Circulating miRNA-21 as a diagnostic biomarker in elderly patients with type 2 cardiorenal syndrome

Circulating miRNAs have attracted attention as serum biomarkers for several diseases. In this study, we aimed to evaluate the diagnostic value of circulating miRNA-21 (miR-21) as a novel biomarker for elderly patients with type 2 cardiorenal syndrome (CRS-2). A total of 157 elderly patients with chronic heart failure (CHF) were recruited for the study. According to an estimated glomerular filtration rate (eGFR) cut-off of 60 ml/min/1.73 m2, 84 patients (53.5%) and 73 patients (46.5%) were assigned to the CRS group and the CHF group, respectively. Expression levels of serum miR-21 and biomarkers for CRS, such as kidney injury factor-1 (KIM-1), neutrophil gelatinase-related apolipoprotein (NGAL), cystatin C (Cys C), amino-terminal pro-B-type natriuretic peptide (NT-proBNP), N-acetyl-κ-D-glucosaminidase (NAG), and heart-type fatty acid–binding protein (H-FABP), were detected. Serum miR-21, KIM-1, NGAL, Cys C, NT-proBNP and H-FABP levels were significantly higher in the CRS group than in the CHF group (P < 0.01), whereas NAG expression was not significantly different between the two groups (P > 0.05). Cys C, H-FABP and eGFR correlated significantly with miR-21 expression, but correlations with miR-21 were not significant for NT-proBNP, NGAL, NAG and KIM-1. Moreover, multivariate logistic regression found that serum miR-21, increased serum Cys C, serum KIM-1, hyperlipidaemia and ejection fraction (EF) were independent influencing factors for CRS (P < 0.05). The AUC of miR-21 based on the receiver operating characteristic (ROC) curve was 0.749, with a sensitivity of 55.95% and a specificity of 84.93%. Furthermore, combining miR-21 with Cys C enhanced the AUC to 0.902, with a sensitivity of 88.1% and a specificity of 83.6% (P < 0.001). Our findings suggest that circulating miR-21 has medium diagnostic value in CRS-2. The combined assessment of miR-21 and Cys C has good clinical value in elderly patients with CRS-2.

The term cardiorenal syndrome (CRS) refers to different clinical conditions in which heart and kidney dysfunction overlap.Ronco et al. proposed classifying CRS into 5 subtypes based on the potential underlying pathophysiologic mechanisms 1 .Among these subtypes, type 2 cardiorenal syndrome (CRS-2) is characterized by chronic abnormalities in cardiac function that result in kidney injury or dysfunction, along with clinical manifestations of CHF and an estimated glomerular filtration rate (eGFR) <60 ml/min/1.73m 2 2 .Furthermore, the incidence of renal insufficiency increases to 45-63% for elderly CHF patients with hypertension, diabetes, atherosclerotic heart disease and lipid metabolism diseases [3][4][5] .Assessment of kidney injury in CHF patients has previously been limited to serum creatinine, urea nitrogen and urinary protein excretion assays, which are prognostic for renal outcomes in CKD patients but not in heart failure patients 6 .However, novel kidney biomarkers (Cys C, NGAL, KIM-1, H-FABP and NAG) have recently been evaluated in CHF patients, and their serum levels have prognostic value for not only renal outcomes but also cardiovascular outcomes 7 .In addition, as an indicator of CHF, www.nature.com/scientificreportswww.nature.com/scientificreports/NT-proBNP correlates negatively with the eGFR level and can be used as an independent predictor of renal damage (i.e., eGFR <60 ml/min/1.73m 2 ) 8 .
MicroRNAs (miRNAs) are endogenous small noncoding RNAs of 21-24 nucleotides that control gene expression post-transcriptionally by inducing target miRNA destabilization or inhibiting protein translation.miRNAs play important roles in several diseases, such as heart failure, hypertension and kidney disease [9][10][11][12] , and many miRNAs have been identified as ideal candidate biomarkers due to their stable expression 12,13 .miR-21, which is encoded on chromosome 17q23.2,plays a regulatory role in several diseases, including cancer 14,15 , cardio-fibrosis 16,17 and kidney disease 18,19 , and combining miR-21 with hs-TnT was demonstrated to significantly increase the diagnostic value in acute coronary syndrome (ACS) patients 20 .Several studies have reported that miR-21 suppresses mitochondrial biogenesis and amplifies kidney disease and fibrosis [21][22][23] .Moreover, molecular miRNA signalling pathways suggest that miR-21 is an excellent biomarker for certain treatments for the early stages of fibrosis and tissue remodelling [24][25][26][27][28][29][30] .Thus, circulating miR-21 has diagnostic and prognostic value for heart and renal diseases, which suggests that serum levels of circulating miR-21 may have value in the diagnosis of cardiorenal syndrome.
The aim of our study was to detect the level of serum circulating miR-21 in elderly patients with CRS-2 and to propose a better diagnostic and therapeutic biomarker for these patients by combining this level with new serum biomarkers.

