MR-proADM is a strong independent predictor of long-term all-cause mortality risk in patients with chronic heart failure: results from the E-INH study

F Kerwagen,F Sahiti,S Sehner,J Albert,V Cejka,N Moser,C Morbach,G Gueder,S Frantz,G Ertl,C E Angermann,S Stoerk
DOI: https://doi.org/10.1093/eurheartj/ehac544.920
IF: 39.3
2022-10-16
European Heart Journal
Abstract:Background Mid-regional proadrenomedullin (MR-proADM) is a blood biomarker indicating critical illness. Its short-term prognostic relevance has been investigated in several conditions including heart failure (HF). Yet, the long-term prognostic utility is unknown. Methods We conducted a post-hoc analysis of the Extended Interdisciplinary Network for Heart Failure (E-INH) study, which investigated the long-term effects of a HF nurse-led remote patient care program (HeartNetCare-HFTM [HNC]). Patients from nine regional centers in Germany hospitalized with HF and a left ventricular ejection fraction (LVEF) <40% were randomized into HNC vs. Usual Care. MR-proADM and other standard biomarkers for disease progression and systemic inflammation were measured from venous blood collected at study inclusion, i.e. during index hospitalization. The prognostic utility was assessed using Kaplan-Meier plots and Cox proportional hazard models, and compared with other biomarkers by ROC curves. Results From 919 out of the 1022 recruited patients (90%), baseline levels of MR-proADM were available: median MR-proADM 0.89 (quartiles 0.63, 1.28) nmol/l; mean age 68±12 years; 28% women; 45% in class III or IV of the New York Heart Association (NYHA) classification. Median LVEF was 31 (25, 37) %. Median levels of NT-proBNP, high sensitive C-reactive protein (hsCRP), tumor necrosis factor (TNF)-a, and interleukin-6 (IL-6) were 3045 (1087, 7759) pg/ml, 9.2 (3.3, 25.2) mg/l, 13.4 (10.4, 17.5) pg/ml, and 4.9 (2.0, 11.4) pg/ml, respectively. Higher levels of MR-proADM at baseline were associated with age, female sex, NYHA class, NT-proBNP, hsCRP, IL-6, and TNF-α, while there was an inverse association with LVEF. In the course of 10 years of follow-up, 691 (68%) patients died. Unadjusted MR-proADM strongly predicted all-cause death when used as a continuous variable (HR 1.31 per nmol/l, 95% CI 1.26–1.37; p<0.001) or when grouped into quartiles (HR 1.85, 95% CI 1.71–2.0; p<0.001). Adjustments for age, sex and NYHA functional class did not materially alter the strong association. Plotting quartiles of MR-proADM in a Kaplan-Meier curve (see Figure 1) confirmed this findings. As shown in Figure 2, MR-proADM had the highest area under the curve (AUC) in ROC analysis when compared to other biomarkers. Conclusion MR-proADM appears to be a strong and independent predictor for long-term all-cause mortality risk in HF with reduced ejection fraction (HFrEF). Therefore, assessing MR-proADM may contribute to better categorization of risk and tailored care. Its clinical utility needs to be investigated in future studies. Funding Acknowledgement Type of funding sources: Public grant(s) – National budget only. Main funding source(s): BMBF
cardiac & cardiovascular systems
What problem does this paper attempt to address?