Treatment of Hyperthyroidism 89 was recognized. Several strategies have evolved but long-term follow-up over 15 to 20 years is required in order to be sure that any apparent improvementin the first few years is maintained. In addition, it must be ascertained that no selection bias, related to the several factors discussed above, will affect the results. Reduction of Initial Dose Estimation of the radiation dose delivered to the thyroid is difficult due to errors in estimating size of the gland by palpation, irregular distribution of radioiodine in the gland, and variations in biological half-life. In early series the activity delivered to the gland tended to be high (150 to 300 wCi/g) resulting in radiation doses in excess of 10,000 rads. The result was a high incidence of hypothyroidism. Goolden and Fraser’? devised a treatment schedule which varied the radioiodine concentration per gram of thyroid depending on the overall size of the thyroid. With this method smaller glands received a lower radiation dose (~ 3500 rads at 60 Ci/g) than larger glands. Their previous schedule called for a uniform dose of 150 «Ci '"I per gram except for glands estimated to be greater than 70 grams, which were dosed to a level of 300 wCi/g. Unfortunately, only 1-year results were reported. These showed a hypothyroidism rate of 5 percent compared to their previous experience of 17 percent at | year. However, at | year 38.5 percent of the group were still toxic and required antithyroid drug therapy. Two other groups have demonstrated a reduced incidence of hypothyroidism at 5 or more years followinginitial '*'I treatment but at the price of prolonging the hyperthyroid status of some patients. Smith and Wilson'® reduced the rate of hypothyroidism from 29.0 percent to 7.4 percent at 5 years by reducingthe estimated radiation dose delivered to the thyroid from 7000 rads to 3500 rads. Further treatment with antithyroid drugs for persistent hyperthyroidism was necessitated in 43 percent of the high-dose group and 64 percent of the low-dose group. Appar- ently, additional }*'I treatment was not administered for persistent disease. In the series of Cevallos et al.,!® patients receiving '"*I at 160 wCi/g had a 5.6 year incidence of hypothyroidism of 45.7 percent while 22.9 percent required further 1'I. In their lower dose group (80 4Ci/g) the hypothyroidism incidenceat 5.5 years was 23.5 percent, and 26.5 percent required additional treatment. In both the series of Smith and Wilson andthat of Cevallos et al. there was a continuing rise in the cumulative incidence of hypothyroidism, but the rate of increase was slowed in comparison with the high-dose group in the Smith and Wilson study. In the series of Cevallos et al. the initial incidence at 1 year showed a marked reduction in the low-dose group, but the rate of increase thereafter was higher than in the high-dose group indicating that at some time in the future the two groups would converge.’® Anothergroup receiving 50 wCi/g *"I was reported by Rapoport et al.!7 These patients also had a low incidence of hypothyroidism but there was a high retreatment rate. Since that report a dosage scheme was devised similar to that of Goolden and Fraser taking into account the increased radiosensitivity of the smaller glands.*4 Results of this plan are not yet available. A rather successful long-term result has been obtained by Jackson!® with the delivering of 3500 rads (~ 40 wCi/g) to a large series of patients. Approximately 38 a0bZ594