ABSTRACT
Background
Heart failure (HF) and chronic kidney disease (CKD) are common comorbidities among patients undergoing coronary angiography. Both conditions are associated with increased risk of adverse cardiovascular outcomes and mortality. However, the joint prognostic impact of HF and CKD in this patient population remains unclear.
Aims
We aimed to evaluate the separate and combined associations of HF and CKD with all‐cause mortality in patients undergoing coronary angiography.
Methods
We analyzed data from the KARDIO registry, an ongoing real‐life clinical database of patients undergoing coronary angiography. Multivariable Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for all‐cause mortality.
Results
Over a median follow‐up of 5.5 years, 11,896 all‐cause deaths were recorded. In multivariable‐adjusted analyses, a history of HF was associated with a higher risk of mortality compared to no HF (HR: 2.19; 95% CI: 2.04−2.34), as was a history of CKD compared to no CKD (HR: 2.06; 95% CI: 1.94−2.19). The associations persisted on mutual adjustment for each exposure. When assessed jointly, and using patients with neither condition (No HF‐No CKD) as the reference group, the adjusted HRs (95% CI) for mortality were 2.18 (2.01−2.36) for HF only, 2.04 (1.90−2.20) for CKD only, and 3.09 (2.77−3.46) for patients with both HF and CKD. Interaction analyses revealed that the associations of HF and CKD with mortality were significantly modified by each other, with the highest risk observed among individuals with both conditions. Subgroup analyses showed consistent directions of association across most categories, although the prognostic impact of HF and CKD varied by age, sex, and cardiovascular disease history.
Conclusions
In patients undergoing coronary angiography, HF and CKD were each independently associated with a twofold increased risk of all‐cause mortality. Interaction effects suggest that the co‐occurrence of HF and CKD synergistically amplifies this risk. These findings highlight the importance of identifying and managing cardiorenal comorbidity to improve outcomes in this high‐risk population.
Keywords: angiography, chronic kidney disease, coronary angiography, heart failure, hospital register, mortality
1. Introduction
Heart failure (HF) is a cardiovascular disease (CVD) which affects over 64 million people globally and remains a major public health concern [1]. Its prevalence is rising due to aging populations and increasing rates of hypertension, diabetes, and coronary artery disease (CAD) [1]. HF is a key driver of hospitalizations and healthcare costs. It substantially increases the risk of adverse cardiovascular outcomes, including all‐cause mortality, despite improvements in treatment [2]. Chronic kidney disease (CKD) affects around 10% of adults worldwide and is similarly associated with growing public health and clinical burdens [3]. Often coexisting with other chronic diseases such as diabetes and hypertension [3, 4], CKD accelerates CVD progression and is a strong predictor of cardiovascular events and mortality [5]. Cardiovascular complications remain the leading cause of death in CKD patients [6]; it has been reported that 50% of deaths from CKD are linked to CVD [7]. This highlights the need for integrated risk assessment.
In the context of CAD, both HF and CKD are prevalent comorbid conditions that independently confer an elevated risk of adverse outcomes. Several studies have shown that patients with HF undergoing invasive coronary angiography (ICA) are at increased risk for mortality [8, 9, 10], likely due to compromised cardiac function and limited physiological reserve. Similarly, impaired renal function has been associated with increased risk of post‐procedural complications and long‐term mortality among patients undergoing ICA [11, 12, 13]. While the individual prognostic implications of HF and CKD in cardiac patients are well recognized, evidence on their combined impact is relatively sparse. Some studies have suggested that the coexistence of HF and CKD exerts a synergistic effect on cardiovascular risk, likely due to overlapping pathophysiological mechanisms such as neurohormonal activation, hemodynamic overload, and systemic inflammation [14, 15]. However, limited data are available on the interplay between HF, CKD, and all‐cause mortality in patients undergoing ICA, particularly in contemporary clinical cardiology settings.
Studying this high‐risk cohort is both clinically and scientifically justified. Patients referred for coronary angiography typically present with advanced cardiovascular risk profiles and are at elevated risk for adverse outcomes. This clinical context provides a valuable opportunity to examine the independent and combined effects of HF and CKD, particularly as the severity of underlying CAD can be objectively assessed through angiographic findings. Given the frequent coexistence of HF and CKD in this population, a better understanding of how these conditions contribute to long‐term mortality may help to inform patient management and guide future research. Using data from the ongoing multicentre Finnish coronary angiography registry—the Kardio study—we aimed to evaluate the separate and combined associations of HF and CKD with all‐cause mortality in patients undergoing coronary angiography.
