PARATHORMON, CALCIUM, PHOSPHORUS, AND LEFT VENTRICULAR STRUCTURE AND FUNCTION IN NORMOTENSIVE HEMODIALYSIS PATIENTS

Abstract

Clinical and experimental data suggest that Parathormon (PTH), calcium, and phosphorus participate in left ventricular hypertrophy (LVH) and affect myocardial contractility in end-stage renal disease. Cellular calcium overload and interstitial fibrosis induced by PTH may lead to impairment of left ventricular diastolic function. Hyperphosphatemia is an independent risk of cardiovascular mortality in dialysis patients. The aim of the study was to estimate the influence of PTH and calcium-phosphorus metabolism on left ventricular structure and function in hemodialysis patients, without hypertension and antihypertensive drug therapy (SBP = 126.2 ± 11.1 DBP = 75.8 ± 6.5 mmHg). Echocardiographic findings in a group of 22 normotensive HD patients had been compared to 43 hypertensive HD patients. Relationships between PTH, calcium-phosphorus metabolism and echocardiography in normotensive group were then evaluated. Left ventricular mass index (LVMI) was lower in normotensive patients: 128.3 ± 46.2 versus 165.8 ± 46.7 (p < 0.01). The prevalence of LVH was 55% in normotensive HD patients compared to 86% in hypertensive group (p < 0.01). In normotensive group we found correlation between PTH and LVMI (r = 0.44; p < 0.05). There were also significant relationships between calcium and posterior wall thickness (r = −0.44; p < 0.05), phosphorus and LVMI (r = 0.47; p < 0.05). A significant correlation was observed between both phosphorus, calcium × phosphorus product and E/A ratio: r = −0.47 and r = −0.43, respectively (p < 0.05 both). Disturbances of calcium-phosphorus metabolism and secondary hyperparathyroidism contributes to left ventricular hypertrophy, and impaired left ventricular diastolic function in normotensive hemodialysis patients.

• Parathormon
• Calcium
• Phosphorus
• Blood pressure
• Echocardiography
• Left ventricular hypertrophy
• Left ventricular diastolic function

INTRODUCTION

Cardiovascular diseases are the main cause of death in dialysis patients [[1]]. Left ventricular hypertrophy is an independent risk factor of cardiovascular mortality in these patients [[2]]. About 70% of hemodialysis patients have LVH [3-7]. Many factors such as hypertension, disturbed elasticity of central arteries, anemia, a-v fistula, volume overload, and hormonal factors such as renin or endothelin are known to contribute to LVH in HD patients [3-6], [8-10]. In the last few years many investigators revealed that PTH participates in pathogenesis of LVH in end stage renal disease patients [[6]], [11-13]. In clinical observation, excess PTH impaired left ventricular systolic function [[13]]. Altered diastolic function is very common in ESRD patients [[5]], [14-15]. PTH increases calcium influx into cardiomyocytes [[16]] and activates fibroblasts [[12]]. Calcium overload and interstitial fibrosis may lead to impairment of left ventricular diastolic function [17-18]. The evidence that calcium and phosphorus affect heart and vessels in ESRD patients is still growing [[4]], [[6]], [19-21]. In a recent study, Block et al. found that hyperphosphatemia was an independent risk factor of cardiovascular mortality in dialysis patients [[19]]. Also, high calcium × phosphorus (Ca × P) product is the marker of elevated cardiac mortality risk. In HD patients, hyperphosphatemia is associated with increased blood pressure, hiperkinetic circulation, and increased cardiac work but not LVH [[21]]. Hypertension is one determinant of LVH in HD patients [[3]], [5-6], [22-23]. Antihypertensive drugs may affect left ventricular mass, left ventricular volume, systolic, and diastolic function [24-26]. Therapy with ACEi leads to reduction of LVM independent of blood pressure decrease [[24]].

The aim of the study was to estimate the influence of PTH and calcium-phosphorus metabolism on left ventricular morphology, systolic, and diastolic function in hemodialysis patients irrespective of hypertension and antihypertensive drug therapy.

