Predictive risk factors for death in elderly patients after hospitalization for acute heart failure in an internal medicine unit

Link to article at PubMed

Intern Emerg Med. 2022 Apr 22. doi: 10.1007/s11739-022-02982-2. Online ahead of print.


To determine the predictive factors of mortality after hospitalization for acute heart failure (AHF) in an internal medicine department. Retrospective observational analysis conducted on 164 patients hospitalized for AHF in 2016-2017. Demographic, clinical and biological characteristics were assessed during hospitalization. The primary endpoint was the occurrence of all-cause death. Multivariate analysis was performed using the Cox model adjusted for age and renal function. The study population was mostly female (n = 106, 64.6%), elderly (82.9 years ± 10.0), with a preserved LVEF (86%). Mean Charlson comorbidity index was 6.5 ± 2.5. After a median follow-up of 17.5 months (IQR 6-38), 109 patients (65%) had died with a median time to death of 14 months (IQR 3-29). In univariate analysis, patients who died were significantly older, had lower BMI and renal function, and higher CCI and NT-proBNP levels (median of 4944 ng/l [2370-14403] versus 1740 ng/l [1119-3503], p < 0.001). In multivariate analysis, risk factors for death were lower BMI (HR 0.69, CI [0.53-0.90], p = 0.005), lower albuminemia (HR 0.77 [0.63-0.94], p = 0.009), higher ferritinemia (HR 1.38 [1.08-1.76], p = 0.010), higher uricemia (HR 1.28 [1.02-1.59], p = 0.030), higher NT-proBNP (HR 2.46 [1.65-3.67], p < 0.001) and longer hospital stay (HR 1.25 [1.05-1.49] p = 0.013). In elderly multimorbid patients, AHF prognosis appears to be influenced by nutritional criteria, including lower BMI, hypoalbuminemia, and hyperuricemia (independently of renal function). These results underline the importance of nutritional status, especially as therapeutic options are available. This consideration paves the way for further research in this field.

PMID:35460014 | DOI:10.1007/s11739-022-02982-2

Leave a Reply

Your email address will not be published.