Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Aug 1.
Published in final edited form as: Br J Haematol. 2011 Jun 21;154(4):512–520. doi: 10.1111/j.1365-2141.2011.08777.x

NT-pro Brain Natriuretic Peptide Levels and the Risk of Death in the Cooperative Study of Sickle Cell Disease

Roberto F Machado 1, Mariana Hildesheim 2,3, Laurel Mendelsohn 2,3, Alan T Remaley 3, Gregory J Kato 2,3, Mark T Gladwin 4,5
PMCID: PMC3206726  NIHMSID: NIHMS300175  PMID: 21689089

Abstract

Epidemiological studies support a hypothesis that pulmonary hypertension (PH) is a common complication of sickle cell disease (SCD) that is associated with a high risk of death and evolves as a complication of haemolytic anaemia. This fundamental hypothesis has been recently challenged and remains controversial. In order to further test this hypothesis in a large and independent cohort of SCD patients we obtained plasma samples from the Cooperative Study of Sickle Cell Disease (CSSCD) for analysis of a biomarker, N-terminal-pro brain natriuretic peptide (NT-proBNP), which is elevated in the setting of pulmonary arterial and venous hypertension. A NT-pro-BNP value previously identified to predict PH in adults with SCD was used to determine the association between the risk of mortality in 758 CSSCD participants (428 children and 330 adults). An abnormally high NT-proBNP level ≥160 ng/l was present in 27.6 % of adult SCD patients. High levels were associated with markers of haemolytic anaemia, such as low haemoglobin level (P<0.001), high lactate dehydrogenase (P<0.001), and high total bilirubin levels (P<0.007). A NT-proBNP level ≥160 ng/l was an independent predictor of mortality (RR 6.24, 95% CI 2.9–13.3, P<0.0001). These findings provide further support for an association between haemolytic anaemia and cardiovascular complications in this patient population.

Keywords: Sickle cell disease, pulmonary hypertension, brain natriuretic peptide, biomarkers, survival

INTRODUCTION

Sickle cell disease (SCD) is amongst the most common monogenetic diseases worldwide. (Weatherall and Clegg 2001) Cardiovascular complications, such as pulmonary hypertension (PH), are common in adults with SCD and proposed as a major cause of morbidity and mortality in these patients. (Ataga, et al 2006, De Castro, et al 2008, Gladwin, et al 2004) In population studies, PH has been evaluated by non-invasive transthoracic Doppler echocardiography, measuring the tricuspid regurgitant jet velocity (TRV) to estimate the pulmonary artery systolic pressure. While the TRV is a useful non-invasive screening tool for PH, definitive diagnosis requires right heart catheterization. In all epidemiological studies conducted to date, Doppler-estimated right ventricular systolic pressure >2 SD above the mean was common in adults with SCD (approximate 30% prevalence) and was associated with a 9.24–15.9 risk ratio for early death.(Ataga, et al 2006, De Castro, et al 2008, Gladwin, et al 2004) Similarly, the levels of N-terminal-pro brain natriuretic peptide (NT-proBNP), a biomarker released from the right or left ventricle under pressure stress, are elevated in 30% of adult SCD patients and a value above 160 ng/l in the National Institutes of Health (NIH) and multi-centre study of hydroxycarbamide (MSH) cohorts prospectively identifies a subgroup with more severe haemolytic anaemia and an increased prospective mortality risk (relative risk [RR] 5.1; 95% confidence interval [CI], 2.1–12.5; P<0.001; 19.5% absolute increase in risk of death).(Machado, et al 2006)

A major risk factor for the development of PH is chronic haemolytic anaemia. Despite the robust basic and epidemiological data suggesting that an increased TRV is common and associated with high mortality in patients with SCD (Ataga, et al 2006, Dasgupta, et al 2010, De Castro, et al 2008, Frei, et al 2008, Gladwin, et al 2004, Hill, et al 2010, Hsu, et al 2007, Kato, et al 2006, Kaul, et al 2004, Machado, et al 2006, Meyer, et al 2010, Minniti, et al 2009, Naoman, et al 2009, Nolan, et al 2006, Nolan, et al 2005, Onyekwere, et al 2008, Reiter, et al 2002, Voskaridou, et al 2007, Yeo, et al 2007, Yeo, et al 2009), the importance of this complication and the mechanistic link to haemolytic anaemia has been recently challenged.(Bunn, et al, 2010) In order to further test our data-driven hypothesis we obtained frozen plasma samples from the Cooperative Study of Sickle Cell Disease (CSSCD), for analysis of a biomarker, NT-proBNP, which is elevated in the setting of pulmonary arterial and venous hypertension or left-sided heart disease. The CSSCD was a prospective multi-centre registry study of 3764 patients with SCD.(Platt, et al 1994) This is the largest cohort of patients with SCD ever assembled with the longest median follow-up. While in the CSSCD population clinical events, such as acute chest syndrome, seizures and renal failure, and laboratory parameters, such as high white blood cell count and low fetal haemoglobin level, predicted early mortality, at the time of the study the presence and associated risk of PH was not appreciated.

Because NT-proBNP is stable in frozen plasma, the measurement of this biomarker enables an indirect but prospective assessment of the prevalence of and associated risk factors for developing PH 30 years ago in the SCD population and allowed us to test the hypothesis that patients with high levels of NT-pro BNP would be at a high risk of death.

METHODS

Study Patients

The design of the CSSCD is described elsewhere.(Gaston and Rosse 1982, Platt, et al 1994) The CSSCD was a prospective study of the clinical course of SCD in which more than 3764 patients were enrolled from birth to 66 years of age, at 23 clinical centres throughout the continental United States. Patients were enrolled at these centres between September 1978 and 1988 and were seen at regular intervals for laboratory evaluation and physical examination. Consent was obtained from all patients, their parents, or their legal guardians. The laboratory results analysed in this report were obtained at routine visits, not during acute illness. All acute and chronic complications were documented at the centres. Deaths were reported on a form that was completed by the centre investigator, regardless of whether deaths occurred at the study centre or elsewhere. The patient population was selected from CSSCD study subjects enrolled between September 1978 and 1988 who participated in Phase 2 (paediatric cohort, n=467) or Phase 2a (adult cohort, n=395), and was based on the availability of one or more serum samples stored at the study repository. A total of 758 patients with SCD (428 or 91.6% of patients in the paediatric cohort and 330 or 83.5% of patients in the adult cohort) who had at least one NT-proBNP level measured between 1989 and 1998 were selected for our analysis. A total of 1,269 samples were available for the analysis (Figure 1).

