Author + information
- Received September 7, 2016
- Revision received September 29, 2016
- Accepted October 4, 2016
- Published online January 2, 2017.
- Leong L. Ng, MDa,b,∗ (, )
- Iain B. Squire, MDa,b,
- Donald J.L. Jones, PhDc,
- Thong Huy Cao, MD, PhDa,b,
- Daniel C.S. Chan, BMedSci, BM BSa,b,
- Jatinderpal K. Sandhu, MPhila,b,
- Paulene A. Quinn, MPhila,b,
- Joan E. Davies, PhDa,b,
- Joachim Struck, PhDd,
- Oliver Hartmann, PhDd,
- Andreas Bergmann, PhDd,
- Alexandre Mebazaa, MD, PhDe,
- Etienne Gayat, PhDe,
- Mattia Arrigo, MDe,
- Eiichi Akiyama, MDe,
- Zaid Sabti, MDf,
- Jens Lohrmann, MDf,
- Raphael Twerenbold, MDf,
- Thomas Herrmann, MDf,
- Carmela Schumacher, MScf,
- Nikola Kozhuharov, MDf,
- Christian Mueller, MDf,
- GREAT Network
- aDepartment of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom
- bNIHR Leicester Cardiovascular Biomedical Research Unit, Glenfield Hospital, Leicester, United Kingdom
- cDepartment of Cancer Studies, University of Leicester, Leicester Royal Infirmary, Leicester, United Kingdom
- dSphingotec GmbH, Hennigsdorf, Germany
- eU942 Inserm; APHP, Hôpitaux Universitaire Saint Louis Lariboisière; Université Paris Diderot, Paris, France
- fCardiovascular Research Institute Basel and Department of Cardiology, University Hospital Basel, Basel, Switzerland
- ↵∗Reprint requests and correspondence:
Dr. Leong L. Ng, Department of Cardiovascular Sciences, Clinical Sciences Wing, Glenfield Hospital, Leicester LE3 9QP, United Kingdom.
Background Proenkephalin A (PENK) and its receptors are widely distributed. Enkephalins are cardiodepressive and difficult to measure directly. PENK is a stable surrogate analyte of labile enkephalins that is correlated inversely with renal function. Cardiorenal syndrome is common in acute heart failure (HF) and portends poor prognosis.
Objectives This study assessed the prognostic value of PENK in acute HF, by identifying levels that may be useful in clinical decisions, and evaluated its utility for predicting cardiorenal syndrome.
Methods This multicenter study measured PENK in 1,908 patients with acute HF (1,186 male; mean age 75.66 ± 11.74 years). The primary endpoint was 1-year all-cause mortality; secondary endpoints were in-hospital mortality, all-cause mortality or HF rehospitalization within 1 year, and in-hospital worsening renal function, defined as a rise in plasma creatinine ≥26.5 μmol/l or 50% higher than the admission value within 5 days of presentation.
Results During 1-year follow-up, 518 patients died. Measures of renal function were the major determinants of PENK levels. PENK independently predicted worsening renal function (odds ratio: 1.58; 95% confidence interval [CI]: 1.24 to 2.00; p < 0.0005) with a model receiver-operating characteristic area of 0.69. PENK was associated with the degree of worsening renal function. Multivariable Cox regression models showed that PENK level was an independent predictor of 1-year mortality (p < 0.0005) and 1-year death and/or HF (hazard ratio: 1.27; 95% CI: 1.10 to 1.45; p = 0.001). PENK levels independently predicted outcomes at 3 or 6 months and were independent predictors of in-hospital mortality, predominantly down-classifying risk in survivors when added to clinical scores; levels <133.3 pmol/l and >211.3 pmol/l detected low-risk and high-risk patients, respectively.
Conclusions PENK levels reflect cardiorenal status in acute HF and are prognostic for worsening renal function and in-hospital mortality as well as mortality during follow-up.
