Abstract
Introduction
Systemic inflammation may play a role in the progression of chronic kidney disease (CKD) and in the development of major cardiovascular events (MACE). Whether this effect is independent of traditional cardiovascular risks is unclear.
Methods
In this study, we compared the incidence of end-stage kidney disease (ESKD) (i.e., chronic dialysis or renal transplantation), MACE, and death in patients with CKD with high sensitivity C-reactive protein (hsCRP) ≥ 2 mg/l versus hsCRP < 2 mg/l, using the TRINETX platform.
Results
Overall, 163,854 subjects with CKD (defined as estimated glomerular filtration rate [eGFR] < 60 ml/min per 1.73 m2, Kidney Disease: Improving Global Outcomes [KDIGO] CKD 3–4) or albumin-to-creatinine ratio > 200 mg/g) (KDIGO A3), hsCRP results, and no concomitant infection were included. After propensity score matching, 27,580 patients with hsCRP ≥ 2 mg/l were compared with 27,580 subjects with hsCRP < 2 mg/l, well-matched for traditional cardiovascular risk factors, including blood pressure, low-density lipoprotein (LDL) cholesterol, baseline treatments, eGFR, and albuminuria. hsCRP ≥ 2 mg/l was associated with a higher risk of MACE (ischemic stroke, myocardial infarction, heart failure, and atrial and ventricular fibrillation) and death (hazard ratio [HR]: 1.941 [95% confidence interval [CI]: 1.853–2.032]) but not a higher risk of ESKD (HR: 1.271 [0.893–1.809]) versus hsCRP < 2 mg/l during follow-up. Among patients with diabetes (21,700 patients), hsCRP ≥ 2 mg/l was associated with higher risks of MACE and death as well as with a higher risk of ESKD (HR: 1.508 [0.997–2.281], P = 0.05).
Conclusion
Systemic inflammation is associated with a higher risk of MACE and death among subjects with CKD 3 and 4 or albuminuria (A3), and a higher risk of ESKD among patients with diabetes mellitus, independently of traditional risk factors.
Keywords: arrhythmia, CKD coronary artery disease, epidemiology, systemic inflammation
Graphical abstract
The estimated number of patients with CKD was 860 million subjects in 2017 and will probably increase in the future.1 This estimation holds true when CKD is defined as increased albuminuria or mildly reduced eGFR.2 In addition to the risk of ESKD, CKD is associated with a high risk of MACE, including coronary heart disease, stroke, peripheral artery disease, arrhythmias, and heart failure.3 However, traditional risk biomarkers of atherosclerotic cardiovascular disease are not optimal to predict MACE and death in this population, even after incorporation of glomerular filtration rate and albuminuria in risk prediction scores.4
Interestingly, it was recently shown that biomarkers of inflammation were significantly associated with MACE and death among patients with CKD but no cardiovascular disease from the Chronic Renal Insufficiency Cohort, and that this association remained significant after adjustment for baseline renal function.5 Moreover, in the CANTOS study, reduction of inflammation using a monoclonal antibody targeting interleukin-1β significantly reduced the risk of recurrent MACE, and this reduction was associated with the magnitude of CRP reduction.6,7 Importantly, this beneficial effect was observed despite no significant change in LDL-cholesterol. In addition, the phase 2 RESCUE trial evaluated ziltivekimab, a monoclonal antibody directed against the interleukin-6 ligand, in patients with systemic inflammation (defined as the elevation of hsCRP > 2 mg/l) and stages CKD 3 and 4.8 In these patients, ziltivekimab at 7.5 mg, 15 mg, and 30 mg every 4 weeks over 24 weeks resulted in hsCRP reduction by −77%, −88%, and −92%, respectively versus −4% on placebo. Consequently, a phase 3 trial has been organized in patients with CKD with hsCRP > 2 mg/l to assess the efficacy of ziltivekimab versus placebo on MACE and to prevent progression of renal disease.
However, whether patients with CKD and systemic inflammation have worse renal and cardiovascular outcomes than other patients with CKD, independently of treatment and comorbid conditions is presently unknown. In the present study, we compared the incidence of ESKD, MACE, and death associated with the presence of hsCRP ≥ 2 mg/l (vs. hsCRP < 2 mg/l) in patients with CKD, and specifically in the subgroup of patients with diabetes mellitus.
Methods
Patient Selection
The data used in this study came from the TriNetX network. It is a global federated health research network providing access to electronic health records, including diagnoses, procedures, medications, laboratory values, and some genomic information for approximately 151 million deidentified patients across 126 large health care organizations in 17 countries worldwide. This analysis included patients with CKD with a significant risk of renal function degradation, defined by eGFR (Modification of Diet in Renal Disease equation) between 15 and 60 ml/min per 1.73 m2 (CKD KDIGO stages 3–4) or urinary albumin-to-creatinine ratio > 200 mg/g (KDIGO A3). Among this selected population, the outcomes of patients with hsCRP ≥ 2 (cohort A) and those with hsCRP < 2 (cohort B) were compared. Patients who had a concomitant infection were excluded from the study.
The analysis process included the following 2 main steps: (i) defining the cohorts through query criteria, and (ii) setting up and running the analysis. Setting up the analysis required definitions for the index event (i.e. hsCRP levels), outcomes criteria, and the time frame. This query was run on the Global Collaborative Network with 128 health care organizations queried and 126 health care organizations responded. A total of 80 providers responded with patients. The index event only included events that occurred up to 20 years ago. The follow-up of patients was up to 6 years after inclusion.
Collected Data
The baseline parameters collected are presented in Table 1. They included demographics, race, and ethnicity; body mass index; blood pressure; cardiovascular comorbidities; acute kidney injury and CKD; diabetes mellitus; overweight and obesity; and noncardiovascular comorbidities, including among others chronic obstructive pulmonary disease, personal history of nicotine dependence, alcohol abuse, and cancers. Laboratory parameters included cholesterol levels, eGFR using creatinine-based formula (Modification of Diet in Renal Disease), albuminuria, and hemoglobin A1c. Collected medication information included presence of beta blockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, angiotensin-receptor blockers, diuretics, antilipemic agents, medications used in diabetes mellitus, anticoagulants, and antiplatelet agents.
Table 1.
