In a prospective randomized trial we investigated the benefit of blood versus asanguinous cardioplegia in routine coronary by-pass grafting. One hundred consecutive adult patients were randomly assigned to two groups: Group A (53 patients) received cold blood cardioplegia and group B (47 patients) had standard St. Thomas solution. In all patients we combined ante- and retrograde administration, in group A warm reperfusion was carried out before aortic declamping. Aortic cross clamp times, amount of cardioplegic solution and blood potassium levels at the end of cardio-pulmonary by-pass were higher in group A than in group B. The need for inotropic support at the end of the operation as well as in the postoperative period was more frequent in the crystalloid cardioplegia group. The occurrence of atrial fibrillation showed no clear correlation to the kind of cardioplegia used. Perioperative infarction was less frequent in the blood cardioplegia group (3.7% vs 6.3%; p < 0.01) and 30-day-mortality was nil (0% vs 3.2%; p < 0.01). Hospital stay did not differ between the two groups. In this clinical trial we could demonstrate the superiority of blood versus crystalloid cardioplegia in routine coronary by-pass grafting. Ante- and retrograde delivery of cold blood cardioplegia followed by warm reperfusion is a safe and simple method yielding satisfactory results.