Introduction: Accurate clinical staging of mediastinal lymph nodes (MLNs) of patients with non-small cell lung cancer (NSCLC) is important in determining therapeutic options and prognoses. Integrated positron emission tomography and computed tomography (PET/CT) scanning is becoming widely used for MLN staging in patients with NSCLC. We performed a bivariate meta-analysis to determine the pooled sensitivity (SEN) and specificity (SPE) of this imaging modality.
Methods: The PubMed/MEDLINE, Embase, and SpringerLink databases were searched for articles related to PET/CT for MLN staging in patients with NSCLC. SEN and SPE were calculated for every study. Hierarchical summary receiver operating characteristic curves were used to summarize overall test performance and assess study quality. Potential between-study heterogeneity was explored by subgroup analyses.
Results: Fourteen of 330 initially identified reports were included in the meta-analysis. When we did not consider the unit of analysis, the pooled weighted SEN and SPE were 0.73 (95% confidence interval [CI]: 0.65-0.79) and 0.92 (95% CI: 0.88-0.94), respectively. In the patient-based data analysis, the pooled weighted SEN was 0.76 (95% CI: 0.65-0.84) and the pooled weighted SPE was 0.88 (95% CI: 0.82-0.92). In the MLN-based data analysis, the pooled SEN was 0.68 (95% CI: 0.56-0.78) and the pooled SPE was 0.95 (95% CI: 0.91-0.97).
Conclusions: Integrated PET/CT is a relatively accurate noninvasive imaging technique, with excellent specificity for MLN staging in patients with NSCLC. Nevertheless, current evidence suggests that we should not depend on the results of PET/CT completely for MLN staging in patients with NSCLC.