Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 1.
Published in final edited form as: JACC Heart Fail. 2025 Feb 5;13(3):467–479. doi: 10.1016/j.jchf.2024.11.007

Outcomes of KDIGO-Defined CKD in U.S. Veterans with HFpEF, HFmrEF and HFrEF

Samir Patel 1,2,*, Venkatesh K Raman 1,3,*, Charles Faselis 1,2,4, Gregg C Fonarow 5, Phillip H Lam 1,3,6, Amiya A Ahmed 7, Paul A Heidenreich 8,9, Stefan D Anker 10,11, Prakash Deedwania 1,12, Charity J Morgan 1,13, Sijian Zhang 1,2, Hans Moore 1,2,3,4, Janani Rangaswami 1,2, George Bakris 14,†, Javed Butler 15,16, Helen M Sheriff 1,2, Richard M Allman 2,17,18, Qing Zeng-Treitler 1,2,§, Wen-Chih Wu 19,20,§, Ali Ahmed 1,2,3
PMCID: PMC12844147  NIHMSID: NIHMS2102107  PMID: 39918536

Abstract

Background:

Chronic kidney disease (CKD) is defined by the Kidney Disease: Improving Global Outcomes (KDIGO) guideline as abnormal kidney structure or function, present for >3 months, with implications for health. KDIGO-defined CKD is associated with poor outcomes in patients with heart failure (HF). Less is known about whether these associations vary by left ventricular ejection fraction (EF).

Objective:

To determine prevalence and outcomes of KDIGO-defined CKD in HF with preserved, mildly reduced, and reduced EF (HFpEF, HFmrEF, and HFrEF).

Methods:

Of the 1,446,053 Veterans who had a HF diagnosis (1991–2017) in the national VA electronic health record data, 365,000 with data on EF had KDIGO-defined CKD or normal kidney function (NKF). CKD was defined as two values measured >90 days apart of estimated glomerular filtration rate (eGFR) <60 mL/min/1.73m2 (categorized into 4 eGFR stages based on the last eGFR: 45–59, 30–44, 15–29, and <15) or urinary albumin-creatinine ratio (uACR) >30mg/g (albuminuria). NKF was defined as two values measured >90 days apart of eGFR ≥60, without eGFR <60 or albuminuria for 3 years before HF diagnosis. Patients were categorized into HFpEF (EF ≥50%; n=85,855), HFmrEF (EF 41–49%; n=39,397), and HFrEF (EF ≤40%; n=139,748). Hazard ratios (HRs) and 95% CIs for 5-year all-cause mortality and HF hospitalization through December 31, 2022, associated with the five CKD groups (vs. NKF) were estimated using Cox regression models.

Results:

Among patients with HF and NKF, mortality occurred in 39%, 37% and 41%, and HF hospitalization occurred in 12%, 15% and 21% of those with HFpEF, HFmrEF and HFrEF, respectively. Compared with NKF, CKD was associated with 16%, 19% and 26% higher multivariable-adjusted risks for death, and 31%, 33% and 32% for HF hospitalization in HFpEF, HFmrEF and HFrEF, respectively. eGFR-associated risks were incrementally higher with decreasing eGFR, except for eGFR <15, likely due to initiation of dialysis during follow-up. Albuminuria was associated with 16%, 10% and 12% higher multivariable-adjusted risks for death and 29%, 30% and 24% for HF hospitalization in HFpEF, HFmrEF and HFrEF, respectively. All associations were statistically significant.

Conclusions:

These findings based on KDIGO-defined CKD and NKF provide new information about the best estimates of true prevalence and outcomes of CKD in HFpEF, HFmrEF, and HFrEF.

Keywords: Heart Failure, Ejection Fraction, Chronic Kidney Disease, KDIGO, Outcomes

Chronic kidney disease (CKD) is a risk factor for incident heart failure (HF) (1–4) and its presence in patients with HF is associated with a higher risk of poor outcomes (5–11). However, less was known about the true estimates of prevalence of CKD in HF and the impact of CKD on outcomes in patients with HF because CKD was traditionally defined using a single serum creatinine or estimated glomerular filtration rate (eGFR) value, often measured during acute HF decompensation in the hospital. The National Kidney Foundation Kidney Disease Outcomes Quality Initiative (KDOQI) and the international Kidney Disease: Improving Global Outcomes (KDIGO) recommend that CKD be defined as abnormal kidney structure or function present for >3 months, with implications for health (12–15). In 2002, KDOQI first proposed that kidney damage or decreased kidney function be persistent for ≥3 months and suggested that proteinuria as evidenced by an urinary albumin to creatinine ratio (uACR) >30 mg/g be used as the principal marker of kidney damage, and eGFR <60 mL/min/1.73m2 as the best measure of decreased kidney function in adults (14,16). In 2012, KDIGO adopted this definition of CKD (13), which was later endorsed by KDOQI (17). However, none of the major studies of CKD in HF have used the KDIGO criteria to define CKD (5–11).

Using national electronic health record (EHR) data from the Department of Veterans Affairs (VA) healthcare system, we have demonstrated that the necessary data to define CKD using KDIGO criteria which require ≥2 ambulatory serum creatinine values measured >90 days apart was available in six of ten patients with a diagnosis of HF who were not receiving kidney replacement therapy (KRT) defined as the receipt of maintenance dialysis or kidney transplantation as documented by the United States Renal Data System (USRDS). Among these patients, seven of ten had either KDIGO-defined CKD or normal kidney function (NKF) (18). We have also demonstrated that when compared with NKF, KDIGO-defined CKD was associated with a higher risk of death and HF hospitalization, which increased with CKD severity. While these estimates and associations would be expected to be generally homogeneous across left ventricular ejection fraction (EF) categories, prior studies suggest potential heterogeneity by EF. For example, in one study, CKD-associated higher risk of death in ambulatory patients with HF was incrementally higher in those with EF 35–55% and >55% than in those with EF <35%, which was statistically significant (8). In contrast, a meta-analysis reported CKD-associated higher risk of death in HFrEF but not in HFpEF (11). A recent study based on a more contemporary HF population found that the CKD-associated risk of death was similar regardless of EF (19).

EF is the basis of most prognostically and therapeutically important classification systems of HF, and accordingly national HF guidelines classify HF into HF with preserved EF (HFpEF), HF with mildly reduced EF (HFmrEF), and HF with reduced EF (HFrEF) (20). While KDIGO-defined CKD is more likely to represent chronic intrinsic kidney disease that behaves similarly in patients with HF regardless of EF, HF is also complicated by cardiorenal syndrome which can be temporary and/or functional (21,22), the nature and effect of which may vary with the neurohormonal and hemodynamic variations associated with changes in EF (23). The objective of the current study was to estimate the prevalence and outcomes of KDIGO-defined CKD in patients with HFpEF, HFmrEF, and HFrEF.

Methods

Data Source and Study Population

The study used data from the Washington DC Veterans Affairs (VA) Medical Center HF (DCVA-HF) study which is a national cohort of 1,446,053 Veterans with HF who received care in the VA healthcare system between October 1, 1999, and December 31, 2017. Using artificial intelligence approaches, the details of which have been previously described (24), we phenotyped 1,031,970 Veterans aged ≥20 years with HF who received care in the VA healthcare system between April 1, 2000, and December 31, 2017. The current study is based on the pre-phenotype population, which was used to define CKD using KDIGO criteria (18). Briefly, from the population of 1,446,053 patients, we identified 828,744 patients who were not receiving KRT and had data on ≥2 ambulatory standardized serum creatinine values measured >90 days apart (18). Of them, 584,514 had either CKD or NKF. The study was approved by the VA Central Institutional Review Board.

Study Exposure

CKD at study baseline was defined using KDIGO criteria (18). Of the 828,744 patients with data necessary to define CKD using KDIGO criteria, 218,551 met KDIGO criteria for CKD, 356,963 had NKF, and 244,230 met neither criteria. CKD was defined by 2 values measured >90 days apart of eGFR <60 or uACR >30 mg/g, in that order. NKF was defined as 2 values measured >90 days apart of eGFR ≥60 without any eGFR <60 or uACR >30 mg/g during the 3-year period before baseline. Patients with eGFR <60 were then categorized into CKD-3A (eGFR 45–59), CKD-3B (eGFR 30–44), CKD-4 (eGFR 15–29), and CKD-5 (eGFR <15) using values proximal to study baseline.