2. Methods
2.1. Patient Population
This study is based on data from the Finnish KARDIO registry, a large, ongoing multicentre database that captures information on individuals undergoing invasive cardiac diagnostic and therapeutic procedures across Finland [16, 17, 18]. Initiated in 2010, the registry was established to generate high‐quality evidence to support improvements in cardiac care. It compiles detailed patient‐level information, including demographic characteristics, comorbid conditions, cardiovascular risk factors, angiographic findings, and procedures such as percutaneous coronary interventions (PCI) and coronary artery bypass grafting (CABG) [16, 17, 18]. The registry includes contributions from seven cardiology centers located in Western, Central, and Northern regions of Finland. Between January 1, 2012, and December 30, 2018, a total of 82,911 individuals aged over 17 years underwent at least one cardiac catheterization procedure, amounting to 149,028 recorded procedures during the study period. Patients included in the registry underwent coronary angiography either for initial diagnostic evaluation or for assessing the severity of previously identified CAD. A substantial proportion of cases involved presentations with acute coronary syndromes (ACS), including ST‐elevation myocardial infarction (STEMI), non‐ST‐elevation ACS (NSTEMI), and unstable angina. The present analysis is limited to patients whose angiography was not primarily indicated for valvular heart disease. After excluding such cases and those with incomplete information, 79,738 individuals with documented HF and CKD status, relevant clinical covariates, and complete outcome data were retained for the final analysis. A notable proportion of the registry data (n = 47,489) had missing values for one or more key variables. The high volume of clinical responsibilities faced by contributing cardiologists may have contributed to these gaps. Missing data were considered to be missing at random (MAR), and Supporting Information S1: Table 1 presents the frequencies of missing values across variables. In accordance with Finnish national regulations on the use of hospital quality registries for research purposes, individual informed consent was not required. Coronary interventions and related clinical decisions were guided by the treating physician's discretion. Ethical approval for the registry and for its linkage to the national death registry was granted by the National Board of Health and Welfare in Finland. This linkage was made possible through the use of each patient's unique Personal Identification Code (PIN), enabling complete follow‐up for mortality outcomes.
2.2. Clinical Data Collection
Clinical data were systematically recorded using a standardized electronic data entry platform by cardiologists at the time of or immediately following the coronary angiography procedure [16, 17, 18]. Information captured included demographic characteristics, cardiovascular risk factors, comorbidities, clinical presentation, diagnostic findings, and procedural details. Baseline demographic and clinical variables included age, sex, body mass index (BMI), smoking status, and histories of hypertension, diabetes mellitus, and dyslipidemia. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m²). Smoking status was categorized as never, former, or current smoker, with current smoking defined as regular use of tobacco products within the month preceding hospital admission. Hypertension was defined by a prior diagnosis, active use of antihypertensive medications, or elevated blood pressure on admission (systolic ≥ 140 mmHg or diastolic ≥ 90 mmHg). Dyslipidemia was defined as the use of lipid‐lowering medications or an LDL cholesterol level above 3.0 mmol/L. A positive family history of CAD was recorded if a first‐degree relative had a diagnosis of myocardial infarction or underwent coronary revascularization before age 55 in men or 65 in women. Data were also collected on clinical symptoms, including angina and functional status as defined by the New York Heart Association (NYHA) classification. Electrocardiographic findings—particularly relevant for those presenting with ACS such as STEMI—were also recorded, alongside a history of myocardial infarction, previous coronary revascularization (PCI or CABG), stroke, and medication use.
HF was identified based on a clinical diagnosis recorded in the registry, as assessed by the attending physician according to prevailing clinical guidelines. CKD was similarly recorded based on clinical diagnosis of renal impairment at the time of hospital presentation (glomerular filtration rate < 60 mL/min). Although laboratory data were available for some patients, CKD classification was primarily derived from clinician‐documented diagnosis. In‐hospital variables, including the final diagnosis, details of coronary anatomy, treatment decisions, and complications, were documented. Decisions regarding coronary interventions and post‐procedural management were guided by the discretion of the treating cardiologist, following local standard‐of‐care practices [16, 17, 18]. All clinical information was collected across seven cardiology centers participating in the KARDIO registry, a nationwide quality improvement and research initiative in Finland.
2.3. Ascertainment of Outcome
The primary outcome for this study was all‐cause mortality. Outcome data were obtained through deterministic linkage of the KARDIO registry with the Finnish National Cause of Death Register. This linkage was made possible using each patient's unique PIN, which is assigned to all Finnish citizens and permanent residents [16, 17, 18]. Follow‐up for mortality began at the time of the patient's initial hospital admission for coronary angiography and extended through to December 31, 2018. The baseline point for clinical evaluation was established at hospital discharge. Mortality status was continuously updated through integration with the national death registry, ensuring complete and accurate capture of death events throughout the follow‐up period. The data linkage was conducted centrally by the National Board of Health and Welfare in Finland. Because of this systematic and comprehensive national registry infrastructure, there was no loss to follow‐up for mortality outcomes in the study cohort.
2.4. Statistical Analysis
Descriptive statistics were used to summarize the baseline characteristics. Continuous variables were presented either as mean (with standard deviation, SD) or median (including the interquartile range, IQR), while categorical data were shown as frequencies (percentages). Missing data was multiply imputed using the aregImpute function from Hmisc R package [19, 20]. The imputation process included 20 rounds and followed the procedure we used in our earlier study [16]. The only exception was the addition of HF, CKD, arteriosclerosis obliterans (ASO), prior myocardial infarction, and stroke into the imputation model. Prevalent HF and CKD status was categorized as Yes or No. For the joint contribution of HF and CKD on mortality risk, participants were categorized into four groups: No HF‐No CKD (reference comparison); Yes HF‐No CKD; No HF‐Yes CKD; and Yes HF‐Yes CKD. Hazard ratios (HRs) with 95% confidence intervals (CIs) for the associations between the exposures and all‐cause mortality were estimated using Cox proportional hazards models with 3 progressive adjustment models: (Model 1) age and hospital (cardiology center); (Model 2) Model 1 plus smoking status, hypertension, family history of CAD, dyslipidemia, invasive procedure urgency, BMI, sex, and sex‐age interaction; and (Model 3) Model 2 plus angiographic finding and number of PCI operations. HF and CKD disease conditions were further adjusted each other in an additional multivariable model (Model 4). For the association of one exposure (HF, CKD) with mortality risk, formal tests of interaction were used to assess if the association was modified by the other exposure and vice versa. We also used interaction tests to assess statistical evidence of effect modification by relevant clinical characteristics such as age, sex, BMI, comorbidities, ASO, family history of CAD, smoking, and CAD severity (angiographic findings) for the associations of HF and CKD with all‐cause mortality. Dyslipidemia, urgency, and angiographic findings were stratified to maintain the proportional hazards assumption, as verified by analyzing scaled Schoenfeld residuals [21]. The variables of BMI and the number of PCI operations underwent P‐spline transformation to appropriately model non‐linear relationships [22]. Rubin's rules were applied to estimate the pooled HRs and corresponding 95% CIs. All analyses and graphics were carried out using R software.