Figure 1. Plasma NT-proBNP Assessment in the Cooperative Study of Sickle Cell Disease.

Figure 1

Open in a new tab

The patient population was selected from CSSCD study subjects enrolled between September 1978 and 1988 who participated in Phase 2 (paediatric cohort, n=467) or Phase 2a (adult cohort, n=395), and was based on the availability of one or more serum samples stored at the study repository. A total of 758 patients with sickle cell disease (428 or 91.6% of patients in the paediatric cohort and 330 or 83.5% of patients in the adult cohort) who had at least one NT-proBNP level measured between 1989 and 1998 were selected for our analysis (w/ BNP). A total of 1,269 samples were available for the analysis.

1Age at entry: < 6 months (infants), 6 months-9 years (children), 10–19 (adolescents), = 20 years (adults)

2Includes samples with unknown sampling dates: n=114 (15%) for infant cohort, n=73 (14%) for adult cohort

N- terminal pro Brain Natriuretic Peptide (NT-pro-BNP) Measurement

Plasma NT–proBNP was measured by a sandwich immunoassay using polyclonal antibodies that recognize epitopes located in the N-terminal segment (1 to 76) of pro-BNP (1 to 108) (Elecsys analyser, Roche Diagnostics, Manheim, Germany). These samples were originally frozen and immediately stored at −−80 °C. The intra-assay and inter-assay coefficients of variation were 1.3%and 4.8%, respectively, and measurements have been previously performed in samples stored at −80°C for as long as15 years. (Kragelund, et al 2005)

Statistical Analysis

Results are presented as median and interquartile range (25th to 75th percentiles), 95% confidence interval (CI), or as a percentage of participants with characteristic, as appropriate. Two-sample Wilcoxon rank-sum tests were used to compare the medians of continuous variables in 2 groups.

An abnormal NT-proBNP level was prospectively defined as ≥ 160 ng/l in the adult cohort as this value has been associated with an increased risk of death in adults with SCD. Given that normal NT-proBNP levels are higher in the paediatric population than in adults, we used a more conservative threshold to define a high NT-proBNP level to be greater than or equal to the age-specific 75th percentile for the paediatric cohort.(Albers, et al 2006, Koch and Singer 2003, Koch, et al 2006, Mir, et al 2006, Mir, et al 2002, Nir, et al 2009) The primary analysis of the relationship between NT-proBNP level and mortality was performed in patients with a first sample with known draw date (n=691) taken before the event of interest (death). Follow-up time for this primary analysis was measured from that sample draw date to the event of interest or the end of follow-up.

Proportional hazards (Cox) regression was used to study relationships between mortality and NT-proBNP levels. Patients were censored at the point of their last contact with study staff if they did not have an event. The regression coefficients were tested for significant difference from zero by a likelihood ratio test. The risk ratio (hazard ratio) and 95% confidence interval (CI) for each predictor are shown. For dichotomous variables, the risk ratio is the ratio of the hazard rates from the proportional hazards model for the two levels of the predictor. For continuous variables, the risk ratio is the ratio of the estimated hazard rate for the 75th percentile compared to the rate for the 25th percentile, calculated as e coefficient x (75th percentile – 25th percentile). This estimate was chosen to provide a comparison among risk ratios for different continuous variables. Kaplan-Meier survival curves were calculated where indicated. Since this was a registry study, prospective power analysis was not performed.

Logistic models were used to investigate associations of a variety of factors with NT-proBNP. Predictors could be either dichotomous or continuous. Goodness of fit of the model was assessed using the Hosmer-Lemeshow statistic (Hosmer and Lemeshow 1989) and using a generalized coefficient of determination (R2). (Cox and Snell 1989, Magge 1990, Nagelkerke 1991) The overall significance of the model (i.e., that at least one predictor was significantly different from zero) was tested with a likelihood ratio test. The significance of a single predictor was tested as a Wald Chi-Square statistic (computed as the square of the ratio of the estimated coefficient to the standard error of the coefficient) compared to a chi-square distribution with one degree of freedom. All analyses were performed using SAS (Version 9.1, SAS Institute, Inc., Cary, NC), Stata (Version 9.0) or GraphPad Prism (Version 4.0, GraphPad Software, San Diego, CA). P-values < 0.05 were considered statistically significant.

RESULTS

The clinical characteristics of the patients evaluated in the study are shown in Table 1. Patients in the paediatric cohort had higher white blood cell counts and higher lactate dehydrogenase (LDH) and aspartate aminotransferase levels than patients in the adult cohort. Further, consistent with a higher burden of chronic organ dysfunction and iron overload, patients in the adult cohort had higher creatinine, total bilirubin, ferritin and uric acid levels than patients in the paediatric cohort.

Table 1.

