The present meta-analysis examined the diagnostic value of novel biomarkers for heart failure (HF), including copeptin, galectin-3, hs-cTnT, MR-proANP, MR-proADM, and ST2. English (EMBASE, Cochrane, and PubMed) and Chinese (Wanfang data, CNKI, SinoMed) databases were searched to identify suitable studies that were published before 1December 2016. Data were extracted using standard forms. Pooled diagnostic statistics were calculated using DerSimonian-Laird random-effects models. The analysis comprised 45 studies. The pooled sensitivities of all biomarkers were 0.80-0.86, along with pooled specificities of 0.60-0.82, positive predictive values (PPVs) of 0.52-0.80, and negative predictive values (NPVs) of 0.70-0.87. Among them, hs-cTnT had the highest sensitivity (0.86 [95% CI: 0.84-0.88]), specificity (0.82 [95% CI: 0.79-0.84]), PPV (0.80 [95% CI: 0.77-0.83]), and NPV (0.87 [95% CI: 0.85-0.89]), while MR-proADM had the lowest sensitivity (0.80 [95% CI: 0.75-0.84]), specificity (0.60 [95% CI: 0.56-0.64]), and PPV (0.52 [95% CI: 0.47-0.56]). Copeptin had the lowest NPV (0.70 [95% CI: 0.66-0.74]). The positive likelihood ratio (LR+) of all biomarkers ranged from 1.97 to 3.21, and the negative likelihood ratio (LR-) from 0.20 to 0.36. MR-proADM had the lowest LR+ and highest LR-; galectin-3 had the highest LR+ and MR-proANP had the lowest LR-. The area under the curve (AUC) was as low as 0.68 for MR-proADM, while AUCs for the other biomarkers ranged from 0.83 to 0.89. The overall diagnostic accuracy of copeptin, galectin-3, hs-cTnT, MR-proANP, and ST2 was relatively good. MR-proADM had apoor capacity to confirm or exclude HF. Improving the diagnostic accuracy of HF by acombination of biomarkers could be considered in the future.