Correlation of Serum Cardiac Markers with Acute Decompensating Events in Liver Cirrhosis

Methods Cirrhotic patients who were consecutively hospitalized between January 2016 and March 2019 were screened. Serum cardiac biomarkers at admission, including N-Terminal pro-B-type natriuretic peptide (NT-pro BNP), high-sensitivity cardiac troponin T (hs-cTnT), creatine kinase (CK), creatine kinase MB (CK-MB), and lactate dehydrogenase (LDH), were collected. Acute decompensating events at admission, primarily including ascites, acute gastrointestinal hemorrhage, and acute-on-chronic liver failure (ACLF), were recorded. Results The NT-pro BNP level was significantly higher in cirrhotic patients with acute decompensating events than in those without any decompensating events (median: 140.75 pg/mL versus 41.86 pg/mL, P < 0.001). The NT-pro BNP level significantly correlated with ascites, acute gastrointestinal hemorrhage, and ACLF. The hs-cTnT level was significantly higher in cirrhotic patients with acute decompensating events than in those without decompensating events (median: 0.008 ng/mL versus 0.006 ng/mL, P = 0.007). The hs-cTnT level significantly correlated with acute gastrointestinal hemorrhage, but not ascites or ACLF. LDH (185.0 U/L versus 173.5 U/L, P = 0.281), CK (71 U/L versus 84 U/L, P = 0.157), and CK-MB (29.5 U/L versus 33.0 U/L, P = 0.604) levels were not significantly different between cirrhotic patients with and without acute decompensating events. Conclusion The elevated NT-pro BNP level seems to be closely related to the development of acute decompensating events in liver cirrhosis.


Introduction
Liver cirrhosis is a state of systemic hyperdynamic circulation characterized by increased cardiac output and decreased peripheral resistance, especially in the presence of decompensating events [1]. This disease activates the sympathetic nervous system (SNS) and the renin-angiotensin-aldosterone system (RAAS), which further increases the myocardial tension and ultimately results in chronic cardiac dysfunction [2][3][4]. Consequently, the levels of atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and troponin I (TnI) are increased [5]. Such chronic cardiac dysfunction in liver cirrhosis presenting with systolic dysfunction, diastolic dysfunction, and electrophysiological changes is known as "cirrhotic cardiomyopathy" [4]. An interaction between liver cirrhosis and cardiac dysfunction suggests that serum cardiac markers may be valuable in evaluating the disease state of liver cirrhosis.
BNP and N-Terminal pro-B-type natriuretic peptide (NT-pro BNP) are secreted in response to increased myocardial stress [6][7][8]. The NT-pro BNP level is significantly increased in liver cirrhosis, which may be related to cardiac dysfunction [9][10][11]. Also, the NT-pro BNP level significantly correlates with the severity of liver dysfunction and prognosis of cirrhotic patients [7,12]. On the other hand, highsensitivity cardiac troponin T (hs-cTnT), another highly specific and sensitive marker of myocardial injury, is also significantly increased in patients with liver cirrhosis. Thus, NTpro BNP as well as hs-cTnT may be valuable for prognostic assessment of liver cirrhosis [12]. However, their correlation with acute decompensating events in liver cirrhosis has never been explored yet. Additionally, the clinical significance of other biomarkers of cardiac injury, such as creatine kinase (CK), creatine kinase MB (CK-MB), and lactate dehydrogenase (LDH), in liver cirrhosis remains unclear.
Therefore, this study is aimed at exploring the relationship between these cardiac markers and decompensating events in cirrhosis.