3. Results
3.1. Patient Characteristics
Patient characteristics overall and by categories of the combination of HF and CKD are shown in Table 1. The majority of patients were male (61.0%) and the overall mean ± SD age was 66 ± 11 years. Participants with a history of both HF and CKD were more likely to be older, males, former smokers, have higher prevalence of comorbidities such as diabetes, hypertension, and stroke, have left main stenosis, and more likely to have had a PCI. Participants with no history of both HF and CKD were more likely to be younger, females, never smokers, have lower BMI and prevalence of comorbidities such as diabetes, hypertension, and stroke, have a prior MI, and more likely to have received medical treatment.
Table 1.
Baseline measurements by heart failure (HF) and chronic kidney disease (CKD).
| Variable | No HF‐No CKD | Yes HF‐No CKD | No HF‐Yes CKD | Yes HF‐Yes CKD | Overall | Missing (n [%]) |
|---|---|---|---|---|---|---|
| Patient, N | 69,409 | 2768 | 2828 | 824 | 79,738 | — |
| Age, years | 65.4 (10.9) | 68.0 (11.4) | 70.6 (11.7) | 72.6 (10.3) | 65.8 (11.0) | 0 (0.0) |
| Body mass index, kg/m2 | 28.1 (4.8) | 29.4 (6.2) | 28.2 (5.2) | 29.2 (6.2) | 28.1 (4.9) | 37,102 (46.5) |
| Ejection fraction, % | 60 (55−67) | 40 (32−45) | 60 (50−65) | 40 (32−47) | 60 (50−65) | 65,912 (82.7) |
| Sex, female, n (%) | 27,712 (39.9) | 969 (35.0) | 952 (33.7) | 227 (27.5) | 31,090 (39.0) | 0 (0.0) |
| Diabetes, n (%) | 20,631 (32.4) | 796 (32.9) | 1215 (47.6) | 378 (53.1) | 23,763 (33.3) | 8279 (10.4) |
| Hypertension, n (%) | 31,542 (49.5) | 1604 (66.3) | 2027 (78.5) | 570 (79.6) | 37,069 (51.7) | 8085 (10.1) |
| HF, n (%) | 0 (0.0) | 2768 (100.0) | 0 (0.0) | 824 (100.0) | 3736 (4.9) | 3357 (4.2) |
| CKD, n (%) | 0 (0.0) | 0 (0.0) | 2828 (100.0) | 824 (100.0) | 3770 (4.9) | 3342 (4.2) |
| Anticoagulation, n (%) | 4922 (9.1) | 561 (37.6) | 402 (21.1) | 205 (45.3) | 6479 (10.8) | 19,829 (24.9) |
| ASO, n (%) | 828 (3.2) | 193 (7.2) | 241 (12.7) | 147 (18.9) | 1480 (4.6) | 47,489 (59.6) |
| NYHA class | 40,184 (50.4) | |||||
| I | 3097 (9.2) | 201 (10.0) | 205 (12.1) | 42 (6.7) | 4655 (11.8) | |
| II | 8171 (24.4) | 558 (27.7) | 533 (31.5) | 123 (19.7) | 9 532 (24.1) | |
| III | 12 567 (37.5) | 516 (25.6) | 459 (27.1) | 169 (27.1) | 13,808 (34.9) | |
| IV | 9717 (29.0) | 743 (36.8) | 496 (29.3) | 290 (46.5) | 11,559 (29.2) | |
| Coronary artery dominant, n (%) | 11,549 (14.5) | |||||
| Balanced | 14,500 (24.2) | 563 (22.7) | 605 (23.4) | 155 (20.8) | 16,695 (24.5) | |
| Right | 30,346 (50.6) | 1498 (60.5) | 1454 (56.4) | 462 (62.1) | 34,908 (51.2) | |
| Left | 15,068 (25.1) | 416 (16.8) | 521 (20.2) | 127 (17.1) | 16,586 (24.3) | |
| Family history of CAD, n (%) | 25,605 (42.5) | 844 (48.7) | 875 (42.5) | 191 (44.6) | 28,630 (43.0) | 13,223 (16.6) |
| Smoking status, n (%) | 16,123 (20.2) | |||||
| Never | 47,426 (82.2) | 924 (55.8) | 1540 (74.9) | 301 (63.0) | 51,120 (80.4) | |
| Former | 3279 (5.7) | 353 (21.3) | 284 (13.8) | 117 (24.5) | 4357 (6.8) | |
| Current | 7014 (12.2) | 380 (22.9) | 232 (11.3) | 60 (12.6) | 8 138 (12.8) | |
| Dyslipidemia, n (%) | 42,663 (68.5) | 1458 (66.9) | 1847 (74.4) | 505 (77.3) | 48,222 (68.9) | 9782 (12.3) |
| Prior stroke, n (%) | 1208 (4.1) | 231 (8.5) | 184 (9.2) | 80 (10.0) | 1811 (5.0) | 43,182 (54.2) |
| Prior MI, n (%) | 44,597 (64.8) | 457 (17.1) | 1108 (39.8) | 195 (24.8) | 46,468 (60.7) | 3245 (4.1) |
| Prior PCI, n (%) | 6514 (9.4) | 184 (6.7) | 309 (11.0) | 65 (7.9) | 7217 (9.4) | 2565 (3.2) |
| Urgency, elective n (%) | 27,820 (49.5) | 1353 (67.0) | 1093 (48.4) | 449 (72.1) | 32,872 (51.1) | 15,417 (19.3) |
