Growth Differentiation Factor 15 and Clinical Outcomes in Japanese Patients With Heart Failure (2024)

Abstract

Background: Heart failure (HF) is an increasing health problem associated with a high mortality rate. Growth differentiation factor (GDF) 15, a stress response cytokine belonging to the transforming growth factor-β superfamily, is associated with poor clinical outcomes in a broad spectrum of cardiovascular diseases. However, the prognostic usefulness of GDF15 in Japanese patients with HF remains unclear.

Methods and Results: We measured serum concentrations of GDF15 and B-type natriuretic peptide (BNP) in 1,201 patients with HF. All patients were prospectively followed for a median period of 1,309 days. In all, 319 HF-related events and 187 all-cause deaths occurred during the follow-up period. Kaplan-Meier analysis demonstrated that, among GDF15 tertiles, the highest tertile group had the greatest risk of HF-related events and all-cause mortality. Multivariate Cox proportional hazard regression analysis demonstrated that the serum GDF15 concentration was an independent predictor of HF-related events and all-cause deaths after adjusting for confounding risk factors. Serum GDF15 improved the prediction capacity for all-cause deaths and HF-related events with a significant net reclassification index and integrated discrimination improvement. Subgroup analysis in patients with HF with preserved ejection fraction also showed the prognostic usefulness of GDF15.

Conclusions: Serum GDF15 concentrations were associated with HF severity and clinical outcomes, indicating that GDF15 could provide additional clinical information to track the health status of patients with HF.

Despite advances in treatment, heart failure (HF) remains a public health problem with high morbidity and mortality.1 Therefore, the early identification and risk stratification of high-risk patients with HF are critical. Biomarkers play an important role in the prevention, diagnosis, prognosis, and risk stratification of patients with HF.2 According to the 2017 Japanese Circulation Society/Japanese Heart Failure Society guidelines for the diagnosis and treatment of acute and chronic heart failure, myocardial injury and inflammation biomarkers are expected to be useful for the assessment of prognosis.3

Growth differentiation factor (GDF) 15 is a member of the transforming growth factor-β superfamily. Although GDF15 is appreciably expressed in the liver and placenta at baseline, circulating GDF15 concentrations are generally low in healthy individuals. GDF15 is a transcriptional target gene for p53 that is induced by several factors, including oxidative stress, inflammation, and mechanical stress, in various tissues4 and is released into the circulation.5^,6 Because circulating GDF15 is reportedly elevated in advanced age, cancer, and cardiovascular disease,7 GDF15 is considered a stress response cytokine. Notably, GDF15 is reported to reflect myocardial cell injury and inflammation in HF and is elevated in patients with HF.8 GDF15 measurement is used clinically for risk stratification of clinical outcomes in Western countries in acute coronary syndrome and chronic HF. However, only limited data regarding GDF15 are available for Japanese patients with HF.9 Furthermore, the prognostic usefulness of GDF15 has not yet been fully elucidated in patients with HF with preserved ejection fraction, which is the major subtype of HF in Asian patients.

The aim of the present study was to examine the association of serum GDF15 concentrations with HF severity and clinical outcomes, such as HF-related events and mortality, in Japanese patients with HF.

Methods

Ethics Statement

All procedures were performed in accordance with the ethical, institutional, and/or national research committee guidelines of the centers at which the studies were conducted, and all procedures complied with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Institutional Ethics Committee of Yamagata University School of Medicine (Yamagata University, 2021-298).

Study Participants

This prospective observational study is part of an ongoing Yamagata cardiovascular and cerebrovascular prognosis survey aimed at elucidating the association between serum GDF15 concentrations and the clinical outcomes of patients with HF. After obtaining written informed consent, blood samples were collected and stored at −80℃ until measurement. All patients were prospectively followed-up twice a year from the day after obtaining informed consent up to 2,500 days, using telephone or medical records.

