Heart rate variability in the clinical assessment of patients with chronic liver disease

Abstract

Autonomic dysfunction (AD) is frequently observed in cirrhotic patients and is associated with poor clinical outcomes and prognoses. Heart rate variability (HRV), a noninvasive tool for assessing autonomic nervous system balance, has been extensively studied in a variety of conditions, including chronic liver disease (CLD); however, no recent reviews have focused on its role in CLD. This article examines the mechanisms of AD in CLD and the foundation for HRV assessment, highlighting its diagnostic, prognostic, and therapeutic applications in CLD, including liver transplantation (LT). Changes in HRV, particularly in patients with cirrhotic complications, and its prognostic significance throughout the natural history of CLD are summarized. We show that HRV is consistently reduced in CLD patients, reflecting AD, and is inversely correlated with liver disease severity. Also, low HRV is associated with complications such as hepatic encephalopathy, ascites, and portal hypertension. Moreover, evidence indicates that reduced HRV is an independent risk factor for mortality and circulatory instability in CLD. Furthermore, treatment with beta-blockers and LT improves HRV, underscoring its potential role in patient management. While further studies are needed, HRV emerges as a promising tool for the comprehensive evaluation and clinical management of patients with CLD, offering insights into disease progression and therapeutic response.

Key Words: Chronic liver disease; Autonomic nervous system; Heart rate variability; Autonomic dysfunction; Complications of liver cirrhosis; Liver transplant

Core Tip: Autonomic dysfunction (AD) is common in cirrhotic patients and is linked to poor outcomes. Heart rate variability (HRV), a noninvasive measure of autonomic balance, could be useful in managing chronic liver disease (CLD) patients. This article explores AD mechanisms, HRV assessment, and its diagnostic, prognostic, and therapeutic relevance in CLD, including liver transplantation.

INTRODUCTION

Most of the health-related problems in chronic liver disease (CLD) patients result from the complications of portal hypertension (PHT) indicating decompensated cirrhosis, or from the development of hepatocellular carcinoma (HCC)[1].

Autonomic dysfunction (AD), common in cirrhotic patients, is associated with PHT. Investigations of heart rate variability (HRV), a noninvasive measure of autonomic nervous system (ANS) balance, have provided evidence of this association. Several studies have shown that HRV, associated with the presence of AD and inversely correlated with the severity of the disease, is diminished in patients with CLD. Contrariwise, HRV improves with some therapies for cirrhosis and PHT. However, there is no up-to-date information about the clinical utility of HRV in CLD.

This article aimed to review the ways in which HRV could be used as a simple tool for evaluating patients with CLD and PHT-related complications and to summarize the utility and limitations of HRV in the clinical assessment of CLD patients.

ANS IN CLD

Neurohumoral activation and ANS

PHT is key in CLD pathophysiology, as it is associated with the extent of disease decompensation and severity[2]. The development of PHT involves a structural and a dynamic component, the latter accounting for approximately 30% of the increased portal pressure[3]. The presence of hepatocellular nodules (such as regenerative nodules and dysplastic or neoplastic nodules) increases hepatic resistance through a structural mechanism. Still, portal pressure also increases due to the increased hepatic production of endothelin and thromboxane, leading to smooth muscle cell and hepatic sinusoid contraction, which is recognized as the dynamic component of the increased hepatic resistance[3-5].

As portal pressure increases, the synthesis of endothelial vasodilator factors such as nitric oxide, carbon monoxide, adrenomedullin, and prostacyclins also increases. These vasodilator factors induce splanchnic and systemic vasodilation, lowering peripheral vascular resistance and blood pressure[6]. In response to low blood pressure, the renin-angiotensin-aldosterone system (RAAS), adenohypophysis-vasopressin system, and sympathetic nervous system (SNS) become activated[7]. Whereas the first two axes induce a reduction in natriuresis to increase the effective circulating volume, the SNS increases the HR, myocardial contractility, and left ventricle ejection fraction (LVEF), thereby increasing the cardiac output (CO) and generating a hyperdynamic circulatory state[5].