Results
Patient demographics.As shown in Table 1, the mean age of the patients and the proportion of males were not significantly different between the CRS and CHF groups (all P > 0.05).The patients in the CRS group had a lower level of haemoglobin (HB) (P < 0.05), whereas no significant differences in white blood cell count (WBC), ejection fraction (EF) or the proportions of diabetes, hypertension, hyperlipidaemia, coronary atherosclerotic heart disease, myocardial infarction, and hypothyroidism were found between the two groups (all P > 0.05).

Discussion
China is a gradually ageing society.CHF is the most common cardiovascular disease and is the most common cause of hospitalization in those older than 65 years.Furthermore, the incidence of CHF is 0.2-0.3% in the general population but increases greatly with age and is as much as 10-fold higher in those over 80 years of age 31 .CRS-2 is defined as renal impairment secondary to CHF, which is one of the leading causes of hospitalization, morbidity and mortality for elderly individuals with CHF and contributes greatly to the direct and indirect costs of CHF.Renal injury was typically an exclusion criterion in previous CHF clinical studies; thus, the reported evidence, diagnosis and management of organ pathology were not sufficient 1 .Accordingly, the detection of biomarkers, early diagnosis and timely treatment are particularly important for CRS-2 patients.
For patients with CHF, we analysed dynamic changes in serum NT-proBNP expression levels to evaluate cardiac function and assessed the expression levels of serum creatinine (Scr) and blood urea nitrogen (BUN) to estimate kidney damage.However, the heart-kidney interaction in CRS-2 patients causes a series of changes in neurohormones and inflammatory factors, and the internal environment changes significantly.Thus, assessment of disease status cannot be limited to the evaluation of a single organ's function.Moreover, when Scr and BUN abnormalities occur in patients, serious renal function damage has already occurred, suggesting that Scr and BUN are relatively late biomarkers.Some new biomarkers, such as NGAL, Cys C, KIM-1, H-FABP and NAG, have been reported to have diagnostic and prognostic value in CRS-2 patients 7 .For example, NGAL is considered an excellent biomarker for the early prediction of CKD 32 , and Bolignano et al. 33 suggested that NGAL could serve as a biomarker of atherosclerosis or CVD as well as CKD progression.Jungbauer et al. 34 followed CHF patients for 5 years and found a strong association of the tubular biomarkers KIM-1 and NAG with CKD progression in CHF, which may suggest their suitability as cardiorenal biomarkers.Cys C also appears to be of value in cardiovascular risk stratification of elderly patients with metabolic syndrome 35 , and H-FABP has diagnostic value in ACS 36 .Moreover, Oezkur et al. identified a correlation between preoperative serum H-FABP and the postoperative incidence of renal impairment in ACS patients 37 .In summary, NGAL, Cys C, KIM-1, H-FABP and NAG are considered to be effective in the early detection of renal impairment and may also reflect cardiac function in CRS-2 patients, contributing to the early diagnosis, prevention and treatment of CRS-2 7,38 .
Many studies to date suggest that circulating miRNAs can be applied as biomarkers for disease diagnosis/ prognosis, as well as therapeutic targets for certain diseases 12,13,39 .miRNAs are endogenous noncoding small RNAs with regulatory functions that recognize target RNAs by complementary base-pairing and guide silencing complexes to degrade the target mRNA or repress its translation.miRNAs also play important roles in the occurrence and development of various diseases and the ageing of organs, and circulating miRNAs stably reflect gene regulation through the internal environment of an individual.Thus, by evaluating circulating miRNAs, the overall assessment of diseases becomes more effective and accurate.Among numerous miRNAs, studies have shown that miRNA-21 plays a regulatory role in both heart and kidney function.Located on chromosome 17, miRNA-21 is a member of the mitochondrial family subtype, with its own specific promoter 40 .miR-21 participates in the pathological and physiological processes of many diseases, including cardiovascular diseases, in which it prevents post-infarction cardiac insufficiency through targeted regulation 16 and participates in the process of atherosclerosis by mediating signalling 41 .One study reported that high expression of miRNA-21 can promote fibroblast activation and accelerate the progression of renal fibrosis.In addition, knockdown of miR-21 expression in animal models through a gene modification technique effectively inhibited the progression of renal fibrosis caused by renal injury 42 .A high level of miRNA-21 downregulates the expression of its targets and accelerates the progression of cardiac fibrosis; therefore, it is an important signalling molecule for cardiac fibrosis 17 .This evidence supports the possibility of using miR-21 for the early diagnosis and assessment of CRS-2.
In our present study, we found significant differences in the abovementioned biomarkers between the CRS and CHF groups.In addition, Cys C, H-FABP and eGFR correlated significantly with miR-21 expression.In multivariate regression analysis, serum miR-21, increased serum Cys C, serum KIM-1, hyperlipidaemia and EF      were independent influencing factors for CRS (P < 0.05), confirming previous research results and our experimental hypothesis.The highest AUC value, 0.888, was observed for Cys C, indicating that Cys C has particular advantages for prediction and early diagnosis.Cys C is a cysteine protease inhibitor, and its half-life is only 1.5 h (for comparison, Scr's half-life is 4 h) 43 .The level of Cys C only depends on the glomerular filtration ability, and experiments have shown that the biological change in Cys C is only 7% of that in Scr 6 .Therefore, compared with other biomarkers, Cys C better reflects changes in renal function 44 .Our results also showed that miR-21 expression was significantly increased in the CRS group, with elevated miR-21 being an independent risk factor for CRS.Furthermore, miR-21 plays a role in the diagnosis of CRS.Among the tested biomarkers, miR-21 was second only to Cys C: the AUC for CRS-2 was 0.749, and its specificity was 84.93%, which was better than that of Cys C (82.19%), although this difference was not statistically significant (P > 0.05).Indeed, the miR-21 AUC for CRS-2 was higher than that of most biomarkers (P < 0.001), except Cys C.These results indicate that miR-21 has an important function in the occurrence and development of CRS disease.Additionally, gene regulation related to miR-21 may be studied to find an effective basis for gene regulation therapy in CRS patients and to explore new treatment schemes, which will slow CRS disease development and contribute to effective intervention.In general, regulation of miR-21 may inhibit or delay the progression of cardiac and renal fibrosis and improve the prognosis of CRS-2.
Regarding our combined diagnostic analysis, although the diagnostic value of miR-21 (AUC 0.749; 95% CI: 0.671-0.827)was not higher than that of Cys C (AUC 0.888; 95% CI: 0.836-0.940),the combined AUC of these biomarkers for CRS-2 was 0.902 (95% CI: 0.854-0.950),with the highest sensitivity in the model, at 88.1%, representing a great improvement in the diagnostic value for CRS-2 compared with that of miR-21 alone (P < 0.001).This suggests that in addition to eGFR, dynamic changes in Cys C and miR-21 may facilitate the diagnosis/evaluation of CRS diseases and the determination of the disease status of these patients.

