The effect of cyclosporin (CsA) on the endotoxin-induced generalized Shwartzman reaction (GSR) was studied in diabetic and nondiabetic rats. After 4.5 wk of diabetes, CsA (20 mg/kg) or intralipid as a control substance was given intraperitoneally daily for 10 days. Next, diabetic rats were given either high-dose (2 mg/kg or low-dose (0.1 mg/kg) endotoxin (Escherichia coli 026:B6 lipopolysaccharide B) as a single injection. The rats were killed at intervals of 1, 4, 8, and 24 h. No significant glomerular thrombi were seen in the nondiabetic control animals, whereas the severity of glomerular thrombi in the diabetic animals was dependent on the presence or absence of CsA, endotoxin dose, and degree of glycemic control. In the diabetic rats, glomerular thrombi occurred maximally at 4 h but were no longer present at 24 h. The CsA/high-dose-endotoxin rats had fewer glomerular thrombi than rats receiving the intralipid/high-dose endotoxin, but this difference was not statistically significant. The CsA/low-dose-endotoxin rats had increased glomerular thrombi compared with the intralipid/low-dose-endotoxin rats (P less than 0.01). Insulin treatment reduced the glomerular capillary thrombi in the CsA/low-dose-endotoxin diabetic animals. Thus, CsA aggravates the GSR with low-dose endotoxin but has no significant effect when high-dose endotoxin is given. Improved glycemic control reduces the GSR in CsA-treated rats. Thus, the interrelationships of diabetes, endotoxin, and CsA on the GSR are complex, and the pathogenesis of these events is unclear.