Predictive markers of asymptomatic atherosclerosis in end-stage renal disease patients

Abstract

Objective: Uremia is associated with accelerated atherosclerosis and increased cardiovascular mortality in patients with end-stage renal disease (ESRD). Cardiac injury markers, such as myoglobin, creatine kinase-MB (CK-MB), or troponins, frequently used to recognize acute coronary events, may be falsely elevated in this patient group. In this study, our aim was to (i) test serum levels of myoglobin, CK-MB, and troponin I (cTnI) in ESRD patients without coronary artery disease (CAD) and compare the results with healthy controls and (ii) to investigate the association between these markers and carotid artery intima–media thickness (CA–IMT), high-sensitive C-reactive protein (hs-CRP), and serum uric acid (SUA) levels in ESRD patients. Materials and methods: Fifty-two ESRD patients (25 hemodialysis and 27 peritoneal dialysis) and 17 healthy controls were included in the study. Serum levels of myoglobin, CK-MB, and cTnI were measured and ultrasonographic CA–IMT was determined in all participants. SUA and hs-CRP levels were only measured in the ESRD group. Results: Serum myoglobin, CK-MB levels, and the mean CA–IMT were significantly higher in ESRD group (p < 0.01), whereas cTnI levels were not different compared to healthy controls (p = 0.70). There was also a positive correlation between CA–IMT and cTnI levels (p = 0.003, r = 0.35) and CA–IMT and hs-CRP (p = 0.03, r = 0.30) or SUA levels (p = 0.003, r = 0.43). Conclusion: cTnI may serve as a more sensitive marker in detecting cardiovascular events in patients with renal failure. Besides the traditional risk factors of atherosclerosis, cTnI, hs-CRP, and SUA may have a predictive role in recognizing premature atherosclerosis in ESRD patients.

Keywords:

• troponin I
• creatine kinase
• MB form
• myoglobin
• intima–media thickness
• kidney failure
• chronic
• uric acid
• C-reactive protein
• inflammation

INTRODUCTION

Cardiovascular diseases are the leading cause of death in patients with end-stage renal disease (ESRD), with an increased mortality 10–30 times of the general population, rendering them the “highest risk group” for subsequent cardiovascular events.¹ There is considerable effort to discover sensitive testing methods for early detection of uremic patients with increased cardiovascular risk.

Myoglobin, creatinine kinase-MB (CK-MB), and cardiac troponins are typical markers of acute myocardial injury in population; however, high rates of false positivity have been reported in patients with renal failure.² Although the exact mechanism is not known, it has been proposed that subclinical myocyte damage and apoptosis due to uremia or decreased clearance due to renal failure may be the underlying cause for the higher levels of cardiac markers in this group of patients.^3,4

Several prospective studies in recent years have documented that high levels of troponins are associated with higher mortality in dialysis patients^5–7 as well as increased left ventricular mass, left ventricular hypertrophy, and/or heart failure.^5,8,9 Majority of these studies tested troponin T (cTnT); however, troponin I (cTnI) was also associated with increased mortality in asymptomatic patients.^10,11

Atherosclerosis is known to be an inflammatory disease¹²; several inflammatory markers have been questioned for their potential in recognizing premature atherosclerosis. Recently, high-sensitive C-reactive protein (hs-CRP) has gained a lot of attention as a potential predictor of cardiovascular disease both in general and in ESRD populations.^13,14 Elevated hs-CRP levels, caused by the persistent inflammation in uremic milieu, have been shown to be an independent predictor of cardiovascular death in dialysis patients.^15,16

Uric acid, the end product of purine metabolism, seems to be associated with endothelial dysfunction through several mechanisms, such as impaired nitric oxide production and vascular smooth muscle proliferation.^17,18 Several epidemiologic studies have suggested that uric acid is related to cardiovascular events.^19,20 However, it is not clear whether there is a causal relationship between hyperuricemia and cardiovascular diseases, or this is a simple coexistence with the components of metabolic syndrome, that explains this increased cardiovascular risk.

