N-terminal fragments of the proatrial natriuretic peptide in patients before and after hemodialysis treatment

Kidney International, Vol. 58 (2000), pp. 374–383

N-terminal fragments of the proatrial natriuretic peptide in patients before and after hemodialysis treatment MARTINA FRANZ, WOLFGANG WOLOSZCZUK, and WALTER H. HO¨RL Division of Nephrology and Dialysis, Department of Internal Medicine, University of Vienna, and Ludwig Boltzmann Institute of Experimental Endocrinology, Vienna, Austria

ease, hemodialysis treatment, dialyzer membrane material, cardiac dysfunction, and hypertension. Therefore, these are not useful markers to accurately estimate volume status in hemodialysis patients.

N-terminal fragments of the proatrial natriuretic peptide in patients before and after hemodialysis treatment. Background. Fragments derived from the prohormone of ␣-human atrial natriuretic peptide (␣-ANP) in patients with cardiac failure are more closely related to the disease state than intact ␣-ANP. Methods. Specific immunoassays have been developed to detect proANP 1-30, proANP 31-67, and proANP 1-98. Plasma concentrations of these fragments were determined in 122 hemodialysis patients with and without cardiac dysfunction, with and without hypertension, as well as with and without dialysisassociated hypotensive episodes either before or after a regularly scheduled hemodialysis session. The effects of different dialyzer membranes were also evaluated. The results of these assays along with other markers of volume regulation such as ␣-ANP and cyclic 3⬘,5⬘ guanosine monophosphate (cGMP) were compared with those of healthy controls. Results. Predialytic and postdialytic plasma concentrations of the proANP fragments were markedly higher in uremic patients than in controls (98-fold for proANP 1-98, 56-fold for proANP 31-67, and 35-fold for proANP 1-30). All proANP fragments, ␣-ANP, and cGMP decreased during hemodialysis. A strong linear correlation was found between predialytic and postdialytic plasma levels. There was no correlation, however, with the amount of fluid removed during hemodialysis. Patients with altered left ventricular hemodynamics displayed significantly higher plasma concentrations of all proANP fragments and ␣-ANP, but not cGMP, than patients with normal cardiac function. Hemodialysis patients with moderate or severe hypertension had higher concentrations of proANP fragments, ␣-ANP, and cGMP than patients with normal blood pressure or patients with only mild hypertension. There was no significant difference in circulating levels of proANP peptides, ␣-ANP, and cGMP between patients with and without frequent dialysisassociated hypotensive episodes. Cellulose-triacetate dialyzers reduced plasma levels of proANP 1-30, proANP 31-67, and proANP 1-98 significantly more than polysulfone dialyzers, but ␣-ANP and cGMP levels were not different. Conclusions. Circulating ␣-ANP and proANP fragments are influenced by a variety of factors such as end-stage renal dis-

Natriuretic peptides constitute a growing family of structurally related but genetically distinct natriuretic peptides that regulate cardiorenal function [1–3]. The circulating hormones are involved in volume homeostasis, blood pressure control, and electrolyte balance. Atrial natriuretic peptide (␣-ANP) is a cardiac hormone stored as a 126-amino acid prohormone (proANP 1-126) within atrial granules [4]. In response to atrial wall tension [5], the prohormone is cleaved by a membrane protease and is released into the circulation as a C-terminal peptide [6], the active ␣-human ANP (proANP 99-126), and a complementary N-terminal peptide, most likely proANP 1-98 [7]. ␣-ANP regulates diuresis and natriuresis through specific renal receptors by activation of the guanylate cyclase/cyclic 3⬘,5⬘ guanosine monophosphate (cGMP) system [8]. It has been found that other fragments derived from the prohormone (N-terminal proANP 1-30, 31-67, and 79-98) are also present in the circulation [9]. It has been further demonstrated that the fragments have biological function based on their ability to increase renal guanylate cyclase activity [10] and to vasodilate the aorta [11]. ␣-Atrial natriuretic peptide is rapidly degraded, primarily by the kidney, but also by other organs, with a plasma half-life of two to four minutes. Pulsatile secretion into the circulation and methodical difficulties caused by instability of the peptide are also limiting factors for a precise determination of ␣-ANP. cGMP is generated when ␣-ANP activates membrane-bound guanylate cyclase. Because cGMP is more stable in plasma than ␣-ANP and because the radioimmunoassay (RIA) for cGMP is somewhat less arduous, it was believed that cGMP would potentially be a better marker for ␣-ANP activity. However, cGMP acts as second messenger for several natriuretic peptides, and thus, specificity is low.

Key words: atrial natriuretic peptide, cyclic GMP, end-stage renal disease, cardiac disease, blood pressure control. Received for publication July 13, 1999 and in revised form January 18, 2000 Accepted for publication February 2, 2000

 2000 by the International Society of Nephrology

374

Franz et al: ProANP peptides during hemodialysis

In 1991, Winters and Vesely published data on plasma concentrations of proANP fragments in patients with chronic renal failure and hemodialysis treatment [12]. They have demonstrated that the C-terminus (that is, active ␣-ANP) decreased during the hemodialysis session, whereas N-terminal fragments (that is, proANP 1-98 and proANP 31-67) increased. However, the authors investigated only 13 patients. After successful kidney transplantation, a normalization of elevated circulating N-terminal and C-terminal ANP prohormone concentrations was reported [13]. No data are available on N-terminal proANP 1-30 or 79-98 in renal failure patients. Furthermore, the decrease of the ANP C-terminus and the increase of the ANP N-terminus observed during hemodialysis in the study of Winters and Vesely are difficult to explain [12]. In nonuremic patients, the significance of circulating proANP fragments has been increasingly investigated during the last years. It has been found that the N-terminal fragment proANP 1-98 significantly correlated with echocardiographic measurements of left ventricular structure and performance, function of aortic and mitral valves. A correlation was also found between proANP 1-98 and mortality in subjects with defined cardiovascular disorders as well as in the total population. Therefore, proANP 1-98 may predict future cardiovascular disorders [14–18]. Successful heart transplantation may normalize the elevated circulating concentrations of proANP 1-98 and proANP 31-67 [19]. Supraventricular and ventricular arrhythmias were also associated with significantly increased levels of proANP 1-98 and proANP 31-67, which decreased after conversion to sinus rhythm [20]. Furthermore, the release of the whole N-terminus (proANP 1-98) and the midportion of the N-terminus (proANP 31-67) is triggered by myocardial necrosis in patients with acute myocardial infarction, but not in patients with unstable angina and ischemia [21]. Natriuretic peptides are elevated in patients with endstage renal disease (ESRD) and decrease during the dialysis session when fluid is removed [22–25]. The accuracy and usefulness of natriuretic peptides as biochemical makers for the estimation of dry body weight in dialysis patients have recently been investigated [25, 26], but have failed to be reliable. Based on these data, a reevaluation of circulating proANP fragments in a large group of patients with ESRD seemed to be indicated. Therefore, the present study was performed to investigate the influence of periodic circulatory volume expansion, as commonly found in patients with chronic renal failure undergoing regular hemodialysis therapy, on plasma concentrations of different proANP fragments, with respect to the presence or absence of cardiac disorders. Furthermore, any potential differences between patients with hypertension or those with dialysis-associated hypotension, as clinical signs for possible volume

