SAT-415 Clinical Performance of Thyroid-Stimulating Immunoglobulin for Graves′ Disease Diagnosis

INTRODUCTION The diagnosis of Graves′disease (GD) is likely when patient presents hyperthyroidism, symmetrical goiter and orbitopathy and further evaluation is unnecessary. Nevertheless patient with nodular thyroid or in the absence of clinical orbitopathy, measurement of TSH receptor antibodies (TRAb) is recommended to distinguish GD from toxic multinodular goiter, toxic adenoma and other etiologies. Radioiodine uptake (RAIU) also helps to diagnosis when TRAb is unavailable. Third generation TRAb assays measure all types of TSH receptor antibodies: stimulating, blocking and neutral with an excellent sensitivity and specificity for GD diagnosis. Recently, an automated bioassay for the thyroid stimulating immunoglobulin (TSI) was introduced to improve GD diagnosis. OBJECTIVE To analyze clinical performance of TSI and compare with TRAb assay for GD diagnosis. MATERIAL AND METHODS Serum samples of 117 patients with thyrotoxicosis due to GD, toxic multinodular goiter and toxic adenoma were runned simultaneously in two analytical system commercially available. TRAb was analyzed by Elecsys Anti-TSHR assay (Roche Diagnostics, Germany) in a Cobase411 analyzer (Roche Diagnostics, Germany) and results, according to the manufacturer interval reference, were negative if < 1.75 IU/L (analytical range: 0.3 to 40 IU/L). TSI was measured by Immulite TSI assay (Siemens Healthcare, UK) in Immulite XPi 2000 and results, according to the manufacturer interval reference, were negative if < 0.55 IU/L (analytical range: 0.1 to 40 IU/L). Clinical diagnoses of thyrotoxicosis were determined according to ATA guideline (2016). Statistical analyses were performed using SPSS and MedCalc softwares. Comparison were evaluated by regression equations and were considered significant when p values were < .05. RESULTS From 2017 to 2019 a total of 312 serum samples from 117 patients (96 females) were evaluated with mean age 49.5 ±15.8 years-old (18 to 90 yrs). We excluded 26 results above the analytical measurement range of both assays. A high degree correlation was observed with a slope of 0.647 and an intercept -.094 IU/L. Spearman correlation coefficient was 0.858 (p < 0.0001, 95% CI 0.825 to 0.886). TSI assay had higher sensibility and negative predictive value compared to TRAb (95.6% vs 88.5% and 83.6% vs 54.8%, respectively) but lower specificity and positive predictive value (90.3% vs 100% and 97.5% vs 100%, respectively). The results of TSI assay showed good agreement with those of the TRAb assay (k=0.74). A total of 31 samples were discordant, favoring to TSI assay in all but one case of GD according to clinical diagnosis (TRAb positive and TSI negative). CONCLUSION TSI assay showed an excellent performance for GD diagnosis with a better sensibility than TRAb assay.