We enrolled 1,201 consecutive patients with HF admitted to our hospital for the diagnosis or treatment of acute and chronic HF exacerbation between 2009 and 2019. Transthoracic echocardiography was performed by physicians and sonographers who were blinded to biochemical data. Left ventricular ejection fraction (LVEF) and left ventricular (LV) end-diastolic diameter were measured. The LV mass index was also calculated according to previous reports.10^,11 The diagnosis of HF was made by 2 cardiologists who used the generally accepted Framingham criteria, including a history of dyspnea and symptomatic exercise intolerance with signs of pulmonary congestion or peripheral edema and radiological or echocardiographic evidence of LV enlargement or dysfunction.12 Diagnoses of hypertension, diabetes, and dyslipidemia were established based on medical records or a history of medical therapy. Exclusion criteria were acute coronary syndrome within 3 months preceding admission, active hepatic disease, malignant disease, and hemodialysis. Demographic and clinical data, including age, sex, New York Heart Association (NYHA) functional class, and medications at discharge, were collected from patients’ medical records and interviews.

Measurement of Serum GDF15 Concentrations

Blood samples were collected in the early morning within 24 h of hospital admission. The samples were centrifuged at 3,000 r.p.m. for 15 min at 4℃, and the resulting blood sample fractions were stored at −80℃ until analysis. The GDF15 assay was performed using an electrochemiluminescent immunoassay (Roche Diagnostics, Japan). Blood samples were also obtained to measure B-type natriuretic peptide (BNP) concentrations and were immediately transferred to chilled tubes containing 4.5 mg EDTA disodium salt and aprotinin (500 U/mL) and centrifuged at 1,000 g at 4℃ for 15 min. The clarified plasma samples were frozen, stored at −70℃, and thawed immediately before the assay. The BNP concentration was measured using a chemiluminescence immunoassay.

Definition of Chronic Kidney Disease (CKD) and Anemia

Urine and venous blood samples were obtained in the early morning within 24 h of admission. Urinary albumin levels were quantitatively measured by immunoturbidimetry in a single spot urine specimen collected early in the morning. Urinary albumin levels were separately corrected for urinary creatinine to the urinary microalbumin-creatinine ratio (UACR). The estimated glomerular filtration rate (eGFR) was calculated using diet modification in the renal disease equation with the Japanese coefficient, as reported previously.13 CKD was defined as reduced eGFR (<60 mL/min/1.73 m²) or microalbuminuria (UACR ≥30 mg/g creatinine) according to the National Kidney Foundation Kidney Disease Outcomes Quality Initiative clinical guidelines.14 Hemoglobin (Hb) levels were measured simultaneously. Anemia was defined as Hb <13 g/dL in men and <12 g/dL in women, according to World Health Organization guidelines.15

Cardiovascular Risk Factors

Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or the use of antihypertensive medication. Diabetes was defined as a fasting blood sugar level ≥126 mg/dL, HbA1c ≥6.5% (NGSP), or the use of antidiabetic medication. Dyslipidemia was defined as high-density lipoprotein cholesterol <40 mg/dL, low-density lipoprotein cholesterol ≥140 mg/dL, triglyceride ≥150 mg/dL, or the use of lipid-lowering medication.

Endpoints and Follow-up Period

All patients were prospectively followed up by telephone or by reviewing their medical records twice a year for a median period of 1,309 days (interquartile range [IQR] 709–2,500 days). The primary endpoint was HF-related events. HF-related events were defined as hospitalizations due to HF progression and death. The secondary endpoint was all-cause deaths.

Statistical Analysis

The normality of continuous variables was confirmed using the Shapiro-Wilk test. Results are expressed as the mean±SD for continuous variables and as numbers (percentages) for categorical variables. Skewed values are presented as the median with IQR. Continuous and categorical variables were compared using t-tests and Chi-squared tests, respectively. Differences in age, eGFR, Hb, LVEF, and LV end-diastolic dimension among the groups were analyzed using analysis of variance (ANOVA) with Tukey’s post hoc test. Differences in BNP, C-reactive protein (CRP), and GDF15 concentrations among groups were analyzed using the Steel-Dwass test. The associations of serum GDF15 concentrations with NYHA functional class and aging per 5-year increase were assessed using the Kruskal-Wallis test. The correlation of serum GDF15 with BNP, Hb, and eGFR was analyzed using simple linear regression analysis. Differences in log₁₀-transformed BNP and GDF15 concentrations among HF subtypes were assessed using ANOVA with Tukey’s post hoc test. Survival curves were plotted using the Kaplan-Meier method and compared using the log-rank test. Owing to the skewed distribution of BNP, CRP, and GDF15 values, we used log₁₀-transformed BNP, CRP, and GDF15 values for univariate and multivariate Cox proportional hazard regression analyses. Cox proportional hazard analysis was used to determine independent predictors of HF-related events and all-cause deaths. Multicollinearity was assessed using the variance inflation factor. Significant predictors selected in the univariate Cox proportional hazard regression analysis were entered into multivariate analysis for HF-related events and all-cause mortality. Receiver operating characteristic (ROC) curves for HF-related events and all-cause mortality were plotted and used to measure the predictive accuracy of serum GDF15 concentration for HF-related events and all-cause mortality. We calculated the net reclassification index (NRI) and integrated discrimination improvement (IDI) to measure the incremental value of serum GDF15 in classifying patients at high or low risk of HF-related events and all-cause mortality. Statistical significance was set at P<0.05. All statistical analyses were performed using the standard software packages JMP^®(version 14.2; SAS Institute, Cary, NC, USA) and EZR (Saitama Medical Center, Jichi Medical University, Shimotsuke, Japan).