Of the 584,514 patients with CKD or NKF, 365,000 had data on EF. Data on EF were extracted by natural language processing (NLP) from echocardiogram reports and values from 365 days before and 90 days after study baseline (first mention of HF in EHR) were used as baseline EF (25). Using baseline EF cutoffs of ≥50%, 41–49%, and ≤40% (20), we defined HFpEF, HFmrEF, and HFrEF in 185,855 (50.9%), 39,397 (10.8%), and 139,748 (38.3%) patients, respectively (Figure 1). Patients with HFpEF, HFmrEF and HFrEF were then categorized into six kidney function groups: NKF, the four eGFR stages, and albuminuria (Figure 1).

Figure 1. Flowchart Displaying the Assembly of the Study Cohorts.

Figure 1.

Open in a new tab

We began by identifying 1,446,053 patients who had ICD codes for HF in VA EHR between October 1, 1999, and December 31, 2017, of whom 1,419,729 were not receiving kidney replacement therapy and 828,744 had data necessary to define CKD using KDIGO criteria. Of them, 584,514 had either KDIGO-defined CKD or normal kidney function (NKF), of whom 365,000 had data on left ventricular ejection fraction. Of them, 185,855 had HFpEF (EF ≥50%), 39,397 had HFmEF (EF 41–49%), and 139,748 had HFrEF (EF ≤40%). CKD was defined by eGFR <60 twice >90d apart or uACR >30 mg/g twice >90d apart. CKD was categorized into eGFR 45–59, 30–44, 15–29 and <15, and albuminuria. NKF was defined as eGFR ≥60 twice >90d apart, without any eGFR <60 or uACR >30. ICD codes used to identify HF: ICD-9 codes 398.91, 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 428.x and ICD-10 codes I09.81, I11.0, I13.0, I13.2, I50., I50.1, I50.20, I50.21, I50.22, I50.23, I50.3, I50.30, I50.31, I50.32, I50.33, I50.4, I50.40, I50.41, I50.42, I50.43, I50.9.

Study Covariates

Data on baseline characteristics were collected from the VA EHR. Baseline data on vital signs, laboratory values, and medications were collected from both inpatient and outpatient settings within ±90 days of study baseline. Comorbidities were defined using International Classification of Diseases (ICD) codes at or before baseline.

Study Outcomes

Study outcomes are all-cause mortality and HF hospitalization during 5 years of follow-up from study baseline, up to December 31, 2022. Information about mortality was obtained from VA Death Ascertainment File (DAF), which is over 98% concordant with National Death Index (26). Information on HF hospitalization was obtained from EHR, and was complemented by information on non-VA hospitalization obtained from VA-linked Medicare data.

Statistical Analysis

All statistical analyses were conducted separately for HFpEF, HFmrEF, and HFrEF, except for the interaction test (described below). Key baseline characteristics for NKF, the four eGFR stages, and albuminuria were estimated as number (%) or mean (SD) as appropriate and are presented in Table 1, and detailed data for HFpEF, HFmrEF, and HFrEF are presented in Supplemental Tables 1a, 1b, and 1c. Absolute standardized differences (ASD), which are not sensitive to sample size, were estimated to assess between-group imbalances in baseline characteristics (27). We generated Kaplan-Meier survival plots for 5-year all-cause mortality for the six kidney function groups. We used Cox regression models to estimate hazard ratios (HR) and 95% confidence intervals (CI) for both outcomes during five years of follow-up associated with the five CKD groups, using NKF as the reference. For non-death outcomes, time was calculated as time to first event for those who had the events and for those without the event of interest, time was defined as time to death or study end. Covariates used in the models are listed in the footnote of Tables 2 and 3. We checked for interactions to assess if the CKD-outcomes associations in HFmrEF and HFrEF are significantly different from those in HFpEF. We checked the proportional hazard assumption by visual examination of log-log plots. Heat maps were generated to display rates and risks for both outcomes associated with CKD in the three EF groups. Missing values of continuous variables were replaced with values generated using linear regression program. Considering that a diagnosis of HF was not centrally adjudicated, which is more likely to misclassify HFpEF than HFrEF, we repeated our mortality analysis in a subset of 132,472 patients with HFpEF whose HF was phenotyped using artificial intelligence approaches (24). All analytic statistical tests were 2-tailed, and a 95% CI was used as the threshold for determining statistical significance. SAS 9.4, SPSS 29, and R 4.0, all for Windows, were used for statistical analyses and graphics generation.

Table 1.

Baseline characteristics of patients with HF and KDIGO-defined CKD and normal kidney function, by EF

