Effect of physiological levels of atrial natriuretic peptide on hormone secretion: inhibition of angiotensin-induced aldosterone secretion and renin release in normal man

Large doses of atrial natriuretic peptide (ANP) inhibit renin and aldosterone secretion in normal man, but the effect of physiological levels is unknown. We, therefore, studied the effect of a low infusion rate of alpha-human ANP (alpha hANP; 0.5 microgram/min for 180 min) on the plasma corticosteroid response to graded physiological doses of angiotensin II (0.5, 1.0, 2.0, and 4.0 ng/kg X min, each for 30 min) and ACTH (6.25, 12.5, 25, and 50 mIU, each for 30 min) in six normal men eating a low salt diet (10 mmol sodium and 100 mmol potassium daily). The angiotensin II and ACTH infusions were given from 0900-1100 h on separate days, during which randomized infusions of placebo or alpha hANP were given from 0800-1100 h according to a single blind protocol. Plasma immunoreactive ANP levels were less than 10 pmol/L on the placebo day compared to 30-50 pmol/L during the alpha hANP infusions, and were not altered by either ACTH or angiotensin II. Compared with the control observations, there was no significant change in arterial pressure or heart rate during either the alpha hANP or angiotensin II infusions. ACTH infusions evoked an incremental response in plasma aldosterone and cortisol, and the dose-response relationship was unaltered by alpha hANP. In contrast, while an incremental and significant increase in plasma aldosterone in response to angiotensin II occurred with the placebo infusion, no significant increase occurred in response to angiotensin during the alpha hANP infusion. The slope of the angiotensin II/aldosterone regression line was significantly less during all alpha hANP infusions compared to that during the placebo infusion (P less than 0.02). In addition, on the ACTH infusion day significant suppression of both PRA (P less than 0.05) and plasma angiotensin II (P less than 0.008) occurred during the alpha hANP infusion compared to that during the placebo infusion, whereas PRA was equally suppressed by angiotensin II in the presence or absence of alpha hANP. alpha hANP also increased urine volume [176 +/- 31 (+/- SEM) vs. 113 +/- 19 mL/mmol creatinine with placebo; P less than 0.03] and sodium excretion (2.14 +/- 0.48 vs. 0.58 +/- 0.22 mmol/mmol creatinine with placebo; P less than 0.004) on the ACTH infusion days. With angiotensin II, urine volume was also significantly increased by alpha hANP (150 +/- 27 vs. 81 +/- 15 mL/mmol creatinine with placebo; P less than 0.03), and urine sodium excretion doubled.(ABSTRACT TRUNCATED AT 400 WORDS)     

Differential effects of atrial natriuretic factor on angiotensin II- and adrenocorticotropin-stimulated aldosterone secretion

The possible role of atrial natriuretic factor (ANF) in the regulation of adrenal sensitivity to angiotensin II (AII) was investigated in vivo and in vitro by analyzing the characteristics of the inhibitory effect of ANF on aldosterone production stimulated by AII and other stimuli. In isolated adrenal glomerulosa cells, ANF caused a dose-dependent inhibition of basal and stimulated aldosterone production by submaximal concentrations of ACTH, AII, and potassium, with an ED50 of about 1 nM for ANF and complete inhibition with 10 nM ANF. ANF increased the ED50 for ACTH from 14.6 +/- 3.2 to 376 +/- 104 pM with no significant decrease in the maximum aldosterone response. In contrast, ANF inhibited the aldosterone responses to all doses of AII, decreasing maximal aldosterone production by 75%, with a small increase in the ED50 for AII. In conscious rats, ANF infusion (100 ng/min) markedly decreased the plasma aldosterone response to AII infusion (5-10 ng/min). With higher AII doses (50 and 100 ng/min), which increased plasma corticosterone (and presumably ACTH secretion), the inhibitory effect of ANF was less marked. When the rise in ACTH secretion was prevented by dexamethasone treatment, ANF decreased the aldosterone response to 100 ng/min AII by 85%. Similarly, ANF had a minor although significant inhibitory effect on the primary ACTH-mediated increases in plasma aldosterone after stress by immobilization for 15 min. The data demonstrate a prominent inhibitory effect of ANF on AII-stimulated aldosterone production in vivo and in vitro. Since plasma ANF levels are increased during atrial distension, these observations support a regulatory role of ANF in the control of the adrenal sensitivity to AII during alterations of extracellular volume.     

