Endocrine Abstracts

Introduction

Prediction of the efficacy of antithyroid medication in Graves' disease would be of great value. There is no consensus if TSH levels at the end of treatment accurately predict relapse. Previous studies have not carefully defined Graves' disease nor used the same treatment protocol in all patients.

Aim

To study patients with Graves' disease treated with antithyroid medication to confirm or refute an association between TSH level at the end of treatment and the risk of relapse over the subsequent year.

Methods

We reviewed the notes of 40 consecutive patients with Graves' disease treated in our unit over the previous 3 years (only patients with at least 2 of the following: goitre, TSH-R, TPO Antibodies, dysthyroid eye disease). Only patients treated with a carbimazole dose titration schedule and taking 5mg carbimazole prior to stopping were included. Final TSH on the clinic visit prior to discontinuation of carbimazole was recorded along with evidence of relapse over the following year.

Results

24 patients (19 female, [mean age +/- SEM] 44.5 +/- 2.2 years) were in remission; 16 patients had relapsed (9 female, 42.1 +/- 4.3 years) during follow-up ( 3.3 +/- 0.7 months). Mean TSH for the remission group (1.45 +/- 0.27 mU/l) and relapse group (0.39 +/- 0.20 mU/l) were significantly different (p<0.001). There was no significant difference in presenting free T4 concentration, dose of carbimazole used or length of treatment, presence of eye disease or size of goitre between groups. TSH >1.1 mmol/l gave a predictive value for remission of 92%.

Discussion

The significant difference in TSH levels at the end of treatment for Graves' disease may allow prediction of likelihood of relapse and recommendation of early discharge from clinic. This is consistent with the concept of TSH autofeedback. Larger prospective studies are necessary to confirm these findings.