CA–IMT is a measure of the thickness of the intima and media layer of the carotid artery, assessed by ultrasound. It is a simple, reliable, and non-invasive method, widely used in clinical trials for detecting asymptomatic atherosclerosis.²¹ The presence of plaques and increased intima–media thickness in the carotid arteries is a strong predictor for cardiovascular events in general population.²² CA–IMT is increased in patients with renal failure and may help in predicting patients that are at a higher risk for future cardiovascular events.^23,24

The first aim of this study was to determine whether there was a difference in serum myoglobin, CK-MB, and cTnI levels between asymptomatic patients with ESRD and healthy controls. Secondly, we aimed to analyze whether these markers showed correlation with CA–IMT, lastly we analyzed whether there was a correlation between serum uric acid (SUA), hs-CRP levels, and CA–IMT in ESRD patients.

MATERIALS AND METHODS

Patient population and study design

A total of 52 patients with ESRD (chronic kidney disease stage 5) and 17 healthy controls were enrolled in the study. ESRD patients were composed of 25 hemodialysis (HD) patients (10 females, 15 males) undergoing standard bicarbonate dialysis routinely three times a week and 27 continuous ambulatory peritoneal dialysis (CAPD) patients (16 females, 11 males). The average urea Kt/V in HD and CAPD patients were 1.29 ± 0.18 and 1.66 ± 0.43, respectively. Patients without symptoms or history of cardiac disease were selected. None of the HD patients were experiencing significant hypotension or anginal episodes during the intradialytic period. Healthy controls (n = 17; 10 females, 7 males) were non-diabetic, normotensive volunteers with normal serum creatinine and without symptoms or history of cardiac disease. Demographic and clinical data were collected at entry.

The study was approved by the local ethical committee and written informed consent was obtained from all patients prior to study entry. The study was conducted according to the declaration of Helsinki.

Measurement of biochemical parameters, cTnI, CK-MB, and hs-CRP

Venous blood from study and control groups were collected in the morning (before the dialysis session in HD group, before the first peritoneal dialysis exchange of the day in CAPD group) from a peripheral vein. The serum and EDTA plasma aliquots were stored at −70°C until analysis. Hemoglobin, serum levels of myoglobin, CK-MB, cTnI, and biochemical parameters were measured. Routine biochemical variables were measured by standardized methods using autoanalyzers. Myoglobin, cTnI, and CK-MB tests were assayed using ELISA test kits (DRG Instruments GmbH, Marburg/Lahn, Germany; EIA-3955 myoglobin, EIA-2952 cTnI, EIA-4361 CK-MB). The hs-CRP levels were measured by nephelometric method.

Measurement of carotis intima–media thickness

B-mode ultrasonographic imaging of the carotid artery was performed using a high-resolution, real-time ultrasonograph (ATL HDI 5000, Philips Medical, Bothell, WA, USA) with a 7.0-MHz linear array transducer. The transducer was manipulated so that the near and far walls of the common carotid artery (CCA) were parallel to the transducer footprint, and the lumen diameter was maximized in the longitudinal plane. Examination and image analysis were performed by trained sonographers kept unaware of other data. The measurements were obtained from the segment 2 cm proximal to the CCA bulbus where no plaque was present. The CA–IMT was measured in the far wall of the vessel as the distance from the leading edge of the lumen–intima interface to the leading edge of the media–adventitia interface. The arithmetic mean value of the measurements of right and left CCA measurements was taken as the mean CA–IMT.^25,26

Statistical analysis

Descriptive statistics for all the identified variables (age, sex, and dialytic age) were performed. The results were given as mean ± SD (normally distributed data). The differences between the three groups were assessed either by parametric (one-way ANOVA and Turkey's test) or by non-parametric (chi-square test) test. For the comparisons of the groups, Mann–Whitney U-test was used for non-parametric variables and chi-square test for parametric variables. Correlations between characteristics and CA–IMT were made using Pearson's correlation test. p-Values less than 0.05 were considered statistically significant. The Statistical package for Social Sciences (version 11.0) (SPSS, Inc., Chicago, IL, USA) was used to perform all statistical calculations.