In recent years, many advances have been made in the understanding of pathophysiology and the management of chronic heart failure (HF). However, the understanding and treatment of acute HF has remained incomplete and broadly unchanged during this period. Accordingly, prognosis remains poor, with 1-year mortality rate exceeding 25% (1). Neurohormonal activation and worsening renal function (WRF) play important roles in the pathogenesis of fluid redistribution, leading to acute decompensation (2). Use of biomarkers might help characterize different phenotypes in acute HF associated with different outcomes that may prompt specific and expedited therapies. Although activation of the natriuretic peptide system is recognized, its value in predicting death at first presentation with acute HF is suboptimal (3), and better tools are needed.
The endogenous opioids (enkephalins, endorphins, dynorphins), extensively studied in nociception and anesthesia, also have roles in cardiovascular regulation (4). Proenkephalin A (PENK) is widely expressed, and cardiac cells secrete enkephalins, which have local effects on opioid receptors. Cardiodepressive through a negative inotropic effect and lower blood pressure and heart rate (5), opioid receptors, especially the δ receptor that binds enkephalins, are widely distributed, with highest densities in the kidney (6).
The possible relationship between endogenous opioid systems and prognosis was suggested by previous studies. Data from ADHERE (Acute Decompensated Heart Failure National Registry) demonstrated that opiate administration in acute HF has been associated with poor outcomes (7). Fontana et al. (8) reported elevated met-enkephalin levels in severe acute HF compared with less severe acute HF.
In several acute disease conditions, elevated plasma levels of a PENK fragment (amino acids 119 through 159) have been associated with renal dysfunction and poor outcomes. For example, we previously demonstrated PENK to be an independent predictor of major adverse cardiac events, including death, reinfarction, and rehospitalization for HF in patients presenting with acute myocardial infarction (9). This also has been shown more recently for stable ambulatory patients with HF (10). PENK predicts acute kidney injury after cardiac surgical procedures (11) and in patients with sepsis (12), and it has been linked to death and major adverse cerebrocardiovascular events in acute stroke (13).
In the present study, we investigated the relationship of the enkephalin system with WRF and worsening prognosis in acute HF. Renal impairment profoundly influences prognosis in HF (14), and development of acute kidney injury is common in acute HF, the so-called cardiorenal syndrome type 1 (15). We therefore examined the utility of PENK in assessing WRF in acute HF. Previous studies were hindered by the instability of met-enkephalin. we used a more recently developed assay (penKid assay, Sphingotec GmbH, Hennigsdorf, Germany) for PENK (16), with epitopes on the proenkephalin molecule that are stable in whole blood for at least 48 h, thus enabling a study of this system in acute HF. The utility of PENK for prediction of short-term and long-term outcomes and inpatient mortality was examined in combination with various clinical risk scores developed for inpatient mortality, namely ADHERE (17), GWTG-HF (Get With the Guidelines Heart Failure) (18), and OPTIMIZE-HF (Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients With Heart Failure) (19).
Three cohorts of unselected patients with acute HF who presented with acute dyspnea to the emergency department of the participating university hospitals in 3 countries (United Kingdom, France, and Switzerland) were recruited. Acute HF was defined, according to the guidelines of the European Society of Cardiology (20), as progressive worsening or new-onset of shortness of breath, along with clinical signs of pulmonary or peripheral edema and elevated jugular venous pressure requiring intensification of diuretic and/or vasodilator therapy. Inclusion was independent of renal function, although patients with terminal renal failure who were receiving established renal replacement therapy were excluded. These studies complied with the declaration of Helsinki, and ethics approval was granted from the respective research ethics committees. All patients provided written informed consent.
Following signed informed consent, venous blood was withdrawn from recumbent patients and collected in pre-chilled tubes containing ethylenediaminetetraacetic acid as an anticoagulant. The intervals for obtaining this admission sample were up to 4 h (Paris), 1 h (Basel), and 12 h (Leicester). Plasma was stored at −80°C until analysis in a single batch.
