Saudi Journal of Kidney Diseases and Transplantation

: 2010  |  Volume : 21  |  Issue : 2  |  Page : 262--268

Clinical significance of N-Terminal Pro-B-type natriuretic peptide (NT-proBNP) in hemodialysis patients

Imed Helal1, Raja Belhadj2, Amira Mohseni2, Lilia Bazdeh3, Habiba Drissa3, Fethi ELyounsi1, Taieb Ben Abdallah1, Jaouida Abdelmoula2, Adel Kheder1,  
1 Department of Nephrology (Pr Kheder), Charles Nicolle Hospital, Tunis, Tunisia
2 Department of Biochemical, Charles Nicolle Hospital, Tunis, Tunisia
3 Department of Cardiology, Charles Nicolle Hospital, Tunis, Tunisia

Correspondence Address:
Imed Helal
Department of Nephrology, Charles Nicolle Hospital, Boulevard 9, Avril1006, Tunis


Circulating biomarkers play a major role in the early detection of cardiovascular di­sease. The purpose of this study was to determine levels of N-Terminal Pro-B-type Natriuretic Pep­tide (NT-proBNP) in hemodialysis (HD) patients and to examine the relationship of this marker to left ventricular hypertrophy and to cardiac dysfunction. Plasma NT-proBNP concentrations were measured in patients undergoing chronic HD, who did not any clinical evidence of heart failure, (n=32; mean age 43.14 ± 12 years; sex-ratio 1.8) as well as healthy volunteers (n=32; mean age 45.84 ± 1.9 years; sex-ratio 1). In addition, the correlation between plasma NT-proBNP concentration and parameters of echocardiography was examined. The plasma NT-proBNP levels in the HD patients were significantly higher (14422.6 ± 13757.8 pg/mL) than those in healthy volunteers (39.21 pg/mL) (P< 10 -3 ). In addition, the area under the receiver operating characteristic curve (ROC) re­vealed that the cut-off level of NT-proBNP was 288 pg/mL. On univariate analysis, the plasma NT­proBNP concentrations, in patients on HD, correlated positively with age (P= 0.004; r=0.5), systolic (P= 0.046; r= 0.36) and diastolic blood pressures (P= 0.037; r= 0.37), residual diuresis (P= 0.09; r= 0.3), the left atrial diameter (LAD) (P= 0.006; r= 0.55), left ventricular mass index (LVMI) (P= 0.01; r= 0.44) and negatively with albumin (P= 0.01; r= -0.44). However, there was no correlation bet­ween plasma levels of NT-proBNP and gender, body mass index (BMI), mean period on dialysis, pulse pressure, dry weight and left ventricular dysfunction. On multivariate analysis, only age (P= 0.014, RR= 2.8) was associated with significantly increased levels of NT-proBNP. Further studies are needed to carefully assess the diagnostic accuracy and prognostic value of NT-proBNP in patients on HD.

How to cite this article:
Helal I, Belhadj R, Mohseni A, Bazdeh L, Drissa H, ELyounsi F, Abdallah TB, Abdelmoula J, Kheder A. Clinical significance of N-Terminal Pro-B-type natriuretic peptide (NT-proBNP) in hemodialysis patients.Saudi J Kidney Dis Transpl 2010;21:262-268

How to cite this URL:
Helal I, Belhadj R, Mohseni A, Bazdeh L, Drissa H, ELyounsi F, Abdallah TB, Abdelmoula J, Kheder A. Clinical significance of N-Terminal Pro-B-type natriuretic peptide (NT-proBNP) in hemodialysis patients. Saudi J Kidney Dis Transpl [serial online] 2010 [cited 2022 Jan 26 ];21:262-268
Available from:

