Cappelli Carlo, Gandossi Elena, Castellano Maurizio, Pizzocaro Claudio, Agosti Barbara, Delbarba Andrea, Pirola Ilenia, De Martino Elvira, Rosei Enrico Agabiti
Department of Medical and Surgical Sciences, Internal Medicine and Endocrinology Unit, University of Brescia, Italy.
Endocr J. 2007 Dec;54(5):713-20. doi: 10.1507/endocrj.k06-069. Epub 2007 Aug 3.
In most trials, at least 30-60% of patients with Graves' disease treated with antithyroid drugs relapse within 2 years after therapy withdrawal. At present, there are no prognostic parameters available early in treatment to indicate patients likely to achieve long-term remission. Because thyrotropin receptor autoantibodies (TRAb) are specific for Graves' disease, we evaluated the ability of their levels and of their rate of change to predict long-term prognosis. In our study 216 consecutive patients with newly diagnosed Graves' disease started a therapy with methimazole. Patients were treated until they achieved euthyroidism and TRAb were measured at 6-month intervals throughout a follow up of 120 months. Our study demonstrated that at the onset of hyperthyroidism patients' age, sex, fT4 levels and goiter size had no prognostic value in predicting long-term prognosis (respectively p = 0.79; p = 0.98; p = 0.83; p = 0.89). On the contrary, at the time of diagnosis TRAb titer was a good predictor of the final outcome (p<0.001); a titer equal to (or) more than 46.5 UI/L could identify patients who had never achieved long-term remission with a sensitivity of 52% and a specificity of 78%. Also fall rate of TRAb at 6 months of follow up and after therapy withdrawal were useful to predict the final outcome (p<0.001). At 6 months of follow up the time of therapy withdrawal, a decrease of TRAb lower than 52.3% or even its increase could identify patients who had never achieved permanent remission with a sensitivity of 55% and a specificity of 79.1%. No single parameter among TRAb, satisfactory identified a sub-set of patients who achieved long remission. Accordingly to our data, the best result in predicting long term remission is probably given by the presence of at least one of the two features evaluated at 6 months (TRAb titer and/or percentage of TRAb fall rate), with a sensitivity of 63% and specificity of 88%. TRAb titers evaluated both at the onset of hyperthyroidism that at 6 months of therapy or their rate of fall at 6 months and at ATD withdrawal are predictors of outcome. However, the presence of at least one, between titers of TRAb or their rate of fall at six months, resulted to be the best predictor of remission with the higher sensitivity and specificity.
在大多数试验中，至少30% - 60%接受抗甲状腺药物治疗的格雷夫斯病患者在停药后2年内复发。目前，治疗早期没有可用的预后参数来表明患者可能实现长期缓解。由于促甲状腺素受体自身抗体（TRAb）对格雷夫斯病具有特异性，我们评估了其水平及其变化率预测长期预后的能力。在我们的研究中，216例新诊断的格雷夫斯病患者开始接受甲巯咪唑治疗。患者接受治疗直至甲状腺功能恢复正常，并在120个月的随访期间每隔6个月测量一次TRAb。我们的研究表明，在甲状腺功能亢进症发作时，患者的年龄、性别、游离甲状腺素水平和甲状腺肿大小在预测长期预后方面没有预后价值（分别为p = 0.79；p = 0.98；p = 0.83；p = 0.89）。相反，在诊断时TRAb滴度是最终结局的良好预测指标（p<0.001）；滴度等于（或）高于46.5 UI/L可以识别出从未实现长期缓解的患者，敏感性为52%，特异性为78%。随访6个月时以及停药后TRAb的下降率也有助于预测最终结局（p<0.001）。在随访6个月即停药时，TRAb下降低于52.3%甚至升高可以识别出从未实现永久缓解的患者，敏感性为55%，特异性为79.1%。TRAb中没有一个单一参数能够令人满意地识别出实现长期缓解的患者亚组。根据我们的数据，预测长期缓解的最佳结果可能是在6个月时评估的两个特征中至少出现一个（TRAb滴度和/或TRAb下降率百分比），敏感性为63%，特异性为88%。在甲状腺功能亢进症发作时以及治疗6个月时评估的TRAb滴度或其在6个月时和停用抗甲状腺药物时的下降率是结局的预测指标。然而，TRAb滴度或其6个月时下降率中至少有一个存在，是缓解的最佳预测指标，具有更高的敏感性和特异性。
