The effect of sulfinpyrazone on endothelial injury induced by homocysteine has been studied both in vitro, using cultured human umbilical vein endothelial cells, and in vivo, using a primate model of homocysteine-induced arteriosclerosis. Oral sulfinpyrazone (250 mumol/kg body weight per day in three divided doses) in eight chronically homocystinemic baboons (0.14 +/- 0.04 mM plasma homocystine) decreased the extent of aortic endothelial injury as measured morphometrically by silver staining techniques, compared with six untreated comparably homocystinemic animals (denuded surface averaged 0.5% with range 0-2.1 vs 7.7 +/- 1.6%, respectively; P less than 0.001). Sulfinpyrazone therapy to homocystinemic baboons also normalized platelet survival and turnover measurements (5.1 +/- 0.4 days and 70,000 +/- 11,000 platelets/microliter per day vs. 2.8 +/- 0.6 days and 179,000 +/- 19,000 platelets/microliter per day in untreated homocystinemic controls; P less than 0.001). Sulfinpyrazone therapy also reduced the size and frequency of homocysteine-induced intimal lesion formation (P less than 0.001). Although sulfinpyrazone reduced the amount of specific 51Cr release from cultured human umbilical vein endothelial cells induced by 10 mM homocysteine after 24 hours of co-incubation, no effect was observed in assays of endothelial cell detachment when sulfinpyrazone (10(-5) M) or its thioether metabolite were pre- or co-incubated during 24 hours with homocysteine (2.5-10 mM). These data suggest that sulfinpyrazone may protect endothelial cells from injury in vivo by some apparently indirect mechanism.