Diagnostic value of novel biomarkers for heart failure : A meta-analysis

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 1 December 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 a poor capacity to confirm or exclude HF. Improving the diagnostic accuracy of HF by a combination of biomarkers could be considered in the future.