| Angiographic findings, n (%) | 4804 (6.0) | |||||
| < 50% | 23,676 (35.8) | 922 (33.5) | 737 (26.5) | 178 (21.8) | 26,205 (35.0) | |
| 1−3‐VD | 39,556 (59.8) | 1 561 (56.7) | 1 865 (67.0) | 543 (66.6) | 45,122 (60.2) | |
| Left main stenosis | 2548 (3.9) | 165 (6.0) | 141 (5.1) | 50 (6.1) | 3015 (4.0) | |
| Other | 356 (0.5) | 105 (3.8) | 40 (1.4) | 44 (5.4) | 592 (0.8) | |
| Treatment decision, n (%) | 2040 (2.6) | |||||
| No treatment for CAD | 4505 (6.7) | 264 (9.9) | 235 (8.6) | 59 (7.5) | 5465 (7.0) | |
| Medical treatment | 28,627 (42.3) | 1055 (39.7) | 1063 (38.9) | 289 (36.8) | 31,600 (40.7) | |
| PCI | 27,723 (40.9) | 1049 (39.5) | 1162 (42.5) | 348 (44.3) | 32,994 (42.5) | |
| CABG | 5832 (8.6) | 277 (10.4) | 244 (8.9) | 81 (10.3) | 6546 (8.4) | |
| Aortic valve surgery | 19 (0.0) | 4 (0.2) | 1 (0.0) | 2 (0.3) | 31 (0.0) | |
| Mitral valve surgery | 880 (1.3) | 1 (0.0) | 24 (0.9) | 1 (0.1) | 910 (1.2) | |
| CAD (PCI/CABG) + valve surgery | 24 (0.0) | 2 (0.1) | 0 (0.0) | 0 (0.0) | 28 (0.0) | |
| Other surgery | 86 (0.1) | 4 (0.2) | 7 (0.3) | 5 (0.6) | 111 (0.1) | |
| Hybrid operation | 13 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 13 (0.0) | |
| All cause death, n (%) | 9096 (13.1) | 695 (25.1) | 831 (29.4) | 334 (40.5) | 11,896 (14.9) | 0 (0.0) |
Note: Data are n (%), median (Q1–Q3) or mean (SD).
Abbreviations: ASO, arteriosclerosis obliterans; CABG, coronary artery by‐pass grafting; CAD, coronary artery disease; MI, myocardial infarction; NYHA, New York Heart Association class; PCI, percutaneous coronary intervention; VD, vessels disease.
3.2. Associations of HF and CKD With All‐Cause Mortality
During a median follow‐up of 5.5 years (IQR 2.5−8.6), encompassing 445,641 person‐years at risk, a total of 11,896 all‐cause deaths were recorded. Table 2 presents the HRs and 95% CIs for the associations of HF and CKD, both individually and in combination, with all‐cause mortality.
Table 2.
Separate and combined associations of heart failure and CKD with the risk of all‐cause mortality.
| Model 1 | Model 2 | Model 3 | |||||
|---|---|---|---|---|---|---|---|
| Events/total | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | |
| HF status | |||||||
| No | 10,019/72,645a | Ref. | Ref. | Ref. | |||
| Yes | 1102/3736a | 2.54 (2.37−2.72) | 1.8 × 10−136 | 2.18 (2.04−2.34) | 1.8 × 10−90 | 2.19 (2.04−2.34) | 7.9 × 10−96 |
| CKD status | |||||||
| No | 9854/72,626b | Ref. | Ref. | Ref. | |||
| Yes | 1202/3770b | 2.27 (2.14−2.41) | 8.0 × 10−148 | 2.06 (1.94−2.20) | 2.9 × 10−109 | 2.06 (1.94−2.19) | 4.6 × 10−109 |
| HF and CKD status combination | |||||||
| No HF‐No CKD | 9096/69,409c | Ref. | Ref. | Ref. | |||
| Yes HF‐No CKD | 695/2768c | 2.48 (2.30−2.69) | 2.1 × 10−100 | 2.17 (2.00−2.35) | 2.0 × 10−68 | 2.18 (2.01−2.36) | 1.2 × 10−71 |
| No HF‐Yes CKD | 831/2828c | 2.19 (2.04–2.36) | 2.6 × 10−94 | 2.04 (1.90−2.20) | 1.4 × 10−72 | 2.04 (1.90−2.20) | 2.3 × 10−72 |
| Yes HF‐Yes CKD | 334/824c | 3.78 (3.39−4.21) | 4.2 × 10−118 | 3.09 (2.77−3.45) | 1.1 × 10−81 | 3.09 (2.77−3.46) | 2.4 × 10−79 |
Note: Model 1: adjusted for age and hospital (cardiology center); Model 2: model 1 plus hypertension, diabetes, smoking, family history of coronary artery disease, sex, dyslipidemia, invasive procedure urgency, body mass index and age‐sex‐interaction; Model 3: model 2 plus angiographic finding.
Abbreviations: CI, confidence interval; CKD, chronic kidney disease; HF, heart failure; ref, reference.
Excluding patients with missing HF status;
Excluding patients with missing CKD status;
Excluding patients with missing HF and CKD status combination.