RESULTS

Characteristics of patients

A total of 52 patients with ESRD and 17 healthy controls were included in the analysis. Table 1 summarizes the baseline characteristics of the participants. The mean age was 53.26 ± 8.18, 49.29 ± 11.75, and 49.35 ± 7.47 in the HD, CAPD, and control groups, respectively (p = 0.27). There was a female dominance in the CAPD and control groups (59.3 and 58.8%) compared to the HD group (40.0%), but the difference was not statistically significant (p = 0.31) (Table 1).

TABLE 1.  The characteristics of the study population

	CKD stage 5 patients		Control group (n = 17)	p	p**
	HD group (n = 25)	CAPD group (n = 27)
Age (year)	53.26 ± 8.18	49.29 ± 11.75	49.35 ± 7.47	0.44	0.27
Gender (F/M)	10/15	16/11	10/7	0.16	0.31
Primary disease
Hypertension	5	7	NA	0.57	NA
Diabetes mellitus	7	6
Glomerulonephritis	2	0
Unknown	4	8
Others	7	6
Duration of dialysis (mo)	50.32 ± 24.73	49.22 ± 48.47	NA	0.33	NA
Baseline laboratory values
Hemoglobin (g/dL)	11.35 ± 1.45	10.96 ± 1.49	NA	0.32	NA
Albumin (g/dL)	3.92 ± 0.49	3.89 ± 0.48	NA	0.45	NA
Calcium (mg/dL)	8.55 ± 0.96	9.01 ± 0.86	NA	0.08	NA
Inorganic phosphate (mg/dL)	4.61 ± 1.43	5.17 ± 1.45	NA	0.13	NA
Total cholesterol (mg/dL)	163 ± 39	181.29 ± 38.37	NA	0.18	NA
HDL (mg/dL)	41.52 ± 10.97	45.29 ± 9.59	NA	0.10	NA
LDL (mg/dL)	88.68 ± 33.49	102.29 ± 27.75	NA	0.12	NA
Triglyceride (mg/dL)	170.88 ± 101.72	162.29 ± 82.69	NA	0.83	NA
Uric acid (mg/dL)	6.56 ± 1.25	5.42 ± 1.03	NA	0.001	NA
Hs-CRP (mg/dL)	0.89 ± 1.06	1.14 ± 1.1	NA	0.23	NA
Fasting plasma Glucose (mg/dL)	109.32 ± 35.97	115.18 ± 37.90	NA	0.52	NA

Notes: *p for HD vs. CAPD;

**p for patient group vs. control.

Abbreviations: CKD, chronic kidney disease; HD, hemodialysis; CAPD, continuous ambulatory peritoneal dialysis; CIMT, carotid intima–media thickness; F, female; M, male; HDL, high-density lipoprotein; LDL, low-density lipoprotein; hs-CRP, high-sensitive C-reactive protein; NA, not available.

There was no significant difference between HD and CAPD groups for the etiology of renal disease (p = 0.57) and the mean duration of dialysis (p = 0.33) (Table 1).

Mean hemoglobin, serum albumin, calcium, phosphate, total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglyceride, fasting plasma glucose, and hs-CRP levels were similar between HD and CAPD groups. SUA level was 6.56 ± 1.25 mg/dL in the HD group and 5.42 ± 1.03 mg/dL in the CAPD group, and the difference was statistically significant (p = 0.001) (Table 1).

Comparison of the predictive markers of myocardial injury

There was no statistically significant difference between cTnI levels of patients and controls (p = 0.70), whereas myoglobin and CK-MB levels were significantly higher in ESRD group (p = 0.0001 and p = 0.003, respectively) (Table 2). Within the patient group, the mean values of myoglobin (p = 0.74), CK-MB (p = 0.19), and cTnI (p = 0.97) were not significantly different between HD and CAPD patients. The mean value of CA–IMT was significantly higher in ESRD group compared to controls (p = 0.001); meanwhile there was no significant difference between HD and CAPD groups (p = 0.51) (Table 2).