Characteristics of Study Participants

Paediatric Cohort Adult Cohort Overall
Characteristic N Median (IQR) N Median (IQR) P* N Median (IQR)
Age, years 428 4 (2–6) 330 36 (33–42) <0.001 758 8 (4–35)
Female, N (%) 428 198 (46.3) 330 210 (63.6) <0.001 758 408 (53.8)
Haemoglobin, g/l 427 91 (77–104) 330 91 (79–108) 0.2 757 91 (78–105)
White blood cell count, × 109/l 427 11.4 (8.3–14.8) 330 9.9 (7.7–12.8) <0.001 757 10.6 (8.0–13.8)
Reticulocyte count, × 109/l 419 218.4 (115.2–380.4) 330 236.4 (134–362.7) 0.4 749 225.4 (127.4–369.8)
Fetal haemoglobin, % 417 5.7 (2.6–10.2) 264 2.8 (1.5–5.9) <0.001 681 4.5 (2.0–8.8)
Creatinine, μmol/l 411 35.4 (27–44) 330 70.7 (62–88) <0.001 741 53 (35–71)
Blood urea nitrogen, μmol/l 307 2.5 (1.7–3.6) 330 3.2 (2.1–4.2) <0.001 637 2.9 (2.1–3.9)
Lactate dehydrogenase, u/l 126 402 (290–522) 328 331.5 (220–484) 0.001 454 352 (239–494)
Alanine aminotransferase, u/l 305 22 (14–29) 329 21 (14–43) 0.09 634 22 (14–33)
Aspartate aminotransferase, u/l 406 47 (34–64) 330 43 (27–65) 0.01 736 45.5 (31–64)
Alkaline phosphatase, u/l 401 193 (153–233) 330 84.5 (62–118) <0.001 731 148 (85–207)
Total bilirubin, μmol/l 405 27 (15–44) 330 32 (21–51) <0.001 735 29 (17–46)
Ferritin, μg/l 428 159 (89–286) 330 431 (143–1012) <0.001 758 210 (102–514)
Uric acid, μmol/l 271 262 (208–309) 330 363 (274–440) <0.001 601 297 (238–375)
Open in a new tab
•

From Wilcoxon or z-test between paediatric and adult cohorts.

Prevalence and Incidence of High NT-proBNP Levels

The prevalence of an abnormal NT-proBNP level ≥ 160 ng/l was 27.6 % in patients in the adult CSSCD cohort, which is consistent with previous reports.(Machado, et al 2006) Similarly, the prevalence of a NT-proBNP level ≥ 160 ng/l was 27.4 % in the paediatric cohort, which is at least partly explained by known higher normal NT-proBNP levels in children, especially during the first year of life (Figure 2).(Albers, et al 2006, Koch and Singer 2003, Koch, et al 2006, Mir, et al 2006, Mir, et al 2002, Nir, et al 2009) The incidence of the development of a high NT-proBNP level in subjects with a normal baseline level was 6 % over a mean follow-up time of 5.9 years (incidence rate per 100-person years of 1.03) in the paediatric cohort and 16.5 % over a mean follow-up time of 1.9 years in the adult cohort (incidence rate per 100-person years of 8.59).

Figure 2. Distribution of Patients with a High NT-proBNP Level ≥ 160 ng/l According to Age.

Figure 2

Open in a new tab

The prevalence of a high NT-proBNP level increases with age in adults. In the paediatric cohort this trend does not occur given the physiologically higher normal NT-proBNP levels in children, especially during the first year of life.

Association Between NT-proBNP Levels and Clinical and Biochemical Markers Provide Insight into the Mechanisms for Elevated NT-proBNP in Patients with SCD

In the paediatric cohort, univariate associations with a high NT-proBNP level included (Table 2) a low haemoglobin level (P <0.001), high reticulocyte count (P <0.001), high LDH (P = 0.001), high aspartate aminotransferase (P = 0.04), and high total bilirubin levels (P < 0.001) suggesting an association between NT-proBNP and the intensity of haemolytic anaemia. High NT-proBNP levels in the paediatric group were also associated with a high white blood cell count (P = 0.005), high ferritin level (P = 0.001) and a history of acute chest syndrome (P = 0.001). Similarly, in the adult cohort high NT-proBNP levels were also associated with markers of haemolytic anaemia, such as low haemoglobin level (P <0.001), high LDH (P < 0.001), and high total bilirubin levels (P < 0.007). The lack of correlation between NT-proBNP and reticulocyte count in adults might reflect dysregulated erythropoietin secretion due to renal dysfunction and suppressive effects of transfusions in the most severely affected adult patients that confounds the relationship between haemoglobin and reticulocyte count. Other variables associated with a high NT-proBNP level in the adult cohort included a high white blood cell count (P = 0.004), a higher fetal haemoglobin level (P = 0.02), high blood urea nitrogen (P = 0.007), high alkaline phosphatase (P = 0.006), high ferritin (P = 0.001) and high uric acid levels (P = 0.003). Interestingly, a history of stroke (P < 0.001), sepsis (P = 0.002) or leg ulcers (P = 0.01) were also associated with a high NT-proBNP level in the adult cohort. It is notable that elevations in NT-proBNP show correlations with many of the classically defined mortality risk factors in SCD.

Table 2.

Univariate Associations with NT-proBNP* by Cohort

Paediatric Cohort Adult Cohort
Low BNP High BNP Low BNP High BNP
Characteristic N Median N Median P† N Median N Median P†
Age, years 302 4.0 126 4.0 0.8 221 35 109 37 0.2
Haemoglobin, g/l 302 94 125 83 <0.001 221 97 109 82 <0.001
White blood cell count, × 109/l 302 10.8 125 12.4 .005 221 9.4 109 10.5 0.004
Reticulocyte count, × 109/l 298 187.6 121 279.8 <0.001 221 221.8 109 265.9 0.07
Fetal haemoglobin, % 295 5.9 122 5.5 0.5 174 2.5 90 3.7 0.02
Creatinine, μmol/l 291 35.3 120 35.3 0.1 221 70.7 109 70.7 0.2
Blood urea nitrogen, μmol/l 220 2.9 87 2.5 0.1 221 2.9 109 3.2 0.007
Lactate dehydrogenase, u/l 86 362 40 457.5 0.001 219 300 109 412 <0.001
Alanine aminotransferase, u/l 214 21 91 22 0.4 220 22 109 21 0.2
Aspartate aminotransferase, u/l 288 45.5 118 52 0.04 221 41 109 46 0.3
Alkaline phosphatase, u/l 284 194.5 117 180 0.1 221 81 109 93 0.006
Total bilirubin, μmol/l 288 24 117 36 <0.001 221 29 109 39 0.007
Ferritin, μg/l 302 145 126 199 0.001 221 337 109 665 0.001
Uric acid, μmol/l 193 256 78 274 0.2 221 345 109 375 0.003
Event History N % N % P§ N % N % P§
Pain crisis 240 79.5 107 84.9 0.2 160 72.4 88 80.7 0.1
Acute chest syndrome 177 58.6 95 75.4 0.001 73 33.0 43 39.5 0.3
Seizures 10 3.3 6 4.8 0.5 6 2.7 8 7.3 0.05
Sepsis 63 20.9 31 24.6 0.4 13 5.9 18 16.5 0.002
Leg ulcer 0 0 0 0 -- 28 12.7 26 23.9 0.01
Priapism (males only) 13 7.9 7 10.8 0.5 2 2.3 2 6.3 0.3
Open in a new tab
*

Dichotomized using 160 ng/l (adults) or age-specific 75th percentile (paediatric)

†

From Wilcoxon test comparing medians between low and high BNP groups.

