Abstract

Background. Accelerated atherosclerosis and vascular calcification are common in chronic haemodialysis (HD) patients. In this study, we aimed to investigate the relationship between left ventricular hypertrophy (LVH) in HD patients and atherosclerosis and vascular calcification measured by electron beam computed tomography (EBCT).

Methods. In a cohort of 118 HD patients (52 male, 66 female, mean age: 46±13 years), we measured biochemical parameters, including BUN, creatinine, albumin, haemoglobin, C-reactive protein and fibrinogen levels, and performed echocardiography, high-resolution B-mode carotid ultrasonography and EBCT in 85 of them. The degree of stenosis was measured at four different sites (communis, bulbus, interna and externa) in both carotid arteries. Carotid plaque scores were calculated by summing the degrees of stenosis measured at all locations.

Results. LVH was detected in 89 of the patients (75%). Plaque-positive patients had higher left ventricular mass index (LVMI) than plaque-negative patients (175±59 vs 143±46 g/m2, P = 0.003). LVMI was correlated with systolic blood pressure (r = 0.62, P<0.001), pulse pressure (r = 0.58, P<0.001), haemoglobin levels (r = – 0.25, P = 0.008), carotid plaque score (r = 0.32, P = 0.001) and coronary (CACS) and aortic wall calcification score (AWCS) (r = 0.34, P = 0.002 and r = 0.43, P<0.001, respectively). Multiple linear regression analysis (model r = 0.76) showed the independent factors related to LVMI to be systolic blood pressure, pulse pressure, CACS and presence of carotid plaques.

Conclusion. Extra-coronary atherosclerosis and vascular calcification are associated with LVH in HD patients. Whether the treatment of atherosclerosis or vascular calcification may cause regression of or even prevent LVH in HD patients remains to be seen.

Introduction

Cardiovascular mortality in chronic haemodialysis (HD) patients is much higher than in an age-matched general population, especially for younger individuals. Possible factors behind this observation are the early development of left ventricular hypertrophy (LVH), atherosclerotic heart disease and arrhythmias in HD patients [1]. LVH, a strong independent predictor of cardiovascular mortality [2], stands out among those factors as being the one that is potentially preventable and reversible [3]. A number of factors present in HD patients—including hypertension, hypervolaemia, anaemia, arteriovenous shunts and increased sympathetic activation—are implicated in the development of LVH [4].

Attention has been drawn previously to the close association between arterial changes and cardiac hypertrophy. London et al. [5] indicated that a correlation exists between arterial elasticity and left ventricular mass index (LVMI) in patients with end-stage renal disease (ESRD). Similarly, a correlation exists between LVMI and the intima-media thickness (IMT) of the carotid artery, another marker of atherosclerosis [6]; and we recently have reported a moderate correlation between LVMI and endothelial dysfunction [7]. Taken together, these findings suggest an association between atherosclerosis and LVH in HD patients. While such an association has been reported previously in non-uraemic patients [8], to our knowledge, the association between atherosclerosis and LVH in HD patients has not been studied before.

In addition to atherosclerosis, another common arterial change in uraemic patients that is related to arterial stiffness is vascular calcification. Guerin et al. [9] showed a correlation between vascular calcification and increased aortic stiffness, which is implicated in the development of LVH in HD patients. They found that LVMI increased with rising vascular calcification scores, but not to a statistically significant degree. Since they evaluated arterial calcifications with a non-invasive and semi-quantitative B-mode ultrasonographic method, a more sensitive and quantitative assessment of vascular calcification might have given a more definitive answer. Electron beam computed tomography (EBCT) is a commonly used technique to measure coronary calcification. A number of studies using EBCT have shown an increased risk of vascular calcification in HD patients [10,11]. However, there currently are no available data on the association of vascular calcification, measured by EBCT, with LVMI in HD patients. Thus, we undertook this cross-sectional study to investigate the possible association of LVH with atherosclerosis and arterial calcification as measured by EBCT in HD patients.

Patients and methods

Patients

In this cross-sectional study, we enrolled 118 HD patients (52 male, 66 female; between the ages of 18 and 65 years, mean age 46±13 years), who had been on HD for >6 months for a mean time on HD of 68±51 months (range 6–187 months). Table 1 depicts the demographic characteristics and the biochemical and echocardiographic parameters of the patients. At the time of the study, 101 patients were dialysed three times a week and 17 patients were dialysed twice weekly for 4 h per session, using blood flow rates of 250–300 ml/min and dialysate flows of 500 ml/min. All patients were dialysed with a standard bicarbonate-containing dialysate bath (Na, 138 mmol/l; K, 1.25 mmol/l; HCO3, 33 mmol/l; Ca, 1.75 mmol/l; Mg, 0.75 mmol/l). The aetiologies of ESRD in the cohort were: primary glomerular diseases (n = 21), tubulointerstitial nephritis (n = 24), diabetic nephropathy (n = 11), other causes (n = 25) and unknown (n = 37). Of the 118 patients, 49 (42%) were hypertensive, on antihypertensive medications that included: angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (n = 18), calcium channel blockers (n = 29), β-blockers (n = 7) and others (α-blocker, nitrates or α-methyl dopa; n = 30). Of the cohort, 88 patients (75%) were receiving calcium acetate (Phosex tablet 250 mg, 750–2000 mg/day) as a phosphate binder; 75 (64%) were receiving erythropoietin at the time of the study. Our examinations of the patients conformed to good medical and laboratory practices and the recommendations of the Declaration of Helsinki on Biomedical Research involving Human Subjects.

