Diagnostic value of C-reactive protein in hepatocellular carcinoma: a meta-analysis

Early diagnosis of hepatocellular carcinoma (HCC) is crucial for the treatment of patients. In the current study, our purpose was to appraise the diagnostic value of C-reactive protein (CRP) for HCC in both the normal population and patients with liver cirrhosis (LC). Multiple databases were searched for literature retrieval. Diagnostic meta-analysis was conducted to determine the namely sensitivity, specificity, diagnostic odds ratio (DOR), negative predictive value (NPV), positive predictive value (PPV), the area under receiver operating characteristic (AUROC), likelihood ratios, positive post-test probability (PPP) and negative post-test probability (NPP). In the normal population, the pooled sensitivity and specificity were 0.99 and 0.79, respectively, with favorable results of NPV and PPV (NPV = 0.97; PPV = 0.74); the DOR and AUROC were 530 and 0.90, respectively; the positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were 4.6 and 0.01, respectively; with the predefined pre-test probability of disease (PPD), the PPP and NPP were 69% and 0%, respectively. In patients with LC, the pooled sensitivity and specificity were 0.59 and 0.63, respectively; the NPV and PPV were 0.57 and 0.58, respectively; the DOR and AUROC were 2 and 0.64, respectively; the PLR and NLR were 1.6 and 0.65, respectively; with the predefined PPD, the PPP and NPP were 44% and 24%, respectively. Our diagnostic meta-analysis suggests that CRP can be considered as a biomarker for the diagnosis of HCC with high diagnostic accuracy in the normal population, whereas the diagnostic value of CRP for HCC in LC patients is extremely low.