ABSTRACT
The purpose of the present study is to examine the diagnostic and predictive accuracy of the thyroglobulin (Tg) to thyroid stimulating hormone (TSH) and TSH/Tg ratios in normothyroid patients with differentiated thyroid cancer (DTC). We conducted a retrospective cohort study evaluating the diagnostic accuracy of the serum Tg/TSH and TSH/Tg ratios in normothyroid patients with thyroid nodules. We also systematically searched the international literature using the Medline, Cochrane's CENTRAL, Scopus, Clinicaltrials.gov, EMBASE, and Google Scholar databases for evidence concerning the diagnostic and predictive accuracy of these ratios. Overall, 374 patients were identified in our cohort study of whom 240 were treated for benign disease and 134 were treated for DTC. Significant differences were noted in the Tg/TSH and TSH/Tg values among cases with malignant and benign disease (P=0.020). However, the diagnostic ROC curve did not confirm these results (Tg/TSH=0.572 and TSH/Tg=0.428). After searching the international literature, we identified 8 studies. The majority of the included data reported significant differences among patients with benign/malignant disease and those with successful iodine therapy compared to those with disease relapse. However, the clinical relevance was clearer among studies that investigated the usefulness of these ratios in predicting recurrent disease. The findings of our study support that the Tg/TSH ratio increases in patients with DTC and can, thus, become useful in the future as a predictive marker of ablative 131I therapy success. However, given the significant variability of Tg its diagnostic accuracy remains to date minimal; thus, the actual cut-off value that can be used to discriminate cancer cases from benign disease has not been determined yet.