In advanced CLD with cirrhosis, this hyperdynamic circulation is decompensated. Both liver insufficiency and portal-systemic shunts increase the levels of vasodilator factors and decrease the density and reactivity of beta-adrenergic receptors to vasoconstriction. Moreover, the increased plasma concentration of angiotensin II leads to dysfunction of the parasympathetic nervous system (PSNS) and, consequently, an imbalance of the ANS. If the hyperdynamic circulation persists, downregulation of beta-adrenergic receptors occurs, impairing vasoconstriction and generating oxidative stress, myocardial damage, and endothelial damage. Finally, despite the increased HR and RAAS activation, CO decreases at this stage, a condition known as cirrhotic cardiomyopathy[8].

For further details on the pathophysiological mechanisms of AD in CLD, please consult the following references[9,10].

AD in CLD

AD is a common finding in cirrhotic patients, with a prevalence of 30% to 67%[11,12]. Patients with CLD frequently have AD and ANS imbalance, even in the early stages of the disease[11-14]. The clinical manifestations of AD include orthostatic hypotension, gastrointestinal discomfort, bladder dysfunction, erectile dysfunction, cardiovascular complications, and cognitive decline. Patients with AD have a hyporeactive circulation that responds inappropriately to major events such as variceal bleeding and sepsis[15]. As AD has a non-specific clinic presentation, its diagnosis can easily be missed.

Although AD is strictly related to neurohumoral activation, its pathogenesis in CLD has not been completely elucidated. Many factors have been suspected to cause AD in CLD, including decreased baroreceptor sensitivity to hypotension, direct axonal damage by alcohol in alcoholic cirrhosis, altered lipid metabolism, vitamin E deficiency, immune mechanisms, and toxic metabolites[15]. However, AD is closely related to hyperdynamic circulation and PHT due to cirrhosis per se rather than the etiology of cirrhosis[5]. Therefore, AD manifestations, such as HRV, would not be related to the cause of the cirrhosis.

The relevance of AD in CLD lies in its prognostic role: It worsens with an increase in the severity of CLD, enhancing the risk of mortality in such patients[12]. Almost 30 years ago, Hendrickse et al[16] reported a mortality rate of 30% in patients with cirrhosis and AD but only 6% in cirrhotic patients without AD. The ways in which ANS imbalance influences CLD are discussed below.

HRV IN CLD

Assessing ANS function through HRV

HRV is broadly used to assess ANS function in different settings, representing the dynamic balance between the SNS and PSNS of the cardiovascular system, which is reflected as variance in time between the consecutive heartbeats (i.e., the inter-beat intervals)[17,18]. These fluctuations can be evaluated using an electrocardiographic register by analyzing the RR interval (Figure 1). After appropriate analysis of the inter-beat intervals, different HRV parameters are generated in the time (SDNN, RMSSD, and pNN50) and frequency domains [high-frequency (HF), low-frequency (LH), and the LF/HF ratio][19]. In recent years, nonlinear markers of HRV have been developed based on chaos theory, which allow assessment of the regularity and complexity of ANS balance. The most common HRV parameters and their interpretations with respect to ANS balance are explained in Table 1.

Figure 1 Typical electrocardiogram recording and obtention of heart rate variability measurements. Starting from a typical electrocardiogram (ECG) recorded using a Holter monitor, after digitalization and edition, the R waves (RR intervals) file or inter-beat interval is obtained, which is used to determine the variation in RR with time and for creating the heart rate variability (HRV) waveform. Three different measurements can be obtained: (1) Time domain HRV: Based on different descriptive statistics and distribution of RR data; (2) Frequency domain HRV: The original HRV trace is separated into a spectrum of frequencies as described in Table 1 (very low frequencies, low frequencies, and high frequencies); and (3) Nonlinear parameters: Based on the complexity and regularity of RR data. ECG: Electrocardiogram; HRV: Heart rate variability; IBI: Inter-beat interval.

Table 1 Most common measurements of heart rate variability.

