Journal Logo

Original Clinical Science—General

Pretransplant NT-proBNP, Dialysis Vintage, and Posttransplant Mortality in Kidney Transplant Recipients

Yeung, Stanley M.H. MD1; van Londen, Marco MD, PhD1; Nakshbandi, Uzma; Said, M. Yusof MD1; Eisenga, Michele F. MD1; Hepkema, Bouke G. PhD1; Nolte, Ilja M. PhD1; Berger, Stefan P. MD, PhD1; de Borst, Martin H. MD, PhD1; Bakker, Stephan J.L. MD, PhD1

Author Information
doi: 10.1097/TP.0000000000003125



Kidney transplantation is the treatment of choice for end-stage kidney disease (ESKD) patients, as transplantation improves quality of life and life expectancy compared with dialysis.1,2 Preemptive transplantation generally leads to better outcomes compared with transplantation after a period of dialysis.3,4 Nevertheless, due to the generalized shortage in donor organs, patients with ESKD often need to start with dialysis.5 Dialysis vintage is a significant and prognostic marker of mortality in ESKD patients and kidney transplant recipient (KTR) with or without underlying cardiovascular disease.6-8 Several mechanisms are proposed to underlie the detrimental effects derived from dialysis on the cardiovascular system. Dialysis causes significant stress on the cardiovascular system by accelerated atherosclerosis, intradialytic volume overload, and subsequent progression of left ventricular hypertrophy.9 Volume overload causes ventricular wall stress resulting in the release of N-terminal probrain natriuretic peptide (NT-proBNP) by the ventricular cells.10,11 High serum levels of NT-proBNP are strongly associated with volume overload and NT-proBNP is a well-established biomarker to diagnose and assess the severity of congestive heart failure.12-14 High plasma NT-proBNP is also independently associated with mortality in general population, ESKD patients, and KTR.15-19 Moreover, it has been shown that NT-proBNP is independently associated with dialysis vintage.20 However, it is not known if pretransplant NT-proBNP levels are associated with posttransplantation mortality and if NT-proBNP plays a role in the association of dialysis vintage and mortality in KTR.

The aim of this study is to assess whether dialysis vintage and pretransplant NT-proBNP are independently associated with mortality in KTR, and in this respect, we also aimed to investigate whether variation in pretransplant NT-proBNP influence the association between dialysis vintage and mortality in KTR.


Study Population

This study was conducted in the Transplant Lines Genetics cohort; details of the cohort have been published previously.21 From a total of 912 KTR who underwent transplantation between January 1995 and December 2005 in the University Medical Center Groningen, the Netherlands, we prospectively selected 686 KTR with available pretransplant serum NT-proBNP levels. We excluded 28 KTR who had a primary nonfunction graft, defined as graft function inadequate to prevent the need for dialysis in the absence of rejection or surgical etiologies of graft failure for 3 months post-transplantation. This results in 658 KTR available for analyses. The Institutional Review Board of the University Medical Center Groningen approved the study protocol (METc 2014/077). All KTR gave written informed consent. All procedures were conducted in accordance with the declarations of Helsinki and Istanbul.

Recipient Characteristics and Follow-up

Relevant donor, recipient, and transplant characteristics were extracted from the Groningen Kidney Transplant Database, which contains information of all kidney transplantations that have been performed at our center since 1968. In addition, the database contains the outcomes of outpatient visits (eg, cause of death, and other relevant variables) after transplantation. Missing patient characteristics, transplantation-related parameters, and pretransplantation clinical and laboratory data were complemented with data from medical records. These data included pretransplantation characteristics, dialysis vintage, and serum creatinine (sample was taken before transplantation and before dialysis if this was necessary before the transplantation).

Pretransplant plasma samples were stored at −80°C until assessment of biochemical measures for this study. Plasma NT-proBNP levels were measured afterward by immunoassay on an Elecsys 2010 analyzer instrument (ELECSYS proBNP, Roche Diagnostics, Mannheim, Germany).22 The body surface area (BSA) was calculated as BSA = (W0.425 (kg) × H0.725 (cm)) × 0.00718423 and body mass index (BMI) as weight (kg) by height (m) squared. Smoking behavior was classified into 2 groups as never or former smoking and current smoking. History of cardiovascular disease was considered positive if there was an event of previous angina pectoris, myocardial infarction, objectified cardiac valve disease, coronary artery disease, left ventricular hypertrophy, congestive heart failure, peripheral artery disease including aortic aneurysms. History of cerebrovascular accident (CVA) was considered as a previous transient ischemic attack or CVA. Diabetes was defined as the use of antidiabetic medication or a fasting plasma glucose >7.0 mmol/L.