Imaging and assays
Transthoracic echocardiography was performed using standard techniques, and left ventricular ejection fraction was calculated using the biplane method of discs formula.
The assay for stable PENK (molecular weight 4,586 Da) was previously described (16), and it has since been modified (9). In brief, 2 mouse monoclonal anti-PENK antibodies were developed by immunization with PENK peptide. Standards or samples (50 μl plasma) were immobilized by the capture antibody (2 μg coated on polystyrene tubes). The detector antibody was labeled with methylacridinium ester, and bound chemiluminescence was measured. The normal range was mean ± SEM of 46.6 ± 14.1 pmol/l, with a median of 45 pmol/l (range 9 to 518 pmol/l) (12).
We used an immunoassay to measure troponin I, which has a coefficient of variation of 10% at 0.03 μg/l with a 99th percentile of 0.04 μg/l. Plasma high-sensitivity troponin T was measured in patients enrolled in Basel. The 99th percentile upper reference limit was 0.014 μg/l. All samples were analyzed in a central laboratory in a blinded manner.
In patients from Leicester, United Kingdom, plasma N-terminal pro–B-type natriuretic peptide (NT-proBNP) was quantified using a sandwich immunoassay as described previously (21). In Paris, plasma brain natriuretic peptide (BNP) was measured using Abbott kits (Abbott Diagnostics, Rungis Cedex, France). The Elecsys NT-proBNP assay (Roche Diagnostics GmbH, Mannheim, Germany) was used in Basel.
The primary endpoint was 1-year all-cause mortality. Secondary endpoints included within-hospital all-cause mortality in the entire cohort and all-cause mortality or HF rehospitalization within 1 year, by using follow-up data for the Leicester and Basel cohorts only. HF rehospitalization was defined as a hospital readmission for which HF was the primary cause, requiring treatment with diuretics, intravenous inotropes, or nitrates. Endpoints were obtained from hospital records and electronic databases. Patients surviving to discharge were followed up for at least 1 year after the initial hospitalization. When patients had multiple events, the time to first event was counted as the censored outcome.
Another endpoint was WRF, defined as a rise in plasma creatinine of ≥26.5 μmol/l or 50% higher than the admission value (whichever was smaller), within 5 days of presentation (14,22). We did not use the definition with urine volumes because administration of diuretics could cause large variations in this parameter. Degree of WRF was analyzed as the absolute increase in plasma creatinine from the admission value and also as an ordinal scale defined as follows: 0 (rise in plasma creatinine <26.5 μmol/l or <1.5-fold); 1 (rise in plasma creatinine of ≥26.5 μmol/l or 1.5- to 2-fold); 2 (>2.0- to 3.0-fold); and 3 (>3.0-fold) (23). The modified diet in renal disease formula was used to obtain the estimated glomerular filtration rate (eGFR).
Statistical analyses were performed using SPSS version 22 (IBM Corp., Armonk, New York) and Stata 13 (Statacorp LP, College Station, Texas) software. Assuming an event rate of 25% and that the covariates predict up to 30% of the biomarker variance, a sample size of 1,000 patients would be powered (99% at p < 0.01) to detect a hazard ratio (HR) of the biomarker of 1.5, using the command stpower cox in Stata 13. All biomarker levels were log10 transformed and normalized to 1 SD increment.
Gaussian data were analyzed using analysis of variance and general linear models, and nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test, and Spearman [rs] correlations) were used against non-Gaussian data. Chi-square tests were used for categorical variables. Independent predictors of PENK levels were assessed using general linear models with coefficients and p values reported for 2,000 bootstrap samples.
Because all 3 cohorts used different natriuretic peptides assays, we combined the data in this composite study by normalizing BNP and NT-proBNP for each center, by using log transformation and calculating the z-transform (thus expressing natriuretic peptides normalized to 1 SD of the log-transformed biomarker for each center) before pooling the z-transformed natriuretic peptides for all centers.
