Context: Serum thyroglobulin (Tg) is an indicator of differentiated thyroid cancer (DTC) relapse.
Objective: Our objective was to conduct a meta-analysis of published data about the diagnostic performance of highly sensitive serum Tg (hsTg) during levothyroxine therapy in DTC follow-up.
Data sources: We performed a comprehensive literature search of PubMed/MEDLINE and Scopus for studies published until July 2013.
Study selection: Studies investigating the diagnostic performance of basal hsTg in monitoring DTC were eligible. Exclusion criteria were 1) articles not within the field of interest; 2) reviews, letters, or conference proceedings; 3) articles evaluating serum Tg measurement with a functional sensitivity >0.1 ng/mL; 4) overlap in patient data; and 5) insufficient data to reassess diagnostic performance of basal serum hsTg.
Data extraction: Information was collected concerning basic study data, patient characteristics, and technical aspects. For each study, the number of true-positive, false-positive, true-negative, and false-negative findings for basal hsTg, considering stimulated Tg measurement as a reference standard, were recorded.
Data synthesis: Pooled data demonstrated that the negative predictive value of hsTg was 97% and 99% considering a stimulated Tg measurement >1 ng/mL and >2 ng/mL as cutoffs for positivity, respectively. Despite the high pooled sensitivity of basal hsTg, the pooled specificity, accuracy, and positive predictive value were insufficient to completely substitute for a stimulated Tg measurement.
Conclusions: Basal hsTg measurement has a very high negative predictive value but an insufficient positive predictive value for monitoring DTC patients. Therefore, a Tg stimulation test can be avoided in patients with an undetectable basal hsTg, whereas a stimulated Tg measurement should be considered when hsTg levels are detectable.