375

overload or underhydration, were investigated. Finally, the effects of different dialyzer membranes on the proANP fragments were evaluated. METHODS Patients We studied 122 regular hemodialysis patients, 55 women and 67 men, with end-stage renal failure caused by various underlying diseases. The age of the patients ranged from 21 to 86 years, with a mean (⫾ SD) age of 54 ⫾ 16 years. Patients with altered left ventricular hemodynamics (N ⫽ 31), such as congestive heart failure, cardiomyopathy, moderate to severe valve dysfunction, or atrial fibrillation, were also investigated. The patients were subjected to hemodialysis for 9 to 15 h/week. One hundred and seven patients were treated with high-flux cellulose-triacetate (N210; Nipro, Osaka, Japan) or polysulfone dialyzers (F60 and F60S; Fresenius, Oberursel, Germany). Fifteen patients were dialyzed with low-flux cellulose-triacetate (N120 and N150; Nipro). Blood flow was usually between 250 and 350 mL/min with a constant dialysate flow rate of 500 mL/min. Ultrafiltration was varied according to the actual weight gain of the patients. All patients underwent electrocardiogram and echocardiography. Some patients had additional radionuclide ventriculography or cardiac catheterization to exclude or confirm cardiac dysfunction. Cardiac dysfunction was defined as the presence of congestive heart failure or cardiomyopathy with significantly reduced left ventricular ejection fraction, existence of moderate- or highgrade valve dysfunction, or current atrial fibrillation. Dry body weight was assessed for each patient on the basis of chest x-ray and clinical observations such as peripheral edema, dyspnea, hypertension, or dialysis-associated hypotension and/or cramps. Additional assessment of the vena cava inferior diameter (VCID) by sonography, as described by Cheriex et al [27], was performed in the majority of the patients. Therefore, the patients have been considered to be on an adequate dry body weight. Besides normal blood pressure control (no antihypertensive drugs), hypertension was defined as mild (one drug), moderate (two or three drugs), or severe (four drugs or more). Systolic and diastolic blood pressures were measured before, during, and at the end of hemodialysis treatment. For statistical analysis, mean arterial blood pressure (MAP) was calculated. Dialysis-associated hypotension was defined as a drop of systolic blood pressure below 100 mm Hg, and frequency was calculated over the period of the last four weeks, graduating in no episodes or frequent episodes (three or more). As a control group, 16 healthy volunteers (hospital staff) with a mean ⫾ SD age of 36 ⫾ 10 years (range 23 to 54) were studied. No drugs were taken by these

376

Franz et al: ProANP peptides during hemodialysis

volunteers except hormonal contraceptives in some females. Analytical methods Blood samples were obtained by puncture of blood access after supine rest of at least 15 minutes before the start of a regularly scheduled hemodialysis session. After hemodialysis, blood samples were obtained from the atrial line of the hemodialysis system immediately before discontinuation of the extracorporeal circulation. In 30 patients, additional blood samples were taken during the hemodialysis procedure after one, two, three, and four hours. In controls blood was drawn from an antecubital vein after supine rest of at least 10 minutes. All samples were collected in chilled tubes containing ethylenediaminetetraacetic acid (EDTA) and aprotinin, immediately placed on ice, and centrifuged within 10 minutes. Plasma was separated and stored at –70⬚C until analysis. Plasma concentrations of proANP 1-30 and proANP 31-67 were determined by competitive enzyme immunoassays (Biomedica GmbH, Vienna, Austria) [28]. The detection limits were 2.5 and 10 pmol/L. ProANP 1-98 was measured by a noncompetitive sandwich-type immunoassay using two antibodies specific for proANP 1019 and proANP 85-90 and synthetic proANP 1-98 as standard. The assay has a detection limit of 50 pmol/L (Biomedica GmbH). Plasma ␣-ANP concentration was measured by a specific sandwich-type immunoradiometric assay (Shionogi & Co., LTD, Osaka, Japan) and cGMP by RIA (Immunotech, Luminy, France). Statistical analysis Results are given as means ⫾ SD. Statistical analyses were performed using the Student’s t-test of paired data when comparing differences between consecutive values in the same individuals, differences between groups were examined by unpaired t-tests. A simple correlation analysis was performed by one-way analysis of variance (ANOVA). A P value less than 0.05 was considered the level of significance. RESULTS Plasma concentrations of proANP fragments in hemodialysis patients versus healthy controls In healthy volunteers, the plasma concentrations of proANP 1-30, proANP 31-67, and proANP 1-98 varied between 0.12 and 0.41 nmol/L. In hemodialysis patients, predialytic levels of proANP 1-98 were 98-fold, while proANP 31-67 were 56-fold, and proANP 1-30 were 35fold elevated, as compared with controls (Fig. 1). During the hemodialysis session, the plasma concentrations of proANP 1-98, proANP 31-67, and proANP 1-30 decreased. At the end of hemodialysis, the concentrations

Fig. 1. Plasma concentrations of different N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67, and proANP 1-98) in controls (䊏) as compared with regular hemodialysis (HD) patients, measured before (䊐) and after ( ) hemodialysis treatment. The fragments were determined by RIA as described in the Methods section. The graph shows the mean ⫾ SD (N ⫽ 122). *P ⬍ 0.0001 compared with controls (N ⫽ 16).

