Estimated protein intake and prognosis in hospitalised heart failure: A focus on patients with and without chronic kidney disease.

Background and aims: In patients with heart failure (HF), differences in the association between estimated protein intake (PI) and prognosis in those with and without chronic kidney disease (CKD) remain to be clarified. This study aimed to investigate whether the prognostic effects of the estimated PI differ between patients with HF with and without CKD.

Methods: We included patients who required hospitalisation owing to worsening HF between 2015 and 2019 and assessed the estimated PI based on the adjusted Maroni formula using the body mass index and urinary urea nitrogen level. Patients were stratified into the higher and lower estimated PI groups according to the median value of the estimated PI at the time of admission. The primary outcome was all-cause mortality.

Results: Among the 694 enrolled patients, 286 had CKD. A lower estimated PI was independently associated with a worse nutritional status. During a median follow-up period of 17.2 months, 175 all-cause deaths occurred, including 99 and 76 in the lower and higher estimated PI groups, respectively. In the Kaplan-Meier curves, the lower estimated PI group was associated with higher overall mortality. However, lower estimated PI was significantly associated with all-cause mortality in patients without CKD, but not in those with CKD, after adjustment for covariates. A significant interaction in terms of prognostic effect was observed between the presence or absence of CKD and estimated PI.

Conclusions: Among patients with HF, prognostic effect of the estimated PI may differ between patients with and without CKD.