Risk Assessment of Microplastics in Humans: Distribution, Exposure, and Toxicological Effects

Abstract

Microplastics are widely present in the environment, and their potential risks to human health have attracted increasing attention. Research on microplastics has exhibited exponential growth since 2014, with a fast-growing focus on human health risks. Keyword co-occurrence networks indicate a research shift from environmental pollution toward human exposure and health effects. Additionally, Trend Factor analysis reveals emerging research topics such as reproductive toxicity, neurotoxicity, and impacts on gut microbiota. This meta-analysis included 125 studies comprising 2977 data samples. The results demonstrated that cytotoxicity in experimental systems was primarily concentrated in Grade I (non-toxic, 62.8%) and Grade II (mildly toxic, 27.6%). Notably, inhibitory effects on cells were significantly enhanced when microplastic concentrations exceeded 40 μg/mL or particle sizes were smaller than 0.02 μm. The Gradient Boosting Decision Tree (GBDT) model was applied to predict cell viability, achieving an R² value of 0.737 for the test set and a classification accuracy of 81.5%. Furthermore, reproductive- and circulatory-system cells exhibited the highest sensitivity to microplastics, whereas connective-tissue cells had the lowest survival rates. The study also identified an overuse of polystyrene (PS) polymers and spherical particles in experimental designs, deviating from realistic exposure scenarios.

Keywords: cell viability; humans; machine learning; meta-analysis; microplastics; trend factor.

Figure 1

Flow diagram of data screening.

Figure 2

The temporal trend in the number of publications on microplastic impacts on human health.

Figure 3

Co-occurrence network of author keywords for research on microplastic impacts on human health (keyword frequency > 23); node size is proportional to keyword frequency.

Figure 4

Evolutionary analysis of trends in research on microplastic impacts on human health (T > 0 indicates that the keyword has exhibited an upward trend over the past three years; T < 0 indicates a downward trend).

Figure 5

Distribution of the effects of key exposure parameters on cell viability: (a) concentration, (b) particle size, (c) exposure duration, and (d) combined concentration–size synergistic impact on cytotoxicity. (Each data point represents a data point in a study, where the colors in (a–c) represent the density distribution of the data points, with red to blue representing an increasing trend in the distribution concentration of the data points. The color in (d) represents the cytotoxicity level of the data points, with red to blue representing an increasing trend in the cytotoxicity level).

Figure 6

Effects of microplastic exposure on cell viability in cell lines derived from different organ systems.

Figure 7

Effects of microplastic exposure on cell viability across different cell types.

Figure 8

Effects of microplastic characteristics on cell viability (CV) distributions: (a) by surface properties, (b) by polymer type, (c) by particle morphology, and (d) by viability assay method.

Figure 9

A Sankey diagram illustrating the impact of various classification parameters on cellular toxicity.

Figure 10

Chord diagrams illustrating interrelationships among classification parameters: (a) interaction between organ systems and cell classifications, and (b) interaction between assay methods and microplastic classifications.

Figure 11

The test-set regression prediction results for the three machine learning models (a) GBDT, (b) CatBoost, and (c) LightGBM.

Figure 12

The distribution of feature importances for the three regression models: (a) GBDT, (b) CatBoost, and (c) LightGBM.

Figure 13

Confusion matrices for the three classification models: (a) GBDT on the training set, (b) GBDT on the test set, (c) CatBoost on the training set, (d) CatBoost on the test set, (e) LightGBM on the training set, and (f) LightGBM on the test set.

Figure 14

Feature-importance rankings and ROC–AUC curves for the three classification models: (a) GBDT, (b) CatBoost, and (c) LightGBM; (d) ROC–AUC curves.