Abstract
BACKGROUND
Increases in cardiac and stress biomarkers may be associated with loss of kidney function through shared mechanisms involving cardiac and kidney injury. We evaluated the associations of cardiac and stress biomarkers [N-terminal pro-B-type natriuretic peptide (NT-proBNP), high-sensitivity troponin T (hsTnT), growth differentiation factor 15 (GDF-15), soluble ST-2 (sST-2)] with progression of chronic kidney disease (CKD).
METHODS
We included 3664 participants with CKD from the Chronic Renal Insufficiency Cohort study. All biomarkers were measured at entry. The primary outcome was CKD progression, defined as progression to end-stage renal disease (ESRD) or 50% decline in estimated glomerular filtration rate (eGFR). Cox models tested the association of each biomarker with CKD progression, adjusting for demographics, site, diabetes, cardiovascular disease, eGFR, urine proteinuria, blood pressure, body mass index, cholesterol, medication use, and mineral metabolism.
RESULTS
There were 1221 participants who had CKD progression over a median (interquartile range) follow-up of 5.8 (2.4–8.6) years. GDF-15, but not sST2, was significantly associated with an increased risk of CKD progression [hazard ratios (HRs) are per SD increase in log-transformed biomarker]: GDF-15 (HR, 1.50; 95% CI, 1.35–1.67) and sST2 (HR, 1.07; 95% CI, 0.99–1.14). NT-proBNP and hsTnT were also associated with increased risk of CKD progression, but weaker than GDF-15: NT-proBNP (HR, 1.24; 95% CI, 1.13–1.36) and hsTnT (HR, 1.11; 95% CI, 1.01–1.22).
CONCLUSIONS
Increases in GDF-15, NT-proBNP, and hsTnT are associated with greater risk for CKD progression. These biomarkers may inform mechanisms underlying kidney injury.
Cardiac and stress biomarkers have been shown to be associated with poor clinical outcomes in patients with and without chronic kidney disease (CKD).10 More specifically, N-terminal pro-B-type natriuretic peptide (NT-proBNP) and high-sensitivity troponin T (hsTnT) are associated with major cardiovascular events and mortality (1, 2). NT-proBNP is secreted from cardiac myocytes in response to stimuli such as pressure or volume overload (3). hsTnT concentrations rise in response to myocardial injury, remodeling, or left ventricular hypertrophy (4). Several studies have also found a link between increased concentrations of growth differentiation factor (GDF) 15 and soluble ST2 (sST2) and clinical outcomes, including cardiovascular disease (5–7). GDF-15 is a member of the transforming growth factor β cytokine superfamily, whose production is induced in response to conditions associated with cellular stress (8). GDF-15 is widely produced in many tissues including kidney tubular cells, cardiomyotcytes, adipocytes, macrophages, endothelial cells, and others (9–12) sST2 is the soluble (fraction of) suppression of tumorigenticity 2 and a member of the interleukin 1 (IL-1) receptor family that is also widely produced and thought to induce inflammation, myocardial hypertrophy, and fibrosis, among other biological insults (7, 13). Increases in GDF-15 and sST2 are linked with several diseases (including cardiovascular and kidney disease) and likely reflect non–tissue-specific stress and injury (8, 13).
The association of these cardiac and stress biomarkers with loss of kidney function is not well established. Prior studies have identified possible associations of NT-proBNP, hsTnT, GDF-15, and sST2 with loss of kidney function (14–18); however, these studies have been limited by inclusion of non-CKD populations, the use of clinical trial populations that may lack generalizability, and small sample size. Therefore, in this study, we evaluated the associations of NT-proBNP, hsTnT, GDF-15, and sST2 with CKD progression in a large, longitudinal cohort of CKD patients.
Methods
STUDY POPULATION
We studied adults with mild-to-moderate CKD [estimated glomerular filtration rate (eGFR), 20–70 mL/min/1.73 m2] enrolled in the Chronic Renal Insufficiency Cohort (CRIC) Study. A total of 3939 participants were enrolled into the CRIC study between June 2003 and August 2008 at 7 clinical centers across the US (19, 20). Inclusion and exclusion criteria have been previously described (19). Participants on maintenance dialysis or with a kidney transplant were not included at cohort entry. CRIC also excluded participants with advanced heart failure, defined as New York Heart Association class III or IV, on cohort entry. All participants enrolled in the study had annual inperson study visits in which detailed interviews were conducted, brief physical examination performed, laboratory measures done, and cardiovascular testing performed. All study participants provided written informed consent, and the study protocol was approved by institutional review boards at each site.
For the present analysis, we excluded participants who were not able to have all 4 biomarkers measured concurrently from stored samples. After applying these exclusions, 3664 participants were analyzed.
CARDIAC AND STRESS BIOMARKERS
GDF-15 and sST2 were measured from EDTA plasma stored at −70 °C from samples at baseline (n = 3664) and at year 2 in a random subcohort (n = 947) in batch at the University of Pennsylvania Laboratory. All assays were measured in duplicate. GDF-15 and sST2 were measured with ELISA (R&D Systems). For GDF-15, the quantitative range was 23.4–1500 pg/mL, with a lower limit of detection of 2.0 pg/mL. At a concentration of 98.8 pg/mL, the intraassay CV was 7.2%; at a concentration of 624 pg/mL, 4.5%. For sST2, the quantitative range was 0.63–40 ng/mL, with a limit of detection of 0.1 ng/mL. At a concentration of 2.6 ng/mL, the intraassay CV was 11.2%; at 0.94 ng/mL, 8.5%.
hs-cTnT and NT-proBNP were measured at baseline in 2008 from EDTA plasma stored at −70 °C, both by chemiluminescent microparticle immunoassay (Roche Diagnostics) on the Elecsys 2010. hs-cTnT was measured with a high-sensitivity assay with a range of values from 3 to 10000 ng/L (21). The limit of blank was 3 ng/L, and limit of detection was <5 ng/L. For hsTnT, the intraassay CV was 3% at a concentration of 30 ng/L and 5.8% at 2213 ng/L. The value at the 99th percentile cutoff from a healthy reference population was 13 ng/L for hsTnT with a 10% intraassay CV (21). The range of values for NT-proBNP was from 114 to 5900 ng/L, and the intraassay CV was 4.25% at a concentration of 132 ng/L and 5.3% at 4640 ng/L.
