An article published on August 5, 2019, in the peer-reviewed journal Environmental Science and Technology reported on in vitro toxicity and chemical composition of plastic consumer products made of different polymers. The study was performed by Lisa Zimmermann and colleagues from the Department of Aquatic Ecotoxicology, Goethe University Frankfurt am Main, Germany, two other German Institutes, and the Department of Biology, Norwegian University of Science and Technology, Trondheim, Norway. The scientists analyzed 34 articles made of eight major polymer types, namely polypropylene (PP), low-density and high-density polyethylene (LDPE, HDPE), polyvinyl chloride (PVC), polyethylene terephthalate (PET), polyurethane (PUR), polystyrene (PS), and bio-based polylactic acid (PLA). For each polymer type four to five items were chosen that covered both food contact materials (FCM), the majority of which represented packaging (20 in total), and non-food contact applications including various articles (14 in total). The in vitro assays covered baseline toxicity in bacteria, oxidative stress, cytotoxicity in mammalian cells and yeast, estrogenicity, and anti-androgenicity.

The authors summarized that PVC and PUR extracts “induced the highest toxicity” across various assays, while PET and HDPE “caused no or low toxicity.” High baseline toxicity similar to PVC and PUR was also observed in all articles made of PLA bioplastics, FCM and non-FCM alike, as well as in three out of four LDPE products (two of them FCM). PS and PP articles showed varying toxicity across the samples tested. Estrogenicity was detected in only four out of 34 products tested, and anti-androgenicity was detected in 9 samples, which included “all PUR extracts, three PVC extracts and one extract [each] from PP and HDPE.”

Comparison of pooled data for food contact and non-food contact articles did not reveal any significant difference for baseline toxicity or estrogenicity, however, non-food contact articles “induced a significantly higher oxidative stress response and anti-androgenicity.” The authors cautioned, however, that “this was not generally true as some individual FCMs were more toxic than non-FCMs made of the same plastic type (e.g., in [the] case[s] of PP, PET and PVC). In addition, there was “a high toxicity for specific food contact articles, including a food wrap (PVC 1, baseline toxicity and anti-androgenicity), a yoghurt  cup, a food tray and a coffee cup lid (PLA 1, 2 and 4, baseline toxicity), a gummi candy packaging (PP 3, oxidative stress response) and another yoghurt cup (PP 2, anti-androgenicity).”

The authors also analyzed all 34 samples with non-target, high resolution gas chromatography-mass spectrometry (GC-QTOF-MS). This resulted in detection of 1411 features in total, of which 213 chemicals were tentatively identified. The four most commonly detected chemicals were butylated hydroxytoluene (CAS 128-37-0), which was detected in seven samples, and 1,7-di-iso-propylnaphthalene (CAS 94133-80-9), methyl isostearate (CAS 68517-10-2), and methyl di-t-butyl hydroxyhydrocinnamate (CAS 6386-38-5), detected in six samples each. With regard to the correlation between the number of detected features and toxicity observed, the authors summarized that, “while there was a general trend for an increased toxicity with higher chemical complexity, chemical and toxicological signatures do not match. Accordingly, it is not possible to predict the toxicity of a polymer based on chemical analysis.”

From the 213 tentatively identified chemicals, the authors further prioritized 25 substances based on abundance in the analyzed extracts as well as toxicity reported in ToxCast database. Among these chemicals, the authors highlighted alpha-pinene (CAS 7785-70-8), alpha-terpineol (CAS 10482-56-1), and octadecanoic acid (CAS 57-11-4) as the “most estrogenic,” butylated hydroxytoluene (CAS 128-37-0) and 1,2-benzenedicarboxylic acid, dinonyl ester (CAS 84-76-4) as “the most potent antiandrogens,” iodopropynyl butylcarbamate (CAS 55406-53-6) and isodecyl diphenyl phosphate (CAS 29761-21-5) as the most potent inducers of oxidative stress, and oleic acid (CAS 112-80-1), isodecyl diphenyl phosphate, and triethyl phosphate (CAS 78-40-0) as the “most cytotoxic” (all estimations based on ToxCast data).

Overall, the study “demonstrate[d] that consumer plastics contain compounds that are toxic in vitro but remain largely unidentified,” but also that “products not inducing toxicity are already on the market.” The authors emphasized that “acknowledging [the] chemical complexity [of plastics] is the first step towards developing new scientific and regulatory approaches to improve their safety.”

Read More

Health and Environmental Alliance (September 17, 2019). “New study on widely used plastic products confirms toxicity of chemical content– Health groups call on new European Commission to make addressing chemical pollution a priority.”

Andrew Turley (September 19, 2019). “Plastics cause in vitro toxicity via unknown mixtures, study finds.Chemical Watch

Daily Mail (September 19, 2019). “The toxic plastic lurking in YOUR home: Three out of four yogurt cups, bath sponges and coffee cup lids contain a cocktail of potentially dangerous chemicals, study finds.”


Zimmermann, L., et al. (2019). “Benchmarking the in vitro toxicity and chemical composition of plastic consumer products.Environmental Science and Technology (published August 5, 2019).