of the determined proANP fragments were still significantly higher than in controls (Fig. 1). Plasma concentrations of ␣-ANP and cGMP In patients on maintenance hemodialysis, plasma concentrations of ␣-ANP and cGMP were also markedly higher than in the control group. Before hemodialysis, the plasma level (mean ⫾ SD) of ␣-ANP was 82.5 ⫾ 80 pmol/L and thus 217-fold elevated as compared with controls (0.38 ⫾ 0.27 pmol/L, P ⬍ 0.0001). Predialytic plasma concentration (mean ⫾ SD) of cGMP was 45.5 ⫾ 26 nmol/L and 379-fold elevated as compared with controls (0.12 ⫾ 0.07 nmol/L, P ⬍ 0.0001). ␣-ANP and cGMP decreased significantly during hemodialysis, but the postdialytic value of ␣-ANP (37.4 ⫾ 37.1 pmol/L) and of cGMP (16.2 ⫾ 8.4 nmol/L) were found to be still markedly higher than in normal persons (98-fold for ␣-ANP and 135-fold for cGMP, P ⬍ 0.0001). Plasma concentrations of cGMP, ␣-ANP, and proANP fragments in hemodialysis patients with or without cardiac dysfunction and changes during hemodialysis treatment Table 1 demonstrates that ␣-ANP and proANP fragments were found to be significantly higher in patients with cardiac dysfunction than in patients with normal cardiac function, before as well as after hemodialysis. Plasma levels of cGMP, however, did not differ between the two groups before hemodialysis. The relative changes in the concentrations of cGMP, ␣-ANP, and proANP were less pronounced when cardiac dysfunction existed (relative decrease of cGMP 55.3 vs. 66.5%, ␣-ANP 52.7 vs. 56%, proANP 1-30 7.5 vs. 29.4%, proANP 31-67 12.8

Franz et al: ProANP peptides during hemodialysis Table 1. Plasma concentrations of cyclic GMP, ␣-ANP and N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67 and proANP 1-98) in regular hemodialysis patients with normal cardiac function (CN) as compared to hemodialysis patients with cardiac dysfunction (CD), measured before and after a regular scheduled hemodialysis session

Before hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L After hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L

CN N ⫽ 91

CD N ⫽ 31

Significance

43.4 ⫾ 25.2 51.6 ⫾ 28.3 6.8 ⫾ 3.6 8.9 ⫾ 3.7 14.1 ⫾ 9.1

41.9 ⫾ 16.4 122.2 ⫾ 107.7 13.2 ⫾ 5.6 14.0 ⫾ 3.8 31.9 ⫾ 16.7

NS P ⬍ 0.0001 P ⬍ 0.0001 P ⬍ 0.0001 P ⬍ 0.0001

14.5 ⫾ 7.5 22.9 ⫾ 18 4.8 ⫾ 3.6 7.0 ⫾ 3.7 7.9 ⫾ 7.1

18.7 ⫾ 6.6 57.7 ⫾ 38.7 10.7 ⫾ 5.4 12.2 ⫾ 4.2 22.2 ⫾ 12.8

P ⬍ 0.01 P ⬍ 0.0001 P ⬍ 0.0001 P ⬍ 0.0001 P ⬍ 0.0001

Data are mean ⫾ SD. Abbreviations are: ANP, atrial natriuretic peptide; cyclic GMP, cyclic 3⬘,5⬘ guanosine monophosphate.

vs. 21.3%, proANP 1-98 30.4 vs. 43.9%). Patients with or without normal cardiac function were comparable with respect to MAP (predialytic 98 ⫾ 14 vs. 99 ⫾ 17 mm Hg; postdialytic 94 ⫾ 15 vs. 99 ⫾ 16 mm Hg; P ⫽ NS), volume removal during hemodialysis (2.4 ⫾ 1.0 vs. 2.2 ⫾ 1.0 kg; P ⫽ NS), and dialyzer membrane material. The etiology of altered left ventricular hemodynamics, such as atrial fibrillation, moderate to severe valve dysfunction, and congestive heart failure or cardiomyopathy was not associated with significant differences of circulating ␣-ANP, cGMP, and proANP fragments (Table 2). Correlations between N-terminal proANP fragments before and after hemodialysis treatment The plasma concentrations of proANP 1-30, proANP 31-67, and proANP 1-98 were well correlated to each other, before and after hemodialysis (Table 3). The correlation was highly significant in hemodialysis patients with normal cardiac function. Overall, there was also a positive correlation between predialytic or postdialytic concentrations of N-terminal proANP fragments and the corresponding levels of ␣-ANP [correlation coefficient (r) between 0.52 and 0.62, P ⬍ 0.0001]. Plasma levels of proANP fragments and cGMP were less closely correlated or did not correlate at all (Table 3). Influence of the volume status on plasma concentrations of cGMP, ␣-ANP, and proANP fragments The interdialytic weight gain ranged between 0 and 5.6 kg with a mean value ⫾ SD of 2.4 ⫾ 1.0 kg in patients with normal cardiac function, and between 0.4 and 5.0 kg with a mean value of 2.2 ⫾ 1.0 kg in patients with cardiac dysfunction (P ⫽ NS). In both groups, there was no significant correlation between predialytic plasma

377

levels of cGMP, ␣-ANP, proANP 1-30, proANP 31-67, or proANP 1-98 and interdialytic weight gain or between the respective postdialytic plasma concentrations and volume removal during hemodialysis (r between 0.01 and 0.36, P ⫽ NS). Also, the serial evaluation of the peptides measured in 30 patients after one, two, three, and four hours during the hemodialysis session did not show a correlation between the ultrafiltration rate and changes of the respective plasma concentrations (r between 0.02 and 0.28, P ⫽ NS). For changes in vascular volume, hemoconcentration as indicated by the variation of hematocrit did not significantly change during hemodialysis when comparing predialysis and postdialysis values (34.5 ⫾ 4.6 vs. 35.1 ⫾ 4.4%; P ⫽ NS). Furthermore, the VCID, determined after the hemodialysis procedure, did not correlate either with the respective plasma concentrations of proANP 1-30, proANP 31-67, and proANP 1-98, or with ␣-ANP and cGMP (r between 0.03 and 0.25, P ⫽ NS). Effect of hypertension and intradialytic blood pressure on plasma concentrations of cGMP, ␣-ANP, and proANP fragments In hemodialysis patients with normal cardiac function and normal blood pressure or mild hypertension (group A; N ⫽ 58), MAP before hemodialysis was 94 ⫾ 15 mm Hg and thus significantly lower than in patients with moderate or severe hypertension (group B; N ⫽ 33; MAP 105 ⫾ 8 mm Hg, P ⬍ 0.001). Both groups were comparable with respect to interdialytic weight gain (2.5 ⫾ 0.9 kg vs. 2.8 ⫾ 1.1 kg, P ⫽ NS). At the end of hemodialysis, MAP was comparable in both groups (92 ⫾ 17 vs. 97 ⫾ 12 mm Hg, P ⫽ NS). Table 4 demonstrates that before the hemodialysis procedure, plasma levels of ␣-ANP, cGMP, and proANP fragments were significantly higher in patients with more severe hypertension (group B) as compared with patients with normotension or only mild hypertension (group A). After hemodialysis, proANP 31-67 remained significantly (P ⬍ 0.02) elevated in group B. A trend for higher proANP 1-98 concentrations (P ⬍ 0.08) was also observed, but the levels of ␣-ANP, cGMP, and proANP 1-30 were not different between groups A and B (Table 4). Patients who never presented with dialysis-associated hypotension (group I, N ⫽ 46) did not significantly differ from those patients with frequent episodes (group II, N ⫽ 16) with respect to their plasma concentrations of the determined peptides and propeptide fragments, before as well as after hemodialysis (Table 5). In group I, MAP was significantly higher than in group II, before (102 ⫾ 9 vs. 91 ⫾ 17 mm Hg, P ⬍ 0.01) as well as after hemodialysis treatment (99 ⫾ 11 vs. 82 ⫾ 18 mm Hg, P ⬍ 0.001). Out of the 30 patients who underwent serial evaluation

