Reliability of real-time elastography to diagnose thyroid nodules previously read at FNAC as indeterminate: a meta-analysis.

The main limit of thyroid fine-needle aspiration cytology (FNAC) is represented by indeterminate report. Recently, real-time elastography (RTE) has been described in the management of these cases. Here, we performed a meta-analysis of published studies specifically focused on the use of RTE in indeterminate thyroid nodules. A comprehensive literature search of PubMed/MEDLINE and Google Scholar databases was conducted by using the combination of the terms "thyroid" and "indeterminate" and "elastography." Pooled sensitivity, specificity, accuracy, PPV and NPV of RTE as predictor of malignancy in thyroid nodules with indeterminate FNAC were calculated, including 95% confidence intervals (95% CI). The area under the summary ROC curve (AUC) was also assessed. Databases found 572 papers, and eight were included in the meta-analysis. Of these, six studies had prospective design and two were retrospective. Pooled malignancy rate was 31%. As common denominator, all studies set the prevalence of hardness within the nodule as risk factor for malignancy of the lesion. Sensitivity of RTE ranged from 11 to 89% (pooled estimate of 69%; 95% CI 55-82%), specificity varied from 6 to 100% (pooled estimate of 75%; 95% CI 42-96%), and accuracy was comprised between 35 and 94% (pooled estimate of 73%; 95% CI 54-89%). The AUC was 0.77. RTE has suboptimal diagnostic accuracy to diagnose thyroid nodules previously classified as indeterminate. Then, RTE alone should not be used for selecting these patients for surgery or not. We advice for further studies using other elastographic approaches and combined RTE and B-mode ultrasonography.