contents