Mapping the environmental-health interface in micro- and nanoplastics research: a bibliometric analysis, 2009–2025

Introduction Research on micro- and nanoplastics (MNPs) connects environmental exposure with biological mechanisms and health-related outcomes. Bibliometric searches at this interface can retrieve ecotoxicological, model-organism, mechanistic, and human-health records and should not be interpreted as inventories of confirmed disease studies. We mapped the growth, collaboration, and thematic structure of this query-defined environmental-health interface. Methods English-language articles and reviews published in 2009–2025 were retrieved from the Web of Science Core Collection (WoSCC) using MNP and health-related terms. The query deliberately retained mechanistic and ecotoxicological evidence, and the health-related terms derived largely from experimental and toxicological studies rather than direct clinical evidence in humans. No title/abstract or full-text eligibility screening was used to classify records as disease studies. The final corpus comprised 2,068 records. VOSviewer (1.6.21) and CiteSpace (7.0. R0) were used for bibliometric analyses. Sensitivity analyses excluded 242 records from two high-volume journals, and cross-database concordance was assessed with PubMed. Results Records from 2024 to 2025 accounted for 59.0% of the WoSCC corpus. China contributed the most publications (n = 1,008), whereas the United States had the highest country-network total link strength (289). The ten most productive journals were in environmental science or toxicology. Major keyword clusters involved environmental exposure, ecotoxicology, oxidative stress, gut and intestinal biology, regulated cell death, and nanoplastic neurotoxicity. Health- and mechanism-related terms tended to have recent average publication years, while the strongest bursts remained environmental or marine. The PubMed set contained 1,613 records and showed a broadly similar increase in annual publication counts to the WoSCC series (Spearman ρ = 0.989). DOI matching identified 1,427 shared records (88.5% of PubMed); neither this overlap nor the temporal correlation validates record-level precision or thematic equivalence. Discussion Health-facing concepts were visible in recent publication years but remained embedded in environmental and experimental toxicology. This analysis maps publication venues, terms, and citation domains; it neither estimates the proportion of direct human evidence nor demonstrates disciplinary transition. Stronger human inference will require validated exposure measurement, prospective studies with prespecified outcomes, and evidence syntheses that directly classify study design.