Determinants of Coronary Artery Disease in Nondiabetic Hemodialysis Patients: A Matched Case-Control Study

Abstract

Background/Aims. The aim of this matched case-control study was to evaluate the determinants of coronary artery disease (CAD) other than conventional risk factors in nondiabetic hemodialysis (HD) patients. Methods. Among 312 consecutive patients on regular HD, 26 nondiabetic patients with angiographically defined coronary artery disease (20 men, 6 women; mean age 57.0 ± 13 years) constituted the case group (group 1). A subject group of the same gender, smoking status, and hypertension with similar ages and body mass indexes who had normal electrocardiography and myocardial perfusion scintigraphy served as controls (20 men, 6 women; mean age 54.1±12 years, group 2). Demographics, high sensitivity C-reactive protein (hs-CRP), erythrocytes dimentation rate (ESR), hematocrit-corrected ESR, beta-2 microglobulin, cardiac troponin I, parathyroid hormone, albumin, calcium (Ca), phosphorus (P), Ca × P, and lipid profiles were compared between the groups. Results. Patients in group 1 had higher hs-CRP and troponin I (18.0±12 vs. 7.2±5 mg/L, p < 0.001; 0.36±0.16 vs. 0.22±0.05 ng/mL, p < 0.001, respectively) and lower HDL cholesterol levels than group 2 (37.0±10mg/dL vs. 46.3±17mg/dL, p = 0.02). Backwards stepwise logistic regression analysis revealed that high hs-CRP and troponin I levels (p = 0.03 and p = 0.01) and low HDL cholesterol levels (p = 0.02) were independently related with CAD. Conclusion. According to these results, in nondiabetic patients on regular hemodialysis, high hs-CRP, troponin I levels and low HDL-cholesterol were the determinants of CAD.

Keywords:

• coronary artery disease
• cardiac troponin I
• C-reactive protein
• hs-CRP
• hemodialysis
• nondiabetic

INTRODUCTION

Cardiovascular (CV) disease is a major cause of morbidity and mortality in patients with end-stage renal disease. Traditional risk factors for CV disease include hypertension, smoking, diabetes, dyslipidemia, left ventricular hypertrophy, and male sex in the general population. Although hemodialysis patients have a high prevalence of many of these factors, they also have nontraditional, or uremia-related, specific factors such as anemia, altered calcium-phosphorus metabolism, inflammation, oxidative stress, hypoalbuminemia, and hyperhomocysteinemia.[1],[2]

Diabetes is not only a traditional risk factor for CV disease, but also one of the most common causes of end-stage renal disease.[3],[4] Thus, epidemiologic studies are less available about the predictive factors of CV disease in nondiabetic hemodialysis (HD) patients. This study sought to investigate the potential determinants of coronary artery disease (CAD) in nondiabetic HD patients.

METHODS

Patients with a fasting plasma glucose <110 mg/dL on more than one occasion without using antidiabetic drugs and a previous diagnosis of diabetes mellitus were defined as nondiabetic. Each of the study groups consisted of 26 nondiabetic patients on regular HD, either with CAD (group 1) or without CAD (group 2). Subjects in the group 1 (mean time on HD, 42.1±27.8 months) and group 2 (mean time on HD, 39.2±30.5 months) were selected from a total of 312 consecutive patients at the Adana Hemodialysis Center at Baskent University Hospital. Group 1 consisted of patients with angiographic evidence of CAD (with >50 % coronary artery stenosis determined by selective coronary angiography) and/or a history of coronary artery bypass graft or percutaneous transluminal coronary angioplasty (20 men, 6 women, mean age 57.0 ± 13 years). Patients on at least one antihypertensive medications and with a history of hypertension were recorded as hypertensive patients. Subjects who were not taking any antihypertensive agent with a mean blood pressure under 140/90 mm/Hg and did not have any history of hypertension were recorded as nonhypertensive. The mean body mass index (BMI) was defined as the weight in kilograms divided by square of height in meters (kg/m²). Smoking status was recorded as current or former smokers or never-smokers. Two patients had peripheral arterial disease and one patient had a history of transient ischemic attack in group 1.

For every patient in group 1, a subject was identified from the database of transplantation waiting list with the following characteristics: same gender, smoking status, and hypertension; a similar age (±5 years) and BMI (±2 kg/m² ); and a normal electrocardiography and myocardial perfusion scintigraphy, with no history or evidence of cerebrovascular or peripheral vascular disease. Patients whose chest radiographs showed calcified plaques in the aortic arch, ventricular wall motility disorders, or valvular calcifications on echocardiography were excluded. These subjects constituted the control group (20 men, 6 women, mean age 54.1±12 years, group 2). BMI was 23.4±3 in group 1 and 22.3±3 kg/m² in group 2.

Patients with conditions known to be associated with acute-phase responses (those who had been hospitalized and/or treated during the previous two months for acute infection, malignancy, tuberculosis, chronic inflammatory disease, or liver disease) were excluded.