In age‐and sex‐adjusted analyses, participants with a history of HF had a significantly higher risk of mortality compared to those without HF (HR = 2.54; 95% CI: 2.37–2.72). This association remained robust after adjustment for multiple clinical and demographic covariates, including hypertension, diabetes, smoking status, family history of CAD, sex, dyslipidemia, urgency of procedure, BMI, and age‐sex interaction (adjusted HR = 2.18; 95% CI: 2.04–2.34). In Model 4, the respective HR for mortality persisted on additional adjustment for CKD 1.95 (95% CI: 1.82−2.09). The association between HF and all‐cause mortality was not significantly modified by several clinically relevant characteristics, except for age, family history of CAD, and stroke (Figure 1). The associations were stronger for younger patients (< 65 years) versus older patients (≥ 65 years), patients with a family history of CAD versus those without, and patients without a history of stroke versus those with a history (p for all < 0.05).
Figure 1.
Associations of heart failure with the risk of all‐cause mortality according to several patient level characteristics. HRs are adjusted for age, hospital (cardiology center), hypertension, diabetes, smoking, family history of CAD, sex, invasive procedure urgency, BMI, and age‐sex‐interaction. *p values are for interaction (Interaction analyses for dyslipidemia and angiographic findings were conducted by stratifying the data set, as these variables violated the proportional hazards assumption and could not be included as standard covariates in the Cox regression models. Consequently, formal interaction terms and corresponding p values could not be estimated). ASO, arteriosclerosis obliterans; BMI, body mass index; CAD, coronary artery disease; CI, confidence interval; HR, hazard ratio; MI, myocardial infarction.
Similarly, CKD was associated with increased mortality risk in both age and sex‐adjusted (HR = 2.27; 95% CI: 2.14–2.41) and addtionally adjusted models (HR = 2.06; 95% CI: 1.94–2.20). Further adjustment for angiographic findings did not materially alter these associations. Following adjustment for HF, the respective risk for CKD was 1.85 (95% CI: 1.74−1.98) (Model 4). The association between CKD and all‐cause mortality was not significantly modified by several clinically relevant characteristics, except for age, sex, family history of CAD, MI, and stroke (Figure 2). The associations were stronger for younger patients (< 65 years) versus older patients (≥ 65 years), females versus males, patients with a family history of CAD versus those without, patients with a history of MI versus those without, and patients without a history of stroke versus those with a history (p for all < 0.05).
Figure 2.
Associations of chronic kidney disease with the risk of all‐cause mortality according to several patient level characteristics. HRs are adjusted for age, hospital (cardiology center), hypertension, diabetes, smoking, family history of CAD, sex, invasive procedure urgency, BMI, and age‐sex‐interaction. *p values are for interaction (Interaction analyses for dyslipidemia and angiographic findings were conducted by stratifying the data set, as these variables violated the proportional hazards assumption and could not be included as standard covariates in the Cox regression models. Consequently, formal interaction terms and corresponding p values could not be estimated). ASO, arteriosclerosis obliterans; BMI, body mass index; CAD, coronary artery disease; CI, confidence interval; HR, hazard ratio; MI, myocardial infarction.
Interaction tests showed that the association between HF and mortality risk was significantly modified by CKD (2.18 [95% CI: 2.01−2.36] for No CKD and 3.09 [95% CI: 2.77−3.46] for Yes CKD; p for interaction = 1.1 × 10−8] and the association between CKD and mortality risk was significantly modified by HF (2.04 [95% CI: 1.90–2.20] for No CKD and 3.09 [95% CI: 2.77–3.46] for Yes CKD; p for interaction = 9.2 × 10−12].
When assessing the combined effect of HF and CKD, and using participants with neither condition (No HF‐No CKD) as the reference group, the fully adjusted HRs for all‐cause mortality were 2.14 (95% CI: 1.97−2.33) for those with HF only, 2.03 (95% CI: 1.89−2.19) for those with CKD only, and 3.07 (95% CI: 2.74−3.44) for those with both HF and CKD (Table 2).
4. Discussion
In this large cohort of patients undergoing coronary angiography, those with both HF and CKD were generally older, more likely to be male and former smokers, and had a greater burden of comorbidities and adverse angiographic features compared to individuals without either condition. When evaluated separately, both HF and CKD were independently associated with more than a twofold increased risk of all‐cause mortality, even after comprehensive adjustment for demographic, clinical, and angiographic factors. However, HF appeared to be a stronger risk indicator of mortality than CKD within the same patient population. The associations were attenuated but remained significant on mutual adjustment for each exposure. Importantly, the co‐occurrence of HF and CKD was associated with the highest risk, with an over threefold elevation in mortality compared to patients with neither condition. Interaction analyses revealed that the associations of HF and CKD with mortality were significantly modified by each other, with the highest risk observed among individuals with both conditions, indicating a synergistic effect on all‐cause mortality. In further analyses, we assessed whether key clinical characteristics modified the associations of HF and CKD with all‐cause mortality. No significant effect modification was observed for most variables, including the severity of CAD. However, the association between HF and mortality was significantly stronger among patients younger than 65 years, those with a family history of CAD, and those without a prior history of stroke. Similarly, the association between CKD and mortality was more pronounced in younger individuals, females, those with a family history of CAD, patients with a history of myocardial infarction, and those without prior stroke.