Mean ±SD, n (%). Normal kidney function Albuminuria
(uACR >30mg/g)		 eGFR Stage-3A
(eGFR 45–59)		 eGFR Stage-3B
(eGFR 30–44)		 eGFR Stage-4
(eGFR 15–29)		 eGFR Stage-5
(eGFR <15)
HFpEF (n=185,855) (N=114335) (N=12337) (N=22275) (N=23317) (N=11025) (N=2566)
Age, years 67.9 (±10.8) 69.1 (±9.2) 76.9 (±9.7) 77.1 (±10.1) 74.6 (±11.0) 69.9 (±11.0)
Women 3967 (3.5%) 250 (2.0%)* 636 (2.9%)* 648 (2.8%)* 317 (2.9%)* 70 (2.7%)*
African American 14108 (12.3%) 1593 (12.9%)* 3799 (17.1%) 4296 (18.4%) 2856 (25.9%) 1035 (40.3%)
LVEF,2 % 59.2 (±6.8) 58.9 (±6.5)* 59.1 (±6.7)* 59.1 (±6.6)* 59.2 (±6.6)* 59.4 (±6.6)*
eGFR, mL/min/1.73m2 87.3 (±13.8) 76.1 (±18.1) 51.7 (±4.2) 38.1 (±4.2) 23.6 (±4.2) 10.9 (±2.8)
Albuminuria 0 (0%) 12337 (100%) 2332 (10.5%) 3005 (12.9%) 1670 (15.1%) 366 (14.3%)
HF hospitalization (index) 8145 (7.1%) 1356 (11.0%) 2315 (10.4%) 2741 (11.8%) 1668 (15.1%) 336 (13.1%)
Hypertension 100533 (87.9%) 12143 (98.4%) 21531 (96.7%) 22767 (97.6%) 10885 (98.7%) 2546 (99.2%)
Coronary artery disease 68034 (59.5%) 8319 (67.4%) 15108 (67.8%) 15773 (67.6%) 7148 (64.8%) 1493 (58.2%)*
Diabetes mellitus 47088 (41.2%) 11768 (95.4%) 12609 (56.6%) 14667 (62.9%) 7663 (69.5%) 1861 (72.5%)
Serum albumin, mg/dL 3.6 (±0.7) 3.6 (±0.6) 3.5 (±0.6) 3.5 (±0.6) 3.3 (±0.6) 3.3 (±0.6)
Hemoglobin, g/L 13.0 (±2.1) 12.6 (±2.1) 12.2 (±2.0) 11.7 (±1.9) 10.9 (±1.8) 10.3 (±1.7)
ACE inhibitors 62598 (54.7%) 7954 (64.5%) 12602 (56.6%)* 12380 (53.1%)* 4861 (44.1%) 906 (35.3%)
ARBs 12047 (10.5%) 2872 (23.3%) 3607 (16.2%) 4236 (18.2%) 2178 (19.8%) 454 (17.7%)
ARNI 2189 (1.9%) 452 (3.7%) 601 (2.7%)* 706 (3%)* 365 (3.3%)* 80 (3.1%)*
Spironolactone 8671 (7.6%) 1207 (9.8%)* 1576 (7.1%)* 1657 (7.1%)* 737 (6.7%)* 81 (3.2%)
Beta blockers 75718 (66.2%) 9894 (80.2%) 16527 (74.2%) 17859 (76.6%) 9049 (82.1%) 2169 (84.5%)
HFmrEF (n=39,397) (N=25323) (N=2587) (N=4597) (N=4441) (N=1966) (N=483)
Age, years 67.7 (±10.7) 68.4 (±9.1) 76.5 (±9.6) 77.1 (±9.9) 74.3 (±10.9) 69.7 (±10.6)
Women 380 (1.5%) 22 (0.9%)* 60 (1.3%)* 65 (1.5%)* 29 (1.5%)* 8 (1.7%)*
African American 3272 (12.9%) 339 (13.1%)* 728 (15.8%)* 776 (17.5%) 438 (22.3%) 181 (37.5%)
LVEF,2 % 45.0 (±1.8) 45.1 (±1.8)* 45.1 (±1.8)* 45.0 (±1.7)* 45.0 (±1.8)* 45.1 (±1.8)*
eGFR, mL/min/1.73m2 86.8 (±13.8) 76.8 (±17.5) 51.7 (±4.2) 38.2 (±4.2) 23.8 (±4.3) 11.0 (±2.9)
Albuminuria 0 (0%) 2587 (100%) 458 (10.0%) 535 (12.0%) 254 (12.9%) 67 (13.9%)
HF hospitalization (index) 1692 (6.7%) 269 (10.4%) 475 (10.3%) 516 (11.6%) 298 (15.2%) 64 (13.3%)
Hypertension 22231 (87.8%) 2530 (97.8%) 4394 (95.6%) 4297 (96.8%) 1933 (98.3%) 480 (99.4%)
Coronary artery disease 17685 (69.8%) 1990 (76.9%) 3529 (76.8%) 3452 (77.7%) 1459 (74.2%) 313 (64.8%)
Diabetes mellitus 10351 (40.9%) 2462 (95.2%) 2539 (55.2%) 2683 (60.4%) 1323 (67.3%) 345 (71.4%)
Serum albumin, mg/dL 3.7 (±0.6) 3.6 (±0.6) 3.6 (±0.6) 3.5 (±0.6) 3.4 (±0.6) 3.3 (±0.6)
Hemoglobin, g/L 13.3 (±2.0) 12.7 (±2.1) 12.4 (±2.0) 11.8 (±1.9) 11.1 (±1.8) 10.5 (±1.7)
ACE inhibitors 16861 (66.6%) 1783 (68.9%)* 2856 (62.1%)* 2514 (56.6%) 884 (45%) 187 (38.7%)
ARBs 2879 (11.4%) 607 (23.5%) 706 (15.4%) 755 (17.0%) 336 (17.1%) 72 (14.9%)
ARNI 599 (2.4%) 114 (4.4%) 153 (3.3%)* 152 (3.4%)* 59 (3%)* 15 (3.1%)*
Spironolactone 1965 (7.8%) 284 (11%) 348 (7.6%)* 346 (7.8%)* 133 (6.8%)* 19 (3.9%)
Beta blockers 19523 (77.1%) 2223 (85.9%) 3710 (80.7%)* 3663 (82.5%) 1684 (85.7%) 417 (86.3%)
HFrEF (n=139,748) (N=93032) (N=8034) (N=15594) (N=15486) (N=6361) (N=1241)
Age, years 67.2 (±10.7) 68.8 (±9.5) 76.7 (±9.7) 77.3 (±9.9) 75.4 (±10.9) 71.1 (±10.9)
Women 1471 (1.6%) 68 (0.8%)* 203 (1.3%)* 236 (1.5%)* 96 (1.5%)* 20 (1.6%)*
African American 13443 (14.4%) 1077 (13.4%)* 2630 (16.9%)* 2611 (16.9%)* 1328 (20.9%) 435 (35.1%)
LVEF,2 % 30.3 (±8.4)* 30.9 (±8.2)* 30.8 (±8.3)* 31.0 (±8.3)* 31.1 (±8.3)* 31.9 (±8.1)
eGFR, mL/min/1.73m2 86.6 (±13.8) 76.9 (±17.7) 51.7 (±4.2) 38.3 (±4.2) 23.9 (±4.2) 11.1 (±2.9)
Albuminuria 0 (0%) 8034 (100%) 1263 (8.1%) 1490 (9.6%) 716 (11.3%) 141 (11.4%)
HF hospitalization (index) 9052 (9.7%) 1771 (11.4%) 1933 (12.5%) 1004 (15.8%) 175 (14.1%) 966 (12%)
Hypertension 22231 (87.8%) 2530 (97.8%) 4394 (95.6%) 4297 (96.8%) 1933 (98.3%) 480 (99.4%)
Coronary artery disease 17685 (69.8%) 1990 (76.9%) 3529 (76.8%) 3452 (77.7%) 1459 (74.2%) 313 (64.8%)
Diabetes mellitus 10351 (40.9%) 2462 (95.2%) 2539 (55.2%) 2683 (60.4%) 1323 (67.3%) 345 (71.4%)
Serum albumin, mg/dL 3.7 (±0.6) 3.6 (±0.6) 3.6 (±0.6) 3.5 (±0.6) 3.4 (±0.6) 3.3 (±0.6)
Hemoglobin, g/L 13.3 (±2.0) 12.7 (±2.1) 12.4 (±2.0) 11.8 (±1.9) 11.1 (±1.8) 10.5 (±1.7)
ACE inhibitors 16861 (66.6%) 1783 (68.9%)* 2856 (62.1%)* 2514 (56.6%) 884 (45%) 187 (38.7%)
ARBs 2879 (11.4%) 607 (23.5%) 706 (15.4%) 755 (17.0%) 336 (17.1%) 72 (14.9%)
ARNI 599 (2.4%) 114 (4.4%) 153 (3.3%)* 152 (3.4%)* 59 (3%)* 15 (3.1%)*
Spironolactone 1965 (7.8%) 284 (11%) 348 (7.6%)* 346 (7.8%)* 133 (6.8%)* 19 (3.9%)
Beta blockers 19523 (77.1%) 2223 (85.9%) 3710 (80.7%)* 3663 (82.5%) 1684 (85.7%) 417 (86.3%)
Open in a new tab

Due to large sample size, all p values were <0.001.

An asterisk (*) in a cell indicates an ASD of <10% which suggest adequate balance. CKD was defined by eGFR <60 mL/min/1.73m2 or uACR >30mg/g twice >90 days apart. Albuminuria was defined as ≥1 uACR >30mg/g. Normal kidney function was defined as eGFR ≥60 twice >90 days apart. CKD was categorized based on the eGFR proximal HF diagnosis.

Abbreviations: ACE, angiotensin–converting enzyme; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor/neprilysin inhibitor; ASD, absolute standardized difference; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate (ml/min/1.73m2); HF, heart failure; HFpEF, heart failure with preserved ejection fraction; KDIGO, Kidney Disease Improving Global Outcomes; LVEF, left ventricular ejection fraction; uACR, urinary albumin creatinine ratio (mg/g); VA, Veterans Affairs.

Table 2.