In 2017, we added year 2 measures of NT-proBNP and hsTnT and remeasured a subset of baseline samples to calibrate the measures. The new measurements in 2017 were performed on the Roche E601. We remeasured NT-proBNP in 100 random samples from baseline and all the year 2 samples (n = 947). We developed and applied a Deming regression (22) to calibrate the 2008 baseline NT-proBNP measures with the 2018 NT-proBNP measures. Deming regression is an extension of linear regression in which both the “X” and “Y” variables are presumed to have random measurement error but does not assume that there is the same amount of measurement error in both “X” and “Y.”
Similarly, for hsTnT, we remeasured any baseline hsTnT measure with a value <5 ng/L, using the newer Roche E601 instrument, which had a limit of blank of 2.5 ng/L and limit of detection of 3 ng/L. At a concentration of 13.5 ng/L, the intraassay CV was 1.9%; at 4831 ng/L, the CV was 0.8% with the newer instrument. We also measured a random subset of 100 samples at baseline and all samples at the year 2 visit (n = 947). We developed and applied a Deming regression (similar methods as used for NT-proBNP) to calibrate the 2008 baseline hsTnT measures with the 2018 hsTnT measures.
PROGRESSION OF CKD
The primary study outcome was progression of CKD through March 31, 2013, defined as either (a) a decline in eGFR by 50% or (b) progression to end-stage renal disease (ESRD), defined as a need for dialysis or kidney transplant. Participants were censored at the occurrence of the primary outcome, end of administrative follow-up, loss to follow-up, or death. eGFR was calculated with serum creatinine measured at annual study visits and the CKD-EPI equation (23). ESRD was identified through self-report, medical records review, and data from the US Renal Data System. Deaths were identified by report from next of kin, retrieval of death certificates or obituaries, review of hospital or outpatient records, and searching Social Security Death vital status.
COVARIATES
Participants provided information on their sociodemographic characteristics, medical history, medication usage, and lifestyle behaviors. The clinical center was defined as the geographic site of recruitment. Race/ethnicity was by self-report. Baseline cardiovascular disease status was determined by self-report and was defined as history of coronary artery disease, heart failure, or stroke. Blood pressure and body mass index (BMI) were assessed with standard protocols (24). Additional assays measured hemoglobin, serum phosphorus, 24-h urine total protein, glucose, LDL cholesterol, HDL cholesterol, fibroblast growth factor (FGF) 23, and total parathyroid hormone (PTH). Echocardiograms were performed 1 year after enrollment and provided data on left ventricular ejection fraction, left ventricular mass index (LVMI, indexed to height), which were adjusted for in a sensitivity analysis.
STATISTICAL ANALYSES
Summary statistics and distributions of NT-proBNP, hsTnT, GDF-15, and sST2 were generated. Study variables were described overall and across categories of hsTnT (4 categories for hsTnT) and quintiles for NT-proBNP, GDF-15, and sST2.
Each biomarker was modeled continuously [per SD of the natural log(biomarker)] and in categories (quintiles with the exception of hsTnT). For hsTnT, participants with concentrations <10 ng/L were designated as the referent group to ensure stability in the statistical models, given the low numbers of participants in the <3 and 3–9.9 ng/L ranges. Participants with hsTnT >10 ng/L were divided into tertiles (for a total of 4 categories of hsTnT) to again allow sufficient number of participants and events in each category for statistical analyses. For the continuous modeling of hsTnT, because the lower limit of detection was 3 ng/mL, for the n = 39 participants who had a hsTNT value <3 ng/L, we set their hsTnT value to 1.5 ng/L and then calculated standard deviation as if the variable were continuous, assuming this (1.5 ng/L) was their “true” value.
Crude rates of progression to ESRD or 50% decline in eGFR were calculated across categories of each biomarker as specified above. Cox proportional hazards models were fit for CKD progression, and follow-up was censored at the end of administrative follow-up, loss to follow-up, or death, whichever occurred first. We performed a series of nested Cox proportional hazard models with sequential adjustment for potential confounders as follows. Model 1 adjusted for demographic factors including age, sex, race/ethnicity, clinical center, and traditional cardiovascular risk factors including diabetes status, self-reported cardiovascular disease at baseline, current smoking, 24-h urine total protein excretion, eGFR, systolic blood pressure, BMI, LDL, and HDL levels. Model 2 included the factors in model 1, pertinent medication use, and markers of mineral metabolism such as use of angiotensin-converting enzyme inhibitors/angiotensis II receptor blockers (ACEi/ARBs), diuretics, and β-blockers; serum phosphorus; PTH; and FGF-23 levels. Missing covariates were multiply imputed with chained equations (25). The multiple analyses over the imputations were combined with Rubin's rules to account for the variability in the imputation procedure (26). Missingness in covariates was as follows: 24-h urinary protein (n = 175), systolic blood pressure (SBP) (n = 1), BMI (n = 9), LDL (n = 6), HDL (n = 2), ACEi/ARBs (n = 25), diuretics (n = 25), β blockers (n = 25), phosphate (n = 54), PTH (n = 32), FGF-23 (n = 31).
In a sensitivity analysis, we excluded 1207 participants with prevalent cardiovascular disease at cohort entry (defined as heart failure, coronary heart disease, or stroke) and repeated our analyses. The rationale of this analysis was to determine whether the observed associations persisted even in the absence of clinical cardiovascular disease.
We performed 5 secondary analyses. In the first, we adjusted for the alternative 3 biomarkers to determine whether the observed associations were independent of other biological pathways. In a second analysis, we adjusted for subclinical measures of heart failure, left ventricular ejection fraction (LVEF), and left ventricular mass index (LVMI), determined by echocardiogram. In a third secondary analysis, we modeled each biomarker as a time-updated exposure, thus using the year 2 measure that was available in a subset of participants. The rationale for this analysis was to determine whether a closer measure of each biomarker strengthened the observed associations with CKD progression. In a fourth secondary analysis, we calculated the c statistic of a baseline clinical model (model 1) and then calculated the c statistic of the model when each biomarker was added to the model individually (27). The rationale for this analysis was to assess whether the biomarkers improved discrimination of CKD progression. In a fifth secondary analysis, for the biomarkers that were found to be statistically significant, we modeled the association of combinations of those biomarkers with risk of CKD progression. Combinations of biomarkers were defined as low/low (reference), high/low, low/high, and high/high, with “low” defined as the lowest 4 quintiles of the biomarkers and “high” as the top quintile. All analyses were performed with the R 3.4.0 (R Foundation for Statistical Computing) software environment.