TABLE 2.  The comparison of CA–IMT and markers of myocardial injury between the groups

	CKD stage 5 patients		Control group (n = 17)	p	p**
	HD group (n = 25)	CAPD group (n = 27)
Myoglobin (ng/mL)	175.2 ± 100.83	170.55 ± 102.26	10.64 ± 4.91	0.74	0.0001
CK-MB (ng/mL)	2.96 ± 2.18	2.42 ± 1.98	0.93 ± 0.35	0.19	0.003
cTnI (ng/mL)	0.54 ± 0.84	0.44 ± 0.73	0.34 ± 0.52	0.97	0.70
CA–IMT (mm)	1.13 ± 0.49	1.00 ± 0.33	0.68 ± 0.18	0.51	0.001

Notes: *p for HD vs. CAPD;

**p for patient group vs. control.

Abbreviations: CKD, chronic kidney disease; HD, hemodialysis; CAPD, continuous ambulatory peritoneal dialysis; CIMT, carotid intima–media thickness; CK-MB, creatinine kinase-MB; cTnI, troponin I.

Factors associated with CA–IMT

SUA (p = 0.003, r = 0.43), hs-CRP (p = 0.03, r = 0.30), creatinine (p = 0.003, r = 0.34), and cTnI levels (p = 0.003, r = 0.35) were positively correlated with CA–IMT. There were no correlations between CA–IMT and myoglobin (p = 0.51, r = 0.09) or CK-MB (p = 0.40, r = 0.11) levels. Serum albumin levels (p = 0.01, r = −0.28) were negatively correlated with CA–IMT (Table 3).

TABLE 3.  Factors associated with the degree of CA–IMT

	Coefficient (r)	p-Value
Age (year)	0.08	0.48
Duration of dialysis (mo)	0.28	0.07
Baseline laboratory values
Hemoglobin (g/dL)	−0.11	0.35
Creatinine (mg/dL)	0.34	0.003
Uric acid (mg/dL)	0.43	0.003
Albumin (g/dL)	−0.28	0.01
Calcium (mg/dL)	0.05	0.70
Inorganic phosphate (mg/dL)	0.18	0.07
Fasting serum glucose (mg/dL)	0.04	0.77
Total cholesterol (mg/dL)	0.18	0.17
LDL (mg/dL)	0.19	0.17
Triglyceride (mg/dL)	0.18	0.19
Hs-CRP (mg/dL)	0.30	0.03
Predictors of myocardial injury
Myoglobin (ng/mL)	0.09	0.51
CK-MB (ng/mL)	0.11	0.40
cTnI (ng/mL)	0.35	0.003

Abbreviations: LDL, low-density lipoprotein; hs-CRP, high-sensitive C-reactive protein; CK-MB, creatinine kinase-MB; cTnI, troponin I.

DISCUSSION

Renal insufficiency alters the metabolism of many proteins; thus the clinical significance of elevated cardiac biomarkers in patients with ESRD is hard to interpret. In this study, we demonstrated that there are significant elevations of myoglobin and CK-MB in asymptomatic ESRD patients. False positive increases of both myoglobin and CK-MB have previously been reported in the literature.^27–29 Troponins are recognized to be more sensitive markers compared to myoglobin and CK-MB in that sense, but still the data on value of different subgroups of troponins display conflicting results. Several studies suggest that cTnT is more frequently elevated than cTnI in dialysis patients in the absence of myocardial damage and cTnI seems to be a more specific marker than CK-MB or cTnT in diagnosis of acute MI.^3,30 In a summation of studies including 350 chronic renal disease patients, cTnT was falsely elevated in 47% whereas cTnI in 2.6%.³¹ The elevation of troponins in asymptomatic patients has been attributed to impaired renal clearance,⁴ microinfarction,³² or left ventricular hypertrophy.³³ The difference between cTnT and cTnI may be explained by a shorter half-life for cTnI,³⁴ different expressions in skeletal and cardiac muscle cells,^35,36 and the undetermined net effect of dialysis on troponins: decreased cTnI and elevated cTnT.^37,38 Urso et al., reported that cTnI levels were not modified by HD.² In our study, the discordant elevation of myoglobin and CK-MB in asymptomatic dialysis patients, in contrast to the similarity of cTnI levels between the patient and the control groups, strengthens the cardio-specificity of cTnI in diagnosing cardiovascular events in this specific patient group.

