Review of acute chemical incidents as a first step in evaluating the usefulness of physiologically based pharmacokinetic models in such incidents.

CONTEXT. Acute chemical incidents can have substantial public health consequences in terms of morbidity and mortality. OBJECTIVE. We aimed to characterize acute chemical incidents and near-misses in the Netherlands and compare the results with previous studies. This review is a first step in evaluating whether Physiologically Based Pharmacokinetic (PBPK) models can be of value in acute chemical incidents. MATERIAL AND METHODS. Government, regional, municipal and University Hospital Institutes involved in the management of acute chemical incidents in the Netherlands were contacted, and they provided data between 2008 and 2010 on the characteristics and consequences of the incidents. The study is a retrospective epidemiological study based on data from five institutes. Incidents involving biological agents or radiation were excluded. RESULTS. A total of 764 reports were available which involved 722 incidents after cross-matching the different sources of data. Forty incidents were excluded, leaving 682 incidents for which information was available in accordance with the inclusion criteria. Of the 682 incidents included in this study, most occurred in non-industrial buildings (37%) or industrial sites (34%). The most frequently observed event types were loss of containment (60%) and fire (36%), leading to gas emission (54%), followed by spill of liquid or solid chemicals (36%). The chemicals involved were most often products of combustion (e.g. smoke, soot, particles, 25%) and volatile organic compounds (e.g. solvents, styrene, xylene, 23%), followed by inorganic gases (e.g. carbon monoxide, hydrogen, hydrogen sulphide, 13%). A minimum of 847 people experienced adverse health effects following exposure during a chemical incident, and 10 fatalities were reported. The most frequently reported symptoms were respiratory (27%), due to irritant chemicals. The number of incidents related to fire and the number of injured people were higher in this study than in previous studies; 49% of the injured were transported to hospital. DISCUSSION. This study helps to identify which chemicals are frequently involved in acute chemical incidents in the Netherlands. The results will be used in future to assess whether PBPK models may be useful for risk assessment of chemicals often involved in acute chemical incidents and for which human toxicological and kinetic data are scarce.