Results

Baseline Characteristics of All Patients and Comparisons of Clinical Characteristics Among GDF15 Tertiles

The baseline characteristics of the 1,201 patients are summarized in Table 1. The etiology of HF was ischemic heart disease, dilated cardiomyopathy, and other in 423 (35%), 157 (13%), and 621 (52%) patients, respectively. Atrial fibrillation (AF) was identified in 418 (34.8%) patients. There were 508, 367, and 326 patients with NYHA functional classes II, III, and IV, respectively. Hypertension, diabetes, dyslipidemia, and obesity were identified in 933 (77.7%), 463 (38.6%), 645 (53.7%), and 311 (28.1%) patients, respectively. There were 674 (56.1%) patients with anemia and 767 (63.9%) with CKD. The median BNP, CRP, and GDF15 concentrations were 315 pg/mL, 0.31 mg/L, and 2,631 pg/mL, respectively.

As shown in Figure 1, patients were divided into 3 groups according to serum GDF15 tertiles: the first tertile (T1; GDF15 <1,944 pg/mL; n=400), the second tertile (T2; GDF15 1,944–3,802 pg/mL; n=400), and the third tertile (T3; GDF15 >3,802 pg/mL; n=401). The T3 group had a more severe NYHA functional class and a higher prevalence of diabetes, anemia, and CKD, and a lower prevalence of dilated cardiomyopathy and obesity compared with the other 2 groups. In addition, the T3 group had lower Hb, eGFR, and LVEF and higher BNP and CRP concentrations than the other 2 groups. The T3 group took fewer angiotensin-converting enzyme inhibitors and/or angiotensin receptor blockers, and more diuretics and mineralocorticoid receptor antagonists than the other groups. The T3 group was older than the T1 group (Table 1).

The T2 group was older and had a higher prevalence of AF, diabetes, dyslipidemia, anemia, and CKD than the T1 group. Furthermore, the T2 group had lower Hb, eGFR, and LVEF and higher BNP concentrations than the T1 group (Table 1).

Association of Serum GDF15 Concentrations With HF Severity and HF Subtypes

Serum GDF15 concentrations increased with advancing NYHA functional class (Figure 2A). The median serum GDF15 concentrations in NYHA classes II, III, and IV were 1,943, 3,269, and 3,493 pg/mL, respectively. Serum GDF15 concentrations increased with age in patients with HF (Figure 2B). As shown in Figure 2C–E, serum GDF15 concentrations were weakly but significantly correlated with BNP, Hb, and eGFR. In contrast, there was a very weak correlation between GDF15 and CRP (R²=0.1618, P<0.0001). There was no significant correlation between GDF15 and the LV mass index (R²=0.0031, P=0.2119).

The basic principles for the treatment of HF differ by HF subtype.16 Therefore, we compared BNP and GDF15 concentrations among HF subtypes, namely HF with reduced ejection fraction (HFrEF), HF with mildly reduced ejection fraction, and HF with preserved ejection fraction (HFpEF), in patients with moderate and severe HF. Interestingly, BNP concentrations were affected by the HF subtype (Figure 3A,B), and BNP concentrations were lower in patients with HFpEF than in those with other HF subtypes. Although GDF15 concentrations were lower in patients with HFpEF than in those with other subtypes of moderate HF (Figure 3C), GDF15 levels did not differ by HF subtype in patients with severe HF (Figure 3D).