Standard immunosuppression consisted of cyclosporine microemulsion (Neoral, Novartis, Pharma b.v., Arnhem, the Netherlands; 10 mg/kg; trough-levels idem) and prednisolone from 1993 to 1996. Mycophenolate mofetil (Cellcept, Roche, Nederland b.v., Woerden, the Netherlands; 2 g/d) was added from May 1997 to date. Since 1998, standard immunosuppression consisted of triple therapy with tacrolimus (Prograft or Advagraf, Astellas Pharma b.v. Leiden, the Netherlands); initial trough level 8–12 ng/mL, followed by 6–10 ng/mL (>2 mo) and 4–6 ng/mL (>6 mo) or cyclosporine microemulsion (Neoral, Novartis Pharma b.v. Arnhem, the Netherlands; 2 times 4 mg/kg daily); initial trough level 200–500 μg/L, 150–500 μg/L (>2 mo), and 75–125 μg/L (>6 mo), in combination with mycophenolate mofetil (Cellcept, Roche b.v. Woerden, the Netherlands; 2 g daily or Myfortic, Novartis Pharma, b.v. Arnhem, the Netherlands; 1440 mg daily), and prednisolone. The current immunosuppressive medication was extracted from the medical records.

Study Outcome

The primary outcome of this study was all-cause mortality censored for graft failure. Graft failure was defined as ESKD requiring dialysis or retransplantation. The outpatient program of the hospital uses continuous surveillance systems by the municipal registration of death to ensure up-to-date information on patient status (alive or deceased). Secondary outcome was cardiovascular mortality censored for noncardiovascular mortality and graft failure. Information on the cause of death was derived from patients’ medical records and was assessed by a nephrologist. Cardiovascular mortality was defined as death due to cerebrovascular disease, ischemic heart disease, heart failure, or sudden circulatory death, and coded according to a previously specified list of International Classification of Diseases, Ninth Revision (codes 410–447) as described previously.24

Statistical Analyses

Normally distributed variables are presented as mean ± standard deviation and skewed distributed variables as median (interquartile range). Non-normally distributed variables were transformed for subsequent analyses.

To identify independent correlations of NT-proBNP with relevant patient factors, univariable and subsequent multivariable linear regression analyses were performed. Multivariable linear regression models were constructed using backward selection (Pout > 0.05) including variables that were significantly associated with NT-proBNP in univariable analysis and clinically relevant variables. Kaplan-Meier survival analyses were performed to study the associations of tertiles of NT-proBNP and dialysis vintage with all-cause mortality and differences between tertiles were tested with the log-rank test.

To investigate the association of NT-proBNP and dialysis vintage with the primary study outcome, Cox regression analyses were performed on both variables. NT-proBNP was log2-transformed to obtain the best-fitting model and a 2 base was used to allow for expression of the hazard ratios (HRs) per doubling of NT-proBNP. Similarly, dialysis vintage was square-root transformed to obtain the best-fitting model. Both transformed values of NT-proBNP and dialysis vintage were standardized to compare these 2 values with each other in the following Cox regression models. Model 1 shows crude HR for the standardized value of either standardized values of log2-transformed NT-proBNP or square root-transformed dialysis vintage. In model 2, we first adjusted for age and sex. In model 3, we additionally adjusted for pretransplant plasma creatinine, history of cardiovascular disease and CVA, diastolic blood pressure, diabetic nephropathy, and BSA; in model 4, we additionally adjusted for transplantation variables such as cold ischemic time (CIT), living donor, delayed graft function, and cytomegalovirus (CMV) status of recipient; in model 5, we additionally adjusted for dialysis modality (hemodialysis). Finally, in the last model, we adjusted for standardized values of log2-transformed NT-proBNP or standardized values of square root-transformed dialysis vintage dependent on the first model. We also tested the variables for hazard proportionality over time with the Schoenfeld residuals test. Similar analyses were performed for the secondary outcome. Because of lower numbers of events, these were restricted to analyses with continuous log2-transformed NT-proBNP and square root–transformed dialysis vintage. To visualize the HR for the primary outcome for patients with elevated NT-proBNP, we plotted a linear spline, fitted on the fully adjusted Cox regression model in which the reference point of NT-proBNP 1200 ng/L was used as reference value as it was found to be the optimal diagnostic cut point for congestive heart failure in chronic kidney disease patients with severe kidney insufficiency (estimated glomerular filtration rate of 44 mL/min/1.73 m2).25 We compared the percentage change in the final model with addition of NT-proBNP or dialysis vintage. Percentage change in HR was calculated as (HR before adjustment − HR after adjustment)/(HR before adjustment − 1) × 100. Effect moderation was analyzed using Cox regression with all-cause mortality as the end point, and NT-proBNP as the condition, and their 2-way interaction as predictors. Effect moderators included in these analyses were age, sex, BSA, and plasma creatinine.

Sensitivity analyses for the primary outcome were performed by excluding all preemptive transplantations.

All statistical analyses were performed with SPSS software, version 22.0 for Windows (IBM, Armonk, NY), R version 3.5.2 (Vienna, Austria) (, and STATA 12.0 (STATA Corp., College Station, TX). In all analyses, a 2-sided P < 0.05 was considered significant.