Patients.
We have prospectively collected the demographic, clinical, and laboratory data of cirrhotic patients who were consecutively admitted to the Department of Gastroenterology of the General Hospital of Northern Theater Command (formerly General Hospital of Shenyang Military Area) and treated by an attending physician (XQ) since January 1, 2016. Until March 31, 2019, there were a total of 761 admissions. We retrospectively screened the patients who had undergone the evaluation of laboratory data regarding serum cardiac markers during the hospitalizations. Exclusion criteria were as follows: (1) confirmed or suspected diagnosis of malignancy, (2) severe renal insufficiency (estimated glomerular filtration rate ðeGFRÞ < 30 mL/min/1.73 m 2 ), (3) cardiac diseases (i.e., heart failure, coronary atherosclerotic heart disease, and atrial fibrillation), (4) ischemic stroke, and (5) absence of data regarding serum cardiac markers detected during the hospitalizations. Repeated admissions of the same patient were not deliberately excluded, because this study focused on the in-hospital outcome and decompensating events during the hospitalizations, but not on long-term follow-up outcomes. If a patient had multiple measurements of serum cardiac markers during the same hospitalization, we selected the data obtained at the first time of blood collection. The study protocol was approved by the Medical Ethics Committee of our hospital. We primarily collected demographic data, etiology of liver cirrhosis, decompensating events at admissions, and laboratory tests including serum cardiac markers.

Definitions and Diagnosis.
Acute gastrointestinal hemorrhage was defined as previously described [13,14]. The Child-Pugh score was calculated [15]. Model for end-stage liver disease (MELD) and MELD with sodium (MELD-Na) scores were calculated according to an equation updated by If Na + < 125 mmol/L, it is set to 125; if Na + > 137, it is set to 137.
The grade of ascites was defined according to the consensus of the International Ascites Club [17]. Patients with acute-on-chronic liver failure (ACLF) were identified by the recommendations of the Asian Pacific Association for the Study of the Liver (APASL) consensus [18,19]. Severe renal insufficiency was defined as eGFR < 30 mL/min/1.73 m 2 [20]. The eGFR was calculated using the simplified equation [21], as follows: 2.4. Measurement of Serum Cardiac Markers. All serum cardiac markers were measured at the Department of Laboratory of our hospital. They included NT-pro BNP detected by enzyme-linked immunosorbent assay (ELISA) (double antibody sandwich method) with a normal range of 0-125 pg/mL, hs-cTnT by ELISA (double antibody sandwich method) with a normal range of 0-0.05 ng/mL, CK by coupled-enzyme assay with a normal range of 38-174 U/L, CK-MB by immune inhibition assay with a normal range of 0-24 U/L, LDH by the spectrophotometric method with a normal range of 109-245 U/L, and hs-CRP by latex immune turbidimetry with a normal range of 0-3 mg/L. Only the data obtained at the first time of measurement were selected, thus avoiding the influence of drugs used during hospitalization.

Statistical
Analyses. Continuous data were expressed as mean ± standard deviation and median (quartiles) and were compared by using the Wilcoxon rank-sum test. Categorical data were expressed as frequency (percentage) and were compared by using the chi-square test. Considering that age and gender are important factors influencing serum NT-2 Gastroenterology Research and Practice pro BNP and hs-cTnT levels [22][23][24][25][26], partial correlation analysis was adjusted for age and gender to analyze the correlation of serum NT-pro BNP and hs-cTnT with liver disease conditions. Pearson or Spearman tests were performed to analyze the correlation between disease conditions and other serological cardiac markers, such as CK, CK-MB, and LDH. Multivariate linear regression analysis was performed to analyze the correlation of serological cardiac markers with categorical variables. A two-sided P < 0:05 was considered to be statistically significant. SPSS statistics software version R23.0.0.0 was employed to perform all statistical analyses.

Comparison between Cirrhotic Patients with and without
Decompensating Events. Cirrhotic patients with decompensating events had significantly higher levels of NT-pro BNP (P < 0:001) and hs-cTnT (P = 0:007) than those without decompensating events ( Figure 2), but the differences in the levels of CK, CK-MB, and LDH were not significant between them (Table 1). Cirrhotic patients with ascites had significantly higher levels of NT-pro BNP (P < 0:001) and hs-cTnT (P = 0:002) than those without decompensating events ( Figure 2), but the differences in the levels of CK, CK-MB, and LDH were not significant between them (Table 2).
Cirrhotic patients with acute gastrointestinal hemorrhage had significantly higher levels of NT-pro BNP (P < 0:001) and hs-cTnT (P = 0:003) than those without decompensating events ( Figure 2), but the differences in the levels of CK, CK-MB, and LDH were not significant between them ( Table 2).
Cirrhotic patients with ACLF had a significantly higher level of NT-pro BNP (P < 0:001) than those without decompensating events (Figure 2), but the differences in the levels of CK, CK-MB, and LDH were not significant between them ( Table 2).