Serum GDF15 Concentrations and Clinical Outcomes in Patients With HF

During the follow-up period, 319 HF-related events and 187 all-cause deaths occurred. As shown in Figure 4, Kaplan-Meier analysis revealed that the T3 group had the greatest risk of HF-related events and all-cause mortality among the 3 groups.

ROC analysis demonstrated that the C index of serum GDF15 for HF-related events was significantly larger than that of BNP and Hb. In addition, the C index of serum GDF15 for all-cause deaths was significantly larger than that of Hb and eGFR (Table 2). Subgroup analysis in patients with HFpEF, demonstrated that the C index of serum GDF15 for HF-related events was significantly larger than that of BNP, Hb, and eGFR. In addition, the C index of serum GDF15 for all-cause deaths was significantly larger than that of Hb and eGFR (Table 2).

Univariate Cox proportional hazard regression analysis demonstrated that serum GDF15 concentrations were significantly associated with HF-related events and all-cause mortality. Furthermore, age, NYHA functional class, anemia, CKD, log₁₀-transformed BNP, log₁₀-transformed CRP, and LVEF were associated with HF-related events and all-cause mortality (Table 3). In this study, AF was found to be associated with HF-related events. Multivariate Cox proportional hazard regression analysis demonstrated that the serum GDF15 concentration was an independent predictor of HF-related events after adjusting for age, NYHA functional class, AF, anemia, CKD, log₁₀-transformed BNP, log₁₀-transformed CRP, and LVEF (hazard ratio [HR] 1.44; 95% confidence interval [CI] 1.24–1.67; P<0.0001; Table 3). Multivariate Cox proportional hazard regression analysis demonstrated that the serum GDF15 concentration was an independent predictor of all-cause mortality after adjusting for age, NYHA functional class, anemia, CKD, log₁₀-transformed BNP, log₁₀-transformed CRP, and LVEF (HR 1.42; 95% CI 1.16–1.72; P=0.0005; Table 3).

The prognostic usefulness of serum GDF15 in patients with HFpEF has not yet been fully elucidated. Thus, we performed subgroup analysis in patients with HFpEF. Univariate Cox proportional hazard regression analysis demonstrated that serum GDF15 concentrations were significantly associated with HF-related events and all-cause mortality. Furthermore, age, NYHA functional class, anemia, CKD, log₁₀-transformed BNP, and log₁₀-transformed CRP were also associated with HF-related events and all-cause mortality. In this study, AF was found to be associated with HF-related events. Multivariate Cox proportional hazard regression analysis demonstrated that the serum GDF15 concentration was an independent predictor of HF-related events after adjusting for age, NYHA functional class, AF, anemia, CKD, log₁₀-transformed BNP, and log₁₀-transformed CRP (HR 2.02; 95% CI 1.58–2.57; P<0.0001; Table 3). Multivariate Cox proportional hazard regression analysis demonstrated that the serum GDF15 concentration was an independent predictor of all-cause mortality after adjusting for age, NYHA functional class, anemia, CKD, log₁₀-transformed BNP, and log₁₀-transformed CRP (HR 2.01; 95% CI 1.50–2.69; P<0.0001; Table 3).

Improvement of Reclassification by Adding Serum GDF15 to Predict HF-Related Events and All-Cause Deaths

To examine whether model fit and discrimination improved when the serum GDF15 concentration was added to the baseline model, we evaluated improvements in the C index, NRI, and IDI. The baseline models included the selected predictors for HF-related events and all-cause deaths in the univariate Cox proportional hazard regression analysis. The addition of serum GDF15 to the baseline model significantly improved the C index, with significant NRI and IDI for HF-related events and all-cause deaths (Table 4). Subgroup analysis in patients with HFpEF also showed that the addition of GDF15 to the baseline model significantly improved the C index, with significant NRI and IDI for HF-related events and all-cause deaths (Table 4).

Discussion

The novel findings of the present study are as follows: (1) serum GDF15 concentrations increased with advancing NYHA functional class and age; (2) serum GDF15 concentrations were correlated with BNP, Hb, and eGFR in patients with HF; (3) Kaplan-Meier analysis demonstrated that the highest GDF15 tertile had the greatest risk for HF-related events and all-cause deaths; (4) multivariate Cox proportional hazard regression analysis demonstrated that serum GDF15 concentrations were independent predictors of HF-related events and all-cause deaths; and (5) serum GDF15 improved prognosis for HF-related events and all-cause deaths. Similar results were obtained in patients with HFpEF.