Baseline Characteristics

We analyzed 658 KTR with a mean age of 46.8 ± 13.7 years, of which 59.4% of the recipients were males. Median dialysis vintage was 3.10 [1.76–4.59] years and median pretransplant NT-proBNP concentration was 2019 (787–5218) ng/L. Baseline characteristics according to tertiles of NT-proBNP are summarized in Table 1. KTR with higher concentrations of NT-proBNP were older, more often male, and were more likely to have diabetes, diabetic nephropathy, and cardiovascular diseases before transplantation. Moreover, KTR with higher levels of NT-proBNP had a longer dialysis vintage with hemodialysis being the most frequently used dialysis modality compared with peritoneal dialysis or no dialysis. Furthermore, KTR with higher NT-proBNP levels had a longer CIT and experienced a higher incidence of delayed graft function and CMV seropositivity. In contrast, BMI, BSA, diastolic blood pressure, living donors, and preemptive transplantation were inversely related to NT-proBNP. Pretransplant plasma creatinine and 1-year posttransplant estimated glomerular filtration rate were not associated with pretransplant NT-proBNP. Multivariable linear regression analyses showed that recipient age, history of cardiovascular disease, dialysis vintage, diabetic nephropathy, hemodialysis, CIT, CMV seropositivity in recipients, and living donors were positively associated with NT-proBNP. BSA was inversely associated with NT-proBNP (Table S1, SDC,

TABLE 1. - Baseline characteristics according to tertiles of NT-proBNP in 658 kidney transplant recipients
Tertiles of (NT-pro)BNP (ng/L)
Overall I II III NT-proBNP (St. β) P
NT-proBNP, (ng/L) 2019 (787–5218) 588 (361–786) 2019 (1571–2756) 8429 (5224–22 112)
Recipient demographics
 Age (y) 46.8 ± 13.7 43.3 ± 13.4 47.9 ± 13.4 49.1 ± 13.6 0.22 <0.001
 Male, n (%) 391 (59.4) 138 (63.0) 136 (61.8) 117 (53.4) 0.10 0.01
 BMI (kg/m2) 24.1 (21.6–27.0) 24.4 (21.9–27.3) 24.9 (21.8–27.4) 23.6 (21.2–25.7) −0.12 0.003
 BSA (m2) 1.88 ± 0.19 1.91 ± 0.19 1.91 ± 0.20 1.83 ± 0.17 −0.17 <0.001
 Systolic blood pressure (mm Hg) 142.9 ± 17.3 142.5 ± 16.7 143.9 ± 17.7 142.3 ± 17.6 −0.01 0.72
 Diastolic blood pressure (mm Hg) 83.8 ± 10.9 85.0 ± 11.8 83.7 ± 10.5 82.9 ± 10.3 −0.09 0.02
 Pretransplant plasma creatinine (μmol/L) 916 ± 284 896 ± 295 952 ± 291 900 ± 265 0.02 0.63
 1 y posttransplant eGFR (mL/min/1.73 m2) 48.4 ± 15.6 48.4 ± 15.1 47.8 ± 15.1 48.9 ± 16.6 −0.03 0.40
 Diabetes, n (%) 28 (4.3) 6 (2.7) 8 (3.6) 14 (6.4) 0.09 0.02
 History of cardiovascular disease, n (%) 68 (10.3) 9 (4.1) 20 (9.1) 39 (17.8) 0.18 <0.001
 History of cerebrovascular accident, n (%) 12 (1.8) 3 (1.4) 5 (2.3) 4 (1.8) 0.01 0.72
 Smoking at transplantation, n (%) 29 (4.4) 8 (3.7) (n = 154) 14 (6.4) (n = 138) 7 (3.2) (n = 117) 0.17 <0.001
 Dialysis vintage before transplantation (y) 3.10 (1.76–4.59) 1.92 (0.94–3.37) 3.28 (2.03–4.43) 4.09 (2.55–5.28) 0.38 <0.001
Dialysis modality, n (%)
 Hemodialysis 359 (54.6) 72 (32.9) 126 (57.3) 161 (73.5) 0.35 <0.001
 Peritoneal dialysis 267 (40.6) 118 (53.9) 91 (41.4) 58 (26.5) −0.22 <0.001
 No dialysis/preemptive transplantation 32 (4.9) 29 (13.2) 3 (1.4) 0 (0) −0.30 <0.001
Primary kidney disease, n (%)
 Primary glomerular disease 176 (26.7) 64 (29.2) 60 (27.3) 52 (23.7) −0.05 0.25
 Glomerulonephritis 42 (6.4) 13 (5.9) 15 (6.8) 14 (6.4) −0.01 0.75
 Tubular interstitial disease 82 (12.5) 34 (15.5) 20 (9.1) 28 (12.8) −0.02 0.70
 Polycystic kidney disease 109 (16.6) 39 (17.8) 39 (17.7) 31 (14.2) −0.03 0.49
 Dysplasia and hypoplasia 16 (2.4) 6 (2.7) 6 (2.7) 4 (1.8) −0.07 0.09
 Renovascular disease 51 (7.8) 18 (8.2) 13 (5.9) 20 (9.1) −0.01 0.73
 Diabetic nephropathy 17 (2.6) 4 (1.8) 3 (1.4) 10 (4.6) 0.11 0.004
 Other or unknown cause 165 (25.1) 41 (18.7) 64 (29.1) 60 (27.4) 0.08 0.05
Donor demographics
 Donor age (y) 43.1 ± 14.1 44.2 ± 12.8 42.6 ± 14.5 42.6 ± 14.9 −0.02 0.56
 Male, n (%) 333 (50.6) 105 (47.9) 107 (48.6) 113 (51.6) −0.008 0.84
Transplantation details
 Living donor, n (%) 150 (22.8) 70 (32.0) 47 (21.4) 33 (15.1) −0.21 <0.001
 Acute rejection, n (%) 234 (35.6) 69 (31.5) 83 (37.7) 82 (37.4) 0.04 0.37
 Delayed graft function, n (%) 169 (25.7) 43 (19.6) 63 (28.6) 63 (28.8) 0.10 0.01
 CIT (h) 18.0 (10.4–23.0) 15.6 (3.0–21.0) 19.0 (11.1–23.1) 19.9 (14.0–24.0) 0.24 <0.001
 CIT excluding living donors (h) 20.0 (16.0–24.0) 18.0 (15.5–23.0) 20.5 (16.0–25.0) 21.0 (17.0–25.7) 0.13 0.003
 Warm ischemic time (min) 35 (30–44) 36.0 (30–45) 35 (30–43) 36 (30–43) −0.02 0.57
CMV status
 CMV seropositivity recipient, n (%) 292 (44.4) 78 (35.6) 96 (43.6) 118 (53.9) 0.14 <0.001
 CMV seropositivity donor, n (%) 320 (48.6) 100 (45.7) 111 (50.5) 110 (50.2) 0.01 0.80
 Second transplantation, n (%) 43 (6.5) 12 (5.5) 14 (6.4) 17 (7.8) 0.03 0.42
 Third transplantation, n (%) 3 (0.5) 0 (0) 1 (0.5) 2 (0.9) 0.06 0.15
 Use of corticosteroids, n (%) 653 (99.2) 219 (100) 216 (98.2) 218 (99.5) −0.01 0.78
 Use of calcineurin inhibitor, n (%) 642 (97.6) 215 (98.2) 212 (96.4) 215 (98.2) 0.02 0.70
 Use of proliferation inhibitor, n (%) 544 (82.7) 182 (83.1) 174 (79.1) 188 (85.8) 0.03 0.38
 Use of mTOR, n (%) 35 (5.3) 12 (5.5) 16 (7.3) 7 (3.2) −0.05 0.24
 Other immunosuppressants, n (%) 207 (31.5) 69 (31.5) 72 (32.7) 66 (30.1) 0.006 0.89
Data are presented as n (%), mean ± standard deviation, or median (interquartile range) for nominal, normally distributed, and non-normally distributed data, respectively. The P value represents the P for trend in univariable linear regression analysis. Bold indicates statistical significance.
BMI, body mass index; BSA, body surface area; CIT, cold ischemic time; CMV, cytomegalovirus; eGFR, estimated glomerular filtration rate; mTOR, mammalian target of rapamycin; NT-proBNP, N-terminal prohormone of brain natriuretic peptide.

