Grilled Roadside Fish in South Africa Carries Microplastics and Elevated Metals

Every piece of grilled fish bought from nine roadside stalls in Thohoyandou, South Africa, contained microplastics in the muscle tissue that consumers actually eat, according to a new study published in Archives of Environmental Contamination and Toxicology. The research, led by Alice R. Tshifura and Tatenda Dalu, examined ready-to-eat braaied fish — fish grilled over an open wood or charcoal fire — sold by informal vendors along the R524 road between Thohoyandou and Nandoni Dam in Limpopo Province. Mean microplastic burdens in the edible flesh ranged from roughly 105 particles per kilogram at the least contaminated stall to about 470 particles per kilogram at the worst, and the authors estimate that a single meal could deliver between approximately 180 and 790 plastic particles to the person eating it. Alongside the plastic, the same tissue carried measurable concentrations of twelve elements, several of which approached or exceeded internationally recommended guideline values for fish at particular stalls.

The study focused on a food chain that food safety research has largely overlooked. Informal roadside vendors supply an affordable and culturally embedded source of animal protein across much of the Global South, yet most published work on contaminants in fish has concentrated on formal supply chains such as aquaculture operations, fisheries landing sites and supermarkets. Between August 2025, the researchers purchased four fish from each of the nine stalls trading along the road network on their sampling dates — a complete census of the accessible vendors rather than a random subsample. The fish were mostly hybrids of Nile tilapia and Mozambique tilapia, bought already grilled and displayed uncovered on open grills or trays. Back in the laboratory, the team dissected a standardised portion of dorsolateral white muscle from each fish using acid-washed stainless steel instruments, then split each sample in two: one portion for microplastic extraction and one for element analysis.

The microplastic workflow relied on alkaline digestion with potassium hydroxide to break down the tissue without degrading the synthetic polymers, followed by density separation in a zinc chloride solution and filtration onto fine membrane filters. Particles were counted and classified by morphology and colour under a stereomicroscope, and a representative subset — 60 percent of all visually counted particles — was confirmed by micro-Fourier transform infrared spectroscopy, with 98 percent of those analysed proving to be genuine synthetic polymers. Procedural blanks processed alongside every batch averaged just two particles each, and these background counts were subtracted from the samples. Fibres and films dominated the particle assemblage overall, while fragments appeared almost exclusively at one stall near the busier eastern section of the road, where they reached a mean of roughly 127 particles per kilogram. Transparent particles outnumbered all coloured categories combined, with black, blue, red, yellow and green particles each generally remaining below 50 particles per kilogram.

Statistical analysis confirmed that these differences among stalls were not random noise. Total microplastic abundance differed significantly across the nine sites, and fibres, fragments and foam each showed strong site-related variation, although films and beads did not. Spectroscopic identification revealed a polymer profile dominated by low-density plastics typical of packaging and consumer goods: polyethylene accounted for 40.5 percent of the particles analysed, polypropylene for 23.8 percent, polystyrene for 18.3 percent and polyvinyl chloride for 7.4 percent. This composition points to weathered waste materials — carrier bags, takeaway containers and discarded packaging — as major contributors, consistent with the plastic-rich environment created by street vending, littering around transport nodes and limited waste management infrastructure in the peri-urban settlements surrounding Thohoyandou.

To translate these measurements into exposure, the researchers multiplied the measured concentrations by a typical portion size, then normalised the result to an average adult body weight of 71 kilograms. The highest estimated intake occurred at the stall closest to the busiest road section, at roughly 790 particles per meal or about 11 particles per kilogram of body weight, while the lowest stalls yielded fewer than 200 particles per meal. Using South African per capita fish consumption of 6.6 kilograms per year — equivalent to about 18 grams per person per day — the authors argue that regular consumers of street-vended fish face repeated exposure. Because intake was expressed per kilogram of body weight, children and other low-body-mass consumers carry a disproportionately greater exposure burden. Importantly, no toxicological reference dose currently exists for microplastic particles, so these figures are screening-level estimates of exposure rather than a formal characterisation of health risk.

