Correlation of Serum Adropin Levels with Risk Factors of Cardiovascular Disease in Hemodialysis Patients

Abstract

Background:

Many preclinical studies have shown that adropin has physiological effects such as regulating glucose, lipid, and energy metabolism, protecting endothelial cells and antiatherosclerosis. Our aim is to explore whether adropin is correlated with risk factors of cardiovascular disease (CVD) in hemodialysis (HD) patients.

Methods:

We recruited 170 HD patients and 120 healthy controls. The serum adropin concentration and clinical characteristics were measured.

Results:

The serum adropin concentration in HD patients was significantly lower than that in healthy controls and which in HD patients with CVD or diabetes mellitus (DM) was significantly lower than that in patients without CVD or DM. The correlation analysis showed that serum adropin levels were correlated negatively with Age, CVD history, DM history, C-reactive protein, type B natriuretic peptide, phosphorus, intact parathyroid hormone, carotid artery plaque amount and carotid intima-media thickness (CIMT), left ventricular septal thickness (LVSTd), and left ventricular posterior wall thickness, whereas it was correlated positively with albumin, hemoglobin, serum creatinine and Kt/V, and ejection fraction value. Partial correlation analysis verified that serum adropin levels were correlated negatively with CIMT, and multiple linear regression analysis revealed that low serum adropin levels may be one independent predictors of CIMT. However, the partial correlation analysis and multiple linear regression analysis did not identify the significant correlation between serum adropin levels and LVSTd.

Conclusions:

Our study revealed that serum adropin level is significantly correlated with risk factors of CVD and low serum adropin levels may be a potential predictor of CVD in HD patients.

Introduction

Cardiovascular disease (CVD) is highly prevalent among patients with end-stage renal disease (ESRD), especially those on hemodialysis (HD), which is the leading cause of mortality and morbidity in HD.¹ The Hemodialysis (HEMO) study reported that 40% of dialysis patients had CVD at entry and CVD is responsible for ∼63% of the cause of admission to hospitals in the United States.² The pathophysiological changes of CVD in HD patients are often manifested as impaired endothelial cell function, atherosclerosis, and vascular calcification, which lead to structural and functional changes of the heart, such as left ventricular septal thickening, left ventricular hypertrophy (LVH), valvular calcification (VC), and valvular regurgitation (VR), resulting in acute coronary syndrome (ACS), arrhythmia, and heart failure.³ Hence, discovering the new pathogenic pathways and/or protective factors for understanding the pathophysiology of CVD in such population and eventually finding new treatments are needed.

Adropin protein is a new identified metabolic hormone containing 76 amino acid encoded by ENHO gene, which is involved in regulating energy homeostasis, insulin resistance, glucose, and lipid metabolism.⁴ Adropin is widely expressed in human organs and tissues, such as brain, cerebellum, liver, kidney, heart, pancreas, small intestine, endocrine cells, and so on.^4,5 Adropin has been reported to be closely related to the occurrence and development of diseases such as diabetes mellitus (DM), CVD, metabolic syndrome (MS), central nervous system diseases, fatty liver, and chronic kidney disease (CKD) and may be used as an independent risk factor of disease prognosis.^4,6
–8

A large number of basic studies have proved that adropin has physiological effects such as regulating lipid metabolism, energy metabolism, protecting endothelial cells, and antiatherosclerosis.⁹ Adropin can induce nitric oxide (NO) release, reduce vascular endothelial permeability, inhibit mononuclear macrophage-endothelial cell adhesion, block vascular endothelial inflammatory factor expression and proliferation, and migration of vascular smooth muscle cells.^9,10 In animal models of cardiac ischemia-reperfusion, adropin can increase myocardial cell activity, reduce apoptosis and caspase-3 activity, and protect myocardial cells during reperfusion.¹¹ Adropin can also improve cardiomyocytes mitochondrial energy metabolism in vitro.¹² Therefore, clinical studies further identified the protective effects of adropin in CVD. It has been proved that serum adropin concentration is negatively correlated with the development and prognosis of coronary heart disease such as ACS (including acute myocardial infarction, unstable angina pectoris) and stable angina pectoris and is also closely related to the structural changes and cardiac functions of atrial fibrillation and heart failure,^13
–15 which indicated that it may be a potential serological marker of CVD.

For the population of ESRD especially on HD, CVD can be due to a myriad of factors originating from the disease pathology itself and from inadvertent effects of HD, there may exist the interaction of these co-morbid diseases. Several risk factors are involved in CVD development and prognosis among HD patients such as traditional risk factors, including hypertension, male, smoking, dyslipidemia, and insulin resistance and uremia-related risk factors, including insufficient HD, mineral bone disorders, malnutrition, and inflammation.³ Therefore, due to its potential cardioprotective effect, it is necessary to explore whether adropin is associated with these risk factors of CVD in HD patients. In the present study, we investigate the association between serum adropin levels and risk factors of CVD in patients on HD.