378

Franz et al: ProANP peptides during hemodialysis

Table 2. Plasma concentrations of cyclic GMP, ␣-ANP and N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67 and proANP 1-98) in regular hemodialysis patients with cardiac dysfunction

Before hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L After hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L

Atrial fibrillation N⫽8

Valve dysfunction N⫽7

Congestive heart failure/ cardiomyopathy N ⫽ 16

Significance

40.0 ⫾ 14.1 114.3 ⫾ 53.2 13.1 ⫾ 6.5 14.3 ⫾ 4.0 29.5 ⫾ 15.2

44.1 ⫾ 26.9 120.7 ⫾ 60.0 11.9 ⫾ 5.2 13.2 ⫾ 3.0 34.1 ⫾ 17.1

41.5 ⫾ 17.4 131.6 ⫾ 135.1 14.6 ⫾ 6.3 14.5 ⫾ 3.4 32.1 ⫾ 18.3

NS NS NS NS NS

19.2 ⫾ 9.7 54.4 ⫾ 30.6 10.6 ⫾ 4.7 12.2 ⫾ 4.1 19.8 ⫾ 12.7

18.3 ⫾ 14.6 53.3 ⫾ 33.8 10.7 ⫾ 2.2 12.3 ⫾ 2.4 22.1 ⫾ 10.4

18.6 ⫾ 5.4 65.4 ⫾ 41.6 10.8 ⫾ 6.4 12.1 ⫾ 4.6 24.7 ⫾ 13.3

NS NS NS NS NS

Data are mean ⫾ SD. Comparisons are between patients with atrial fibrillation, valve dysfunction, and congestive heart failure or cardiomyopathy, measured before and after a regularly scheduled hemodialysis session. NS is not significant.

Table 3. Correlation coefficient (r) of predialytic and postdialytic plasma concentrations of N-terminal proANP fragments and cyclic GMP in regular hemodialysis patients with normal cardiac function (CN) as compared to hemodialysis patients with cardiac dysfunction (CD)

Predialytic ProANP 1-30 vs. proANP 31-67 ProANP 1-30 vs. proANP 1-98 ProANP 1-30 vs. cyclic GMP ProANP 31-67 vs. proANP 1-98 ProANP 31-67 vs. cyclic GMP ProANP 1-98 vs. cyclic GMP Postdialytic ProANP 1-30 vs. proANP 31-67 ProANP 1-30 vs. proANP 1-98 ProANP 1-30 vs. cyclic GMP ProANP 31-67 vs. proANP 1-98 ProANP 31-67 vs. cyclic GMP ProANP 1-98 vs. cyclic GMP

CN

Significance

CD

r ⫽ 0.69 r ⫽ 0.70 r ⫽ 0.29 r ⫽ 0.68 r ⫽ 0.25 r ⫽ 0.01

P ⬍ 0.0001 P ⬍ 0.0001 P ⬍ 0.01 P ⬍ 0.0001 P ⬍ 0.02 NS

r ⫽ 0.61 r ⫽ 0.52 r ⫽ 0.31 r ⫽ 0.47 r ⫽ 0.07 r ⫽ 0.26

P ⬍ 0.01 P ⬍ 0.01 NS P ⬍ 0.02 NS NS

r ⫽ 0.68 r ⫽ 0.54 r ⫽ 0.21 r ⫽ 0.68 r ⫽ 0.16 r ⫽ 0.24

P ⬍ 0.0001 P ⬍ 0.0001 NS P ⬍ 0.0001 NS NS

r ⫽ 0.66 r ⫽ 0.55 r ⫽ 0.27 r ⫽ 0.72 r ⫽ 0.20 r ⫽ 0.42

P ⬍ 0.001 P ⬍ 0.01 NS P ⬍ 0.0001 NS P ⬍ 0.05

of proANP fragments, 6 patients had persistent intradialytic hypertension, and another 6 patients actually displayed a hypotensive episode. The baseline plasma concentration of proANP 1-30 was significantly higher in the hypertensive patients (11.1 ⫾ 1.6 vs. 5.2 ⫾ 2.2, mean ⫾ SD; P ⬍ 0.04), whereas their plasma levels of proANP 31-67 and proANP 1-98 were comparable to the other six patients with normal or low blood pressure (proANP 31-67, 12.9 ⫾ 1.1 vs. 12.1 ⫾ 6.3; proANP 1-98, 21.9 ⫾ 3.5 vs. 20.0 ⫾ 3.1; P ⫽ NS). After one and two hours of hemodialysis, there was no significant difference of the proANP peptides between the two groups. After three hours and when the hypotensive episode occurred in the six patients, their respective plasma concentration of proANP 31-67 was significantly lower compared with those patients with persistent hypertension (6.2 ⫾ 2.5 vs. 11.4 ⫾ 5.1 mol/L, P ⬍ 0.05). No such difference was found for circulating proANP 1-30 (2.5 ⫾ 1.2 vs. 4.0 ⫾ 1.6, P ⫽ NS) and proANP 1-98 (18.0 ⫾ 3.1 vs. 21.5 ⫾ 1.9, P ⫽ NS). After four hours of hemodialysis treatment,

