Premium
Performance of a third‐generation TSH‐receptor antibody in a UK clinic
Author(s) -
Theodoraki A.,
Jones G.,
Parker J.,
Woolman E.,
Martin N.,
Perera S.,
Thomas M.,
Bunn C.,
Khoo B.,
Bouloux P. M.,
Vanderpump M. P. J.
Publication year - 2011
Publication title -
clinical endocrinology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.055
H-Index - 147
eISSN - 1365-2265
pISSN - 0300-0664
DOI - 10.1111/j.1365-2265.2011.04022.x
Subject(s) - trab , medicine , prospective cohort study , retrospective cohort study , graves' disease , receiver operating characteristic , anti thyroid autoantibodies , thyroid disease , autoantibody , gastroenterology , antibody , disease , thyroid , immunology
Summary Background UK national guidelines recommend the measurement of TSH receptor antibodies (TRAb) in certain clinical scenarios. A commercial third‐generation TRAb autoantibody M22‐biotin ELISA assay was introduced in May 2008 in our centre. Objective To evaluate the diagnostic performance of a TRAb assay in a retrospective and subsequently a prospective cohort in a UK centre. Design A retrospective review of patients with thyroid disease followed by a prospective observational study in consecutive patients with newly found suppressed serum thyrotrophin (TSH). Patients and Measurements Medical records of 200 consecutive patients with thyroid disorders who had TRAb measured since the introduction of the assay. In a prospective study 44 patients with newly identified hyperthyroidism (TSH < 0·02 mIU/l) had sera assayed for TRAb prior to their clinic appointment at which a final diagnosis was sought. Results In the retrospective cohort, the manufacturer’s cut‐off point of TRAb ≥0·4 U/l resulted in a positive predictive value (PPV) of 95%, sensitivity 85%, specificity 94% and negative predictive value (NVP) 79% to diagnose Graves’ disease using defined criteria. Receiver operating characteristic (ROC) analysis determined an optimal cut‐off point of TRAb ≥3·5 U/l with a 100% specificity to exclude patients without Graves’ disease at the cost though of a lower sensitivity (43%). In the prospective study, the sensitivity, PPV, specificity and NPV were all 96% using the ≥0·4 U/l cut‐off. When combining hyperthyroid patients from both cohorts the assay sensitivity and specificity at ≥0·4 U/l cut‐off were 95% and 92% respectively. A positive TRAb result increased the probability of Graves’ disease for a particular patient by 25–35% and only six (2·5%) patients had a diagnosis of hyperthyroidism of uncertain aetiology after TRAb testing. Conclusions The assay studied specifically identifies patients with Graves’ disease. It is a reliable tool in the initial clinical assessment to determine the aetiology of hyperthyroidism and has the potential for cost‐savings.