Risk Stratification in Acute Heart Failure

We would like to congratulate Demissei et al. on their important work.1 However, may we ask that they provide some additional data in order to better put their findings into clinical perspective? First, the clinical prediction model used seems incomplete. Clinical experience as well as data from previous studies suggest that the following four routinely available variables substantially impact post-discharge events and should therefore be included in the model: blood pressure at discharge; worsening renal function during hospitalization; frailty; and the total dose of i.v. diuretics administered during the hospital stay. Second, surprisingly, the paper by Demissei et al.1 makes no reference to the two most extensively studied biomarkers in patients with acute heart failure: high-sensitivity cardiac troponin T and NT-proBNP. It is possible that these would outperform all biomarkers reported.2-5 Were the authors also able to measure these? Third, we ask that the authors please report findings in the overall cohort, including patients hospitalized for more than 17 days, who were excluded from the current analysis.