Fluorinated hydrocarbons used as refrigerants : toxicokinetics and effects in humans

Abstract

Hydrofluorocarbons (HFCs) and hydrochlorofluorocarbons (HCFCs) have replaced the chlorofluorocarbons (CFCs) in refrigeration installations and air-conditioning applications. As substitution progressed, some refrigeration mechanics experienced symptoms of inflammation. The new refrigerants were well studied in animals, but data on toxicology and toxicokinetics in humans were scarce. The aim of this thesis was to study human toxicokinetics and effects of HFC-134a and HFC-143a. Moreover, autoimmune responses due to exposure to halogenated hydrocarbons used as refrigerants were investigated and a case of myocardial infarction was discussed.

Toxicokinetics and Effects (Study I and II): Ten male volunteers were exposed to 500 ppm HFC-134a and/or HFC-143a in an exposure chamber (2 h, workload of 50 W). Blood, exhaled air, and urine were collected up to 19 h post-exposure and analysed for HFC with gas chromatography. The experimental data was described by a physiologically-based toxicokinetic (PBTK) model. Markers of inflammation (C‑reactive protein, serum amyloid A protein, D-dimer, fibrinogen) and uric acid were analysed in plasma collected before and 21 h post-exposure. The volunteers rated symptoms related to irritation and CNS-symptoms on a visual analogue scale before, during and after exposure.

The increase in blood upon exposure to HFC was fast, and apparent steady-states (9.4 µM for HFC-134a, 4.8 µM for HFC143a) were reached within minutes. The post-exposure decrease in blood was fast as well and parallel to that of exhaled air. The urinary excretion of HFC‑134a and HFC‑143a was less than 0.002% of the inhaled amount with half-times of 58 and 53 min. The experimental data fitted well within a PBTK framework, and the relative respiratory uptake was estimated to 3.7 % for HFC‑134a and 1.6% for HFC-143a. No effect due to exposure was seen on symptom ratings or in the electrocardiographic recordings. Fibrinogen plasma concentration had increased after exposure to both HFCs, whereas none of the other inflammatory markers or uric acid had increased significantly. Further studies are needed to confirm or reject this finding.

Autoimmune Response (Study III): Serum samples from 44 Swedish men, occupationally exposed to refrigerants, were screened for antibodies against CYP2E1 with enzyme-linked immunosorbent assay. The men participating in the study were selected from a cohort of 280 refrigeration mechanics. One group comprised 30 men with asthma, joint or influenza-like symptoms and another exposed group comprised 14 men with no such symptoms. Unexposed, healthy subjects from Sweden (n=35) and Italy (n=26) constituted control groups.No increase in CYP2E1 antibody titer was seen among occupationally exposed persons compared to unexposed controls. Neither were there any differences between exposed, symptomatic subjects and healthy exposed or unexposed ones. Further, no relation between antibody titer and severity of symptoms, smoking, inhaled anesthetics, age, or year in occupation was detected.

Case Report (Study IV): A case of myocardial infarction was reported. Possible associations between welding, exposure to decomposed HCFC-22, respiratory symptoms and the myocardial infarction were discussed.

In conclusion, the respiratory uptake of HFC-134a and HFC-143a is low and the post-exposure decrease in blood and exhaled air is fast. There may be an inflammatory response at 500 ppm exposure level (i.e. at the Swedish occupational exposure limit, 8‑h time-weighted average), but further studies are needed to confirm this response. Exposure to halogenated hydrocarbons could not be associated to CYP2E1-related autoimmunity.