Serum GDF15 Concentrations in Patients With HF

GDF15 values reportedly differ according to age and the type of cardiovascular disease.17^,18 In accordance with previous results in healthy individuals, serum GDF15 also increased with age in patients with HF. The upper limit of GDF15 in healthy individuals was reported to be 1,200 pg/mL.19 In the present study, the median serum GDF15 concentration was 2,631 pg/mL (IQR 1,621–4,775 pg/mL), and 86% of HF patients had GDF15 values greater than 1,200 pg/mL. Kempf et al reported that the median GDF15 concentration was 1,949 pg/mL, and that patients with chronic HF and GDF15 ≥2,000 pg/mL had a high mortality rate.20 The Valsartan Heart Failure Trial (Val-HeFT) study reported a median GDF15 concentration of 2,040 pg/mL, and demonstrated that the combination of GDF15 ≥2,040 pg/mL and BNP ≥97 pg/mL stratifies symptomatic HF patients at high risk of all-cause death.21 The Singapore Heart Failure Outcomes and Phenotypes (SHOP) study found a median GDF15 concentration of 2,581 pg/mL, and that GDF15 tertiles could classify the different risks of patients with HF.22 The Relaxin for the Treatment of Acute Heart Failure study reported that the median GDF15 concentrations at baseline and 2 days were 4,013 and 3,608 pg/mL, respectively, in patients with acute HF.23 Considering these findings, the GDF15 values in Japanese patients with HF were almost equal to those in previous reports of studies performed in Western countries and Singapore.

The major sources of circulating GDF15 in HF are not completely understood. Experimental studies have demonstrated that mechanical stretch, ischemia, and neurohormones such as angiotensin II and inflammatory cytokines augment GDF15 expression in cultured cardiomyocytes.24^,25 Beyond production by cardiomyocytes, GDF15 upregulation has been shown to occur in endothelial cells, vascular smooth muscle cells, adipocytes, and macrophages in response to stress.26 An experimental study demonstrated that GDF15 concentrations depended on extracardiac tissues, despite an increase in the expression of GDF15 mRNA in cardiac tissue in HF model mice.27 Furthermore, a report focusing on patients with end-stage non-ischemic cardiomyopathy showed that GDF15 expression in the left ventricle was very low.28 In the present study, patients with dilated cardiomyopathy had lower serum GDF15 concentrations than those with ischemic heart disease and other conditions (Supplementary Figure). These findings suggest that circulating GDF15 concentrations represent the accumulated secretion of GDF15 from multiple sources in patients with HF. Thus, we explored the factors involved in GDF15 elevation and found that serum GDF15 concentrations were significantly correlated with BNP, Hb, and eGFR, suggesting that anemia and kidney dysfunction may be involved in the increase in serum GDF15 concentrations in patients with HF, as well as mechanical stretch and ischemia in the heart. Elevated circulating GDF15 concentrations are correlated with mRNA expression in the kidney and are associated with CKD progression in patients with CKD.29 Anemia involved in renal dysfunction and chronic inflammation is reportedly associated with high GDF15 concentrations, whereas iron deficiency anemia is not.30^,31 An inverse correlation between serum GDF15 concentrations and Hb has been reported in patients with HF and heart allograft recipients.32 Although HF is reportedly associated with elevated circulating GDF15 concentrations,22 it is plausible that several factors, such as aging, anemia, and kidney dysfunction, modify GDF15 concentrations in HF.

BNP and N-terminal pro B-type natriuretic peptide (NT-proBNP) concentrations were reported to be significantly lower in patients with HFpEF than in those with HFrEF, whereas there was no significant difference in GDF15 concentrations between patients with HFpEF and HFrEF.33^,34 Similarly, in the present study, the measured GDF15 increased to a similar degree in HFrEF and HFpEF patients in NYHA functional class III/IV. This feature makes serum GDF15 measurement attractive for risk prediction in patients with decompensated HF.