To assess the prognostic value of biomarkers, a “base” model was generated using Cox survival analysis, which included variables that were associated with study outcomes on univariable analysis at p < 0.10 or had been associated with poor outcome in other studies (age, sex, previous history of HF, ischemic heart disease, hypertension or diabetes, plasma urea, sodium, eGFR, hemoglobin, and biomarkers [log NT-proBNP or BNP]). Because different natriuretic peptide assays were used, these were log normalized and then z-transformed (divided by 1 SD increment) before analysis. PENK was added to this base model to evaluate its relative prognostic value with all variables entered simultaneously, with added value assessed using the increment in log-likelihood chi-square test for nested regression models.
To predict WRF, we used logistic, linear, and ordinal regression models containing clinical variables, plasma biomarkers, and use of therapies associated with WRF such as diuretics and angiotensin-converting enzyme inhibitors or angiotensin receptor blockers.
A logistic regression model was used to assess the relative prognostic power of these biomarkers and clinical risk scores (ADHERE, GWTG-HF, and OPTIMIZE-HF) to predict in-hospital all-cause mortality. The probability of an outcome calculated from the logistic base model and the added value of natriuretic peptides and/or PENK was evaluated by reclassification analysis with calculation of category-free net reclassification improvement as described by Pencina et al. (24). A similar approach to assess net reclassification improvement was adopted for prediction of 1-year mortality or the composite of 1-year death and/or HF.
We also constructed classification trees using chi-square automatic interaction detection, which chooses the biomarker at each step that has the strongest interaction with the dependent variable.
According to patient characteristics seen in the cohorts (Leicester, Paris, and Basel) (Table 1), patients at the 3 sites were similar in sex, but varied in age, renal function, body mass index, and comorbidities. The outcomes (death and death or HF at 1 year) were similar among sites. The PENK levels were similar among sites, with no differences in the normalized natriuretic peptide distributions.
The characteristics of the combined Leicester-Paris-Basel cohort are shown in Table 2, according to PENK quartiles. Patients with higher PENK levels were older, had a lower body mass index, were more often female, and had comorbidities such as histories of hypertension, ischemic heart disease, HF, and renal impairment; their initial systolic blood pressures (SBP) and heart rates were also lower. With increasing PENK quartiles, renal function deteriorated, and natriuretic peptide levels increased. Higher PENK also was associated with more frequent prescription of loop diuretics and aldosterone antagonists.
Correlation analysis and effects of changes in proenkephalin A
Spearman analysis (rs, p value) showed that PENK was correlated with age (0.366; p < 0.0005), eGFR (−0.752; p < 0.0005), plasma creatinine (0.668; p < 0.0005), plasma urea (0.641; p < 0.0005), heart rate (−0.165; p < 0.0005), SBP (−0.100; p < 0.0005), troponin T (0.373; p < 0.0005), and z-score of log natriuretic peptide (0.419; p < 0.0005). There were nonsignificant correlations with plasma sodium. A univariate general linear model indicated the following independent predictors of PENK level, in descending order according to variance accounted for in the model (Table 3): eGFR, plasma urea, natriuretic peptide levels, age, sex, past history of renal impairment, SBP, and heart rate. These variables accounted for 60.6% of the variance of PENK levels, and of these, 2 measures of renal function (eGFR and plasma urea) accounted for 46.8% of the model.
Of the 1,714 patients with data on plasma creatinine within 5 days of hospitalization, 264 had developed a rise in plasma creatinine of ≥26.5 μmol/l or 50% higher than the admission value. Using clinical variables, use of nephrotoxic drugs on admission (angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, diuretics), natriuretic peptides, and PENK, the independent predictors of WRF were past history of renal disease, plasma sodium, SBP, and PENK (Figure 1). The receiver-operating characteristic (ROC) area for the full model was 0.69 (95% confidence interval [CI]: 0.65 to 0.73) compared with 0.67 (95% CI: 0.63 to 0.71) for a model without PENK (p value for difference in ROC areas = 0.054).
