Heart failure (HF) is a severe disorder associated with high morbidity and mortality.1 Renal dysfunction as assessed by creatinine or estimated glomerular filtration rate (eGFR) has been associated with worse outcomes2–4 in patients with HF. However, creatinine is a late marker of renal function and early markers of the disorder are required.

Neutrophil gelatinase-associated lipocalin (NGAL) is an early marker of kidney injury and has been found to be elevated in both plasma and urine in patients with HF.5–7 NGAL is a 25 kDa protein covalently bound to matrix metalloproteinase-9 that was first isolated from neutrophils. It is produced by a wide variety of cells including respiratory and intestinal epithelial cells,8–10 endothelial cells, renal tubular cells and cardiomyocytes.7,11 Due to its association with kidney injury, inflammation and matrix remodeling, NGAL has been proposed as a marker of prognosis in patients with HF. However, data on patients under optimal treatment for chronic HF due to multiple etiologies are scarce. Additionally, the ability of NGAL to predict a decrease in eGFR over time has not been assessed. The aim of this study was to assess the prognostic role of plasma NGAL in patients with chronic HF and its relationship with traditional biomarkers of renal function and also with B-type natriuretic peptide (BNP), a marker of myocardial wall stress.

The mean age of the study population was 61.6±11 years and 33 (54%) were male. Forty-seven (77%) patients were in NYHA functional class I or II, 14 (23%) in class III. Etiologies of HF were ischemic in 12 (19.7%), hypertensive in 32 (52.4%), idiopathic dilated cardiomyopathy (CM) in 10 (16.4%), alcoholic CM in 2 (3.3%), myocarditis in two (3.3%), peripartum CM in two (3.3%), and chemotherapy-induced CM in one (1.6%). Median NGAL in the whole cohort was 119 ng/ml (interquartile range 78.2–315). Baseline characteristics are shown in Table 1. Figure 1 depicts the correlation of NGAL with other biomarkers. As can be seen, there was a positive correlation with creatinine and albuminuria and a negative correlation with eGFR, but no correlation was observed between NGAL and BNP.

Figure 1.

Correlations of plasma neutrophil gelatinase-associated lipocalin with creatinine (a), estimated glomerular filtration rate (b), albuminuria (c), and B-type natriuretic peptide (d). BNP: B-type natriuretic peptide; NGAL: neutrophil gelatinase-associated lipocalin.

(0.14MB).

Fifteen (24.6%) patients reached the primary endpoint. Baseline NGAL was higher in these patients, as depicted in Figure 2. Patients with events were also more likely to have worse renal function at baseline as assessed by creatinine, eGFR and microalbuminuria. Univariate comparison of patients with and without events is shown in Table 2. The best NGAL cutoff as determined by ROC curve analysis was 179 ng/ml, with sensitivity of 73.3%, specificity of 80.4% and area under the curve (AUC) of 0.73 (p=0.007). Event-free survival was lower in patients with NGAL above this cutoff as shown in Figure 3. Figure 4 depicts ROC curves of NGAL and eGFR in the prediction of the primary endpoint. NGAL added to eGFR, with a 14% increase in the AUC. Using Cox proportional hazards models, only NGAL (hazard ratio [HR] 1.0035, 95% confidence interval [CI] 1.0019–1.0052, p<0.0001) and male gender (HR 5.9, 95% CI 1.22–28.6, p=0.028) were independently related to time to first event.

Figure 2.

Plasma neutrophil gelatinase-associated lipocalin levels in patients with and without events in follow-up. NGAL: neutrophil gelatinase-associated lipocalin.

(0.04MB).

Figure 3.

Event-free survival rate in patients below and above the neutrophil gelatinase-associated lipocalin cutoff determined by receiver operating characteristic curve analysis. NGAL: neutrophil gelatinase-associated lipocalin.

(0.05MB).

Figure 4.

Receiver operating characteristic curve analysis of neutrophil gelatinase-associated lipocalin and estimated glomerular filtration rate for predicting cardiovascular mortality, heart failure hospitalization or emergency department visit for heart failure. AUC: area under the curve; eGFR: estimated glomerular filtration rate; NGAL: neutrophil gelatinase-associated lipocalin.

(0.06MB).