Baseline characteristics of patients before and after propensity score matching
| Characteristics | Before propensity score matching |
After propensity score matching |
||||||
|---|---|---|---|---|---|---|---|---|
| hsCRP ≥ 2 |
hsCRP ≤ 2 |
P-Value | Std diff. (%) | hsCRP ≥ 2 |
hsCRP ≤ 2 |
P-Value | Std diff. (%) | |
| (n = 104,908) | (n = 29,473) | (n = 27,580) | (n = 27,580) | |||||
| Age, yr (%) | 73.0 ± 12.0 | 71.7 ± 11.4 | < 0.001 | 10.6 | 72.7 ± 11.7 | 72.0 ± 11.3 | < 0.001 | 6.1 |
| Men, n (%) | 56,281 (53.6%) | 16,012 (54.3%) | 0.039 | 1.4 | 14,923 (54.1%) | 14,972 (54.3%) | 0.675 | 0.4 |
| Systolic BP (mm Hg), mean ± SD | 126.8 ± 24.2 | 130.2 ± 23.4 | < 0.001 | 14.2 | 128 ± 24 | 130 ± 23 | < 0.001 | 9.8 |
| Diastolic BP (mm Hg), mean ± SD | 69.1 ± 14.6 | 71.4 ± 13.2 | < 0.001 | 16.6 | 70.1 ± 14.0 | 71.4 ± 13.2 | < 0.001 | 9 |
| Body mass index (kg/m2), mean ± SD | 28.9 ± 6.9 | 26.8 ± 5.4 | < 0.001 | 33.5 | 27.3 ± 6.0 | 27.0 ± 5.5 | < 0.001 | 6.2 |
| White, n (%) | 44,158 (42.1%) | 10,646 (36.1%) | < 0.001 | 12.3 | 10,430 (37.8%) | 10,609 (38.5%) | 0.117 | 1.3 |
| Black or African American, n (%) | 7409 (7.1%) | 1184 (4%) | < 0.001 | 13.3 | 1182 (4.3%) | 1183 (4.3%) | 0.983 | 0 |
| Asian, n (%) | 5439 (5.2%) | 5634 (19.1%) | < 0.001 | 43.6 | 4114 (14.9%) | 3821 (13.9%) | < 0.001 | 3 |
| Hispanic or Latino, n (%) | 3221 (3.1%) | 486 (1.6%) | < 0.001 | 9.4 | 485 (1.8%) | 484 (1.8%) | 0.974 | 0 |
| Unknown race, n (%) | 45,488 (43.4%) | 11,370 (38.6%) | < 0.001 | 9.7 | 11,218 (40.7%) | 11,336 (41.1%) | 0.307 | 0.9 |
| Comorbid conditions | ||||||||
| Hypertension, n (%) | 60,503 (57.7%) | 16,833 (57.1%) | 0.086 | 1.1 | 15,635 (56.7%) | 15,661 (56.8%) | 0.823 | 0.2 |
| Diabetes mellitus, n (%) | 34,126 (32.5%) | 9689 (32.9%) | 0.265 | 0.7 | 8883 (32.2%) | 8889 (32.2%) | 0.956 | 0 |
| Smoker, n (%) | 10525 (10%) | 1798 (6.1%) | < 0.001 | 14.5 | 1847 (6.7%) | 1786 (6.5%) | 0.295 | 0.9 |
| Overweight or obesity, n (%) | 14,039 (13.4%) | 2415 (8.2%) | < 0.001 | 16.8 | 2428 (8.8%) | 2398 (8.7%) | 0.651 | 0.4 |
| Dyslipidemia, n (%) | 44,725 (42.6%) | 13,962 (47.4%) | < 0.001 | 9.5 | 12,878 (46.7%) | 12,769 (46.3%) | 0.352 | 0.8 |
| Alcohol-related diagnoses, n (%) | 1962 (1.9%) | 310 (1.1%) | < 0.001 | 6.8 | 295 (1.1%) | 307 (1.1%) | 0.623 | 0.4 |
| Heart failure, n (%) | 26,983 (25.7%) | 5700 (19.3%) | < 0.001 | 15.3 | 5620 (20.4%) | 5542 (20.1%) | 0.408 | 0.7 |
| Coronary artery disease, n (%) | 64,981 (61.9%) | 18,680 (63.4%) | < 0.001 | 3 | 17,555 (63.7%) | 17,493 (63.4%) | 0.583 | 0.5 |
| Myocardial infarction, n (%) | 8617 (8.2%) | 2603 (8.8%) | 0.001 | 2.2 | 2415 (8.8%) | 2480 (9%) | 0.33 | 0.8 |
| Dilated cardiomyopathy, n (%) | 1643 (1.6%) | 329 (1.1%) | < 0.001 | 3.9 | 350 (1.3%) | 313 (1.1%) | 0.148 | 1.2 |
| Ischemic stroke, n (%) | 25,320 (24.1%) | 6807 (23.1%) | < 0.001 | 2.4 | 6312 (22.9%) | 6379 (23.1%) | 0.498 | 0.6 |
| Intracranial hemorrhage, n (%) | 2045 (1.9%) | 614 (2.1%) | 0.145 | 1 | 602 (2.2%) | 575 (2.1%) | 0.426 | 0.7 |
| Atrial fibrillation or flutter, n (%) | 21,952 (20.9%) | 4862 (16.5%) | < 0.001 | 11.4 | 4711 (17.1%) | 4673 (16.9%) | 0.667 | 0.4 |
| Kidney disease, n (%) | 29,734 (28.3%) | 6248 (21.2%) | < 0.001 | 16.6 | 60,42 (21.9%) | 5823 (21.1%) | 0.023 | 1.9 |
| Lung disease, n (%) | 42,446 (40.5%) | 9798 (33.2%) | < 0.001 | 15 | 9428 (34.2%) | 9304 (33.7%) | 0.265 | 0.9 |
| COPD, n (%) | 10,364 (9.9%) | 1926 (6.5%) | < 0.001 | 12.2 | 1907 (6.9%) | 1856 (6.7%) | 0.389 | 0.7 |
| Sleep apnea syndrome, n (%) | 9259 (8.8%) | 2156 (7.3%) | < 0.001 | 5.5 | 2100 (7.6%) | 2083 (7.6%) | 0.785 | 0.2 |
| Peripheral vascular disease, n (%) | 10440 (10%) | 1848 (6.3%) | < 0.001 | 13.5 | 1834 (6.6%) | 1817 (6.6%) | 0.771 | 0.2 |
| Previous cancer, n (%) | 24,016 (22.9%) | 7363 (25%) | < 0.001 | 4.9 | 6898 (25%) | 6774 (24.6%) | 0.221 | 1 |
| Anemia, n (%) | 13,485 (12.9%) | 2936 (10%) | < 0.001 | 9.1 | 2929 (10.6%) | 2813 (10.2%) | 0.106 | 1.4 |
| Biology measurements | ||||||||
| Total cholesterol (mg/dl), mean ± SD | 160.6 ± 50.2 | 164.1 ± 47.3 | < 0.001 | 7.2 | 164.5 ± 48.7 | 163.6 ± 47.7 | 0.06 | 1.9 |
| LDL cholesterol (mg/dl), mean ± SD | 89.4 ± 38.5 | 89.5 ± 37.4 | 0.846 | 0.2 | 91.0 ± 37.4 | 89.1 ± 37.7 | < 0.001 | 5.2 |
| HDL cholesterol (mg/dl), mean ± SD | 47.3 ± 17.0 | 51.6 ± 17.7 | < 0.001 | 24.4 | 50.2 ± 17.1 | 51.4 ± 17.9 | < 0.001 | 7 |
| Triglyceride (mg/dl), mean ± SD | 136.9 ± 149.1 | 131.8 ± 225.7 | < 0.001 | 2.7 | 133.6 ± 147.1 | 131.9 ± 227.0 | 0.392 | 0.9 |
| Hemoglobin A1c (%), mean ± SD | 6.5 ± 1.6 | 6.2 ± 1.4 | < 0.001 | 20.1 | 6.3 ± 1.5 | 6.2 ± 1.4 | < 0.001 | 8.4 |
| Estimated GFR (MDRD, ml/min), mean ± SD | 50.0 ± 19.3 | 53.1 ± 15.7 | < 0.001 | 17.3 | 51.9 ± 17.9 | 52.8 ± 15.9 | < 0.001 | 5.3 |
| Albuminuria (mg/g), mean ± SD | 293.0 ± 939.1 | 228.6 ± 900.9 | 0.034 | 7 | 187.7 ± 597.2 | 231.8 ± 913.8 | 0.164 | 5.7 |
| Albuminuria > 200 mg/g, n (%) | 1055 (1%) | 274 (0.9%) | 0.244 | 0.8 | 255 (0.9%) | 267 (1%) | 0.598 | 0.4 |
| hsCRP (mg/L), mean ± SD | 14.8 ± 24.0 | 1.0 ± 7.5 | < 0.001 | - | 12.8 ± 21.2 | 1.0 ± 7.8 | < 0.001 | - |
| Hemoglobin (g/dl), mean ± SD | 12.2 ± 2.4 | 13.0 ± 2.1 | < 0.001 | 37.8 | 12.6 ± 2.3 | 13.0 ± 2.1 | < 0.001 | 18.6 |
| Baseline treatments | ||||||||
| Beta blockers, n (%) | 54,391 (51.8%) | 13,079 (44.4%) | < 0.001 | 15 | 12,659 (45.9%) | 12,579 (45.6%) | 0.494 | 0.6 |