5–year all-cause mortality in patients with HFpEF, HFmrEF and HFrEF and KDIGO-defined CKD

Mean age, years Mean eGFR, mL/min/1.73m2 Events (%) Hazard ratio (95% confidence interval)
Unadjusted Age-adjusted MV-adjusted
HFpEF (n=185,855)
 NKF (61.5%) 67.9 (±10.8) 87.3 (±13.8) 44410 (38.8%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (38.5%) 75.0 (±10.5) 45.7 (±19.5) 41645 (58.2%) 1.73 (1.71–1.75) 1.31 (1.29–1.32) 1.16 (1.14–1.18)
Albuminuria (6.7%) 69.1 (±9.2) 76.1 (±18.1) 5650 (45.8%) 1.21 (1.18–1.25) 1.18 (1.15–1.21) 1.16 (1.12–1.20)
eGFR 45–59 (12.0%) 76.9 (±9.7) 51.7 (±4.2) 12181 (54.7%) 1.57 (1.54–1.60) 1.09 (1.07–1.12) 1.07 (1.05–1.09)
eGFR 30–44 (12.5%) 77.1 (±10.1) 38.1 (±4.2) 14412 (61.8%) 1.90 (1.87–1.94) 1.31 (1.28–1.34) 1.17 (1.15–1.20)
eGFR 15–29 (5.9%) 74.6 (±11.0) 23.6 (±4.2) 7689 (69.7%) 2.35 (2.30–2.41) 1.83 (1.78–1.87) 1.36 (1.32–1.40)
eGFR <15 (1.4%) 69.9 (±11.0) 10.9 (±2.8) 1713 (66.8%) 2.14 (2.04–2.25) 2.09 (1.99–2.19) 1.28 (1.22–1.35)
HFmrEF (n=39,397)
 NKF (64.3%) 67.7 (±10.7) 86.8 (±13.8) 9399 (37.1%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (35.7%) 74.7 (±10.4) 46.7 (±19.6) 8390 (59.6%) 1.93 (1.88–1.99) 1.44 (1.39–1.48) 1.19 (1.14–1.23)
Albuminuria (6.6%) 68.4 (±9.1) 76.8 (±17.5) 1136 (43.9%) 1.23 (1.15–1.31) 1.21 (1.14–1.29) 1.10 (1.02–1.20)
eGFR 45–59 (11.7%) 76.5 (±9.6) 51.7 (±4.2) 2564 (55.8%) 1.74 (1.67–1.82) 1.19 (1.13–1.24) 1.08 (1.03–1.13)
eGFR 30–44 (11.3%) 77.1 (±9.9) 38.2 (±4.2) 2923 (65.8%) 2.29 (2.19–2.38) 1.53 (1.47–1.60) 1.24 (1.18–1.30)*
eGFR 15–29 (5.0%) 74.3 (±10.9) 23.8 (±4.3) 1432 (72.8%) 2.74 (2.59–2.90) 2.09 (1.97–2.21) 1.42 (1.33–1.51)*
eGFR <15 (1.2%) 69.7 (±10.6) 11.0 (±2.9) 335 (69.4%) 2.44 (2.19–2.73) 2.36 (2.11–2.63) 1.28 (1.14–1.44)
HFrEF (n=139,748)
 NKF (66.6%) 67.2 (±10.7) 86.6 (±13.8) 37725 (40.6%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (33.4%) 75.2 (±10.4) 46.7 (±19.0) 30269 (64.8%) 1.99 (1.96–2.02) 1.45 (1.43–1.48) 1.26 (1.23–1.28)
Albuminuria (5.7%) 68.8 (±9.5) 76.9 (±17.7) 3997 (49.8%) 1.30 (1.26–1.35) 1.24 (1.20–1.28) 1.12 (1.07–1.17)
eGFR 45–59 (11.2%) 76.7 (±9.7) 51.7 (±4.2) 9526 (61.1%) 1.79 (1.75–1.83) 1.23 (1.20–1.26) 1.15 (1.12–1.18)*
eGFR 30–44 (11.2%) 77.3 (±9.9) 38.3 (±4.2) 10851 (70.1%) 2.28 (2.13–2.33) 1.53 (1.49–1.56) 1.31 (1.28–1.34)*
eGFR 15–29 (4.6%) 75.4 (±10.9) 23.9 (±4.2) 4952 (77.9%) 2.89 (2.81–2.98) 2.14 (2.07–2.20) 1.52 (1.47–1.57)*
eGFR <15 (0.9%) 71.1 (±10.9) 11.1 (±2.9) 943 (76.0%) 2.75 (2.58–2.94) 2.53 (2.38–2.70) 1.50 (1.40–1.60)*
Open in a new tab

HFpEF, HFmrEF, and HFrEF were defined as EF ≥50%, 41–49%, and ≤40%, respectively. CKD was defined by eGFR <60 mL/min/1.73m2 on two separate occasions >90 days apart, and among those with eGFR <60 mL/min/1.73m2 by albuminuria on two separate occasions >90 days apart. Albuminuria was defined as ≥1 uACR >30mg/g. Normal kidney function was defined as eGFR ≥60 mL/min/1.73m2 on two occasions >90 days apart, without any eGFR <60 or albuminuria during the 3–year period. Using the proximal eGFR, eGFR was categorized into 4 stages: 3A (eGFR 45–59), 3B (eGFR 30–44), 4 (eGFR 15–29) and 5 (eGFR <15). eGFR was calculated using the 2021 CKD–EPI (without race) using ambulatory serum creatinine calibrated to be traceable to an isotope dilution mass spectrometry (IDMS) reference standard.

*

These MV-adjusted risks in HFmrEF and HFrEF were statistically significantly different from those in HFpEF (p value for interaction, <0.05)

The multivariable model was adjusted for age, sex, race, and the 21 other variables: (1) left ventricular ejection fraction, (2) first HF diagnosis as principal hospital discharge diagnosis, (3) first HF diagnosis as primary outpatient encounter diagnosis, (4) albuminuria (only for the 4 eGFR groups), (5) smoking, (6) hypertension, (7) diabetes mellitus, (8) coronary artery disease, (9) atrial fibrillation, (10) stroke, (11) peripheral vascular disease, (12) asthma, (13) chronic obstructive pulmonary disease, (14) body mass index, (15) pulse, (16) systolic blood pressure, (17) diastolic blood pressure, (18) serum sodium, (19) serum potassium, (20) serum albumin, (21) hemoglobin, (22) RASIs (angiotensin–converting enzyme inhibitors, angiotensin II receptor blockers, angiotensin receptor/neprilysin inhibitor), (23) MRAs (spironolactone and eplerenone), (24) loop diuretics (furosemide, torsemide, bumetanide, and ethacrynic acid), (25) beta blockers (bisoprolol, carvedilol, metoprolol, acebutolol, atenolol, betaxolol, carteolol, labetalol, nadolol, nebivolol, penbutolol, pindolol, propranolol, timolol, and sotalol), and anti–hypertensive drugs (26) calcium channel blockers, (27) hydralazine, and (28) thiazide diuretics.

Median follow-up time for mortality among those with HFpEF, HFmrEF and HFpEF were 4.91, 5.00, and 4.77 years respectively.

Abbreviations: CI, confidence interval; CKD, chronic kidney disease; CKD-EPI, chronic kidney disease-epidemiology; eGFR, estimated glomerular filtration rate (ml/min/1.73m2); HF, heart failure; HFmrEF, HF with mildly-reduced EF; HFpEF, HF with preserved EF; HFrEF, HF with reduced EF HR, hazard ratio; KDIGO, Kidney Disease Improving Global Outcomes; MRA, mineralocorticoid receptor antagonists; MV, multivariable; NKF, normal kidney function; RASI, renin-angiotensin system inhibitors.

Table 3.

5–year HF hospitalization in patients with HFpEF, HFmrEF and HFrEF and KDIGO-defined CKD