We also evaluated the relationship of PENK with the severity of development of renal impairment by performing linear regression of these variables and the absolute change in plasma creatinine from the admission level. The significant predictors are listed in Table 4.
Plasma samples taken before and after treatment of the acute episode of HF were available in 1,012 patients of the Leicester and Basel cohorts, with post-treatment levels obtained approximately 5 days following admission. Overall, median levels (interquartile range) before and after treatment were 97.2 pmol/l (66.9 to 146.8 pmol/l) and 98.7 pmol/l (66.6 to 141.8 pmol/l), respectively, (p = NS; Wilcoxon signed rank test). However, in comparing plasma PENK between admission and after therapy for those patients who did or did not develop WRF by using a repeated measures design, there was a significant interaction between the pre-therapy and post-therapy PENK levels and WRF development (p < 0.0005). Patients with WRF showed a higher level of PENK on admission, which increased further after therapy (Figure 2).
To explore whether PENK levels changed before creatinine levels, we calculated PENK-to-creatinine ratios in patients according to WRF status. In patients without WRF, PENK-to-creatinine ratios did not change between admission and post-therapy samples, a finding suggesting that both analytes changed in tandem, whereas in patients with WRF, ratios were initially elevated and fell significantly with time (p < 0.0005). There was a significant interaction between pre-therapy and post-therapy PENK-to-creatinine ratios and WRF development (p < 0.0005) (Figure 2).
During a minimum follow-up of 1 year, there were 518 deaths in the whole cohort (N = 1,908) and 699 death or HF endpoints in the Leicester and Basel cohorts (N = 1,694). Patients with elevated PENK levels had more deaths during follow-up (Table 2). Figure 3A illustrates a graded increase in the cumulative incidence of all-cause mortality with increasing PENK quartiles (p < 0.0005). Comparison of PENK quartiles revealed significant differences among all of them (p < 0.001), except comparing quartile 1 versus quartile 2 (p = 0.009). Figure 3B shows a graded increase in event rates for death and/or HF hospitalization with increasing PENK quartiles (p < 0.0005). Apart from quartile 1 versus 2 (p = 0.015), all other quartile comparisons were statistically different (p < 0.0005).
Figure 4A illustrates the univariable hazard ratios for factors affecting the outcome of all-cause mortality at 1 year, by using Cox proportional hazard survival analysis. Model 1 (Figure 4B) included relevant clinical variables and z-transformed natriuretic peptide levels, with independent predictors being age, past history of hypertension, SBP, plasma urea, sodium, eGFR, and natriuretic peptide levels. Addition of PENK to this base model (Figure 4C) showed that it had independent predictive value for death, its added value being statistically significant (p < 0.0005) using the increment in log likelihood ratio chi-square for nested regression models. For the endpoint of death at 3 and 6 months, the multivariable adjusted HR for PENK remained significant for both time points (3 months: HR: 1.49; 95% CI: 1.20 to 1.85; p < 0.0005; 6 months: HR: 1.40; 95% CI: 1.17 to 1.68; p < 0.0005).
The C statistic for 1-year mortality was 0.741 (in the base model using the foregoing demographic and clinical chemistry variables), and it rose to 0.754 (p = 0.021) and 0.751 (p = 0.051) with addition of natriuretic peptide and PENK, respectively, and to 0.759 (p = 0.007) with the addition of both biomarkers.