§

From Chi-squared test of independence between event history and BNP.

Table 3 illustrates the results of a multivariate logistic regression analysis of independent clinical and laboratory factors associated with a high NT-proBNP level. In the paediatric cohort these factors included low haemoglobin, high reticulocyte count and total bilirubin levels, while in the adult cohort these factors included increasing age, low haemoglobin level, high blood urea nitrogen level and high LDH level. When data from both cohorts were combined a low haemoglobin level and a high blood urea nitrogen level were independently associated with a high NT-proBNP level.

Table 3.

Logistic Regression Analysis of NT-proBNP* by Cohort

Independent Variable OR† (95% CI) P‡ P§
Paediatric
 Haemoglobin, g/l 0.60 (0.4–0.9) 0.03 0.003
 Reticulocyte count, × 109/l 1.37 (0.9–1.9) 0.08 0.006
 Total bilirubin, μmol/l 1.35 (1.0–1.7) 0.03 0.002
Adult
 Age, years 1.37 (1.0–1.9) 0.05 0.0007
 Haemoglobin, g/l 0.35 (0.2–0.6) <0.0001 <0.0001
 Blood urea nitrogen, μmol/l 1.53 (1.1–2.1) 0.006 <0.0001
 Lactate dehydrogenase (u/l) 1.52 (1.0–2.2) 0.03 0.02
Overall
 Haemoglobin, g/l 0.38 (0.3–0.5) <0.0001 <0.0001
 Blood urea nitrogen, μmol/l 1.31 (1.1–1.6) 0.006 <0.0001
Open in a new tab
*

Dichotomized using 160 ng/l (adults) or age-specific 75th percentile (paediatric)

†

Expressed as 75th percentile relative to 25th percentile. OR is adjusted for the other variables listed in the model

‡

From Wald χ2 test on (estimated coefficient/estimated standard error) from logistic regression model of dichotomized NT-proBNP

§

From Student’s t-test of Ho (estimated coefficient equals zero) from linear regression model of continuous NT-proBNP

Relationship Between NT-proBNP and Mortality

During a median follow up time of 41 months in the overall cohort, 14 deaths (3.3 % of patients) occurred in the paediatric cohort and 27 deaths (8.2 % of patients) occurred in the adult cohort. Table 4 illustrates a proportional hazards regression analysis of the risk of death according to NT-proBNP level in the study cohorts. A high NT-proBNP level ≥ 160 ng/l was a univariate predictor of mortality by proportional hazards regression analysis (RR 6.24, 95% CI 2.9–13.3, P < 0.001) in the whole cohort (Figure 3A). The relationship was also significant when age of death was used in the proportional hazards regression analysis (RR 4.58, 95% CI 2.1–9.8, P < 0.001; Figure 3B). Similarly, in a multivariate model including the cohort-specific risk factors for death (Table 4), the risk ratio for death of patients with a NT-proBNP level ≥ 160 ng/l when compared with those with levels < 160 ng/l was 6.94 (95% CI 3.0–16.2, P < 0.001) when time to death was used in the analysis, and 5.38 (95% CI 2.3–12.7, P < 0.001) when age of death was used in the analysis.

Table 4.

Cox Proportional Hazards Regression Analysis of Mortality Associated with High NT-proBNP

High NT-proBNP Level*
Unadjusted Adjusted†
RR (95% CI) P‡ RR (95% CI) P‡
Paediatric Cohort
 Time to event§ 2.34 (0.8–7.0) 0.1 2.24 (0.8–6.7) 0.1
 Age at event 2.34 (0.8–7.0) 0.1 2.20 (0.7–6.6) 0.2
Adult Cohort
 Time to event§ 9.64 (3.7–24.8) <0.001 7.72 (2.9–20.4) <0.001
 Age at event 5.77 (2.2–14.8) <0.001 6.69 (2.4–18.5) <0.001
Overall Cohort
 Time to event§ 6.24 (2.9–13.3) <0.001 6.87 (3.0–16.0) <0.001
 Age at event 4.58 (2.1–9.8) <0.001 5.42 (2.3–12.9) <0.001
Open in a new tab
*

≥ 75th percentile (age specific) for paediatric cohort; ≥ 160 ng/l for adult cohort

†

Adjusted for risk factors of death: sepsis (paediatric cohort); age, creatinine, white cell count, acute chest syndrome, seizures, stroke (adult cohort)

‡

From Wald χ2 test on estimated coefficient/estimated standard error

§

Months from first sample date for analysis of first sample; months from end of Phase I for analysis of any sample

Figure 3. Kaplan–Meier Survival Curves According to NT-proBNP Level.

Figure 3

Open in a new tab

The survival rate (A) and age of death (B) were significantly higher among patients with a low plasma NT-proBNP level than among those with a high plasma NT-proBNP level (P< 0.001).

As shown in Table 4, the association between a high NT-proBNP level ≥ 160 ng/l and mortality was stronger in the adult than in the paediatric cohort. This association was strengthened across all cohorts when any available NT-proBNP level was used in the analysis, this was probably related to a larger sample size and a greater number of deaths in this larger patient sample.