Role of cyclic nucleotides in the inhibition of growth-hormone secretion by somatostatin

The effects of somatostatin on GH secretion, cyclic AMP and cyclic GMp concentrations in dispersed bovine anterior pituitary cells were studied following activation of adenylate cyclase with cholera toxin and inhibition of phosphodiesterase with isobutylmethylxanthine (IBMX). Cholera toxin (10(-5)M) increased intracellular cyclic AMP concentration 10-fold and cyclic GMP concentration 3-fold relative to control, and stimulated the secretion of GH. IBMX (10(-4) M) als increased intracellular concentrations of both cyclic AMP and cyclic GMP and the secretion of GH and potentiated the actions of cholera toxin particularly in raising intracellular cyclic AMP concentrations which were elevated 40-fold in the presence of cholera toxin and IBMX. Somatostatin (5 X 10(-7) M) completely prevented GH secretion elicited by cholera toxin and/or IBMX. Somatostatin was without effect on control cyclic AMP and cyclic GMp concentrations and on the increases in both cyclic AMP and cyclic GMP caused by cholera toxin and by IBMX alone, or in combination. The data suggest that bovine GH secretion is increased when concentrations of either or both cyclic nucleotides are elevated within the cells, although incubation of cells with extracellular concentrations of cyclic AMP and cyclic GMP derivatives up to 2 x 10(-3) M caused only small changes in GH release. We suggest that somatostatin inhibits cholera toxin-induced bovine GH secretion by preventing activation of the secretory process by either cyclic AMP or cyclic GMP.     

Effects of atrial natriuretic factor and vasopressin on cyclic nucleotides in cultured kidney cells

The cellular mechanism of the action of atrial natriuretic factor (ANF) is thought to involve activation of guanylate cyclase. Increasing evidence shows a direct tubular effect of ANF. Part of the ANF-induced diuresis has been suggested to be due to inhibition of the action of arginine vasopressin (AVP) in the cortical collecting tubule. In this study we investigated the effect of ANF on cyclic nucleotide production in primary cultures of cortical collecting tubule cells immunodissected with a monoclonal antibody. ANF caused a dose-dependent stimulation in cyclic guanosine 3',5'-monophosphate (cGMP) production; the half-maximal stimulation was observed at approximately 1 nM of ANF. ANF (0.01-100 nM) had no effect on cyclic adenosine 3',5'-monophosphate (cAMP) accumulation in cortical collecting tubule cultures. AVP caused a dose-dependent increase in cAMP production, and this effect was not altered by the simultaneous addition of ANF (100 nM). Similarly, ANF-induced cGMP stimulation was not influenced by AVP (10 nM). We conclude that 1) ANF has a direct stimulatory action on cGMP production by cultured cortical collecting tubule cells and 2) any interaction between ANF and AVP is likely to occur at steps distal to cyclic nucleotide formation.     

Changes in the levels of atrial natriuretic factor, aldosterone, cyclic nucleotides and beta 2-microglobulin in patients with myocardial infarction after ultrasonic therapy]

A total of 43 patients were examined: 21 patients with large myocardial infarction (MI) and 22 donors. In MI patients without signs of left ventricular failure, the level of atrial natriuretic factor was higher than that in the controls, which resulted in inhibition of aldosterone production; whereas in those with the signs, it was slightly lower than that in healthy individuals. All the patients with MI were found to have elevated blood plasma levels of cyclic nucleotides and beta 2-microglobulin. Ultrasound application to the heart was demonstrated to normalize secretion of atrial natriuretic factor, to increase plasma cyclic nucleotide levels and to produce no effects on beta 2-microglobulin quantities.     

