Serum NT-proBNP levels are a prognostic predictor in patients with advanced heart failure

Abstracts

OBJECTIVE: To verify if the determination of NT-proBNP values would help predict the prognosis in advanced heart failure (HF) patients. METHODS: One hundred and five subjects with average age of 52.4 years were evaluated, 66.6% of them males. Thirty-three (32.0%) subjects were outpatients and 70 (67.9%) were inpatients (functional class III/IV) admitted to the hospital for cardiac compensation. All patients had left ventricular systolic dysfunction and a mean ejection fraction of 0.29. The NT-proBNP levels were measured in all patients and they were followed-up over a period from 2 to 90 days (average 77 days). A ROC curve was drawn to determine the best cut-off point, as well as the corresponding Kaplan-Meyer survival curves. RESULTS: During the follow-up period, 22 patients died. The average NT-proBNP value of the patients who remained alive was 6,443.67±6,071.62 pg/ml, whereas that of those who died was 14,609.66±12,165.15 pg/ml (p=0.001). The ROC curve identified a cut-off point at 6,000 pg/ml with 77.3% sensitivity (area under the curve: 0.74). The survival curve for values below and above 6,000 pg/ml was significantly different (p=0.002): patients with values below 6,000 pg/ml had a 90.2% 90-day survival, and those patients with values above, a 66% survival. CONCLUSION: Patients with advanced HF, especially those admitted to the hospital for cardiac compensation, had much higher NT-proBNP values, with a two-fold increase among those who died during the follow-up period. Values above 6,000 pg/ml identify the patients most likely to die within 90 days after hospital discharge.

Advanced heart failure; BNP; prognosis; decompensated heart failure

OBJETIVO: Verificar se a dosagem de NT-proBNP seria de auxílio na predição do prognóstico de pacientes com insuficiência cardíaca (IC) avançada. MÉTODOS: Foram estudados 105 pacientes: 33 (32,0%) do ambulatório e 70 (67,9%) em classe funcional III/IV, hospitalizados para compensação cardíaca, com média de idade de 52,4 anos, dos quais, 66,6% homens. Todos tinham disfunção sistólica do ventrículo esquerdo sendo a fração de ejeção média de 0,29. Em todos dosou-se o NT-proBNP e foram acompanhados por um período de 2 a 91 dias (média 77 dias). Construiu-se a curva ROC para determinação do melhor nível de corte e curvas de sobrevida Kaplan-Meyer de acordo com esse nível. RESULTADOS: Durante o período de seguimento, 22 pacientes (20,9%) morreram. O NT-proBNP médio dos pacientes vivos foi de 6.443,67±6.071,62 pg/ml e dos que morreram foi de 14.609,66±12.165,15 pg/ml (p=0,001). A curva ROC identificou nível de corte de 6.000 pg/ml com sensibilidade de 77,3% (área da curva de 0,74). A curva de sobrevida para valores abaixo e acima de 6.000 pg/ml diferiu significantemente (p=0,002) com os pacientes com valores abaixo de 6.000 pg/ml apresentando sobrevida de 90,2% em 90 dias e os pacientes com valores superiores, sobrevida de 66,6%. CONCLUSÃO: Os pacientes. com IC avançada, especialmente os internados para compensação, apresentam valores muito aumentados de NT-proBNP, sendo estes duas vezes mais elevados entre os que morreram no seguimento. Valor acima de 6.000 pg/ml identifica grupo de pacientes com alta probabilidade de morrer em 90 dias após a alta hospitalar.

Insuficiência cardíaca avançada; BNP; prognóstico; insuficiência cardíaca descompensada

ORIGINAL ARTICLE

Serum NT-proBNP levels are a prognostic predictor in patients with advanced heart failure

Antonio Carlos Pereira-Barretto; Mucio Tavares de Oliveira Junior; Célia Cassaro Strunz; Carlos Henrique Del Carlo; Airton Roberto Scipioni; José Antonio Franchini Ramires

Instituto do Coração do Hospital das Clínicas – FMUSP - São Paulo, SP - Brazil

^{Mailing Address}

ABSTRACT

OBJECTIVE: To verify if the determination of NT-proBNP values would help predict the prognosis in advanced heart failure (HF) patients.