Full Text


Circulating biomarkers play a major role in the early detection of cardiovascular disease such as heart failure. In recent years, natriuretic pep­tides have become promising candidates in this respect; [1] particularly, B-type natriuretic peptides (BNP) have attracted much attention. The circulating half-life of BNP is 23 mins, whereas the inactive terminal fragment (NT-proBNP) has a much longer half-life (60-120 mins), which is of relevance for its use as a diagnostic tool. [1] The clinical benefit of using these markers to screen for cardiovascular (CV) risk has been well docu­mented in the general population. [1],[2],[3],[4] Patients with chronic kidney disease (CKD), including hemo­dialysis (HD) patients, have one of the highest CV risk scores. In this population, the clinical benefit of NT-proBNP measurements has not been well established. [5],[6],[7] The fact that HD pa­tients have significantly elevated BNP and NT­ proBNP levels [5],[7],[8] has been considered a major limitation for its use as a diagnostic tool. Eleva­ted levels were attributed to chronic fluid over­load, rendering measurement of NT-proBNP le­vels unreliable. However, considering the CV burden of patients with CKD, [9],[10],[11],[12],[13],[14],[15],[16] there is a de­finite need to explore the diagnostic value of NT-proBNP levels and to establish the valid nor­mal ranges in these patients.

In this study, we measured serum NT-proBNP levels in stable patients on chronic HD without clinical signs of progressive heart failure, and investigated its pathophysiological significance. In particular, our aim was to determine a cut-off NT-proBNP value adjusted for HD patients.

 Patients and Methods

Patient population

We studied 32 stable patients who had been on HD for at least six months at the Charles Ni­colle hospital. All patients were clinically stable and gave informed consent. Patients with acute cardiac disease and/or acute cardiac failure were excluded from the study. Clinical characteristics and laboratory data of the patients are shown in [Table 1]. All HD patients received regular dialy­sis using the cellulose-triacetate or polysulphone dialyzer three times per week in sessions lasting four hours each. The dialysate was bicarbonate buffered, the dialysate flow rate was 500 mL/ min, and blood flow ranged from 250 to 300 mL/min.

Plasma NT-proBNP concentrations were also measured in 32 control subjects. These subjects were carefully screened to exclude cardiac or other systemic disease.


In all patients, trans-thoracic echocardiogram­phy was performed on an inter-dialytic day in the evaluation phase.

M-mode and two-dimensional images as well as spectral- and colour-flow Doppler recordings were obtained (ATL). Left ventricular end-dia­stolic diameter (LVDd), left ventricular end-sy­stolic diameter (LVDs), inter-ventricular septal thickness (IVST), left atrial diameter (LAD) and left ventricular posterior wall thickness (LVP­WT) were measured.

Left ventricular ejection fraction (LVEF) was calculated by standard techniques. Left ventri­cular mass (LVM) was calculated by the re­gression equation described by Devreux and Reichek; LVM = 1.04 x [(IVST+LVPWT+LV­Dd) 3 - LVDd 3 ] - 13.6. This value was divided by the body surface area to determine the left ventricular mass index (LVMI), presented in g/m 2 . LVMI was defined as LVMI > 47 g/m 2 for females and > 50 g/m 2 for males.

Biochemical measurements

We collected blood samples from all patients before the commencement of a dialysis session, from the venous line. The samples were collec­ted in chilled tubes containing ethylenediamine tetraacetic acid (EDTA) and were immediately centrifuged at 4°C. Plasma was separated and stored at -20°C until assays were performed. NT-proBNP was measured using a comercially available electro-chemoluminescence immuno­assay that was performed on a Roche E2010 modular analytics system. All other biochemical parameters were measured with routine labora­tory methods.

 Statistical Analysis

We used the SPSS statistical software 11.5 (Lead Technologies Inc., Chicago, USA) for statistical analysis. Data are expressed as mean ± SD. Each factor was contrasted by univariate analysis, using either unpaired student's t test for continuous variables, or by K2 analysis for non-continuous data. Significant factors by uni­variate analysis were then compared by multi­variate analysis and the relative risk (RR) for each was calculated. We calculated a receiver operating characteristic (ROC) curve in order to establish an NT-proBNP cut-off value that dis­criminates HD patients from control subjects. Because of its abnormal distribution, the plasma NT-proBNP values were logarithmically trans­formed before regression analysis was perfor­med. Statistical significance was defined by P value less than 0.05.


General Patient Characteristics

The baseline characteristics of the 32 patients included in the study are shown in [Table 1]; the mean age 43.14 ± 12 years, sex-ratio was 1.8 and there was no clinical evidence of heart fai­lure. Among the healthy volunteers, the mean age was 45.84 ± 1.9 years and sex-ratio was 1. [Table 2] depicts the echocardiograph findings, LV mass and LV function in the patient popu­lation. Twenty-three patients (72%) had LVH on echocardiography. Diastolic dysfunction was present in 19 patients (59%).