Materials and Methods

Subjects

The present study was a cross-sectional study, which recruited 170 patients on maintenance HD at People's Hospital of Pudong New District, Shanghai, China, and Shanghai East Hospital, Tongji University School of Medicine, Shanghai, China, between January 2018 and December 2018. Exclusion criteria covered pregnancy, malignancy, protein-losing enteropathies, diarrhea or signs of severe infection, missing more dialysis sessions than one per month, acute illness within 3 months before enrollment, ACS and/or cerebral stroke in 6 months before the study commencement, and treatment with statins or fibrates 6 weeks before commencement into the study. The clinical diagnoses of primary renal disease included chronic glomerulonephritis (n = 78), diabetic nephropathy (n = 55), hypertensive nephrosclerosis (n = 25), polycystic kidney disease (n = 7), and other diseases (n = 5). All HD patients had been dialyzed for at least 3 months before entering the study. The patient population included adults with an age range of 24–88 years. Patients were hemodialyzed using single-use hollow fiber dialyzers equipped with polysulfone membranes. The dialyzate used in all patients was a standard ionic composition and bicarbonate-based buffer. Dialysis efficiency was evaluated according to the Kidney Disease Outcomes Quality Initiative guidelines. Individuals who formerly or currently smoked >10 cigarettes per day for at least 2 years were defined as smokers. DM was defined as a fasting glucose level >7.0 mmol/L or use of any hypoglycemic medication. Hypertension was considered present if the patient received medical therapy for the condition or if blood pressure was >140/90 mmHg. For each HD subject, history of CVD, DM, and smoking were recorded. The definition of CVD included angina, class III or IV congestive heart failure, a history of myocardial infarction, cerebrovascular accident, or amputation due to peripheral vascular disease. None of the subjects received antibiotics, corticosteroids, anti-inflammatory drugs, cytotoxic drugs, or vitamins during the study period.

The study also included 120 unrelated control subjects randomly selected from outpatients who underwent regular physical examinations during the same time in the same hospital. Excluded from the control group were subjects with a history of significant concomitant diseases, including CVD, DM, hypertension, CKD, dyslipidemia and malignant diseases, angina, symptoms or signs of other atherosclerotic vascular diseases, or an abnormal electrocardiogram. All subjects enrolled in this study were of Han Chinese origin and resided in Shanghai City, China. The study complies with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of the People's Hospital of Pudong New District in Shanghai (No. PRYLW2018-01) and Shanghai East Hospital (No. 2018-002), and informed consent was obtained from all subjects participating in the study.

Clinical and biochemical parameters

Age, gender, history of DM, CVD, smokers, and ESRD causes were collected by reviewing medical records. Blood were sampled immediately before the midweek dialysis treatment using the slow flow/stop pump technique. Blood urea nitrogen (Bun), serum creatinine (Scr), serum total calcium (Ca), phosphorus (P), intact parathyroid hormone (iPTH), albumin (Alb), triglycerides (TG), total cholesterol (TCH), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), C-reactive protein (CRP), type B natriuretic peptide (BNP), and hemoglobin (Hb) amounts were measured by standard laboratory methods. All routine assays were performed at the central laboratory of the hospital. Then serum samples were stored at −70°C until further analysis. Serum adropin levels were measured twice before an HD session, using an enzyme-linked immunosorbent assay (ELISA), following which the results were averaged. The Human adropin (ENHO) Elisa Kit (Lifespan Biotech Ltd., Seattle, WA) was used. The detection limit range of the assay kit is 15.6–1000 pg/mL. According to the manufacturer, the intra-assay and interassay coefficients of variation are ≤7% and ≤10%, respectively. Blood samples from healthy controls in a fasting state were collected at physical examination center of People's Hospital of Pudong New District and Shanghai East Hospital, Tongji University School of Medicine. All measurements were performed by automatic biochemical analyzer (Roche Cobas c702, Basel, Switzerland). All ELISAs and nonroutine assays were performed by the same operator.

Determination of carotid artery plaque and carotid intima-media thickness

Carotid artery plaque (CAP) amount and carotid intima-media thickness (CIMT) were measured by carotid artery ultrasound. Carotid artery ultrasound scanning was performed using a high-resolution ultrasonography device (Toshiba High-resolution B-mode ultrasonography; ARTIDA, Tokyo, Japan) provided with a 7.5-MHz linear transducer. All measurements were done blindly by the same operator. Patients were examined in a supine position with the neck extended and the probe in the anterolateral position. All measurements of CIMT were made in the longitudinal plane at the point of maximum thickness on the far wall of the common carotid artery along a 1 cm section of the artery proximal to the carotid bulb in the diastolic phase. CIMT was defined as the distance between the inner echogenic line representing the intima–blood interface and the outer echogenic line representing the adventitia–media junction. Measurements were repeated three times, and CIMT was expressed as the mean of six measurements (three on each side).

Determination of ultrasound cardiogram

Four-dimensional and Doppler echocardiography (Philips, IE33, Amsterdam, Netherlands) examination were performed by an experienced sonographer at the dialysis interphase for HD patients. M-mode and 4D measurements were conducted by trained sonographers in accord with methods recommended by the American Society of Echocardiography. M-mode measurements included aortic root diameter, left atrial diameter (LAD), left ventricular end-diastolic diameter (LVIDd), left ventricular terminal diameter (LVIDs), left ventricular septal thickness (LVSTd), left ventricular posterior wall thickness (LVPWTd), VC, and VR. Ejection fraction (EF) and LAD were determined from apical two- and four-chamber views by the Simpson's biplane formulae.

