Abstract Background Placebo effects contribute substantially to outcome in most fields of medicine. While clinical trials typically try to control or minimize these effects, the potential of placebo mechanisms to improve outcome is rarely used. Patient expectations about treatment efficacy and outcome are major mechanisms that contribute to these placebo effects. We aimed to optimize these expectations to improve outcome in patients undergoing coronary artery bypass graft (CABG) surgery. Methods In a prospective three-arm randomized clinical trial with a 6 month follow-up, 124 patients scheduled for CABG surgery were randomized to either a brief psychological pre-surgery intervention to optimize outcome expectations (EXPECT); or a psychological control intervention focusing on emotional support and general advice, but not on expectations (SUPPORT); or to standard medical care (SMC). Interventions were kept brief to be feasible with a heart surgery environment; “dose” of therapy was identical for both pre-surgery interventions. Primary outcome was disability 6 months after surgery. Secondary outcomes comprised further clinical and immunological variables. Results Patients in the EXPECT group showed significantly larger improvements in disability (−12.6; −17.6 to −7.5) than the SMC group (−1.9; −6.6 to +2.7); patients in the SUPPORT group (−6.7; −11.8 to 1.7) did not differ from the SMC group. Comparing follow-up scores and controlling for baseline scores of EXPECT versus SUPPORT on the variable disability only revealed a trend in favor of the EXPECT group ( P = 0.09). Specific advantages for EXPECT compared to SUPPORT were found for mental quality of life and fitness for work (hours per week). Both psychological pre-surgery interventions induced less pronounced increases in pro-inflammatory cytokine concentrations reflected by decreased interleukin-8 levels post-surgery compared to changes in SMC patients and lower interleukin-6 levels in patients of the EXPECT group at follow-up. Both pre-surgery interventions were characterized by great patient acceptability and no adverse effects were attributed to them. Considering the innovative nature of this approach, replication in larger, multicenter trials is needed. Conclusions Optimizing patients’ expectations pre-surgery helps to improve outcome 6 months after treatment. This implies that making use of placebo mechanisms has the potential to improve long-term outcome of highly invasive .