the malformations that showed increasing trends from 1979 to 1987 (Edmonds and James, 1990). Recorded time trends for such anomalies are likely to include artifacts, such as improved case finding and other improvements in diagnosis, but merit further study.
Although causality cannot be assumed from a single study, a study that used a rat model identified a variety of cardiac defects caused by the administration of trichloroethylene and dichloroethylene in solutions delivered through a catheter into the gravid uterus from an intraperitoneal osmotic pump (Dawson et al., 1990). There was a dose-response relationship: 9 percent and 12.5 percent of congenital cardiac anomalies were found in the lower-dose trichloroethylene and dichloroethylene groups; 14 percent and 21 percent, respectively, were found in the higher-dose groups compared with 3 percent in the control group. Experimental studies have also found that TCE applied through a catheter directly into the gravid uterus during the period of heart development induces cardiac anomalies in both chicks (Loeber et al., 1988) and rats (Dawson et al., 1990). Dichloroacetic acid (DCA) and trichloroacetic acid (TCA) are both by-products of chlorine disinfection of water containing natural organic material; TCA is a metabolite of TCE. DCA or TCA exposure of pregnant Long-Evans rats by oral intubation produced dose-related cardiac malformations in fetuses (Smith et al., 1989a,b).
Taken together the preceding evidence strengthens the empirical basis for concluding that TCE induces cardiac anomalies in humans. The grounds for this inference include evidence that the relationship is biologically plausible, in that exposure to TCE or its metabolites experimentally induces cardiac teratogenesis in exposed animals (Loeber et al., 1988; Dawson et al., 1990). Moreover, these same effects have been significantly detected in exposed humans, with some evidence of a dose-response relationship in the animal studies. In addition, markers of exposure to TCE have been detected in humans and animals, and the findings are statistically significant. Thus, a chain of evidence links TCE to cardiac anomalies, although additional studies need to be conducted to confirm the association. In addition, extensive studies in animals demonstrate a range of other effects of TCE from reproductive impairment to effects on DNA (ATSDR, 1989), for which corroborating human data do not exist.
One cross-sectional study of health problems compared rates of morbidity and mortality from 1980-1985 for a number of chronic diseases, including heart disease, anemia, skin cancer, hypertension, stroke, and chronic kidney disease, in residents of three towns surrounding an abandoned Superfund site in Galena, Kansas, with those of residents in two control towns (Neuberger et al., 1990). Environmental