DISCUSSION

Vascular complications have clearly emerged as a major threat to the well being of individuals with SCD. Here we show that the cardiovascular marker NT-proBNP identifies patients with SCD at high risk of death. Further, our results confirm that NT-proBNP elevation is common in patients with SCD, and is strongly associated with increased the intensity of haemolytic anaemia, age, renal insufficiency and iron overload. These associations are the same as those we have reported associated with high estimated pulmonary artery systolic pressures in adult patients with SCD.(Gladwin, et al 2004) Remarkably, our data suggest that NT-proBNP levels provide additional prognostic information that is independent of that of the traditional risk factors established previously in the CSSCD.

Brain natriuretic peptide (BNP) is a hormone released from the cardiac ventricles in response to cardiomyocyte stretch, and high levels reflect cardiac chamber volume and pressure overload. Synthesized as a prohormone that is split into an active component and an inactive N-terminal fragment (NT-proBNP), BNP produces systemic vasodilation and natriuresis, inhibits the sympathetic nervous system and the renin-angiotensin-aldosterone system and promotes natriuresis.(Hall 2004, Levin, et al 1998) The prognostic importance of this hormone has been demonstrated in several cardiovascular disorders.(Di Angelantonio, et al 2009) Further, we have previously shown in two different well-characterized groups of patients with SCD that a high NT-proBNP level is a major predictor of risk of death in these patients.(Machado, et al 2006) In our study, NT-proBNP levels correlated with markers of pulmonary arterial hypertension and right ventricular dysfunction to a greater extent than markers of left ventricular dysfunction. Taken together with other reports,(Anthi, et al 2007) these data suggest that although the aetiology of PH is likely to be multifactorial, pulmonary vascular disease is the predominant pathophysiological process responsible for the PH observed in patients with SCD. Furthermore, in the absence of left ventricular dysfunction and renal insufficiency (Wang and Lai 2008), elevations in NT-proBNP levels are likely to reflect pulmonary vascular disease.

The associations between NT-proBNP levels and certain laboratory and clinical parameters provide insight into the mechanisms associated with the development of cardiovascular complications and an increased risk of death in patients with SCD. Our data suggest that in the adult cohort the number of episodes of vaso-occlusive crisis and the acute chest syndrome are not independently associated with NT-proBNP elevations. This is consistent with previous observations suggesting that PH, a major NT-proBNP-linked risk factor for death in adults with SCD, is not associated with the rate of vaso-occlusive events. (Gladwin, et al 2004, Machado, et al 2006) Overall, the strongest laboratory associations with NT-proBNP elevations included markers of the severity of haemolytic anaemia, such as low haemoglobin level and high levels of serum LDH and bilirubin, and in the paediatric cohort, a high reticulocyte count. Further, analysis of variables associated with the development of a high NT-proBNP identified markers of severity of haemolysis together with increasing age and markers of renal and hepatic dysfunction as risk factors for the development of an abnormal NT-proBNP level. Our current analysis of CSSCD data is consistent with our previous mechanistic data and models suggesting that the presence of high-grade haemolysis is a major risk factor for death in patients with SCD. These observations are also consistent with the results of a study of 84 patients with sickle cell/β-thalassemia demonstrating that high a NT-proBNP level was associated with increased reticulocyte count and ferritin level. (Voskaridou, et al 2007)

The importance of PH in the sickle cell population, in terms of prevalence and associated morbidity/mortality, and the mechanistic link to haemolytic anaemia has been recently severely challenged by Bunn et al. (2010) Despite the fact that more than 18 cohort studies have consistently associated the severity of haemolytic anaemia with increasing TRV or risk of death, including the NIH-PH cohort,(Gladwin, et al 2004) the Duke cohort,(De Castro, et al 2008) the University of North Carolina cohort,(Ataga, et al 2006) the MSH cohort,(Machado, et al 2006) the PUSH cohort,(Minniti, et al 2009, Naoman, et al 2009, Onyekwere, et al 2008) and a recently published Greek study (Voskaridou, et al. 2010), Bunn et al. (2010) have challenged the results of these larger studies by citing selected underpowered studies. For example, they suggest that De Castro et al. (2008) examined 125 adults and saw no association between LDH and TRV, however this finding was actually in a smaller subgroup of 75 HbSS patients and the haemoglobin levels were significantly lower (7.4 vs. 8.5; P=0.002) in the high TRV group.(De Castro, et al 2008) They cite a study of Pashankar et al. (2008) who examined 62 children and showed significantly higher reticulocytosis suggestive of higher haemolytic rate; follow-up with added patients also showed a strong correlation with low haemoglobin as well (p=0.003).(Pashankar, et al 2009)

While haemolytic anaemia is an independent and contributing risk factor for vasculopathic complications of SCD, such as PH, sudden death, leg ulcers, and priapism, we recognize that this does not occur in isolation. We have duly noted that priapism and leg ulcers occur more frequently in patients with SCD than in patients with other haemolytic diseases such as paroxysmal nocturnal haemoglobinuria, spherocytosis, and pyruvate kinase deficiency, representing the contribution of sickle vaso-occlusion and inflammatory injury to haemolytic injury. This is particularly evident in priapism, where an association with indices of haemolysis and inflammation is present.(Nolan, et al 2005) While not the only mechanism for vascular disease, haemolytic anaemia represents a major independent attributable risk factor.

The performance of NT-proBNP as a marker of increased risk of death was different between the paediatric and the adult cohorts as the association between a high NT-proBNP level and the risk of death was only significant in the adult cohort. These dissimilarities probably reflect the differences between the major causes of death between children and adults with SCD. Multiple paediatric cohort studies suggest that death in children in SCD has been most commonly related to bacterial infection or severe vaso-occlusive events, such as acute chest syndrome and multiorgan failure syndrome.(Gill, et al 1995, Lee, et al 1995, Quinn, et al 2004) In contrast, the risk of death in adulthood appears to be more commonly linked to cardiovascular complications such as PH,(Ataga, et al 2004, Gladwin, et al 2004, Machado, et al 2006, Steinberg, et al 2003) which are more likely to be reflected by changes in NT-proBNP levels. Our study could also be underpowered to detect significant differences in mortality given the small number of deaths in the infant/paediatric cohort. Finally, relatively few children in CSSCD were followed until death, and blood samples were not available from the adolescent cohort, limiting the power of our study to establish childhood risk markers of future mortality.