Mean age, years Mean eGFR, mL/min/1.73m2 Events (%) Hazard ratio (95% confidence interval)
Unadjusted Age-adjusted MV-adjusted
HFpEF (n=185,855)
 NKF (61.5%) 67.9 (±10.8) 87.3 (±13.8) 13227 (11.6%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (38.5%) 75.0 (±10.5) 45.7 (±19.5) 14906 (20.8%) 2.12 (2.07–2.17) 1.84 (1.80–1.89) 1.31 (1.27–1.35)
Albuminuria (6.7%) 69.1 (±9.2) 76.1 (±18.1) 2451 (19.9%) 1.83 (1.75–1.91) 1.79 (1.71–1.87) 1.29 (1.22–1.37)
eGFR 45–59 (12.0%) 76.9 (±9.7) 51.7 (±4.2) 4235 (19.0%) 1.86 (1.79–1.92) 1.55 (1.50–1.61) 1.24 (1.19–1.29)
eGFR 30–44 (12.5%) 77.1 (±10.1) 38.1 (±4.2) 5142 (22.1%) 2.32 (2.24–2.39) 1.93 (1.86–1.99) 1.38 (1.33–1.43)
eGFR 15–29 (5.9%) 74.6 (±11.0) 23.6 (±4.2) 2635 (23.9%) 2.77 (2.66–2.89) 2.44 (2.34–2.55) 1.43 (1.36–1.50)
eGFR <15 (1.4%) 69.9 (±11.0) 10.9 (±2.8) 443 (17.3%) 1.84 (1.67–2.02) 1.81 (1.64–1.99) 0.97 (0.87–1.07)
HFmrEF (n=39,397)
 NKF (64.3%) 67.7 (±10.7) 86.8 (±13.8) 3661 (14.5%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (35.7%) 74.7 (±10.4) 46.7 (±19.6) 3477 (24.7%) 2.09 (2.00–2.20) 1.83 (1.74–1.92) 1.33 (1.25–1.41)
Albuminuria (6.6%) 68.4 (±9.1) 76.8 (±17.5) 581 (22.5%) 1.67 (1.53–1.82) 1.64 (1.50–1.79) 1.30 (1.15–1.48)
eGFR 45–59 (11.7%) 76.5 (±9.6) 51.7 (±4.2) 1003 (21.8%) 1.76 (1.65–1.89) 1.49 (1.38–1.60) 1.21 (1.12–1.30)
eGFR 30–44 (11.3%) 77.1 (±9.9) 38.2 (±4.2) 1223 (27.5%) 2.51 (2.35–2.68) 2.10 (1.38–2.24) 1.48 (1.37–1.59)
eGFR 15–29 (5.0%) 74.3 (±10.9) 23.8 (±4.3) 567 (28.8%) 2.85 (2.61–3.12) 2.52 (2.31–2.76) 1.47 (1.33–1.63)
eGFR <15 (1.2%) 69.7 (±10.6) 11.0 (±2.9) 103 (21.3%) 1.93 (1.58–2.34) 1.88 (1.55–2.29) 0.91 (0.74–1.12)
HFrEF (n=139,748)
 NKF (66.6%) 67.2 (±10.7) 86.6 (±13.8) 19231 (20.7%) 1.00 (Reference) 1.00 (Reference) 1.00 (Reference)
 CKD (33.4%) 75.2 (±10.4) 46.7 (±19.0) 12904 (27.6%) 1.64 (1.60–1.67) 1.53 (1.49–1.57) 1.32 (1.29–1.36)
Albuminuria (5.7%) 68.8 (±9.5) 76.9 (±17.7) 2100 (26.1%) 1.37 (1.31–1.44) 1.36 (1.30–1.42) 1.24 (1.16–1.32)
eGFR 45–59 (11.2%) 76.7 (±9.7) 51.7 (±4.2) 4120 (26.4%) 1.49 (1.44–1.54) 1.38 (1.34–1.43) 1.25 (1.21–1.30)
eGFR 30–44 (11.2%) 77.3 (±9.9) 38.3 (±4.2) 4456 (28.8%) 1.78 (1.73–1.84) 1.64 (1.59–1.70) 1.38 (1.33–1.44)
eGFR 15–29 (4.6%) 75.4 (±10.9) 23.9 (±4.2) 1953 (30.7%) 2.14 (2.04–2.24) 2.01 (1.92–2.11) 1.51 (1.43–1.59)
eGFR <15 (0.9%) 71.1 (±10.9) 11.1 (±2.9) 275 (22.2%) 1.49 (1.32–1.67) 1.46 (1.29–1.64) 0.95 (0.84–1.07)
Open in a new tab

HFpEF, HFmrEF, and HFrEF were defined as EF ≥50%, 41–49%, and ≤40%, respectively. CKD was defined by eGFR <60 mL/min/1.73m2 on two separate occasions >90 days apart, and among those with eGFR <60 mL/min/1.73m2 by albuminuria on two separate occasions >90 days apart. Albuminuria was defined as ≥1 uACR >30mg/g. Normal kidney function was defined as eGFR ≥60 mL/min/1.73m2 on two occasions >90 days apart, without any eGFR <60 or albuminuria during the 3–year period. Using the proximal eGFR, eGFR was categorized into 4 stages: 3A (eGFR 45–59), 3B (eGFR 30–44), 4 (eGFR 15–29) and 5 (eGFR <15). eGFR was calculated using the 2021 CKD–EPI (without race) using ambulatory serum creatinine calibrated to be traceable to an isotope dilution mass spectrometry (IDMS) reference standard. The variables used in the multivariable model are the same as those listed in Table 2.

Median follow-up time for HF hospitalization among those with HFpEF, HFmrEF and HFpEF were 4.30, 4.29, and 3.57 years respectively.

Abbreviations: CI, confidence interval; CKD, chronic kidney disease; CKD-EPI, chronic kidney disease-epidemiology; eGFR, estimated glomerular filtration rate (ml/min/1.73m2); HF, heart failure; HFmrEF, HF with mildly-reduced EF; HFpEF, HF with preserved EF; HFrEF, HF with reduced EF HR, hazard ratio; KDIGO, Kidney Disease Improving Global Outcomes; MRA, mineralocorticoid receptor antagonists; MV, multivariable; NKF, normal kidney function; RASI, renin-angiotensin system inhibitors.

Results

Baseline Characteristics

Patients with HFpEF, HFmrEF and HFrEF who had NKF had mean ages of 67.9 (±10.8), 67.7 (±10.7), 67.2 (±10.7) years, mean EF of 59 (±7), 45 (±2), and 30 (±8) %, and mean eGFR of 87 (±14), 87 (±14), and 87 (±14), respectively (Tables 1, Supplemental Table 1a, 1b, 1c). Nearly 95% of the patients with albuminuria had diabetes mellitus in all three EF groups. All between-group differences were statistically significant (p <0.001), but inconsequential differences defined as ASD <10% are marked by asterisks. The median (25 to 75 percentile) time of staging before study baseline were 71 (22–182), 65 (20–65) and 59 (20–148) days for HFrEF, HFmrEF and HFpEF, respectively.

CKD and Outcomes in HF patients with EF data

Five-year all-cause mortality occurred in 38.8%, 37.1%, and 40.6% of the patients with NKF and 58.2%, 59.6%, and 64.8% of the patients with KDIGO-defined CKD in HFpEF, HFmrEF and HFrEF, respectively (Table 2). Compared with NKF, CKD was associated with 73% (95% CI, 71–75%), 93% (95% CI, 88–99%), and 99% (95% CI, 96–102%) higher risk of death in patients with HFpEF, HFmrEF, and HFrEF, respectively. All associations were attenuated but remained significant after multivariable adjustment (Table 2). CKD was also associated with higher risks of HF hospitalization in all three EF groups (Table 3).

CKD Stages and Mortality in HFpEF, HFmrEF and HFrEF

Kaplan Meier curves for 5-year total mortality for the six kidney function groups in HFpEF, HFmrEF and HFrEF are displayed in Figure 2. Compared with NKF, CKD stage eGFR-3A was associated with 57% (95% CI, 54–60%), 74% (95% CI, 67–82%), and 79% (95% CI, 75–83%) higher risks of death in HFpEF, HFmrEF, and HFrEF, respectively, which were attenuated to 7% (95% CI, 5–9%), 8% (95% CI, 3–13%), and 15% (95% CI, 12–18%) after multivariable adjustment (Table 2; Central Illustration; Supplemental Figure). This risk was significantly greater in HFrEF than in HFpEF (interaction p, <0.001). Baseline characteristics explained 88% (57%–7%/57%), 89% (74%–8%/74%), and 81% (79%–15%/79%) of the risks in HFpEF, HFmrEF, and HFrEF, respectively (Supplemental Table 2). Associations of eGFR-3B, eGFR-4 and eGFR-5 with death are displayed in Table 2 and Central Illustration.

Figure 2. Kaplan-Meier Plots for 5-Year Mortality by KDIGO-Defined CKD in patients with HFpEF (a), HFmrEF (b), and HFrEF (c).

Figure 2.