Table 1.

The clinical characteristics and the results of biochemical and echocardiographic examinations in the study cohort (n = 118)

Clinical features  
    Hypertensive n (%) 49 (42%) 
    Smoker n (%) 38 (32%) 
    Diabetic n (%) 11 (9%) 
    Time on dialysis (months) 67.7±51.1 
    BMI (kg/m221.9±4.2 
    Ultrafiltration (kg) 3.02±1.11 
    Dose of calcium acetate (mg/day) 1431±397 
Biochemical parameters  
    BUN (mg/dl) 68±17 
    Creatinine (mg/dl) 10.1±2.6 
    Kt/V 1.21±0.18 
    Albumin (g/dl) 4.1±0.4 
    Ca (mg/dl) 9.2±0.9 
    P (mg/dl) 6.3±1.9 
    Alkaline phosphatase (IU/l) 320±232 
    iPTHa (ng/dl) 427±48 
    Total cholesterol (mg/dl) 169±51 
    Triglycerides (mg/dl) 180±133 
    HDL-cholesterol (mg/dl) 37±11 
    LDL-cholesterol (mg/dl) 92±38 
    Haemoglobin (g/dl) 11.3±1.9 
    Fibrinogen (mg/dl) 415±137 
    hs-CRPa (mg/dl) 1.19±0.14 
Echocardiographic measurements  
    Left atrial diameter (cm) 3.57±0.60 
    LVEDD (cm) 4.76±0.66 
    IVSd (cm) 1.30±0.36 
    LVPW (cm) 1.19±0.25 
    IVRT (s) 130±25 
    EF (%) 65±9 
    LVMI (g/m2164±59 
    Carotid plaque present 56% 
    Coronary calcification present 65% 
    Coronary artery calcification score median (25–75%) 82 (0–589) 
    Aortic wall calcification score median (25–75%) 0.5 (0–61) 
Clinical features  
    Hypertensive n (%) 49 (42%) 
    Smoker n (%) 38 (32%) 
    Diabetic n (%) 11 (9%) 
    Time on dialysis (months) 67.7±51.1 
    BMI (kg/m221.9±4.2 
    Ultrafiltration (kg) 3.02±1.11 
    Dose of calcium acetate (mg/day) 1431±397 
Biochemical parameters  
    BUN (mg/dl) 68±17 
    Creatinine (mg/dl) 10.1±2.6 
    Kt/V 1.21±0.18 
    Albumin (g/dl) 4.1±0.4 
    Ca (mg/dl) 9.2±0.9 
    P (mg/dl) 6.3±1.9 
    Alkaline phosphatase (IU/l) 320±232 
    iPTHa (ng/dl) 427±48 
    Total cholesterol (mg/dl) 169±51 
    Triglycerides (mg/dl) 180±133 
    HDL-cholesterol (mg/dl) 37±11 
    LDL-cholesterol (mg/dl) 92±38 
    Haemoglobin (g/dl) 11.3±1.9 
    Fibrinogen (mg/dl) 415±137 
    hs-CRPa (mg/dl) 1.19±0.14 
Echocardiographic measurements  
    Left atrial diameter (cm) 3.57±0.60 
    LVEDD (cm) 4.76±0.66 
    IVSd (cm) 1.30±0.36 
    LVPW (cm) 1.19±0.25 
    IVRT (s) 130±25 
    EF (%) 65±9 
    LVMI (g/m2164±59 
    Carotid plaque present 56% 
    Coronary calcification present 65% 
    Coronary artery calcification score median (25–75%) 82 (0–589) 
    Aortic wall calcification score median (25–75%) 0.5 (0–61) 

aData expressed as mean±SEM.

Methods

Blood pressure measurements were made manually using an Erka sphygmomanometer after a 5 min rest. Patients with pre-dialytic systolic blood pressures >140 mmHg or diastolic blood pressures >90 mmHg, or both, were considered to be hypertensive, as were those receiving any antihypertensive medication. The biochemical variables were: blood urea nitrogen (BUN), serum creatinine, calcium, phosphorus, intact parathyroid hormone (iPTH), magnesium, total protein, albumin, total cholesterol, high-density lipoprotein (HDL)-cholesterol, triglycerides and complete blood counts. Serum cholesterol and triglycerides were measured by standard enzymatic procedures. HDL-cholesterol levels were determined after the precipitation of apoB-containing lipoproteins by phosphotungstic acid. Low-density lipoprotein (LDL)-cholesterol levels were calculated according to the formula described by Friedewald et al. [12] which is valid in patients who have triglyceride levels <400 mg/dl:  

\begin{eqnarray*}&&\mathrm{LDL-cholesterol}\ {=}\ \mathrm{total\ cholesterol}\ {-}\ [(\mathrm{triglyceride{/}5})\\&&{\ }{+}\mathrm{HDL-cholesterol}]\end{eqnarray*}

Serum iPTH levels were measured by immunoradiometric assay (IRMA) with a commercially available kit (DSL-8000, Diagnostic Systems Laboratories, Inc. Webster, TX). High sensitive C-reactive protein (hs-CRP) levels were measured by the nephelometric method (Dade Behring, Germany, Catalogue No: 0QIY). Blood samples for biochemical parameters and serum iPTH levels were drawn immediately prior to HD in a post-absorptive state.

