Diagnostic accuracy and association with lymph node metastasis of the systemic immune-inflammation index in thyroid cancer: a systematic review and meta-analysis.

BACKGROUND

The Systemic Immune-Inflammation Index (SII), a composite biomarker integrating neutrophil, platelet, and lymphocyte counts, has been increasingly investigated in thyroid cancer. However, its diagnostic accuracy and association with lymph node metastasis (LNM) remain uncertain. This study aimed to systematically evaluate the diagnostic performance of SII for thyroid cancer and its association with LNM.

METHODS

A systematic literature search was conducted across PubMed, Scopus, and Web of Science. A diagnostic meta-analysis using the bivariate Reitsma model was performed to pool sensitivity, specificity, likelihood ratios, and the area under the summary receiver operating characteristic curve (AUC). The association between SII and LNM was assessed by pooling mean differences (MD) using an inverse-variance random-effects model. Heterogeneity was evaluated using I² statistics and Cochran's Q test. This systematic review was registered in PROSPERO (PROSPERO 2025 CRD420251233710).

RESULTS

Ten studies were included. The diagnostic meta-analysis (6 studies; 2,209 participants) yielded a pooled sensitivity of 76.8% (95% CI: 63.8-86.1%), specificity of 71.2% (95% CI: 54.9-83.4%), and AUC of 0.805. The positive and negative likelihood ratios were 2.78 and 0.34, respectively. The LNM meta-analysis (5 studies; 2,073 participants) demonstrated significantly higher SII in patients with LNM (MD = 102.87; 95% CI: 58.86-146.89; p < 0.00001; I² = 56%). Leave-one-out sensitivity analyses confirmed the robustness of both findings. Sensitivity analysis excluding one outlier study yielded consistent results with reduced heterogeneity (MD = 105.72; I² = 32%).

CONCLUSION

SII demonstrates moderate diagnostic accuracy for thyroid cancer and is significantly elevated in patients with lymph node metastasis, supporting its potential as an adjunctive biomarker for diagnostic triage and preoperative risk stratification. Given substantial between-study heterogeneity, threshold variability, and partial verification bias affecting specificity, the pooled diagnostic estimates should be interpreted as preliminary and require confirmation in prospective studies with standardized cut-offs and uniform histopathological reference standards. Further large-scale prospective studies are warranted.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/prospero/, identifier CRD420251233710.