Correlation of Serum Cardiac
Markers with Child-Pugh and MELD Scores in Patients with Liver Cirrhosis. Partial correlation analyses demonstrated that the NT-pro BNP level significantly correlated with Child-Pugh and MELD scores. These correlations were observed in all the cirrhotic patients and in those with decompensating events, but not in those without decompensating events (Table 3). Partial correlation analyses demonstrated that the hs-cTnT level had no significant correlation with Child-Pugh and MELD scores in cirrhotic patients regardless of the presence of decompensating events (Supplementary Table 1). Correlation analyses demonstrated that CK (Supplementary Table 2) and CK-MB (Supplementary Table 3) levels did not significantly correlate with Child-Pugh and MELD scores in cirrhotic patients. On the other hand, the LDH level significantly correlated with Child-Pugh and MELD scores in cirrhotic patients (Supplementary Table 4).

Correlation between Serum Cardiac Markers and
Decompensating Events in Cirrhotic Patients. Age-or gender-adjusted multivariate linear regression analyses demonstrated that the NT-pro BNP level significantly correlated with overall acute decompensating events, ascites, acute gastrointestinal hemorrhage, and ACLF (Table 4); the hs-cTnT level significantly correlated with overall acute decompensating events and acute gastrointestinal hemorrhage, but not ascites or ACLF (Supplementary Table 5); CK and

Gastroenterology Research and Practice
CK-MB levels did not significantly correlate with overall acute decompensating events, ascites, acute gastrointestinal hemorrhage, or ACLF; the LDH level significantly correlated with ascites, acute gastrointestinal hemorrhage, and ACLF, but not overall acute decompensating events (Supplementary Table 6).

Discussion
In this retrospective observational study, we rigorously screened the participants by excluding the confounding factors, which makes our statistical results more reliable. Additionally, we included a relatively large number of cirrhotic patients, which makes our conclusions more representative. The major findings are as follows: (i) the NT-pro BNP level was significantly higher in decompensated cirrhosis. (ii) The NT-pro BNP level also significantly correlated with Child-Pugh and MELD scores in cirrhosis with acute decompensation, but not in those without decompensation. (iii) The hs-cTnT level was elevated in cirrhosis with acute decompensation but was unrelated to the liver disease severity. (iv) The LDH level significantly correlated with Child-Pugh and MELD scores in cirrhosis, but was unrelated to decompensating events. (v) CK and CK-MB levels were neither significantly increased in cirrhosis with decompensation nor correlated with Child-Pugh and MELD scores.
4.1. NT-pro BNP. NT-pro BNP, a prohormone of BNP, is secreted into the systemic circulation by cardiac ventricles in response to myocardial hypertrophy and is involved in the regulation of cardiac volume homeostasis [27][28][29]. Thus, the NT-pro BNP level is often considered as an effective and useful marker for screening of early stages of cardiac dysfunction [30]. As we know, cirrhotic cardiomyopathy is a chronic cardiac systolic and diastolic dysfunction in cirrhotic patients in the absence of prior heart disease [31,32]. There is no obvious abnormal change of cardiac function in the resting state; besides, a decreased afterload in cirrhosis often results in normal or even increased left ventricular ejection fraction [31]. Thus, noninvasive cardiac biomarkers are potentially useful to reflect the slight change of pressure state of enddiastolic wall stress and intracardiac filling pressures.
The serum NT-pro BNP level is significantly higher in patients with liver cirrhosis [7,9,10,12,33], probably because it is often associated with hyperdynamic circulation, such as increased heart rate and cardiac output, thereby impairing cardiac contractility [34][35][36]. Our study for the first time found that the serum NT-pro BNP level was significantly higher in cirrhotic patients who suffer an acute decompensation, such as ascites, gastrointestinal hemorrhage, and ACLF, when the values were compared to those without decompensation. Interestingly, we also found that the NT-pro BNP level significantly correlated with Child-   Gastroenterology Research and Practice