PATIENTS AND METHODS

We studied 22 HD patients: F = 13, M = 9, ages 46.2 ± 11.1 (range 26–74 years), on HD for 52.0 ± 41.6 (range 7–136 months). The underlying renal diseases were: glomerulonephritis in 12, interstitial nephritis in two, policystic kidney disease in five, and unknown in three patients. The patients were eligible for entry to the study if they stayed without antihypertensive drugs for at least one year and had: systolic blood pressure (SBP) before HD <140 mmHg and diastolic blood pressure (DBP) before HD <90 mmHg. In 10 patients, renal disease was not accompanied by hypertension; another 12 had been hypertensive in the terminal phase of renal disease and became normotensive on hemodialysis. Blood pressure was measured with a mercury sphygnomanometer in a sitting position after 15 minutes rest. SBP and DBP before and after HD from two monthly periods over one year preceding echocardiography (26 HD charts) were averaged. In the same way, weight changes during HD session (UF) were averaged. The study group was selected from the population of 65 HD patients without history of previous myocardial infarction, valvular heart disease, large pericardial effusion, diabetes, or parathyreoidectomy in whom echocardiography had been performed. The remaining 43 patients were the control group (hypertensive group). Thirteen of 22 patients with PTH ≥275 pg/mL were treated with alphacalcidol (1αOHD₃), eight patients were given erythropoetin (EPO). 1αOHD₃ and EPO therapy had started at least 12 months before echocardiographic examination.

Echocardiography with American Society of Echocardiography standard M-mode measurement [[27]] was performed after HD session using Sonos 2000 (Hewlett-Packard) equipment with 2.5/2.0 mHz transducer. Interventricular septum thickness (IVS), left ventricular internal diameter (LVID), and posterior wall thickness (PW) were measured in end-diastole. Left ventricular mass (LVM) was calculated using the formula by Devereux et al.; LVM = 0.8 [1.04 (IVS + LVID + PW)³ − LVID³] + 0.6 g [[28]]. End-diastolic volume (EDV) and ejection fraction (EF) was calculated from 2–D apical four-chamber view with modified Simpson rule [[29]]. LVM and EDV were indexed to body surface area: (LVMI) and (EDVI), respectively. The definition of left ventricular hypertrophy (LVH) was sex related: LVMI > 100 g/m² in women and LVMI > 131 g/m² in men [[4]], [[6]], [[8]]. Diastolic function assessment was done by pulse Doppler echocardiography of transmitral flow from the apical four-chamber view. The peak early mitral flow velocity (E) and peak atrial flow velocity (A) were measured with sample volume positioned at the tips of mitral valve leaflets. E/A ratio was calculated to assess left ventricular diastolic function. Echocardiographic data from three consecutive cycles were averaged.

Blood was sampled for intact PTH (normal range: 12–72 pg/mL), hematocrit (Ht), hemoglobin (Hb), calcium (Ca), phosphorus (P), creatinine, and urea (expressed as BUN). Kt/V (delivered dose of dialysis) was calculated according to NKF-DOQI recommendation [[30]]. Data presented in Table 1 are averaged values of all measurements from six months before echocardiography.

Table 1. Clinical and Biochemical Characteristic of Normotensive and Hypertensive HD Patients

	Normotensive Group (n = 22)	Hypertensive Group (n = 43)	p
Age (years)	46.2 ± 11.1	49.8 ± 12.1	ns
Gender F/M	13/9	22/21	ns
Duration of HD (months)	52.0 ± 41.6	30.7 ± 23.9	<0.05
Body weight after HD (kg)	64.7 ± 10.0	62.4 ± 8.8	ns
BSA (m^2)	1.74 ± 0.22	1.70 ± 0.15	ns
SBD before HD (mmHg)	126.2 ± 11.1	144.5 ± 11.0	<0.001
DBP before HD (mmHg)	75.8 ± 6.5	81.7 ± 5.7	<0.001
SBP after HD (mmHg)	111.6 ± 12.0	133.1 ± 12.0	<0.001
DBP after HD (mmHg)	70.4 ± 7.1	78.3 ± 6.3	<0.001
UF (kg)	2.17 ± 0.57	2.12 ± 0.71	ns
PTH (pg/mL)	525.3 ± 418.5	362.7 ± 477.2	ns
Calcium (mmol/L)	2.26 ± 0.18	2.22 ± 0.21	ns
Phosphorus (mmol/L)	2.13 ± 0.54	2.22 ± 0.39	ns
Calcium × phosphorus product	4.80 ± 1.16	4.89 ± 0.87	ns
Hb (g/dL)	10.7 ± 1.4	9.6 ± 1.2	<0.001
Ht (%)	32.2 ± 4.3	28.9 ± 3.8	<0.01
Albumin (g/dL)	4.30 ± 0.45	4.29 ± 0.37	ns
BUN before HD (mg/dL)	70.0 ± 13.3	72.8 ± 13.5	ns
Creatinine before HD (mg/dL)	10.9 ± 1.9	10.7 ± 1.8	ns
Kt/V	1.11 ± 0.16	1.14 ± 0.18	ns