Echocardiographic examination

Echocardiography was performed the day after HD using the Vingmed System Five (Norway) echocardiographic system equipped with 2.5 MHz transducers. M-Mode and 2D measurements were done in accordance with methods recommended by the American Society of Echocardiography. Criteria for LVH were LVMI >134 g/m2 for males and >110 g/m2 for females. Cardiac mass was calculated using the formula derived by Reichek and Devereux [13]:  

\begin{eqnarray*}&&\mathrm{Left}\ \mathrm{ventricle}\ \mathrm{mass}\ (\mathrm{g})\ {=}\ \mathrm{1}\mathrm{.04}\ {\times}\ [(\mathrm{LVID}_{\mathrm{d}}\ {\pm}\ \mathrm{IVS}_{\mathrm{d}})\\&&{\ }{\pm}\ \mathrm{LVPW}_{\mathrm{d}})^{3}\ {-}\ (\mathrm{LVID}_{\mathrm{s}})^{3}]\\&&{\ }{-}\ \mathrm{13}\mathrm{.6}\end{eqnarray*}

Cardiac mass was standardized by dividing it by the body surface area, which was calculated using the Du Bois formula [14]:  

\begin{eqnarray*}&&\mathrm{Body}\ \mathrm{surface}\ \mathrm{area}\ {=}\ \left(\right.\mathrm{weight}^{0.425}\ {\times}\ \mathrm{height}^{\mathrm{0.725}}\\&&{\ }{\times}\ \mathrm{0.007184}\right)\end{eqnarray*}

Relative wall thickness (RWT) was calculated with the following formula:  

\begin{eqnarray*}&&\mathrm{RWT}\ {=}\ [(\mathrm{2}\ {\times}\ \mathrm{LV}\ \mathrm{posterior}\ \mathrm{wall}\ \mathrm{thickness})\\&&{\ }{/}\mathrm{LV}\ \mathrm{cavity}\ \mathrm{diameter}]\end{eqnarray*}

B-mode high-resolution carotid Doppler examination

Doppler ultrasonographic examinations were performed on 113 patients (96%) using an echocardiographic system equipped with 10 mHz linear transducers (Vingmed System Five). A trained sonographer scanned the common carotid arteries, the carotid bulbs and the first 2 cm of the internal and external carotid arteries, all bilaterally. The sonographer visualized the vessels in multiple planes, and then focused on the interfaces required for measuring the IMT and any areas of focal plaque formation. A local thickening of the intima >1 mm in thickness was defined as a plaque. IMT measurements were performed at a distance of 20 mm from the bifurcation in a plaque-free area of the common carotid. Measurements were done 3–4 times on both right and left sides, and the mean of these measurements was calculated and used. The intra-observer variability for IMT measurements was 4.2%, and the absolute difference between duplicate IMT measurements was 0.03 mm. Carotid artery stenosis was measured in four different locations (communis, bulbus, interna and externa) bilaterally. Carotid plaque scores (CPS) were calculated by summing up the degrees of stenosis in all locations in both sides. All measurements were recorded on VHS videotape for subsequent off-line analysis.

Electron beam computed tomography examination

EBCT studies were performed in 85 patients (72%) using an Ultrafast C-150 scanner (GE-Imatron, CA). The scan acquisition consisted of 40 contiguous transverse two-dimensional images of 3 mm thick sections extending from a level above the origins of the coronary arteries to the cardiac apex. The duration of exposure was 0.1 s per tomographic level, with a radiation intensity of 130 kVp, 630 mA. The images were acquired during diastole, with the ECG triggering at the 71% point of the R–R interval, using a 26 cm2 field of view and a 512 × 512 reconstruction matrix, and without a contrast agent.

Calcification was defined as a minimum of two adjacent pixels (>0.52 mm2) with a density >130 Hounsfield units. The peak intensity (Hounsfield unit) and area (mm2) of individual calcifications were calculated. As proposed by Agatston et al. [15], a calcium score was obtained by multiplying the area of each region of interest with a factor indicating the peak density within the individual area. In addition, aortic wall calcification scores (AWCS) were calculated using images taken from the segment of aorta beginning 4 cm above the Valsalva sinus in the ascending aorta and ending in the descending aorta at the level of the diaphragm. Image quality and scoring accuracy were assessed by a single radiologist, blinded to the patients' clinical and other laboratory results, who carefully inspected each image and made vessel-by-vessel analyses of the calcific foci (Figure 1).

Fig. 1.

Electron beam computed tomography image displays calcifications of the wall of the ascending aorta and three-vessel coronary artery calcifications.

Fig. 1.

Electron beam computed tomography image displays calcifications of the wall of the ascending aorta and three-vessel coronary artery calcifications.