In conclusion, our results suggest that a high NT-proBNP level is a major risk factor for death in patients with SCD. Our findings also provide further support for a mechanistic link between haemolytic anaemia and the development of cardiovascular complications in this patient population. This adverse prognostic marker might be considered as a potential eligibility criterion for investigational or high-risk interventions, such as haematopoietic stem cell transplantation or gene therapy.

Acknowledgments

This study was supported by National Institutesof Health Intramural Research Funds. Roberto Machado receives grant support from the NHLBI (K23HL098454). Mark T. Gladwin has received research support in the form of a Collaborative Research and Development Agreement between the US Government and Ikaria INO Therapeutics and is listed as a co-inventor on a US Government Patent for the use of nitrite salts for cardiovascular indications. Dr. Gladwin receives grant support from the Institute of Transfusion Medicine, the Hemophilia Center of Western Pennsylvania and Federal funding by the NHLBI, NIDDK and NIAMS of the National Institutes of health (NIH grants R01HL098032, RO1HL096973, RC1DK085852, P30AR058910).

Footnotes

Author Contributions

Study concept and design: Machado, Kato, Gladwin.

Acquisition of data: Machado, Mendelsohn, Gladwin.

Analysis and interpretation of data: Machado, Mendelsohn, Hildesheim, Remaley, Kato, Gladwin.

Drafting of the manuscript: Machado, Hildesheim, Gladwin.

Critical revision of the manuscript for important intellectual content: Machado, Hildesheim, Mendelshon, Kato, Reamley, Gladwin.

Statistical analysis: Hildesheim.

Conflict of Interest

The authors have no conflicts of interest to disclose.