Statistical analysis

All statistical analyses were performed with SPSS 26.0 for Windows. P values <0.05 were considered statistically significant. Normally distributed variables were expressed as mean ± standard deviation and nonnormally distributed variables as median with 25th and 75th percentiles. Differences between groups were compared by Student's t-test or Mann–Whitney U test. Categorical data were presented as percentage, and assessed by the chi-square test. The Spearman or Pearson's method was used to estimate the correlation between adropin and other parameters. A partial correlation analysis was carried out to determine the correlation of adropin with CIMT and LVSTd by the control of other associated parameters. A stepwise multiple linear regression analysis was applied to examine the relationship between CIMT or LVSTd and a set of clinical parameters, including age, gender, BMI, DM history, time on HD treatment, smoking, TG, TC, LDL-C, HDL, BNP, Hb, albumin, CRP, calcium, phosphate, iPTH, Kt/V_urea, and adropin. The significant independent variables were ordered according to their standardized effect, defined as regression coefficient/standard error of the regression (β).

Results

Clinical characteristics of participants

The clinical characteristics and biochemical data of control subjects and HD patients are summarized in Table 1. The two groups were statistically different by definition in terms of requirement for renal replacement therapy. In addition, there were significant differences between HD patients and controls in Age, SBP, DBP, HbA1C, Scr, and BUN, and HD patients had lower level of Alb, Hb, TC, LDL-C, and HDL-C, which reflected the real disease status of HD patients compared to healthy controls.

Table 1.

Baseline Characteristics of the Studied Subjects

Characteristics	HD subjects n = 170	Controls n = 120	P	HD subjects with CVD n = 80	HD subjects without CVD n = 90	P	HD subjects with DM n = 55	HD subjects without DM n = 115	P
Age (years)	63.2 ± 12.1	40.1 ± 5.7	<0.001^*	66.7 ± 9.6	60.2 ± 13.2	<0.001^#	63.2 ± 10.2	63.2 ± 12.9	0.997
Smoker (n)	56	—	—	28	29	0.746	26	28	0.003^$
DM history (n)	55	—	—	45	10	<0.001^#	55	0	<0.001^$
CVD history (n)	80	—	—	80	0	<0.001^#	45	35	<0.001^$
HD months	48.5 ± 51.9	—	—	45.2 ± 52.5	51.5 ± 51.6	0.427	25.6 ± 24.8	59.5 ± 57.7	0.001^$
BMI	22.7 ± 4.1	22.4 ± 2.7	0.520	23.1 ± 4.8	22.4 ± 3.3	0.254	23.3 ± 3.2	22.4 ± 4.4	0.153
SBP (mmHg)	142 ± 18	117 ± 7	<0.001^*	143 ± 20	141 ± 15	0.353	143 ± 20	141 ± 15	0.372
DBP (mmHg)	81 ± 13	72 ± 7	<0.001^*	82 ± 15	80 ± 11	0.419	82 ± 15	80 ± 11	0.425
HbA1c (%)	6.3 ± 1.6	4.5 ± 0.5	<0.001^*	6.7 ± 2.0	5.8 ± 0.9	<0.001^#	7.5 ± 2.0	5.5 ± 0.8	<0.001^$
TG (mmol/L)	2.0 ± 1.1	1.4 ± 4.0	0.091	2.2 ± 1.2	1.7 ± 0.9	0.002^#	2.5 ± 1.2	1.7 ± 0.9	<0.001^$
TC (mmol/L)	3.6 ± 1	4.7 ± 7.0	0.044^*	3.8 ± 1.0	3.5 ± 1.0	0.035^#	3.9 ± 1.0	3.5 ± 1.0	0.045^$
LDL-C (mmol/L)	2.3 ± 0.9	2.7 ± 0.7	<0.001^*	2.5 ± 1.0	2.1 ± 0.7	0.018^#	2.6 ± 1.0	2.1 ± 0.8	0.001^$
HDL-C (mmol/L)	1.1 ± 0.3	1.3 ± 0.3	<0.001^*	1.08 ± 0.4	1.03 ± 0.3	0.453	0.9 ± 1.1	1.1 ± 0.3	0.001^$
Alb (g/L)	40.2 ± 3.6	44.7 ± 3.1	<0.001^*	39.8 ± 3.7	41.7 ± 3.4	0.001^#	40.3 ± 3.3	41.0 ± 3.8	0.265
Hb (g/L)	108.0 ± 14.8	135.5 ± 16.1	<0.001^*	107.0 ± 12.7	112.2 ± 11.0	0.004^#	106.6 ± 17.4	110.2 ± 13.2	0.410
CRP (mg/L)	6.5 ± 9.7	—	—	9.4 ± 12.9	3.9 ± 4.0	<0.001^#	12.6 ± 14.5	3.6 ± 3.7	<0.001^$
BNP (ng/L)	7542.2 ± 983.3	—	—	10132.3 ± 10710.7	5284.3 ± 8278.9	0.001^#	8945.2 ± 11073.0	5688.6 ± 6955.9	0.021^$
Ca (mmol/L)	2.32 ± 0.21	—	—	2.3 ± 0.2	2.3 ± 0.2	0.142	2.2 ± 0.2	2.4 ± 0.2	<0.001^$
P (mmol/L)	1.83 ± 0.51	—	—	1.9 ± 0.519	1.7 ± 0.5	0.016^#	1.9 ± 0.5	1.7 ± 0.4	0.002^$
iPTH (pg/mL)	354.5 ± 316.1	—	—	334.2 ± 207.0	278.0 ± 166.8	0.048^#	310.8 ± 156.5	301.5 ± 202.0	0.772
Scr (μmol/L)	913.4 ± 244.3	68.0 ± 14.2	<0.001^*	978.4 ± 262.9	888.8 ± 245.6	0.023^#	957.9 ± 260.2	891.1 ± 250.0	0.397
Bun (mmol/L)	25.3 ± 7.0	5.1 ± 1.1	<0.001^*	25.7 ± 5.7	25.4 ± 7.5	0.779	26.2 ± 6.5	25.3 ± 6.6	0.115
Kt/V	1.45 ± 0.26	—	—	1.4 ± 0.2	1.5 ± 0.3	0.027^#	1.3 ± 0.2	1.5 ± 0.3	<0.001^$
CAP amount	1.36 ± 1.73	—	—	2.1 ± 1.8	0.7 ± 1.3	<0.001^#	2.1 ± 1.8	1.0 ± 1.5	<0.001^$
CIMT	0.91 ± 0.24	—	—	1.1 ± 0.2	0.8 ± 0.2	<0.001^#	1.1 ± 0.3	0.8 ± 0.2	<0.001^$
Adropin (pmol/L)	269.2 ± 153.0	580.1 ± 84.8	<0.001^*	190.9 ± 107.0	338.9 ± 154.5	<0.001^#	233.1 ± 132.5	286.5 ± 159.6	0.030^$
UCG
ARD (mm)	28.9 ± 3.1	—	—	29.0 ± 3.4	28.8 ± 2.7	0.804	29.0 ± 3.2	28.8 ± 3.0	0.702
LAD (mm)	41.0 ± 5.6	—	—	42.2 ± 5.1	40.0 ± 5.8	0.008^#	42.8 ± 4.7	40.2 ± 5.8	0.005^$
LVIDd (mm)	49.7 ± 6.3	—	—	50.3 ± 5.8	49.2 ± 6.8	0.032^#	50.5 ± 5.5	49.3 ± 6.6	0.202
LVIDs (mm)	33.0 ± 5.5	—	—	33.5 ± 5.5	32.5 ± 5.5	0.062	33.9 ± 5.1	32.6 ± 5.7	0.136
LVSTd (mm)	10.2 ± 1.7	—	—	11.0 ± 1.7	9.7 ± 1.6	<0.001^#	11.2 ± 1.8	9.9 ± 1.7	<0.001^$
LVPWTd (mm)	10.0 ± 1.6	—	—	10.5 ± 1.6	9.6 ± 1.4	<0.001^#	10.8 ± 1.7	9.6 ± 1.4	<0.001^$
EF ()	63.0 ± 7.8	—	—	59.6 ± 9.1	66.0 ± 4.6	0.010^#	59.3 ± 9.9	64.8 ± 5.7	<0.001^$
VC (n)	86	—	—	60	26	<0.001^#	37	48	0.003^$
VR (n)	144	—	—	75	69	<0.001^#	51	93	0.033^$