Results
CHARACTERISTICS OF STUDY POPULATION
Among 3664 eligible participants, those with higher concentrations of GDF-15 were older in age, less likely to be women, more likely to be black, have lower eGFR, higher urine proteinuria, and more likely have a history of cardiovascular disease (Table 1). Similarly, participants in the highest categories of sST2, NT-proBNP, and hsTnT also had greater morbidity (see Tables 1a–c in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol65/issue11).
Table 1.
Baseline characteristics by quintile of baseline GDF-15 concentration (pg/mL) in adults with CKD (N = 3664).a
| ≤906 | 907–1280 | 1281–1720 | 1721–2410 | >2410 | |
|---|---|---|---|---|---|
| N | 733 | 733 | 732 | 733 | 733 |
| Age, years | 51.6 (11.6) | 57.0 (10.9) | 59.1 (10.5) | 60.6 (9.7) | 60.7 (9.4) |
| Women | 370 (50) | 340 (46) | 327 (45) | 331 (45) | 305 (42) |
| Race/ethnicity | |||||
| Non-Hispanic white | 398 (54) | 323 (44) | 316 (43) | 286 (39) | 221 (30) |
| Non-Hispanic black | 258 (35) | 301 (41) | 297 (41) | 315 (43) | 334 (46) |
| Hispanic | 45 (6) | 74 (10) | 98 (13) | 101 (14) | 149 (20) |
| Other | 32 (4) | 35 (5) | 21 (3) | 31 (4) | 29 (4) |
| eGFR, mL/min/1.73 m2 | 58.7 (13.3) | 48.0 (11.6) | 42.5 (11.2) | 38.2 (12.0) | 33.9 (11.7) |
| Median 24-h urine protein, g/d | 0.1 (0.0–0.1) | 0.1 (0.0–0.4) | 0.1 (0.1–0.7) | 0.3 (0.1–1.2) | 0.6 (0.1–2.8) |
| Diabetes mellitus | 131 (18) | 289 (39) | 373 (51) | 469 (64) | 523 (71) |
| History of cardiovascular disease | 95 (13) | 189 (26) | 248 (34) | 337 (46) | 338 (46) |
| History of heart failure | 26 (4) | 49 (7) | 59 (8) | 98 (13) | 116 (16) |
| History of atrial fibrillation | 89 (12) | 104 (14) | 114 (16) | 148 (20) | 154 (21) |
| Current smoker | 45 (6) | 67 (9) | 92 (13) | 121 (17) | 139 (19) |
| Alcohol use | 574 (78) | 500 (68) | 435 (59) | 409 (56) | 394 (54) |
| Use of ACEi/ARB | 399 (54) | 524 (71) | 532 (73) | 553 (75) | 501 (68) |
| Use of diuretics | 269 (37) | 400 (55) | 457 (62) | 506 (69) | 532 (73) |
| Use of beta blockers | 229 (31) | 329 (45) | 378 (52) | 421 (57) | 438 (60) |
| BMI, kg/m2 | 31.1 (7.3) | 32.5 (8.1) | 32.6 (7.8) | 32.6 (8.0) | 31.9 (8.1) |
| Systolic blood pressure, mmHg | 120.6 (18.1) | 125.1 (19.7) | 128.3 (21.5) | 132.6 (22.9) | 136.6 (24.3) |
| Diastolic blood pressure, mmHg | 73.8 (11.7) | 72.6 (12.7) | 71.0 (12.6) | 70.4 (13.5) | 69.8 (13.2) |
| Hemoglobin, g/dL | 13.6 (1.5) | 13.0 (1.6) | 12.6 (1.6) | 12.1 (1.7) | 11.7 (1.7) |
| LDL cholesterol, mg/dL | 111.6 (33.2) | 104.8 (34.5) | 102.4 (35.1) | 99.5 (36.7) | 96.8 (35.3) |
| HDL cholesterol, mg/dL | 50.2 (16.0) | 48.4 (15.2) | 47.1 (15.1) | 46.4 (15.4) | 46.3 (15.6) |
| Median fibroblast growth factor-23, RU/mL | 90.7 (67.1–125.9) | 122.1 (88.3–175.4) | 147.1 (105.9–219.0) | 186.9 (123.5–279.9) | 236.1 (152.1–391.6) |
| Serum phosphorus, mg/dL | 3.5 (0.5) | 3.6 (0.6) | 3.7 (0.6) | 3.8 (0.7) | 4.0 (0.8) |
| Median total parathyroid hormone, pg/mL | 39.1 (29.0–55.4) | 45.4 (32.4–69.0) | 55.0 (36.0–84.5) | 68.6 (41.8–107.1) | 84.1 (48.4–151.0) |
| LVEF, % | 55.1 (7.5) | 55.0 (7.9) | 54.7 (8.4) | 53.4 (9.1) | 53.2 (9.5) |
| Left ventricular mass, indexed to height, g/m2.7 | 60.6 (15.3) | 70.4 (22.4) | 71.8 (26.0) | 78.0 (23.6) | 77.1 (21.7) |
Entries are mean (SD) or N (%), except as noted.
GDF-15 AND RISK OF CKD PROGRESSION
The median (interquartile range) length of follow-up time was 5.8 (2.4–8.6) years. There were 1221 participants whose eGFR declined by 50% or progressed to ESRD during follow-up. The incidence rates of CKD progression increased across higher concentrations of GDF-15 (Fig. 1).
Fig. 1. Unadjusted incidence rates of ESRD/50% eGFR decline by cardiac and stress biomarkers in adults with CKD.
In multivariable models, every standard deviation higher GDF-15 was associated with a 50% increased risk of CKD progression [hazard ratio (HR), 1.50 per SD increase of log GDF-15; 95% CI, 1.35–1.67] (Table 2, model 2). There was also an association of higher concentrations of GDF-15 (when modeled in categories) with risk of CKD progression in which the highest quintile was associated with 3-fold higher risk of CKD progression vs the lowest quintile (HR, 3.47; 95% CI, 2.50–4.82, for quintile 5 vs 1, model 2).
Table 2.