| Calcium channel blockers, n (%) | 37,339 (35.6%) | 10619 (36%) | 0.166 | 0.9 | 9771 (35.4%) | 9615 (34.9%) | 0.164 | 1.2 |
| ACE inhibitors, n (%) | 30,785 (29.3%) | 7298 (24.8%) | < 0.001 | 10.3 | 7152 (25.9%) | 7090 (25.7%) | 0.546 | 0.5 |
| Angiotensin 2 inhibitors, n (%) | 26,801 (25.5%) | 8654 (29.4%) | < 0.001 | 8.6 | 7883 (28.6%) | 7688 (27.9%) | 0.065 | 1.6 |
| Digitalis glycosides, n (%) | 4626 (4.4%) | 846 (2.9%) | < 0.001 | 8.2 | 1033 (3.7%) | 807 (2.9%) | <0.001 | 4.6 |
| Diuretics, n (%) | 49,490 (47.2%) | 10,846 (36.8%) | < 0.001 | 21.1 | 10,659 (38.6%) | 10,353 (37.5%) | 0.007 | 2.3 |
| Lipid-lowering drugs, n (%) | 59,401 (56.6%) | 16,773 (56.9%) | 0.378 | 0.6 | 15,501 (56.2%) | 15,487 (56.2%) | 0.904 | 0.1 |
| Glucose-lowering therapy, n (%) | 38,039 (36.3%) | 8663 (29.4%) | < 0.001 | 14.7 | 8225 (29.8%) | 8064 (29.2%) | 0.133 | 1.3 |
| Insulin, n (%) | 29,279 (27.9%) | 5160 (17.5%) | < 0.001 | 25 | 5284 (19.2%) | 5091 (18.5%) | 0.035 | 1.8 |
| Noninsulin glucose-lowering therapy, n (%) | 20,592 (19.6%) | 5804 (19.7%) | 0.807 | 0.2 | 5313 (19.3%) | 5218 (18.9%) | 0.303 | 0.9 |
| Metformin, n (%) | 13,453 (12.8%) | 4111 (13.9%) | < 0.001 | 3.3 | 3625 (13.1%) | 3616 (13.1%) | 0.91 | 0.1 |
| Sulfonylureas, n (%) | 7491 (7.1%) | 2022 (6.9%) | 0.098 | 1.1 | 1843 (6.7%) | 1808 (6.6%) | 0.549 | 0.5 |
| GLP-1 receptor agonists, n (%) | 1826 (1.7%) | 537 (1.8%) | 0.347 | 0.6 | 487 (1.8%) | 513 (1.9%) | 0.407 | 0.7 |
| DPP4 inhibitors, n (%) | 5383 (5.1%) | 1796 (6.1%) | < 0.001 | 4.2 | 1554 (5.6%) | 1546 (5.6%) | 0.882 | 0.1 |
| SGLT2 inhibitors, n (%) | 3997 (3.8%) | 1208 (4.1%) | 0.023 | 1.5 | 1118 (4.1%) | 1103 (4%) | 0.745 | 0.3 |
| Thiazolidinediones, n (%) | 1440 (1.4%) | 596 (2%) | < 0.001 | 5 | 426 (1.5%) | 518 (1.9%) | 0.003 | 2.6 |
| Antiplatelet therapy, n (%) | 57,773 (55.1%) | 14,945 (50.7%) | < 0.001 | 8.7 | 14,377 (52.1%) | 14,187 (51.4%) | 0.105 | 1.4 |
| Anticoagulant, n (%) | 8238 (7.9%) | 1188 (4%) | < 0.001 | 16.2 | 1261 (4.6%) | 1182 (4.3%) | 0.102 | 1.4 |
ACE, angiotensin-converting enzyme; BP, blood pressure; COPD, chronic obstructive pulmonary disease; DPP4, dipeptidyl peptidase 4; GFR, glomerular filtration rate; GLP1, glucagon-like peptide-1; HDL, high-density lipoprotein; hsCRP, high sensitivity C-reactive protein; LDL, low-density lipoprotein; MDRD, Modification of Diet in Renal Disease formula; SGLT2, sodium glucose cotransporter-2; Std diff., standard difference.
Sample Size Calculation and Propensity Score Matching
Considering that this was an observational, retrospective real-world evidence study based on the TriNetX Global Collaborative Network, no a priori sample size calculation was performed. All eligible patients meeting the predefined inclusion criteria across participating health care organizations were included to maximize statistical power and external validity.
The propensity scores were estimated using logistic regression incorporating all demographic, clinical, laboratory, and medication variables available in TriNetX (57 covariates, as listed in the Supplementary Material). Matching was performed 1:1 using greedy nearest-neighbor matching without replacement, with a caliper of 0.1 of the SD of the logit of the propensity score, as implemented in the TriNetX platform. Balance was assessed using standardized mean differences (< 10% for all variables after matching). The exact number of patients before and after matching was double-checked and corrected, where necessary (27,580 vs. 27,580 in each arm for the main analysis). Propensity score matching was performed on all collected parameters for the whole population, and separately among patients with diabetes mellitus (Supplementary Material).
Statistical Analyses and Outcome Measures
Results are presented with mean and SD or median and interquartile range for quantitative parameters and number and percentages for qualitative parameters. Kaplan–Meier survival analyses and log-rank tests to compare the risk of these events during follow-up in patients with hsCRP ≥ 2 vs < 2 mg/l were applied for all time-to-event outcomes, with HRs and 95% CIs derived from univariable Cox proportional hazards models. The outcomes were evaluated from 1 day after the index event to 6 years, in accordance with the TriNetX analysis window. For outcomes with chronic baseline conditions, we indicated whether TriNetX excluded patients with previous events (as listed in the Supplementary Appendix generated by the platform). All these clarifications have been added to the revised Methods and Supplementary Material.
Incidence of ESKD (defined as chronic dialysis or renal transplantation), cardiovascular events, and death was assessed over a maximum follow-up of 6 years. For all analyses, we first studied the whole population, and then studied patients with diabetes mellitus.
Results
Baseline Characteristics Before and After Propensity Score Matching in the Whole Population
Overall, 134,381 patients had CKD and hsCRP results (104,908 with hsCRP ≥ 2 mg/l and 29,473 with hsCRP < 2 mg/l) (Figure 1). These 2 groups were different for most clinical diagnoses, race, ethnicity, sex ratio, blood pressure, body mass index, comorbid conditions, biological measurements, and treatments (Table 1).
Figure 1.
Flow chart. Selected population before and after propensity score matching. CKD, chronic kidney disease; hsCRP, high sensitivity C-reactive protein.