^* HD subjects versus control.

HD subjects with CVD versus HD subjects without CVD, P < 0.05.

HD subjects with DM versus HD subjects without DM, P < 0.05.

Alb, albumin; ARD, aortic root diameter; BMI, body mass index; BNP, type B natriuretic peptide; BUN, blood urea nitrogen; Ca, calcium; CAP, carotid artery plaque; CIMT, carotid intima-media thickness; CRP, C-reactive protein; CVD, cardiovascular disease; DBP, diastolic blood pressure; DM, diabetes mellitus; Hb, hemoglobin; HbA1c, hemoglobin A1C; HD, hemodialysis, HDL-C, high-density lipoprotein cholesterol; iPTH, intact parathyroid hormone; LAD, left atrial diameter; LDL-C, low-density lipoprotein cholesterol; LVIDd, left ventricular end-diastolic diameter; LVIDs, left ventricular terminal diameter; LVSTd, left ventricular septal thickness; LVPWTd, left ventricular posterior wall thickness; P, phosphorus, SBP, systolic blood pressure; Scr, serum creatinine; TC, total cholesterol; TG, triglycerides; UCG, ultrasound cardiogram; EF, ejection fraction; VC, valvular calcification, VR, valvular regurgitation.

When HD patients were divided into two groups according to CVD history and DM history, we found that the presence of DM history, and Age, HbA1c, TG, TC, LDL-c, P, iPTH, Scr, CRP, BNP, CAP, CIMT, LAD, LVIDd, LVSTd, LVPWTd, the presence of VC and VR in HD patients with CVD history were higher than patients without CVD (P < 0.05, perspectively). In contrast, HD patients with CVD had lower levels of Kt/V, Alb, Hb, and EF value (P < 0.05, respectively). Furthermore, we also found that the presence of male, smokers and CVD history, and HbA1c, TG, TC, LDL-c, P, CRP, BNP, CAP, CIMT, LAD, LVSTd, LVPWTd, the presence of VC and VR in HD patients with DM history were higher than patients without DM (P < 0.05, perspectively), while HD patients with DM had lower levels of HDL-c, Ca, Kt/V, and EF value (P < 0.05, perspectively).

Serum adropin concentration in participants

When analyzed by serum adropin concentration (Table 1), we found that the serum adropin concentration in HD patients was significantly lower than that in healthy controls (P < 0.001). Meanwhile, the serum adropin concentration in HD patients with CVD or DM history was significantly lower than that in patients without CVD or DM (P < 0.05, perspectively).