Associations of cardiac and stress biomarkers and incident ESRD/50% eGFR decline in adults with CKD.
| Cardiac biomarker | N at risk | N events | Model 1a |
Model 2b |
Model 3 (sensitivity analysis 1)c |
Model 4 (sensitivity analysis 2)d |
||||
|---|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |||
| Continuous predictors | ||||||||||
| Log(NT-proBNP) per 1 SD (1.71) increase | 1.32 (1.21–1.44) | <0.0001 | 1.24 (1.13–1.36) | <0.0001 | 1.17 (1.06, 1.29) | 0.002 | 1.22 (1.11, 1.34) | <0.0001 | ||
| Log(hs-cTnT) per 1 SD (0.82) increase | 1.14 (1.04–1.25) | 0.005 | 1.11 (1.01–1.22) | 0.02 | 1.04 (0.95–1.15) | 0.41 | 1.09 (1.00–1.20) | 0.06 | ||
| Log(GDF-15) per 1 SD (0.59) increase | 1.54 (1.39–1.71) | <0.0001 | 1.50 (1.35–1.67) | <0.0001 | 1.49 (1.34–1.66) | <0.0001 | 1.50 (1.35–1.67) | <0.0001 | ||
| Log(sST-2) per 1 SD (0.56) increase | 1.08 (1.01–1.16) | 0.03 | 1.07 (0.99–1.14) | 0.08 | 1.03 (0.96–1.1) | 0.38 | 1.06 (0.99–1.14) | 0.08 | ||
| Categorical predictors | ||||||||||
| NT-proBNP, ng/L | ||||||||||
| (Reference: ≤37.1) | 733 | 109 | ||||||||
| 37.2–92.5 | 733 | 204 | 1.64 (1.29–2.08) | <0.0001 | 1.58 (1.24–2.01) | <0.0001 | 1.56 (1.23–1.99) | 0.0005 | 1.57 (1.23–2.00) | <0.0001 |
| 92.6–199 | 732 | 236 | 1.70 (1.34–2.14) | 1.62 (1.27–2.05) | 1.56 (1.23–1.98) | 1.60 (1.26–2.03) | ||||
| 200–497 | 733 | 308 | 1.93 (1.50–2.47) | 1.75 (1.35–2.25) | 1.62 (1.25–2.09) | 1.72 (1.33–2.23) | ||||
| >497 | 733 | 364 | 2.36 (1.83–3.04) | 2.05 (1.57–2.67) | 1.81 (1.38–2.38) | 1.98 (1.51–2.61) | ||||
| hsTnT, ng/L | ||||||||||
| (Reference: <10) | 1154 | 226 | ||||||||
| 10.1–16.1 | 825 | 233 | 1.09 (0.90–1.32) | 0.02 | 1.08 (0.89–1.30) | 0.07 | 1.06 (0.88–1.28) | 0.40 | 1.06 (0.88–1.29) | 0.12 |
| 16.2–27.8 | 842 | 303 | 1.01 (0.83–1.25) | 0.98 (0.80–1.20) | 0.94 (0.77–1.15) | 0.96 (0.78–1.17) | ||||
| >27.8 | 843 | 459 | 1.33 (1.06–1.66) | 1.24 (0.99–1.55) | 1.09 (0.87–1.38) | 1.19 (0.95–1.49) | ||||
| GDF-15, pg/mL | ||||||||||
| (Reference: ≤906) | 733 | 73 | ||||||||
| 907–1280 | 733 | 164 | 1.54 (1.16–2.03) | <0.0001 | 1.54 (1.16–2.05) | <0.0001 | 1.55 (1.16–2.06) | <0.0001 | 1.55 (1.16–2.06) | <0.0001 |
| 1281–1720 | 732 | 245 | 1.95 (1.46–2.60) | 1.93 (1.44–2.58) | 1.92 (1.43–2.58) | 1.92 (1.43–2.57) | ||||
| 1721–2410 | 733 | 329 | 2.89 (2.14–3.90) | 2.80 (2.06–3.80) | 2.79 (2.06–3.79) | 2.79 (2.05–3.79) | ||||
| >2410 | 733 | 410 | 3.64 (2.64–5.00) | 3.47 (2.50–4.82) | 3.39 (2.43–4.71) | 3.46 (2.49–4.81) | ||||
| sST-2, ng/mL | ||||||||||
| (Reference: ≤10.5) | 733 | 192 | ||||||||
| 10.6–13.6 | 733 | 203 | 0.99 (0.79–1.23) | 0.05 | 0.97 (0.78–1.21) | 0.17 | 0.98 (0.79–1.21) | 0.71 | 0.98 (0.78–1.22) | 0.15 |
| 13.7–17.2 | 734 | 213 | 0.95 (0.76–1.19) | 0.95 (0.76–1.19) | 0.95 (0.76–1.18) | 0.95 (0.76–1.19) | ||||
| 17.3–22.9 | 731 | 269 | 1.07 (0.86–1.33) | 1.02 (0.82–1.27) | 0.97 (0.78–1.21) | 1.02 (0.82–1.27) | ||||
| > 22.9 | 733 | 344 | 1.27 (1.01–1.58) | 1.19 (0.95–1.50) | 1.09 (0.87–1.36) | 1.20 (0.96–1.51) | ||||
Bold font indicates P < 0.05.
Model 1: age, sex, race, site, diabetes, CVD, smoking, 24-h urinary protein, eGFR, SBP, BMI, LDL, HDL.
Model 2: Model 1 + ACEi/ARBs, diuretics, beta blockers, phosphate, PTH, FGF-23.
Model 3: Model 2 + other biomarkers (NT pro-BP, hsTnT, GDF-15, sST-2).
Model 4: Model 2 + year 1 LVM indexed to height, year 1 ejection fraction.
In the first secondary analysis, when further adjusted for NT-proBNP, hsTnT, and sST2, the associations remained strong and statistically significant (Table 2, model 3). In a second analysis, when adjusted for LVEF and LVMI, the associations were similar as well (Table 2, model 4). Furthermore, when GDF-15 was modeled as a time-updated exposure, the observed associations were similar to the main analysis (see Table 2 in the online Data Supplement). The c statistic of the baseline model 1 was 0.848. With the addition of GDF-15, the c statistic was 0.850.
In sensitivity analyses, the strong association of GDF-15 with CKD progression persisted when excluding participants with prevalent cardiovascular disease (Table 3, model 2).
Table 3.