After propensity score matching, 27,580 patients with hsCRP ≥ 2 mg/l were compared with 27,580 patients with hsCRP < 2 mg/l (Table 1, Figure 1). These 2 groups were well-matched (standard difference of all parameters between the 2 groups < 10%) on baseline characteristics (Table 1). In these 2 groups, mean age was 72 years and men represented 54% of the patients (Table 1). No meaningful differences were noted with regard to race and ethnicity, hypertension (57%), diabetes mellitus (32%), body mass index, systolic and diastolic blood pressure, comorbid conditions, smoking habits, and treatments (Table 1). In addition, these 2 populations were well-matched for LDL-cholesterol and high-density lipoprotein cholesterol levels, triglyceride levels, hemoglobin A1c, eGFR, and albuminuria (Table 1). The use of antihypertensive or other cardiovascular medications, glucose-lowering therapy, anticoagulants, and antiplatelet therapy was well-matched (Table 1).
ESKD, MACE, and Death According to hsCRP Levels During Follow-Up in the Whole Population
The median follow-up was 501 (interquartile range: 1276) days in patients with hsCRP ≥ 2 mg/l and 633 (1428) years in patients with hsCRP ≤ 2 mg/l. As shown in Table 2, patients with hsCRP ≥ 2 mg/l had significant greater risks of all-cause death (yearly rate: 1.94% vs. 10.1%, HR: 1.941 [95% CI: 1.853–2.032]), ischemic strokes, thromboembolism, atrial fibrillation, cardiac arrest, myocardial infarction, and hospitalization for heart failure (Table 2). The rate of ESKD was not statistically different in patients with baseline hsCRP ≥ 2 mg/l versus hsCRP < 2 mg/l (HR: 1.271 [0.893–1.809], P = 0.18); however, the risk was low in the 2 groups (yearly rate: 0.21 vs. 0.23%).
Table 2.
Clinical outcomes during follow-up in the matched population
| Outcomes | hsCRP ≥ 2 mg/l |
hsCRP < 2 mg/l |
Hazard ratio (95% CI) | P-value | ||
|---|---|---|---|---|---|---|
| (n = 27,580) |
(n = 27,580) |
|||||
| Number of events | Yearly rate, % | Number of events | Yearly rate, % | |||
| Death | 5012 | 5.81 | 2840 | 3.62 | 1.941 (1.853–2.032) | < 0.0001 |
| ESKD | 67 | 0.07 | 57 | 0.08 | 1.271 (0.893–1.809) | 0.18 |
| MACE and other cardiovascular events | ||||||
| Ischemic stroke or thromboembolism | 173 | 0.23 | 118 | 0.16 | 1.619 (1.281–2.046) | < 0.0001 |
| Acute MI | 142 | 0.19 | 101 | 0.13 | 1.524 (1.180–1.967) | 0.001 |
| Atrial fibrillation | 1603 | 2.55 | 1217 | 1.91 | 1.444 (1.341–1.556) | < 0.0001 |
| VT/VF/cardiac arrest | 1147 | 1.34 | 839 | 0.90 | 1.458 (1.334–1.594) | < 0.0001 |
| Cardiac arrest | 521 | 0.60 | 224 | 0.26 | 2.482 (2.122–2.903) | < 0.0001 |
| VT or VF | 707 | 0.86 | 668 | 0.71 | 1.13 (1.016–1.256) | 0.02 |
| MI/stroke/heart failure/death | 6617 | 6.90 | 4059 | 4.53 | 1.797 (1.728–1.869) | < 0.0001 |
| MI, ischemic stroke, or heart failure | 2053 | 1.89 | 1389 | 1.26 | 1.558 (1.455–1.668) | < 0.0001 |
| Incident heart failure | 1069 | 1.95 | 929 | 1.61 | 1.28 (1.173–1.398) | < 0.0001 |
| Hospitalization for heart failure | 2063 | 3.37 | 1669 | 2.69 | 1.355 (1.270–1.445) | < 0.0001 |
CI, confidence interval; ESKD, end-stage kidney disease; hsCRP, high sensitivity C-reactive protein; MACE, major cardiovascular event; MI, myocardial infarction; VF, ventricular fibrillation; VT, ventricular tachycardia.
Baseline Characteristics Before and After Propensity Score Matching Among Patients With Diabetes
Overall, 53,078 patients with diabetes had CKD and hsCRP results (41,148 with hsCRP ≥ 2 mg/l and 11,930 with hsCRP < 2 mg/l). Before propensity score matching, most parameters were different between the 2 groups (Table 3). After propensity score matching, 10,850 patients with hsCRP ≥ 2 mg/l were compared with 10,850 patients with hsCRP < 2 mg/l (Table 3). These 2 groups were well-matched (standard difference of all parameters between the 2 groups < 10% for almost all parameters) on baseline characteristics (Table 3). In these 2 groups, mean age was 72 years, and men represented 56% of the patients (Table 3). Patients were similar with regard to race and ethnicity, hypertension, body mass index, systolic and diastolic blood pressure, comorbid conditions, smoking habits, and treatments (Table 3). They were also well-matched for LDL cholesterol and high-density lipoprotein cholesterol levels, triglyceride levels, hemoglobin A1c (6.9% ± 1.6% vs. 6.7 ± 1.6%), eGFR (50.4 ± 18.5 vs. 51.1 ± 16.7 ml/min per 1.73 m2), albuminuria (232.8 ± 746.4 vs. 234.7 ± 853.8), and the rate of patients with albuminuria > 200 mg/g (1.9% vs. 2.1%) (Table 3). The use of antihypertensive or other cardiovascular medications, glucose-lowering therapy (including sodium-glucose cotransporter -2 inhibitors and glucagon-like peptide-1 receptor agonists), anticoagulants, and antiplatelet therapy was well-matched (Table 3).
Table 3.