Figure 5A reports the HRs for the outcome of death or HF at 1 year in the Leicester and Basel cohorts. Model 1 (Figure 5B) is a multivariable model that included the independent predictors: age, past history of HF, hypertension, ischemic heart disease, diabetes, SBP, plasma urea, and natriuretic peptide levels. Addition of PENK (model 2) showed that it had independent predictive value for death or HF (p = 0.003) (Figure 5C), and the increment in log likelihood ratio chi-square was statistically significant (p = 0.001). For the endpoint of death or HF at 3 and 6 months, the multivariable adjusted HR for PENK remained significant for both time points (3 months HR: 1.27; 95% CI: 1.06 to 1.53; p = 0.011; 6 months HR: 1.32; 95% CI: 1.13 to 1.54; p < 0.0005).
Using the base model, the C statistic for 1-year death or HF was 0.692, and it rose to 0.702 (p = 0.079) and 0.700 (p = 0.09) with addition of natriuretic peptide and PENK, respectively, and to 0.706 (p = 0.039) with the addition of both biomarkers.
Predicting inpatient mortality
The use of PENK and/or natriuretic peptide together with clinical risk scores for determining inpatient mortality (n = 82) was examined using logistic regression analysis. Table 5 reports the odds ratios for z-transformed clinical scores individually and with the addition of natriuretic peptide and/or PENK. Using the ADHERE score, PENK was an independent predictor for inpatient mortality (p < 0.0005), with a C statistic of 0.709. PENK remained an independent predictor when used together with either the GWTG-HF score (C statistic 0.718) or the OPTIMIZE score (C statistic 0.729). Natriuretic peptide levels were not independent predictors in all 3 clinical scores, when used with PENK. The increment in ROC areas when biomarkers were added to clinical scores did not achieve conventional levels of statistical significance.
Category-free reclassification analyses, using the continuous net reclassification improvement index (>0) (Table 6), calculated for the biomarkers added to different clinical risk scores (ADHERE, GWTG-HF, and OPTIMIZE-HF) for predicting inpatient mortality, showed that natriuretic peptides up-classified risk in those patients endpoints, whereas PENK predominantly down-classified risk in patients without endpoints, with a smaller effect on up-classifying risk in patients who died (for ADHERE and GWTG-HF risk scores). Adding PENK to models consisting of risk scores and natriuretic peptide confirmed that PENK predominantly down-classified risk in survivors with less effect on up-classifying risk in patients with events (for ADHERE and GWTG-HF risk scores).
Decision tree analysis
To determine optimal cutpoints for PENK, we constructed decision trees using PENK, natriuretic peptides, and clinical risk scores (ADHERE, GWTG-HF, or OPTIMIZE-HF) to classify patients into survivors or those who died in the hospital. Only PENK was selected from these variables using chi-square automatic interaction detection, with the cutoff points and associated in-hospital death rates (Figure 6). Patients with PENK levels <133.3 pmol/l had an in-hospital mortality rate of 2.1%, whereas for patients with PENK >211.3 pmol/l, the mortality rate was 13.1%.
In this observational multicenter cohort study, we described the use of a recent plasma PENK assay for assessing WRF and risk stratification following acute HF, by measuring an analyte that is stable in plasma for at least 48 h, unlike previous assays of labile enkephalins. Plasma PENK was strongly correlated with renal function (eGFR), and the majority of its variance was accounted for by 2 measures of renal function, eGFR and plasma urea. Plasma PENK was an independent predictor of WRF, and levels increased over time while renal function declined. The temporal pattern of change of PENK-to-creatinine ratios differed in patients who had WRF compared with patients who did not, a finding suggesting that the dynamics of PENK and creatinine differed.
During follow-up, PENK was a strong independent predictor of death and the composite endpoint of death or HF, for both short-term (3-month) and longer-term (1-year) follow-up. We had previously demonstrated that PENK predicted outcomes such as death, myocardial infarction, and HF following acute myocardial infarction and was strongly linked to renal function (9). This present work reinforced our earlier findings and extended it to another acute cardiovascular presentation: acute HF. For prediction of in-hospital mortality, PENK remained significant when used with a variety of risk scores developed for this purpose (ADHERE, OPTIMIZE-HF, and GWTG-HF). However, the increment in ROC areas was modest and not significant. Previous analyses of data relying on increment in ROC areas could wrongly conclude that a biomarker has no added value (because of its very conservative power) even though logistic regression, which relies on increment in log-likelihood ratio, could indicate otherwise (25,26). We used reclassification as an additional method to demonstrate the utility of PENK, especially in down-classifying risk in patients without endpoints.
