eGFR Slope as Predictor of Mortality in Heart Failure Patients

Figures

• 

  Figure 1. Survival probabilities of HF patients stratified by CKD history. ‘Number at risk’ refers to the number of surviving patients at the defined timepoints. The P‐values for each criterion are shown on the plots (P < 0.05 considered statistically significant). CKD, chronic kidney disease.

• 

  Figure 2. (A) eGFR trends in survivors versus non‐survivors by CKD status. Calculated linear regression lines (blue) for the mean eGFR values over time for survivors and non‐survivors, based on the absence (‘no CKD’) or presence (‘CKD’) of chronic kidney disease in the patients' medical history. White dots represent distinct eGFR values in mL/min/1.73 m2, measured over time (x axis, in years). (B) Dynamic prediction of mortality according to eGFR trends. Survival probabilities are calculated using eGFR trends in a single patient with stable (bottom mid), increasing (bottom right) and decreasing eGFR (bottom left). On the left‐hand side of each plot (in blue), each asterisk represents an eGFR measurement and the line represents the eGFR trajectory modelled over time. On the right‐hand side of each plot (in red), the joint model updates the survival probability (and 95% CI) over 12 months in an individual patient with stable, increasing and decreasing eGFR.