The element analysis told a more nuanced story. Iron was the dominant trace element at every stall, with mean concentrations reaching approximately 105 milligrams per kilogram at the highest site, followed by zinc at roughly 20 to 45 milligrams per kilogram. Sodium, chromium, manganese, iron, selenium and molybdenum all varied significantly among stalls, while magnesium, phosphorus, potassium, calcium, copper and zinc did not. Mean concentrations of chromium, copper, selenium and molybdenum exceeded the screening values conventionally applied to fish at several stalls. The authors note, however, that no statutory maximum level in fish exists for any of the twelve elements measured — international food standards cover only lead, cadmium, inorganic arsenic and methylmercury — so an exceedance here signals a concentration unusual for edible fish tissue rather than a breach of a legal threshold. Manganese exceeded its benchmark at every stall, but the researchers caution that this element is homeostatically regulated in vertebrates and that the benchmark itself is weakly founded.

The most striking chemical finding concerned the essential macro-elements, which appeared at concentrations one to two orders of magnitude above published ranges for uncontaminated freshwater fish muscle. Sodium, normally 50 to 500 milligrams per kilogram in fish tissue, reached between 10,000 and 22,000 milligrams per kilogram, while calcium, which typically sits between 1,000 and 5,000 milligrams per kilogram, climbed above 60,000 in some samples. A single 100-gram portion of fish from these stalls would therefore deliver 1.0 to 2.2 grams of sodium — at or above the entire World Health Organization daily limit for adults. The plausible contributors are largely post-harvest: heavy salting before grilling, mineral-rich dust deposition at roadside stalls, leaching from calcareous ash during open-fire cooking, and contact with metal cookware or degraded grilling surfaces. For regular consumers, the authors suggest, this sodium load may be the more immediate health concern, particularly in a population where dietary salt intake is already high and hypertension is prevalent.

Disentangling where the contamination originated proved harder than detecting it. The fish were purchased at the point of sale only, and the provenance of individual specimens could not be verified, so differences among stalls may reflect the environments in which the fish were produced or captured as much as roadside vending conditions. No uncooked or pre-display control fish were analysed, meaning the post-harvest pathway — airborne microplastics from tyre wear and road dust, plastic packaging contact, synthetic handling materials, dust settling during cooking — remains a plausible rather than a demonstrated explanation. The study also found that heat-based cooking does not remove microplastics; research cited by the authors indicates that grilling can fragment polymer particles further, potentially increasing particle counts. Regression analyses showed that larger fish contained more particles in absolute terms but fewer coloured particles per unit body mass, and that small fish carried higher concentrations of chromium, iron and molybdenum, hinting that much of the contaminant load may be recent or externally derived rather than solely accumulated over a lifetime.

The wider implications are considerable. Microplastics are known to adsorb metals on their surfaces — experimental work shows that lead and cadmium bind readily to oxidised polyethylene and polypropylene, the very polymers identified here — raising the possibility that some fraction of the metal load in the tissue travels with ingested plastic, although the study did not test this correlation directly. Chronic exposure to toxic metals is associated with renal dysfunction, developmental impairment, neurotoxicity and cardiovascular disease, while the health effects of ingested microplastics remain under active investigation, with concerns centring on inflammation, immune dysregulation and chemical exposure from sorbed pollutants. With four fish per stall, the statistical design had limited power to detect small differences, and the authors stress that non-significant results should not be read as evidence of no difference. Their recommendations are correspondingly practical: improved waste management, covered display of food, cleaner preparation surfaces and vendor training could reduce contaminant loads without undermining the livelihoods that depend on informal fish marketing. The necessary next step, they argue, is tracing fish along the entire value chain from production to consumer, so that the contributions of aquatic environments and roadside handling can finally be separated — and so that food safety monitoring reaches the markets where millions of people actually buy their dinner.

Subject of Research: Microplastic and element contamination in grilled fish sold by informal roadside vendors in Limpopo Province, South Africa

Article Title: Assessment of Microplastic and Element Contamination in Fish Sold by Informal Roadside Vendors in Limpopo Province, South Africa

Article References: Tshifura, A. R., Themba, N. N., Teixeira-de-Mello, F., Munyai, L. F., Dondofema, F., Matam, N. Y., & Dalu, T. (2026). Assessment of Microplastic and Element Contamination in Fish Sold by Informal Roadside Vendors in Limpopo Province, South Africa. Archives of Environmental Contamination and Toxicology, 91(3), Article 28. https://doi.org/10.1007/s00244-026-01223-3

Image Credits: AI Generated

DOI: 10.1007/s00244-026-01223-3

Keywords: microplastics, roadside fish vendors, food safety, South Africa, Limpopo, heavy metals, braaied fish, informal markets, tilapia, dietary exposure, sodium, polyethylene