Dialysis Vintage, All-cause Mortality, and Cardiovascular Mortality

During 12.7 (7.8–15.6) years of follow-up, 248 (37.7%) of 658 KTR died before development of graft failure, with 68 (27.4%) deaths being classified as of cardiovascular origin. Of the patients who died, 45 (18.1%) had a history of known cardiovascular disease compared with 23 (5.6%) of patients who survived (P < 0.001). Median pretransplant NT-proBNP concentrations of the patients who died during follow-up were higher (2913 [IQR, 1394–8522] ng/L) compared with those of patients who survived (1594 [IQR, 647–3985] ng/L) (P < 0.001).

Dialysis vintage was associated with all-cause mortality (HR: 1.33; 95% confidence interval [CI], 1.18-1.51; P < 0.001 per SD increase, Table 2). This association remained independent of adjustment for age, sex, pretransplant serum creatinine, BSA, history of CVD and diabetes, CIT, living donor, delayed graft function, CMV status of recipient, and dialysis modality (hemodialysis) (Table 2, model 5: HR [95% CI]: 1.22 [1.03-1.43]; P = 0.02). When standardized values of log2NT-proBNP were added to the model, the HR of dialysis vintage with all-cause mortality decreased by 40% and this association became nonsignificant (Table 2, model 6: HR [95% CI]: 1.14 [0.96-1.34], P = 0.14). Furthermore, dialysis vintage was associated with cardiovascular mortality as a secondary outcome; however, this association did not remain significant in the adjusted models (Table S2, SDC,

