BACKGROUND: The best surgical treatment for hyperthyroidism caused by Graves' disease remains a controversial subject. METHODS: Seven hundred fourteen consecutive patients underwent total or near-total thyroidectomy for Graves' disease in a 13-year period. In a first analysis, postoperative rates of suffocating hematoma, wound infection, recurrent laryngeal nerve (RLN) palsy, hypoparathyroidism, and persistence or recurrence of hyperthyroidism, were studied and compared with the same parameters in 4,426 patients who underwent bilateral thyroid gland resection for other conditions. A second analysis identified factors associated with postoperative complications among Graves' disease patients. RESULTS: Comparing Graves' disease patients with patients who had bilateral thyroid resection for other conditions, the transient morbidity rate was 13.3% versus 8.2% (p < 0.0001), with 10.2% versus 5.0% (p < 0.0001) hypoparathyroidism, 2.2% versus 1.7% (p = 0.35) RLN palsy, 1.7% versus 0.9% (p < 0.05) suffocating hematoma, and 0.3% versus 0.4% (p = 0.67) wound infection, respectively. Permanent morbidity rate was 2% versus 2.2% (p = 0.72), including 0.4% versus 0.6% RLN palsy and 1.5% versus 1.7% hypoparathyroidism. Among the Graves' disease patients, univariate analysis revealed that those who experienced postoperative complications had a higher weight resected thyroid gland (odds ratio = 1.5; 95% CI, 1.0-2.3) and a higher rate of total thyroidectomy (24.4% versus 19.5%, odds ratio = 2.2; 95% CI, 1.4-3.4) than patients without complications. In the multivariable model, these two factors remained independent. There was no recurrence of hyperthyroidism with a median followup of 6.7 years (interquartile range 4.1 to 10.1 years). Persistent hyperthyroidism developed in three patients. CONCLUSIONS: Total or near-total thyroidectomy is an effective and safe treatment for Graves' disease when performed by an experienced surgeon.