The physiological role of atrial natriuretic hormone in the regulation of aldosterone and salt and water metabolism

To investigate the mechanisms by which small changes in plasma levels of atrial natriuretic hormone (ANH) affect aldosterone, 10 normal young men were infused for 2 h with 0.6 pmol/kg.min human [Ser,Tyr28]ANH under 3 study conditions: 1) high salt diet (H), 2) low salt diet (L), and 3) low salt diet plus pretreatment with the angiotensin-converting enzyme inhibitor enalapril (LE). Baseline ANH levels were higher on H than on L or LE. A postural drop in ANH was observed when subjects went from standing to sitting. Plasma ANH levels increased during infusion by up to 4.5 pmol/L (H, 7.0 +/- 1.3 to 11.5 +/- 1.4; L, 4.3 +/- 0.6 to 8.7 +/- 1.1; LE, 4.2 +/- 0.5 to 8.6 +/- 1.5). At all time points, plasma ANH was well within the normal range. Plasma aldosterone did not change during H, decreased by about 60% for both low salt conditions, and remained suppressed at 1 h of recovery for L, but not for LE. This suggests that ANH can suppress aldosterone by both indirect and direct mechanisms, although the indirect mechanism appears to predominate. A prompt increase in urine flow was seen during ANH infusion and was sustained at 1 h of recovery, but little change was seen in urinary sodium or potassium excretion, heart rate, or blood pressure. The difference between the natriuretic and diuretic effects of ANH was seen under all conditions. These results support the hypothesis that within the normal physiological range, ANH is a regulator of salt and water metabolism in normal man.     

The role of atrial pressure in secreting atrial natriuretic polypeptides

The role of atrial pressure and catecholamines in secreting human atrial natriuretic polypeptides (hANP) were investigated in patients with acute or old myocardial infarction (AMI or OMI). No differences were observed in hANP levels obtained from brachial vein, right atrium, pulmonary artery and femoral artery, suggesting that hANP degradation in the pulmonary circulation has little clinical significance in hANP measurement. Plasma hANP levels in 10 patients with AMI were higher than those in controls and significantly correlated to both mean right atrial pressure (mRA) and pulmonary wedge pressure (PCW) (r = 0.76, p less than 0.05; r = 0.95; p less than 0.01, respectively). In 12 patients with OMI, plasma hANP levels were normal at rest and were significantly increased (p less than 0.01) with bicycle ergometer exercise associated with significant elevations of plasma catecholamine levels, mRA, and PCW. However, the increments of hANP correlated only to those of PCW. These results suggest that the elevation of atrial (principally left atrial) pressure rather than catecholamine stimulates hANP secretion.     

Modulatory role of angiotensin-II in the secretion of atrial natriuretic factor in rabbits

This study was designed to investigate whether the increase in circulating atrial natriuretic factor (ANF) levels produced by angiotensin II (Ang II) is a consequence of the hemodynamic changes or whether it occurs also in the absence of pressor changes. For this purpose in anesthetized and awake rabbits we evaluated the effects of Ang II (0.1 micrograms/kg.min) alone or during the simultaneous infusion of sodium nitroprusside (NP) at a dose titrated to abolish the pressor effects. Systemic blood pressure increased from 76 +/- 4 to 113 +/- 5 mm Hg (P less than 0.001) during Ang II and from 76 +/- 2 to 75 +/- 3 mm Hg (P = NS) during Ang II plus NP. The alpha-adrenergic agonist phenylephrine, used as a control, raised blood pressure from 65 +/- 2 to 101 +/- 8 mm Hg (P less than 0.001), and its pressor effect was abolished by the concomitant infusion of NP (64 +/- 2 to 61 +/- 1 mm Hg; P = NS). The increase in plasma ANF levels produced by Ang II alone (from 36.5 +/- 5 to 237 +/- 57 pg/ml; P less than 0.001) was not different from that observed during Ang II plus NP (from 46 +/- 10 to 207 +/- 88 pg/ml; P less than 0.001). In contrast, the stimulatory effect on ANF release of phenylephrine (from 56.1 +/- 9 to 202 +/- 40 pg/ml; P less than 0.001) was completely abolished when its pressor effects were prevented by the combined infusion of NP (from 58.5 +/- 15 to 42.3 +/- 10 pg/ml; P = NS). These results show that the stimulatory effect of Ang II on ANF release can be clearly dissociated from its pressor effect, whereas the increase in plasma ANF levels caused by phenylephrine is strictly related to its hemodynamic effect. Therefore, Ang II is capable of modulating ANF secretion in a manner that is independent of its pressor actions. In addition, our results suggest that ANF release is not solely linked to myocyte stretch.     