METHODS: One hundred and five subjects with average age of 52.4 years were evaluated, 66.6% of them males. Thirty-three (32.0%) subjects were outpatients and 70 (67.9%) were inpatients (functional class III/IV) admitted to the hospital for cardiac compensation. All patients had left ventricular systolic dysfunction and a mean ejection fraction of 0.29. The NT-proBNP levels were measured in all patients and they were followed-up over a period ranging from 2 to 90 days (average 77 days). A ROC curve was drawn to determine the best cut-off point, as well as the corresponding Kaplan-Meyer survival curves.

RESULTS: During the follow-up period, 22 patients died. The average NT-proBNP value of the patients who remained alive was 6,443.67±6,071.62 pg/ml, whereas that of those who died was 14,609.66±12,165.15 pg/ml (p=0.001). The ROC curve identified a cut-off point at 6,000 pg/ml with 77.3% sensitivity (area under the curve: 0.74). The survival curve for values below and above 6,000 pg/ml was significantly different (p=0.002): patients with values below 6,000 pg/ml had a 90.2% 90-day survival, and those patients with values above, a 66% survival.

CONCLUSION: Patients with advanced HF, especially those admitted to the hospital for cardiac compensation, had much higher NT-proBNP values, with a two-fold increase among those who died during the follow-up period. Values above 6,000 pg/ml identify the patients most likely to die within 90 days after hospital discharge.

Key words: Advanced heart failure, BNP, prognosis, decom­pensated heart failure.

Heart failure is known as an impairing condition with a poor prognosis, especially during the advanced stage when patients progress with shorter survival rates and mortality rates higher than 50% during the first year of follow-up^1-4. Early identification of those patients who might experience the worst progression or premature death is vital, as this would allow intervention in an attempt to change the natural course of their disease.

In evaluating heart failure populations the magnitude of clinical manifestations has proved to be a valuable tool for prognosis stratification^5-9. Several studies have reported that less symptomatic patients (NYHA functional classes I and II) have a better course of the disease than those who are more symptomatic (FC III and IV)^4-10. Although symptoms may help stratify prognosis, they are not useful when more homogeneous groups are evaluated, such as FC IV patients admitted for cardiac compensation, since their clinical states are very similar^5,6,8. Because of their symptoms, these patients need hospitalization and form a poor prognosis population, although tests may help identify which ones will have the worst clinical progression.

Several variables have shown this capacity for selection^4-10, such as the degree of heart remodeling, the intensity of neurohormonal stimulation through plasma concentrations of noradrenalin and sodium, and the level of kidney dysfunction.

Recently, the measurement of B-type natriuretic peptide was added to this list of variables because of comparison studies that identified it as the best laboratory variable for the stratification of prognosis^11-14.

In this article, we report on our experience with NT-proBNP measurement in a population consisting mostly of advanced HF patients. Our objective was to verify its role in stratifying these patients regarding the identification of who would be at risk for early death.

METHODS

One hundred and five patients with congestive heart failure and an ejection fraction under 40% were enrolled in the study. One third of the patients were referred from the hospital’s cardiac outpatient unit, and two thirds were patients who had been admitted for cardiac compensation. Most of them were classified as III/IV patients (68.6%). Outpatients were compensated cardiac subjects. Criteria for hospital admission were hypotension, evident anasarca, indication for inotropic therapy, and lack of compensation after having received intravenous medication in the emergency room, which are all characteristics of the advanced form of the disease.

Blood samples were drawn from all patients, either at hospital admission or at the outpatient unit, to determine NT-proBNP concentrations by commercially available immunoassays (Roche Diagnostics, Mannheim, Germany). Patients were treated by their physicians and followed as to the risk of mortality during an average period of 77 days (2 to 91 days).

Statistical Analysis - Continuous variables are presented as means ± standard deviations, whereas categorical variables are presented as rates and percentages. T-Student or U-Mann-Whitney tests were used to compare the continuous variables. A ROC ("Receiver Operating Characteristic") curve was drawn to estimate the cut-off point for the natriuretic peptide value (NT-proBNP) as a predictor of mortality. The likelihood of death was estimated by computing the relative risk and a 95% confi dence interval for the variables studied. All tests used were two-tailed and the p<0.05 value was considered statistically significant.

