A Higher Cutoff for TSI Would Better Predict Recurrence in Patients With Graves’ Disease?

Abstract

Abstract Introduction: GD is an autoimmune disease mediated by immunoglobulins (Igs) that activate TSH receptor (rTSH). Relapse after withdrawal of antithyroid drugs (ATD) can reach 60%. Measurement of TSH receptor antibodies (TRAb) and thyroid stimulating immunoglobulin (TSI) could be an indirect indicator of GD activity. TRAb assays measures thyroid-stimulating, thyroid-blocking and neutral Igs; TSI assays measures only stimulating Igs. Objetive: Evaluate, prospectively, autoimmunity before and after ATD therapy for thyrotoxicosis through TSI measurement. Methods: Patients were evaluated at the first visit and at the time of ATD withdrawal. TSH, thyroid hormones, TPO antibody, thyroglobulin antibody, and TRAb were measured using eletrochemiluminescent assays Roche Diagnostics; TSI was determined by chemiluminescent assay Siemens Diagnostics. According to manufacturers, TRAb < 1.75 IU/L and TSI < 0.55 IU/L were negative. Results: Sixty-seven patients mean age 45,7±2,45 years, 65 women, were evaluated: 50 at the first visit, 40 (80%) with GD, and 10 (20%) with toxic multinodular goiter (TMNG). TSI diagnostic sensitivity (Sen%) and specificity (Spe%) to diagnose GD were 90% and 100% respectively, similar to that of TRAb, of 89% and 100%. Thirty-six patients were evaluated for recurrence after suspension of ATD (19 of them also had the initial assessment): 21 (58.3%) did not present recurrence in an mean period of 9.5±2.1 months (3-18); and 15 (41.7%) relapsed in 4.4±2.6 months (2-12). In 10/21 patients who did not relapse, and whose TRAb was negative, TSI was positive at low levels, which was responsible for the low Spe% of this test. Assessing possible other cutoff points for the TSI in the recurrence assessment, an adjustment to 1.4 (TSI <1.4 IU/L = negative) raised the Spe% to 86%. Conclusions: In this group, TSI and TRAb were equivalent for GD diagnosis. Many clinical factors have been suggested and TRAb measurement is known to be useful for predicting GD relapse because of the active pathogenic role of TRAb. For predicting recurrence, with the proposed cutoff point proposed by the kit manufacturer for TSI, a better sensitivity was obtained when compared with TRAb (93% versus 67%), despite very low specificity (38%); by raising the cutting point to 1.4 specificity could be increased to 86% without reduced sensitivity. A larger sample in needed to support a higher TSI cutoff point in the clinical routine for the assessment of GD recurrence after ATD.