References

  1. Albers S, Mir TS, Haddad M, Laer S. N-Terminal pro-brain natriuretic peptide: normal ranges in the pediatric population including method comparison and interlaboratory variability. Clin Chem Lab Med. 2006;44:80–85. doi: 10.1515/CCLM.2006.016. [DOI] [PubMed] [Google Scholar]
  2. Anthi A, Machado RF, Jison ML, Taveira-Dasilva AM, Rubin LJ, Hunter L, Hunter CJ, Coles W, Nichols J, Avila NA, Sachdev V, Chen CC, Gladwin MT. Hemodynamic and functional assessment of patients with sickle cell disease and pulmonary hypertension. Am J Respir Crit Care Med. 2007;175:1272–1279. doi: 10.1164/rccm.200610-1498OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ataga K, Jones S, Olajide O, Strayhorn D, Lail A, Orringer EP. The Relationship of Pulmonary Hypertension and Survival in Sickle Cell Disease. Blood. 2004;104:463a. [Google Scholar]
  4. Ataga KI, Moore CG, Jones S, Olajide O, Strayhorn D, Hinderliter A, Orringer EP. Pulmonary hypertension in patients with sickle cell disease: a longitudinal study. Br J Haematol. 2006;134:109–115. doi: 10.1111/j.1365-2141.2006.06110.x. [DOI] [PubMed] [Google Scholar]
  5. Bunn HF, Nathan DG, Dover GJ, Hebbel RP, Platt OS, Rosse WF, Ware RE. Pulmonary hypertension and nitric oxide depletion in sickle cell disease. Blood. 2010;116:687–692. doi: 10.1182/blood-2010-02-268193. [DOI] [PubMed] [Google Scholar]
  6. Cox DR, Snell EJ. The analysis of binary data. Chapman and Hall; London: 1989. [Google Scholar]
  7. Dasgupta T, Fabry ME, Kaul DK. Antisickling property of fetal hemoglobin enhances nitric oxide bioavailability and ameliorates organ oxidative stress in transgenic-knockout sickle mice. Am J Physiol Regul Integr Comp Physiol. 2010;298:R394–402. doi: 10.1152/ajpregu.00611.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. De Castro LM, Jonassaint JC, Graham FL, Ashley-Koch A, Telen MJ. Pulmonary hypertension associated with sickle cell disease: clinical and laboratory endpoints and disease outcomes. Am J Hematol. 2008;83:19–25. doi: 10.1002/ajh.21058. [DOI] [PubMed] [Google Scholar]
  9. Di Angelantonio E, Chowdhury R, Sarwar N, Ray KK, Gobin R, Saleheen D, Thompson A, Gudnason V, Sattar N, Danesh J. B-type natriuretic peptides and cardiovascular risk: systematic review and meta-analysis of 40 prospective studies. Circulation. 2009;120:2177–2187. doi: 10.1161/CIRCULATIONAHA.109.884866. [DOI] [PubMed] [Google Scholar]
  10. Frei AC, Guo Y, Jones DW, Pritchard KA, Jr, Fagan KA, Hogg N, Wandersee NJ. Vascular dysfunction in a murine model of severe hemolysis. Blood. 2008;112:398–405. doi: 10.1182/blood-2007-12-126714. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gaston M, Rosse WF. The cooperative study of sickle cell disease: review of study design and objectives. Am J Pediatr Hematol Oncol. 1982;4:197–201. [PubMed] [Google Scholar]
  12. Gill FM, Sleeper LA, Weiner SJ, Brown AK, Bellevue R, Grover R, Pegelow CH, Vichinsky E. Clinical events in the first decade in a cohort of infants with sickle cell disease. Cooperative Study of Sickle Cell Disease. Blood. 1995;86:776–783. [PubMed] [Google Scholar]
  13. Gladwin MT, Sachdev V, Jison ML, Shizukuda Y, Plehn JF, Minter K, Brown B, Coles WA, Nichols JS, Ernst I, Hunter LA, Blackwelder WC, Schechter AN, Rodgers GP, Castro O, Ognibene FP. Pulmonary hypertension as a risk factor for death in patients with sickle cell disease. N Engl J Med. 2004;350:886–895. doi: 10.1056/NEJMoa035477. [DOI] [PubMed] [Google Scholar]
  14. Hall C. Essential biochemistry and physiology of (NT-pro)BNP. Eur J Heart Fail. 2004;6:257–260. doi: 10.1016/j.ejheart.2003.12.015. [DOI] [PubMed] [Google Scholar]
  15. Hill A, Rother RP, Wang X, Morris SM, Jr, Quinn-Senger K, Kelly R, Richards SJ, Bessler M, Bell L, Hillmen P, Gladwin MT. Effect of eculizumab on haemolysis-associated nitric oxide depletion, dyspnoea, and measures of pulmonary hypertension in patients with paroxysmal nocturnal haemoglobinuria. Br J Haematol. 2010;149:414–25. doi: 10.1111/j.1365-2141.2010.08096.x. [DOI] [PubMed] [Google Scholar]
  16. Hosmer DW, Lemeshow S. Applied logistic regression. John Wiley & Sons; New York: 1989. [Google Scholar]
  17. Hsu LL, Champion HC, Campbell-Lee SA, Bivalacqua TJ, Manci EA, Diwan BA, Schimel DM, Cochard AE, Wang X, Schechter AN, Noguchi CT, Gladwin MT. Hemolysis in sickle cell mice causes pulmonary hypertension due to global impairment in nitric oxide bioavailability. Blood. 2007;109:3088–3098. doi: 10.1182/blood-2006-08-039438. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kato GJ, McGowan V, Machado RF, Little JA, Taylor Jt, Morris CR, Nichols JS, Wang X, Poljakovic M, Morris SM, Jr, Gladwin MT. Lactate dehydrogenase as a biomarker of hemolysis-associated nitric oxide resistance, priapism, leg ulceration, pulmonary hypertension, and death in patients with sickle cell disease. Blood. 2006;107:2279–2285. doi: 10.1182/blood-2005-06-2373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaul DK, Liu XD, Chang HY, Nagel RL, Fabry ME. Effect of fetal hemoglobin on microvascular regulation in sickle transgenic-knockout mice. J Clin Invest. 2004;114:1136–1145. doi: 10.1172/JCI21633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart. 2003;89:875–878. doi: 10.1136/heart.89.8.875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Koch AM, Rauh M, Zink S, Singer H. Decreasing ratio of plasma N-terminal pro-B-type natriuretic peptide and B-type natriuretic peptide according to age. Acta Paediatr. 2006;95:805–809. doi: 10.1080/08035250500466647. [DOI] [PubMed] [Google Scholar]
  22. Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-B-type natriuretic peptide and long-term mortality in stable coronary heart disease. N Engl J Med. 2005;352:666–675. doi: 10.1056/NEJMoa042330. [DOI] [PubMed] [Google Scholar]
  23. Lee A, Thomas P, Cupidore L, Serjeant B, Serjeant G. Improved survival in homozygous sickle cell disease: lessons from a cohort study. Bmj. 1995;311:1600–1602. doi: 10.1136/bmj.311.7020.1600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med. 1998;339:321–328. doi: 10.1056/NEJM199807303390507. [DOI] [PubMed] [Google Scholar]
  25. Machado RF, Anthi A, Steinberg MH, Bonds D, Sachdev V, Kato GJ, Taveira-DaSilva AM, Ballas SK, Blackwelder W, Xu X, Hunter L, Barton B, Waclawiw M, Castro O, Gladwin MT. N-terminal pro-brain natriuretic peptide levels and risk of death in sickle cell disease. Jama. 2006;296:310–318. doi: 10.1001/jama.296.3.310. [DOI] [PubMed] [Google Scholar]
  26. Magge L. Measures based on wald and likelihood ratio joint significance tests. American Statistician. 1990;44:250–253. [Google Scholar]