Modulatory role of vasopressin in secretion of atrial natriuretic polypeptide in conscious rats

To investigate whether vasopressin is involved in the secretory mechanism of atrial natriuretic polypeptide (ANP), effects of arginine-vasopressin (AVP) administered iv on plasma ANP levels were studied in conscious, unrestrained rats. The administration of 100 ng and 1 microgram of AVP caused a dose-dependent increase of the plasma ANP level, which was blocked by a V1-receptor antagonist of AVP, and was attenuated by 5 ml blood volume reduction before the stimulation. The injection of less than 10 ng of AVP induced no significant effects on ANP secretion. However, the administration of 5 ng of AVP significantly enhanced ANP secretion induced by intravascular volume expansion with 3 ml saline infusion. These results suggest the possible physiological significance of AVP as a modulator rather than a direct stimulator of ANP secretion from the heart.     

Effect of atrial natriuretic peptide on potassium-stimulated aldosterone secretion: potential relevance to hypoaldosteronism in man.

Atrial natriuretic peptide (ANP) has been shown to suppress aldosterone secretion under certain circumstances, although the physiological significance of this is uncertain. We wondered if ANP would suppress potassium-stimulated aldosterone secretion in man and, if so, whether we might find high circulating levels of ANP in patients with the syndrome of acquired hypoaldosteronism. We studied seven healthy young subjects under two conditions: 1) infusion of KCl (0.5 mmol/kg) over 45 min, and 2) KCl infused with ANP (0.01 microgram/kg.min) for 60 min. We also evaluated ANP levels in eight elderly subjects with the syndrome of acquired hypoaldosteronism, as defined by hyperkalemia (mean serum K+, 5.3 +/- 0.1 mmol/L) associated with inappropriately low aldosterone levels (216 +/- 50 pmol/L). In the normal subjects, ANP almost completely suppressed the aldosterone response to KCl infusion (P less than 0.001, by analysis of variance) despite a similar rise in the serum potassium level with KCl alone (0.70 +/- 0.07 mmol/L) and KCl plus ANP (0.75 +/- 0.09 mmol/L). PRA fell slightly during KCl plus ANP treatment, but did not change during the infusion of KCl alone. ANP levels were approximately 800 pmol/L during the ANP infusion studies. Endogenous ANP levels in the hyperkalemic patients with hypoaldosteronism were markedly elevated at 1186 +/- 340 pmol/L (compared to 93 +/- 10 pmol/L in healthy elderly controls), a level that would be capable of suppressing the potassium-mediated aldosterone response. Exogenous infusion of ANP suppressed the aldosterone response to hyperkalemia, and ANP levels were found to be markedly elevated in a group of patients with hyperkalemia and hypoaldosteronism. We suggest that ANP may contribute to clinically significant hypoaldosteronism and hyperkalemia in the syndrome of acquired hypoaldosteronism.     

Effect of synthetic human atrial natriuretic peptide on aldosterone secretion by dispersed aldosterone-producing adenoma cells in vitro

The effect of synthetic alpha-human atrial natriuretic peptide (alpha hANP), a potent natriuretic and vasorelaxant polypeptide recently isolated from human atria, on aldosterone secretion was studied in vitro in collagenase-dispersed adrenal adenoma cells from a patient with primary aldosteronism. alpha hANP (3.2 X 10(-7) M) significantly inhibited both basal and potassium (16 mM)-stimulated aldosterone secretion, whereas it had little or no effect on aldosterone secretion submaximally or maximally stimulated by ACTH (3.4 X 10(-10)-3.4 X 10(-9) M) or angiotensin II (10(-8)-10(-9) M). The less potent effect of alpha hANP on aldosterone secretion by dispersed human adrenal tumor cells compared to that in in vitro animal studies may reflect decreased affinity and/or number of specific receptors for ANP on the tumor cells. Whether ANP plays a physiological role in regulation of aldosterone secretion in humans in vivo remains to be determined.     

Atrial natriuretic peptide inhibition of calcium ionophore A23187-stimulated aldosterone secretion in rat adrenal glomerulosa cells.

