Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2010) 21 P389

1Department of Endocrinology, Royal Free Hampstead NHS Trust, London NW3 2QG, UK; 2Department of Clinical Biochemistry, Royal Free Hampstead NHS Trust, London NW3 2QG, UK; 3Department of Clinical Immunology, Royal Free Hampstead NHS Trust, London NW3 2QG, UK.


Background: The BTA guidelines for the use of thyroid function tests recommend the measurement of TSH receptor antibodies (TRAb) when investigating hyperthyroidism of uncertain aetiology, in suspected Graves’ ophthalmopathy and in pregnant women with Graves’ disease. An in-house TSH receptor autoantibody ELISA assay (TRAb) was introduced in 2008. This study has audited the assay performance and evaluated its clinical usefulness in a tertiary centre.

Methods: From May 2008 until July 2009 the Clinical Immunology Department received 251 requests for TRAb. Samples with inadequate clinical information (12%) or duplicates (3%) were excluded. Hospital medical records were available and reviewed for 94% of the processed samples (n=200; mean age 46 years, women 80%). FT4, FT3, TSH, TPOAb were also measured. The TRAb assay titre cut off for a positive sample was ≥0.4 U/l.

Results: Of the 200 patients identified, 63% (n=125) had Graves’ disease. From the remaining 75 patients: 10% had a toxic multinodular goitre (MNG), 7.5% thyroiditis, 3.5% hypothyroidism, 3% hyperemesis gravidarum, 0.5% toxic nodule and 11.5% other diagnosis. TRAb antibodies were detectable in 83% patients with Graves’ disease. In those without Graves’ disease, TRAb were positive in 9% (three patients with autoimmune thyroiditis, one post-partum thyroiditis, one hypothyroidism and two toxic MNG). With a cut-off point of TRAb ≥0.4 U/l the positive predictive value to diagnose Graves’ disease was 95%, sensitivity 83%, specificity 91% and negative predictive value 76%. ROC curve analysis determined an optimal cut-off point of TRAb ≥3.5 U/l with a 99% specificity to diagnose Graves’ disease.

TRAb cut-off (U/l)Sensitivity (%)Specificity (%)PPV (%)NPV (%)
0.483919477
172949567
256959557
3.542999852
72410010045

Conclusion: The assay studied has a high positive predictive value for diagnosing Graves’ disease and a high specificity when TRAb ≥3.5 U/l. However the sensitivity is less than optimal and a negative result does not exclude the presence of Graves’ disease.

Article tools

My recent searches

No recent searches.

My recently viewed abstracts