Our findings align with prior studies that have consistently demonstrated an elevated risk of mortality among patients with HF or CKD undergoing coronary angiography. HF has been associated with increased vulnerability to procedural complications and poor long‐term outcomes [8, 9, 10], while impaired renal function has similarly been linked to adverse prognoses following ICA [11, 12, 13]. The current study reinforces these established associations within a large, real‐world cohort among patients with ICA. However, evidence on the combined prognostic impact of HF and CKD in this setting has been limited. While a few studies have reported that the coexistence of HF and CKD contributes to worse outcomes in broader cardiovascular patient populations [23, 24], data specifically evaluating their joint effect in patients undergoing ICA are sparse. Our study extends the literature by demonstrating the prognostic value of the co‐occurrence of HF and CKD, with a threefold mortality risk when both conditions occur in the same patient.
Several biological mechanisms may underlie the observed associations between HF, CKD, and increased mortality risk in patients undergoing coronary angiography. Based on previous evidence, HF may be a stronger risk indicator of mortality than CKD among these patients due to its direct impact on cardiac output and end‐organ perfusion, leading to progressive hemodynamic compromise [25, 26]. In advanced HF, neurohormonal activation—including the renin‐angiotensin‐aldosterone system (RAAS) and sympathetic nervous system—drives systemic inflammation, fluid overload, and arrhythmogenesis [27, 28], all of which contribute to adverse cardiovascular events and death. Additionally, HF is frequently accompanied by a pro‐thrombotic state [29] and impaired response to medical therapy [30], which may further elevate mortality risk. The presence of both HF and CKD likely reflects a more advanced stage of systemic disease, with overlapping and mutually reinforcing pathophysiological processes. These include volume overload with fluid retention, oxidative stress, endothelial dysfunction, and chronic low‐grade inflammation [31, 32, 33, 34], which accelerate CVD progression. Moreover, impaired renal clearance in CKD can lead to the accumulation of cardiotoxic metabolites and suboptimal pharmacologic dosing, further worsening cardiac function. Together, these interactions create a vicious cycle—often referred to as the cardiorenal syndrome [32, 35]—that amplifies the risk of adverse outcomes beyond the individual contributions of each condition. Vascular calcification in CKD results in the pathological deposition of calcium phosphate crystals within both the intimal and medial layers of the arterial wall, contributing to the progression of atherosclerosis [7]. Evidence indicates that cardiovascular calcification in patients with CKD is more severe, widespread and progresses more rapidly than in individuals without renal disease [36]. Moderate to severe cardiovascular calcification commonly affects the aorta, cardiac valves, and peripheral arteries in CKD. Intimal calcification is associated with atherosclerosis, while both intimal and medial calcifications are frequently observed in this population [37]. Consequently, widespread and extensive coronary atherosclerosis is a hallmark feature in patients with CKD [7]. The subgroup analyses demonstrated that the associations of HF and CKD with all‐cause mortality were directionally consistent across all examined subgroups, reinforcing the robustness of the overall findings. However, the results also indicated that the prognostic impact of these conditions may differ by certain clinical characteristics. For instance, the stronger associations observed among younger patients may reflect the lower competing risk of non‐cardiovascular mortality in this group, thereby amplifying the relative contribution of HF and CKD to mortality. The more pronounced effects in those with a family history of CAD or prior MI could suggest underlying genetic susceptibility or greater cumulative vascular burden. Additionally, the stronger association of CKD with mortality in women may be related to sex‐specific differences in renal pathophysiology, cardiovascular risk profiles, or disparities in healthcare utilization and disease recognition. These observations warrant further investigation and may have implications for identifying higher‐risk patient subgroups.
These findings highlight the need for increased clinical vigilance and comprehensive risk stratification in patients undergoing coronary angiography, particularly those with HF, CKD, or both. The substantially elevated mortality risk associated with the coexistence of HF and CKD underscores the importance of early identification, integrated management, and close follow‐up of this high‐risk subgroup. Tailored therapeutic strategies addressing both cardiac and renal dysfunction may be essential to improving outcomes in this patient population.
4.1. Strength and Limitations
A key strength of this study is the use of data from the KARDIO registry, a large, multicentre, prospectively maintained clinical database that captures real‐world information on patients undergoing ICA across Finland′s publicly funded healthcare system [16].
The registry′s broad geographic coverage, substantial sample size, and inclusion of patients with diverse clinical characteristics enhance the generalizability of the findings to routine clinical practice. The availability of detailed data on demographics, comorbidities, lifestyle factors, and angiographic findings allowed for robust adjustment for potential confounders. However, the study also has limitations. As an observational analysis, causal inferences cannot be made. The outcome was limited to all‐cause mortality due to the unavailability of cause‐specific mortality data. As such, we were unable to distinguish whether deaths were attributable to cardiovascular causes, renal complications, or other comorbid conditions. This limits our ability to explore disease‐specific mechanisms and may underestimate or overestimate the direct contributions of HF and CKD to cardiovascular mortality. Nevertheless, all‐cause mortality remains a clinically meaningful and unbiased outcome that captures the overall burden of these comorbidities in this high‐risk population. Although we adjusted for several important covariates, residual confounding remains possible because we were unable to adjust for certain clinically relevant confounders such as specific medical treatments (e.g., revascularization procedures or pharmacotherapy), lifestyle behaviors (e.g., physical activity), socioeconomic status, and cardiorespiratory fitness, due to the lack of availability in the registry [38, 39, 40]. Given the high proportion of missingness for left ventricular ejection fraction, we did not include this variable in the imputation model or the primary multivariable analyses. Additionally, data on the use of contemporary treatments for HF and CKD, including sodium–glucose co‐transporter 2 inhibitors, were not available, which may influence the observed associations. Despite these limitations, the consistency of our findings with existing evidence supports their validity and relevance to clinical practice.
