Diagnostic value of pleural fluid neuron-specific enolase for malignant pleural effusion

Neuron-specific enolase (NSE) in pleural fluid has been proposed as a promising diagnostic biomarker. However, existing studies on the diagnostic accuracy of NSE have reported inconsistent results. This study aimed to assess the accuracy of NSE in differentiating malignant pleural effusion (MPE) from benign pleural effusion (BPE) and investigate potential sources of heterogeneity in the diagnostic performance of NSE reported in previous studies.

We prospectively enrolled patients with undiagnosed pleural effusion from two centers in China (Hohhot and Changshu) and blindly measured their pleural fluid NSE level using an electrochemiluminescence assay. The diagnostic accuracy of NSE was assessed using a receiver operating characteristic (ROC) curve and decision curve analysis (DCA). We used the published studies to analyze the association between the prevalence of heart failure (HF) in the studied cohort and the diagnostic accuracy of NSE.

The Hohhot center enrolled 153 patients (66 MPEs, 87 BPEs), and the Changshu center enrolled 58 patients (26 MPEs, 32 BPEs). MPE patients exhibited significantly higher levels of NSE compared to BPE patients in the Hohhot cohort. The areas under the curve (AUCs) for NSE were 0.68 [95% confidence interval (CI): 0.59-0.77] for the Hohhot cohort and 0.65 (95% CI: 0.51-0.79) for the Changshu cohort. The sensitivity and specificity of NSE in the Hohhot cohort were 0.50 (95% CI: 0.38-0.62) and 0.79 (95% CI: 0.70-0.86), respectively, at the 13.92 ng/mL threshold. In the Changshu cohort, the sensitivity and specificity of NSE were 0.42 (95% CI: 0.26-0.61) and 0.84 (95% CI: 0.68-0.93), respectively, at the 62.50 ng/mL threshold. The DCA of NSE was near the reference lines in both cohorts. HF prevalence was positively correlated with AUC in published studies.

The current evidence does not support that NSE serves as a useful diagnostic marker for MPE. The prevalence of HF patients in the studied cohort affects the diagnostic accuracy of NSE.