Associations of cardiac and stress biomarkers and incident ESRD/50% eGFR decline in the CRIC Study, among participants without prevalent cardiovascular disease at baseline (N = 2457).
| Cardiac biomarker | N at risk | N events | Model 1a |
Model 2b |
||
|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |||
| Continuous predictors | ||||||
| Log(NT-proBNP) per 1 SD (1.71) increase | 1.32 (1.18–1.48) | <0.0001 | 1.25 (1.11–1.41) | 0.0003 | ||
| Log(hsTnT) per 1 SD (0.82) increase | 1.06 (0.95–1.19) | 0.27 | 1.04 (0.93–1.16) | 0.47 | ||
| Log(GDF-15) per 1 SD (0.59) increase | 1.53 (1.34–1.74) | <0.0001 | 1.49 (1.31–1.70) | <0.0001 | ||
| Log(sST-2) per 1 SD (0.56) increase | 1.06 (0.97–1.16) | 0.18 | 1.05 (0.96–1.14) | 0.28 | ||
| Categorical predictors | ||||||
| NT-proBNP, ng/L | ||||||
| (Reference: ≤37.1) | 633 | 97 | ||||
| 37.2–92.5 | 582 | 157 | 1.58 (1.21–2.07) | <0.0001 | 1.51 (1.15–1.99) | 0.0002 |
| 92.6–199 | 522 | 178 | 1.79 (1.38–2.33) | 1.69 (1.29–2.21) | ||
| 200–497 | 430 | 187 | 1.91 (1.41–2.57) | 1.73 (1.27–2.35) | ||
| >497 | 290 | 153 | 2.48 (1.82–3.38) | 2.15 (1.55–2.99) | ||
| hsTnT, ng/L | ||||||
| (Reference: <10) | 952 | 189 | ||||
| 10.1–16.1 | 591 | 162 | 1.10 (0.88–1.38) | 0.75 | 1.07 (0.85–1.33) | 0.77 |
| 16.2–27.8 | 491 | 179 | 0.99 (0.77–1.28) | 0.94 (0.73–1.21) | ||
| >27.8 | 423 | 242 | 1.08 (0.81–1.44) | 1.01 (0.76–1.34) | ||
| GDF-15, pg/mL | ||||||
| (Reference: ≤906) | 638 | 65 | ||||
| 907–1280 | 544 | 135 | 1.64 (1.21–2.23) | <0.0001 | 1.62 (1.18–2.22) | <0.0001 |
| 1281–1720 | 484 | 158 | 1.82 (1.31–2.53) | 1.77 (1.27–2.48) | ||
| 1721–2410 | 396 | 187 | 2.77 (1.94–3.94) | 2.65 (1.85–3.82) | ||
| >2410 | 395 | 227 | 3.33 (2.29–4.84) | 3.16 (2.15–4.65) | ||
| sST-2, ng/mL | ||||||
| (Reference: ≤10.5) | 555 | 138 | ||||
| 10.6–13.6 | 503 | 132 | 1.06 (0.81–1.39) | 0.18 | 1.04 (0.80–1.37) | 0.29 |
| 13.7–17.2 | 504 | 130 | 0.96 (0.72–1.27) | 0.96 (0.72–1.27) | ||
| 17.3–22.9 | 453 | 161 | 1.05 (0.80–1.37) | 0.98 (0.74–1.30) | ||
| >22.9 | 442 | 211 | 1.31 (0.98–1.76) | 1.24 (0.92–1.67) | ||
Model 1: age, sex, race, site, diabetes, smoking, 24-h urinary protein, eGFR, SBP, BMI, LDL, HDL.
Model 2: Model 1 + ACEi/ARBs, diuretics, beta blockers, phosphate, PTH, FGF-23.
sST-2 AND RISK OF CKD PROGRESSION
The incidence rate for CKD progression was highest for the highest quintile of sST2 (Fig. 1). In models adjusted for cardiovascular risk factors, there was a modest association of higher sST2 with CKD progression (HR, 1.08 per SD increase of log sST2; 95% CI, 1.01–1.16) (Table 2, model 1). However, with adjustment for receipt of cardiovascular medications and mineral metabolism markers, this association was attenuated and no longer statistically significant.
There was no statistically significant association when models were adjusted for the alternative biomarkers or echocardiographic measures (Table 2, Models 3 and 4), Similarly, the association of sST2 was not statistically significant when the population was restricted to participants without cardiovascular disease (Table 3) or when sST2 was modeled as a time-updated exposure in adjusted models (see Table 2 in the online Data Supplement). The c statistic of the baseline model 1 was 0.848. With the addition of sST2, the c statistic was 0.846.
NT-proBNP AND RISK OF CKD PROGRESSION
The incidence rates of CKD progression were higher among participants with higher concentrations of NT-proBNP (Fig. 1).
In models adjusted for cardiovascular risk factors, every standard deviation higher NT-proBNP was associated with 32% greater risk of CKD progression (HR, 1.32 per SD higher in log NT-proBNP; 95% CI, 1.21–1.44) (Table 2, model 1). This association remained strong when further adjusted for cardiovascular medication use and mineral metabolism markers (Table 2, model 2).
In a secondary analysis, when adjusting for alternative biomarkers and echocardiographic measures, we found similar associations of NT-proBNP and CKD progression (Table 2, Models 3 and 4). Time-updated NT-proBNP was also similarly associated with CKD progression compared with the main results (see Table 2 in the online Data Supplement). The c statistic of the baseline model 1 was 0.848. With the addition of NT-proBNP, the c statistic was 0.848.
When we excluded participants with prevalent cardiovascular disease, the association of NT-proBNP with CKD progression remained statistically significant (Table 3, model 2).
hsTnT AND RISK OF CKD PROGRESSION
The incidence rates of CKD progression were higher among participants with higher concentrations of hsTnT (Fig. 1).
When adjusted for cardiovascular risk factors, there was a modest association of higher hsTnT with risk of CKD progression (Table 2, model 1). After further adjusted for use of cardiovascular medications and mineral metabolism markers, this association remained statistically significant (HR, 1.11 per SD increase of log hsTnT; 95% CI, 1.01–1.22, model 2).
In secondary analyses that adjusted for NT-proBNP, GDF-15, and sST2 or LVMI and LVEF, the association of hsTnT with CKD progression was not statistically significant (Table 2). In contrast, time-updated hsTnT was significantly associated with risk of CKD progression (see Table 2 in the online Data Supplement), whereas it was not among the subset of participants without prevalent cardiovascular disease (Table 3). The c statistic of the baseline model 1 was 0.848. With the addition of hsTnT, the c statistic was 0.846.