The effect of atrial natriuretic peptide (ANP) on calcium ionophore A23187-stimulated aldosterone secretion was investigated using collagenase-dispersed rat adrenal glomerulosa cell suspensions. A23187 treatment induced a dose-dependent stimulation of aldosterone secretion, exhibiting an EC50 of approximately 75 nM. In agreement with the presumed action of A23187 as a Ca2+ ionophore, stimulation was dependent on the extracellular Ca2+ concentration, being completely inhibited in nominally Ca(2+)-free medium. In such Ca(2+)-free medium, stimulation of aldosterone secretion by bath applied 25-hydroxycholesterol was not inhibited, indicating that cells and biosynthetic pathway enzymes were not inhibited by low extracellular Ca2+ levels. A23187-induced aldosterone secretion was also inhibited by more than 90% when cells were simultaneously treated with ANP. Maximal ANP inhibition of A23187-stimulated aldosterone secretion was not overcome by concentrations of A23187 up to 10 microM or by increasing the extracellular Ca2+ concentration from 1.25 to 5 mM in the presence of A23187 and ANP. Addition of A23187 to ACTH-, angiotensin II-, or K(+)-stimulated glomerulosa cells did not overcome ANP-induced inhibition of aldosterone secretion stimulated by these secretagogues. In contrast to ANP inhibition of Ca(2+)-dependent A23187 stimulation of aldosterone secretion, ANP inhibition of dBcAMP-stimulated aldosterone secretion was readily overcome by increasing the dBcAMP concentration. These results indicated that ANP selectively and noncompetitively inhibited an intracellular step necessary for Ca(2+)-dependent stimulation of the early pathway of aldosterone biosynthesis in rat adrenal glomerulosa cells.     

The role of frequency of atrial contraction versus atrial pressure in atrial natriuretic peptide release

This study was designed to investigate the role of frequency of atrial contraction compared to acute increases in right atrial pressure in the regulation of atrial natriuretic peptide (ANP) release in humans. The studies were performed in patients undergoing electrophysiological study. In group 1 (n = 12) the rate of atrial contraction was increased by continuous rapid right atrial pacing at a rate of 120 beats/min (bpm; group 1A; n = 6) or 176 bpm (group lb; n = 6) for 5 min. No increases in atrial pressure or circulating ANP occurred in response to atrial tachycardia. In contrast, continuous rapid right ventricular pacing (group II: n = 12) at ventricular rates of 120 bpm (group IIa; n = 6) and 150 bpm (group IIb; n = 6) increased both right atrial pressure and circulating ANP. These results demonstrate that, in contrast to studies in vitro, increases in the frequency of atrial contraction in the absence of increases in atrial pressure do not release atrial natriuretic peptide. These studies, therefore, support the conclusion that atrial pressure is the primary physiological stimulus for ANP.     

Mechanisms of inhibition of aldosterone secretion by adrenocorticotropin

The mechanisms by which prolonged administration of ACTH causes a decrease in aldosterone secretion were studied in the rat. After 6 days of treatment with ACTH (2 U/day), blood corticosterone was elevated and plasma aldosterone was decreased in rats maintained on either a normal or low sodium diet. PRA was also decreased, probably secondary to increased sodium and/or fluid retention. In collagenase-dispersed glomerulosa cells from adrenals of ACTH-treated rats, angiotensin II receptors were markedly decreased, as were the in vitro aldosterone responses to angiotensin II, ACTH, 8-bromo-cAMP, and potassium. However, the production of deoxycorticosterone and precursor steroids was increased, indicating the presence of a block in the late aldosterone biosynthetic pathway. Measurement of the activity of biosynthetic enzymes of the steroidogenic pathway in isolated mitochondria revealed an 80% increase in side-chain cleavage enzyme in both glomerulosa and fasciculata mitochondria from ACTH-treated rats. Although ACTH injection also increased 11-hydroxylase activity in the fasciculata zone, this enzyme was reduced by 50% in capsular mitochondria. The 18-hydroxylase activity in adrenal capsular mitochondria was markedly decreased by ACTH treatment in both normal and sodium-restricted animals. The importance of ACTH-induced steroidogenesis in the development of altered glomerulosa cell function was indicated by the ability of aminoglutethimide to prevent the inhibitory effects of ACTH on angiotensin II receptors and PRA. It is likely that the observed inhibition of the renin-angiotensin system is responsible for the decrease in angiotensin II receptors and 18-hydroxylase, since both are highly dependent on the trophic effect of angiotensin II. The specific lesions produced in adrenal glomerulosa cells by long term ACTH treatment include decreased levels of angiotensin II receptors, 11-hydroxylase, and 18-hydroxylase. These changes are secondary to the suppression of renin-angiotensin activity and are responsible for the impaired aldosterone secretion that results from prolonged treatment with ACTH. (Endocrinology 108: 522, 1981)     