Statistics

For statistical analysis, we used the Statistical Package for Social Sciences for Windows, version 10.0 (SPSS Inc, Chicago, IL). Between-group comparisons of continuous data for two groups were performed using the Student t-test or the Mann–Whitney U-test when appropriate. The χ2 test with Yates correction and Fisher's exact test were used for 2 × 2 contingency tables for non-numerical data, when appropriate. Correlations between numerical parameters with non-normal distribution were analysed with Spearman's rho correlation test. Comparisons of more than two groups were made with one-way analysis of variance (ANOVA) using the Tukey HSD test for post hoc analysis. In the stepwise linear regression analysis for predicting LVMI, the following parameters were included in the model as independent variables: systolic blood pressure, pulse pressure, haemoglobin, CACS and presence of plaque in the carotid system. Results are expressed as mean±SD unless otherwise stated. All tests of significance were two sided, and differences were considered statistically significant when the P-value was <0.05.

Results

On echocardiography, 89 of the 118 HD patients (75%) had LVH. A comparison of patients with or without LVH is presented in Table 2. Patients with LVH had higher systolic blood pressure, diastolic blood pressure and pulse pressure than patients without LVH. In addition, the overall prevalence of hypertension was higher and haemoglobin levels were lower in patients with LVH. It is noteworthy that patients with LVH also had significantly higher CPS, CACS and AWCS than those without LVH. However, there were no differences in gender distribution, biochemical parameters (including calcium, phosphorus and iPTH), inflammatory markers (hs-CRP and fibrinogen levels), time on dialysis or weekly dialysis frequency between patients with and without LVH; in addition, the prevalence of the use of calcium acetate and the daily dose of calcium acetate were similar between patients with and without LVH.

Table 2.

Comparisons between HD patients with and without LVH

 LVH (+) n = 89 LVH (–) n = 29 P-value 
Age (years) 47±13 43±12 NS 
Gender (male/female) 36/53 16/13 NS 
Time on dialysis (months) 66±51 73±52 NS 
Ultrafiltration (kg) 3.04±1.13 2.95±1.05 NS 
Systolic blood pressure (mmHg) 134±24 107±22 <0.001 
Diastolic blood pressure (mmHg) 85±13 76±15 0.007 
Pulse pressure (mmHg) 50±16 38±10 0.001 
Prevalence of hypertension (%) 47% 21% 0.004 
Ca (mg/dl) 9.2±0.9 9.1±0.8 NS 
P (mg/dl) 6.3±2.1 6.4±1.8 NS 
iPTH (ng/dl) 438±56 391±93 NS 
Total cholesterol (mg/dl) 167±44 172±68 NS 
Triglycerides (mg/dl)a 163±8 230±43 NS 
Haemoglobin (g/dl) 11.1±1.8 12.1±1.7 0.008 
hs-CRPa (mg/dl) 1.17±0.16 1.23±0.23 NS 
Fibrinogen (mg/dl) 412±143 425±119 NS 
Prevalence of carotid plaque % 62% 48% NS 
CPS median (25–75%) 20 (0–58) 1 (0–33) 0.04 
CACS median (25–75%) 89 (0–1076) 34 (0–268) 0.003 
AWCS median (25–75%) 2.3 (0–179) 0 (0–0) <0.001 
LVMI (g/m2184±53 103±22 <0.001 
 LVH (+) n = 89 LVH (–) n = 29 P-value 
Age (years) 47±13 43±12 NS 
Gender (male/female) 36/53 16/13 NS 
Time on dialysis (months) 66±51 73±52 NS 
Ultrafiltration (kg) 3.04±1.13 2.95±1.05 NS 
Systolic blood pressure (mmHg) 134±24 107±22 <0.001 
Diastolic blood pressure (mmHg) 85±13 76±15 0.007 
Pulse pressure (mmHg) 50±16 38±10 0.001 
Prevalence of hypertension (%) 47% 21% 0.004 
Ca (mg/dl) 9.2±0.9 9.1±0.8 NS 
P (mg/dl) 6.3±2.1 6.4±1.8 NS 
iPTH (ng/dl) 438±56 391±93 NS 
Total cholesterol (mg/dl) 167±44 172±68 NS 
Triglycerides (mg/dl)a 163±8 230±43 NS 
Haemoglobin (g/dl) 11.1±1.8 12.1±1.7 0.008 
hs-CRPa (mg/dl) 1.17±0.16 1.23±0.23 NS 
Fibrinogen (mg/dl) 412±143 425±119 NS 
Prevalence of carotid plaque % 62% 48% NS 
CPS median (25–75%) 20 (0–58) 1 (0–33) 0.04 
CACS median (25–75%) 89 (0–1076) 34 (0–268) 0.003 
AWCS median (25–75%) 2.3 (0–179) 0 (0–0) <0.001 
LVMI (g/m2184±53 103±22 <0.001 

aData expressed as mean±SEM.

iPTH = intact parathyroid hormone; hs-CRP = high sensitive C-reactive protein; LVMI = left ventricular mass index; CPS = carotid plaque score; CACS = coronary artery calcification score; AWCS = aortic wall calcification score.