SECONDARY ANALYSIS: COMBINATION OF NT-proBNP AND GDF-15
For the strongest biomarkers, NT-proBNP and GDF-15, we modeled whether combinations of the biomarkers were differentially associated with CKD progression. Participants with high concentrations of both biomarkers (defined as the top quintile) had nearly 2-fold increased risk of CKD progression compared with those with low concentrations of both biomarkers or high concentration of only 1 biomarker (see Table 3 in the online Data Supplement).
Discussion
In this large study of over 3000 adults with CKD, we found strong associations of increases in GDF-15, NT-proBNP, and hsTnT (but not sST2) with CKD progression. The findings for GDF-15 and NT-proBNP (but not hsTnT) were robust when adjusting for important potential confounders; alternative biomarkers or measures of subclinical heart failure. These findings highlight possible shared cardiac, stress, and kidney mechanisms that contribute to the progression of kidney disease.
The association of increased GDF-15 with CKD progression was strong and greater than that observed with NT-proBNP. GDF-15 is a member of transforming growth factor β cytokine superfamily, which is widely produced in cardiomyocytes, adipocytes, and macrophages and may reflect an early response protein induced after tissue injury (11). In the heart, GDF-15 activates pathways responsible for cardioprotection (9–11). Preclinical studies show protective effects of GDF-15 in the kidney as well. In mice models of type 1 and type 2 diabetes, GDF-15 knock out mice displayed increased tubular damage, with evidence of glucosuria and polyuria, and increased interstitial damage. Further data have suggested that GDF-15 modulates the renal extracellular matrix production and also enhances cellular proliferation of tubular epithelial cells (28). Intrarenal GDF-15 may also prevent organ damage by minimizing recruitment of inflammatory cells. Despite the “protective” role of GDF-15 production in preclinical studies, clinical studies suggest that higher concentrations of GDF-15 are associated with worse outcomes (29, 30). We previously reported a strong correlation of intrarenal GDF1511 expression (in the tubulointerstitial compartment) with circulating GDF-15 concentrations in patients with CKD; in this analysis, higher circulating GDF-15 was also significantly associated with CKD progression (14). Among Framingham participants, every log increase in GDF-15 was associated with increased 2-fold greater odds of incident CKD and rapid decline of kidney function (15). Our study supports and extends the findings of previous studies by studying a larger cohort of CKD patients. Further translational work is needed to bridge preclinical and clinical data to fully elucidate the role of GDF-15 in the kidney.
We did not find a significant association between sST2 with CKD progression. Models have shown that sST2, an IL-1-like cytokine receptor for IL-33, is produced by cardiac myocytes (among other cell types) in response to myocardial infarction (31) and mediates injury by neutralizing the cardioprotective effects of IL-33, which reduces cardiac hypertrophy, fibrosis, and apoptosis (13, 32). A clinical study showed a trend toward an association of higher sST2 with incident microalbuminuria in participants without CKD (15). Furthermore, our previous study of older adults did not find a significant association of higher sST2 with kidney function decline (33) but was limited largely to participants without CKD, which differs from our present analysis.
Higher NT-proBNP concentrations reflect a wide range of cardiac pathophysiology including changes in myocardial pressure. Similar to our study, others have reported strong associations between higher NT-proBNP and decline in kidney function (34, 35). In the TREAT trial, participants with CKD in the highest quartile of NT-proBNP had a nearly 4-fold risk of ESRD compared to those in the lowest quartile (18). We also found an association of baseline and time-updated hsTnT with CKD progression in our study. Data from prior studies show conflicting results. In the Framingham offspring cohort, TnT was not associated with rapid kidney function decline or incident CKD (15). In contrast, a study of older adults reported a strong association of hsTnT with incident CKD (16). Further work is needed to determine whether reductions in NT-proBNP and hsTnT can slow or prevent CKD progression (36–40).
It is less likely that NT-proBNP and GDF-15 may discriminate patients at the highest risk of CKD progression because there was not a substantial improvement in the c statistic with addition of these biomarkers to a baseline clinical model. However, these data may promote additional studies to understand the molecular mechanisms that underlie CKD progression, which may lead the way for novel therapeutics.
Our study had several strengths. We studied a well-characterized diverse cohort of participants with CKD. Changes in eGFR were determined with standardized, annual serum creatinine measures. We also recognize some limitations. It is possible that participants had nonsustained fluctuations in eGFR or acute kidney injury, which could have misclassified the outcome of CKD progression. The accuracy of the chosen assays for our study have not been compared to other assays in CKD patients nor are there established clinical cutoffs in this population. Echocardiograms were performed one year after baseline so were not concurrent with the biomarker measures. We cannot determine causality or mechanisms from this observational study. The study was conducted among volunteer research participants, so the findings may not be generalizable to all CKD patients.
In conclusion, increases in GDF-15 and NT-proBNP are associated with greater risk for CKD progression among participants with CKD. hsTnT was also associated with CKD progression, but only among participants with underlying cardiovascular disease. Cardiac and stress biomarkers may identify novel mechanisms underlying kidney injury and may be promising tools for disease monitoring and development of targeted interventions to slow the progression of CKD.
10 Nonstandard abbreviations
- CKD
chronic kidney disease
- NT-proBNP
N-terminal pro-B-type natriuretic peptide
- hsTnT
high-sensitivity troponin T
- GDF-15
growth differentiation factor-15
- sST2
soluble (fraction of) suppression of tumorigenicity 2
- IL
interleukin
- CRIC
Chronic Renal Insufficiency Cohort
- ESRD
end-stage renal disease
- BMI
body mass index
- FGF
fibroblast growth factor
- PTH
parathyroid hormone
- SBP
systolic blood pressure
- LVMI
left ventricular mass index
- ACEi/ARBs
angiotensin-converting enzyme inhibitors/angiotensis II receptor blockers
- LVEF
left ventricular ejection fraction
- HR
hazard ratio.
11 Human gene
- GDF15
growth differentiation factor 15.
Footnotes
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.
N. Bansal, financial support, statistical analysis, administrative support, provision of study material or patients; L. Zelnick, statistical analysis; A. Anderson, financial support; J. Kusek, financial support; A.S. Go, provision of study material or patients.
Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:
Employment or Leadership: R.H. Christenson, American College of Cardiology, The Journal of Applied Laboratory Medicine, AACC; C. deFilippi, The Journal of Applied Laboratory Medicine, AACC.
Consultant or Advisory Role: R.H. Christenson, Roche Diagnostics, Siemens Healthineers, Quidel, Beckman Coulter; C. deFilippi, Roche Diagnostics, Ortho Diagnostics, Siemens Healthineers; H. Feldman, Kyowa Hakko Kirin Co, Ltd.
Stock Ownership: None declared.
Honoraria: R.H. Christenson, Roche Diagnostics, Siemens Healthineers, Beckman Coulter, Becton Dickinson; H. Feldman, University of Wisconsin Madison Grand Rounds, The Gloria and Burton Rose lecture: Medicine Grand Rounds at NYU, Health Research and Policy Departmental Review at Stanford University, 2018 Congress of Nephrology and Dialysis in Dalat, Vietnam.
Research Funding: N. Bansal, R01 DK103612; A. Anderson, NIDDK; R.H. Christenson, Roche Diagnostics, Siemens Healthineers, Beckman Coulter, and Quidel to institution; S. Seliger, Roche Diagnostics, Abbott Diagnostics; A.S. Go, National Institute of Diabetes, Digestive and Kidney Diseases. Roche Diagnostics provided partial funding for the NT-proBNP and hsTnT assays. Funding for the CRIC Study was obtained under a cooperative agreement from National Institute of Diabetes and Digestive and Kidney Diseases (U01DK060990, U01DK060984, U01DK061022, U01DK061021, U01DK061028, U01DK060980, U01DK060963, and U01DK060902). In addition, this work was supported in part by: the Perelman School of Medicine at the University of Pennsylvania Clinical and Translational Science Award NIH/NCATS UL1TR000003, Johns Hopkins University UL1 TR-000424, University of Maryland GCRC M01 RR-16500, Clinical and Translational Science Collaborative of Cleveland, UL1TR000439 from the National Center for Advancing Translational Sciences (NCATS) component of the National Institutes of Health and NIH roadmap for Medical Research, Michigan Institute for Clinical and Health Research (MICHR) UL1TR000433, University of Illinois at Chicago CTSA UL1RR029879, Tulane COBRE for Clinical and Translational Research in Cardiometabolic Diseases P20 GM109036, Kaiser Permanente NIH/NCRR UCSF- CTSI UL1 RR-024131.
Expert Testimony: None declared.
Patents: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, preparation of manuscript, or final approval of manuscript.
References
- 1. Kistorp C, Raymond I, Pedersen F, Gustafsson F, Faber J, Hildebrandt P. N-terminal pro-brain natriuretic peptide, C-reactive protein, and urinary albumin levels as predictors of mortality and cardiovascular events in older adults. JAMA 2005;293:1609–16. [DOI] [PubMed] [Google Scholar]
- 2. Saunders JT, Nambi V, de Lemos JA, Chambless LE, Virani SS, Boerwinkle E, et al. Cardiac troponin T measured by a highly sensitive assay predicts coronary heart disease, heart failure, and mortality in the Atherosclerosis Risk in Communities Study. Circulation 2011;123:1367–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Yasue H, Yoshimura M, Sumida H, Kikuta K, Kugiyama K, Jougasaki M, et al. Localization and mechanism of secretion of B-type natriuretic peptide in comparison with those of a-type natriuretic peptide in normal subjects and patients with heart failure. Circulation 1994;90:195–203. [DOI] [PubMed] [Google Scholar]
- 4. de Lemos JA, Drazner MH, Omland T, Ayers CR, Khera A, Rohatgi A, et al. Association of troponin T detected with a highly sensitive assay and cardiac structure and mortality risk in the general population. JAMA 2010;304:2503–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Daniels LB, Clopton P, Laughlin GA, Maisel AS, Barrett-Connor E. Growth-differentiation factor-15 is a robust, independent predictor of 11-year mortality risk in community-dwelling older adults: the Rancho Bernardo Study. Circulation 2011;123:2101–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Wang TJ, Wollert KC, Larson MG, Coglianese E, McCabe EL, Cheng S, et al. Prognostic utility of novel biomarkers of cardiovascular stress: the Framingham Heart Study. Circulation 2012;126:1596–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Weinberg EO, Shimpo M, Hurwitz S, Tominaga S, Rouleau JL, Lee RT. Identification of serum soluble ST2 receptor as a novel heart failure biomarker. Circulation 2003;107:721–6. [DOI] [PubMed] [Google Scholar]
- 8. Emmerson PJ, Duffin KL, Chintharlapalli S, Wu X. GDF15 and growth control. Front Physiol 2018;9:1712. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bonaterra GA, Zugel S, Thogersen J, Walter SA, Haberkorn U, Strelau J, Kinscherf R. Growth differentiation factor-15 deficiency inhibits atherosclerosis progression by regulating interleukin-6-dependent inflammatory response to vascular injury. J Am Heart Assoc 2012;1:e002550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kempf T, Eden M, Strelau J, Naguib M, Willenbockel C, Tongers J, et al. The transforming growth factor-beta superfamily member growth-differentiation factor-15 protects the heart from ischemia/reperfusion injury. Circ Res 2006;98:351–60. [DOI] [PubMed] [Google Scholar]
- 11. Preusch MR, Baeuerle M, Albrecht C, Blessing E, Bischof M, Katus HA, Bea F. GDF-15 protects from macrophage accumulation in a mouse model of advanced atherosclerosis. Eur J Med Res 2013;18:19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Wollert KC, Kempf T, Wallentin L. Growth differentiation factor 15 as a biomarker in cardiovascular disease. Clin Chem 2017;63:140–51. [DOI] [PubMed] [Google Scholar]
- 13. Pusceddu I, Dieplinger B, Mueller T. ST2 and the ST2/IL-33 signalling pathway-biochemistry and pathophysiology in animal models and humans. Clin Chim Acta 2019;495:493–500. [DOI] [PubMed] [Google Scholar]
- 14. Nair V, Robinson-Cohen C, Smith MR, Bellovich KA, Bhat ZY, Bobadilla M, et al. Growth differentiation factor-15 and risk of CKD progression. J Am Soc Nephrol 2017;28:2233–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Ho JE, Hwang SJ, Wollert KC, Larson MG, Cheng S, Kempf T, et al. Biomarkers of cardiovascular stress and incident chronic kidney disease. Clin Chem 2013;59:1613–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bansal N, Katz R, Dalrymple L, de Boer I, DeFilippi C, Kestenbaum B, et al. NT-probnp and troponin T and risk of rapid kidney function decline and incident CKD in elderly adults. Clin J Am Soc Nephrol 2015;10:205–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tung YC, Chang CH, Chen YC, Chu PH. Combined biomarker analysis for risk of acute kidney injury in patients with st-segment elevation myocardial infarction. PLoS One 2015;10:e0125282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Desai AS, Toto R, Jarolim P, Uno H, Eckardt KU, Kewalramani R, et al. Association between cardiac biomarkers and the development of ESRD in patients with type 2 diabetes mellitus, anemia, and CKD. Am J Kidney Dis 2011;58:717–28. [DOI] [PubMed] [Google Scholar]
- 19. Feldman HI, Appel LJ, Chertow GM, Cifelli D, Cizman B, Daugirdas J, et al. The Chronic Renal Insufficiency Cohort (CRIC) study: design and methods. J Am Soc Nephrol 2003;14:S148–53. [DOI] [PubMed] [Google Scholar]
- 20. Lash JP, Go AS, Appel LJ, He J, Ojo A, Rahman M, et al. Chronic Renal Insufficiency Cohort (CRIC) study: baseline characteristics and associations with kidney function. Clin J Am Soc Nephrol 2009;4:1302–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 2010;56:254–61. [DOI] [PubMed] [Google Scholar]
- 22. Linnet K. Necessary sample size for method comparison studies based on regression analysis. Clin Chem 1999;45:882–94. [PubMed] [Google Scholar]
- 23. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey anthropometry procedures manual. Centers for disease control and prevention (serial online) 2000. https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf (Accessed June 2019).
