ing data from gas-uptake studies did not adequately describe the rapid disappearance of HFC-236fa from exposure chambers. Inhalation uptake of HFC-236fa by rats was biphasic with a rapid equilibration phase of up to 30 min followed by a slow linear uptake phase. The pattern of loss was inconsistent with either first-order or saturable metabolism. Experiments have revealed that HFC-236fa has low blood/tissue partition coefficients.

Acute and subchronic inhalation studies in rats have shown HFC-236fa to be of low toxicity. The principal effects of HFC-236fa appear to be attenuation of alerting response, narcosis, and cardiac sensitization. There is some evidence suggesting accommodation to decrements in alerting response resulting from HFC-236fa exposure. Exposure-related alterations in some clinical chemistry measurements have been shown following acute and subchronic exposure of rats to high concentrations of HFC-236fa, but the changes are of questionable biological or toxicological relevance. Reductions in hepatic β-oxidation were observed in rats following 2-week discontinuous exposure to 50,000 ppm, but there were no histopathological correlates to confirm peroxisome proliferation. There was no evidence of peroxisome proliferation in rats following 90-day discontinuous exposure to HFC-236fa at concentrations as high as 50,000 ppm. Studies in dogs indicate that HFC-236fa might have the potential for cardiac sensitization at high exposure concentrations (i.e., ≥150,000 ppm). There have been no studies assessing the carcinogenic potential of HFC-236fa. NOAEL and LOAEL values for noncancer end points are summarized in Table 2-1.

Results of developmental toxicity studies in rats and rabbits suggest that HFC-236fa at concentrations as high as 50,000 ppm is not a developmental toxicant. Signs of maternal toxicity were evident at concentrations of 20,000 and 50,000 ppm in rats but not in rabbits.

Genotoxicity studies of HFC-236fa are negative.

There are no currently available reproductive toxicity or carcinogenicity studies of HFC-236fa.

The Navy proposes to use the same exposure guidance levels for HFC-236fa that were set for chlorofluorocarbons CFC-12 and CFC-114 (1-hr EEGL of 2,000 ppm, 24-hr EEGL of 1,000 ppm, and 90-day CEGL of 100 ppm), but did not provide an adequate rationale for doing this. To evaluate the validity the Navy's proposed guidance levels, the subcommittee reviewed the available toxicity data on HFC-236fa to determine what levels would be ade-

quately protective of submariner health. A comparison of those results is presented below.

The subcommittee evaluated the Navy's proposed 1-hr EEGL of 2,000 ppm for HFC-236fa by considering an acute toxicity study in rats (Keller 1994) and a cardiac sensitization study in dogs (HRC 1994). In the study by Keller (1994), the NOAEL for a 4-hr exposure was 134,000 ppm based on the end point of narcosis in rats. In the HRC (1994) study, the NOAEL for a 5-min exposure was 100,000 ppm based on cardiac sensitization in dogs. The subcommittee divided those NOAELs by an uncertainty factor of 10 to account for interspecies differences, yielding values of 13,400 ppm based on the Keller (1994) study and 10,000 ppm based on the HRC (1994) study. The Navy's proposed 1-hr EEGL of 2,000 ppm is well below those values, and, therefore, the subcommittee concludes that it is adequately protective of submariner health.

To evaluate the Navy's proposed 24-hr EEGL of 1,000 ppm for HFC-236fa, the subcommittee considered a 2-week toxicity study (Valentine 1995) and a 14-week toxicity study (Valentine 1996) conducted in rats. The 2-week study involved a cumulative exposure of 60 hr and reported a LOAEL of 20,000 ppm and a NOAEL of 5,000 ppm based on alerting response. In contrast, the 14-week study reported a NOAEL of 20,000 ppm based on the same end point. The subcommittee questioned the use of alerting response for determining a NOAEL, because the effects were transient in both studies. In addition, alerting responses are subjectively analyzed and vary among strains and individual animals, and it is unclear how or if such effects are applicable to humans. The subcommittee considered using a NOAEL of 5,000 ppm found in the 2-week study (Valentine 1995) but decided that that value was probably too conservative. Having no data on HFC-236fa at concentrations between 5,000 ppm and 20,000 ppm, the subcommittee believed that 20,000 ppm was a better choice for a NOAEL. The subcommittee