Conclusion
Serum circulating miR-21 is increased in patients with CRS-2.Circulating miR-21 has medium diagnostic value; therefore, it may become a factor in the clinical pathophysiological activation and diagnosis of CRS-2.The combined assessment of miR-21 and Cys C has good clinical value for CRS-2, with a potential role in the clinic for better monitoring the status of CRS patients.Study participants.We applied the following inclusion criteria: (i) age over 65 years; (ii) NYHA II-IV; (iv) occurrence of cardiac insufficiency earlier than renal insufficiency in CRS patients.Exclusion criteria: patients with autoimmune diseases, cancer, liver diseases, haematological diseases, recent surgery or trauma, acute and/ or chronic inflammatory state, and CHF induced by primary kidney disease or CKD (i.e., CRS-4).Data for a total of 157 patients diagnosed with CHF at Hebei General Hospital between November 2017 and March 2019 were collected, including 74 males and 83 females between 65 and 97 years old.Fresh blood samples were analysed using standard hospital assays to measure serum levels of creatinine, NGAL, Cys C, NT-proBNP.The estimated glomerular filtration rate (eGFR) was calculated by the CKD-EPI Creatinine 2009 Equation, and the 157 patients were divided into a CRS group (eGFR < 60 ml/min/1.73m 2 ) and a CHF group (eGFR ≥ 60 ml/min/1.73m 2 ).

Sample preparation.
We collected 3-4 ml peripheral blood from each patient into a procoagulation tube (HEBEI XINLE SCI&TECH CO., Ltd, Hebei, China), and the sample was centrifuged at 2000 rpm for 10 min.

Figure 4 .
Figure 4. ROC analysis of the individual/combined biomarkers.Combined miR-21+CysC was used to predict CRS in elderly CHF patients compared to miR-21 alone.

Ethics statement.
All experimental procedures were approved by the Ethics Committee of Hebei General Hospital [clinical ethics approval number 2017203].The study was conducted in accordance with the Declaration of Helsinki and with Good Clinical Practice guidelines, as defined by the International Conference on Harmonisation.All patients provided written informed consent before enrolment.A data and safety monitoring committee was established to provide oversight of safety and efficacy considerations.The study was registered on ResMan [approval number ChiCTR1800016858].

Table 1 .
Clinical characteristics of individuals CHF and CRS group.Data are expressed as Median (Quartile range) WBC: white blood cell count, EF: ejection fraction, HB: haemoglobin.

Table 4 .
AUC for individual/combination biomarkers.Receiver operating characteristic (ROC) curve analysis was used to calculate the area under the curve (AUC) of individual and combined biomarkers for diagnosing CRS.