Serum GDF15 Concentrations and Clinical Outcomes

There is evidence indicating that GDF15 predicts mortality and HF development in a broad spectrum of cardiovascular diseases and the general population.18^,35^–37 Previous reports have demonstrated a significant association between high GDF15 concentrations and mortality and HF rehospitalization in patients with HF.38 In accordance with these reports, we clearly demonstrated that serum GDF15 could also identify and stratify Japanese patients with HF at high risk of HF-related events and all-cause mortality. Of note, similar results were obtained in subgroup analysis in patients with HFpEF. Chan et al reported that GDF15 is associated with clinical outcomes in patients with HFpEF similar to those with HFrEF using SHOP study data including 186 HFpEF patients.22 Izumiya et al reported the usefulness of GDF15 in predicting the composite of all-cause mortality, non-fatal myocardial infarction, stroke, and hospitalization for HF decompensation in patients with diastolic dysfunction including 73 HFpEF patients.9 Because we included 643 HFpEF patients in the present study, we clearly demonstrated the prognostic usefulness of serum GDF15 in patients with HFpEF.

In general, GDF15 has a cardioprotective role, such as antihypertrophic and antiapoptotic effects, through diverse mechanisms.39 Because the present study was a prospective observational study, we could not determine the causal relationship between serum GDF15 concentrations and clinical outcomes. Further studies are required to reveal the precise mechanism by which GDF15 leads to a deterioration in clinical outcomes in patients with HF. Importantly, we demonstrated that the C indices of GDF15 for HF-related events and all-cause deaths were the highest among BNP, Hb, and eGFR, suggesting the prognostic usefulness of GDF15. Furthermore, we showed that the prediction models for HF-related events and all-cause mortality improved significantly after adding serum GDF15 concentrations, as evidenced by an improvement in C indices, NRI, and IDI. This result confirmed the clinical importance of GDF15 in clinical outcomes and raised the possibility that GDF15 could be an additional clinical factor to consider in patients with HF.

Study Limitations

First, we measured GDF15 and BNP concentrations at a single center. Second, the prescription rate of HF medications was relatively low in the present study owing to the high prevalence of HFpEF and CKD. However, it was almost similar to the results from the Japanese Registry Of Acute Decompensated Heart Failure (JROADHF).40 Third, the optimal cut-off value for serum GDF15 concentrations remains unclear. Further studies with larger sample sizes are required to confirm the abnormal cut-off value of serum GDF15. Finally, because the study population was enrolled before angiotensin receptor-neprilysin inhibitor and sodium-glucose cotransporter 2 inhibitors were clinically administered for HF in Japan, we could not examine the impact of these drugs on serum GDF15 concentrations and clinical outcomes.

Conclusions

Our study found several factors, including aging, HF severity, volume overload, anemia, and kidney dysfunction, modified serum GDF15 concentrations in Japanese patients with HF. Serum GDF15 concentrations improved the predictive capacity for HF-related events and all-cause mortality. Therefore, the serum GDF15 concentration could be a feasible marker for adverse clinical outcomes in Japanese patients with HF.

Acknowledgments

The serum GDF15 assay was provided by Roche, which had no role in the study’s design, data analysis, manuscript preparation, or decision to submit the manuscript for publication.

Sources of Funding

This study did not receive any specific funding.

Disclosures

The authors declare that there are no conflicts of interest.

IRB Information

This study was approved by the Ethical Review Committee of Yamagata University Faculty of Medicine (No. 2021-298).