- 25. Royston P. Multiple imputation of missing values. Stata Journal 2004;4:227–41. [Google Scholar]
- 26. Rubin DB. Multiple imputation for nonresponse in surveys. New York (NY): Wiley; 1987. [Google Scholar]
- 27. Heagerty PJ, Zheng Y. Survival model predictive accuracy and ROC curves. Biometrics 2005;61:92–105. [DOI] [PubMed] [Google Scholar]
- 28. Mazagova M, Buikema H, van Buiten A, Duin M, Goris M, Sandovici M, et al. Genetic deletion of growth differentiation factor 15 augments renal damage in both type 1 and type 2 models of diabetes. Am J Physiol Renal Physiol 2013;305:F1249–64. [DOI] [PubMed] [Google Scholar]
- 29. Kahli A, Guenancia C, Zeller M, Grosjean S, Stamboul K, Rochette L, et al. Growth differentiation factor-15 (GDF-15) levels are associated with cardiac and renal injury in patients undergoing coronary artery bypass grafting with cardiopulmonary bypass. PLoS One 2014;9:e105759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lajer M, Jorsal A, Tarnow L, Parving HH, Rossing P. Plasma growth differentiation factor-15 independently predicts all-cause and cardiovascular mortality as well as deterioration of kidney function in type 1 diabetic patients with nephropathy. Diabetes Care 2010;33:1567–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Weinberg EO, Shimpo M, De Keulenaer GW, MacGillivray C, Tominaga S, Solomon SD, et al. Expression and regulation of ST2, an interleukin-1 receptor family member, in cardiomyocytes and myocardial infarction. Circulation 2002;106:2961–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Schmitz J, Owyang A, Oldham E, Song Y, Murphy E, McClanahan TK, et al. Il-33, an interleukin-1-like cytokine that signals via the IL-1 receptor-related protein ST2 and induces T helper type 2-associated cytokines. Immunity 2005;23:479–90. [DOI] [PubMed] [Google Scholar]
- 33. Bansal N, Katz R, Seliger S, DeFilippi C, Sarnak MJ, Delaney JA, et al. Galectin-3 and soluble ST2 and kidney function decline in older adults: the Cardiovascular Health Study (CHS). Am J Kidney Dis 2016;67:994–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Yasuda K, Kimura T, Sasaki K, Obi Y, Iio K, Yamato M, et al. Plasma B-type natriuretic peptide level predicts kidney prognosis in patients with predialysis chronic kidney disease. Nephrol Dial Transplant 2012;27:3885–91. [DOI] [PubMed] [Google Scholar]
- 35. Spanaus K-S, Kronenberg F, Ritz E, Schlapbach R, Fliser D, Hersberger M, et al. B-type natriuretic peptide concentrations predict the progression of nondiabetic chronic kidney disease: the Mild-to-Moderate Kidney Disease Study. Clin Chem 2007;53:1264–72. [DOI] [PubMed] [Google Scholar]
- 36. Pfisterer M, Buser P, Rickli H, Gutmann M, Erne P, Rickenbacher P, et al. BNP-guided vs symptom-guided heart failure therapy: the trial of intensified vs standard medical therapy in elderly patients with congestive heart failure (TIME-CHF) randomized trial. JAMA 2009;301:383–92. [DOI] [PubMed] [Google Scholar]
- 37. Jourdain P, Jondeau G, Funck F, Gueffet P, Le Helloco A, Donal E, et al. STARS-BNP multicenter study. J Am Coll Cardiol 2007;49:1733–9. [DOI] [PubMed] [Google Scholar]
- 38. Ledwidge M, Gallagher J, Conlon C, Tallon E, O'Connell E, Dawkins I, et al. Natriuretic peptide-based screening and collaborative care for heart failure: the STOP-HF randomized trial. JAMA 2013;310:66–74. [DOI] [PubMed] [Google Scholar]
- 39. Zhan F, Zeng XY, Zhang X, Zhou S, Xu Y, Zhang JP, Jiao YJ. [Preparation and identification of monoclonal antibody against human growth differentiation factor 15]. Xi bao yu fen zi mian yi xue za zhi = Chinese J Cell Mol Immunol 2011;27:539–41, 44. [PubMed] [Google Scholar]
- 40. Lee HY, Rhee CK, Kang JY, Byun JH, Choi JY, Kim SJ, et al. Blockade of IL-33/ST2 ameliorates airway inflammation in a murine model of allergic asthma. Experimental lung research 2014;40:66–76. [DOI] [PubMed] [Google Scholar]