Conclusion
	The diagnostic and therapeutic use of 131I in thyroid disorders is wide spread due to the fact that few adverse effects have been reported. A large amount of studies have addressed the hazards of 131I and some conclusions could be drawn. Firstly, the number of individuals under study in the diagnostic and hyperthyroid cohorts are probably sufficiently large to give reliable results. Secondly, the only patients, with the exception of stomach cancer among hyperthyroid patients, that show significantly elevated risks after 131I exposure, are those treated for a thyroid cancer. The dose to organs other than thyroid from diagnostic activities of 131I for medical reasons to draw any firm conclusions. In the Swedish study, approximately 2,400 individuals were younger than 20 years of age at exposure (4) and three thyroid cancers were found, giving a non-significant increased risk of 1.69 (95% CI 0.35-4.93). There is an increased thyroid cancer risk associ-ated with low doses of external gamma radiation in those exposed at younger ages and the risk decreases with increasing age, but so far no study has found a significantly increased risk of thyroid cancer in any population above the age of 30 years at exposure, regardless of type of exposure. There seems to be little evidence of a carcinogenic effect of 131I after treatment for hyperthyroidism despite organ doses shown to be sufficient to induce cancer in the A bomb survivors. This could be explained by the lower carcinogenic effect of 131I compared to external gamma radiation as shown in epidemiological and experimental studies (10,40). Other studies contradict these results at least for doses lower than 4 Sv (41). Iodine-131 also seems to have a lower effect than short-lived isotopes as 132I, which emit higher energy beta particles at higher dose-rates (42). The lack of an increased risk of thyroid cancer among those treated for hyperthyroidism is probably explained by the high doses having a cell-killing, rather than carcinogenic effect on the thyroid gland. The studies of adverse effects of 131I on thyroid cancer patients are small and difficult to interpret. Use of 131I was not convincingly linked to elevated risks of leukemia or cancer since we were unable to detect a increased SIRs over time and with increasing organ doses. The overall risk in the Swedish study was 1.43 (95% CI 1.07-1.87) for the first 10 years and did not change for the following years (32). The same pattern was seen for organs considered to receive quit high doses from 131I (i.e. salivary gland, stomach, intestines, bladder). A significant excess was seen for salivary glands and these findings were based on three cases, one of whom also received external radiotherapy towards the neck region. An increased risk of tumors in the adrenal gland was noted for both patients treated with 131I and those treated with surgery, suggesting an underlying predisposition or common etiological factor. Seven of the 11 endocrine tumors in the study were pheochromocytomas and three parathyroid adenoma. In 1961, Sipple (43) noted the coexistence of medullary thyroid carcinoma, pheochromocytoma and parathyroid adenoma. Since five of the patients with pheochromocytoma had a medullary carcinoma, this might indicate that the knowledge among physicians to search for other endocrine tumors in thyroid cancer patients could contribute to our findings. The role of individual differences in genetic background as a modifier of risk is yet to be discovered. The risk was significantly elevated for kidney cancer (SIR=3.00, n=7; Table 1) but not for bladder cancer (SIR=1.61, n=4) despite the fact that the dose to the bladder is approximately eight times higher than for the kidney, and one explanation could be the screening for pheochromocytomas. The bone marrow is known to be on of the most radiosensitive tissues in the body, besides breast, lung, and thyroid (7). Several epidemiological studies have found an association between ionizing radiation and leukemia, including reports on patients with cervical, endometrial, and breast cancer, and patients with menstrual disorders and ankylosing spondylitis (44). The association of leukemia and 131I treatment for thyroid cancer reported in several studies is, although based on few patients, noteworthy (Table 1) (29-32). The risk must, however, be discussed in the light of the therapeutically benefits of the treatment as always in cancer treatment regardless of underlying disease. As a conclusion, the present results should not cause any restrictions to the use of 131I for patients with suspected or diagnosed thyroid disorders. The only concern should be younger patients since we have little information on the carcinogenic potential of 131I in children. It could also be concluded that the sharply increased mortality among hyperthyroid patients the first year after diagnosis is not related to the 131I treatment but the underlying disease, urging the physicians to start treatment as soon as possible. As always radiotherapy should follow the ALARA principle, i.e. the administered dose should be As Low As Reasonably Achievable.