To analyze the long term outcome after multimodality therapy for acromegaly, a retrospective review was performed on 162 patients who underwent transsphenoidal surgery at Massachusetts General Hospital between 1978 and 1996. The surgical cure rate for microadenomas was 91%, that for macroadenomas was 48%, and it was 57% overall. The surgical cure rate was significantly dependent on tumor size, but was not dependent on age or sex. An improvement in the surgical cure rate was noted over the course of the review, from 45% before 1987 to 73% since 1991. Long term follow-up was obtained in 99% of U.S. residents (149 of 151), with a mean follow-up period of 7.8 yr. Adjuvant radiation and/or pharmacological therapy was given to 61 patients. Of the entire group, 83% (124 of 149) were in biochemical remission as determined by normalization of serum insulin-like growth factor I levels or by GH suppression after oral glucose tolerance testing at last contact or at death. The recurrence rate was 6% at 10 yr and 10% at 15 yr after surgery in those patients who initially met the criteria for surgical cure. The 10-yr survival rate was 88%, and there were 12 deaths at postoperative intervals of 2-12 yr, with the most common cause of death being cardiovascular disease. A Cox proportional hazards model showed that patient-years with persistent disease carried a 3.5-fold [95% confidence interval (CI), 1.0-12; P = 0.02] relative mortality risk compared to those patient-years in remission. A Poisson person-years regression analysis showed no significant difference in survival between those 86 patients cured at operation and an age- and sex-matched sample from the U.S. population [standardized mortality ratio (SMR), 0.84; 95% CI, 0.3-2.2; P = 0.35]. A similar analysis on the entire group of 149 patients showed no significant difference in survival from that in a control sample (SMR, 1.16; 95% CI, 0.66-2.0; P = 0.3). Mortality risk for patient-years with persistent active disease after unsuccessful treatment vs. that in the U.S. population sample remained increased (SMR, 1.8; 95% CI, 0.9-3.6; P = .05). This analysis suggests that the decreased survival previously reported to be associated with acromegaly can be normalized by successful surgical and adjunctive therapy.