<0.001
High-sensitivity C-reactive protein (mg/L) Pugh and MELD scores in cirrhotic patients with acute decompensating events, but not in those without decompensation, which would suggest that NT-pro BNP can reflect the insidious change of cardiac dysfunction in advanced cirrhosis with cardiac dysfunction.
In addition, BNP is a natriuretic hormone released from myocardial cells in response to volume expansion, enddiastolic wall stress, and possibly increased intracardiac filling pressures [8,37]. Hypertrophy of the left ventricle, left-atrial dilatation, and increased end-diastolic and endsystolic left-ventricular volume are frequently observed in liver cirrhosis [1,32,38,39], which are potentially the main causes for an increase of NT-pro BNP.
4.2. hs-cTnT. hs-cTnT, a protein complex regulating the contraction of striated muscle, is released when myocardial ischemia induces nonreversible injury of myocardial tissue [40]. hs-cTnT is a specific and sensitive biomarker of myocardial damage and is being widely used for clinical screening in patients with suspected acute myocardial infarction [41,42], but not for evaluating the change of myocardial contractility. The hs-cTnT level can be also elevated in some  [43][44][45][46], tachycardia, pericarditis, vigorous exercise [47], and atrial fibrillation [48][49][50]. The present study has rigorously excluded these conditions. The hs-cTnT level seems to be related to the severity and survival of cirrhotic patients [12,14]. Our results also showed that the hs-cTnT level was significantly higher in decompensated cirrhotic patients than those without decompensation. This association was mainly attributed to the effect of acute gastrointestinal hemorrhage, but not to ascites or ACLF (Supplementary Table 5). A possible explanation for this finding could be that acute gastrointestinal bleeding in cirrhotic patients may lead to hypovolemic hypotension, which is a significant risk factor for myocardial damage [51][52][53], thereby increasing the levels of myocardial damage biomarkers [54]. The pathophysiological link of the association remains unexplained, and it needs further research to clarify its mechanism.
4.3. CK and CK-MB. We did not find any significant difference in CK and CK-MB levels, comparing compensated and decompensated cirrhotic patients. Moreover, there was no correlation of CK and CK-MB levels with Child-Pugh and MELD scores in cirrhosis. These analyses were performed in the groups with and without decompensating events. Traditionally, CK-MB is helpful for estimating the infarct size in acute myocardial infarction and is highly specific to heart tissue [55,56] while CK is used for assessing myocardial damage in acute myocardial infarction [57]. None of them is a good indicator of cardiac volume overload. Our study suggested that CK and CK-MB levels did not correlate with the severity of cirrhosis.
4.4. LDH. LDH, a cytoplasmic enzyme, exists in a wide range of tissues and is elevated when cells are damaged. LDH is not specific for the diagnosis of a disease. There are five types of serum LDH isoenzymes. Among them, LDH1 is mainly derived from the heart and LDH5 from the liver [58]. However, LDH5 have lower specificity and sensitivity than ALT for diagnosing and evaluating liver diseases [59]. LDH significantly correlated with Child-Pugh and MELD scores in cirrhotic patients. However, there was no significant difference in LDH levels between cirrhotic patients with and without decompensation. These results indicate that LDH might not be sensitive to early cardiac dysfunction caused by cirrhosis.

4.5.
Limitations. First, the number of patients with ACLF was small in our cohort and the relationship between serum cardiac markers and ACLF needs further clarification. Second, a reasonable and convenient approach for quantifying blood loss volume during acute gastrointestinal hemorrhage and volume of ascites was unavailable. Third, healthy controls may make the results more comprehensive. Further studies should further consider the effects of the severity of such decompensating events on the long-term prognosis.

Conclusion
An elevated NT-pro BNP level might be useful to identify the cardiac volume overload caused by acute decompensating events in advanced cirrhosis. Additionally, the hs-cTnT level was elevated in cirrhosis with acute decompensating events.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon reasonable request.

Disclosure
The abstract was published in the Asian Pacific Association for the Study of the Liver (APASL) 2020 Conference as a poster presentation. Please see the following link: https:// link.springer.com/content/pdf/10.1007/s12072-020-10030-4 .pdf.

Conflicts of Interest
The authors declare that they have no conflicts of interest.

Supplementary Materials
Supplementary