Previous studies of fixed-dose radioiodine therapy (RIT) for Graves' disease (GD) have utilized a variety of techniques and reported differing success rates. This study sought to compare the efficacy of RIT using two fixed-dose protocols and to estimate the optimal radioiodine (RAI) activity for the treatment of GD. This retrospective trial enrolled 658 patients with GD who received RIT between January 2014 and December 2021. Participants were divided into two groups: protocol 1, which utilized a thyroid size-specific RAI dose, and protocol 2, which employed a modified dose calculation approach. The primary outcome assessed was the presence of euthyroidism or hypothyroidism at the 6-month follow-up. The success rates of RIT were compared between the two protocols. The RIT success rate was marginally lower for protocol 2 than for protocol 1 (63.6% vs. 67.2%); however, the risk of treatment failure did not differ considerably between the groups (relative risk, 1.1089; 95% confidence interval, 0.8937 to 1.3758; P=0.3477). The median RAI activity associated with protocol 2 was lower than that for protocol 1 (10.7 mCi vs. 15.0 mCi, P=0.0079), and the frequency of hypothyroidism was significantly lower in the protocol 2 group (39.0% vs. 48.9%, P=0.0117). The success rate of the modified dose calculation protocol was comparable to that of the thyroid size-specific RAI dose protocol. The former approach reduced RAI activity and the incidence of hypothyroidism following RIT without compromising the success rate.