Correlations of serum adropin levels with characteristics of HD patients

In the present study, the correlations of serum adropin concentration with characteristics of HD patients were analyzed (Table 2). There were significant negative correlations between serum adropin concentrations and Age, CVD and DM history, CRP, BNP, P, iPTH, CAP, CIMT, LVSTd, and LVPWTd (P < 0.05, perspectively). Meanwhile, serum adropin concentrations were positively correlated with Alb, Hb, Scr, Kt/V, and EF value (P < 0.05, perspectively).

Table 2.

Correlation of Serum Adropin Levels with Clinical Characteristics and Ultrasound Cardiogram Data of Hemodialysis Patients

Characteristics	Correlation coefficient (r)	P
Age (years)	−0.297	0.001^*
Male/female	−0.026	0.739
CVD history	−0.125	0.048^*
DM history	−0.217	0.001^*
BMI	−0.137	0.075
HbA1c (%)	−0.098	0.206
TG (mmol/L)	−0.027	0.729
TC (mmol/L)	−0.079	0.304
LDL (mmol/L)	−0.106	0.169
HDL (mmol/L)	−0.043	0.578
Alb (g/L)	0.293	<0.001^*
Hb (g/L)	0.170	0.026^*
CRP (mg/L)	−0.203	0.008^*
BNP (ng/L)	−0.204	0.008^*
Ca (mmol/L)	−0.013	0.864
P (mmol/L)	−0.181	0.018^*
iPTH (pg/mL)	−0.172	0.025^*
Scr (μmol/L)	0.205	0.007^*
Bun (mmol/L)	0.006	0.938
Kt/V	0.206	0.007^*
CAP amount (n)	−0.257	0.001^*
CIMT (mm)	−0.446	<0.001^*
UCG
ARD (mm)	−0.097	0.207
LAD (mm)	−0.070	0.364
LVIDd (mm)	−0.111	0.15
LVIDs (mm)	−0.136	0.077
LVSTd (mm)	−0.278	<0.001^*
LVPWTd (mm)	−0.198	0.01^*
EF (%)	0.253	0.001^*
VC (n)	−0.085	0.179
VR (n)	−0.026	0.74

^* P < 0.05.

Partial correlation analysis of serum adropin and CIMT in HD patients

Arterial atherosclerosis and vascular calcification are the precursors of CVD.¹ Our previous study supports the consensus that insufficient HD, abnormal calcium-phosphate and lipid metabolism, malnutrition status, DM, and inflammation are related to development and progression of atherosclerosis, which are the risk factors of CVD in HD paitents.¹⁶ When these indices (shown in Table 3) were used as controlled variables together, we verified the association between serum adropin and CIMT by partial correlation analysis, which verified the negative correlation between serum adropin levels and CIMT.

Table 3.

Partial Correlation Analysis of Serum Adropin and Carotid Intima-Media Thickness in Hemodialysis Patients

Controlled variables	Coefficient (r) between Adropin and CIMT	P
Age, HD months, BMI, TG, TC, LDL, HDL, HbA1c, Ca, p.iPTH, Kt/v, Alb, Hb, CRP, BNP, CAP	−0.225	0.005^*

^*P < 0.05.

Multiple linear regression analysis for defining the independent determinants of CIMT in HD patients

Furthermore, multiple stepwise linear regression analysis was used to define independent determinants of CIMT in HD patients. The linear regression model which incorporated adropin and other index (shown in Table 4) revealed that low serum adropin concentration, high levels of Age, CRP, and BNP, and presence of DM were independent predictors of CIMT (P < 0.05, perspectively).

Table 4.

Multiple Linear Regression Analysis for Defining Serum Adropin as the Independent Determinants of Carotid Intima-Media Thickness in Hemodialysis Patients

Independent variables	β (coefficient)	Standard coefficient (β)	t	P
Constant variable	0.419		1.296	0.044^*
Adropin	−0.000	−0.178	−2.839	0.005^*
Age	0.004	0.189	3.012	0.003^*
HD time	−0.003	−0.060	−0.783	0.435
BMI	0.002	0.034	0.558	0.578
TG (mmol/L)	0.021	0.090	1.206	0.230
TC (mmol/L)	−0.028	−0.112	−0.943	0.347
LDL (mmol/L)	0.050	0.169	1.527	0.129
HDL (mmol/L)	0.087	0.117	1.615	0.108
HbA1c (%)	−0.003	−0.020	−0.283	0.778
Ca (mmol/L)	0.020	0.017	0.272	0.786
P (mmol/L)	−0.019	−0.033	−0.554	0.581
iPTH (pg/mL)	−6.568E-05	−0.049	−0.746	0.457
Kt/V	−0.111	−0.116	−1.914	0.057
Alb (g/L)	0.000	0.006	0.093	0.926
Hb (g/L)	−0.001	−0.029	−0.487	0.627
CRP (mg/L)	0.003	0.133	2.147	0.033^*
BNP (ng/L)	4.226E-06	0.143	2.230	0.027^*
CAP amount (n)	0.060	0.392	6.136	<0.001^*

^*P < 0.05.