Baseline characteristics of patients with diabetes before and after propensity score matching
| Characteristics | Before propensity score matching |
After propensity score matching |
||||||
|---|---|---|---|---|---|---|---|---|
| hsCRP ≥ 2 |
hsCRP < 2 |
P-Value | Std diff. (%) | hsCRP ≥ 2 |
hsCRP < 2 |
P-Value | Std diff. (%) | |
| (n = 41,148) | (n = 11,930) | (n = 10,850) | (n = 10,850) | |||||
| Age, yr, n (%) | 71.7 ± 11.6 | 71.4 ± 10.7 | 0.01 | 2.7 | 72.3 ± 11.1 | 71.7 ± 10.7 | < 0.001 | 5.6 |
| Men (%) | 23,083 (56.1%) | 6697 (56.1%) | 0.941 | 0.1 | 6114 (56.4%) | 6080 (56%) | 0.642 | 0.6 |
| Systolic BP (mm Hg), mean ± SD | 128.5 ± 24.3 | 131.4 ± 23.4 | < 0.001 | 12.4 | 130 ± 24 | 131 ± 23 | < 0.001 | 7.5 |
| Diastolic BP (mm Hg), mean ± SD | 69.0 ± 14.7 | 71.0 ± 13.0 | < 0.001 | 14.8 | 70 ± 15 | 71 ± 13 | 0.041 | 4 |
| Body mass index (kg/m2), mean ± SD | 30.1 ± 7.3 | 27.5 ± 5.7 | < 0.001 | 40.6 | 28.1 ± 6.4 | 27.8 ± 5.9 | 0.003 | 5.2 |
| White, n (%) | 17,045 (41.4%) | 3241 (27.2%) | < 0.001 | 30.4 | 3127 (28.8%) | 3232 (29.8%) | 0.117 | 2.1 |
| Black or African American, n (%) | 3952 (9.6%) | 566 (4.7%) | < 0.001 | 18.9 | 586 (5.4%) | 566 (5.2%) | 0.545 | 0.8 |
| Asian, n (%) | 2907 (7.1%) | 3016 (25.3%) | < 0.001 | 51.1 | 2224 (20.5%) | 2069 (19.1%) | 0.008 | 3.6 |
| Hispanic or Latino, n (%) | 1852 (4.5%) | 264 (2.2%) | < 0.001 | 12.7 | 268 (2.5%) | 264 (2.4%) | 0.861 | 0.2 |
| Unknown Race, n (%) | 16,137 (39.2%) | 4857 (40.7%) | 0.003 | 3.1 | 4662 (43%) | 4735 (43.6%) | 0.317 | 1.4 |
| Comorbid conditions | ||||||||
| Hypertension, n (%) | 28,675 (69.7%) | 7929 (66.5%) | < 0.001 | 6.9 | 7236 (66.7%) | 7170 (66.1%) | 0.343 | 1.3 |
| Smoker, n (%) | 4935 (12%) | 745 (6.2%) | < 0.001 | 20.1 | 748 (6.9%) | 740 (6.8%) | 0.83 | 0.3 |
| Overweight or obesity, n (%) | 8324 (20.2%) | 1298 (10.9%) | < 0.001 | 26 | 1333 (12.3%) | 1290 (11.9%) | 0.371 | 1.2 |
| Dyslipidemia, n (%) | 23,009 (55.9%) | 6722 (56.3%) | 0.407 | 0.9 | 6057 (55.8%) | 6003 (55.3%) | 0.461 | 1 |
| Alcohol-related diagnoses, n (%) | 797 (1.9%) | 100 (0.8%) | < 0.001 | 9.4 | 116 (1.1%) | 98 (0.9%) | 0.216 | 1.7 |
| Heart failure, n (%) | 12,895 (31.3%) | 3058 (25.6%) | < 0.001 | 12.7 | 2911 (26.8%) | 2905 (26.8%) | 0.927 | 0.1 |
| Coronary artery disease, n (%) | 27,658 (67.2%) | 7827 (65.6%) | 0.001 | 3.4 | 7192 (66.3%) | 7143 (65.8%) | 0.482 | 1 |
| Myocardial infarction, n (%) | 3675 (8.9%) | 1045 (8.8%) | 0.562 | 0.6 | 990 (9.1%) | 952 (8.8%) | 0.366 | 1.2 |
| Dilated cardiomyopathy, n (%) | 716 (1.7%) | 145 (1.2%) | < 0.001 | 4.3 | 131 (1.2%) | 138 (1.3%) | 0.668 | 0.6 |
| Ischemic stroke, n (%) | 9025 (21.9%) | 2851 (23.9%) | < 0.001 | 4.7 | 2544 (23.4%) | 2551 (23.5%) | 0.911 | 0.2 |
| Intracranial hemorrhage, n (%) | 816 (2%) | 311 (2.6%) | < 0.001 | 4.2 | 246 (2.3%) | 282 (2.6%) | 0.113 | 2.2 |
| Atrial fibrillation or flutter, n (%) | 8901 (21.6%) | 1917 (16.1%) | < 0.001 | 14.3 | 1855 (17.1%) | 1806 (16.6%) | 0.374 | 1.2 |
| Kidney disease, n (%) | 15,898 (38.6%) | 3381 (28.3%) | < 0.001 | 21.9 | 3168 (29.2%) | 3087 (28.5%) | 0.225 | 1.6 |
| Lung disease, n (%) | 18,782 (45.6%) | 4383 (36.7%) | < 0.001 | 18.2 | 4247 (39.1%) | 4093 (37.7%) | 0.032 | 2.9 |
| COPD, n (%) | 4810 (11.7%) | 871 (7.3%) | < 0.001 | 15 | 843 (7.8%) | 824 (7.6%) | 0.628 | 0.7 |
| Sleep apnea syndrome, n (%) | 5248 (12.8%) | 1073 (9%) | < 0.001 | 12.1 | 1028 (9.5%) | 1029 (9.5%) | 0.982 | 0 |
| Peripheral vascular disease, n (%) | 5256 (12.8%) | 842 (7.1%) | < 0.001 | 19.2 | 838 (7.7%) | 824 (7.6%) | 0.721 | 0.5 |
| Previous cancer, n (%) | 9921 (24.1%) | 3365 (28.2%) | < 0.001 | 9.3 | 3005 (27.7%) | 3004 (27.7%) | 0.988 | 0 |
| Anemia, n (%) | 6592 (16%) | 1402 (11.8%) | < 0.001 | 12.4 | 1386 (12.8%) | 1320 (12.2%) | 0.175 | 1.8 |
| Biology measurements | ||||||||
| Total cholesterol (mg/dl), mean ± SD | 157.9 ± 51.1 | 163.4 ± 47.3 | < 0.001 | 11.2 | 163.6 ± 49.7 | 163.3 ± 48.0 | 0.704 | 0.6 |
| LDL cholesterol (mg/dl), mean ± SD | 86.4 ± 39.1 | 88.7 ± 37.2 | < 0.001 | 6 | 90.2 ± 38.2 | 88.4 ± 37.6 | 0.004 | 4.7 |
| HDL cholesterol (mg/dl), mean ± SD | 45.2 ± 16.1 | 50.4 ± 17.1 | < 0.001 | 31.3 | 49.0 ± 16.7 | 50.1 ± 17.3 | < 0.001 | 6.3 |
| Triglyceride (mg/dl), mean ± SD | 152.6 ± 160.7 | 144.6 ± 224.8 | 0.001 | 4.1 | 145.5 ± 120.1 | 144.9 ± 227.2 | 0.841 | 0.3 |
| Hemoglobin A1c (%), mean ± SD | 7.2 ± 1.8 | 6.7 ± 1.5 | < 0.001 | 26 | 6.9 ± 1.6 | 6.7 ± 1.6 | < 0.001 | 9.5 |
| Estimated GFR (MDRD, ml/min), mean ± SD | 48.3 ± 20.3 | 51.5 ± 16.5 | < 0.001 | 17.1 | 50.4 ± 18.5 | 51.1 ± 16.7 | 0.002 | 4.2 |
| Albuminuria (mg/g), mean ± SD | 360.0 ± 1081.2 | 225.5 ± 835.6 | 0.001 | 13.9 | 232.8 ± 746.4 | 234.7 ± 853.8 | 0.962 | 0.2 |
| Albuminuria >200 mg/g, n (%) | 853 (2.1%) | 233 (2%) | 0.415 | 0.9 | 204 (1.9%) | 229 (2.1%) | 0.225 | 1.6 |
| CRP (mg/l), mean ± SD | 14.3 ± 21.4 | 1.0 ± 7.8 | < 0.001 | 82.5 | 12.4 ± 19.9 | 1.1 ± 8.1 | < 0.001 | - |
| Hemoglobin (g/dl), mean ± SD | 12.0 ± 2.4 | 12.9 ± 2.1 | < 0.001 | 37.4 | 12.4 ± 2.3 | 12.8 ± 2.1 | < 0.001 | 18 |
| Baseline treatments | ||||||||