Open in a new tab

During 5 years of follow-up, when compared with normal kidney function (NKF), KDIGO-defined CKD was associated with a higher risk of death in patients with HFpEF, HFmrEF and HFrEF. CKD was defined as GFR <60 mL/min/1.73m2 or uACR >30 mg/g, each twice, >90d apart, and was categorized into eGFR 45–59, 30–44, 15–29, and <15. NKF was defined as eGFR ≥60 twic >90d apart, without any eGFR <60 or uACR >30 during a 3-year period before baseline.

Central Illustration. Heat Map Displaying the Events and Risks of Death and HF Hospitalization During 5 Years of Follow-up, by KDIGO-Defined CKD in HFpEF, HFmrEF, and HFrEF.

Central Illustration.

Open in a new tab

Incidence and multivariable-adjusted risks for death and HF hospitalization during 5 years of follow-up in patients with HFpEF, HFmrEF and HFrEF, and KDIGO-defined CKD. CKD was defined as GFR <60 mL/min/1.73m2 or uACR >30 mg/g, each twice, >90d apart, and was categorized into eGFR 45–59, 30–44, 15–29, and <15. NKF was defined as eGFR ≥60 twice >90d apart, without any eGFR <60 or uACR >30 during a 3-year period before baseline.

Compared with NKF, albuminuria was associated with 21% (95% CI, 18–25%), 23% (95% CI, 15–31%), and 30% (95% CI, 26–35%) significantly higher risks of death in HFpEF, HFmrEF, and HFrEF, respectively, which were attenuated to 16% (95% CI, 12–20%), 10% (95% CI, 2–20%), and 12% (95% CI, 7–17%), after multivariable adjustment (Table 2; Central Illustration; Supplemental Figure). Age and other baseline characteristics explained 24% (21–16/21), 57% (23–10/23), and 60% (30–12/30) of the risks of death in HFpEF, HFmrEF, and HFrEF, respectively (Supplemental Table 1). Associations of CKD with death in 132,472 patients with phenotyped HFpEF are consistent with those based on 185,855 patients with HFpEF included in our study (Supplemental Table 3).

CKD Stages and HF Hospitalization in HFpEF, HFmrEF and HFrEF

Compared with NKF, eGFR-3A was associated with 86% (95% CI, 79–92%), 76% (95% CI, 65–89%), and 49% (95% CI, 44–54%) higher risk of HF hospitalization, which were attenuated to 24% (95% CI, 19–29%), 21% (95% CI, 12–30%), and 25% (95% CI, 21–30%) after multivariable adjustment (Table 3; Central Illustration). The risk increased in eGFR-3B and eGFR-4 in all three EF groups and a similar pattern of risk attenuation was observed after multivariable adjustment. However, compared with NKF, the risk was lower in eGFR-5 and disappeared after multivariable adjustment (Table 3; Central Illustration; Supplemental Figure). In all three EF groups, CKD-associated risk of HF hospitalization was numerically greater than that of death in all CKD categories except for eGFR-5 (Supplemental Table 4). Associations of albuminuria with HF hospitalization in the three EF groups are presented in Table 3, Central Illustration and Supplemental Figure.

Discussion

The findings from our study demonstrate that when CKD is defined using KDIGO guideline criteria, CKD is associated with generally similar higher risks of death and HF hospitalization in patients with HFpEF, HFmrEF and HFrEF, which were attenuated after multivariable adjustment for other baseline characteristics. To the best of our knowledge, this is the first study to examine the association of KDIGO-defined CKD and outcomes in patients with HFpEF, HFmrEF and HFrEF that provide the best estimates of the true prevalence and associated risks of CKD in these patients.

Worse outcomes in patients with CKD can be broadly explained by age, other baseline characteristics, and CKD-associated neurohormonal and hemodynamic changes (18,28). Findings from our study suggest that potential explanations of these associations may vary by EF and eGFR stages. Heterogeneity in baseline risk is known to influence the magnitude of benefit and harm (29). Despite a baseline (in NKF) higher death rate in HFrEF than in HFpEF, CKD-associated risk of death was greater in HFrEF. This is unlikely to be explained by age, as mean ages of patients with NKF (67.9 vs. 67.2 years) and eGFR-3A (76.9 vs. 76.7 years) were similar in HFrEF and HFpEF. When age and other baseline characteristics were adjusted for, the proportion of eGFR-3A-associated risk attributable to abnormal kidney function was numerically greater in HFrEF than in HFpEF (19% and 12%, respectively; Supplemental Table 2). A potential explanation for this might be a greater neurohormonal activation in HFrEF than in HFpEF (23,30), which is overburdened by the added neurohormonal activation associated with CKD, contributing to the higher risk of death in HFrEF (11,31–33). The risk of death associated with other eGFR stages was also significantly higher in HFrEF than in HFpEF. However, as the eGFR dropped, the proportion of risk explained by age dropped and the proportion explained by abnormal kidney function increased. In contrast, the risk associated with eGFR 15–29 that can be attributed to abnormal kidney function was similar in HFrEF and HFpEF (28% and 27%, respectively; Supplemental Table 2).

While the risk of death increased with decreasing eGFR, there was no incremental increase in risk associated with eGFR-5, likely due to a survivor cohort effect and initiation of dialysis during follow-up. Patients in the eGFR-5 group in our study were not receiving maintenance dialysis at baseline. Because most (67.1%) Veterans with eGFR-5 receive maintenance dialysis (34), patients with eGFR-5 in our study who were not receiving KRT at baseline likely had early, asymptomatic or slowly-progressing kidney failure. However, as kidney failure worsened during follow-up, most patients with eGFR-5 likely received dialysis (34,35), which attenuated the risk to null. Routine removal of fluid during maintenance dialysis may explain the elimination of the risk of HF hospitalization in patients with eGFR-5 in all three EF groups. Although precipitating causes for HF hospitalization cannot always be determined and are often multifactorial (36), these findings suggest that in patients with HF and eGFR-5, fluid retention is the primary cause of hospitalization for HF, regardless of EF. A potential explanation why the benefit of dialysis did not extend to death is that sudden cardiac death is the predominant mode of death which is unlikely to be affected by dialysis (37,38).

To the best of our knowledge, this is the first study of CKD in HF that has defined CKD using KDIGO criteria which require documented duration of abnormal eGFR and/or albuminuria for >3 months and the new information obtained by our study has important clinical implication. Our findings suggest that the actual prevalence of CKD in HF is lower and that between-EF variation in the prevalence and outcomes of CKD is smaller than previously reported in the literature. In a cohort of mostly ambulatory HF patients, we found that 38%, 36% and 33% of the patients with HFpEF, HFmrEF and HFrEF had KDIGO-defined CKD. In a predominantly hospitalized HF patients in the Swedish Heart Failure Registry, CKD defined as eGFR ≤60 was present in 56%, 48%, and 45% of the patients with HFpEF, HFmrEF and HFrEF, respectively (39). Despite the use of patients without CKD as the reference, CKD was associated with 32%, 51% and 49% higher adjusted risk of death in those with HFpEF, HFmrEF and HFrEF, respectively (39). In contrast, the corresponding risks in our study were more modest at 16%, 19% and 26% despite the use of a strictly defined NKF group that had no eGFR <60 or uACR >30 in a 3-year period before baseline. Findings from our study show that the risk of death was even lower (between 7 and 12%) in patients with eGFR 45–59 who represent nearly 4 out of 10 patients with eGFR <60. These findings suggest that while even a slight decrease in eGFR may significantly increase the risk of death, the magnitude of this risk is modest. Renin angiotensin system inhibitors (RASIs) have been shown to lower the risk of CKD progression and development of kidney failure in patients with CKD (40–43). Future prospective studies need to examine CKD progression and risk of kidney failure in patients with HF and KDIGO-defined CKD, and also examine the effectiveness of RASIs and other HF medications on clinical and kidney outcomes in these patients.