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-23-0088

References

• 1.Bui AL, Horwich TB, Fonarow GC. Epidemiology and risk profile of heart failure. Nat Rev Cardiol 2011; 8: 30–41.
• 2.Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Colvin MM, et al. 2017 ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Failure Society of America. Circulation 2017; 136: e137–e161.
• 3.Tsutsui H, Isobe M, Ito H, Ito H, Okumura K, Ono M, et al. JCS 2017/JHFS 2017 guideline on diagnosis and treatment of acute and chronic heart failure: Digest version. Circ J 2019; 83: 2084–2184.
• 4.Osada M, Park HL, Park MJ, Liu JW, Wu G, Trink B, et al. A p53-type response element in the GDF15 promoter confers high specificity for p53 activation. Biochem Biophys Res Commun 2007; 354: 913–918.
• 5.Baek SJ, Eling T. Growth differentiation factor 15 (GDF15): A survival protein with therapeutic potential in metabolic diseases. Pharmacol Ther 2019; 198: 46–58.
• 6.Bootcov MR, Bauskin AR, Valenzuela SM, Moore AG, Bansal M, He XY, et al. MIC-1, a novel macrophage inhibitory cytokine, is a divergent member of the TGF-beta superfamily. Proc Natl Acad Sci USA 1997; 94: 11514–11519.
• 7.Corre J, Hebraud B, Bourin P. Concise review: Growth differentiation factor 15 in pathology: A clinical role? Stem Cells Transl Med 2013; 2: 946–952.
• 8.Stahrenberg R, Edelmann F, Mende M, Kockskamper A, Dungen HD, Luers C, et al. The novel biomarker growth differentiation factor 15 in heart failure with normal ejection fraction. Eur J Heart Fail 2010; 12: 1309–1316.
• 9.Izumiya Y, Hanatani S, Kimura Y, Takashio S, Yamamoto E, Kusaka H, et al. Growth differentiation factor-15 is a useful prognostic marker in patients with heart failure with preserved ejection fraction. Can J Cardiol 2014; 30: 338–344.
• 10.Levy D, Savage DD, Garrison RJ, Anderson KM, Kannel WB, Castelli WP. Echocardiographic criteria for left ventricular hypertrophy: The Framingham Heart Study. Am J Cardiol 1987; 59: 956–960.
• 11.Otaki Y, Takahashi H, Watanabe T, Kadowaki S, Narumi T, Honda Y, et al. Electrocardiographic left ventricular hypertrophy Cornell product is a feasible predictor of cardiac prognosis in patients with chronic heart failure. Clin Res Cardiol 2014; 103: 275–284.
• 12.McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of congestive heart failure: The Framingham study. N Engl J Med 1971; 285: 1441–1446.
• 13.Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009; 53: 982–992.
• 14.Kidney Disease: Improving Global Outcomes (KDIGO) Diabetes Work Group. KDIGO 2020 clinical practice guideline for diabetes management in chronic kidney disease. Kidney Int 2020; 98: S1–S115.
• 15.Nutritional anaemias: Report of a WHO scientific group. World Health Organ Tech Rep Ser 1968; 405: 5–37.
• 16.Tsutsui H, Ide T, Ito H, Kihara Y, Kinugawa K, Kinugawa S, et al. JCS/JHFS 2021 guideline focused update on diagnosis and treatment of acute and chronic heart failure. Circ J 2021; 85: 2252–2291.
• 17.Wollert KC, Kempf T, Giannitsis E, Bertsch T, Braun SL, Maier H, et al. An automated assay for growth differentiation factor 15. J Appl Lab Med 2017; 1: 510–521.
• 18.Wollert KC, Kempf T, Wallentin L. Growth differentiation factor 15 as a biomarker in cardiovascular disease. Clin Chem 2017; 63: 140–151.
• 19.Kempf T, Horn-Wichmann R, Brabant G, Peter T, Allhoff T, Klein G, et al. Circulating concentrations of growth-differentiation factor 15 in apparently healthy elderly individuals and patients with chronic heart failure as assessed by a new immunoradiometric sandwich assay. Clin Chem 2007; 53: 284–291.
• 20.Kempf T, von Haehling S, Peter T, Allhoff T, Cicoira M, Doehner W, et al. Prognostic utility of growth differentiation factor-15 in patients with chronic heart failure. J Am Coll Cardiol 2007; 50: 1054–1060.
• 21.Anand IS, Kempf T, Rector TS, Tapken H, Allhoff T, Jantzen F, et al. Serial measurement of growth-differentiation factor-15 in heart failure: Relation to disease severity and prognosis in the Valsartan Heart Failure Trial. Circulation 2010; 122: 1387–1395.