Causes of renal failure in group 1 and 2, respectively, were polycystic kidney disease in 2 and 1 patients (7.7%, 3.9%), glomerulonephritis in 2 and 3 patients (7.7%, 11.5%), hypertension (HT) in 6 and 5 patients (23.1%, 19.2%), infectious or obstructive uropathies in 5 and 6 patients (19.2%, 23.1%), and unknown causes in 11 and 11 patients (42.3% in both groups).

All patients were on a three times per week HD program, with bicarbonate dialysate and a Hemophan hollow fiber membrane (Idemsa, Spain) for 4–5 hours. The dialysis prescription sought to achieve a value of ≥ 0.65 for the urea reduction ratio and a value of Kt/V ≥ 1.2. The study protocol was reviewed and accepted by the ethical committee of the Baskent University, and informed consent was obtained from each patient.

In the morning of a routine dialysis day, venous blood samples were taken from a peripheral vein under fasting conditions. High-sensitivity serum C-reactive protein (hs‐CRP); erythrocyte sedimentation rate (ESR), beta 2 microglobulin (β2M), cardiac troponin I (cTnI), intact parathyroid hormone (iPTH), albumin, calcium (Ca), phosphorus (P), total and high-density lipoprotein (HDL) cholesterol, and triglyceride levels were measured.

Hs-CRP was measured by immunonephelometry using BN* Systems (CardioPhase, Dade Behring, Marburg, Germany), ESR by the Westergren method, β2M and iPTH were by microenzyme immunoassay (AxSYM, Abbott Diagnostics, Wiesbaden Germany; Immulite, Barcelona, Spain), cTnI by immunometric assay (Immulite Analyzer, DPC, Los Angeles, California, USA), albumin by immunochemical reaction (Behring Nephelometers, Marburg, Germany), and Ca, P, total and HDL cholesterol, and triglyceride levels were measured by routine laboratory techniques using an automated analyzer (Hitachi 912, Roche-Hitachi Modular Analytics, Roche Diagnostics GmbH, Mannheim, Germany). Low-density lipoprotein (LDL) cholesterol was calculated using the Friedewald formula. Hematocrit-corrected ESR was calculated according to ESRx Hct/45 formula, as previously described.[5] Expected values for healthy individuals in the literature are ≤ 3 mg/L for hs-CRP. Intra- and inter-assay coefficients of variance (CVs) for hs-CRP were 4.0% at 15 mg/L and 2.6% at 14 mg/L. Expected values for healthy individuals were below 1.0 ng/mL for cTnI. Intra and inter CVs for cTnI were 5.8% at 0.8±0.04 ng/mL and 8.4% at 0.8 ± 0.067 ng/mL. Epidemiologic and analytical characteristics were compared between the two groups.

Statistical Analysis

Data were collated and analyzed with SPSS 12.0 computer program. The numeric variables were expressed as the mean±SD, the categorical variables as percentage. Values for the parameters not conforming to a normal distribution were presented as median (min-max). Risk factors between the groups were compared using Mann-Whitney U test. Backward stepwise logistic regression analysis was performed to evaluate independent determinants of CAD in the study population. The variables in the regression model were selected according to the results of the univariate analysis (with a p value less than 0.10). Variables in the model include hsCRP, ESR, hematocrit-corrected ESR, cTnI, β2M, Ca, and HDL cholesterol. Bivariate Pearson's correlations were examined between all the variables. A p value < 0.05 was considered statistically significant.

RESULTS

Of individuals in group 1 and group 2, 76.9% were male, 61.5% were hypertensive, and 30.8% were current or former smokers. Age, BMI and time on HD were not statistically different between the groups (p > 0.05 for all). Patients in group 1 had higher hs-CRP and troponin I levels and lower HDL cholesterol levels than group 2. Although Ca, ESR, and hematocrit-corrected ESR levels were higher in group 1 than group 2, the differences were not statistically significant. Table 1 shows baseline and analytical characteristics of groups 1 and 2.