All data are expressed as mean ± SD. Statistical analysis was performed using Student's t-test. If the variable was not normally distributed, U Mann-Whitney test was used. Qualitative data were compared with the χ²-test. Simple linear regression analysis was performed to correlate variables; p < 0.05 was considered statistically significant.

RESULTS

Table 1 shows clinical characteristics of the normotensive HD patients and its comparison to the hypertensive group. Normotensive patients stay longer on HD therapy than the hypertensive group: 52.0 ± 41.6 versus 30.7 ± 23.9 months (p < 0.05). SBP and DBP before and after HD were significantly higher in hypertensive patients (Tab. 1) despite they had been treated with 1–4 blood pressure medications (mean 1.8 ± 1.3 drug/patient). Anemia was less pronounced in normotensive HD patients than in hypertensive group: Hb 10.7 ± 1.4 versus 9.6 ± 1.2 (p < 0.001), Ht 32.2 ± 4.3 versus 28.9 ± 3.8 (p < 0.01), respectively. There was no significant difference in age, gender, body surface area, PTH, calcium, phosphorus, Ca × P product, albumin, and dialysis adequacy between both groups.

Echocardiography

IVS and PW thickness and also LVMI were significantly lower in normotensive patients (Tab. 2). The prevalence of left ventricular hypertrophy was 55% in normotensive HD patients and 86% in the hypertensive HD population (χ² = 7.78; p < 0.01). Left ventricular internal diameter did not differ between the normotensive and hypertensive group, but EDVI was significantly lower in normotensive patients (Tab. 2). Only one patient in the normotensive group had left ventricular dilatation: EDVI > 71.9 mL/m². Four (18%) patients from the investigated group had impaired left ventricular systolic function (EF < 55%). There were no statistically significant differences in EF and E/A ratio between patients with and without hypertension.

Table 2. Echocardiographic Characteristic of Normotensive and Hypertensive HD Patients

	Normotensive Group (n = 22)	Hypertensive Group (n = 43)	p
IVS (mm)	12.4 ± 2.1	15.3 ± 3.0	<0.001
LVID (mm)	48.0 ± 7.0	47.7 ± 6.7	ns
PW (mm)	11.1 ± 2.0	12.8 ± 2.5	<0.01
EDVI (mL/m^2)	44.8 ± 14.0	52.0 ± 12.4	<0.02
LVMI (g/m^2)	128.3 ± 46.2	165.8 ± 46.7	<0.01
LVH	12 (55%)	37 (86%)	<0.01
EF (%)	64.0 ± 10.0	67.3 ± 6.5	ns
E/A	0.88 ± 0.20	0.93 ± 0.31	ns

Blood Pressure and UF

In the normotensive group we did not find a statistically significant correlation between SBP before HD and echocardiographic data. DBP before HD was correlated with EDVI (r = 0.43; p < 0.05). We found a positive correlation between SBP after HD and PW thickness (r = 46; p < 0.05), EDVI (r = 0.55; p < 0.01), and LVMI (r = 0.53, p < 0.02). DBP after HD were significantly correlated with IVS thickness (r = 0.43; p < 0.05) and EDVI (r = 0.47; p < 0.05). Negative correlation was observed between EF and DPB before HD (r = −0.44), SBP and DBP after HD (r = −0.44 and r = −0.43, respectively; p < 0.05 for all). Negative correlation between E/A ratio and blood pressure did not reach statistical significance. IVS and PW thickness was related to UF (r = 0.45; p < 0.05 and r}=0.54; p < 0.01, respectively). Also negative correlation between EF and UF was found (r = −0.49; p < 0.05).

