Background: Coronary bypass surgery often leads to short-term cognitive dysfunction, whereas coronary angioplasty does not. Perioperative cognitive dysfunction usually resolves, although a subgroup of surgical patients may continue to exhibit long-term cognitive dysfunction. The purpose of this study was to compare cognitive function 5 years after randomization to a strategy of either initial coronary surgery or initial angioplasty. Methods and Results: Five centers in the Bypass Angioplasty Revascularization Investigation participated in this ancillary study. Patients with multivessel coronary disease randomized to angioplasty or surgery were eligible at the time of their 5-year clinic visit. A battery of five measures previously shown to be sensitive to perioperative changes in cognitive function was administered, including the Logical and Figural Memory Scales from the Wechsler Memory Scale, the Digit Symbol and Digit Span subtests from the Wechsler Adult Intelligence Scale, and Part B of the Reitan Trail Making Test. The 125 study patients were generally similar to the 133 patients who were eligible but did not participate, although study participants were significantly younger (P=.003). The 64 patients randomly assigned to angioplasty had baseline characteristics similar to those of 61 patients randomly assigned to surgery. Cognitive function scores were not significantly different between angioplasty or surgery patients in an intention-to-treat analysis (P=.57). There also was no difference in cognitive function scores when the data were analyzed according to whether the patient had ever undergone bypass surgery (P=.59). Conclusions: Long- term cognitive function is similar after coronary bypass surgery and coronary angioplasty in the majority of patients.