TABLE 2. - Associations of continuous standardized log2 NT-proBNP (ng/L), standardized square root dialysis vintage (y), and NT-proBNP in tertiles with all-cause mortality in 658 stable kidney transplant recipients
Log2 NT-proBNP in Z score Square root dialysis vintage in Z score NT-proBNP I NT-proBNP II NT-proBNP III
Model HR (95% CI) HR (95% CI) 588 (361–786) 2019 (1571–2756) 8429 (5224–22112)
1 1.67 (1.48-1.89)**** 1.33 (1.18-1.51)**** 1.0 (ref) 1.93 (1.37–2.72)**** 2.91 (2.09–4.04)****
2 1.49 (1.31-1.68)**** 1.28 (1.11-1.47)**** 1.0 (ref) 1.47 (1.04–2.08)* 2.26 (1.62–3.15)****
3 1.44 (1.27-1.64)**** 1.35 (1.17-1.55)**** 1.0 (ref) 1.49 (1.05–2.11)* 2.14 (1.51–3.02)****
4 1.41 (1.23-1.61)**** 1.30 (1.10-1.52)*** 1.0 (ref) 1.45 (1.02–2.05)* 2.00 (1.40–2.84)****
5 1.34 (1.16-1.55)**** 1.22 (1.03-1.43)* 1.0 (ref) 1.32 (0.93–1.89) 1.73 (1.19–2.50)**
6 1.31 (1.13-1.52)**** 1.14 (0.96-1.34) 1.0 (ref) 1.28 (0.89–1.82) 1.60 (1.10–2.33)*
Data are presented as HR; 95% CI; NT-proBNP; Z score, standardized score; P value is shown as: *≤0.05, **≤0.01, ***0.001, ****<0.001. Model 1 = crude standardized values of log2 NT-proBNP/square root of dialysis days. Model 2 = as model 1 and additionally adjusted for age and sex. Model 3 = as model 2 and additionally adjusted for pretransplant serum creatinine, history of cardiovascular disease and cerebrovascular accident, diastolic blood pressure, diabetic nephropathy, and body surface area. Model 4 = as model 3 and additionally adjusted for cold ischemic time, living donor, delayed graft function, and cytomegalovirus status of recipient. Model 5 = as model 4 and additionally adjusted for dialysis modality (hemodialysis). Model 6 = as model 5 and additionally adjusted for crude standardized values of log2 NT-proBNP/square root of dialysis year.
CI, confidence interval; HR, hazard ratio; NT-proBNP, N-terminal prohormone of brain natriuretic peptide.

All-cause Mortality According to Tertiles of NT-proBNP

The mortality rates in NT-proBNP tertiles were as follow: 53 (24.2%) in the first, 87 (39.5%) in the second, and 108 (49.3%) in the third tertile (Figure 1: log-rank P < 0.001). In univariable Cox regression analysis of NT-proBNP as a continuous variable, NT-proBNP was significantly associated with all-cause mortality (Table 2: HR [95% CI]: 1.67 [1.48-1.89]; P < 0.001 per SD increase). This association remained significant after adjustment for the same variables as done with the dialysis vintage models (Table 2, model 5: HR [95% CI]: 1.34 [1.16-1.55]; P < 0.001). When standardized value of square root transformed dialysis vintage was added to the model, the HR of NT-proBNP with all-cause mortality was reduced by 9% (Table 2, model 6: HR [95% CI]: 1.31 [1.13-1.52], P < 0.001), but the strong association remained significant. Similar results, albeit with a slightly stronger point estimate of the HR, were found in analyses with cardiovascular mortality as a secondary outcome (Table S2, SDC, When analyzed according to tertiles of NT-proBNP, patients in the highest tertile had the highest mortality risk (Table 2, model 6: HR [95% CI]: 1.60 [1.10-2.33], P = 0.02) compared with patients in the first tertile. A fully adjusted linear spline for the association between NT-proBNP and risk of mortality is shown in Figure 2. Furthermore, we found no significant effect modification by age, sex, BSA, or plasma creatinine for the association between NT-proBNP levels and all-cause mortality. Schoenfeld residuals test revealed proportional hazards.

Kaplan-Meier curve for all-cause mortality censored for graft failure among 658 kidney transplant recipients according to tertiles of baseline NT-proBNP (ng/L). Tertile I = 588 (361–786), tertile II = 2019 (1571–2756), tertile III = 8429 (5224–22 112). NT-proBNP, N-terminal probrain natriuretic peptide.
Associations between NT-proBNP and all-cause mortality in 658 KTR. Data were fit by a Cox proportional hazards regression model based on restricted cubic splines and adjusted for age, sex, pretransplant serum creatinine, history of CVD and CVA, diastolic blood pressure, diabetic nephropathy, BSA, cold ischemic time, living donor, delayed graft function, CMV status of recipient, dialysis modality (hemodialysis), and square root of dialysis vintage (y). Data are shown on a standardized log2 scale. Reference standard was 1200 ng/L defined as an optimal cut-off for diagnosis of heart failure in patients with GFR <44 mL/min/1.73 m2. The grey area represents the 95% CI. Z scores on the x-axis corresponds with the following NT-proBNP values (ng/L): −2 = 125; −1 = 535; 0 = 2290; 1 = 9797; 2 = 42 055; 3 = 180 295. BSA, body surface area; CI, confidence interval; CMV, cytomegalovirus; CVA, cerebrovascular accident; CVD, cardiovascular disease; GFR, glomerular filtration rate; KTR, kidney transplant recipient; NT-proBNP, N-terminal probrain natriuretic peptide.

