BACKGROUNDSeveral studies have examined the potential of adrenomedullin (ADM), pro-adrenomedullin (Pro-ADM), and mid-regional-pro-ADM (MR-Pro-ADM) as biomarkers for diagnosing and assessing the severity of heart failure (HF), with conflicting results. We aimed to investigate their diagnostic utility and their correlation with HF severity based on the New York Heart Association (NYHA) classification.METHODSWe searched PubMed, EMBASE, and Scopus using a predefined search string. The quality assessment of included studies was conducted using the Newcastle Ottawa Scale (NOS), and the primary outcome was the mean difference (MD) in serum levels of ADM, Pro-ADM, and MR-Pro-ADM, in addition to the area under the curve (AUC).RESULTSA total of 28 articles fulfilled our inclusion criteria and were included in our qualitative and quantitative synthesis, with a total of 15,405 subjects. Significant MD in ADM levels in HF patients vs. controls (6.024 [95 % CI 1.691, 10.356]), NYHA I vs. controls (-1.202 [95 % CI -2.111, -0.292]), NYHA IV vs. controls (-7.536 [95 % CI -12.680, -2.393]), NYHA III vs. NYHA IV (-4.438 [95 % CI -7.612, -1.263]), and NYHA I-II vs NYHA III-IV (-2.351 [95 % CI -4.361, -0.341]) were observed. Moreover, a significant MD was observed in pro-ADM levels in NYHA I-II patients vs. controls (-0.960 [95 % CI -1.479, -0.440]), NYHA III-IV vs. controls (-1.979 [95 % CI -2.958, -1.000]), and NYHA I-II vs. NYHA III-IV (-0.966 [95 % CI -1.407, -0.526]). Furthermore, MR-Pro-ADM levels were significantly different in NYHA I-II vs. NYHA III-IV (-0.428 [95 % CI -0.492, -0.365]). MR-Pro-ADM predicted HF with an AUC of 0.781 (95 % CI 0.755, 0.806).CONCLUSIONSAmong HF patients, there was a significant increase in ADM levels compared to control subjects, and these levels increased with the progression of NYHA classes. Similarly, both Pro-ADM and MR-Pro-ADM displayed higher concentrations in NYHA class III-IV HF patients compared to those in NYHA class I-II. However, MR-Pro-ADM exhibited lower accuracy in predicting HF compared to established biomarkers.