Limitations

As in any observational study, bias due to unmeasured confounders is possible. However, adjustment for some potential mediators such as hemoglobin may have attenuated the observed associations. We did not have data on serum cystatin C levels. Although eGFR based on serum cystatin C (vs. creatinine) is a better predictor of death (44–46), few defined CKD using KDGIO criteria and none were in patients with HF (47). Our study is based on patients who had HF between 1999 and 2017. It is unknown how changes in HF therapy, especially for HFrEF, during this period may have confounded CKD-outcomes associations observed in our study. RASIs improve outcomes in patients with CKD (40–43) and are associated with improved outcomes in HF (49–51). Although RASIs have been used in HF since early 1990s (52) and the management of CKD in HF has not changed substantially during 1999–2017, bias due to other changes in HF therapy is possible. The study is based on predominantly male Veterans which may limit generalizability to other populations. Another limitation of our study is that HF was defined using ICD codes which likely misclassified HFpEF. However, the consistency of the findings our sensitivity analysis in subset of HFpEF whose HF was phenotyped using artificial intelligence approaches (24) suggest that the association of CKD and outcomes observed in our study was not affected by the how HF was adjudicated. A recurrent event analysis was not performed and may have provided a more complete picture of the associations being studied. Finally, despite our attempt to capture non-VA HF hospitalizations from the CMS Medicare data, considering that CKD groups were older, it is possible that HF hospitalization is overestimated in those with CKD.

Conclusions

The findings of our study demonstrate that CKD, defined using KDIGO criteria that require that abnormal kidney structure or function be present for >3 months, is a marker of poor outcomes in patients with HF, regardless of EF. CKD-associated risk of death increased with decreasing EF, but the risk of HF hospitalization was greater and similar across the EF groups. To the best of our knowledge this is the first study of KDIGO-defined CKD in HFpEF, HFmrEF, and HFrEF that provides new information about the true estimates of the prevalence and outcomes of CKD in these patients.

Supplementary Material

Supp.

Clinical Perspectives.

Clinical Competency in Medical Knowledge:

When CKD is defined using KDIGO criteria (abnormal kidney structure or function present for >3 months with implications for health), CKD-associated risk of death was greater in HFrEF than in HFpEF, but the risk of HF hospitalization was similar across the three EF groups.

Translational Outlook:

Future studies need to examine whether CKD defined using a single eGFR is characteristically and prognostically different from CKD defined using KDIGO criteria in patients with HFpEF, HFmrEF and HFrEF.

Funding:

This work was supported by a grant from the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development Service (I01HX002422) to the Washington DC VA Medical Center. Support for CMS and USRDS Data provided by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development, Health Services Research and Development Service, VA Information Resource Center (Project Numbers SDR 02-237 and 98-004). The funding organization or sponsor played no role in the design, analysis, or interpretation of the current study or in the preparation, review, approval, or the decision to submit the manuscript for publication.

Author Conflict of Interest Disclosures:

AA, WW and QZ declares receipt of research grant funding from National Institutes of Health, Department of Veterans Affairs and Department of Defense. GLB declares receipt of consulting fees from Bayer, Janssen, KBP Biosciences, Ionis, Alnylam, Novo Nordisk, Janssen, InREGEN. JB declares the following relationships: Consultant to Abbott, American Regent, Amgen, Applied Therapeutic, AskBio, Astellas, AstraZeneca, Bayer, Boehringer Ingelheim, Boston Scientific, Bristol Myers Squibb, Cardiac Dimension, Cardiocell, Cardior, CSL Bearing, CVRx, Cytokinetics, Daxor, Edwards, Element Science, Faraday, Foundry, G3P, Innolife, Impulse Dynamics, Imbria, Inventiva, Ionis, Lexicon, Lilly, LivaNova, Janssen, Medtronics, Merck, Occlutech, Owkin, Novartis, Novo Nordisk, Pfizer, Pharmacosmos, Pharmain, Prolaio, Regeneron, Renibus, Roche, Salamandra, Sanofi, SC Pharma, Secretome, Sequana, SQ Innovation, Tenex, Tricog, Ultromics, Vifor, and Zoll. GCF declares receipt of consulting fees from Abbott, Amgen, AstraZeneca, Bayer, Cytokinetics, Egnite, Janssen, Medtronic, Merck, Novartis, Pfizer, Urovant, and payment or honoraria for lectures from Novartis.

Abbreviations

CKD

chronic kidney disease

CKD–EPI

CKD–Epidemiology Collaboration

EF

ejection fraction

eGFR

estimated glomerular filtration rate (ml/min/1.73m2)

HF

heart failure

HFmrEF

heart failure with mildly reduced ejection fraction

HFpEF

heart failure with preserved ejection fraction

HFrEF

heart failure with reduced ejection fraction

NKF

normal kidney function

uACR

urinary albumin creatinine ratio (mg/g)

VA

Veterans Affairs

Footnotes

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.

References:

