This prospective cross-sectional study examined the long-term effects of early-life drinking water arsenic exposure on pulmonary function and metabolic profiles in exhaled breath condensate (EBC) in a cohort of adults from Antofagasta, Chile. Participants were exposed to arsenic-contaminated drinking water in utero and during early childhood (1958-1970), when arsenic concentrations in the Antofagasta municipal water supply reached approximately 870 µg/L before declining after treatment plant installation. The study recruited 150 adults (75 males, 75 females) aged 47-58 years at the time of study (2017-2019) and classified them into high arsenic exposure (HighAE, drinking water arsenic >=870 µg/L at birth) and low arsenic exposure (LowAE, ⇐110 µg/L) groups.
Key numbers
- n = 150 participants (75 males, 75 females); HighAE n = 75, LowAE n = 75.
- Early-life drinking water arsenic: HighAE group >=870 µg/L; LowAE group ⇐110 µg/L.
- HighAE participants had significantly shorter stature than LowAE participants (a known marker of early-life arsenic toxicity).
- Forced vital capacity (FVC): significantly reduced in HighAE vs LowAE in both males and females.
- Forced expiratory volume in 1 second (FEV1): significantly reduced in HighAE vs LowAE in both males and females.
- Males in HighAE group also had reduced red blood cell concentrations compared to LowAE.
- 8-hydroxy-2-deoxyguanosine (8-OH-2dG), a marker of oxidative DNA damage: significantly elevated in HighAE males vs LowAE (p = 0.030); females in HighAE showed reductions in 8-OH-2dG (sex-divergent pattern noted).
- 8-iso-prostaglandin F2alpha (8-iso-PGF2alpha), a marker of lipid peroxidation and oxidative stress: significantly elevated (p = 0.019) in HighAE participants.
- Untargeted EBC metabolomics: 201 peaks uniquely associated with arsenic exposure status across three analytical methods (EBC-HILIC, EBC-RP, ethanol-RP).
- PLS-DA model discriminating HighAE from LowAE: AUC = 0.97 for the merged untargeted dataset.
- Subgroup of untargeted markers showed concentrations proportional to level of arsenic exposure.
- Current urinary arsenic levels in both groups were similar (exposure ended decades prior), confirming the metabolic profiles reflect early-life rather than current exposure.
Methods (brief)
Cross-sectional study with prospective exposure classification based on historical water arsenic records for Antofagasta. EBC collected using a custom condenser glass tube device (25 min collection). Targeted EBC analysis by LC-MS/MS for oxidative stress markers: 8-OH-2dG, 8-iso-PGF2alpha, o-tyrosine, 11-dh-TXB2, LXA4, 15-HETE. Untargeted EBC metabolomics by liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS). Inflammatory cytokines (IL-6, IL-8, TNF-alpha) and matrix metalloproteinase-1 (MMP-1) measured by immunoassay. Pulmonary function by spirometry per American Thoracic Society/European Respiratory Society guidelines. Statistical analysis: comparative tests for HighAE vs LowAE; PLS-DA for metabolomic discrimination. Historical water arsenic records from Chilean and municipal sources used to classify exposure.
Evidence fitness
Supports: direct evidence that early-life arsenic exposure in drinking water produces lifelong impairment of pulmonary function (FVC and FEV1) and elevated oxidative stress markers in exhaled breath condensate, persisting approximately 47 years after the exposure ended. The EBC metabolomic signature (AUC 0.97) demonstrates that early-life arsenic exposure leaves a durable metabolic footprint in airway fluid. This is not dietary arsenic evidence (the exposure route is drinking water) and does not provide food concentration data.
Limitations
Cross-sectional design; historical water arsenic classification based on municipal records (individual-level historical exposure not directly measured); current arsenic exposures cannot be entirely excluded as contributing; EBC collection and metabolomics are exploratory methods; sex-divergent patterns in some oxidative stress markers require mechanistic explanation; sample size (n = 150) limits subgroup analyses.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.