Individual studies and meta-analyses of HF cohorts have indicated the importance of renal impairment in determining prognosis (14), but there are few biomarkers that predict cardiorenal syndrome type 1 (15). Creatinine itself might weakly predict WRF in acute HF, but our findings suggest that PENK levels might contribute to a history of renal impairment, SBP, and plasma sodium, to provide modest accuracy in WRF prediction. The temporal patterns of plasma PENK in patients who developed WRF were different from those in patients who were spared this complication, with rising levels of PENK and falling PENK-to-creatinine ratios seen in patients with WRF.
The pathophysiological link between PENK and prognosis in acute HF may be related to the effects of opioids on the cardiovascular system (Central Illustration). Enkephalin excess may have a cardiodepressive effect (4), by lowering BP and reducing organ perfusion (including kidney perfusion, which may explain the strong link between PENK and eGFR). Fontana et al. (8) demonstrated a rise in atrial natriuretic peptide and norepinephrine levels, together with responses in heart rate and blood pressure, following nonspecific opioid blockade using naloxone in severe cases of acute HF. However, in less severe cases of acute HF, naloxone showed an opposite depressive effect on atrial natriuretic peptide and no effect on heart rate and blood pressure. The data suggest that endogenous opioids could be suppressing atrial natriuretic peptide secretion in severe acute HF. Peacock et al. (7) also described poorer outcomes in patients with acute HF who were administered opiates within the ADHERE registry. Similarly, enkephalins may exert a direct effect on renal function because Sezen et al. (27) reported that δ opioid agonists stimulate urine flow and sodium excretion, and, thus, in a situation where kidney function declines, the increase in enkephalin release may be a counter-regulatory measure. Enkephalins are widely expressed, and major sources of plasma enkephalin and proenkephalin include the heart, adrenal glands, skeletal muscle, and kidney (6).
On the basis of our findings indicating links between PENK and outcomes, there may be potential to use high PENK levels to select patients for intensified therapy settings or low PENK levels to rule out such care. The links of elevated PENK levels with poor outcome in acute HF agreed with findings in other acute emergencies such as acute myocardial infarction (9) and stroke (13). Better risk prediction may allow clinicians to improve allocation of treatment and resources, including placement of patients within the hospital, frequency of monitoring of renal function, and initiation and speed of up-titration of HF therapies associated with WRF (e.g., loop diuretics, angiotensin-converting enzyme inhibitors, and aldosterone receptor blockers). Some of these possibilities may need to be addressed by well-designed clinical trials.
First, the basis of our findings was a large number of patients prospectively enrolled and hospitalized in 3 European countries. Additional studies are warranted to validate these findings in non-European populations and in general practice settings. Furthermore, different natriuretic peptide and cardiac troponin assays were used in the 3 sites. We attempted to mitigate this by z-transforming the log-transformed peptide values. Additionally, we cannot comment on patients with terminal renal failure who were undergoing long-term hemodialysis because they were excluded. Finally, troponin results were not available on all patients.
Prospective studies on the clinical effectiveness of using PENK for management strategies, whether directed at low-risk or high-risk groups, need to be performed. Moreover, PENK as a predictor of WRF should be compared with other potential biomarkers, such as neutrophil gelatinase–associated lipocalin or cystatin C, for which a multimarker panel could be investigated. Sequential sampling of PENK should also be investigated to assess how quickly levels rise before WRF development.
Following acute HF, circulating PENK levels reflect cardiorenal status and provide short-term and long-term prognostic information on both mortality and cardiovascular morbidity. PENK predicted WRF and could be used in conjunction with different clinical risk scores for in-hospital mortality.