5. Conclusions
In this large, contemporary cohort of patients undergoing coronary angiography, HF and CKD were each independently associated with a more than twofold increased risk of all‐cause mortality. Their coexistence conferred an even greater risk, with evidence of a statistically significant interaction suggesting synergistic effects on mortality. These results highlight the prognostic significance of cardiorenal comorbidity in this high‐risk population and emphasize the need for integrated clinical approaches that address both HF and CKD to improve long‐term outcomes following coronary angiography.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supplementary Table 1: Baseline measurements by heart failure and kidney failure missingness.
Acknowledgments
We thank the staff of all hospitals (Tampere, Jyväskylä, Oulu, Rovaniemi, Kajaani, Kokkola, Pori) which have been collecting data for the KARDIO register. Open access publishing facilitated by Ita‐Suomen yliopisto, as part of the Wiley ‐ FinELib agreement.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Savarese G., Becher P. M., Lund L. H., Seferovic P., Rosano G. M. C., and Coats A. J. S., “Global Burden of Heart Failure: A Comprehensive and Updated Review of Epidemiology,” Cardiovascular Research 118, no. 17 (2023): 3272–3287. [DOI] [PubMed] [Google Scholar]
- 2. Yancy C. W., Jessup M., Bozkurt B., et al., “2013 ACCF/AHA Guideline for the Management of Heart Failure,” Journal of the American College of Cardiology 62, no. 16 (2013): e147–e239. [DOI] [PubMed] [Google Scholar]
- 3. Kovesdy C. P., “Epidemiology of Chronic Kidney Disease: An Update 2022,” Kidney International Supplements (2011 12, no. 1 (2022): 7–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kazancioğlu R., “Risk Factors for Chronic Kidney Disease: An Update,” Kidney International Supplements 3, no. 4 (2013): 368–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Zoccali C., Mallamaci F., Adamczak M., et al., “Cardiovascular Complications in Chronic Kidney Disease: A Review From the European Renal and Cardiovascular Medicine Working Group of the European Renal Association,” Cardiovascular Research 119, no. 11 (2023): 2017–2032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sarnak M. J., Levey A. S., Schoolwerth A. C., et al., “Kidney Disease as a Risk Factor for Development of Cardiovascular Disease: A Statement From the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention,” Circulation 108, no. 17 (2003): 2154–2169. [DOI] [PubMed] [Google Scholar]
- 7. Dube P., DeRiso A., Patel M., et al., “Vascular Calcification in Chronic Kidney Disease: Diversity in the Vessel Wall,” Biomedicines 9, no. 4 (2021): 404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Anker N., Olesen K. K. W., Thrane P. G., et al., “Coronary Artery Disease Doubles Excess Mortality in Patients With Heart Failure With Reduced Ejection Fraction,” Journal of the American Heart Association 14, no. 9 (2025): e037915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Nielsen R. R., Pryds K., Olesen K. K. W., et al., “Coronary Artery Disease is a Stronger Predictor of All‐Cause Mortality Than Left Ventricular Ejection Fraction Among Patients With Newly Diagnosed Heart Failure: Insights From the WDHR,” Journal of the American Heart Association 13, no. 14 (2024): e9771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Bollano E., Redfors B., Rawshani A., et al., “Temporal Trends in Characteristics and Outcome of Heart Failure Patients With and Without Significant Coronary Artery Disease,” ESC Heart Failure 9, no. 3 (2022): 1812–1822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Weight N., Moledina S., Ullah M., et al., “Impact of Chronic Kidney Disease on the Processes of Care and Long‐Term Mortality of Non‐ST‐Segment‐Elevation Myocardial Infarction: A Nationwide Cohort Study and Long‐Term Follow‐Up,” Journal of the American Heart Association 13, no. 16 (2024): e032671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Zeng G., Yuan D., Wang P., et al., “Mild Renal Function Impairment and Long‐Term Outcomes in Patients With Three‐Vessel Coronary Artery Disease: A Cohort Study,” Cardiorenal Medicine 13, no. 1 (2023): 354–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Limpijankit T., Chandavimol M., Srimahachota S., et al., “Dose‐Dependent Effect of Impaired Renal Function on All‐Cause Mortality in Patients Following Percutaneous Coronary Intervention,” Clinical Cardiology 45, no. 8 (2022): 882–891. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ramirez‐Guerrero G., Ronco C., and Reis T., “Cardiorenal Syndrome and Inflammation: A Forgotten Frontier Resolved by Sorbents?,” Cardiorenal Medicine 14, no. 1 (2024): 454–458. [DOI] [PubMed] [Google Scholar]
- 15. Ndumele C. E., Rangaswami J., Chow S. L., et al., “Cardiovascular‐Kidney‐Metabolic Health: A Presidential Advisory From the American Heart Association,” Circulation 148, no. 20 (2023): 1606–1635. [DOI] [PubMed] [Google Scholar]
- 16. Laukkanen J. A., Kunutsor S. K., Hernesniemi J., et al., “Underweight and Obesity are Related to Higher Mortality in Patients Undergoing Coronary Angiography: The Kardio Invasive Cardiology Register Study,” Catheterization and Cardiovascular Interventions: Official Journal of the Society for Cardiac Angiography & Interventions 100, no. 7 (2022): 1242–1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Laukkanen J. A., Kunutsor S. K., Immonen J., et al., “Diabetes and Mortality Risk in Patients Undergoing Coronary Angiography: The KARDIO Study,” Catheterization and cArdiovascular Interventions: Official Journal of the Society for Cardiac Angiography & Interventions 104, no. 4 (2024): 743–750. [DOI] [PubMed] [Google Scholar]
- 18. Laukkanen J. A., Pohjonen T., Kurl S., and Kunutsor S. K., “Associations of High‐Density Lipoprotein Cholesterol Levels With the Presence and Severity of Coronary Atherosclerosis,” Atherosclerosis (2024): 118629. [DOI] [PubMed] [Google Scholar]
- 19. Frank E. H. J.. Hmisc: Harrell Miscellaneous. R package version 4.5‐0, https://CRAN.R-project.org/package=Hmisc.