Partial correlation and multiple linear regression analysis of serum adropin and LVSTd in HD patients

Left ventricular septal thickening is an important structural pathological changes of CVD in HD patients, which was influenced by length of HD duration, hypertension, insufficient HD, abnormal calcium-phosphate and lipid metabolism, malnutrition status, DM, and inflammation.³ Our previous data had manifested that serum adropin has significant negative correlation with LVSTd, therefore, we use partial correlation analysis to verify this association when we make these indices (shown in Table 5) as controlled variables together. Our results manifested that there was no significant correlation between serum adropin levels and LVSTd. Finally, we use multiple linear regression analysis, which incorporated adropin and other index (shown in Table 6) to define independent determinants of LVSTd in HD patients, and the results revealed that all variables, including serum adropin, were not independent predictors of LVSTd in our selected HD patients (P > 0.05, perspectively).

Table 5.

Partial Correlation Analysis of Serum Adropin and Left Ventricular Septal Thickness in Hemodialysis Patients

Controlled variables	Coefficient (r) between adropin and LVSTd	P
Age, HD history, BMI, TG, TC, LDL, HDL, HbA1c, Ca, p.iPTH, Kt/v, Alb, Hb, CRP, BNP, CAP, CIMT	−0.075	0.358

Table 6.

Multiple Linear Regression Analysis for Serum Adropin and LVIDd in Hemodialysis Patients

Independent variables	β (coefficient)	Standard coefficient (β)	t	P
Constant variable	10.096		3.574	0.000^*
Adropin	−0.001	−0.073	−0.922	0.358
Age	0.015	0.099	1.246	0.215
HD time	0.020	0.048	0.510	0.611
BMI	0.055	0.126	1.683	0.094
TG (mmol/L)	0.142	0.086	0.934	0.352
TC (mmol/L)	−0.128	−0.072	−0.489	0.625
LDL (mmol/L)	0.310	0.150	1.088	0.279
HDL (mmol/L)	0.217	0.041	0.458	0.648
HbA1c (%)	−0.058	−0.052	−0.595	0.553
Ca (mmol/L)	0.715	0.715	0.715	0.715
P (mmol/L)	0.715	0.715	0.715	0.715
iPTH (pg/mL)	0.715	0.715	0.715	0.715
Kt/V	0.715	0.715	0.715	0.715
Alb (g/L)	0.030	0.060	0.719	0.473
Hb (g/L)	−0.006	−0.039	−0.530	0.597
CRP (mg/L)	0.010	0.054	0.698	0.486
BNP (ng/L)	6.047E-05	0.291	3.614	0.000^*
CAP amount (n)	0.186	0.172	1.955	0.052
CIMT (mm)	0.033	0.005	0.046	0.963

^*P < 0.05.

Discussion

Adropin is a new identified secreted protein, which acts as a regulatory polypeptide involved in energy homeostasis, insulin resistance, glucose, and lipid metabolism.⁴ Adropin is widely expressed in human organs and tissues, and in most of the published study populations, serum adropin concentrations in most diseases were significantly lower than that in healthy populations, which may be closely related to the occurrence and development of many diseases.⁴ Adropin concentrations take no change during HD sessions due to its molecular weight of 4,999.9D.¹⁷ Up to date, there is few published studies on the correlation between adropin and clinical manifestations of HD patients in small populations with controversial conclusions. Małgorzata found that there is no significant difference in serum adropin concentrations between ESRD patients and healthy controls.¹⁸ However, our present study found that serum adropin concentration in HD patients was significantly lower than that in healthy controls, which is consistent with the study from Alicja.¹⁹ It is the first time that the serum adropin concentration of Chinese HD patients were detected in a single HD center in China, which is inevitably biased. Therefore, the real range of serum adropin levels in HD patients needs enlarging sample sizes and enrolling different ethnic populations for further validation.

CVD is the leading cause of mortality and morbidity in patients with ESRD, which is partly explained by the fact that 40%–70% of patients receiving dialysis have significant coronary artery disease (CAD).¹ Of our HD patients, 47.1% had CVD history, which was higher than in general population. We further found that the serum adropin concentration in HD patients with CVD were significantly lower than that in patients without CVD. This difference is close to most of other conclusions in general populations with CVD or not. A recent meta-analysis of seven clinical studies involving 945 patients with CAD found that serum adropin levels were significantly lower in patients with ACS than in healthy controls.¹³ This phenomenon also existed in patients with atrial fibrillation¹⁴ and heart failure patients with low EF compared with healthy controls.⁹ Therefore, it suggests that serum adropin may be an important potential biomarker for the diagnosis and prediction of CVD in HD patients.

In addition to the traditional risk factors, the high incidence and poor prognosis of CVD in HD patients has unique nontraditional risk factors, including insufficient HD, hyperphosphatemia, secondary hyperparathyroidism, anemia, malnutrition, and inflammation.³ In the present study, when comparing the clinical data of HD patients with CVD or not, we verified that HD patients with CVD hold more above-mentioned risk factors than those without CVD. With the increase of age, the decrease of serum adropin concentration may be related to the poor diet and nutritional status of the elderly. As we all know, all these nontraditional risk factors may lead to atherosclerosis, arteries and valves calcification, myocardial remodeling, and hypertrophy eventually promote the development and mortality of CVD in HD patients.³ Our results from the correlation analysis further showed that lower serum adropin concentration was companied with higher levels of CRP, P, and iPTH and with lower levels of Alb, Hb, and Kt/V. It indicated that similar to these nontraditional risk factors, lower serum adropin levels may be the potential risk factor of CVD in HD patients. Further study is necessary to reveal the interaction between adropin and these risk factors, and the regulatory mechanism influencing the development of CVD.