Significance

there was again no significant difference in the plasma concentrations of proANP peptides between both groups. Effect of different dialyzer membranes on plasma concentrations of cGMP, ␣-ANP, and proANP fragments The majority of patients with normal cardiac function were treated with high-flux membrane dialyzers (N ⫽ 80). The decrease of plasma concentrations of proANP 1-30, proANP 31-67, and proANP 1-98 during the hemodialysis procedure was significantly higher with the use of cellulose-triacetate dialyzers than with polysulfone dialyzers (Table 6). No differences were found for plasma concentrations of ␣-ANP or cGMP. DISCUSSION In the present study, we investigated circulating N-terminal proANP peptides in 122 patients with ESRD who were undergoing regular hemodialysis treatment. Anti-

379

Franz et al: ProANP peptides during hemodialysis Table 4. Plasma concentrations of cyclic GMP, ␣-ANP and N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67 and proANP 1-98) in regular hemodialysis patients with normal cardiac function

Table 6. Plasma concentrations of cyclic GMP, ␣-ANP and N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67 and proANP 1-98) in regular hemodialysis patients with normal cardiac function

Group A N ⫽ 58

PS N ⫽ 51

CT N ⫽ 29

Significance

45.2 ⫾ 27.4 51.6 ⫾ 29.3 6.5 ⫾ 3.6 9.1 ⫾ 4.0 14.1 ⫾ 9.2

44.2 ⫾ 27.9 47.4 ⫾ 29.0 6.2 ⫾ 4.2 8.1 ⫾ 3.4 13.6 ⫾ 12.7

NS NS NS NS NS

16.0 ⫾ 9.9 23.1 ⫾ 19.8 5.1 ⫾ 3.4 7.2 ⫾ 3.3 9.3 ⫾ 8.0

14.2 ⫾ 6.0 24.7 ⫾ 22.3 2.7 ⫾ 1.6 4.4 ⫾ 2.2 4.1 ⫾ 3.1

NS NS P ⬍ 0.014 P ⬍ 0.004 P ⬍ 0.014

Before hemodialysis cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L After hemodialysis cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L

Group B N ⫽ 33

Significance

37.5 ⫾ 23.8 45.5 ⫾ 29.3 5.8 ⫾ 3.2 7.8 ⫾ 2.9 11.8 ⫾ 6.6

53.2 ⫾ 24.4 60.3 ⫾ 24.5 7.7 ⫾ 4.2 10.2 ⫾ 4.1 16.6 ⫾ 11.5

P P P P P

⬍ ⬍ ⬍ ⬍ ⬍

13.8 ⫾ 7.9 21.9 ⫾ 19.6 4.1 ⫾ 3.4 6.1 ⫾ 3.2 6.6 ⫾ 6.3

15.3 ⫾ 6.9 24.2 ⫾ 15.8 5.2 ⫾ 3.3 8.0 ⫾ 3.8 9.4 ⫾ 8.0

NS NS NS P ⬍ 0.02 NS

0.007 0.025 0.035 0.004 0.02

Data are mean ⫾ SD. Comparisons are between normotensive or mild hypertensive patients (Group A) and hemodialysis patients with moderate or severe hypertension (Group B) before and after a regularly scheduled hemodialysis session.

Table 5. Plasma concentrations of cyclic GMP, ␣-ANP and N-terminal fragments of the proatrial natriuretic peptide (proANP 1-30, proANP 31-67 and proANP 1-98) in regular hemodialysis patients with normal cardiac function

Before hemodialysis cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L After hemodialysis cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L

Group I N ⫽ 46

Group II N ⫽ 16

Significance

48.2 ⫾ 28.2 55.8 ⫾ 29.6 7.6 ⫾ 4.1 9.6 ⫾ 4.2 16.0 ⫾ 11.2

44.4 ⫾ 19.1 51.6 ⫾ 22.2 5.8 ⫾ 2.5 8.4 ⫾ 2.8 12.4 ⫾ 6.8

NS NS NS NS NS

15.3 ⫾ 7.1 25.5 ⫾ 19.0 4.7 ⫾ 3.2 7.5 ⫾ 4.1 9.6 ⫾ 9.2

17.0 ⫾ 11.2 24.2 ⫾ 23.2 4.3 ⫾ 2.9 6.2 ⫾ 2.9 5.6 ⫾ 4.6

NS NS NS NS NS

Data are mean ⫾ SD. Comparisons are between patients without episodes of dialysis-associated hypotension (Group I) and hemodialysis patients with frequent episodes (Group II), measured before and after a regular scheduled hemodialysis session.

sera have been developed that recognize specific regions within residues of the amino acid sequence 1-98. These antisera possess negligible cross-reactivity with the C-terminal portion of the prohormone [28]. These antisera were then used to measure the amounts of proANP 1-30, proANP 31-67, and proANP 1-98 in plasma. The results of the study demonstrate the following: (1) N-terminal proANP fragments (proANP 1-30, proANP 31-67, and proANP 1-98) are highly elevated in hemodialysis patients as compared with healthy controls. (2) The relative decrease during hemodialysis is less pronounced for proANP fragments than for ␣-ANP and cGMP. (3) Uremic patients with cardiac dysfunction display significantly higher plasma concentrations of proANP fragments than

Before hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L After hemodialysis Cyclic GMP nmol/L ␣-ANP pmol/L ProANP 1-30 nmol/L ProANP 31-67 nmol/L ProANP 1-98 nmol/L

Data are mean ⫾ SD. Comparisons are between high-flux polysulfone dialyzers (PS) and high-flux cellulose triacetate dialyzers (CT) before and after a regular scheduled hemodialysis session.

those hemodialysis patients with normal cardiac function. (4) Hypertension, but not dialysis-induced hypotension, is associated with significant differences of circulating proANP fragments. (5) Hemodialysis treatment using cellulose-triacetate dialyzers reduced plasma levels of proANP 1-30, proANP 31-67, and proANP 1-98 more than hemodialysis therapy using polysulfone dialyzers. In uremic patients, plasma concentrations of the Cterminus of the prohormone (that is, ␣-ANP) are determined by the level of the intravascular filling volume and the resulting atrial distension. Elevated plasma concentrations in patients with chronic renal failure could also be explained by the disappearance of the hormone’s clearance by the kidney, either by excretion or by metabolism. On the other hand, elimination of hormones across dialyzers may contribute to changes in plasma concentrations and symptoms occurring during hemodialysis treatment. The contribution of the kidney to the whole body metabolic clearance rate for ␣-ANP is small (approximately 14%) so that diminished excretory function of the kidney alone cannot explain the significant rise of plasma levels. Since only a small amount of ␣-ANP was detected in dialysis fluid [23], the elimination by diffusion across the dialyzer membrane is negligible. It is postulated that the increased circulating volume in chronic renal failure stimulates ␣-ANP secretion and that the decrease during hemodialysis treatment is directly correlated to fluid removal and to the loss in body weight. However, we could not find a direct correlation between the amount of fluid removed during the hemodialysis session and ␣-ANP plasma concentrations. Periodic secretion as well as instability of the hormone may influence the analysis and thus the reproducibility. Since other peptides of the prohormone, the whole N-terminus (proANP 1-98), the midportion of the N-ter-

