Okamoto, Takahiro1; Nagai, Erin1; Tokumitsu, Hiroki1; Noguchi, Ei-ichiro1; Sakamoto, Akiko1
1 Department of Surgery II, Tokyo Women’s Medical University, Tokyo, Japan
Background / Purpose: Clinical utility of measuring serum thyroglobulin (Tg) in the differential diagnosis of follicular thyroid neoplasm has been controversial. The evidence on the test performance was critically appraised.
Methods: A systematic review of the literature was conducted. MEDLINE (1966 – 2016) was searched for relevant studies. Bibliographies of the retrieved articles were perused for additional citations. Inclusion criteria for the systematic review were as follows: a study population was patients with surgically proven follicular thyroid neoplasm, the test examined was serum Tg, the reference standard was histopathology, and the outcomes were sensitivity/specificity. Selected studies were appraised their risk of bias using QUADAS (Quality Assessment of Diagnostic Accuracy Studies) -2.
Results: 3 retrospective studies met the inclusion criteria. 2 studies using the cut-off value of 1,000 ng/ml reported sensitivity of 0.57 (95%CI: 0.29 – 0.82) and 0.25 (0.17 -0.34), and specificity of 0.87 (0.79 – 0.93) and 0.84 (0.81 – 0.86), respectively. 3rd study estimated sensitivity to be 0.72 (0.55 – 0.85) and specificity to be 0.80 (0.66 – 0.91) with the cut-off of 500 ng/ml. 1 study limited its study population to solitary nodule while others included multiple nodules. None of the studies reported reproducibility of histopathological diagnoses. It was unclear whether verification bias was avoided in all 3 studies.
Discussion & Conclusion: Quality of the evidence was unsatisfactory, yet serum Tg > 1,000 ng/ml may be suggestive of follicular thyroid carcinoma since its specificity was high. Further research with methodological rigor is needed to clarify the usefulness.