Thyroid hypoechogenicity after methimazole withdrawal in Graves' disease: a useful index for predicting recurrence?

Objective: A characteristic thyroid ultrasonographic picture with diffuse or scattered low echogenicity has been described in Graves' disease (GD). Thyroid hypoechogenicity in GD at onset has been considered a prognostic index of relapse after medical treatment; moreover, thyroid hypoechogenicity is regularly observed in GD at the onset, but not in patients with 'burned-out' disease. The aim of this study was to evaluate the usefulness of thyroid hypoechogenicity changes in predicting GD relapse.
Design: Longitudinal prospective study of previously untreated patients with GD.
Patients: Thirty-nine consecutive patients aged 10-72 years were treated with methimazole (MMI) for 12-24 months on a titration regimen. Evaluation of patients in remission or with relapse was done 12 and 24 months after MMI withdrawal.
Measurements: Thyroid ultrasonography and TSH receptor antibodies (TRAb) were evaluated in basal conditions and then one month after MMI withdrawal. Thyroid hypoechogenicity score (assessed by the same observer with the same equipment) was graded as: 0 absent; 1 mild; 2 moderate; 3 marked. At the withdrawal evaluation a score < 2 and a TRAb value < 10 U/l were considered as normal.
Results: Twelve and 24 months after withdrawal, there were 10 (25.6%) and 17 (44.7%) relapses, respectively. Neither thyroid hypoechogenicity score nor TRAb values evaluated in basal conditions, showed significant differences between patients remaining euthyroid and those who became again hyperthyroid. In the whole group, the thyroid hypoechogenicity score was significantly lower at the withdrawal than in basal conditions (1.1 +/- 1.1 vs 2 +/- 0.8; P < 0.0001); it was significantly lower in patients in remission (P < 0.001), but not in those who relapsed. The thyroid hypoechogenicity score at withdrawal was normal in 23/29 (79.3%) of patients still euthyroid and in 4/10 (40%) of those who relapsed up to the 12th month (P < 0.05); it was normal in 19/21 (90.4%) of patients still euthyroid and in 7/17 (41.2%) of those who relapsed up to the 24th month (P < 0.05). A normal thyroid hypoechogenicity score at withdrawal of MMI had a higher specificity (0.95) and sensitivity (0.59) with respect to TRAb values (0.86 and 0.53, respectively) for the prediction of the relapse of hyperthyroidism at the 24th month.
Conclusions: Basal thyroid hypoechogenicity cannot be used as an index of relapse of GD. MMI treatment induces evident changes in thyroid hypoechogenicity, mainly in patients who subsequently go into remission. The absence or a low grade of thyroid hypoechogenicity after MMI treatment seems to be a favourable prognostic index of remission of hyperthyroidism in GD.