Supplementary Material for: Risk of hospital admissions and death in patients with heart failure and chronic kidney disease: findings from a novel multidisciplinary clinic

Introduction:
Patients with heart failure (HF) and chronic kidney disease (CKD) are often sub-optimally treated due to concerns of hyperkalaemia, declining kidney function, and hypotension. They commonly suffer from fluid overload which can lead to frequent hospitalisations and death. This research aims to determine the characteristics associated with hospital admissions and death in patients with CKD and HF.
Methods:
Consecutive patients with CKD stage 3 to 5 and HF (regardless of ejection fraction) attending a large, specialised CKD-HF clinic between 12/Sept/2019 and 11/Nov/2021 were identified and data were collected on demographic factors, renal and heart function, medications, hospitalisations, and death. Multinomial and Cox regressions determined the characteristics of patients requiring hospitalisation and their risk of death, respectively.
Results:
A total of 667 admissions were attributable to 318 patients, 201 admissions were for HF. Men were less likely than women to have been admitted to hospital for HF (RR 0.43, 95% CI 0.20, 0.94) and non-HF causes (RR 0.21, 95% CI 0.10, 0.47). A serum haemoglobin level greater than 100 g/L was associated with fewer HF and non-HF admissions compared to a serum haemoglobin less than 100 g/L (RR 0.26, 95% CI 0.09, 0.74; RR 0.17, 95% CI 0.06, 0.47). Compared to CKD stage 3, CKD stage 4 was associated with an increased risk of HF and non-HF admissions (RR 4.01, 95% CI 1.04, 15.5; RR 4.33, 95% CI 1.13, 16.5). Having a HF admission (HR 2.41, 95% CI 1.27, 4.60), HFrEF (HR 2.18, 95% CI 1.30, 3.63)), CKD stage 4 (HR 1.91, 95% CI 1.16, 3.16), and loop diuretic use (HR 2.24, 95% CI 1.14, 4.40) were associated with a significantly increased risk of death compared to people with no admissions, with HFpEF, CKD stage 3, and no diuretic use, respectively. The use of RAAS inhibitors halved the risk of death compared to non-prescribed patients (HR 0.44, 95% CI 0.27, 0.72).
Conclusion:
Hospital admissions among CKD-HF patients were common, particularly in those with lower serum haemoglobin levels and advanced CKD stage. The risk of death was higher in those with HF admissions, the presence of HFrEF, advanced CKD stage, loop diuretic use, and those not prescribed RAAS inhibitors.