Detection of Various Microplastics in Human Stool

In the past century, global production of plastics has grown exponentially to more than 350 million tons per year, part of which ends up littering the environment (1, 2). Microplastics are defined as plastic particles smaller than 5 mm (3). They either are produced in these dimensions or result from fragmentation of larger plastic structures. Microplastics are concerning because they are increasingly polluting aqueous (4), terrestrial (5), and airborne (6) environments. Moreover, there have been several reports of microplastics in food (7), particularly seafood (8, 9), sea salt (1012), and drinking water (1214). In the field, microplastics are detected primarily in the gastrointestinal tract of marine animals (15), whereas in experimental settings, cellular uptake and tissue accumulation of microplastics and, more importantly, nanoplastics have been shown (1619). Within tissue, plastic particles are seen as foreign bodies and may trigger local immunoreactions (20). Moreover, microplastics can serve as a vector for other chemicals, such as environmental pollutants or plastic additives, which may leach out and cause exposure to hazardous substances (18, 21). Scientists and public authorities have raised concerns about microplastics in food, potential intake by humans, and consequences for health (7, 22, 23), but data are scarce. There are reports on inhalation of microplastics in humans (6, 24), but the gastrointestinal burden has not yet been investigated in humans despite the findings of microplastics in food and in the gastrointestinal tract of marine animals. We thus conducted this prospective case series, which was approved by the Ethics Committee of the Medical University of Vienna (EK Nr. 1866/2017) and performed according to the revised Declaration of Helsinki. Methods Study Participants and Procedures Eight volunteers residing in Tokyo, Japan; Krasnoyarsk, Russia; Groningen, the Netherlands; Birmingham, United Kingdom; Sassari, Italy; Toru, Poland; Enonteki, Finland; and Vienna, Austria, were recruited to represent different geographic regions and dietary patterns. Participants had to be healthy and could not meet any of the exclusion criteria (Table). They were provided with a stool sampling kit and asked to document their food intake (without any dietary restrictions) 6 to 7 days before sampling. The ingredients of toothpaste, cosmetic products, and chewing gum were recorded. Stool sampling was performed by the participants according to predefined instructions to avoid contamination with plastics or synthetic fibers. Metal spoons were used to sample stool, which was then placed in labeled, pseudonymized, and preweighed glass bottles containing an antibacterial aqueous solution (sections 1.1 to 1.4 of the Supplement and Supplement Figure 1). Table. Descriptive Statistics for the Study Cohort Supplement. Supplementary Materials Samples were shipped to Vienna, and an aliquot was analyzed at the Environment Agency Austria laboratory by researchers who were blinded to the sample origin. The samples were chemically pretreated to dissolve natural organic matter (section 1.5 of the Supplement). Remaining microplastics and residues of nondigestible material were filtered using a 50-m metal sieve. After resuspension in ultrapure water, an aliquot was transferred to a filter via a vacuum system and dried. The composition of microparticles (>50 m) was characterized using Fourier-transform infrared (FT-IR) microspectroscopy in imaging mode (Spotlight 400 [PerkinElmer]). The acquired IR map of 1 scanned filter contained about 1 million IR spectra, which were compared with an in-house library. An automated correlation analysis highlighted potential microplastics, which were crosschecked for the presence of characteristic IR peaks to prevent false-positive results (Figure [panels A to C]; sections 1.6 and 1.7 of the Supplement). We focused on 10 common plastics (2): polyethylene, polypropylene, polyvinyl chloride, polystyrene, polyethylene terephthalate, polyamide, polyurethane, polycarbonate, polymethyl methacrylate, and polyoxymethylene. Figure. Identification and relative frequency of microplastics in stool. Microplastics were characterized via FT-IR microspectroscopy and spectra correlation analysis. FT-IR= Fourier-transform infrared; PA= polyamide; PC= polycarbonate; PE= polyethylene; PET= polyethylene terephthalate; POM= polyoxymethylene; PP= polypropylene; PS= polystyrene; PU= polyurethane; PVC= polyvinyl chloride. A. Microscopic image of arbitrary particles and a fiber to provide an overview. B. Chemical composition of solids was determined via FT-IR microspectroscopy in imaging mode. A heat map displays the correlation value with PET, indicating the presence of 1 microplastic fiber and 1 fragment. C. The acquired IR spectrum correlated strongly with the PET reference spectrum. D. Relative frequencies of 9 microplastics detected in 8 stool samples. * Present in all samples and