RESULTS

Table 1 displays the main clinical characteristics of the study population.

Ischemic cardiomyopathy was the most frequent etiology observed, followed by the Chagasic and idiopathic forms of the disease.

Patient ages ranged from 18 to 89 years (mean age 52.44±11.98 years); 70 (66.6%) were males and 35 (33.3%) were females. The average left ventricular ejection fraction was 0.29, and the left ventricular diastolic diameter was 6.33 cm.

The NT-proBNP concentration ranged from 70.19 pg/ml to 48,001.93 pg/ml (8,187.86 ± 8,407.90 pg/ml).

During the follow-up period, the 22 (20.9%) patients who died had an average NT-proBNP concentration of 14,609.66±12,165.15 pg/ml, whereas those who were still alive had an average value of 6,443.67±6,071.62 pg/ml (p = 0.001).

The ROC curve identified the 6,000 pg/ml value as the best cut-off point to stratify the population as to risk of mortality. Area under the curve was 0.74 (figure 1).

Patients with NT-proBNP values above 6,000 pg/ml were 3.6 times more likely to die (CI 95 %: 1.4 – 9.0; p=0.003) than those with values below this level (figure 2).

No patient with NT-proBNP values under 2,000 pg/ml (21 patients) died during the 90-day follow-up period.

DISCUSSION

Heart failure is considered an increasingly prevalent modern epidemic despite the new drugs and procedures currently available^1-3. In spite of the advances in recent years, HF patients still are at a higher risk of mortality than patients with some types of cancer¹.

The severe symptomatology of FC III or IV patients, added to the presence of cardiogenic shock and the need for hospitalization for cardiac compensation, all identify those who will have the worst disease progression and the highest mortality risk⁴. However, even in a population of patients with the most severe form of the disease, not all of them will die; hence, the early identification of those with the worst clinical progression would allow early referrals to special treatments or transplantation and, hopefully, change the bleak outlook of the disease⁹ .

The search for markers of the worst prognosis is universal, especially by health units that treat heart failure. Cardiac impairment identified by an ejection fraction of less than 25% (measured by radioisotopic ventriculography) and oxygen uptake lower than 12 ml/kg/min, along with being symptomatic, represent criteria used by HF units to indicate heart transplantation¹⁵.

BNP determination should soon be added to these variables, since several studies have documented its value for prognostic stratification.^(11-14) The high levels detected at the first measurement indicate that patients are decompensated, and the maintenance of high levels identifies which patients have a very poor prognosis¹¹.

The introduction of BNP among the variables that indicate prognosis is based on comparison studies conducted with other variables that have shown that this measurement is superior to the evaluation of symptoms, ejection fractions, and even oxygen uptake¹².

The values that predict the worst course of the disease vary according to the severity of the cases in the study population and the measurement method used. Levels of B-type natriuretic peptide can be evaluated by measuring BNP or NT-proBNP. Studies comparing techniques have shown that there is a good equivalence between both methods, although NT-proBNP values in pg/ml are approximately 8-fold higher than those of BNP¹⁶.

For FC III or IV patients, the values identified as mortality predictors are always over 1,000 pg/ml. Bittencourt found 4,137 pg/ml, Logeart 1,015 pg/ml, Hartman 1,767 pg/ml, and Gardner, 1,490 pg/ml (mean values)^11-14.

We analyzed a population with advanced HF and found average values of 7,433.70 pg/ml, characterizing a population with a very severe form of the disease.

By analyzing the values of patients who died, we found results of up to 14,609.66 pg/ml, i.e., approximately twice as high as those of patients who were still alive at the end of the follow-up. This result was similar to that reported by Gardner, who also found values twice as high as those of patients with the worst course of the disease, although the levels he found were lower than ours since he studied a population of less severe cases (3,052 pg/ml for those who died and 1,222 pg/ml for those who lived)¹².

Using the ROC curve, we identified 6,000 pg/ml as the best cut-off point for stratifying the population as to the risk of death. Ninety point-three percent of the patients with values below this level were alive after one year of follow-up, against 72.1% of those with higher values.

