Serum thyroid stimulating hormone level for predicting utility of thyroid uptake and scan.

Abstract

Background: Thyroid uptake and scan (TUS) is a clinical tool used for differentiation of thyrotoxicosis etiologies. Although guidelines recommend ordering a TUS for evaluation of low TSH levels, no specific value is defined. This study aimed to determine a TSH cutoff at which TUSs yield a greater likelihood of successful determination of etiology to avoid unnecessary testing.

Methods: This was a retrospective study on 137 patients seen by an endocrinologist who underwent TUS for evaluation of low TSH (<0.4 μU/mL). A receiver operating curve analysis was performed to determine the TSH cutoff with maximal sensitivity and specificity for prediction of diagnostic utility.

Results: Ninety percent of TUSs (n = 123) led to a diagnosis, while 10% (n = 14) were inconclusive or normal. Diagnoses included Graves' diseases (52%), toxic multinodular goiter (19%), thyroiditis (12%), and solitary toxic adenoma (7%). The median TSH value was 0.008 μU/mL (IQR 0.005, 0.011), and the median free T4 value was 1.7 μU/mL (IQR 1.3, 2.8). The ROC analysis produced an area under the curve of 0.86. The optimal TSH cutoff value was 0.02 μU/mL (sensitivity 80%, specificity 93%) for prediction of diagnostic yield.

Conclusion: This study demonstrates that TSH is a useful predictor of the utility of TUS in yielding an etiology of thyrotoxicosis. Our analysis showed that TUS had a greater likelihood of determining an etiology when TSH was ≤0.02 μU/mL. This information can help clinicians avoid unnecessary cost and patient time burden when TUS is unlikely to aid in determining the etiology of thyrotoxicosis.