ROC Analysis

The plasma NT-proBNP levels in the HD po­poulation were significantly higher (14422.6 ± 13757.8 pg/mL) than those among healthy sub­jects (39.21 pg/mL) (P -3 )

In addition, by creating ROC curves, we iden­tified a cut-off value for NT-proBNP that dis­criminates between HD patients and control sub­jects [Figure 1].

An NT-proBNP cut-off value of 288 pg/mL resulted in a 100% specificity and sensitivity.

 Correlation Study

Univariate analysis

On univariate analysis, plasma NT-proBNP con­centrations among HD patients correlated posi­tively with age (P= 0.007; r= 0.41), pre-HD systolic (P= 0.046; r= 0.36) and diastolic blood pressures (P= 0.037; r= 0.37), residual diuresis (P= 0.09; r=0 .53), the LAD (P= 0.009; r= 0.53), LVMI (P= 0.01; r= 0.44) and negatively with BMI (P= 0.04; r= -0.32) and albumin (P= 0.01; r= -0.44).

However, there was no correlation between plasma levels of NT-proBNP and gender, mean period on dialysis, pulse pressure, dry weight and left ventricular dysfunction.

Multivariate Analysis

On multivariate analysis, only age of the study patients (P= 0.014, RR= 2.8) was associated with significantly elevated NT-proBNP levels under­lining that it is an independent factor [Figure 2]. NT-proBNP levels were also statistically signifi­cantly associated with BMI (P = 0.07, RR= -2).


The measurement of the plasma concentration of cardiac natriuretic peptides, particularly BNP, in uremic patients on chronic dialysis, has proved to be useful for the diagnosis of LVH and for excluding the presence of LV dysfunction. Left ventricular hypertrophy and LV dysfunction are currently considered the strongest predictors of cardiovascular and total mortality in the dialysis population. [17],[18] LVH is a notoriously pervasive complication in end-stage renal disease with a prevalence rate ranging from 60 to 80%. LV systolic dysfunction is relatively much less fre­quent, with its prevalence being approximately 15%. [18] Our study again further confirms in a dialysis population the high prevalence of these alterations. If these alterations are to be a target for intervention, simple and reliable methods, specifically validated in the dialysis population are needed. Echocardiography is undoubtedly of proven value, and most agree that serial echo­cardiographic studies in patients entering renal replacement therapy is a better system to iden­tify and treat alterations in LV mass and func­tion in these patients. However, hospital-based echocardiographic services are often stretched, and for this reason, in routine clinical practice, this technique is applied much less often than desirable. The situation is probably similar to that of general practice. [19] Echocardiography apart, there are no simple and validated methods to clinically diagnose these abnormalities in dia­lysis patients. Echocardiogram, which is a widely available method in the general population, has a sensitivity that may reach at most 40%, [20] and its use in dialysis patients poses additional pro­blems. Thus, the opportunity for intervention on LVH and systolic dysfunction is in part limited by the availability of simple, easily accessible diagnostic tools.

In the present study, we investigated whether the natriuretic peptide NT-proBNP may be of help in the assessment of LVH in patients on maintenance HD, since its diagnostic potential has been extensively studied in patients with normal kidney function. [1],[2],[3],[6] We found clearly elevated NT-proBNP levels in our HD patients; in fact, even patients without a diagnosis of LVD showed values more than ten-fold higher than the normal range given by the manufacturer. Our finding is in line with repeatedly reported increased levels of natriuretic peptides as a re­sult of reduced renal excretion and chronic vo­lume overload in CKD patients without LVD. [5],[6],[21],[22],[23] We have to point out, however, that a recommendation for specific reference values in order to use NT-proBNP routinely as a diagnos­tic tool to separate LVD from over-hydration in these patients is not available yet.