| Beta blockers, n (%) | 22,774 (55.3%) | 5057 (42.4%) | < 0.001 | 26.1 | 4853 (44.7%) | 4763 (43.9%) | 0.219 | 1.7 |
| Calcium channel blockers, n (%) | 16,894 (41.1%) | 4999 (41.9%) | 0.098 | 1.7 | 4445 (41%) | 4325 (39.9%) | 0.097 | 2.3 |
| ACE inhibitors, n (%) | 12,667 (30.8%) | 2702 (22.6%) | < 0.001 | 18.5 | 2596 (23.9%) | 2558 (23.6%) | 0.544 | 0.8 |
| Angiotensin 2 inhibitors, n (%) | 12,661 (30.8%) | 4225 (35.4%) | < 0.001 | 9.9 | 3669 (33.8%) | 3606 (33.2%) | 0.365 | 1.2 |
| Digitalis glycosides, n (%) | 1817 (4.4%) | 353 (3%) | < 0.001 | 7.7 | 368 (3.4%) | 325 (3%) | 0.097 | 2.3 |
| Diuretics, n (%) | 20,947 (50.9%) | 4533 (38%) | < 0.001 | 26.2 | 4349 (40.1%) | 4193 (38.6%) | 0.03 | 2.9 |
| Lipid-lowering drugs, n (%) | 26,139 (63.5%) | 7082 (59.4%) | < 0.001 | 8.6 | 6375 (58.8%) | 6275 (57.8%) | 0.169 | 1.9 |
| Glucose-lowering therapy, n (%) | 25,703 (62.5%) | 6121 (51.3%) | < 0.001 | 22.7 | 5660 (52.2%) | 5522 (50.9%) | 0.061 | 2.5 |
| Insulin, n (%) | 18,819 (45.7%) | 3159 (26.5%) | < 0.001 | 40.9 | 3202 (29.5%) | 3100 (28.6%) | 0.127 | 2.1 |
| Noninsulin glucose-lowering therapy, n (%) | 17,359 (42.2%) | 5016 (42%) | 0.783 | 0.3 | 4427 (40.8%) | 4424 (40.8%) | 0.967 | 0.1 |
| Metformin, n (%) | 11,652 (28.3%) | 3623 (30.4%) | < 0.001 | 4.5 | 3188 (29.4%) | 3117 (28.7%) | 0.288 | 1.4 |
| Sulfonylureas, n (%) | 6825 (16.6%) | 1883 (15.8%) | 0.037 | 2.2 | 1671 (15.4%) | 1653 (15.2%) | 0.734 | 0.5 |
| GLP-1 receptor agonists, n (%) | 1667 (4.1%) | 490 (4.1%) | 0.785 | 0.3 | 451 (4.2%) | 456 (4.2%) | 0.865 | 0.2 |
| DPP4 inhibitors, n (%) | 5081 (12.3%) | 1733 (14.5%) | < 0.001 | 6.4 | 1473 (13.6%) | 1460 (13.5%) | 0.796 | 0.4 |
| SGLT2 inhibitors, n (%) | 2933 (7.1%) | 949 (8%) | 0.002 | 3.1 | 848 (7.8%) | 838 (7.7%) | 0.8 | 0.3 |
| Thiazolidinediones, n (%) | 1379 (3.4%) | 585 (4.9%) | < 0.001 | 7.8 | 409 (3.8%) | 492 (4.5%) | 0.005 | 3.8 |
| Antiplatelet therapy, n (%) | 24,002 (58.3%) | 5884 (49.3%) | < 0.001 | 18.1 | 5549 (51.1%) | 5412 (49.9%) | 0.063 | 2.5 |
| Anticoagulant, n (%) | 3706 (9%) | 463 (3.9%) | < 0.001 | 21 | 524 (4.8%) | 460 (4.2%) | 0.037 | 2.8 |
ACE, angiotensin-converting enzyme; BP, blood pressure; COPD, chronic obstructive pulmonary disease; DPP4, dipeptidyl peptidase 4; GFR, glomerular filtration rate; GLP1, glucagon-like peptide-1; HDL, high-density lipoprotein; hsCRP, high sensitivity C-reactive protein; LDL, low-density lipoprotein; MDRD, Modification of Diet in Renal Disease formula; SGLT2, sodium glucose cotransporter-2; Std diff., standard difference.
hsCRP Levels and Major Outcomes During Follow-Up in the Subgroup of Patients With Diabetes
The median follow-up was 589 (interquartile range: 1366) days in patients with hsCRP ≥ 2 mg/l and 763 (1550) in patients with hsCRP < 2 mg/l. As shown in Table 4 and Figure 2, patients with hsCRP ≥ 2 mg/l had significant greater risks of all-cause death (yearly rate: 15.5% vs. 9.3%, HR: 1.899 [95% CI: 1.769–2.038]), ischemic strokes, thromboembolism, atrial fibrillation, cardiac arrest, myocardial infarction, and hospitalization for heart failure (Table 4). In addition, the rate of ESKD tended to be higher in patients with baseline hsCRP ≥ 2 mg/l versus ≤ 2 mg/l (HR: 1.508 [0.997–2.281], P = 0.05) (Table 4).
Table 4.
Clinical outcomes during follow-up in the matched population of patients with diabetes
| hsCRP ≥ 2 mg/l |
hsCRP < 2 mg/l |
Hazard ratio (95% CI) | P-value | |||
|---|---|---|---|---|---|---|
| (n = 10,850) |
(n = 10,850) |
|||||
| Number of events | Yearly rate, % | Number of events | Yearly rate, % | |||
| Death | 2077 | 15.05 | 1225 | 9.30 | 1.899 (1.769–2.038) | < 0.0001 |
| ESKD | 53 | 0.47 | 39 | 0.34 | 1.508 (0.997–2.281) | 0.05 |
| MACE and other cardiovascular events | ||||||
| Ischemic stroke or thromboembolism | 74 | 0.65 | 38 | 0.30 | 2.186 (1.478–3.233) | < 0.0001 |
| Acute MI | 63 | 0.58 | 39 | 0.29 | 1.785 (1.197–2.663) | 0.004 |
| Atrial fibrillation | 678 | 6.59 | 508 | 4.83 | 1.493 (1.331–1.675) | < 0.0001 |
| VT/VF/cardiac arrest | 416 | 3.16 | 338 | 2.24 | 1.331 (1.153–1.537) | < 0.0001 |
| Cardiac arrest | 170 | 1.37 | 94 | 0.66 | 1.979 (1.538–2.546) | < 0.0001 |
| VT or VF | 277 | 2.06 | 275 | 1.81 | 1.084 (0.917–1.281) | 0.34 |
| MI/stroke/heart failure/death | 2659 | 17.53 | 1562 | 10.91 | 1.916 (1.800–2.040) | < 0.0001 |
| MI, ischemic stroke, or heart failure | 753 | 4.47 | 400 | 2.31 | 2.013 (1.783–2.273) | < 0.0001 |
| Incident heart failure | 934 | 10.64 | 861 | 9.40 | 1.212 (1.105–1.330) | < 0.0001 |
| Hospitalization for heart failure | 415 | 2.50 | 235 | 1.38 | 1.875 (1.597–2.200) | < 0.0001 |
CI, confidence interval; ESKD, end-stage kidney disease; hsCRP, high sensitivity C-reactive protein; MACE, major cardiovascular event; MI, myocardial infarction; VF, ventricular fibrillation; VT, ventricular tachycardia.
Figure 2.
Event-free curve for all-cause death in the 2 groups. Event-free curve for all-cause death in matched patients with hsCRP < 2 mg/l (blue) versus patients with hsCRP ≥ 2 mg/l (red). hsCRP, high sensitivity C-reactive protein.