380

Franz et al: ProANP peptides during hemodialysis

minus (proANP 31-67) that is proteolytically cleaved from the N-terminus, as well as proANP 1-30 and proANP 79-98 have been found in the circulation of healthy subjects and in human pathologies, their role in dialysis patients has been investigated. The better stability of N-terminal fragments of proANP than the C-terminus (for example, ␣-ANP) in the plasma should be associated with more reliable test results. Our data showed that plasma levels of proANP 1-30, proANP 31-67, and proANP 1-98 were markedly higher in patients undergoing regular hemodialysis treatment than in healthy controls (Fig. 1). However, their relative increase was less pronounced than for ␣-ANP and cGMP. Overall plasma concentrations of proANP 1-30, proANP 31-67, and proANP 1-98 decreased during hemodialysis (Fig. 1). It is unknown whether the kidney clears proANP fragments. It may be that in the uremic state the kidney is responsible for a higher percentage of metabolic clearance of proANP fragments. It may also be likely that nonrenal clearance of these petides decreases in uremia and therefore contributes to the increased levels. So far, it has been demonstrated that N-terminal proANP peptides have biological functions, but the effects of these peptides are not well characterized in pathophysiological situations. A recent study performed by Villarreal et al indicates that proANP 31-67 has important hemodynamic and renal effects in hypertensive rats due to the hypotensive and natriuretic actions of the peptide [29]. It was concluded that circulating levels of proANP 31-67 may represent one adaptive mechanism, which is independent of ␣-ANP and is involved in the regulation of systemic hemodynamics and renal function in hypertension [29]. Further investigations, particularly in hypertensive humans, are necessary to give answers to this question. Winters and Vesely obtained higher plasma levels of ␣-ANP and proANP 31-67, whereas lower proANP 1-98 concentrations were found as compared with our data. ProANP 1-30 has not been investigated by this group. The relative decrease of ␣-ANP in the study of Winters and Vesely during hemodialysis is comparable to our results [12]. Those authors, however, reported a significant rise of circulating proANP 1-98 (P ⬍ 0.002) and proANP 31-67 (P ⬍ 0.003) during the hemodialysis session. This effect was explained by the respective abilities of propeptides to cross the dialyzer membrane. It was found that only 1.5% of proANP 1-98 and proANP 31-67 was cleared by the dialyzer [12]. We could not confirm these findings, however. The differences may result from the relative small group without information on cardiac function of the patients studied by Winters and Vesely and by differences in dialyzer membranes used [12]. We have studied the effects of different dialyzer membranes on circulating proANP fragments in our patient

population with normal cardiac function, and have found that high-flux cellulose triacetate dialyzers reduced all proANP peptides significantly more than high-flux polysulfone dialyzers (Table 6). According to their molecular weight and size, respectively, elimination across the dialyzer membrane is unlikely. We suggest that differences in membrane adsorption for proANP fragments may be a possible explanation for the observed variations of proANP peptides during hemodialysis treatment. Lowflux dialyzer membranes were used in only some patients. The heterogeneity of membrane material as well as important differences in patient characteristics did not justify a comparison of low-flux with high-flux membrane dialyzers. Similar results to the study of Winters and Vesely were obtained in some of the patients studied by our group [12]. Although the peptide levels changed during hemodialysis in the vast majority of patients, in several patients there was no decrease in peptide fragments during dialysis. In patients with normal cardiac function, proANP 1-30 did not decline in 25%, proANP 31-67 in 22%, and proANP 1-98 in 10% of the patients investigated. We have investigated various factors that might explain this phenomenon, for example, ultrafiltration rate, type of dialyzer membrane, blood pressure control, or medical intervention during hemodialysis. However, we cannot give reasons for this observation. It is also of interest that in the majority of these patients the plasma concentration of only one proANP fragment did not decline, whereas the other proANP peptides did. Our study demonstrates that predialytic and postdialytic plasma levels of proANP fragments correlate to each other (Table 3). These correlations were also significant between plasma propeptides and ␣-ANP, but proANP fragments and cGMP were less closely correlated or did not correlate at all. Concentrations of circulating proANP fragments, ␣-ANP, and cGMP, however, were not correlated to the extent of interdialytic weight gain or fluid removal during hemodialysis, even in patients with normal cardiac function. A precise serial evaluation of ␣-ANP, cGMP, and proANP peptides together with calculations of the individual ultrafiltration rate in 30 patients throughout the hemodialysis session confirmed our data. Also, the VCID indicating the hydration state of patients did not correlate with the respective plasma levels of the peptides. Therefore, factors other than volume overload alone may contribute to the increase of natriuretic peptides and propeptides. Recent investigations on proANP fragments in patients with altered left ventricular hemodynamics have documented elevated concentrations of circulating propeptides due to increased synthesis and secretion. Therefore, we have compared hemodialysis patients displaying normal cardiac function with those patients suffering from cardiac dysfunction. There was a significant differ-