The physiological function of adropin is regulating lipid, glucose, and energy metabolism, and it has proved that adropin is closely related to DM, hyperlipidemia, and MS.⁴ In our selected HD patients, serum adropin concentrations in patients with DM were significantly lower than that without DM. This result is similar to other previous studies in general DM population, which had suggested that adropin is negatively correlated with glucose and HbA1C, along with the different conclusions about the relationship between adropin and BMI or lipid.^20
–22 However, different results still were reported in other populations.^23,24 In our study, HD patients with CVD had poorer glucose control and lipid metabolism disorders than patients without CVD. However, we found adropin had no significant correlation with BMI, HbA1C, and lipid in our HD patients. This conclusion is similar to other studies involved HD patients. Alicja found that insulin resistance and dyslipidemia are frequently observed in HD patients, however, plasma adropin concentrations were not related to these abnormalities, along with the negative relationship with BMI.¹⁹ Małgorzata also found that the same results in ESRD patients.¹⁸ Malnutrition status and diversity of diet in most HD patients may interfere with adropin and lipid levels. In general CVD patients, adropin was proved to have no significant correlation with lipid levels, although it is closely related to the clinical characteristics and mortality of CVD.^9,13 Therefore, we speculate that the correlation between low adropin levels and CVD status in HD patients may not be due to the decline of its regulating glucose and lipid metabolism, but of its direct effect on vascular endothelium and cardiomyocytes.

Arteriosclerosis is predominant in ESRD patients, characterized by the intima of the arteries thicken, calcification, and plaque formation, which are important pathological change of CVD.³ In this study, we found that CAP and CIMT in HD patients with CVD were higher than patients without CVD, and serum adropin concentrations were negatively related to CIMT when risk factors were controlled, which contributes to thickening of blood vessels and plaque formation.³ This may be due to the physiological effect of adropin to protect blood vessels and inhibit plaque formation. It has reported that, in the apoE^−/− mice, adropin can induce endothelial nitric oxide synthase (eNOS) expression level, increase the release of NO, inhibit the infiltration of mononuclear macrophages, endothelial cell adhesion, vascular endothelial inflammatory factors and VSMC proliferation, migration, and reduce atherosclerotic plaque size.^9,10 Coskun found that serum adropin concentration is positively related with blood NO levels in patients with MS.²⁵ Therefore, we speculated that low serum adropin level may be the independent risk factor of atherosclerosis in HD patients due to its protective effect on vascular endothelium.

The other reason for the high incidence and mortality of CVD in HD patients is the abnormal cardiac structure and function, and LVH is present in 74% of ESRD patients in one cohort study.³ In this study, we found that HD patients with CVD had LVH with lower EF value and higher BNP levels, and serum adropin levels were negatively correlated with BNP, LVSTd, and LVPWTd, along with positive correlation with EF value. However, the results from partial correlation and multiple linear regression analysis did not verify the correlation between adropin and LVSTd. Several studies have manifested that adropin is negatively correlated with BNP and LAD and positively correlated with EF value in patients with atrial fibrillation and chronic heart failure.^9,14 In ESRD patients, Małgorzata also found that adropin was positively correlated with BNP and LVIDs, and negatively correlated with relative wall thickness (RWT).¹⁷ It suggested the relationship between adropin with cardiac structural and function is unclear. However, the results from basic studies have proved that adropin can increase myocardial cell activity, reduce apoptosis, improve cardiomyocytes mitochondrial energy metabolism, and protect myocardial cells during reperfusion.^11,12 Therefore, we speculate that adropin may have cardioprotective effect by direct manner, which needs to be further verified by expanding the sample size and longer duration of dynamic echocardiography observation.

The major limitation of this study lies in the small amount of samples, and it might overestimate the magnitude of correlations. Furthermore, our study is only one cross-sectional study and absent of dynamic observation and analysis. Expansion of the sample size and ethnicity, further dynamic study on the correlation between adropin and CVD mortality will provide more evidence to identify its role and relationship with CVD in HD patients.

Conclusions

In summary, our study manifested that serum adropin levels in HD patients were significantly lower than that in healthy controls, and it in HD patients with CVD was significantly lower than that in patients without CVD. Further analysis revealed that serum adropin levels was significantly correlated with several risk factors of CVD, cardiac structure and function, which indicate that low serum adropin may be a potential predictor of cardiac status in patients on HD.

Footnotes

Authors' Contributions

F.L., B.C., and H.Q. designed experiments. X.Z., Y.W., H.Q., Y.G., H.W., M.L., S.Z., J.S., X.S., W.L., S.M., and A.Z. conducted the experiments. F.L. and B.C. analyzed data. F.L., B.C., and H.Q. prepared the article.

Acknowledgments

The sponsor and investigators thank the patients and their families for their participation and support for this clinical study. We thank the medical staff at the Department of Nephrology, Shanghai East Hospital, Tongji University School of Medicine, and Department of Nephrology, the People's Hospital of Pudong New District in Shanghai. We acknowledge the laboratory support provided by Department of Laboratory and Ultrasound, Shanghai East Hospital, and the People's Hospital of Pudong New District in Shanghai.

Author Disclosure Statement

No conflicting financial interests exist.

Funding Information

This study was supported by the grants from the Academic Leader Training Plan of Health System of Pudong New District of Shanghai, China (PWRd2019-13 to F.L.), Science, Technology and Economic Committee of Pudong New District of Shanghai, China (PKJ2017-Y19 to F.L.), and Leading Talent Training Plan of Health System of Pudong New District of Shanghai, China (PWRl2019-08 to H.Q.).