COMPETENCY IN MEDICAL KNOWLEDGE: PENK is inversely related to prevalent renal function, predicts WRF in acute HF, and provides an assessment of short-term and long-term prognosis independent of renal function. It may be used in conjunction with clinical scores for predicting in-hospital mortality.
TRANSLATIONAL OUTLOOK: Further studies should assess the utility of PENK to risk stratify and guide therapy for patients with acute HF in comparison with other biomarkers.
This work was supported by the John and Lucille van Geest Foundation and the National Institute for Health Research Leicester Cardiovascular Biomedical Research Unit. Dr. Bergmann holds ownership in Sphingotec GmbH, which manufactures the penKid assay; and is a member of the board of directors of Sphingotec GmbH. Drs. Hartmann and Struck are employees of Sphingotec GmbH. Dr. Alexandre Mebazaa has received speaker honoraria from Abbott, Novartis, Orion, Roche, and Servier; and has received fees as a member of advisory boards and/or steering committees from Cardiorentis, Adrenomed, MyCartis, NeuroTronik, ZS Pharma, and Critical Diagnostics. Dr. Twerenbold has received speaker honoraria from Roche; and has received a research grant from the Swiss National Science Foundation. Dr. Mueller has received research grants from the Swiss National Science Foundation, the Swiss Heart Foundation, the European Union, the Cardiovascular Research Foundation Basel, the University Hospital Basel, Abbott, AstraZeneca, Beckman Coulter, BG Medicine, bioMérieux, BRAHMS, Critical Diagnostics, Nanosphere, Roche, Siemens, Singulex, and Sphingotec; and has received speaker or consulting honoraria from Abbott, Alere, AstraZeneca, bioMérieux, Bristol-Myers Squibb, Boehringer Ingelheim, BRAHMS, Cardiorentis, Eli Lilly, Novartis, Roche, Sanofi, Siemens, and Singulex. Dr. Squire has received research grants from Novartis and Servier; and has received speaker or consulting honoraria from Novartis. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- B-type natriuretic peptide
- confidence interval
- estimated glomerular filtration rate
- heart failure
- hazard ratio
- N-terminal pro–B-type natriuretic peptide
- proenkephalin A
- receiver-operating characteristic
- systolic blood pressure
- worsening renal function
- Received September 7, 2016.
- Revision received September 29, 2016.
- Accepted October 4, 2016.
- American College of Cardiology Foundation
- Mebazaa A.,
- Gayat E.,
- Lassus J.,
- et al.,
- GREAT Network
- Cohen-Solal A.,
- Laribi S.,
- Ishihara S.,
- et al.
- Peacock W.F.,
- Hollander J.E.,
- Diercks D.B.,
- et al.
- Fontana F.,
- Bernardi P.,
- Pich E.M.,
- et al.
- Ng L.L.,
- Sandhu J.K.,
- Narayan H.,
- et al.
- Arbit B.,
- Marston N.,
- Shah K.,
- et al.
- Marino R.,
- Struck J.,
- Hartmann O.,
- et al.
- Doehner W.,
- von Haehling S.,
- Suhr J.,
- et al.
- Damman K.,
- Valente M.A.,
- Voors A.A.,
- et al.
- Ronco C.,
- Haapio M.,
- House A.A.,
- et al.
- Peterson P.N.,
- Rumsfeld J.S.,
- Liang L.,
- et al.
- Abraham W.T.,
- Fonarow G.C.,
- Albert N.M.,
- et al.,
- OPTIMIZE-HF Investigators and Coordinators
- Ponikowski P.,
- Voors A.A.,
- Anker S.D.,
- et al.
- Omland T.,
- Persson A.,
- Ng L.,
- et al.
- Damman K.,
- Tang W.H.,
- Testani J.M.,
- McMurray J.J.