Serum Irisin Predicts Mortality Risk in Acute Heart Failure Patients Open Access

Acute heart failure (AHF) is a complex and heterogeneous clinical syndrome and a major public health burden worldwide [1-3]. AHF leads to about 676,000 annual emergency visits in the United States, and over 80% AHF patients need hospitalization [4]. AHF has a high mortality and always requires immediate medical attention and urgent therapy [1-3]. Identification of novel prognostic biomarkers for AHF long-term mortality risk will help make appropriate management decision and improve clinical therapy for patients [5, 6].

Irisin is a peptide hormone cleaved from a plasma membrane protein fibronectin type III domain containing protein 5 (FNDC5), which was firstly reported to be increased by exercise training and capable of triggering brown-fat-like development of white fat and increasing energy expenditure [7]. Emerging studies have demonstrated a close relationship between circulating irisin levels and many major chronic diseases such as obesity, type 2 diabetes, nonalcoholic fatty liver disease, chronic kidney disease, cancer, and cardiovascular diseases [8-20]. Of note, increased circulating irisin levels have been associated with the development of major adverse cardiovascular events (MACE) in coronary artery disease patients after percutaneous coronary interventions [17]. However, the role of serum irisin as a predictor for mortality risk in acute heart failure (AHF) patients is not known.

In the present study, a total of 161 AHF patients were enrolled and followed up for 1 year. Serum irisin level as determined by enzyme-linked immunosorbent assay (ELISA) was associated with mortality risk. It was found that serum irisin was significantly higher in patients deceased in 1-year follow-up and its level was associated with mortality based on both univariate and multivariate logistic regression analyses. Collectively, our study identified serum irisin as a predictive biomarker for 1-year all-cause mortality in AHF patients though large multicenter studies are highly desired.

All investigations conformed to the principles of the Declaration of Helsinki, and the study was approved by the ethics committee of Nanjing Medical University. All participants signed the informed consent before enrollment in the present study.

A total of 161 AHF patients were recruited in this study dated from April 2012 to October 2015. The inclusion and exclusion criteria have been stated in our previous report [21]. In brief, the diagnosis of AHF was confirmed by two senior cardiology physicians separately based on clinical and biological parameters according to the ACCF/AHA Guidelines for the Management of Heart Failure. Eligible patients were those who were hospitalized with new-onset or worsening preexisting heart failure as the primary cause of admission. Subjects with a history of malignant tumors, severe mental illness, cognitive dysfunction or dementia, or other uncontrolled systemic diseases were excluded.

The patients were followed up by monthly visits or telephone interview for 1 year after the day of enrollment. All-cause mortality within 1 year after AHF diagnosis was set as the endpoint of this study. All enrolled AHF patients received standard treatment and management according to the guidelines during the follow-up period, and information concerning the death was obtained from the hospital medical records or family members.

Venous blood of enrolled patients was collected at admission in serum collecting tubes, after centrifugation (4°C at 3000× g for 15 min). The supernatant was transferred to RNase/DNase-free tubes and stored at -80°C. Serum irisin concentrations were measured by ELISA kits (Phoenix Pharmaceuticals, Burlingame, CA, USA) in accordance with the manufacturer’s instructions. The sensitivity of the assay was 0.1 ng/ml, and the linear range of the standard was 0.1-1,000 ng/ml.

The SPSS software package (version 20.0, Chicago, IL) was used for statistical analyses. The results were shown as the mean ± standard deviation (SD) for the number of assays indicated, and as count (percentage) for categorical variables. Comparisons between the groups were conducted using the independent sample t test. Univariate analysis and multivariate analysis with odds ratios and 95% CI were performed to examine the associations of the clinical factors with survival. Kaplan–Meier method was used to estimate overall survival within survival and deceased group. Survival data were collected until 12 months after the last patient enrolled in the study and survival was assessed every 6 months thereafter. Receiver-operator characteristic (ROC) curve was used to present the results for assessing the predictive accuracy of serum irisin. MedCalc software (Mariakerke, Belgium) was used to compare the areas between ROC curves. A P value of <0.05 was taken as statistically significant.

During the 1-year follow-up period, a total of 42 (26.1%) patients deceased. Serum irisin levels in survived AHF patients were compared to those deceased ones, and it was found that serum irisin level was significantly higher in deceased AHF patients during the 1-year follow-up period (Fig. 1), indicating that serum irisin level might correlate with AHF mortality risk.

AHF patients were divided into 2 groups based on the median of serum irisin levels (high-level group and low-level group). In high serum irisin group, higher American Society of Anesthesiologists (ASA) class, more renal insufficiency, less cardiomyopathy, and lower plasma sodium level were found (Table 1). However, binary logistic regression indicated that serum irisin level was not associated with most baseline features except for AHA class and cardiomyopathy (Table 2), confirming that serum irisin was not a marker for other comorbidities.