Carotid plaques were present in 66 of the 113 (58%) patients. As can be seen in Table 3, patients with carotid plaques were older, and had higher total cholesterol, LDL-cholesterol, pulse pressures, LVMI, CACS and AWCS than patients without carotid plaques. The CPS was correlated with CACS (r = 0.40, P<0.001) and AWCS (r = 0.51, P<0.001, Spearman's rho test). When patients were classified according to tertiles of CPS [group 1 (no plaque), group 2 (0<CPS<30) and group 3 (CPS >30)], there was a gradual increase in LVMI with increasing CPS, with the lowest LVMI in patients without carotid plaques and the highest in patients with a plaque score >30 (144±46 g/m2 for group 1, 165 ± 63 g/m2 for group 2 and 182±56 g/m2 for group 3; P = 0.007). A post hoc analysis showed a significant difference between group 1 and group 3 (P = 0.005, Figure 2).

Fig. 2.

The results of comparisons of LVMI among the three groups with different carotid plaque scores: group I (no plaque), group II (0<CPS<30) and group III (CPS >30) (P = 0.007, one-way ANOVA).

Fig. 2.

The results of comparisons of LVMI among the three groups with different carotid plaque scores: group I (no plaque), group II (0<CPS<30) and group III (CPS >30) (P = 0.007, one-way ANOVA).

Table 3.

Comparisons between HD patients with and without coronary calcification

 Coronary calcification
 
 Carotid plaque
 
 
 (−) n = 30 (+) n = 55 (−) n = 47 (+) n = 66 
Age (years) 38±12 49±11 39±12 51±11 
Gender (male/female) 14/16 27/28 22/25 28/38 
Time on dialysis (months) 51±48 76±53 73±35 63±49 
Systolic BP (mmHg) 130±29 127±24 124±26 130±25 
Diastolic BP (mmHg) 89±16 81±12 82±16 83±13 
Pulse pressure (mmHg) 45±16 48±15 44±14 50±16 
Calcium (mg/dl) 9.0±0.9 9.1±0.9 9.2±0.9 9.1±0.8 
Phosphorus (mg/dl) 5.8±1.6 6.8±2.3 6.3±2.0 6.4±2.0 
iPTHa (ng/dl) 406±88 479±77 529±95 359±51 
Total cholesterol (mg/dl) 157±54 171±44 155±46 179±50 
Triglycerides (mg/dl) 169±104 199±170 174±109 172±70 
LDL-cholesterol (mg/dl) 85±41 92±32 82±34 101±39 
Haemoglobin (g/dl) 11.2±2.0 11.4±1.7 11.0±1.8 11.5±1.9 
hs-CRPa 0.88±0.15 1.31±0.25 1.00±0.18 1.32±0.21 
Fibrinogen (mg/dl) 375±122 451±149 388±112 435±156 
LVMI (g/m2143±48 178±57 143±46 175±59 
CPS median (25–75%) 0 (0–10.0) 28.5 (0–65.3) – 41 (18–71) 
CACSa median (25–75%) – 268 (89–1670) 1.3 (0–162) 131 (6.6–1394) 
AWCSa median (25–75%) 0 (0–0.25) 13.2 (0–205.6) 0 (0–1.2) 16.3 (0–218.9) 
 Coronary calcification
 
 Carotid plaque
 
 
 (−) n = 30 (+) n = 55 (−) n = 47 (+) n = 66 
Age (years) 38±12 49±11 39±12 51±11 
Gender (male/female) 14/16 27/28 22/25 28/38 
Time on dialysis (months) 51±48 76±53 73±35 63±49 
Systolic BP (mmHg) 130±29 127±24 124±26 130±25 
Diastolic BP (mmHg) 89±16 81±12 82±16 83±13 
Pulse pressure (mmHg) 45±16 48±15 44±14 50±16 
Calcium (mg/dl) 9.0±0.9 9.1±0.9 9.2±0.9 9.1±0.8 
Phosphorus (mg/dl) 5.8±1.6 6.8±2.3 6.3±2.0 6.4±2.0 
iPTHa (ng/dl) 406±88 479±77 529±95 359±51 
Total cholesterol (mg/dl) 157±54 171±44 155±46 179±50 
Triglycerides (mg/dl) 169±104 199±170 174±109 172±70 
LDL-cholesterol (mg/dl) 85±41 92±32 82±34 101±39 
Haemoglobin (g/dl) 11.2±2.0 11.4±1.7 11.0±1.8 11.5±1.9 
hs-CRPa 0.88±0.15 1.31±0.25 1.00±0.18 1.32±0.21 
Fibrinogen (mg/dl) 375±122 451±149 388±112 435±156 
LVMI (g/m2143±48 178±57 143±46 175±59 
CPS median (25–75%) 0 (0–10.0) 28.5 (0–65.3) – 41 (18–71) 
CACSa median (25–75%) – 268 (89–1670) 1.3 (0–162) 131 (6.6–1394) 
AWCSa median (25–75%) 0 (0–0.25) 13.2 (0–205.6) 0 (0–1.2) 16.3 (0–218.9) 

BP = blood pressure; iPTH = intact parathyroid hormone; hs-CRP = high sensitive C-reactive protein; LVMI = left ventricular mass index; CPS = carotid plaque score; CACS = coronary artery calcification score; AWCS = aortic wall calcification score.

aData expressed as mean±SEM.