Anemia

In normotensive HD patients, we found moderate negative correlation between Hb and PW thickness (r = −0.43; p < 0.05), Ht and PW (r = −0.42; NS), Hb and LVMI (r = −0.38; NS) and Ht and LVMI (r = −0.32; NS). Only the first one was statistically significant.

PTH

Significant positive correlation between PTH and LVMI (r = 0.44; p < 0.05) in normotensive patients was revealed (Fig. 1). This association was seen in normotensive HD patients with LVH (n = 12; r = 0.71; p < 0.01) but not in patients with normal LVMI (n = 10; r = 0.13; NS) (Fig. 2). IVS and PW were weakly related to PTH (r = 0.38, r = 0.36, respectively); these relationships did not reach statistical significance (Tab. 3).

Figure 1. Correlation between PTH concentration and left ventricular mass index (LVMI) in 22 normotensive HD patients (r = 0.44; p < 0.05; y = 0.048x + 103).

Figure 2. Correlation between PTH and LVMI in normotensive patients with LVH [•] (r = 0.71; p < 0.01; y = 0.072x + 113.5). No correlation was found between PTH and LVMI in patients without LVH [○] (r = 0.13; NS).

Table 3. Correlation Coefficients between PTH, Calcium and Phosphorus Concentrations and Echocardiographic Data in 22 Normotensive HD Patients

	IVS	LVID	PW	EDVI	LVMI	EF	E/A
PTH	0.38	0.22	0.36	0.27	0.44^a	−0.01	−0.33
Calcium	−0.32	−0.29	−0.44^a	−0.10	−0.39	0.16	0.16
Phosphorus	0.27	0.27	0.28	0.23	0.47^a	−0.03	−0.47^a
Ca × P	0.16	0.15	0.11	0.18	0.34	0.05	−0.43^a
1αOHD3 (n = 13)	−0.31	−0.14	−0.55	−0.05	−0.29	0.09	−0.19

^ap < 0.05; ^bp < 0.02; ^cp < 0.01.

Calcium and Phosphorus

There were significant relationships between calcium and PW thickness (r = −0.44; p < 0.05) (Fig. 3), phosphorus and LVMI (r = 0.47; p < 0.05) (Fig. 4) in normotensive HD patients. Significant relationships were also observed between phosphorus and E/A ratio (r = −0.47; p < 0.05) (Fig. 5) Ca × P product and E/A ratio (r = −0.43; p < 0.05) (Fig. 6).

Figure 3. Correlation between serum calcium concentration and left ventricular posterior wall thickness (PW) in 22 normotensive HD patients (r = −0.44; p < 0.05; y = −0.48x + 2.21).

Figure 4. Correlation between serum phosphorus concentration and left ventricular mass index (LVMI) in 22 normotensive HD patients (r = 0.47; p < 0.05; y = 40.7x + 41.5).

Figure 5. Correlation between serum phosphorus concentration and E/A ratio in 22 normotensive HD patients (r = −0.47; p < 0.05; y = −0.17x + 1.24).

Figure 6. Correlation between calcium-phosphorus product and E/A ratio in 22 normotensive HD patients (r = −0.43; p < 0.05; y = −0.07x × 1.22).

DISCUSSION

Hypertension is one of the major risk factors of LVH in dialysis patients [[3]], [5-6], [[8]], [22-23]. Because the aim of the study was to split the effect of high blood pressure from calcium and phosphorus metabolism on left ventricular structure and function, we enrolled normotensive patients in the study. IVS, PW thickness, and LVMI were lower in normotensive patients than in the hypertensive group. Similar data were published by Facchin et al. [[22]]. The frequency of LVH in normotensive group was high—55%. The frequency of LVH in HD patients is 65–76% [3-7]. LVH was also seen in normotensive HD patients [[22]], [[31]]. We found relations between blood pressure and left ventricular wall thickness and LVMI in normotensive HD patients, with the strongest correlation between SBP after HD and LVMI. These findings raise a question: what should be the target blood pressure in HD patients? We enrolled HD patients in the study with predialysis blood pressure <140/90 mmHg and without antihypertensive drug therapy. We do believe that in some of these HD patients further reduction of blood pressure may be profitable in term reduction of LVM. The target blood pressure, the lowest well tolerated by the patient, has to be achieved slowly under strict clinical control.