accounted for almost 80% of detected microplastics. The quality control sample (1 procedural blank without stool) was processed and analyzed together with the stool samples to cover all potential contamination sources, such as sample containers, laboratory equipment, chemicals, sample digestion and filtration, and analytic measurements (section 1.7 of the Supplement). Data were processed using Prism 7.00 (GraphPad Software), and results are presented as medians and interquartile ranges (IQRs). Role of the Funding Source This work was performed as part of each researcher's individual employment contract. No specific funding was received for the study. Results Three men and 5 women aged 33 to 65 years participated in the study (Table). None of the participants were vegetarian, and 6 consumed seafood during the observation period. Food was commonly wrapped, packed, or stored in plastic. Seven participants (87.5%) drank from plastic bottles daily, and 3 used cosmetic products containing synthetic polymers (such as shower gel, face wash, or hand cream). However, according to the labels, all toothpastes and chewing gums were microplastic-free (Table). Each participant provided 1 stool sample, with a median weight of 34 g (IQR, 8 to 39 g), of which a median aliquot of 7 g (IQR, 3 to 11 g) was analyzed via FT-IR. All 8 samples contained microplastics, ranging in size from 50 to 500 m. We detected no plastic particle larger than 500 m, and particles smaller than 50 m were not investigated because of methodological limitations. The identified microplastics were mostly shaped as fragments and films and rarely as spheres or fibers. The median microplastic concentration was 20 pieces (IQR, 18 to 172 pieces) per 10 g of stool. In each stool sample, 3 to 7 plastic types were found, and overall, 9 plastic types (out of 10 analyzed) were detected (Supplement Table 1). Polypropylene and polyethylene terephthalate were present in all 8 samples, with relative frequencies of 62.8% and 17.0%, respectively. Polymethyl methacrylate was the only plastic not observed in any sample. All detected microplastics and their relative abundances are shown in the Figure (panel D). Of note, the quality control sample did not contain any of the 10 investigated plastics. Discussion Microplastics are an emerging environmental concern and have already entered the food chain (7). This study is, to our knowledge, the first to provide evidence for the presence of microplastics in human stool, indicating involuntary ingestion. The daily stool excretion of an average adult is approximately 100 g (25), and in our study, we detected a median of 20 microplastic particles per 10 g of fecal matter. We identified polypropylene and polyethylene terephthalate in all stool samples, which accounted for almost 80% of the total microplastic burden. Moreover, we detected 7 other types of microplastic, suggesting that the potential sources of ingestion are manifold. Apart from food products and water, microplastics may also originate from food processing, packaging, or preparation and from airborne fallout. The estimated annual intake of microplastics is 123 to 11000 particles from shellfish (8, 9, 26), 37 to 1000 particles from salt (1012), and 4400 to 5800 particles from tap water (12), whereas airborne fallout accounts for 13731 to 68415 swallowed particles per year (26). A recent meta-analysis (27) concluded that total microplastic consumption ranges from 39000 to 52000 particles per year. Moreover, bottled water seems to be a strong contributor to microplastic ingestion, with an average of 118 to 325 particles per liter (13, 14), for a total of 90000 microplastics annually if the recommended water intake comes entirely from bottled sources (27). In our study, we aimed to characterize everyday domestic life of the participants. Although seafood was rarely consumed, food and drinks were often stored in plastic containers. In contrast, the reported personal care products did not contain any of the 10 investigated plastics. Our study is limited by its small sample, and larger investigations are needed to establish correlations between types and amounts of fecal microplastics and geographic area, food intake, and other potential sources of microplastic ingestion. To analyze microplastics in human stool, we needed to develop new methods. Stool sampling for such analysis is not standardized and might be prone to contamination. The participants were thus provided clean tools and step-by-step instructions to collect approximately 30 g of feces. No sampling difficulties were reported, but we received 2 low-weight samples, from which smaller aliquots were analyzed. These 2 samples contained the highest microplastic concentrations. The ideal sample weight to balance analytic feasibility with the need to avoid sampling bias must be determined in future trials. We used FT-IR microspectroscopy, which is the current analytic standard to identify the type and number of plastic particles (28). Although thi