We performed a prospective, multicenter, observational study in which patients were assigned to groups (one group with plaque in which MNPs were detected and one group with plaque in which MNPs were not detected) after enrollment. Patients were recruited from Hospital Cardarelli, Ospedale del Mare, and the University of Salerno from August 1, 2019, to July 31, 2020. Consecutive patients with asymptomatic carotid artery stenosis (as classified by the North American Symptomatic Carotid Endarterectomy Trial¹⁴) for whom intervention was indicated were screened for this study. A total of 447 consecutive patients were approached to participate, and 312 agreed to undergo screening. Patients with asymptomatic disease were selected for participation, in order to maximize the chances of surviving the postprocedure period and to minimize interpatient variation in plaque phenotypes.^15,16 Baseline clinical examinations were conducted and electronic health records were evaluated to collect demographic, clinical, and therapeutic-intervention data. Blood samples were obtained according to standard procedures after overnight fasting for analysis of biochemical variables. After undergoing carotid endarterectomy, the patients were followed to monitor the incidence of nonfatal myocardial infarction, nonfatal stroke, and death from any cause until July 1, 2023. Follow-up visits were according to common clinical practice and were not scheduled. Participants without visits during the follow-up period were considered to be lost to follow-up. Events were adjudicated by means of review of electronic health records by investigators who were unaware of the results of pyrolysis–gas chromatography–mass spectrometry (and thus to group assignments). The protocol was approved by the local ethics review committee (Università Vanvitelli Caserta). The participants provided written informed consent, and we adhered to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines checklist for reporting of data.¹⁷

Specimens of the atheromatous plaque that were surgically excised from the carotid artery bifurcation region at atherectomy were collected in glass tubes and then frozen in liquid nitrogen, fixed in 10% buffered formalin, or fixed in 2.5% electronic microscopy–grade glutaraldehyde for subsequent analyses. The abundance of 11 different MNPs was measured with pyrolysis–gas chromatography–mass spectrometry (a quantitative technique that measures MNPs in combination and does not distinguish between microplastics and nanoplastics), and results were corroborated with the use of electron microscopy and stable isotope analysis (see the Supplementary Methods section and Fig. S1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). Samples were analyzed as they became available to researchers, who were unaware of outcome data. All the authors agreed to submit the manuscript for publication.

Patients were eligible if they were 18 to 75 years of age, had asymptomatic extracranial high-grade (>70%) internal carotid artery stenosis, and were scheduled to undergo carotid endarterectomy. Exclusion criteria were evidence of heart failure, valvular defects, malignant neoplasms, or secondary causes of hypertension. Patients with complications in the postoperative period before discharge, who had incomplete data, or who were lost during follow-up were excluded from the analysis.

The primary end point was a composite of nonfatal myocardial infarction, nonfatal stroke, or death from any cause among patients with plaque containing MNPs and patients with plaque that did not contain those substances. Secondary end points included levels of tissue biomarkers interleukin-18, interleukin-1β, tumor necrosis factor α (TNF-α), interleukin-6, CD68, CD3, and collagen in patients with evidence of MNPs as compared with those without.

There were no previous data with which to gauge a sample size. Therefore, an interim analysis that included the first 100 patients was performed to calculate the sample size. We observed that 61 patients had evidence of MNPs in plaque and that such patients had a higher risk of a primary end-point event by a factor of 2.1, than patients without MNPs. Considering an alpha of 0.05, a beta of 0.2, a baseline event rate of 9 events per 100 person-years in the group without evidence of MNPs, and a planned follow-up of 3 years, we calculated a sample size of 246 patients, with the intention of using a proportional-hazards model in the data analysis. We expected a 20% loss to follow-up and adjusted the sample size to 300 patients.

After analysis of atherosclerotic plaque, patients were assigned to one of two groups: patients with evidence of MNPs (having detectable levels of at least one of the two substances) or patients with no evidence of MNPs. The distribution of variables was assessed with the use of the Shapiro–Wilk test. Continuous variables were compared between the two groups with a t-test for normally distributed data, the Mann–Whitney U test for non-normally distributed data, and Fisher’s exact test for categorical variables. Linear regression analyses were performed to estimate the association of the burden of MNPs with plaque markers. Cox regression analysis was used to examine the association between the presence of MNPs within plaque and the incidence of the composite primary end point and was adjusted for age, sex, body-mass index, total cholesterol, high-density lipoprotein and low-density lipoprotein cholesterol, triglycerides, creatinine, diabetes, hypertension, and previous cardiovascular events. A two-sided P value of less than 0.05 was considered to indicate statistical significance. All calculations were performed with the use of SPSS software, version 12. Figures were prepared with the use of GraphPad Prism, version 9.1.2.