In the present study, we found that plasma NGAL is a powerful predictor of long-term outcomes in patients with chronic HF under optimal medical treatment who were managed by HF specialists. NGAL was a strong predictor of clinical outcomes such as cardiovascular death and HF hospitalizations. It was in fact stronger than BNP, which is considered to be the standard HF biomarker, and performed better than conventional measures of renal function such as creatinine, eGFR, and microalbuminuria. Importantly, our data indicate that even in patients without chronic kidney disease or albuminuria, increased NGAL is associated with unfavorable outcomes.

In contrast with creatinine, NGAL is a tubular damage marker that has been found to be elevated early after acute kidney injury (AKI). Dent et al. demonstrated that NGAL was effective in identifying AKI 2–3 hours after cardiopulmonary bypass surgery as compared with 2–3 days for creatinine.12 In patients admitted for HF, Damman et al. demonstrated that admission serum NGAL levels were associated with increased risk of subsequent worsening of renal function.5 Additionally, in patients with acutely decompensated HF, Maisel et al. found that serum NGAL measured at discharge was a better risk predictor of 30-day events than BNP, creatinine or eGFR.13

In patients with chronic HF, few such studies have been carried out. In the GISSI-HF study, urinary NGAL was measured along with other urinary markers of tubular damage. All of them were associated with a poor clinical outcome, even when eGFR was normal.6 In the CORONA study, the ability of plasma NGAL to predict clinical events was assessed in patients with chronic HF.14 Multivariate analysis revealed that NGAL added significant information when adjusting for clinical variables, but was no longer significant after further adjustment for apolipoprotein A-1, C-reactive protein (CRP), eGFR, and N-terminal pro-brain natriuretic peptide (NT-proBNP). However, there are some important differences between the CORONA study and the present study. The former included patients with ischemic systolic HF, while patients with multiple etiologies were included in our study. Besides, the CORONA study included elderly patients.

Interestingly, NGAL seems to predict cardiovascular outcomes in the community. Plasma NGAL was measured in 1393 Rancho Bernardo Study participants without cardiovascular disease (CVD), who were followed for a mean of 11 years. NGAL was a significant predictor of mortality and CVD in this elderly cohort, independently of traditional risk factors and kidney function and in addition to NT-proBNP and CRP.15

Taken together, these results indicate that plasma NGAL provides important prognostic information and is not merely serving as a surrogate measure of renal function. It has been related to inflammation16 and matrix remodeling17 and is expressed at high levels in the heart and in atherosclerotic plaques.18 There are important differences between plasma and urinary NGAL. NGAL mRNA is markedly up-regulated within the distal nephron during ischemic or toxic renal injury, and the secretion of NGAL into the urine seems to comprise the major fraction of urinary NGAL. By contrast, although plasma NGAL levels increase during acute and chronic renal impairment, the kidney seems not to be the only source of circulating NGAL. Kidney injury and other ischemic and inflammatory processes cause up-regulation of NGAL production in other tissues and cells, which probably represents the main source of plasma NGAL.19,20 Whether such differences may result in significant differences in plasma NGAL performance as compared with urinary NGAL is not known. To date, no head-to-head comparison between these two biomarkers has been carried out.

In the present study NGAL correlated with creatinine, eGFR, and albuminuria but not with BNP. In addition, BNP, a standard HF biomarker, was not associated with outcomes. A possible explanation for this is the fact that all patients were on optimal medical treatment for HF for at least six months. BNP has been shown to respond to HF treatment, and some studies in fact report benefits in using natriuretic peptides to guide treatment.21,22 Median BNP in our population was quite low (61 pg/ml), reflecting excellent medical treatment. These results suggest that in chronic HF patients under optimal treatment, BNP is no longer a good risk marker since it has been reduced in response to treatment. In this scenario markers of renal injury and inflammation such as NGAL seem to be superior to wall stress biomarkers.

Two limitations of the present study must be addressed. This is a small, single-center study with a small number of events, which may limit statistical analysis. Therefore, our data need to be validated in larger prospective studies. Secondly, we studied a heterogeneous population, with multiple etiologies for HF. Since this is a small sample, we could not control for issues related to specific etiologies. Despite these limitations, this was a prospective longitudinal study with valuable information regarding prognosis in stable HF patients, using a simple biomarker that can be easily used in daily clinical practice. Studies in a large population addressing the natural course of NGAL in chronic HF and its potential as a treatment target are warranted.

In this study we found that in patients with chronic HF under optimal medical treatment, NGAL correlates with traditional markers of renal function such as creatinine, eGFR, and albuminuria but not with BNP. NGAL was associated with outcomes, independently of clinical variables, conventional measures of renal function, and BNP.

The authors have no conflicts of interest to declare.