Sensitivity Analyses

In sensitivity analyses, when excluding preemptive transplantations, the HRs for analyses with the primary outcome remained similar: NT-proBNP in the fully adjusted model: HR (95% CI): 1.30 (1.13-1.50); P < 0.001 was associated with all-cause mortality. When adding NT-proBNP in the fully adjusted model of dialysis vintage, the association with all-cause mortality also weakened and became nonsignificant (HR [95% CI]: 1.13 [0.98-1.32]; P = 0.10) (Table S3, SDC,


In this study, we found that dialysis vintage and pretransplant NT-proBNP are independently associated with all-cause mortality in KTR during a median follow-up of 12.7 (7.8–15.6) years. Moreover, the association of dialysis vintage with all-cause mortality was strongly dependent on adjustment for NT-proBNP, with a 40% reduction in HR after adjustment. Results for analyses with cardiovascular mortality as outcome were not materially different for NT-proBNP. However, dialysis vintage was not associated with cardiovascular mortality.

Cardiovascular disease is the leading cause of morbidity and mortality in KTR patients. The majority of the KTR patients have 1 or more cardiovascular risk factors which put them at a 50-fold higher risk of experiencing a cardiac event compared with general population.26 Pretransplant dialysis vintage has earlier been found to be significantly associated with survival rate in KTR.6,8 Dialysis patients are more susceptible to cardiac events and mortality due to specific comorbidities through which patients have resulted in ESKD in the first place (eg, diabetes and hypertension).27 Dialysis by itself can put stress on the heart while it is associated with accelerated coronary atherosclerosis, cardiac arrhythmias, acute heart failure, and sudden circulatory death.9,28,29 Dialysis exerts significant stress especially on the cardiovascular system through the progression of cardiac dilation, left ventricular hypertrophy, and left ventricular dysfunction which are caused by intradialytic volume overload, hypertension, and subclinical ischemia.9,28 Dialysis patients who are ineligible for transplantation have in general more comorbidities and are in worse physical condition compared with dialysis patients who are put on the transplantation waiting list. It is well known that dialysis vintage is positively associated with mortality in ESKD patients and, accordingly, preemptive transplantation leads to a considerable improvement in graft and patient survival compared with dialyzed transplantation recipients.3,4

In keeping with previous studies, we reproduced the association between dialysis vintage and higher risk of mortality after transplantation, independent of adjustment for important potential confounders. However, we identified for the first time that this association is largely dependent on pretransplant NT-proBNP levels. NT-proBNP is a peptide normally released in response to cardiac stretch and is clinically used as a biomarker for diagnosing heart failure and for screening of volume overload and left ventricular dysfunction.12,14 Elevated levels of NT-proBNP are often seen in dialysis patients, which is caused by volume overload and interdialytic weight gain.30 Moreover, high NT-proBNP levels are associated with several other factors, such as diminished kidney function, female sex, and increasing age.16,31 Previously, it has been shown that NT-proBNP during dialysis or posttransplantation is a prognostic marker for cardiovascular morbidity and mortality in dialysis patients and KTR.16,17,19 In the current study, NT-proBNP is found to be independently associated with mortality in KTR. In addition, when NT-proBNP was added to the final model of dialysis vintage, the HR of the association of dialysis vintage and all-cause mortality risk was reduced by 40% and lost significance. Thus, making NT-proBNP to be a potential biomarker for the impact of dialysis on the cardiovascular condition of ESKD patients.

The susceptibility for chronic fluid overload and fluid retention as a result of dialysis can differ among patients. One study found that patients with low BMI and low serum albumin levels are more susceptible to chronic fluid overloading.30 Other study showed that younger patients, males, and diabetic patients are more prone to intradialytic volume changes while undergoing hemodialysis.32 Explanation for these differences are not yet known. Different hemodialysis frequency can also have a different impact on the cardiovascular status; a few studies have shown that frequent intensive daily dialysis and nocturnal home hemodialysis might improve blood pressure control and reduce left ventricular mass.33,34 Therefore, dialysis vintage alone might not be a usable prognostic predictor for mortality in ESKD patients and KTR. We suggest that NT-proBNP, as an indirect assessment of the cardiovascular status, could be used as an additional predictor variable for mortality in KTR.

To date, no clear guidelines for cardiovascular screening and surveillance exist for kidney transplant candidates without cardiac symptoms.35 The aim of preoperative cardiac risk evaluation in kidney transplant candidates is to reduce morbidity and mortality of cardiovascular disease and to determine if the transplant candidate has an active cardiac disease, defined as unstable coronary syndromes, decompensated heart failure, significant arrhythmias, and severe valvular heart disease.35 Potential transplantation recipients who exert cardiac symptoms are referred to a cardiologist for further examination, but different guidelines offer different recommendations about the approach of asymptomatic patients.35 Although our study was initially not designed to test the predictive value of NT-proBNP about mortality in KTR, the current study does, however, prove the significant association of pretransplant NT-proBNP in KTR and their risk of mortality after transplantation.

One of the strengths of this study is the long follow-up time and a relatively large cohort. Some limitations need to be addressed. First, data on serum NT-proBNP of 226 patients (24.7%) are missing. Because of these missing values, these cases were not included in the main analyses resulting in 658 cases with NT-proBNP values instead of potential 884 cases. It is unknown why these samples were missing.