Baseline Characteristics Before and After Propensity Score Matching Among Patients With No Diabetes
Overall, 80,266 patients without diabetes had CKD and hsCRP results (62,763 patients with hsCRP ≥ 2 mg/l and 17,503 patients with hsCRP < 2 mg/l) (Supplementary Table S1). Before propensity score matching, most parameters were different between the 2 groups (Supplementary Table S1). After propensity score matching, 16,530 patients with hsCRP ≥ 2 mg/l were compared with 16,530 patients with hsCRP < 2 mg/l (Supplementary Table S1). These 2 groups were well-matched (standard difference of all parameters between the 2 groups < 10% for almost all parameters) on baseline characteristics (Supplementary Table S1).
hsCRP Levels and Major Outcomes During Follow-Up in the Subgroup of Patients With No Diabetes
The median follow-up was 452 (interquartile range: 1220) days in patients with hsCRP ≥ 2 mg/l and 553 (1332) in patients with hsCRP < 2 mg/l. As shown in Supplementary Table S2, patients with hsCRP ≥ 2 mg/l had significant greater risks of all-cause death (yearly rate: 17% vs. 11%, HR: 1.965 [95% CI: 1.849–2.090]), ischemic strokes, thromboembolism, atrial fibrillation, cardiac arrest, myocardial infarction, and hospitalization for heart failure (Supplementary Table S2). The risk of ESKD did not reach the statistical threshold (HR: 1.247 [0.653–2.382], P = 0.5) (Supplementary Table 2).
Discussion
In the present study conducted in a large population of well-matched patients with CKD stage 3 and 4 or albuminuria (A3) (median eGFR: 52 ml/min per 1.73 m2, albumin-to-creatinine ratio: 210 mg/g), we found that hsCRP ≥ 2 mg/l was associated with significantly higher risks of all-cause death and MACE compared with patients with hsCRP < 2 mg/l, and among them, in those with no diabetes mellitus. Similar results were found for MACE and death risks among patients with diabetes and hsCRP ≥ 2 mg/l, but in addition, the risk of ESKD was higher than in patients with hsCRP < 2 mg/l.
The association between a baseline hsCRP level ≥ 2 mg/l with increased risks of a wide range of major cardiovascular and cerebrovascular events, ventricular and atrial arrythmias, and all-cause death among patients with CKD was observed, though these patients were well-matched for traditional cardiovascular risk factors, including comorbid conditions, a wide range of biological and clinical parameters and baseline treatments. Whether this association between CRP levels and cardiovascular events is causal has been extensively discussed and is still debated. Mendelian randomization studies in 2 cohorts and a meta-analysis of 47 studies (194,418 subjects) did not find an independent association between CRP levels and coronary heart disease.9 In marked contrast, in a recent genome-wide association in half a million subjects, a genetic risk score of CRP was associated with coronary heart disease.10 A recent study from the UK biobank indicated that CRP levels > 10 mg/l (vs. CRP < 0.5 mg/l) were significantly and positively associated with the risk of incident atrial fibrillation.11 Furthermore, they found a continuous positive relationship between baseline CRP levels and incident ventricular arrythmia.11 CRP may promote inflammation and subsequent myocardial fibrosis through the TLR4/NF-κB/TGF-β pathway.12 Another study showed that CRP played a proarrhythmic role by directly affecting calcium homeostasis in cardiomyocytes.13 These studies were mainly conducted in the general population. Our study included patients with CKD and albuminuria. These associations were confirmed in the subgroup of patients with diabetes mellitus. Our results are in accordance with smaller observational studies. In 543 patients with CKD stage 5, markers of inflammation, including elevated CRP were associated with baseline cardiovascular disease and mortality during follow-up.14 In 2399 patients with mainly CKD stage 3a or 3b, markers of inflammation, including elevated CRP were associated with baseline cardiovascular disease and mortality during follow-up.5 An independent association was found in CKD populations and the risk of death,15 including in those with coronary artery disease.16,17
We observed that hsCRP ≥ 2 mg/l (vs. hsCRP < 2 mg/l) was associated with an increased risk of ESKD in subjects with diabetes mellitus, but not in the whole population. The absence of a significant association between elevated hsCRP levels and the risk of ESKD in the whole population was probably because of a lack of power, because very few patients reached ESKD. It was found that hsCRP predicted the risk of developing CKD in a population-based cohort.18 A faster progression to kidney failure associated with elevated hsCRP was noted in another study.5 It was therefore proposed by the authors that “traditional cardiovascular risk estimates could be improved by adding markers of inflammation and measures of kidney function.”5
Whether targeting inflammation is beneficial or detrimental in patients with risk of ESKD and MACE is unclear. Statins have demonstrated antiinflammatory effects.19 However, a recent meta-analysis found no association with CKD progression toward ESKD.20 Targeting interleukin-1β using canakinumab decreased hsCRP and provided cardioprotection, even in a subgroup of patients with mild CKD; however, no nephroprotection was found in a recent clinical trial.21 In a phase 2 trial, ziltivekimab, a monoclonal antibody directed against the interleukin-6 ligand, markedly reduced hsCRP in patients with stages 3 and 4 CKD.8 The randomized clinical trial titled “ZEUS-Effects of Ziltivekimab Versus Placebo on Cardiovascular Outcomes in Participants With Established Atherosclerotic Cardiovascular Disease, Chronic Kidney Disease and Systemic Inflammation” is ongoing.22 This study in 6000 patients with CKD will analyze whether ziltivekimab results in cardioprotection and effective nephroprotection.22 Interestingly, the inclusion criteria for this study are similar to those used in the present cohort. Although it is tempting to speculate that similar results would be observed in the ZEUS clinical trial, they cannot be extrapolated to ZEUS future findings.
The study has limitations. A selection bias is possible, although a propensity matching score was used. It is a real-life study, and therefore, even after careful propensity score matching, differences in the management of these subjects could still be present. We focused on analysis on subjects with mild to moderate KDIGO CKD 3 and 4 or albuminuria > 200 mg/g (KDIGO A3), that is, patients who have the highest risk of progression of renal diseases. It is possible that different results could be found in subjects with more advanced renal dysfunction. We believe that it is unlikely because smaller observational studies confirmed that elevated CRP is present in patients with CKD, regardless of the CKD stage, and that cardiovascular events were associated with elevated hsCRP in mild CKD stages as well as in patients under chronic dialysis.
The strength of this study derives from its size and design. Overall, after propensity score matching, 55,160 subjects for the whole population and 21,700 subjects with diabetes mellitus were evaluated (from an initial population of 163,854 subjects). The 2 subgroups of subjects were well-matched on all traditional cardiovascular risk factors, including blood pressure; comorbid conditions; LDL cholesterol and high-density lipoprotein cholesterol levels; hemoglobin A1c; eGFR and albuminuria; and baseline treatments, including sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists for subjects with diabetes mellitus. Careful analyses were performed in the whole population and in subjects with diabetes mellitus. Cardiovascular outcomes, including atrial and ventricular arrythmias, were thoroughly evaluated and hard renal outcomes were assessed.
In conclusion, the presence of systemic inflammation (defined as hsCRP ≥ 2 mg/l) is associated with elevated risk of ischemic stroke, MACE, atrial and ventricular arrythmias among subjects with mild CKD 3 and 4 or elevated albuminuria (A3), independently of traditional cardiovascular risk factors and treatments, even in the absence of diabetes mellitus. In addition, hsCRP ≥ 2 mg/l tended to be associated with a greater risk of ESKD, independently of baseline blood pressure, renal function, and treatments in the subgroup of patients with diabetes mellitus.