• 22.Chan MM, Santhanakrishnan R, Chong JP, Chen Z, Tai BC, Liew OW, et al. Growth differentiation factor 15 in heart failure with preserved vs. reduced ejection fraction. Eur J Heart Fail 2016; 18: 81–88.
• 23.Cotter G, Voors AA, Prescott MF, Felker GM, Filippatos G, Greenberg BH, et al. Growth differentiation factor 15 (GDF-15) in patients admitted for acute heart failure: Results from the RELAX-AHF study. Eur J Heart Fail 2015; 17: 1133–1143.
• 24.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: 351–360.
• 25.Frank D, Kuhn C, Brors B, Hanselmann C, Ludde M, Katus HA, et al. Gene expression pattern in biomechanically stretched cardiomyocytes: Evidence for a stretch-specific gene program. Hypertension 2008; 51: 309–318.
• 26.Kempf T, Wollert KC. Growth-differentiation factor-15 in heart failure. Heart Fail Clin 2009; 5: 537–547.
• 27.Du W, Piek A, Schouten EM, van de Kolk CWA, Mueller C, Mebazaa A, et al. Plasma levels of heart failure biomarkers are primarily a reflection of extracardiac production. Theranostics 2018; 8: 4155–4169.
• 28.Lok SI, Winkens B, Goldschmeding R, van Geffen AJ, Nous FM, van Kuik J, et al. Circulating growth differentiation factor-15 correlates with myocardial fibrosis in patients with non-ischaemic dilated cardiomyopathy and decreases rapidly after left ventricular assist device support. Eur J Heart Fail 2012; 14: 1249–1256.
• 29.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: 2233–2240.
• 30.Waalen J, von Lohneysen K, Lee P, Xu X, Friedman JS. Erythropoietin, GDF15, IL6, hepcidin and testosterone levels in a large cohort of elderly individuals with anaemia of known and unknown cause. Eur J Haematol 2011; 87: 107–116.
• 31.Theurl I, Finkenstedt A, Schroll A, Nairz M, Sonnweber T, Bellmann-Weiler R, et al. Growth differentiation factor 15 in anaemia of chronic disease, iron deficiency anaemia and mixed type anaemia. Br J Haematol 2010; 148: 449–455.
• 32.Przybyłowski P, Wasilewski G, Bachorzewska-Gajewska H, Golabek K, Dobrzycki S, Małyszko J. Growth differentiation factor 15 is related to anemia and iron metabolism in heart allograft recipients and patients with chronic heart failure. Transplant Proc 2014; 46: 2852–2855.
• 33.Santhanakrishnan R, Chong JP, Ng TP, Ling LH, Sim D, Leong KT, et al. Growth differentiation factor 15, ST2, high-sensitivity troponin T, and N-terminal pro brain natriuretic peptide in heart failure with preserved vs. reduced ejection fraction. Eur J Heart Fail 2012; 14: 1338–1347.
• 34.Kitada S, Kikuchi S, Tsujino T, Masuyama T, Ohte N, J-MELODIC study investigators. The prognostic value of brain natriuretic peptide in patients with heart failure and left ventricular ejection fraction higher than 60%: A sub-analysis of the J-MELODIC study. ESC Heart Fail 2018; 5: 36–45.
• 35.Brown DA, Breit SN, Buring J, Fairlie WD, Bauskin AR, Liu T, et al. Concentration in plasma of macrophage inhibitory cytokine-1 and risk of cardiovascular events in women: A nested case-control study. Lancet 2002; 359: 2159–2163.
• 36.De Haan JJ, Haitjema S, den Ruijter HM, Pasterkamp G, de Borst GJ, Teraa M, et al. Growth differentiation factor 15 is associated with major amputation and mortality in patients with peripheral artery disease. J Am Heart Assoc 2017; 6: e006625.
• 37.Hagstrom E, Held C, Stewart RA, Aylward PE, Budaj A, Cannon CP, et al. Growth differentiation factor 15 predicts all-cause morbidity and mortality in stable coronary heart disease. Clin Chem 2017; 63: 325–333.
• 38.Wollert KC, Kempf T, Wallentin L. Growth differentiation factor 15 as a biomarker in cardiovascular disease. Clinical Chemistry 2017; 63: 140–151.
• 39.Adela R, Banerjee SK. GDF-15 as a target and biomarker for diabetes and cardiovascular diseases: A translational prospective. J Diabetes Res 2015; 2015: 490842.
• 40.Ide T, Kaku H, Matsushima S, Tohyama T, Enzan N, Funakoshi K, et al. Clinical characteristics and outcomes of hospitalized patients with heart failure from the large-scale Japanese Registry Of Acute Decompensated Heart Failure (JROADHF). Circ J 2021; 85: 1438–1450.