0.01 <P<0.05; 0.001<P<0.01; P<0.001

Of the 85 HD patients, 55 (65%) had evidence of coronary artery calcification. These patients were older, were on dialysis longer and they had lower diastolic blood pressures and higher phosphorus and fibrinogen levels, LVMI, AWCS and CPS than patients without coronary calcification (Table 3). LVMI was also significantly higher in patients with aortic wall calcification compared with patients without calcification (196±64 vs 148±44 g/m2, P<0.001). The prevalence of calcium acetate usage (76 vs 67%, P = NS) and the daily dose of calcium acetate (1357±395 vs 1400±392 mg/day, P = NS) were similar between patients with and without coronary calcification, respectively. When patients were divided into groups according to tertiles of CACS [group 1 (no coronary calcification), group 2 (0<CACS<176) and group 3 (CACS >176)], LVMI was directly related to the severity of coronary calcification (144±49 g/m2 for group 1, 163±43 g/m2 for group 2 and 192±66 g/m2 for group 3, P = 0.004). Post hoc analysis revealed a significant difference in LVMI between group 1 and group 3 (P = 0.003). This association was also valid for AWCS (data not shown). The patients with coronary calcification had a similar isovolumetric relaxation time, which is an echocardiographic index of diastolic dysfunction, compared with patients without calcification (127±22 vs 130±22 ms, P>0.05)

The results of bivariate correlation for LVMI are given in Table 4. LVMI had a strong correlation with systolic blood pressure and pulse pressure. In addition, a modest correlation was present for LVMI and diastolic blood pressure, CACS, AWCS and CPS; a weak correlation was seen with IMT and haemoglobin levels (Table 4). There was no association between LV geometry the prevalence carotid plaque, plaque score, CACS or AWCS (data not shown).

Table 4.

Correlations of LVMI with other studied parameters (Spearman's rho test)

 Correlation coefficient (rP-value 
Haemoglobin levels −0.247 0.008 
Systolic BP 0.621 <0.001 
Diastolic BP 0.440 <0.001 
Pulse pressure 0.580 <0.001 
CACS 0.335 0.002 
AWCS 0.434 <0.001 
CPS 0.314 0.001 
IMT 0.279 0.003 
 Correlation coefficient (rP-value 
Haemoglobin levels −0.247 0.008 
Systolic BP 0.621 <0.001 
Diastolic BP 0.440 <0.001 
Pulse pressure 0.580 <0.001 
CACS 0.335 0.002 
AWCS 0.434 <0.001 
CPS 0.314 0.001 
IMT 0.279 0.003 

BP = blood pressure; CACS = coronary artery calcification score; AWCS = aortic wall calcification score; CPS = carotid plaque score; IMT = intima-media thickness of carotid artery.

In stepwise linear regression analysis, pulse pressure, systolic blood pressure, CACS and the presence of plaque in the carotid system were found to be independent predictors of LVMI, with this model explaining 56% of the variability in LVMI in our HD patients. Haemoglobin, however, was not found to be an independent variable in this model (model r = 0.76, P<0.001) (Table 5).

Table 5.

The results of stepwise linear regression analysis for predicting LVMI

 β±SE (95% CI) Standardized β P-value 
Pulse pressure (mmHg) 0.960±0.442 0.277 0.03 
Systolic BP (mmHg) 0.694±0.294 0.299 0.02 
CACS (log) 0.019±0.003 0.440 <0.0001 
Presence of carotid plaque 11.486±4.961 0.184 0.02 
 β±SE (95% CI) Standardized β P-value 
Pulse pressure (mmHg) 0.960±0.442 0.277 0.03 
Systolic BP (mmHg) 0.694±0.294 0.299 0.02 
CACS (log) 0.019±0.003 0.440 <0.0001 
Presence of carotid plaque 11.486±4.961 0.184 0.02 

Model r = 0.76, adjusted r2 = 0.56, F = 24.43, P<0.0001.

In multiple linear regression analysis, age (P<0.001), time on dialysis (P = 0.002), LVMI (P = 0.003), diastolic blood pressure (P = 0.03) and serum phosphorus (P = 0.03) turned out to be independent variables for predicting CACS (model r = 0.74). On linear regression analysis for predicting AWCS, age (P<0.001), serum phosphorus (P = 0.001), CPS (P = 0.026) and LVMI (P = 0.03) were identified as independent variables (model r = 0.75).

Discussion

LVH is common in patients on chronic HD, and it has been implicated as an independent risk factor for cardiovascular mortality in this population. In our study, 75% of the patients had LVH, which is consistent with previous reports [4,16]. In addition to the effects of the well known risk factors, such as hypertension, hypervolaemia and anaemia, arterial wall changes and decreased aortic elasticity have also been described to engender cardiac hypertrophy in uraemic patients [4–6]; however, the exact mechanisms responsible for increased aortic stiffness in HD patients are unclear. In hypertensive patients, aortic stiffness is strongly associated with the presence and severity of atherosclerosis [17,18], which is common and accelerated in HD patients. Therefore, the relationship between atherosclerosis and aortic stiffness, on one hand, and the LVMI described in the non-uraemic population, on the other, may also be true for HD patients. In our study, the patients who had carotid plaques had higher LVMI than the patients who did not; furthermore, their severity of atherosclerosis paralleled the increase in LVMI (Figure 2). In multiple regression analysis, the presence of plaque in the carotid system was an independent predictor of LVMI in HD patients. Together, these findings strongly suggest that atherosclerosis is associated with LVH in HD patients, as it is in the non-uraemic population. Previous data (e.g. correlations of IMT, serum homocysteine levels and endothelial dysfunction with LVMI) lend indirect support to this conclusion.