Disclosure
All the authors declared no competing interests.
Author Contributions
JMH and LF contributed to study conception, statistical analysis, writing of the first draft, and access to the data. VM, JBdF, SR, AB, and SC contributed to corrections of the drafts and interpretation of data. All the authors contributed to the drafting and correction of the manuscript.
Footnotes
Propensity score matching.
Table S1. Baseline characteristics of patients with no diabetes before and after propensity score matching.
Table S2. Clinical outcomes during follow-up in the matched population of patients with no diabetes.
Strobe checklist.
Supplementary Material
Propensity score matching. Table S1. Baseline characteristics of patients with no diabetes before and after propensity score matching. Table S2. Clinical outcomes during follow-up in the matched population of patients with no diabetes. Strobe checklist.
References
- 1.Jager K.J., Kovesdy C., Langham R., Rosenberg M., Jha V., Zoccali C. A single number for advocacy and communication-worldwide more than 850 million individuals have kidney diseases. Kidney Int. 2019;96:1048–1050. doi: 10.1016/j.kint.2019.07.012. [DOI] [PubMed] [Google Scholar]
- 2.De Nicola L., Correa-Rotter R., Navarro-Gonzalez J.F., et al. Projecting the population level burden of CKD progression according to urine albumin-to-creatinine ratio categories. Kidney Int Rep. 2024;9:3464–3476. doi: 10.1016/j.ekir.2024.09.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Matsushita K., Ballew S.H., Wang A.Y., Kalyesubula R., Schaeffner E., Agarwal R. Epidemiology and risk of cardiovascular disease in populations with chronic kidney disease. Nat Rev Nephrol. 2022;18:696–707. doi: 10.1038/s41581-022-00616-6. [DOI] [PubMed] [Google Scholar]
- 4.Matsushita K., Kaptoge S., Hageman S.H.J., et al. Including measures of chronic kidney disease to improve cardiovascular risk prediction by SCORE2 and SCORE2-OP. Eur J Prev Cardiol. 2023;30:8–16. doi: 10.1093/eurjpc/zwac176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Amdur R.L., Feldman H.I., Dominic E.A., et al. Use of measures of inflammation and kidney function for prediction of atherosclerotic vascular disease events and death in patients with CKD: findings from the CRIC study. Am J Kidney Dis. 2019;73:344–353. doi: 10.1053/j.ajkd.2018.09.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ridker P.M., MacFadyen J.G., Everett B.M., et al. Relationship of C-reactive protein reduction to cardiovascular event reduction following treatment with canakinumab: a secondary analysis from the CANTOS randomised controlled trial. Lancet. 2018;391:319–328. doi: 10.1016/S0140-6736(17)32814-3. [DOI] [PubMed] [Google Scholar]
- 7.Ridker P.M., Libby P., MacFadyen J.G., et al. Modulation of the interleukin-6 signalling pathway and incidence rates of atherosclerotic events and all-cause mortality: analyses from the Canakinumab Anti-Inflammatory Thrombosis Outcomes Study (CANTOS) Eur Heart J. 2018;39:3499–3507. doi: 10.1093/eurheartj/ehy310. [DOI] [PubMed] [Google Scholar]
- 8.Ridker P.M., Devalaraja M., Baeres F.M.M., et al. IL-6 inhibition with ziltivekimab in patients at high atherosclerotic risk (RESCUE): a double-blind, randomised, placebo-controlled, phase 2 trial. Lancet. 2021;397:2060–2069. doi: 10.1016/S0140-6736(21)00520-1. [DOI] [PubMed] [Google Scholar]
- 9.Elliott P., Chambers J.C., Zhang W., et al. Genetic Loci associated with C-reactive protein levels and risk of coronary heart disease. JAMA. 2009;302:37–48. doi: 10.1001/jama.2009.954. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Said S., Pazoki R., Karhunen V., et al. Genetic analysis of over half a million people characterises C-reactive protein loci. Nat Commun. 2022;13:2198. doi: 10.1038/s41467-022-29650-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yang X., Zhao S., Wang S., et al. Systemic inflammation indicators and risk of incident arrhythmias in 478,524 individuals: evidence from the UK Biobank cohort. BMC Med. 2023;21:76. doi: 10.1186/s12916-023-02770-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Oikonomou E., Zografos T., Papamikroulis G.A., et al. Biomarkers in atrial fibrillation and heart failure. Curr Med Chem. 2019;26:873–887. doi: 10.2174/0929867324666170830100424. [DOI] [PubMed] [Google Scholar]
- 13.Xie Y., Li Q., Zhang H.F., et al. Effect of C reactive protein on the sodium-calcium exchanger 1 in cardiomyocytes. Exp Ther Med. 2021;22:815. doi: 10.3892/etm.2021.10247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sun J., Axelsson J., Machowska A., et al. Biomarkers of cardiovascular disease and mortality risk in patients with advanced CKD. Clin J Am Soc Nephrol. 2016;11:1163–1172. doi: 10.2215/CJN.10441015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Munoz Mendoza J., Isakova T., Cai X., et al. Inflammation and elevated levels of fibroblast growth factor 23 are independent risk factors for death in chronic kidney disease. Kidney Int. 2017;91:711–719. doi: 10.1016/j.kint.2016.10.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Batra G., Ghukasyan Lakic T., Lindback J., et al. Interleukin 6 and cardiovascular outcomes in patients with chronic kidney disease and chronic coronary syndrome. JAMA Cardiol. 2021;6:1440–1445. doi: 10.1001/jamacardio.2021.3079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Schunk S.J., Triem S., Schmit D., et al. Interleukin-1alpha is a central regulator of leukocyte-endothelial adhesion in myocardial infarction and in chronic kidney disease. Circulation. 2021;144:893–908. doi: 10.1161/CIRCULATIONAHA.121.053547. [DOI] [PubMed] [Google Scholar]
- 18.Shankar A., Sun L., Klein B.E., et al. Markers of inflammation predict the long-term risk of developing chronic kidney disease: a population-based cohort study. Kidney Int. 2011;80:1231–1238. doi: 10.1038/ki.2011.283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Diamantis E., Kyriakos G., Quiles-Sanchez L.V., Farmaki P., Troupis T. The anti-inflammatory effects of statins on coronary artery disease: an updated review of the literature. Curr Cardiol Rev. 2017;13:209–216. doi: 10.2174/1573403X13666170426104611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhao L., Li S., Gao Y. Efficacy of statins on renal function in patients with chronic kidney disease: a systematic review and meta-analysis. Ren Fail. 2021;43:718–728. doi: 10.1080/0886022X.2021.1915799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ridker P.M., MacFadyen J.G., Glynn R.J., et al. Inhibition of interleukin-1beta by canakinumab and cardiovascular outcomes in patients with chronic kidney disease. J Am Coll Cardiol. 2018;71:2405–2414. doi: 10.1016/j.jacc.2018.03.490. [DOI] [PubMed] [Google Scholar]
- 22.Ridker P.M. From RESCUE to ZEUS: will interleukin-6 inhibition with ziltivekimab prove effective for cardiovascular event reduction? Cardiovasc Res. 2021;117:e138–e140. doi: 10.1093/cvr/cvab231. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Propensity score matching. Table S1. Baseline characteristics of patients with no diabetes before and after propensity score matching. Table S2. Clinical outcomes during follow-up in the matched population of patients with no diabetes. Strobe checklist.