The brain natriuretic peptide was first isolated from a porcine brain and has potent biological effects, as does the atrial natriuretic peptide. [24] Plasma brain natriuretic peptide levels are very low or not detectable in normal subjects; the brain natriuretic peptide is a novel cardiac hor­mone secreted mainly from the ventricules in patients with heart failure. [25] It has been reported that high left ventricular end-diastolic pressure or low ejection fraction might stimulate the se­cretion of brain natriuretic peptide. [26],[27] How­ever, there have been only a few reports on the mechanisms involved in the reduction of brain natriuretic peptide levels after HD in patients with chronic renal failure. [28],[29]

he present study confirms previous reports that plasma NT-proBNP levels are increased in dialysis patients. However, the current study does not confirm previous reports that NT­proBNP is a biomarker for LVH in patients with uremia, and is probably influenced by isolated renal dysfunction. Also, we found that only the age of patients was associated with a significant increase of NT-proBNP levels, indicating that other factors and interventions were probably biased by age. Another explanation for the pre­sent data is that aging is a major risk factor for cardiac disease and the increase in the levels of the peptides is dependent on age.

Our study has limitations, with the most im­portant being the small number of our study population. Also, we focused on patients with­out overt heart failure because, we believed that in overt heart failure, echocardiography is al­most always performed to confirm the clinical diagnosis and to incorporate anatomic and he­modynamic information into the clinical deci­sion process. Thus, the diagnostic value of these peptides for overt heart failure, although likely, cannot be extrapolated from our data. The third limitation derives from the fact that we did echocardiographic studies and plasma sampling during the dialysis interval rather than imme­diately before or after dialysis. While this ap­proach is probably ideal from a physiological standpoint (the volume status approximates the individual steady state between dialysis sessions), it demands an additional outpatient appointment during a non-dialysis day. Pre-dialysis sampling tends to shift upwardly the diagnostic thres­holds of these peptides, and it remains to be seen whether this affects the diagnostic power of NT-proBNP. Nonetheless, it is common prac­tice in many dialysis centers to periodically re­examine patients in the dialysis interval, and this may be a good occasion for NT-proBNP testing. Finally, although we analyzed data both on the basis of prospectively preset thresholds and on retrospectively defined "best cut offs," the diagnostic potential of BNP was maximized by the second, retrospective, approach. Thus, these retrospective thresholds, which are speci­fic for patients on dialysis, remain to be pros­pectively tested in other dialysis centers to prove the external validity of our findings.


In this study, we found that only the age of patients was associated with a significant in­crease of NT-proBNP levels, indicating that other factors were probably biased by the age. Fi­nally, we can summarize that NT-proBNP is not a good marker of cardiovascular disease or water overload in HD patients. These low-mole­cular peptides are similar to beta2-microglo­bulin and other low molecular weight peptides and proteins that accumulate in blood due to decreased glomerular filtration rate. Their eli­mination during HD is influenced by the type of HD membrane. Further work is needed to care­fully assess the diagnostic accuracy and prog­nostic value of NT-proBNP in HD patients.