Studies suggests that troponin levels may predict long-term mortality in patients with ESRD. Elevated cTnT levels were found to be significantly associated with increased all-cause mortality in dialysis patients, in a meta-analysis of 28 studies conducted between 1999 and 2004.⁷ However, the interpretation of association between elevated cTnI levels and mortality were unclear from that meta-analysis due to lack of standardization of assays in the literature, although when individually evaluated, studies point out the prognostic value of cTnI in asymptomatic dialysis patients.^10,39–41 The association of elevated cTnT and cTnI levels with coronary artery calcification in asymptomatic dialysis patients may represent this link between troponins and mortality in the vascular frame.⁴²

We observed a significant positive correlation between the CA–IMT and the cTnI levels within the patient group. Hojs et al. demonstrated that CA–IMT values were significantly higher in patients with elevated cTnT levels.⁴³ Consistent with our findings, the study of Zumrutdal et al. showed a similar association between CA–IMT and cTnI.⁴⁴

In our study we found that hs-CRP levels were significantly and independently correlated with CA–IMT. Zoccali et al. reported a similar independent association between CRP and CA–IMT in dialysis patients.¹⁶ This is a finding that draws attention to the link between inflammation and atherosclerosis, and CRP is now getting to be viewed as a non-traditional risk factor for dialysis patients.⁴⁵ Increased cTnT levels have been found in association with proinflammatory cytokines such as CRP or IL-6 in the literature⁴⁶ and Zumrutdal et al. reported a positive correlation between cTnI and CRP.⁴⁴ A prospective study of asymptomatic dialysis patients revealed that combination of increased cTnI and CRP was associated with a dramatic increase in all-cause mortality.⁴⁷ We did not find any correlation between hs-CRP and cTnI; the use of some drugs (such as antibiotics or statins), presence of silent infections or intercurrent events in dialysis patients, or the cross-sectional design of our study may play a role on this outcome.

In this study, we also found that SUA levels were significantly correlated with CA–IMT. Previous studies failed to demonstrate an association between SUA and CA–IMT in dialysis patients.^48,49 However, this association has been observed previously in other patient groups: Tavil et al. reported higher CA–IMT in hyperuricemic hypertensive patients⁵⁰ and uric acid levels were associated with CA–IMT in healthy postmenopausal women,⁵¹ the elderly,⁵² and the hypercholesterolemic.⁵³ Our findings may suggest that SUA levels may serve as a predictive marker for atherosclerosis in dialysis patients as well. There were no correlations between SUA and cTnI or other cardiac markers in our study; the associations between uric acid and cardiac markers in dialysis patients have not been studied in the literature.

The negative correlation between serum albumin and CA–IMT is also a notable finding in our study. Hypoalbuminemia, a marker of malnutrition and underlying inflammation, is associated with increased risk of cardiovascular disease and mortality in patients with ESRD.^54,55 Savage et al. reported that serum albumin was inversely related to intima–media thickness in non-diabetic ESRD patients, though in this study, mean IMT were similar in patients and controls.⁵⁶ Similarly, Oh et al. demonstrated an independent negative correlation between serum albumin and CA–IMT in a small group of peritoneal dialysis patients.⁵⁷ On the other hand, ARIC study investigators found no association between serum albumin and CA–IMT in a large group of non-renal patients.⁵⁸ Our finding is a positive contribution to the highly debated interaction between malnutrition, inflammation, and atherosclerosis in uremic patients.

The main limitations of our study are relatively small sample size and cross sectional design. Larger scale and long follow-up trials are needed to confirm of our findings.

In conclusion, we suggest that cTnI is a more sensitive marker for detecting cardiac ischemia in asymptomatic dialysis patients. cTnI may also serve as a predictor of asymptomatic atherosclerosis in this patient group along with hyperuricemia and elevated hs-CRP.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.