Table 1 Comparison of clinical and analytical characteristics of groups 1 and 2

	Group 1 (n =26)	Group 2 (n = 26)	p value
Age (years)	57.0 ± 13	54.1 ± 12	0.18
Body mass index (kg/m^2)	23.4 ± 3	22.3 ± 3	0.12
Time on hemodialysis (months)	42.1 ± 27.8	39.2 ± 30.5	0.23
Hs-CRP (mg/L)	18.0 ± 12	7.2 ± 5	<0.001
ESR (mm/h)	52 ± 15	44 ± 21	0.09
	57 (10–68)^*	42.5 (10–80)^*
Hct-corrected ESR (mm/h)	39.5 ± 13	32.7 ± 15	0.08
	44 (8–60)^*	33.5 (7–63)^*
Beta-2 microglobulin (mg/dL)	5.4 ± 1.4	4.8 ± 1.5	0.06
	5.4 (2.6–8.0)^*	4.5 (1.8–8)^*
Cardiac troponin I (ng/mL)	0.36 ± 0.16	0.22 ± 0.05	<0.001
Parathyroid hormone (pg/mL)	321.2 ± 231.8	337.2 ± 338.0	0.64
Albumin (gr/dL)	3.8 ± 0.4	4.0 ± 0.5	0.2
Ca (mg/dL)	9.54 ± 0.97	9.0 ± 0.98	0.06
P (mg/dL)	5.5 ± 1.4	5.7 ± 1.6	0.5
Ca × P (mg/dL)	51.4 ± 16.1	53.1 ± 16	0.7
Total cholesterol (mg/dL)	174.5 ± 47.5	176.7 ± 34.7	0.4
	162 (90–341)^*	173 (90–234)^*
LDL cholesterol (mg/dL)	98.1 ± 37	95.0 ± 26	0.9
	92.4 (44–149)^*	91.6 (28–151)^*
HDL cholesterol (mg/dL)	37.0 ± 10.0	46.3 ± 17	0.02
	33.5 (25–61)^*	47 (13–106)^*
Triglyceride (mg/dL)	197.4 ± 77	176.5 ± 81.0	0.3
	200 (76–398)^*	163 (76–400)^*

Values are expressed as mean ± SD.

Abbreviations: Hs-CRP = high-sensitivity C-reactive protein, ESR = erythrocyte sedimentation rate, hct = hematocrit, Ca = calcium, P = phosphorus, LDL = low-density lipoprotein, HDL = high-density lipoprotein.

^*Values are expressed as median (min-max).

In logistic regression analysis, high hs-CRP and cTnI levels (Odds Ratio [OR] 1.415 per increase in 1 mg/L 95% confidence interval [CI], 1.042 to 1.920, p = 0.03; and OR 1.030 per increase in 1 ng/mL 95% CI 1.008 to 1.053, p = 0.01) and low HDL cholesterol levels (OR 0.865 per increase in 1mg/dL, 95% CI 0.756 to 0.977, p = 0.02, respectively) were independently related with CAD. Hs-CRP levels were correlated with age (r = 0.327, p = 0.02), ESR (r = 0.329, p = 0.02), hematocrit-corrected ESR (r = 0.374; p = 0.01), and cTnI (r = 0.283; p = 0.04).

DISCUSSION

This study found that high hs-CRP, troponin I, and low HDL cholesterol were the determinants of CAD in nondiabetic HD patients. Predictive value of CRP in CV risk and mortality in HD patients was shown in numerous studies, and CRP has been a new addition to CV risk factor profiles.[6],[7] All cause and CV mortality and a poorer survival were higher in patients with elevated CRP in patients on HD.[8],[9] Although a wide variety of potential sources may be associated with elevated CRP in this patient population,[8],[10] underlying CAD may be one of the possible links for this elevation. These results were in accordance with the literature linking CRP and CV disease in HD patients.[11]

Elevated serum troponin concentrations frequently are observed in HD patients without evidence of acute coronary disease, and their clinical significance is still under investigation.[12] CTnI is exclusively of cardiac origin and does not express in the skeletal muscle at any developmental stage. Thus, it has been proposed to be a more specific, diagnostic, and prognostic marker of patients with renal failure.[12–14] In this study, even small elevations of cTnI concentration, at levels lower than those traditionally used for the diagnosis of acute cardiac events, were associated with the presence of CAD in asymptomatic dialysis patients, independently. Thus, small and non-specific increases in cTnI levels may reflect underlying CAD in nondiabetic HD patients. Further studies on troponin are needed to establish their role as a new CV risk marker.[15]

Previously, an association between elevated cardiac troponin T and CRP was shown in predialysis patients.[16] In this study, hs-CRP was moderately and positively correlated with age, ESR, hematocrit-corrected ESR, and cTnI levels. These results also suggest a possible association between cTnI and inflammation. Further studies are needed to elucidate the importance of this association.

Dyslipidemia may be present in >90% of HD patients, and the evidence is stronger that low HDL cholesterol level is a risk factor for CV disease in the HD population.[17] This study showed that low HDL cholesterol was one of the independent determinants of CAD in nondiabetic HD patients among other lipid parameters.

There were several limitations in the present study. One major limitation of this study is that this is a case-control study on a small number of patients. A larger study may reveal more conclusive data. Secondly, the exclusion of CAD in the control group was done with myocardial perfusion scintigraphy. However, it is not usually possible to use coronary angiography in clinical trials of dialysis patients who have normal results with non-invasive tests.[18] Another limitation is that although case-control studies are powerful tools for highlighting the differences, their cross-sectional nature do not allow definition of causal relations.

In conclusion, this study documented the determinants of CAD in a substantial subgroup of HD patients. Additional and larger scale studies are needed for the confirmation of these findings.

This article was awarded “The Best Abstract Presented by Young Authors” at XL2 Congress European Renal Association European Dialysis and Transplant Association Congress (ERA-EDTA), June 4–7, 2005, Istanbul, Turkey.