In Germany in 2006 a series of rapidly developing and sometimes severe cases of pulmonary health impairment were observed after normal use of the "Magic Nano Glass & Ceramic" spray and "Magic Nano Bath" spray. In contrast, the previously marketed "Magic Nano" pump spray product (handheld trigger device without propellants) was unobtrusive. Analysis of particles discharged from these products did not reveal stable (solid) nano-sized particles. The precipitous increase of pulmonary health impairment in humans caused by "Magic Nano Sprays" triggered a comparative assessment of the acute inhalation toxicity of "Magic Nano Glass & Ceramic" spray, "Magic Nano Bath" spray, and "Magic Nano" pump spray in rats. The first two test specimens were examined as spray-can aerosols using an intermittent generation principle, whereas the undiluted liquid content of the pump spray was continuously aerosolized. Groups of Wistar rats were nose-only exposed for 4 h. However, due to mortality occurring already during exposure following exposure to Glass & Ceramic spray, the exposure duration was reduced to approximately 2 h in some groups. In addition to endpoints called for by contemporary testing guidelines, respiratory tract injury was also probed by respiratory function measurements during exposure supplemented by analyses in bronchoalveolar lavage (BAL) fluid on the first postexposure day, including lung histopathology in rats exposed to Glass & Ceramic spray. The Glass & Ceramic spray caused mortality at 2269 mg/m(3) and above, the pump spray was in the beginning lethal range at 81222 mg/m(3), while the bath spray was tolerated without mortality up to the maximum tested nominal concentrations of 28100 mg/m(3). The time-adjusted 4-h LC(50) of Glass & Ceramic spray was 5098 mg/m(3). The analysis of respiratory patterns revealed changes indicative of both upper and lower respiratory tract sensory irritation. In addition to clinical signs suggestive of marked lung irritation, especially in the rats exposed to the Glass & Ceramic spray, histopathology revealed lung inflammation, hemorrhages, edema, and focal septal thickening. Lung weights and BAL endpoints (lactate dehydrogenase [LDH], protein, gamma-glutamyltransaminase, and neutrophilic granulocytes) were markedly increased. In summary, this comparative study demonstrates that the conventional OECD 403 protocol is suitable to comparatively assess the potential and potency of these types of consumer products in their end-use configuration. Measurements in BAL were most suitable for the identification of acute lung injury. By contrast, lung function measurements during exposure did not demonstrate any conclusive association with lung injury. It does not appear that the particle size per se is a key determinant in the toxicity of "Magic Nano Glass & Ceramic" spray (approximately 100% mortality occurred at a MMAD of approximately 7 microm; GSD approximately 3). This might suggest that more volatile substances or substances contained in the test articles that co-evaporate with solvents or water may be causative for the findings observed. Hence, the findings of this study support the notion that the assessment of the acute inhalation toxicity of complex end-use products is methodologically challenging and cannot be readily anticipated based on compositional or physical (particle size) data. Accordingly, in the absence of predictive and validated in vitro assays, in vivo inhalation testing of potentially toxic commercialized spray products appears to be indispensable for consumer safety. In order to prevent indiscriminant testing of such products in bioassays, the development of standardized in vitro alternatives should be considered with high priority.