  • 1.Boorsma EM, Ter Maaten JM, Voors AA, van Veldhuisen DJ. Renal Compression in Heart Failure: The Renal Tamponade Hypothesis. JACC Heart Fail 2022;10:175–183. [DOI] [PubMed] [Google Scholar]
  • 2.Bibbins-Domingo K, Chertow GM, Fried LF et al. Renal function and heart failure risk in older black and white individuals: the Health, Aging, and Body Composition Study. Arch Intern Med 2006;166:1396–402. [DOI] [PubMed] [Google Scholar]
  • 3.Ahmed A, Campbell RC. Epidemiology of chronic kidney disease in heart failure. Heart Fail Clin 2008;4:387–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bowling CB, Feller MA, Mujib M et al. Relationship between stage of kidney disease and incident heart failure in older adults. Am J Nephrol 2011;34:135–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Heywood JT, Fonarow GC, Costanzo MR et al. High prevalence of renal dysfunction and its impact on outcome in 118,465 patients hospitalized with acute decompensated heart failure: a report from the ADHERE database. J Card Fail 2007;13:422–30. [DOI] [PubMed] [Google Scholar]
  • 6.Patel RB, Fonarow GC, Greene SJ et al. Kidney Function and Outcomes in Patients Hospitalized With Heart Failure. J Am Coll Cardiol 2021;78:330–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Campbell RC, Sui X, Filippatos G et al. Association of chronic kidney disease with outcomes in chronic heart failure: a propensity-matched study. Nephrol Dial Transplant 2009;24:186–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Ahmed A, Rich MW, Sanders PW et al. Chronic kidney disease associated mortality in diastolic versus systolic heart failure: a propensity matched study. Am J Cardiol 2007;99:393–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Bibbins-Domingo K, Lin F, Vittinghoff E, Barrett-Connor E, Grady D, Shlipak MG. Renal insufficiency as an independent predictor of mortality among women with heart failure. J Am Coll Cardiol 2004;44:1593–600. [DOI] [PubMed] [Google Scholar]
  • 10.Smith GL, Lichtman JH, Bracken MB et al. Renal impairment and outcomes in heart failure: systematic review and meta-analysis. J Am Coll Cardiol 2006;47:1987–96. [DOI] [PubMed] [Google Scholar]
  • 11.Hillege HL, Nitsch D, Pfeffer MA et al. Renal function as a predictor of outcome in a broad spectrum of patients with heart failure. Circulation 2006;113:671–8. [DOI] [PubMed] [Google Scholar]
  • 12.Kidney Disease: Improving Global Outcomes CKDWG. KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int 2024;105:S117–S314. [DOI] [PubMed] [Google Scholar]
  • 13.Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013;3:1–150. [Google Scholar]
  • 14.National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002;39:S1–266. [PubMed] [Google Scholar]
  • 15.Levey AS, Becker C, Inker LA. Glomerular filtration rate and albuminuria for detection and staging of acute and chronic kidney disease in adults: a systematic review. JAMA 2015;313:837–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Levey AS, Coresh J, Balk E et al. National Kidney Foundation practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Ann Intern Med 2003;139:137–47. [DOI] [PubMed] [Google Scholar]
  • 17.Inker LA, Astor BC, Fox CH et al. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for the evaluation and management of CKD. Am J Kidney Dis 2014;63:713–35. [DOI] [PubMed] [Google Scholar]
  • 18.Patel S, Raman V, Zhang S et al. Identification and Outcomes of KDIGO-Defined Chronic Kidney Disease in 1.4 Million U.S. Veterans with Heart Failure. Euro J Heart Fail 2024;26:1251–1260. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Butler J, Packer M, Siddiqi TJ et al. Efficacy of Empagliflozin in Patients With Heart Failure Across Kidney Risk Categories. J Am Coll Cardiol 2023;81:1902–1914. [DOI] [PubMed] [Google Scholar]
  • 20.Heidenreich PA, Bozkurt B, Aguilar D et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol 2022;79:e263–e421. [DOI] [PubMed] [Google Scholar]
  • 21.Zannad F, Rossignol P. Cardiorenal Syndrome Revisited. Circulation 2018;138:929–944. [DOI] [PubMed] [Google Scholar]
  • 22.Rangaswami J, Lo KB, Vaduganathan M, Mathew RO. Eligibility for SGLT2 Inhibitors in Heart Failure Without the Race Coefficient for Kidney Function Estimation. J Am Coll Cardiol 2021;78:1669–1670. [DOI] [PubMed] [Google Scholar]
  • 23.Kitzman DW, Little WC, Brubaker PH et al. Pathophysiological characterization of isolated diastolic heart failure in comparison to systolic heart failure. JAMA 2002;288:2144–50. [DOI] [PubMed] [Google Scholar]
  • 24.Shao Y, Zhang S, Raman VK et al. Artificial intelligence approaches for phenotyping heart failure in U.S. Veterans Health Administration electronic health record. ESC Heart Fail 2024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Garvin JH, DuVall SL, South BR et al. Automated extraction of ejection fraction for quality measurement using regular expressions in Unstructured Information Management Architecture (UIMA) for heart failure. J Am Med Inform Assoc 2012;19:859–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Arnold N VHA Death Ascertainment File (DAF) and Other Data Sources for Mortality Ascertainment. 2023.
  • 27.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Rangaswami J, Bhalla V, Blair JEA et al. Cardiorenal Syndrome: Classification, Pathophysiology, Diagnosis, and Treatment Strategies: A Scientific Statement From the American Heart Association. Circulation 2019;139:e840–e878. [DOI] [PubMed] [Google Scholar]
  • 29.Rothwell PM. Treating individuals 2. Subgroup analysis in randomised controlled trials: importance, indications, and interpretation. Lancet 2005;365:176–86. [DOI] [PubMed] [Google Scholar]
  • 30.Vergaro G, Aimo A, Prontera C et al. Sympathetic and renin-angiotensin-aldosterone system activation in heart failure with preserved, mid-range and reduced ejection fraction. Int J Cardiol 2019;296:91–97. [DOI] [PubMed] [Google Scholar]
  • 31.Schlaich MP, Socratous F, Hennebry S et al. Sympathetic activation in chronic renal failure. J Am Soc Nephrol 2009;20:933–9. [DOI] [PubMed] [Google Scholar]
  • 32.Zoccali C, Mallamaci F, Parlongo S et al. Plasma norepinephrine predicts survival and incident cardiovascular events in patients with end-stage renal disease. Circulation 2002;105:1354–9. [DOI] [PubMed] [Google Scholar]
  • 33.Stella A, Castoldi G. Role of Neurohormonal Activation in the Pathogenesis of Cardiovascular Complications in Chronic Kidney Disease. In: Berbari A, Mancia G, editors. Cardiorenal Syndrome: Mechanisms, Risk and Treatment. Italy: Springer, 2010:279–290. [Google Scholar]
  • 34.Wong SPY, Hebert PL, Laundry RJ et al. Decisions about Renal Replacement Therapy in Patients with Advanced Kidney Disease in the US Department of Veterans Affairs, 2000–2011. Clin J Am Soc Nephrol 2016;11:1825–1833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.United States Renal Data System. 2021. USRDS Annual Data Report: Epidemiology of kidney disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD. [Google Scholar]
  • 36.Fonarow GC, Abraham WT, Albert NM et al. Factors identified as precipitating hospital admissions for heart failure and clinical outcomes: findings from OPTIMIZE-HF. Arch Intern Med 2008;168:847–54. [DOI] [PubMed] [Google Scholar]
  • 37.Zile MR, Gaasch WH, Anand IS et al. Mode of death in patients with heart failure and a preserved ejection fraction: results from the Irbesartan in Heart Failure With Preserved Ejection Fraction Study (I-Preserve) trial. Circulation 2010;121:1393–405. [DOI] [PubMed] [Google Scholar]
  • 38.Carson P, Anand I, O'Connor C et al. Mode of death in advanced heart failure: the Comparison of Medical, Pacing, and Defibrillation Therapies in Heart Failure (COMPANION) trial. J Am Coll Cardiol 2005;46:2329–34. [DOI] [PubMed] [Google Scholar]
  • 39.Lofman I, Szummer K, Dahlstrom U, Jernberg T, Lund LH. Associations with and prognostic impact of chronic kidney disease in heart failure with preserved, mid-range, and reduced ejection fraction. Eur J Heart Fail 2017;19:1606–1614. [DOI] [PubMed] [Google Scholar]
  • 40.Brenner BM, Cooper ME, de Zeeuw D et al. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med 2001;345:861–9. [DOI] [PubMed] [Google Scholar]
  • 41.Hou FF, Zhang X, Zhang GH et al. Efficacy and safety of benazepril for advanced chronic renal insufficiency. N Engl J Med 2006;354:131–40. [DOI] [PubMed] [Google Scholar]
  • 42.Lewis EJ, Hunsicker LG, Clarke WR et al. Renoprotective effect of the angiotensin-receptor antagonist irbesartan in patients with nephropathy due to type 2 diabetes. N Engl J Med 2001;345:851–60. [DOI] [PubMed] [Google Scholar]
  • 43.Parving HH, Lehnert H, Brochner-Mortensen J et al. The effect of irbesartan on the development of diabetic nephropathy in patients with type 2 diabetes. N Engl J Med 2001;345:870–8. [DOI] [PubMed] [Google Scholar]
  • 44.Willey JZ, Moon YP, Husain SA et al. Creatinine versus cystatin C for renal function-based mortality prediction in an elderly cohort: The Northern Manhattan Study. PLoS One 2020;15:e0226509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Gottlieb ER, Estiverne C, Tolan NV, Melanson SEF, Mendu ML. Estimated GFR With Cystatin C and Creatinine in Clinical Practice: A Retrospective Cohort Study. Kidney Med 2023;5:100600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Lees JS, Rutherford E, Stevens KI et al. Assessment of Cystatin C Level for Risk Stratification in Adults With Chronic Kidney Disease. JAMA Netw Open 2022;5:e2238300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Valente MA, Hillege HL, Navis G et al. The Chronic Kidney Disease Epidemiology Collaboration equation outperforms the Modification of Diet in Renal Disease equation for estimating glomerular filtration rate in chronic systolic heart failure. Eur J Heart Fail 2014;16:86–94. [DOI] [PubMed] [Google Scholar]
  • 48.Hou FF, Xie D, Zhang X et al. Renoprotection of Optimal Antiproteinuric Doses (ROAD) Study: a randomized controlled study of benazepril and losartan in chronic renal insufficiency. J Am Soc Nephrol 2007;18:1889–98. [DOI] [PubMed] [Google Scholar]
  • 49.Ahmed A, Fonarow GC, Zhang Y et al. Renin-angiotensin inhibition in systolic heart failure and chronic kidney disease. Am J Med 2012;125:399–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Patel S, Lam PH, Kanonidis EI et al. Renin-Angiotensin Inhibition and Outcomes in HFrEF and Advanced Kidney Disease. Am J Med 2023;136:677–686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Ahmed A, Rich MW, Zile M et al. Renin-angiotensin inhibition in diastolic heart failure and chronic kidney disease. Am J Med 2013;126:150–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Garg R, Yusuf S. Overview of randomized trials of angiotensin-converting enzyme inhibitors on mortality and morbidity in patients with heart failure. Collaborative Group on ACE Inhibitor Trials. JAMA 1995;273:1450–6. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp.
NIHMS2102107-supplement-Supp_.docx (3.3MB, docx)

RESOURCES