Background: Due to shared symptoms (i.p. dyspnoea), acute heart failure (AHF) is difficult to differentiate from COPD exacerbations (AECOPD) in acutely dyspnoeic COPD patients presenting at the hospital.

Aims: Review existing evidence on biomarkers that differentiate AHF from AECOPD.

Methods: A systematic review was conducted including studies on biomarkers detecting AHF in acutely dyspnoeic COPD patients. An AI tool (ASReview) was used for title and abstract screening. Full text screening was performed by two independent reviewers. 20% of the data extracted was checked by a second reviewer, and quality was assessed by two reviewers using the QUADAS-2 tool.

Results: 4,695 abstracts entered screening, of which 142 were relevant. After full text screening, 13 studies were included, of which 12 had high risk of bias due to unclear patient selection and flow, and use of various index and reference tests.

12 blood biomarkers for AHF were evaluated. BNP had a sensitivity (Sn) of 62-100% and negative predictive values (NPV) of 88-100% (4 studies). NT-proBNP showed Sn of 73-92% and NPV of 69-92% (5 studies). One study evaluated additional markers, among which MR-proANP, copeptin, ST2, and adiponectin showed good discriminative performance (Sn 88-96%, NPV 75-91%). 4 studies evaluated imaging markers and vitals, of which 1 study showed promising results for minimum inspiratory inferior vena cava diameter (Sn 95%, NPV 96%).

Conclusions: We found little solid evidence on biomarkers that detect AHF in dyspnoeic COPD patients. BNPs are most promising, but must be interpreted together with imaging and clinical signs. Future studies should validate the found biomarkers in COPD patients and focus on study quality.