Franz et al: ProANP peptides during hemodialysis

ence in predialytic plasma levels of ␣-ANP and proANP fragments between these two groups of patients (Table 1). Plasma levels were much higher in both groups, either before or after hemodialysis, as compared with nondialyzed patients with congestive heart failure. Azizi et al reported mean ␣-ANP plasma levels between 22 and 60 pmol/L and proANP 1-30 concentrations between 0.5 and 1.9 nmol/L according to the NYHA functional classification (class I to class III) [16]. In another study by Lerman et al, mean plasma proANP 1-98 concentration in patients with symptomless left-ventricular dysfunction was 0.24 nmol/L, and hence much lower than in our hemodialysis patients [14]. Thus, it is demonstrated that ESRD may have a stronger impact on elevated circulating ␣-ANP and proANP fragments than cardiac dysfunction, and may be explained by either enhanced accumulation, increased production, or both. The hemodialysis procedure lowered plasma concentrations of cGMP, ␣-ANP, and proANP peptides less in patients with cardiac dysfunction than without cardiac dysfunction. The difference was more pronounced for cGMP and proANP fragments than for ␣-ANP, and was highest for proANP 1-30. Both groups of patients were comparable with respect to fluid removal and intradialytic reduction of body weight, to dialyzer membrane material, and in their blood pressure control, respectively. Baker et al could demonstrate that N-terminal ANP fragments remained elevated at 60-minutes postexercise, reflecting the longer half-life of the N-terminus in the circulation [30]. Thus, we suppose that the longer half-life of proANP peptides may contribute to the differences in the relative decrease during hemodialysis. The smaller decrease of all proANP fragments in hemodialysis patients with cardiac dysfunction, however, attributes rather to cardiac factors. Particularly, mean values of proANP 1-30 decreased during hemodialysis by 7.5% only when cardiac hemodynamics are altered. Etiology of cardiac dysfunction, however, such as congestive heart failure, cardiomyopathy, valve dysfunction, or atrial fibrillation, had no influence on the plasma levels of the respective peptides (Table 2). We speculate that plasma levels of proANP 1-30 might be predicative for cardiovascular disorders and may probably reflect cardiac morbidity and mortality as a prognostic factor. In hemodialysis patients with moderate or severe hypertension but yet normal cardiac function (discussed later in this article), proANP 31-67 remained elevated compared with normotensive or only mild hypertensive patients (Table 4). Increased cardiac stress caused by hypertension might be a possible explanation. Hypertension is frequently associated with elevated proANP and ␣-ANP concentrations. McMurray and Vesely demonstrated that circulating concentrations of proANP 1-30, proANP 31-67, proANP 79-98, and ␣-ANP increase in patients with high blood pressure in an appar-

381

ent attempt to overcome the constriction of blood vessels [31]. Baker et al showed a strong positive correlation of N-terminal ANP fragments with blood pressure and heart rate [30]. Also, in pregnant women with preeclampsia [32] or in patients with pheochromocytomas [33], proANP 1-98 and proANP 31-67 circulate in higher concentrations. Our data are in agreement with the reported results. We found that severity of hypertension, defined according to the prescribed number of antihypertensive drugs, was linked to higher plasma levels of proANP fragments, ␣-ANP, and cGMP (Table 4). The difference, however, was statistically significant for predialytic values only, as long as blood pressure was significantly higher in the hypertensive group. After hemodialysis, MAP was comparable between the two groups. Significant differences were found only for proANP 3167. Thus, it is likely that the actual blood pressure caused by fluid overload stimulates production of ANP peptides and propeptides rather than the overall long-term effect of hypertension. In contrast, we found no significant differences between patients with frequent dialysis-induced hypotensive episodes and those without (Table 5). One would rather expect higher or lower plasma levels in these patients than in patients without such events. Increased plasma concentrations would explain the excessive vasodilatory effect, whereas low concentrations would be indicative for reduced stimulation caused by hypotension. In addition, we have compared the plasma levels of proANP fragments between patients who displayed severe and persistent intradialytic hypertension and patients who actually had a hypotensive episode during the hemodialysis session. Although only six patients of each group were studied, data on serial measurements of the circulating peptides, and their changes during hemodialysis exist. Persistent intradialytic hypertension was not associated with significantly different variations of proANP fragments except baseline values obtained before hemodialysis when compared with hypotensive patients. The higher decrease of proANP 31-67 observed in association with a dialysis-associated hypotensive episode may reflect its possible role in the regulation of systemic hemodynamics, as recently demonstrated in a study by Villarreal et al in the hypertensive rat [29]. However, this hypothesis must be handled with caution because of the small number of patients investigated. In summary, the results of our study demonstrate that plasma concentrations of N-terminal proANP fragments (proANP 1-30, proANP 31-67, and proANP 1-98) are markedly elevated in patients undergoing regular hemodialysis compared with healthy controls. The overall decrease of the proANP fragments during hemodialysis was more pronounced for proANP 1-98 than for proANP 1-30 and proANP 31-67. However, the extent of fluid removal and intradialytic body weight reduction did not

382

Franz et al: ProANP peptides during hemodialysis

significantly correlate with plasma concentrations of proANP fragments. In patients with cardiac dysfunction, the relative decrease of N-terminal proANP fragments during hemodialysis was markedly less compared with patients with normal cardiac function, particularly for proANP 1-30. Circulating proANP fragments were significantly higher in patients with hypertension, but no difference was found between patients with frequent dialysis-induced hypotensive episodes and those patients without such events. The choice of the dialyzer membrane, however, may have a highly significant effect on the reduction of proANP 1-30, proANP 31-67, and proANP 1-98, respectively. Plasma concentrations of N-terminal proANP fragments may partly reflect volume status in hemodialysis patients, but there is evidence that other factors, such as cardiac function and blood pressure control, also determine the synthesis and release of proANP peptides into the circulation. Therefore, the determination of proANP fragments as biochemical parameters for an accurate estimation of dry body weight in patients undergoing regular hemodialysis is not reliable. The same holds true for cGMP and ␣-ANP. However, we hypothesize that periodic calculations of certain circulating proANP peptides, such as proANP 1-30 and proANP 31-67, might be useful for early recognition of cardiovascular disorders and have the potential to be a prognostic factor for intradialytic morbidity and mortality. Reprint requests to Martina Franz, M.D., Division of Nephrology and Dialysis, Department of Internal Medicine, University of Vienna, Wa¨hringer Gu¨rtel 18-20, A-1090 Vienna, Austria. E-mail: [email protected]

REFERENCES 1. Maack T: Role of atrial natriuretic factor in volume control. Kidney Int 49:1732–1737, 1996 2. Nakao K, Itoh H, Saito Y, Mukoyama M, Ogawa Y: The natriuretic peptide family. Curr Opin Nephrol Hypertens 5:4–11, 1996 3. Ogawa Y, Itoh H, Nakao K: Molecular biology and biochemistry of natriuretic peptide family. Clin Exp Pharmacol Physiol 22:49–53, 1995 4. Maki M, Takayanagi R, Misono KS, Pandey KN, Tibbetts C, Inagami T: Structure of rat atrial natriuretic factor precursor deduced from cDNA sequence. Nature 309:722–724, 1984 5. Dietz JR, Nazian SJ, Vesely DL: Release of ANF, proANF 1-98, and proANF 31-67 from isolated rat atria by atrial distension. Am J Physiol 260:H1774–H1778, 1991 6. Currie MG, Geller DM, Cole BR, Siegel NR, Fok KF, Adams SP, Eubanks SR, Galluppi GR, Needleman P: Purification and sequence analysis of bioactive atrial peptides (atriopeptins). Science 223:67–69, 1984 7. Thibault G, Murthy KK, Gutkowska J, Seidah NG, Lazure C, Chretien M, Cantin M: NH2-terminal fragment of rat pro-atrial natriuretic factor in the circulation: Identification, radioimmunoassay and half-life. Peptides 9:47–53, 1988 8. De Zeeuw D, Janssen WM, De Jong PE: Atrial natriuretic factor: Its (patho)physiological significance in humans. Kidney Int 41: 1115–1133, 1992 9. Winters CJ, Sallman AL, Meadows J, Rico DM, Vesely DL: Two new hormones: Prohormone atrial natriuretic peptides 1-30 and 31-67 circulate in man. Biochem Biophys Res Commun 150:231–236, 1988

