Objective estimates of the probability of developing hypothyroidism following radioactive iodine treatment of thyrotoxicosis

Objective: Several risk factors have been shown individually to influence the outcome following radioactive iodine (RAI) therapy in the treatment of hyperthyroidism. However, no attempt has been made to determine their independent prognostic values that could be used in a regression model to provide objective estimates of the probability of developing hypothyroidism. Study design and methods: We audited records of all hyperthyroid patients treated with first dose RAI between 1980 and 1996. Patients were aetiologically categorized into Graves' disease, solitary toxic nodule and toxic multinodular goiter. Following RAI, outcome was categorized as hypothyroidism, euthyroidism and persistent hyperthyroidism. Multiple logistic regression analysis was used to identify significant risk factors, their prognostic values and probability estimates of developing hypothyroidism in the presence of one or more of these factors. Results: The cumulative incidence of hypothyroidism was 55.8% at 1 year and 86.1% at 10 years. Graves' disease (odds ratio: 4.29), presence of thyroid autoantibodies (odds ratio: 3.51), no antithyroid treatment given prior to RAI (odds ratio: 2.50), non-palpable goiter (odds ratio: 2.48) and high RAI dose (odds ratio: 1.90), were identified as significant independent risk factors. We then developed a predictive table that provides objective estimates of developing hypothyroidism. In the absence of all risk factors we can predict an 11.9% probability of developing hypothyroidism; this increases linearly to a 96.4% probability in the presence of all factors. Conclusions: These objective estimates would help in understanding the relative contributions of the known risk factors, and to predict the probability of developing hypothyroidism following RAI treatment. This would not only help patients make an informed consent for a treatment that would lead to life-long replacement therapy but may also prove useful in calculating the RAI dose that may reduce or delay the onset of developing hypothyroidism.