References

Aoki

, Ikari

. Cardiovascular disease in patients with end-stage renal disease on hemodialysis. Ann Vasc Dis, 2017; 10:327–337.

Zoccali

, Mallamaci

, Tripepi

. Traditional and emerging cardiovascular risk factors in end-stage renal disease. Kidney Int Suppl, 2003; 63:S105–S110.

Ahmadmehrabi

, Tang

WHW

. Hemodialysis-induced cardiovascular disease. Semin Dial, 2018; 31:258–267.

Marczuk

, Cecerska-Heryc

, Jesionowska

, et al. Adropin—Physiological and pathophysiological role. Postepy Hig Med Dosw, 2016; 70:981–988.

Aydin

, Kuloglu

, Aydin

, et al. Expression of adropin in rat brain, cerebellum, kidneys, heart, liver, and pancreas in streptozotocin-induced diabetes. Mol Cell Biochem, 2013; 380:73–81.

Kutlu

, Altun

, Dikker

, et al. Serum adropin levels are reduced in adult patients with nonalcoholic fatty liver disease. Med Princ Pract, 2019; 28:463–469.

Shahjouei

, Ansari

, Pourmotabbed

, et al. Potential roles of adropin in central nervous system: Review of current literature. Front Mol Biosci, 2016; 3:25.

, Chen

. Association of serum adropin concentrations with diabetic nephropathy. Mediators Inflamm, 2016; 2016:6038261.

, Xie

, Zheng

, et al. A novel peptide adropin in cardiovascular diseases. Clin Chim Acta, 2016; 453:107–113.

10.

Sato

, Yamashita

, Shirai

, et al. Adropin contributes to anti-atherosclerosis by suppressing monocyte-endothelial cell adhesion and smooth muscle cell proliferation. Int J Mol Sci, 2018; 19:1293.

11.

, Fang

, Yuan

, et al. Adropin reduces hypoxia/reoxygenation-induced myocardial injury via the reperfusion injury salvage kinase pathway. Exp Ther Med, 2019; 18:3307–3314.

12.

Thapa

, Stoner

, Zhang

, et al. Adropin regulates pyruvate dehydrogenase in cardiac cells via a novel GPCR-MAPK-PDK4 signaling pathway. Redox Biol, 2018; 18:25–32.

13.

Zheng

, Liu

, Chen

, et al. Association between serum adropin level and coronary artery disease: A systematic review and meta-analysis. Cardiovasc Diagn Ther, 2019; 9:1–7.

14.

Wang

, Xue

, Shang

, et al. Association of serum adropin with the presence of atrial fibrillation and atrial remodeling. J Clin Lab Anal, 2019; 33:e22672.

15.

Yosaee

, Soltani

, Sekhavati

, et al. Adropin-a novel biomarker of heart disease: A systematic review article. Iran J Public Health, 2016; 45:1568–1576.

16.

Liu

, Zhang

, Qi

, et al. Correlation of serum magnesium with cardiovascular risk factors in maintenance hemodialysis patients—A cross-sectional study. Magnes Res, 2013; 26:100–108.

17.

Kaluzna

, Pawlaczyk

, Schwermer

, et al. Adropin and irisin: New biomarkers of cardiac status in patients with end-stage renal disease? A preliminary study. Adv Clin Exp Med, 2019; 28:347–353.

18.

Kaluzna

, Hoppe

, Schwermer

, et al. Adropin and irisin levels in relation to nutrition, body composition, and insulin resistance in patients with end-stage renal disease on chronic hemodialysis and peritoneal dialysis. Pol Arch Med Wewn, 2016; 126:474–482.

19.

Grzegorzewska

, Niepolski

, Mostowska

, et al. Involvement of adropin and adropin-associated genes in metabolic abnormalities of hemodialysis patients. Life Sci, 2016; 160:41–46.

20.

Butler

, Tam

, Stanhope

, et al. Low circulating adropin concentrations with obesity and aging correlate with risk factors for metabolic disease and increase after gastric bypass surgery in humans. J Clin Endocrinol Metabol, 2012; 97:3783–3791.

21.

Zang

, Jiang

, Cheng

, et al. Serum adropin levels are decreased in Chinese type 2 diabetic patients and negatively correlated with body mass index. Endocrine J, 2018; 65:685–691.

22.

Choi

, Yim

. Plasma adropin as a potential marker predicting obesity and obesity-associated cancer in Korean patients with type 2 diabetes mellitus. J Cancer Prev, 2018; 23:191–196.

23.

Palizban

, Yazdani

, Jahanbani-Ardakani

. Role of rs7903146 polymorphism and adropin serum level in patients with diabetes mellitus; a case-control study from Isfahan, Iran. Arch Physiol Biochem, 2019; 125:1–4.

24.

Dabrowski

, Jarmuzek

, Gondek

, et al. First and third trimester serum concentrations of adropin and copeptin in gestational diabetes mellitus and normal pregnancy. Ginekol Polska, 2016; 87:629–634.

25.

Oruc

, Akpinar

, Dervisoglu

, et al. Low concentrations of adropin are associated with endothelial dysfunction as assessed by flow-mediated dilatation in patients with metabolic syndrome. Clin Chem Lab Med, 2017; 55:139–144.