10. Vesely DL, Bayliss JM, Sallman AL: Human prepro atrial natriuretic factor 26-55, 56-92, and 104-123 increase renal guanylate cyclase activity. Biochem Biophys Res Commun 143:186–193, 1987 11. Vesely DL, Norris JS, Walters JM, Jespersen RR, Baeyens DA: Atrial natriuretic prohormone peptides 1-30, 31-67, and 79-98 vasodilate the aorta. Biochem Biophys Res Commun 148:1540–1548, 1987 12. Winters CJ, Vesely DL: Change in plasma immunoreactive N-terminus, C-terminus, and 4,000-dalton midportion of atrial natriuretic factor prohormone with hemodialysis. Nephron 58:17–22, 1991 13. Pevahouse JB, Flanigan WJ, Winters CJ, Vesely DL: Normalization of elevated circulating N-terminal and C-terminal atrial natriuretic factor prohormone concentrations by renal transplantation. Transplantation 53:1375–1377, 1992 14. Lerman A, Gibbons RJ, Rodeheffer RJ, Bailey KR, McKinley LJ, Heublein DM, Burnett JC Jr: Circulating N-terminal atrial natriuretic peptide as a marker for symptomless left-ventricular dysfunction. Lancet 341:1105–1109, 1993 15. Davidson NC, Naas AA, Hanson JK, Kennedy NS, Coutie WJ, Struthers AD: Comparison of atrial natriuretic peptide B-type natriuretic peptide, and N-terminal proatrial natriuretic peptide as indicators of left ventricular systolic dysfunction. Am J Cardiol 77:828–831, 1996 16. Azizi C, Maistre G, Kalotka H, Isnard R, Barthelemy C, Masson F, Pham P, Pousset F, Eurin J, Lechat P, Komajda M, Carayon A: Plasma levels and molecular forms of proatrial natriuretic peptides in healthy subjects and in patients with congestive heart failure. J Endocrinol 148:51–57, 1996 17. Wallen T, Landahl S, Hedner T, Hall C, Saito Y, Nakao K: Atrial natriuretic peptides predict mortality in the elderly. J Intern Med 241:269–275, 1997 18. Iivanainen AM, Tikkanen I, Tilvis R, Heikkila J, Helenius T, Kupari M: Association between atrial natriuretic peptides, echocardiographic findings and mortality in an elderly population sample. J Intern Med 241:261–168, 1997 19. Weston MW, Cintron GB, Giordano AT, Vesely DL: Normalization of circulating atrial natriuretic peptides in cardiac transplant recipients. Am Heart J 127:129–142, 1994 20. Ngo L, Bissett JK, Winters CJ, Vesely DL: Plasma prohormone atrial natriuretic peptides 1-98 and 31-67 increase with supraventricular and ventricular arrhythmias. Am J Med Sci 300:71–77, 1990 21. Ngo L, Vesely DL, Bissett JK, Murphy ML, Dinh H, Sallman AL, Rico DM, Winters CJ, Wyeth RP: Acute and sustained release of the atrial natriuretic factor prohormone N-terminus with acute myocardial infarction. Am J Med Sci 301:157–164, 1991 22. Leunissen KM, Menheere PP, Cheriex EC, Van den Berg BW, Noordzij TC, Van Hooff JP: Plasma alpha-human atrial natriuretic peptide and volume status in chronic hemodialysis patients. Nephrol Dial Transplant 4:382–386, 1989 23. Kohse KP, Feifel K, Mayer-Wehrstein R: Differential regulation of brain and atrial natriuretic peptides in hemodialysis patients. Clin Nephrol 40:83–90, 1993 24. Lauster F, Heim JM, Drummer C, Fu¨lle HJ, Gerzer R, Schiffl H: Plasma cGMP level as a marker of the hydration state in renal replacement therapy. Kidney Int 43(Suppl 41):S57–S59, 1993 25. Franz M, Pohanka E, Tribl B, Woloszczuk W, Ho¨rl WH: Living on chronic hemodialysis between dryness and fluid overload. Kidney Int 51(Suppl 59):S39–S42, 1997 26. Jaeger JQ, Mehta RL: Assessment of dry weight in hemodialysis: An overview. J Am Soc Nephrol 10:392–403, 1999 27. Cheriex EC, Leunissen KM, Janssen JH, Mooy JM, van Hooff JP: Echography of the inferior vena cava is a simple and reliable tool for the estimation of “dry weight” in haemodialysis patients. Nephrol Dial Transplant 4:563–568, 1989 28. Hartter E, Khalafpour S, Missbichler A, Hawa G, Woloszczuk W: Enzyme immunoassays for fragments (epitopes) of human proatrial natriuretic peptides. Clin Chem Lab Med 38:27–32, 2000 29. Villarreal D, Reams GP, Taraben A, Freeman RH: Hemodynamic and renal effects of proANP 31-67 in hypertensive rats. Proc Soc Exp Biol Med 221:166–170, 1999 30. Baker BJ, Wu WC, Winters CJ, Dinh H, Wyeth R, Sallman

Franz et al: ProANP peptides during hemodialysis AL, Vesely DL: Exercise increases the circulating concentration of the N-terminus of the atrial natriuretic factor prohormone in normal individuals. Am Heart J 122:1395–1402, 1991 31. McMurray RW Jr, Vesely DL: Calorie-restricted weight reduction, blood pressure, and atrial natriuretic peptides. Nutrition 9:178–182, 1993 32. Merkouris RW, Miller FC, Catanzarite V, Quirk JG Jr, Rigg

383

LA, Vesely DL: The N-terminal and C-terminal portions of the atrial natriuretic factor prohormone increase during preeclampsia. Am J Obstet Gynecol 164:1197–1202, 1991 33. Vesely DL, Arnold WC, Winters CJ, Sallman AL, Rico DM: Increased circulating concentration of the N-terminus of the atrial natriuretic factor prohormone in persons with pheochromocytomas. J Clin Endocrinol Metab 71:1138–1146, 1990