Taking advantage of the univariate logistic regression analysis, 18 variables were found to be related to the mortality (Table 3). Among them, ALT, AST, cystatin C, uric acid, NT-pro-BNP, blood urea nitrogen, and serum irisin level were selected as independent variables and were further subjected to the multivariate logistic regression analysis. We found that higher levels of blood urea nitrogen and serum irisin level were associated with an increased AHF mortality risk (OR: 1.181 [1.022-1.366] and OR: 1.287 [1.079-1.537], respectively, P<0.05) (Table 4).

After that, the mortality risk predictive power of both variables, together with NT-pro-BNP, the most commonly used biomarker, was determined by ROC curve analysis. Both blood urea nitrogen and NT-pro-BNP had only moderate prognostic values for AHF mortality risk compared to serum irisin level (Fig. 2). The combination of two or three variables did not show better power than serum irisin level (Fig. 3), indicating that serum irisin level is a good predictor for AHF mortality risk at least in 1-year follow-up period. Finally, Kaplan-Meier survival analysis demonstrated that AHF patients with higher serum irisin levels had significantly higher mortality risk (Fig. 4). Thus, our study suggests that serum irisin is a predictive biomarker for 1-year all-cause mortality in AHF patients though large multicenter studies are highly needed.

AHF is a condition with poor prognosis, and the 30-day and 6-month mortality rate of AHF is approximately 20% and 30%, respectively [22, 23]. An accurate and rapid identification of AHF patients at high mortality risk is extremely important for facilitating clinical management of AHF patients and increasing overall survival [2]. To be useful, biomarkers for AHF should be easily interpretable, rapidly available, and highly cost-effective. Recently, several biomarkers, including copeptin, galectin-3, and soluble suppression of tumorigenicity2 (sST2), have been reported to be associated with AHF diagnosis and may also be used to predict AHF prognosis [5]. Nevertheless, exploring highly sensitive and specific novel biomarkers for AHF prognosis remains a challenging task.

Irisin is a peptide hormone cleaved from FNDC5, which can be released into the circulation [7]. Since its discovery, irisin has been found to be associated with various diseases, including metabolic diseases and cancers [24]. Serum irisin was reported to be decreased in myocardial infarction in both animal experiments and humans [18-20]. In addition, decreased serum irisin was found to be associated with the presence, severity, and higher SYNTAX score of coronary artery disease [15, 16]. Serum irisin has also been found to be independently associated with general Framingham risk profile and is a strong positive biomarker for cardiovascular disease risk within 10 years [25]. Besides that, increased circulating irisin levels were also demonstrated to be associated with MACE in coronary artery disease patients after percutaneous coronary interventions. Of note, the protective effects of irisin in cardiovascular diseases have consistently been reported [26-30]. Thus, the elevated circulating irisin levels in patients with risk for cardiovascular diseases or MACE have been considered as a state of irisin-resistance. In addition, as heart might be a major producer of irisin, the uncoupling properties of irisin may result in more ATP loss and lead to poor prognosis of cardiovascular diseases [31]. Here we also provide evidence that elevated serum irisin levels predict mortality risk in AHF patients though its functional role deserves to be investigated in the future.

This study was subjected to several limitations as follows. Firstly, our sample size was limited to 161 patients in one institution, and undoubtedly, our findings will need to be validated in a larger population and multiple institutions. Secondly, we did not correlate serum irisin levels with other AHF biomarkers such as copeptin, galectin-3, and sST2, due to the lack of availability of these data. Thirdly, the source of serum irisin remains to be determined. Finally, the response of serum irisin to AHF treatment and if serum irisin-guided therapy can achieve additional benefits for AHF patients are unclear.

In summary, this study identified serum irisin as a predictive biomarker for 1-year all-cause mortality in AHF patients though large multicenter studies are highly needed.

This work was supported by the grants from National Natural Science Foundation of China (81570362 and 91639101 to JJ Xiao, 81370332 and 81170201 to XL Li, 81400647 to Y Bei), the development fund for Shanghai talents (to JJ Xiao), and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD20102013 to XL Li), and Twelve-Fifth National Key Technology R&D Program (2011BAL11B08 to XL Li). Dr XL Li is an Associate Fellow at the Collaborative Innovation Center for Cardiovascular Disease Translational Medicine and F Zhou is a Fellow at the Six Talents Peak Project for Jiangsu Province [WSN-031(IB15)].

The authors declare no conflict of interest.

S. Shen, R. Gao and Y. Bei contributed equally to this work.