Radioiodine and thyroid eye disease

Ophthalmopathy is a characteristic feature of Graves’ disease, although it is usually mild or subclinical. Enlargement of the extraocular muscle can be shown by computed tomography or magnetic resonance scanning in most patients with autoimmune hyperthyroidism, but only in 10-25% of cases does this result in clinically important problems such as proptosis, conjunctival oedema, or ophthalmoplegia—and the dreaded complication of optic nerve compression is mercifully rare.¹ When it does occur, however, severe thyroid eye disease is difficult to treat and may result in disfigurement, diplopia, or visual loss. Radioactive iodine (I-131) is widely used to treat the thyrotoxicosis of Graves’ disease, but, despite its demonstrable efficacy and safety,² there have long been concerns about its possible adverse effect on thyroid eye disease. Recently definitive evidence for this link has been presented.³ As a result, all doctors should now be aware that radioiodine should be used with caution in patients with ophthalmopathy.

In their large, well designed study Bartalena et al treated 443 patients with Graves’ hyperthyroidism and mild or no ophthalmopathy with methimazole until euthyroid, then randomly allocated them to continued methimazole treatment, radioiodine, or radioiodine with adjuvant corticosteroid therapy.³ The groups were well matched at baseline, and hypothyroidism or persistent hyperthyroidism after radioiodine treatment was corrected promptly. The results of the study were clear cut. After radioiodine treatment 15% of patients developed new or worsened ophthalmopathy, whereas this occurred in only 3% of patients treated with methimazole and in none treated with radioiodine plus prednisone. In the radioiodine group 24% of those with pre-existing ophthalmopathy suffered an exacerbation, whereas only 8% of patients without eye disease at baseline developed it.

In most cases the adverse effect was transient, lasting two to three months, but eight patients, seven of whom had ophthalmopathy at baseline, required orbital radiotherapy and high dose corticosteroids. The trial excluded patients with pre-existing moderate or severe eye disease, in whom such an exacerbation could have been disastrous. The study conclusively shows an adverse effect of radioiodine on thyroid eye disease compared to methimazole. It confirms the results of a previous randomised trial,⁴ which was criticised on methodological grounds.⁵

The mechanism by which radioiodine exacerbates ophthalmopathy is poorly understood, as is the pathogenesis of thyroid eye disease in general. Two plausible theories have been advanced.¹ The first is that radiation induced thyroid damage releases one or more antigens which are shared by thyroid and retro-orbital tissues, resulting in immune mediated ophthalmopathy. Putative antigens include a 64 kDa protein which has been isolated from eye muscle and thyroid¹,⁶ and the receptor for thyroid stimulating hormone, which is expressed in retro-orbital tissues as well as in thyrocytes.⁷ Recently an animal model for thyroid eye disease has been developed in mice treated with syngeneic lymphocytes sensitised to the human thyroid stimulating hormone receptor.⁷ This strengthens the role of the receptor in thyroid eye disease and should lead to further productive research. The second mechanism by which radioiodine may exacerbate ophthalmopathy is by the rapid induction of hypothyroidism, causing increased secretion of thyroid stimulating hormone.¹ This in turn may stimulate antigen production by thyrocytes or induce proliferation or differentiation in retro-orbital preadipocytes which express the receptor for thyroid stimulating hormone.

We believe that the study by Bartalena et al³ has important implications for clinical practice. Firstly, since radioiodine treatment carries a substantial risk of exacerbating pre-existing thyroid eye disease it should be avoided as far as possible in patients with active or severe ophthalmopathy, in whom medical therapy with a thionamide drug such as carbimazole is preferable. Radioiodine may be used in patients with mild eye disease but adjuvant corticosteroids should be prescribed. Oral prednisone at a dose of 0.4-0.5 mg/kg daily for one month, tapered over the following two months, was effective in this study; lower doses may also be effective but have not been tested.

Secondly, patients without clinical evidence of thyroid eye disease have a small risk (8% in this study) of developing ophthalmopathy and a very low risk (<1%) of developing severe eye disease. It may be prudent to warn all patients of this possible complication, but the risks do not justify denying most patients the benefits of definitive treatment with radioiodine when indicated. In addition, the risks do not justify the routine use of corticosteroids in patients without ophthalmopathy.

Finally, it is known from previous research that smoking, a raised serum tri-iodothyronine concentration, and uncorrected hypothyroidism are also risk factors for thyroid eye disease after radioiodine.¹,⁴ To minimise the risk of thyroid ophthalmopathy as far as possible, patients should therefore be advised not to smoke, be rendered euthyroid with a thionamide before radioiodine, and be followed closely to detect and correct early hypothyroidism or persistent thyrotoxicosis.