We also observed that no patient with NT-proBNP values below 2,000 pg/ml died over the 90-day follow-up period.

Literature data document the usefulness of BNP determination for prognosis stratification. The cut-off point varies according to the severity of the cases in the population studied. The cut-off point in our study was quite high, since our population consisted of patients with extremely severe conditions in whom cardiac compensation was difficult to achieve. This is a population that merits attention, with many potential candidates for cardiac transplantation. We consider values above 6,000 pg/ml strong indicators of the need to ponder this possibility.

Knowledge of NT-proBNP levels allowed us to foresee which patients would have the worst disease progression. NT-proBNP determination is an easily performed test that proved to be an excellent prognosis marker for advanced HF patients.

Potencial Conflict of Interest

No potential conflict of interest relevant to this article was reported.

REFERENCES

Received on 08/05/05

Accepted on 03/17/06

• 1. McMurray JJV, Stewart S. The burden of heart failure. Eur Heart J. 2002; 4 (suppl D): 50-8.
• 2. Fonarow GC, Adams KF, Abraham WT, et al. Risk stratification for in-hospital mortality in acutely decompensated heart failure. Classification and regression tree analysis. JAMA. 2005; 293: 572-80.
• 3. Shahar E, Lee S, Kim J, et al. Hospitalized heart failure: rates and long-term mortality. J Cardiac Fail. 2004; 10: 374-8.
• 4. Wong PS, Davidsson GK, Timeyin J, et al. Heart failure in patients admittd to hospital: mortality is still high. Eur J Intern Med. 2002; 13: 304-10.
• 5. Kearney MT, Foxx KAA, Lee AJ, et al. Predicting death due to progressive heart failure in patients with mild to moderate chronic heart failure. J Am Coll Cardiol. 2002; 40: 1801-8.
• 6. Brophy JM, Dagenais GR, McSherry F, et al. A multivariate model for predicting mortality in patients with heart failure and systolic dysfunction. Am J Med. 2004; 116: 300-304.
• 7. Felker GM, Leiberger JD, Califf RM, et al. Risk stratification after hospitalization for decompensated heart failure. J Cardiac Fail. 2004; 10: 460-6.
• 8. VillaCorta H, Mesquita ET, Cardoso R, et al. Preditores de sobrevida obtidos na unidade de emergência em pacientes atendidos por insuficiência cardíaca descompensada. Rev Port Cardiol. 2003; 22: 495-507.
• 9. Oliveira Jr MT, Canesin MF, Munhoz RT, et al. Principais características clínicas de pacientes que sobrevivem 24 meses ou mais após uma hospitalização devido a descompensação cardíaca. Arq Bras Cardiol. 2005; 84 (2): 161-6
• 10. Bettencourt P, Azevedo A, Pimenta J, et al. N-Terminal-ProBrain Natriuretic peptide predicts outcome after hospital discharge in heart failure patients. Circulation. 2004; 110: 2168-74.
• 11. Gardner RS, Ozalp F, Murday AJ, et al. N-Terminal pro-brain natriuretic peptide. A new gold standard in predicting mortality in patients with advanced heart failure. Eur Heart J. 2003; 24: 1735-43.
• 12. Hartmann F, Packer M, Coats AJS, et al. Prognostic impact of plasma N-Terminal pro-brain natriuretic peptide in severe chronic congestive heart failure. A sub study of the carvedilol prospective randomized cumulative survival (COPERNICUS) trial. Circulation. 2004; 110: 1780-6.
• 13. Logeart D, Thabut G, Jourdain P, et al. Predischarge B-type natriuretic peptide assay for identifying patients at high risk of re-admission after decompensated heart failure. J Am Coll Cardiol. 2004; 43: 635-41.
• 14. II Diretrizes da Sociedade Brasileira de Cardiologia para o diagnóstico e tratamento da insuficiência cardíaca. Arq Bras Cardiol. 1999; 72: 1-30.

Publication Dates

• Publication in this collection
  28 Sept 2006
• Date of issue
  Aug 2006

History

• Received
  05 Aug 2005
• Accepted
  17 Mar 2006