A total of 312 patients who were undergoing carotid endarterectomy were screened. Of the patients screened, 8 had a stroke or died before hospital discharge, and 47 had incomplete data or were lost during follow-up (Figure 1A). Of the 257 patients who completed a mean (±SD) follow-up of 33.7±6.9 months, 150 patients (58.4%) had a detectable amount of polyethylene in excised carotid plaque, and 31 of those (12.1%) also had a measurable amount of polyvinyl chloride in the carotid plaque. Among patients with evidence of these MNPs in plaque, the mean level of polyethylene was 21.7±24.5 μg per milligram of plaque, and the mean level of polyvinyl chloride was 5.2±2.4 μg per milligram of plaque (Figure 1B).

The patients’ characteristics at baseline are summarized in Table 1. Patients with evidence of MNPs were younger; more likely to be men; less likely to have hypertension; more likely to have diabetes, cardiovascular disease, and dyslipidemia; more likely to smoke; and had higher creatinine values than those without evidence of plastics in excised plaque; the other clinical variables appeared similar in the two groups. There were no apparent differences in the incidence of MNPs according to the geographic areas where the patients lived or the centers where they were enrolled (Fig. S2). No substantive differences were found between the patients who were included in the analysis and those who were excluded with regard to the clinical characteristics of the patients or the prevalence and the levels of detected MNPs (Table S1).

To substantiate the results obtained with pyrolysis–gas chromatography–mass spectrometry and to gain preliminary information about the size of the MNP particles, we evaluated plaque samples that were positive for both polyethylene and polyvinyl chloride from 10 randomly selected patients using transmission electron microscopy and scanning electron microscopy. Visualization with transmission electron microscopy showed the presence of particles with jagged edges that were probably of foreign origin within the foamy macrophages present in atheromatous plaque and in the amorphous material of plaque (Figure 2A and Fig. S3A). These particles were almost all smaller than 1 μm and were probably nanometers in size.

Observing the same slices using the back-scattered electrons with scanning electron microscopy, we made spectral x-ray maps from some particles that resembled those seen on transmission electron microscopy with regard to size and shape (Figure 2B). These maps provided evidence of a reduction in the presence of carbon and oxygen in the plaque samples and a greater presence of chlorine (Fig. S3B). To compare the elements detectable in the particles with those present in the background, we obtained x-ray spectra from two identical areas and confirmed that the chlorine concentration was higher inside the particles (Fig. S3B). As a further control, we also compared chlorine content in two adjacent identical areas that were devoid of particles and found the levels of chlorine to be similar (Fig. S3C). Given the probable nonbiologic nature of chlorine at the solid state, these results may confirm deposits of polyvinyl chloride. Another example of such particles is shown in Figure S3D; samples from 4 of the 10 patients tested had chlorine levels detectable with this pattern.

Stable isotope analysis was performed on 26 random patient plaque samples because petroleum-derived plastics have lower δ¹³C values (the ratio of carbon-13 to carbon-12) than human tissues.¹⁸ This analysis identified two distinct clusters of patients — one group with higher isotopic values of δ¹³C and one with lower values — which indicated higher and lower ratios, respectively, of carbon-13 to carbon-12. Lower isotopic values of δ¹³C could be due to contamination with MNPs, since petroleum-derived material has an isotopic signal lower than that of human tissues. Lower isotopic values of δ¹³C were more evident in plaque that had evidence of MNPs (Fig. S4).

Because previous data suggested that MNPs can induce proinflammatory pathways,¹³ four inflammatory markers — interleukin-18, interleukin-1β, interleukin-6, and TNF-α — were measured with the use of enzyme-linked immunosorbent assay (Figure 3A through 3D). Collagen content of plaque samples (Figure 3E) and levels of CD3 (Figure 3F) and CD68 (Figure 3G), two markers of lymphocyte and macrophage infiltration, respectively, were assessed by means of immunohistochemical assay. Linear regression analysis revealed a correlation between the amount of polyethylene present and the expression levels of these markers (Fig. S5). Results of these analyses, stratified according to the presence of polyethylene alone or in combination with polyvinyl chloride, are shown in Figure S6.

A primary end-point event (nonfatal myocardial infarction, nonfatal stroke, or death from any cause), occurred in 8 of 107 patients (7.5%) in the group that did not have evidence of MNPs (2.2 events per 100 patient-years) and in 30 of 150 patients (20.0%) in the group that had evidence of MNPs (6.1 events per 100 patient-years) at 33.7±6.9 months. The incidence of individual components of the composite end point is shown in Table S2. Patients with MNPs in plaque had a higher risk of having a primary end-point event than patients with no evidence of MNPs (hazard ratio, 4.53; 95% confidence interval [CI], 2.00 to 10.27; P<0.001) (Figure 4), as shown by Cox regression analysis with adjustment for risk factors for cardiovascular disease (Table S3). The unadjusted hazard ratio was 2.84 (95% CI, 1.50 to 5.40; P=0.007). When the levels of MNPs were analyzed as a continuous variable, the results showed an association with the primary end point (Table S4).