The prognostic value of brain natriuretic peptide (BNP) in non-cardiac patients with sepsis, ultra-long follow-up

Highlights

• An observational study, investigating BNP as a prognostic factor in patients with sepsis, without congestive heart failure.

• BNP is a good independent prognostic value for in-hospital mortality, 90-days mortality and long term survival.

• BNP is with more accurate prognostic value than SOFA for long term mortality.

• In patients with sepsis and high BNP levels, a close follow up should be considered.

Abstract

Objectives

Sepsis is a multifactorial syndrome with increasing incidence of morbidity and mortality. Identification of outcome predictors is therefore essential. Recently, elevated brain natriuretic peptide (BNP) levels have been observed in patients with septic shock. Little information is available concerning BNP levels in patients with critical illness, especially with sepsis. Therefore, this study aims to evaluate the role of BNP as a biomarker for long-term mortality in patients with sepsis.

Methods

We studied 259 patients with sepsis and absence of heart failure. BNP levels were obtained for all patients. A long-term survival follow-up was done, and survival was evaluated 90 days after admission, and during the subsequent 60 months of follow-up.

Results

Eighty-two patients died during the 90-day follow-up (31.7%), 53 died in the index hospitalization (20.5%).

On multivariate analysis models, elevated values of BNP were a strong predictor of in-hospital mortality, 90-day and 60-month mortality in patients with sepsis. BNP was a better prognostic predictor than the Sepsis-related Organ Failure Assessment (SOFA) score for 90-day mortality, and a better predictor for 60-month mortality in low risk groups.

Conclusion

In the population of hospitalized patients with sepsis, BNP is a strong independent predictor of short- and long-term mortality.

Introduction

Sepsis is a syndrome induced by infection and involving physiological, pathological, and biochemical abnormalities. It is now defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection [1].

This response is characterized by an uncontrolled release of cytokines and pro-inflammatory interleukins, causing extensive tissue and end-organ damage [2].

The severity of sepsis can range from mild to life-threatening. It leads to high rates of morbidity and mortality [2], and requires aggressive management that includes hemodynamic monitoring and appropriate antibiotic coverage [3]. However, about 50% of the patients who suffered from severe sepsis were admitted with a milder degree of sepsis [2]. Since their clinical presentation did not predict such deterioration [4], it is crucial that biomarkers predicting the prognosis of patients with sepsis be found.

Several severity scoring systems have been developed for use in intensive care units (ICU). They include the Acute Physiology and Chronic Health Evaluation (APACHE) and its modifications (APACHE II, III and IV), the Simplified Acute Physiology Score (SAPS) and SAP II and the Mortality Probability Model (MPM). The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3), on the other hand, recommends the use of the SOFA score for predicting in-hospital mortality in patients with sepsis. These scoring systems tend to be complex, so that their use may prove to be impractical, especially in emergency departments (ED).

BNP and Pro-BNP are released from cardiomyocytes, as a result of tension over the ventricular walls, secondary to volume or pressure overload, ischemia, necrosis, or remodeling [5], [6], [7]. The diagnostic [8], [9], [10], [11] and prognostic [12], [13], [14], [15] roles of BNP and N-terminal Pro-BNP in patients with dyspnea, suspected heart failure, or diagnosed heart failure with suspected exacerbation are well established.

The high BNP levels observed in patients with septic shock are not fully understood [15], [16], [17], [18]. Moreover, the use of BNP in sepsis is controversial; while several studies have demonstrated an inverse relationship between BNP levels and outcome [19], [20], [21], [22], [23], [24], others doubted the clinical significance of BNP measurement in sepsis [25], [26]. In a meta-analysis that included 12 prospective studies, Wang et al. recently suggested that an elevated BNP or NT-proBNP level may prove to be a powerful predictor of in-hospital and 28-day mortality in septic patients [27]. Nevertheless, the usefulness of BNP as a long-term predictor of mortality has not been studied.

Therefore, our primary aim was to investigate the association between BNP levels and 90-day all-cause mortality and mortality during a 60-month follow-up in patients with sepsis without known heart failure. The secondary endpoint was all-cause in-hospital mortality.