Systolic function is normal in the majority of HD patients [4-5], [[21]]. In the present study, four (18%) patients had impaired left ventricular systolic function. Similar data were reported by Foley et al. and Covic et al. [4-5], [[8]]. There were no differences in EF and E/A ratio between normotensive and hypertensive patients. These results are in agreement with previous data [[22]]. We found significant negative correlation between blood pressure and EF in normotensive HD patients. In the Foley et al. study, coronary heart disease was the only risk factor for left ventricular systolic dysfunction [[4]]. On the other hand Hara et al. revealed that PTH >200 pg/mL had been detrimental for left ventricular systolic function [[13]]. We did not find a relationship between PTH, calcium and phosphorus serum concentrations, and EF.

Anemia was found to be the risk factor of LVH in HD patients [[9]]. The reasons why association between Hb and LVMI did not reach significance in our study are probably the small number of patients and relatively high Hb concentration (10.7 ± 1.4 g/dL).

Increasing numbers of clinical and experimental studies suggest that secondary hyperparathyroidism contributes to left ventricular remodeling in dialysis patients [[3]], [[6]], [11-12], [[16]], [[18]]. Secondary hyperparathyroidism was associated with LVH [[3]], [[6]] and LV dilatation [[11]]. Therapy of secondary hyperparathyroidism with vitamin D or parathyreoidectomy leads to reduction of left ventricular hypertrophy and dilatation [[12]], [[14]], [33-34], but other authors did not confirm these relationships [35-36]. PTH exerts direct trophic effect on cardiomyocytes and activates fibroblasts leading to interstitial fibrosis [[12]], [[16]], [[18]]. In the present study PTH correlates with LVMI, especially in patients with LVH. We also found a relationship between hypocalcemia and PW thickness, hyperphosphatemia, and LVMI. Foley et al. showed that hypocalcemia and hyperphosphatemia had been risk factors of left ventricular dilatation but not hypertrophy in dialysis patients [[4]]. Serum calcium was inversely correlated with LVMI in the study of Savage et al. [[6]].

Decreased E/A ratio indicates impairment of left ventricular diastolic function [[14]], [[22]], [[37]]. E/A ratio <1 was observed in 77% of patients in our study. Covic et al. found impaired left ventricular diastolic function in 58% of HD patients [[5]] Myocardial relaxation is an energy-dependent process. Excess of PTH in chronic renal failure increases intracellular calcium content and inhibits myocardial energy production [[16]], [[38]]. It leads to slower reuptake of calcium to sarcoplasmatic reticulum. PTH is a permissive factor in the genesis of interstitial fibrosis [[12]], [[17]], which leads to increasing left ventricular stiffness [[22]].

The evidence that calcium and phosphorus metabolism affects the heart and vessels in ESRD patients is still growing [[4]], [[6]], [19-21]. Hyperphosphatemia is an independent risk factor for death in dialysis patients. Also high Ca × P product is the marker of elevated cardiac mortality risk. Hyperphosphatemia may act directly on the heart and vessels; it promotes calcifications, modifies coronary plaque morphology, and influences heart microvasculature [[20]]. Hyperphosphatemia is associated with increased blood pressure and hyperkinetic circulation [[21]], and high-output states determine left ventricular hypertrophy [[9]], [[39]]. Hyperphosphatemia also leads to excessive secretion of PTH, which influences cardiac morphology and function. [[4]], [[20]]. The associations between calcium and phosphorus and echocardiography may be partly explained by secondary hyperparathyroidism [[4]].

In an experimental model, vitamin D deficiency was associated with increased heart weight [[40]]. Pulse therapy with intravenous calcitriol leads to LVH regression in HD patients with secondary hyperparathyroidism [[34]]. We found a negative borderline correlation between the dose of 1αOHD₃ and PW thickness. In our opinion, the relationship between vitamin D and cardiac morphology and function is a very interesting issue, and needs further investigation.

CONCLUSIONS

Blood pressure seems to be one of the major determinants of left ventricular hypertrophy even in normotensive patients on hemodialysis. Also, disturbances of calcium-phosphorus metabolism and secondary hyperparathyroidism contribute to left ventricular hypertrophy, and impaired left ventricular diastolic function in normotensive hemodialysis patients.