An 'injury-time integral' model for extrapolating from acute to chronic effects of phosgene

The present study compares acute and subchronic episodic exposures to phosgene to test the applicability of the 'concentrationxtime' (CxT) product as a measure of exposure dose, and to relate acute toxicity and adaptive responses to chronic toxicity. Rats (male Fischer 344) were exposed (six hours/day) to air or 0.1, 0.2, 0.5 and 1.0 ppm of phosgene one time or on a repeated regimen for up to 12 weeks as follows: 0.1 ppm (five days/week), 0.2 ppm (five days/week), 0.5 ppm (two days/week), or 1.0 ppm (one day/week) (note that the CxT for the three highest exposures was the same). Animals were sacrificed at 4, 8, and 12 weeks during the exposure and after four weeks recovery. Bronchoalveolar lavage (BAL) was performed 18 hours after the last exposure for each time period and the BAL supernatant assayed for protein. Elevated BAL fluid protein was defined as 'acute injury', diminished response after repeated exposure was defined as 'adaptation', and increased lung hydroxyproline or trichrome staining for collagen was defined as 'chronic injury'. Results indicated that exposures that cause maximal chronic injury involve high exposure concentrations and longer times between exposures, not high CxT products. A conceptual model is presented that explains the lack of CxT correlation by the fact that adaptation reduces an 'injury-time integral' as phosgene exposure is lengthened from acute to subchronic. At high exposure concentrations, the adaptive response appears to be overwhelmed, causing a continued injury-time integral, which appears to be related to appearance of chronic injury. The adaptive response is predicted to disappear if the time between exposures is lengthened, leading to a continued high injury-time integral and chronic injury. It has generally been assumed that long, continuous exposures of rodents is a conservative approach for detecting possible chronic effects. The present study suggests that such an approach my not be conservative, but might actually mask effects that could occur under intermittent exposure conditions.