- 20. Rubin D. B.. Multiple Imputation for Nonresponse in Surveys. New York: John Wiley and Sons. 1987
- 21. Therneau T.. A Package for Survival Analysis in S. version 2.38. 2015, https://CRAN.R-project.org/package=survival.
- 22. Eilers P. H. C. and Marx B. D., “Flexible Smoothing With B‐Splines and Penalties,” Statistical Science 11 (1996): 89–102. [Google Scholar]
- 23. Patel R. N., Sharma A., Prasad A., and Bansal S., “Heart Failure With Preserved Ejection Fraction With CKD: A Narrative Review of a Multispecialty Disorder,” Kidney Medicine 5, no. 12 (2023): 100705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Yu A. S., Pak K. J., Zhou H., et al., “All‐Cause and Cardiovascular‐Related Mortality in CKD Patients With and Without Heart Failure: A Population‐Based Cohort Study in Kaiser Permanente Southern California,” Kidney Medicine 5, no. 5 (2023): 100624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Hsu S., Fang J. C., and Borlaug B. A., “Hemodynamics for the Heart Failure Clinician: A State‐of‐the‐Art Review,” Journal of Cardiac Failure 28, no. 1 (2022): 133–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Harjola V. P., Mullens W., Banaszewski M., et al., “Organ Dysfunction, Injury and Failure in Acute Heart Failure: From Pathophysiology to Diagnosis and Management. A Review on Behalf of the Acute Heart Failure Committee of the Heart Failure Association (HFA) of the European Society of Cardiology (ESC),” European Journal of Heart Failure 19, no. 7 (2017): 821–836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Hartupee J. and Mann D. L., “Neurohormonal Activation in Heart Failure With Reduced Ejection Fraction,” Nature Reviews Cardiology 14, no. 1 (2017): 30–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Xanthopoulos A., Skoularigis J., and Triposkiadis F., “The Neurohormonal Overactivity Syndrome in Heart Failure,” Life 13, no. 1 (2023): 250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Shantsila E. and Lip G. Y. H., “Thrombotic Complications in Heart Failure: An Underappreciated Challenge,” Circulation 130, no. 5 (2014): 387–389. [DOI] [PubMed] [Google Scholar]
- 30. Bozkurt B., “Nonresponse to Heart Failure Therapy,” JACC: Heart Failure 11, no. 6 (2023): 729–732. [DOI] [PubMed] [Google Scholar]
- 31. Savira F., Magaye R., Liew D., et al., “Cardiorenal Syndrome: Multi‐Organ Dysfunction Involving the Heart, Kidney and Vasculature,” British Journal of Pharmacology 177, no. 13 (2020): 2906–2922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kumar U., Wettersten N., and Garimella P. S., “Cardiorenal Syndrome: Pathophysiology,” Cardiology Clinics 37, no. 3 (2019): 251–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Baaten C. C. F. M. J., Vondenhoff S., and Noels H., “Endothelial Cell Dysfunction and Increased Cardiovascular Risk in Patients With Chronic Kidney Disease,” Circulation Research 132, no. 8 (2023): 970–992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Jankowski J., Floege J., Fliser D., Böhm M., and Marx N., “Cardiovascular Disease in Chronic Kidney Disease: Pathophysiological Insights and Therapeutic Options,” Circulation 143, no. 11 (2021): 1157–1172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Prastaro M., Nardi E., Paolillo S., et al., “Cardiorenal Syndrome: Pathophysiology as a Key to the Therapeutic Approach in an Under‐Diagnosed Disease,” Journal of Clinical Ultrasound 50, no. 8 (2022): 1110–1124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Sarnak M. J., “Cardiovascular Complications in Chronic Kidney Disease,” supplement, American Journal of Kidney Diseases 41, no. 5 Suppl (2003): 11–17. [DOI] [PubMed] [Google Scholar]
- 37. Chen N. X. and Moe S. M., “Vascular Calcification: Pathophysiology and Risk Factors,” Current Hypertension Reports 14, no. 3 (2012): 228–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Laukkanen J. A., Kurl S., and Salonen J. T., “Cardiorespiratory Fitness and Physical Activity as Risk Predictors of Future Atherosclerotic Cardiovascular Diseases,” Current Atherosclerosis Reports 4, no. 6 (2002): 468–476. [DOI] [PubMed] [Google Scholar]
- 39. Laukkanen J. A., Isiozor N. M., and Kunutsor S. K., “Objectively Assessed Cardiorespiratory Fitness and All‐Cause Mortality Risk,” Mayo Clinic Proceedings 97, no. 6 (2022): 1054–1073. [DOI] [PubMed] [Google Scholar]
- 40. Laukkanen J. A. and Kunutsor S. K., “Revascularization Versus Medical Therapy for the Treatment of Stable Coronary Artery Disease: A Meta‐Analysis of Contemporary Randomized Controlled Trials,” International Journal of Cardiology 324 (2021): 13–21. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1: Baseline measurements by heart failure and kidney failure missingness.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to privacy or ethical restrictions.