Influence of cyclic nucleotides on junctional permeability in atrial muscle

The influence of dB-cAMP (5 X 10(-4) M) and 8-Br-cGMP (5 X 10(-5) M; 5 X 10(-9) M) on the longitudinal diffusion of Lucifer Yellow CH along dog trabeculae was investigated. It was found that dB-cAMP enhances the diffusion coefficient of the dye from 4 +/- 0.63 X 10(-7) cm2/s (control) to 2 +/- 0.53 X 10(-6) cm2/s. Efflux studies showed that the permeability of the surface cell membrane to Lucifer Yellow is negligible ruling out the possibility that the dye moved from cell-to-cell through the extracellular space. The permeability of the nexal membrane (Pnexus = 3 X 4 cm/s) was appreciably enhanced in fibers exposed to dB-cAMP (9.1 X 10(-4) cm/s). No change in the longitudinal redistribution of Lucifer Yellow CH was found with 8-Br-cGMP. The results support the hypothesis that cAMP is a modulator of junctional permeability in the normal heart.     

Studies on the role of calcium and cyclic nucleotides in the control of TSH secretion.

The role of putative mediators in the control of thyrotropin (TSH) secretion has been investigated by monitoring hormone release from isolated anterior cells in response to agents and conditions which modify cyclic nucleotide concentrations or calcium fluxes. The secretory response to 5-min pulses of TRH, raised K⁺ concentration and A-23187 was diminished within 5 min when Ca²⁺-free perifusion was begun 80 sec prior to the pulse. In contrast, the response to theophylline and IBMX was unaffected under these conditions. Both IBMX and dibutyryl cyclic AMP potentiated the effects of TRH and raised K⁺concentration but not that of A-23187. Methoxyverapamil inhibited TSH secretion stimulated by TRH, raised K⁺ concentration and IBMX but not that induced by A-23187. Calcium efflux showed a similar temporal profile to hormone secretion in response to TRH, IBMX and raised K⁺ concentration. It is proposed that one interpretation of these findings is that there is an interrelationship between calcium ions and the cyclic nucleotides in the control of TSH secretion and that cyclic nucleotides may act at the level of the Ca²⁺ channel to modulate Ca²⁺ entry.     

The prognostic role of atrial natriuretic peptides in hemodialysis patients

BACKGROUND: It is well known that plasma atrial natriuretic peptide (ANP) is an indicator of extracellular fluid volume expansion and that plasma ANP is considered to be a marker for setting the proper dry weight of HD patients. Although the plasma ANP is a prognostic predictor of cardiac death, the prognostic role of ANP in HD patients has yet to be elucidated. In this study, we investigated the prognostic role of ANP in HD patients. METHODS: Plasma ANP concentrations were measured in 105 HD patients after HD. Multiple regression analysis was performed to determine the major factors causing increased plasma ANP concentrations. Cardiac mortality was monitored for 24 months after baseline analysis, and the prognostic role of ANP was examined by Cox proportional hazards regression analysis. RESULTS: Multiple regression analysis showed that cardiovascular disease (CD) and age were independent factors for elevated ANP (R2 = 0.298, p < 0.0001). During a 24-month follow-up period, cardiac death occurred in 11 patients. Kaplan- Meier survival estimates of patients from varying plasma ANP levels (<50 and >50 pg/ml) differed between the two groups (p < 0.0001). The group with the higher ANP level (>50 pg/ml) had the lower survival. When compared with patients with ANP <50, the hazard ratios for cardiac death of patients with ANP of >50 pg/ml were 32.0 (95% confidence interval (CI) 4.1 to 252.4). Univariate Cox proportional hazards model showed that ANP, left ventricular ejection fraction (LVEF), LVMI, age, serum albumin and C-reactive protein (CRP) were significantly associated with the risk of cardiac mortality. By stepwise multivariate Cox proportional hazards analysis, only ANP, LVMI and CRP remained powerful independent predictors of cardiac death. The relative risk ratios were 3.483 (95% CI 1.640-7.397) for ln ANP, 1.023 (1.008-1.038) for LVMI, and 1.379 (1.115-1.705) for CRP. CONCLUSION: High plasma ANP level of post-HD were strongly associated with CD and age. Post-HD ANP level may be a reliable parameter for assessing the risk for cardiac death in HD patients by providing prognostic information independent of other variables previously reported.