To evaluate the accuracy of 18F-FDG-PET/CT for the detection of recurrent and/or metastatic diseases in differentiated thyroid cancer (DTC) patients with thyroglobulin elevation and negative iodine scintigraphy. Whether PET/CT with TSH stimulation (sPET/CT) had better diagnostic performance than PET/CT without TSH stimulation (nsPET/CT) in this scenario was also evaluated.
PubMed and Embase databases were searched for eligible studies from January 2001 to December 2018. Only studies with clearly stated reference standard (histopathology confirmation and/or clinical/imaging follow-up) were included. Publication bias was assessed by Deeks funnel plot. The pooled sensitivity, specificity, diagnostic odds ratio (DOR) and the area under the summary receiver-operating characteristics curve (AUC) for PET/CT was determined by random-effect analysis, respectively. sPET/CT and nsPET/CT were compared pairwise for all diagnostic estimate indexes using Z-test.
We included 17 studies with 1195 patients in this meta-analysis. The pooled sensitivity, specificity, DOR and AUC for PET/CT on patient-based data were 0.86 (95% CI: 0.79–0.91), 0.84 (95% CI: 0.72–0.91), 31.00 (95% CI: 12.00–80.00) and 0.91 (95% CI: 0.88–0.93), respectively. There was high heterogeneity (I2 = 80% for sensitivity, I2 = 82% for specificity) and possible publication bias (P = 0.01). Z test did not detect statistically significant difference between sPET/CT and nsPET/CT for all the diagnostic estimate indexes (all P > 0.05).
On patient-based analysis, 18F-FDG-PET/CT has high diagnostic accuracy for the detection of recurrent and/or metastatic diseases in DTC patients with thyroglobulin elevation and negative iodine scintigraphy, but existing studies were limited by high heterogeneity and possible publication bias. The diagnostic performance of sPET/CT may be not superior to nsPET/CT.
Supplement Figure 1. Summary of the quality assessment of each study.
Supplement Figure 2. Results of Deeks funnel plot of asymmetry test for publication bias. Significant slope indicates that possible bias. ESS: effective sample size.