Patients with ESRD have an increased risk of vascular calcification due to impaired divalent ion metabolism. Hyperphosphataemia or an increased calcium–phosphorus product, or both, are implicated as important factors underlying such calcification [11,21]. It is consistent with these data that in our study serum phosphorus levels were significantly higher in the patients with coronary (Table 3) and aortic calcification (7.34±2.28 vs 5.91±1.80 mg/dl, P = 0.002). It is generally accepted that vascular calcification in uraemic patients is localized predominantly in the tunica media of the artery [22]. However, atherosclerotic plaques tend to be calcified, especially in advanced atherosclerosis [23]. Indeed, in a recent morphological study, Schwarz et al. [24] reported that atherosclerotic plaques in uraemic patients were heavily calcified. Therefore, arterial calcification in patients with ESRD is the sum of intimal (atherosclerotic) and medial calcification; and EBCT cannot differentiate between the two. Accordingly, we also found a positive correlation between carotid atherosclerosis and both CACS and AWCS. Vascular calcification in ESRD patients, wherever its location, leads to aortic stiffness [9], which is important for the development of LVH. In our present study, the patients with coronary calcification had higher LVMI than the patients without (Table 3); and the severity of calcification was significantly related to LVMI (Figure 3). Moreover, in multiple linear regression analysis, CACS was an independent variable for predicting LVMI. These findings implicate vascular calcification as an independent risk factor for LVH in HD patients, regardless of its relationship to atherosclerosis. The association between CACS, measured by EBCT, and LVH has been reported recently in patients with essential hypertension [25]. This same study also noted a relationship with LV geometry. In our study, we did not detect any association between CACS and LV geometry. Uraemia-related factors, such as hypervolaemia and anaemia, may strongly influence LV geometry in HD patients and may blur any possible association between CACS and LV geometry.

In our present study, another independent variable influence on LVMI was pulse pressure. On a haemodynamic basis, pulse pressure, the pulsatile component of blood pressure, represents arterial stiffness and function of the large conduit arteries, whereas mean arterial pressure, a steady component of blood pressure, reflects total vascular resistance [26]. Pulse pressure is a significant predictor of cardiovascular mortality and LVH in the non-uraemic population [27]. Recently, both pulse pressure and aortic stiffness were associated with mortality in ESRD patients [28,29]. However, the relationship of pulse pressure to LVH has not been assessed adequately in HD patients. Studying a small group of patients, Brahimi et al. [30] reported a possible association between pulse pressure and LVMI. In our study, HD patients with LVH had higher pulse pressures than those without LVH. LVMI was correlated with pulse pressure in univariate analysis (Figure 4), and linear regression analysis revealed that pulse pressure was an independent predictor of LVH. These findings are consistent with results reported for patients with essential hypertension. Thus, considering pulse pressure in addition to systolic and diastolic blood pressure could be useful in the treatment of LVH in this group of patients.

Fig. 3.

Correlation between CACS and LVMI in the study cohort (r = 0.34, P = 0.002).

Fig. 3.

Correlation between CACS and LVMI in the study cohort (r = 0.34, P = 0.002).

Fig. 4.

Correlation between pulse pressure and LVMI in the study cohort (r = 0.58, P<0.001).

Fig. 4.

Correlation between pulse pressure and LVMI in the study cohort (r = 0.58, P<0.001).

In conclusion, extra-coronary atherosclerosis and calcification of large vessels are independent predictors of LVH in chronic HD patients (in addition to increased systolic blood pressure and pulse pressure). Our findings suggest that the treatment of LVH in HD patients requires a multilateral approach, and not only blood pressure control. Since treatments of vascular structural disorders are emerging (e.g. regression of atherosclerosis by using statins or better control of disordered mineral metabolism), their inclusion in the armamentarium in current use could have further beneficial effects on the control and regression of LVH in HD patients, adding to the effect of improved control of blood pressure, hypervolaemia and anaemia. Early aggressive treatment of atherosclerosis and vascular calcification may even turn out to be a preventive strategy against LVH in HD patients.

The authors would like to thank Edith Simmons, MD, for her careful reading and review of the manuscript. Special thanks to the administration and staff of TEST Tani Merkezi, Istanbul for performing EBCT examinations. This study was supported by the Turkish Kidney Foundation (grant number 2003-A/132).

Conflict of interest statement. None declared.