1Felker GM, Petersen JW, Mark DB. Natriuretic peptides in the diagnosis and management of heart failure. CMAJ 2006;175:611-7.
2Goetze JP, Christoffersen C, Perko M, et al. Increased cardiac BNP expression associated with myocardial ischemia. FASEB J 2003;17: 1105-7.
3Konstam MA. Natriuretic peptides and cardio­vascular events: More than a stretch. JAMA 2007; 297:212-4.
4Sullivan DR, West M, Jeremy R. Utility of brain natriuretic peptide (BNP) measurement in cardio­vascular disease. Heart Lung Circ 2005; 14:78-84.
5Racek J, Kralova H, Trefil L, et al. Brain natriu­retic peptide and N-terminal proBNP in chronic haemodialysis patients. Nephron Clin Pract 2006; 103:c162-72.
6Wang AY, Lam CW, Yu CM, et al. N-terminal probrain natriuretic peptide: An independent risk predictor of cardiovascular congestion, mortality, and adverse cardiovascular outcomes in chronic peritoneal dialysis patients. J Am Soc Nephrol 2007;18:321-30.
7Khan IA, Fink J, Nass C, et al. N-terminal pro-B­type natriuretic peptide and B-type natriuretic peptide for identifying coronary artery disease and left ventricular hypertrophy in ambulatory chronic kidney disease patients. Am J Cardiol 2006;97: 1530-4.
8Vickery S, Price CP, John RI, et al. B-type natriu­retic peptide (BNP) and amino-terminal pro-BNP in patients with CKD: relationship to renal func­tion and left ventricular hypertrophy. Am J Kidney Dis 2005;46:610-20.
9Fort J. Chronic renal failure: A cardiovascular risk factor. Kidney Int Suppl 2005;99:S25-9.
10Das M, Aronow WS, McClung JA, et al. Increased prevalence of coronary artery disease, silent myo­cardial ischemia, complex ventricular arrhythmias, atrial fibrillation, left ventricular hypertrophy, mitral annular calcium, and aortic valve calcium in patients with chronic renal insufficiency. Cardiol Rev 2006;14:14-7.
11Guerin AP, Pannier B, Marchais SJ, et al. Cardiovas­cular disease in the dialysis population: prognostic significance of arterial disorders. Curr Opin Nephrol Hypertens 2006;15:105-10.
12Kielstein JT, Zoccali C. Asymmetric dimethylargi­nine: a cardiovascular risk factor and a uremic toxin coming of age? Am J Kidney Dis 2005;46:186­-202.
13Dikow R, Zeier M, Ritz E. Pathophysiology of cardiovascular disease and renal failure. Cardiol Clin 2005;23:311-7.
14Foley RN, Parfrey PS. Cardiovascular disease and mortality in ESRD. J Nephrol 1998;11:239-45.
15Foley RN, Parfrey PS, Sarnak MJ. Epidemiology of cardiovascular disease in chronic renal di-sease. J Am Soc Nephrol 1998;9(Suppl):S16-23.
16Foley RN, Parfrey PS, Sarnak MJ. Clinical epi­demiology of cardiovascular disease in chronic renal disease. Am J Kidney Dis 1998;32(Suppl3): S112-9.
17Parfrey PS, Harnett JD, Barre PE. The natural history of myocardial disease in dialysis patients. J Am Soc Nephrol 1991;2:2-12.
18Parfrey PS, Foley RN, Harnett DJ, et al. Outcome and risk factors for left ventricular disorders in chronic uraemia. Nephrol Dial Transplant 1996; 11:1277-85.
19Smith H, Pickering RM, Struthers A, et al. Bio­chemical diagnosis of ventricular dysfunction in general practice: Observational study. Br Med J 2000;320:906-8.
20Schillaci G, Verdecchia P, Borgioni C, et al. Im­proved electrocardiographic diagnosis of left ven­tricular hypertrophy. Am J Cardiol 1994;74:714-9.
21Suresh M, Farrington K. Natriuretic peptides and the dialysis patient. Semin Dial 2005;18:409-19.
22Fagugli RM, Palumbo B, Ricciardi D, et al. Asso­ciation between brain natriuretic peptide and extra­cellular water in hemodialysis patients. Nephron Clin Pract 2003;95:c60-6.
23Sharma R, Gaze DC, Pellerin D, et al. Raised plas­ma N-terminal pro-B-type natriuretic peptide con­centrations predict mortality and cardiac disease in end-stage renal disease. Heart 2006; 92:1518-9.
24Sudoh T, Kangawa K, Minamino N, Matsuo H. A new natriuretic peptide in porcine brain. Nature 1988;332:78-81.
25Mukoyama M, Nakao K, Saito Y, et al. Increased human natriuretic peptide in congestive heart failure. N Engl J Med 1990;323:757-8.
26Nakamura M, Kawata Y, Yoshida H, et al. Rela­tionship between plasma atrial natriuretic peptide concentration and hemodynamic parameters during percutaneous transvenous mitral commissurotomy in patients with mitral stenosis. Am Heart J 1992; i24:1283-8.
27Yoshimura M, Yassue H, Okumura K, et al. Diffe­rent secretion patterns of atrial natriuretic peptide and Brain natriuretic peptide in patients with con­gestive heart failure. Circulation 1993; 87:464-9.
28Lang CC, Choy AM, Henderson IS, Coutie WJ. Effect of haemodialysis on plasma levels of brain natriuretic peptide in patients with chronic renal failure. Clin Sci 1992; 82: 127-131.
29Buckley MG, Markandu ND, Sagnella GA, Singer DR, MacGregor GA. Plasma concentration and comparisons of brain natriuretic peptide and atrial natriuretic peptide in normal subjects, cardiac transplant recipients and patients with dialysis­independent or dialysis-de-pendent chronic failure. Clin Sci 1992;83:437-44.