After an ischemic stroke with evidence of atherosclerosis, lipid-lowering treatment with a target LDL cholesterol of <70 mg/dL compared with 100 ± 10 mg/dL reduced the risk of subsequent cardiovascular events. In this analysis, we explored the effect in the subgroup of patients with diabetes compared with the subgroup without, as well as in those with newly diagnosed diabetes. Patients with ischemic stroke in the previous 3 months or transient ischemic attack within the previous 15 days and evidence of cerebrovascular or coronary artery atherosclerosis were randomly assigned at a 1:1 ratio to a target LDL cholesterol of <70 mg/dL or 100 ± 10 mg/dL using statin or ezetimibe. The primary outcome was the composite of ischemic stroke, myocardial infarction, new symptoms requiring urgent coronary or carotid revascularization, and death resulting from vascular disease. We performed a prespecified analysis to evaluate the effect in patients with diabetes. Of 2,860 patients enrolled, 643 had diabetes at baseline, with a mean age of 66.2 years and baseline LDL cholesterol of 127 mg/dL, and were followed for a median of 3 years. The primary composite end point occurred in 27 (8.2%) of 328 patients in the lower-target group and in 44 (14.0%) of 315 patients in the higher-target group (adjusted hazard ratio [HR] 0.56; 95% CI 0.34-0.89; P = 0.016). In patients without diabetes, the HR was 0.87 (95% CI 0.66-1.14; P = 0.31; interaction P = 0.15). In those with diabetes, there were three intracranial hemorrhages in both randomization groups (0.9% vs. 1.0%, respectively). Newly diagnosed diabetes occurred in 98 (9.2%) of 1,070 and in 80 (7.4%) of 1,085 patients in the lower- and higher-target groups, respectively (HR 1.27; 95% CI 0.94-1.71; P = 0.11), and baseline higher HbA1c was the unique multivariable predictor. In conclusion, after an ischemic stroke with evidence of atherosclerosis, targeting an LDL cholesterol of <70 mg/dL compared with 100 ± 10 mg/dL consistently reduced the risk of subsequent stroke and other major vascular events in patients with and without diabetes, but the higher risk in those with diabetes yielded a higher absolute risk reduction, with number needed to treat of 17.