  27. Meyer C, Heiss C, Drexhage C, Kehmeier ES, Balzer J, Muhlfeld A, Merx MW, Lauer T, Kuhl H, Floege J, Kelm M, Rassaf T. Hemodialysis-induced release of hemoglobin limits nitric oxide bioavailability and impairs vascular function. J Am Coll Cardiol. 2010;55:454–459. doi: 10.1016/j.jacc.2009.07.068. [DOI] [PubMed] [Google Scholar]
  28. Minniti CP, Sable C, Campbell A, Rana S, Ensing G, Dham N, Onyekwere O, Nouraie M, Kato GJ, Gladwin MT, Castro OL, Gordeuk VR. Elevated tricuspid regurgitant jet velocity in children and adolescents with sickle cell disease: association with hemolysis and hemoglobin oxygen desaturation. Haematologica. 2009;94:340–347. doi: 10.3324/haematol.13812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mir TS, Flato M, Falkenberg J, Haddad M, Budden R, Weil J, Albers S, Laer S. Plasma concentrations of N-terminal brain natriuretic peptide in healthy children, adolescents, and young adults: effect of age and gender. Pediatr Cardiol. 2006;27:73–77. doi: 10.1007/s00246-005-1022-4. [DOI] [PubMed] [Google Scholar]
  30. Mir TS, Marohn S, Laer S, Eiselt M, Grollmus O, Weil J. Plasma concentrations of N-terminal pro-brain natriuretic peptide in control children from the neonatal to adolescent period and in children with congestive heart failure. Pediatrics. 2002;110:e76. doi: 10.1542/peds.110.6.e76. [DOI] [PubMed] [Google Scholar]
  31. Nagelkerke NJ. A note on a general definition of the coefficient of determination. Biometrika. 1991;78:691–692. [Google Scholar]
  32. Naoman SG, Nouraie M, Castro OL, Nwokolo C, Fadojutimi-Akinsiku M, Diaz S, Willie-Carnegie G, Andrews N, Gordeuk VR. Echocardiographic findings in patients with sickle cell disease. Ann Hematol. 2009;89:61–66. doi: 10.1007/s00277-009-0764-0. [DOI] [PubMed] [Google Scholar]
  33. Nir A, Lindinger A, Rauh M, Bar-Oz B, Laer S, Schwachtgen L, Koch A, Falkenberg J, Mir TS. NT-Pro-B-type Natriuretic Peptide in Infants and Children: Reference Values Based on Combined Data from Four Studies. Pediatr Cardiol. 2009;30:3–8. doi: 10.1007/s00246-008-9258-4. [DOI] [PubMed] [Google Scholar]
  34. Nolan VG, Wyszynski DF, Farrer LA, Steinberg MH. Hemolysis associated priapism in sickle cell disease. Blood. 2005;106:3264–3267. doi: 10.1182/blood-2005-04-1594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nolan VG, Adewoye A, Baldwin C, Wang L, Ma Q, Wyszynski DF, Farrell JJ, Sebastiani P, Farrer LA, Steinberg MH. Sickle cell leg ulcers: associations with haemolysis and SNPs in Klotho, TEK and genes of the TGF-beta/BMP pathway. Br J Haematol. 2006;133:570–578. doi: 10.1111/j.1365-2141.2006.06074.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Onyekwere OC, Campbell A, Teshome M, Onyeagoro S, Sylvan C, Akintilo A, Hutchinson S, Ensing G, Gaskin P, Kato G, Rana S, Kwagyan J, Gordeuk V, Williams J, Castro O. Pulmonary hypertension in children and adolescents with sickle cell disease. Pediatr Cardiol. 2008;29:309–312. doi: 10.1007/s00246-007-9018-x. [DOI] [PubMed] [Google Scholar]
  37. Pashankar FD, Carbonella J, Bazzy-Asaad A, Friedman A. Prevalence and risk factors of elevated pulmonary artery pressures in children with sickle cell disease. Pediatrics. 2008;121:777–782. doi: 10.1542/peds.2007-0730. [DOI] [PubMed] [Google Scholar]
  38. Pashankar FD, Carbonella J, Bazzy-Asaad A, Friedman A. Longitudinal follow up of elevated pulmonary artery pressures in children with sickle cell disease. Br J Haematol. 2009;144:736–741. doi: 10.1111/j.1365-2141.2008.07501.x. [DOI] [PubMed] [Google Scholar]
  39. Platt OS, Brambilla DJ, Rosse WF, Milner PF, Castro O, Steinberg MH, Klug PP. Mortality in sickle cell disease. Life expectancy and risk factors for early death. N Engl J Med. 1994;330:1639–1644. doi: 10.1056/NEJM199406093302303. [DOI] [PubMed] [Google Scholar]
  40. Quinn CT, Rogers ZR, Buchanan GR. Survival of children with sickle cell disease. Blood. 2004;103:4023–4027. doi: 10.1182/blood-2003-11-3758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Reiter CD, Wang X, Tanus-Santos JE, Hogg N, Cannon RO, Schechter AN, Gladwin MT. Cell-free hemoglobin limits nitric oxide bioavailability in sickle-cell disease. Nat Med. 2002;8:1383–1389. doi: 10.1038/nm1202-799. [DOI] [PubMed] [Google Scholar]
  42. Steinberg MH, Barton F, Castro O, Pegelow CH, Ballas SK, Kutlar A, Orringer E, Bellevue R, Olivieri N, Eckman J, Varma M, Ramirez G, Adler B, Smith W, Carlos T, Ataga K, DeCastro L, Bigelow C, Saunthararajah Y, Telfer M, Vichinsky E, Claster S, Shurin S, Bridges K, Waclawiw M, Bonds D, Terrin M. Effect of hydroxyurea on mortality and morbidity in adult sickle cell anemia: risks and benefits up to 9 years of treatment. JAMA. 2003;289:1645–1651. doi: 10.1001/jama.289.13.1645. [DOI] [PubMed] [Google Scholar]
  43. Voskaridou E, Christoulas D, Bilalis A, Plata E, Varvagiannis K, Stamatopoulos G, Sinopoulou K, Balassopoulou A, Loukopoulos D, Terpos E. The effect of prolonged administration of hydroxyurea on morbidity and mortality in adult patients with sickle cell syndromes: results of a 17-year, single-center trial (LaSHS) Blood. 2010;115:2354–2363. doi: 10.1182/blood-2009-05-221333. [DOI] [PubMed] [Google Scholar]
  44. Voskaridou E, Tsetsos G, Tsoutsias A, Spyropoulou E, Christoulas D, Terpos E. Pulmonary hypertension in patients with sickle cell/beta thalassemia: incidence and correlation with serum N-terminal pro-brain natriuretic peptide concentrations. Haematologica. 2007;92:738–743. doi: 10.3324/haematol.11136. [DOI] [PubMed] [Google Scholar]
  45. Wang AY, Lai KN. Use of cardiac biomarkers in end-stage renal disease. J Am Soc Nephrol. 2008;19:1643–1652. doi: 10.1681/ASN.2008010012. [DOI] [PubMed] [Google Scholar]
  46. Weatherall DJ, Clegg JB. Inherited haemoglobin disorders: an increasing global health problem. Bull World Health Organ. 2001;79:704–712. [PMC free article] [PubMed] [Google Scholar]
  47. Yeo TW, Lampah DA, Gitawati R, Tjitra E, Kenangalem E, McNeil YR, Darcy CJ, Granger DL, Weinberg JB, Lopansri BK, Price RN, Duffull SB, Celermajer DS, Anstey NM. Impaired nitric oxide bioavailability and L-arginine reversible endothelial dysfunction in adults with falciparum malaria. J Exp Med. 2007;204:2693–2704. doi: 10.1084/jem.20070819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yeo TW, Lampah DA, Tjitra E, Gitawati R, Kenangalem E, Piera K, Granger DL, Lopansri BK, Weinberg JB, Price RN, Duffull SB, Celermajer DS, Anstey NM. Relationship of cell-free hemoglobin to impaired endothelial nitric oxide bioavailability and perfusion in severe falciparum malaria. J Infect Dis. 2009;200:1522–1529. doi: 10.1086/644641. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES