Records of 231 patients (171 males, 60 females; aged 10 months to 83 years [median 45 years]) who underwent operation for constrictive pericarditis at the Mayo Clinic from 1936 through 1982 were reviewed. All had had hemodynamically significant pericardial constriction preoperatively, and pericardial disease was confirmed at operation. Preoperatively, 69% were in New York Heart Association Class III or IV and 81% had peripheral edema or ascites. Pericardiectomy was performed through a left anterolateral thoracotomy (34%), a median sternotomy (27%), a U incision (Harrington) (21%), or a bilateral anterior thoracotomy (18%). Postoperatively, 28% of patients had evidence of low cardiac output; 70% of the 32 deaths within 30 days of operation were due to low cardiac output. Operative risk was significantly (p less than 0.001) related to preoperative disability (1% for Class I or II; 10% for class III; 46% for Class IV). Median postoperative follow-up was 9 years (longest was 43 years). Probability of survival for patients dismissed alive from the hospital was 84% at 5 years, 71% at 15 years, and 52% at 30 years. Long-term survival (excluding operative mortality) was not significantly influenced by the disability class preoperatively, the operative approach, or the development of low cardiac output in the immediate postoperative period. At the end of the follow-up interval, there were 141 patients in whom functional capacity could be assessed; 140 were in Class I or II. We conclude that a poor hemodynamic result after complete pericardiectomy relates to the preoperative degree of constriction and resultant cardiomyopathy. We recommend early pericardiectomy when pericardial constriction is diagnosed, and we continue to use a left anterolateral thoracotomy as the preferred approach for most patients.