References

1
US Renal Data System. 1998 Annual Report: The National Institute of Diabetes and Digestive and Kidney Diseases.
Am J Kidney Dis
 
1998
;
32 [Suppl 1]
:
S81
–S88
2
Zoccali C, Benedetto FA, Mallamaci F et al. Prognostic impact of the indexation of left ventricular mass in patients undergoing dialysis.
J Am Soc Nephrol
 
2001
;
12
:
2768
–2774
3
London GM, Pannier B, Guerin AP et al. Alteration of left ventricular hypertrophy in and survival of patients receiving hemodialysis: follow-up of an interventional study.
J Am Soc Nephrol
 
2001
;
12
:
2759
–2767
4
London GM. Cardiovascular disease in chronic renal failure: pathophysiologic aspects.
Semin Dial
 
2003
;
16
:
85
–94
5
London GM. Alterations of arterial function in end-stage renal disease.
Nephron
 
2000
;
84
:
111
–118
6
London GM, Guerin AP, Marchais SJ et al. Cardiac and arterial interactions in end stage renal disease.
Kidney Int
 
1996
;
50
:
600
–608
7
Yildiz A, Oflaz H, Pusuroglu H et al. Left ventricular hypertrophy and endothelial dysfunction in chronic hemodialysis patients.
Am J Kidney Dis
 
2003
;
41
:
616
–623
8
Roman MJ, Pickering TG, Schwartz JE, Pini R, Devereux RB. Association of carotid atherosclerosis and left ventricular hypertrophy.
J Am Coll Cardiol
 
1995
;
25
:
83
–90
9
Guerin AP, London GM, Marchais SJ, Metivier F. Arterial stiffening and vascular calcifications in end-stage renal disease
Nephrol Dial Transplant
 
2000
;
15
:
1014
–1021
10
Braun J, Oldendorf M, Moshage W, Heidler R, Zeidler E, Luft FC. Electron beam computed tomography in the evaluation of cardiac calcification in chronic dialysis patients.
Am J Kidney Dis
 
1996
;
27
:
394
–401
11
Goodman WG, Goldin J, Kuizon BD et al. Coronary artery calcification in young adults with end-stage renal disease who are undergoing dialysis.
N Engl J Med
 
2000
;
342
:
1478
–1483
12
Friedewald WT, Levy RI, Fredrickson DS. Estimation of concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge.
Clin Chem
 
1972
;
18
:
499
–502
13
Devereux RB, Reichek N. Echocardiographic determination of left ventricular mass in men. Anatomic validation of the method.
Circulation
 
1977
;
55
:
613
–618
14
Du Bois D, Du Bois EF. The measurement of the surface area of man.
Archives of Internal Medicine
 
1915
;
16
:
868
–881
15
Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Detrano R. Quantification of coronary artery calcification using ultrafast computed tomography.
J Am Coll Cardiol
 
1990
;
15
:
827
–832
16
Middleton RJ, Parfrey PS, Foley RN. Left ventricular hypertrophy in the renal patient.
J Am Soc Nephrol
 
2001
;
12
:
1079
–1084
17
Blacher J, Asmr R, Dijane S et al. Aortic pulse velocity as a marker of cardiovascular risk in hypertensive patients.
Hypertension
 
1999
;
33
:
1111
–1117
18
van Popele NM, Grobbee DE, Bots ML et al. Association between arterial stiffness and atherosclerosis: The Rotterdam Study.
Stroke
 
2001
;
32
:
454
–460
19
Amann K, Ritz C, Adamczak M, Ritz E. Why is coronary heart disese of uraemic patients so frequent and so devastating.
Nephrol Dial Transplant
 
2003
;
18
:
631
–640
20
Locatelli F, Bommer J, London GM et al. Cardiovascular disease determinants in chronic renal failure: clinical approach and treatment.
Nephrol Dial Transplant
 
2001
;
16
:
459
–468
21
Kimura K, Saika Y, Otani H et al. Factors associated with calcification of the abdominal aorta in hemodialysis patients.
Kidney Int
 
1999
;
56 [Suppl 71]
:
S238
–S241
22
Ibels LS, Alfrey AC, Huffer WE et al. Arterial calcification and pathology in uremic patients undergoing dialysis.
Am J Med
 
1979
;
66
:
790
–796
23
Stary HC, Chandler AB, Dinsmore RE et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis.
Arterioscler Thromb Vasc Biol
 
1995
;
15
:
1512
–1531
24
Schwarz U, Buzello M, Ritz E et al. Morphology of coronary atherosclerotic lesions in patients with end-stage renal failure.
Nephrol Dial Transplant
 
2000
;
15
:
218
–223
25
Altunkan S, Erdogan N, Altin L, Budoff MJ. Relation of coronary artery calcium to left ventricular mass and geometry in patients with essential hypertension.
Blood Press Monit
 
2003
;
8
:
9
–15
26
Safar ME. Pulse pressure, arterial stiffness, and cardiovascular risk.
Curr Opin Cardiol
 
2000
;
15
:
258
–263
27
Pannier B, Brunel P, El Aroussy W et al. Pulse pressure and echocardiographic findings in essential hypertension.
J Hypertens
 
1989
;
7
:
127
–132
28
Klassen PS, Lowrie EG, Reddan DN et al. Association between pulse pressure and mortality in patients undergoing hemodialysis.
J Am Med Assoc
 
2002
;
287
:
1548
–1555
29
Blacher J, Guerin A, Pannier B et al. Impact of aortic stiffness on survival in end-stage renal disease
Circulation
 
1999
;
99
:
2434
–2439
30
Brahimi M, Dahan M, Dabire H, Levy BI. Impact of pulse pressure on degree of cardiac hypertrophy in patients with uremia.
J Hypertens
 
2000
;
18
:
1645
–1650

Author notes

Division of 1Nephrology and 3Cardiology, Istanbul School of Medicine, Department of Internal Medicine and 2TEST Tani Merkezi, Radiology Center, Istanbul, Turkey

Comments

0 Comments