Another limitation is that we do not have multiple measurements of NT-proBNP samples. It would be interesting in future studies to assess whether NT-proBNP is usable as a biomarker by following the trend of NT-proBNP from pretransplant evaluation to the actual transplantation and posttransplantation. Previous studies showed a high between-person coefficient of variation of NT-proBNP, which suggests that the use of a single value of NT-proBNP comparing to a reference value may be limited.36,37 Instead, series of NT-proBNP measurements may show large relative change, which may suggest for worsening of the volume status or cardiovascular condition in a patient.37 Furthermore, the lack of other cardiac measurements, such as an ECG, and concentrations of high-sensitive troponin T limit full assessment of the cardiac condition. While we have data about dialysis vintage, we do not have other dialysis factors, such as arteriovenous vascular access, fluctuations in volume status, dialysis frequency, prevalence of anemia, and left ventricular abnormalities, which all might influence the NT-proBNP levels and the cardiovascular system.33,34

For future studies, it would be interesting to include information on the cardiac evaluation of the patient. NT-proBNP could be compared with an evaluation tool for heart function, for example, echocardiography, because it has already been shown that transplantation improves left ventricular ejection fraction and heart failure condition.38 Also, it remains an important issue whether patients with elevated NT-proBNP will still benefit from transplantation compared with staying on dialysis and that our results ask for such studies to be performed.

Finally, other strong markers associated with cardiovascular outcomes in KTR were not available in this study. It would be interesting if NT-proBNP is still an independent prognostic predictor if it was corrected for other known strong prognostic markers, such as fibroblast growth factor 23,39 high-sensitivity troponin T,40 central wave reflection, and aortic stiffness.41


Our study shows that dialysis vintage is a predictor of mortality in KTR and this association might be explained for a considerable part by variation in pretransplant NT-proBNP at the time of transplantation. Future studies are needed to evaluate the potential value of NT-proBNP screening as a check of cardiac patency of patients on the waiting list of kidney transplantation.


This study made use of the infrastructure and framework provided by the TransplantLines Biobank and Cohort study, which is running in the University Medical Center Groningen and which is registered at with number NCT03272841.


1. Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med. 1999; 341:1725–1730
2. Jofré R, López-Gómez JM, Moreno F, et al. Changes in quality of life after renal transplantation. Am J Kidney Dis. 1998; 32:93–100
3. Kasiske BL, Snyder JJ, Matas AJ, et al. Preemptive kidney transplantation: the advantage and the advantaged. J Am Soc Nephrol. 2002; 13:1358–1364
4. Mange KC, Joffe MM, Feldman HI. Effect of the use or nonuse of long-term dialysis on the subsequent survival of renal transplants from living donors. N Engl J Med. 2001; 344:726–731
5. Matas AJ, Smith JM, Skeans MA, et al. OPTN/SRTR 2013 annual data report: kidney. Am J Transplant. 2015; 15Suppl 21–34
6. Cosio FG, Alamir A, Yim S, et al. Patient survival after renal transplantation: i. The impact of dialysis pre-transplant. Kidney Int. 1998; 53:767–772
7. Bleyer AJ, Hartman J, Brannon PC, et al. Characteristics of sudden death in hemodialysis patients. Kidney Int. 2006; 69:2268–2273
8. Resende L, Guerra J, Santana A, et al. Influence of dialysis duration and modality on kidney transplant outcomes. Transplant Proc. 2009; 41:837–839
9. Wanner C, Amann K, Shoji T. The heart and vascular system in dialysis. Lancet. 2016; 388:276–284
10. Hall C. NT-ProBNP: the mechanism behind the marker. J Card Fail. 2005; 115, SupplS81–S83
11. Daniels LB, Maisel AS. Natriuretic peptides. J Am Coll Cardiol. 2007; 50:2357–2368
12. Silver MA, Maisel A, Yancy CW, et al.; BNP Consensus Panel. BNP Consensus Panel 2004: a clinical approach for the diagnostic, prognostic, screening, treatment monitoring, and therapeutic roles of natriuretic peptides in cardiovascular diseases. Congest Heart Fail. 2004; 105 Suppl 31–30
13. Weber M, Hamm C. Role of B-type natriuretic peptide (BNP) and NT-proBNP in clinical routine. Heart. 2006; 92:843–849
14. Felker GM, Petersen JW, Mark DB. Natriuretic peptides in the diagnosis and management of heart failure. CMAJ. 2006; 175:611–617
15. Wang AY, Lai KN. Use of cardiac biomarkers in end-stage renal disease. J Am Soc Nephrol. 2008; 19:1643–1652
16. Sommerer C, Giannitsis E, Schwenger V, et al. Cardiac biomarkers in haemodialysis patients: the prognostic value of amino-terminal pro-B-type natriuretic peptide and cardiac troponin T. Nephron Clin Pract. 2007; 107:c77–c81
17. Oterdoom LH, de Vries AP, van Ree RM, et al. N-terminal pro-B-type natriuretic peptide and mortality in renal transplant recipients versus the general population. Transplantation. 2009; 87:1562–1570
18. Schaub JA, Coca SG, Moledina DG, et al. Amino-terminal pro-B-type natriuretic peptide for diagnosis and prognosis in patients with renal dysfunction: a systematic review and meta-analysis. JACC Heart Fail. 2015; 3:977–989
19. Jarolim P, Claggett BL, Conrad MJ, et al. B-type natriuretic peptide and cardiac troponin I are associated with adverse outcomes in stable kidney transplant recipients. Transplantation. 2017; 101:182–190
20. Bednarek-Skublewska A, Zaluska W, Ksiazek A. The relationship between serum level of N-terminal pro-B-type natriuretic peptide and nutritional status, and inflammation in chronic hemodialysis patients. Clin Nephrol. 2010; 73:14–20
21. Reznichenko A, Snieder H, van den Born J, et al.; REGaTTA (REnal GeneTics TrAnsplantation) Groningen group. CUBN as a novel locus for end-stage renal disease: insights from renal transplantation. PLoS One. 2012; 7:e36512
22. van Hateren KJ, Alkhalaf A, Kleefstra N, et al. Comparison of midregional pro-A-type natriuretic peptide and the N-terminal pro-B-type natriuretic peptide for predicting mortality and cardiovascular events. Clin Chem. 2012; 58:293–297
23. Du Bois D, Du Bois EF. A formula to estimate the approximate surface area if height and weight be known. 1916. Nutrition. 1989; 5:303–311
24. Keyzer CA, de Borst MH, van den Berg E, et al. Calcification propensity and survival among renal transplant recipients. J Am Soc Nephrol. 2016; 27:239–248
25. Anwaruddin S, Lloyd-Jones DM, Baggish A, et al. Renal function, congestive heart failure, and amino-terminal pro-brain natriuretic peptide measurement: results from the proBNP investigation of dyspnea in the emergency department (PRIDE) study. J Am Coll Cardiol. 2006; 47:91–97
26. Ojo AO. Cardiovascular complications after renal transplantation and their prevention. Transplantation. 2006; 82:603–611
27. Saran R, Robinson B, Abbott KC, et al. US renal data system 2016 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis. 2017; 693 Suppl 1A7–A8
28. McIntyre CW. Effects of hemodialysis on cardiac function. Kidney Int. 2009; 76:371–375
29. de Bie MK, van Dam B, Gaasbeek A, et al. The current status of interventions aiming at reducing sudden cardiac death in dialysis patients. Eur Heart J. 2009; 30:1559–1564
30. Antlanger M, Hecking M, Haidinger M, et al. Fluid overload in hemodialysis patients: a cross-sectional study to determine its association with cardiac biomarkers and nutritional status. BMC Nephrol. 2013; 14:266
31. David S, Kümpers P, Seidler V, et al. Diagnostic value of N-terminal pro-B-type natriuretic peptide (NT-proBNP) for left ventricular dysfunction in patients with chronic kidney disease stage 5 on haemodialysis. Nephrol Dial Transplant. 2008; 23:1370–1377
32. Kalantar-Zadeh K, Regidor DL, Kovesdy CP, et al. Fluid retention is associated with cardiovascular mortality in patients undergoing long-term hemodialysis. Circulation. 2009; 119:671–679
33. Chan CT. Cardiovascular effects of home intensive hemodialysis. Adv Chronic Kidney Dis. 2009; 16:173–178
34. McCullough PA, Chan CT, Weinhandl ED, et al. Intensive hemodialysis, left ventricular hypertrophy, and cardiovascular disease. Am J Kidney Dis. 2016; 685S1S5–S14
35. Lentine KL, Costa SP, Weir MR, et al.; American Heart Association Council on the Kidney in Cardiovascular Disease and Council on Peripheral Vascular Disease; American Heart Association; American College of Cardiology Foundation. Cardiac disease evaluation and management among kidney and liver transplantation candidates: a scientific statement from the American Heart Association and the American College of Cardiology Foundation: endorsed by the American Society of Transplant Surgeons, American Society of Transplantation, and National Kidney Foundation. Circulation. 2012; 126:617–663
36. Aakre KM, Røraas T, Petersen PH, et al. Week-to-week biological variation in the N-terminal prohormone of brain natriuretic peptide in hemodialysis patients and healthy individuals. Clin Chem. 2013; 59:1813–1814
37. Fahim MA, Hayen A, Horvath AR, et al. N-terminal pro-B-type natriuretic peptide variability in stable dialysis patients. Clin J Am Soc Nephrol. 2015; 10:620–629
38. Wali RK, Wang GS, Gottlieb SS, et al. Effect of kidney transplantation on left ventricular systolic dysfunction and congestive heart failure in patients with end-stage renal disease. J Am Coll Cardiol. 2005; 45:1051–1060
39. Baia LC, Humalda JK, Vervloet MG, et al.; NIGRAM Consortium. Fibroblast growth factor 23 and cardiovascular mortality after kidney transplantation. Clin J Am Soc Nephrol. 2013; 8:1968–1978
40. Arroyo D, Quiroga B, Panizo N, et al. High-sensitivity troponin T levels in kidney transplant recipients. Transplant Proc. 2012; 44:2545–2547
41. Verbeke F, Maréchal C, Van Laecke S, et al. Aortic stiffness and central wave reflections predict outcome in renal transplant recipients. Hypertension. 2011; 58:833–838

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.