Variable	Description	ANS interpretation
Time domain measurements
SDNN (ms)	The standard deviation of all the NN intervals	SNS and PSNS
RMSSD (ms)	The square root of the mean of the sum of the squares of differences between the adjacent NN intervals	PSNS
NN50	The number of adjacent NN intervals that differ by more than 50 ms	PSNS
pNN50	The NN50 count divided by the total number of all NN intervals	PSNS
Frequency domain measurements
LF (ms2)	Power in the low frequency range (0.04–0.15 Hz)	SNS and PSNS
HF (ms2)	Power in the high frequency range (0.15–0.4 Hz)	PSNS
LF/HF (nu)	Ratio LF/HF	SNS and PSNS
Nonlinear measurements
SD1/SD2	Quotient between standard deviation 1 and 2 from the Poincaré plot	SNS and PSNS
ApEn	Approximate entropy, measures signal complexity and regularity	SNS and PSNS
DFA	Detrended fluctuation analysis from RRnvs RRn+1 correlations; short (α1) and long (α2) fluctuations	SNS and PSNS

ANS: Autonomic nervous system; SNS: Sympathetic nervous system; PSNS: Parasympathetic nervous system; ApEn: Approximate entropy; LF/HF: Low frequency/high frequency ratio DFA: Detrended fluctuation analysis; SDNN: Standard deviation of “NN” intervals; RMSSD: Root mean square of successive differences between normal heartbeats pNN50: Percentage of adjacent NN intervals that differ from each other by more than 50 ms.

In clinical practice, most HRV indexes can be easily obtained. The appropriate electrocardiogram (EKG) signal for calculating the time and frequency domain parameters of HRV can be obtained via 24-hour Holter monitoring. Thus, the same medical test for evaluating cardiac rhythm alterations (ambulatory EKG monitoring) allows physicians to measure HRV. In fact, the standard Holter monitoring report usually includes the results of HRV analysis[20]. The sensitivity and accuracy of HRV are higher than those of other tests evaluating autonomic function[5], allowing a better assessment of the functioning of the PSNS[21].

Given the relation of HRV with SNS activity[15], diminished HRV shows a dominant sympathetic tone due to either SNS overflow or lower PSNS activity[17]. Traditionally, HF, PNN50, and RMSSD have been associated with PSNS activity, and LF or SDNN with SNS activity[22]. However, recent evidence shows that it is incorrect to completely and exclusively associate the function of a specific ANS branch with a specific HRV parameter[23]. Related hypotheses have been refuted, e.g., that sympathovagal balance can be quantified with the LH/HF index[24]. Nevertheless, there is strong evidence that overall reduced HRV is a marker of poor health. AD, as evidenced by reduced HRV, has been associated with worse overall outcomes in patients with a variety of diseases, such as acute myocardial infarction[19,25], unstable angina[26], arterial hypertension[27], diabetes mellitus[28], dyslipidemia[29], atherosclerosis[30] and sepsis[31], among others[19,32].

HRV in healthy subjects and in CLD patients

Since 1990, several studies comparing CLD patients and healthy subjects have demonstrated that HRV is diminished in patients with CLD[32–34]. This decrease in HRV is more prominent in patients with evidence of PHT, such as those with hepatic encephalopathy (HE)[35], ascites[36], or portal vein thrombosis[37]. For example, it has been reported that patients with cirrhosis, especially those with Child class C disease, had a higher mean HR and lower HRV values than the corresponding values in healthy controls[35]. Similar results were found in studies comparing HRV values in patients with primary biliary cholangitis and healthy subjects[38]. Furthermore, a cohort study of healthy subjects with a median follow-up of 4.2 years showed that HRV as assessed by SDNN and RMSSD was inversely associated with the risk of incident metabolic dysfunction-associated liver disease and fibrosis[39].

HRV as a severity marker in CLD

Several studies have shown that lower HRV values are associated with increased CLD severity when evaluated using the Child-Pugh score[12,16,32,36,40,41]. In particular, those patients with serum hypoalbuminemia (a component of the Child-Pugh score) present significantly lower HRV values in some time and frequency domains[42]. In a study that mainly included Child-Pugh B and C cirrhotic patients, those who died during a 2-year follow-up had lower HRV than did the survivors[32]. In recent studies, HRV has been identified as a predictor of mortality in patients with cirrhosis, exhibiting an effect that is independent of the model for end-stage liver disease (MELD) score[40,43]. Furthermore, the association of reduced HRV with malnutrition in CLD increases the cardiovascular risk and is associated with a poor prognosis before and after liver transplantation (LT)[44]. Moreover, it has been reported that there is a correlation between SDNN, prothrombin activity, and serum albumin concentration, such that as AD increases, prothrombin activity and albumin levels decrease[33]. It has also been demonstrated that the SDNN parameter is reduced in cirrhotic patients as they progress toward decompensation and is further decreased in patients with acute-on-chronic liver failure, potentially serving as a predictor of 90-day mortality[45].

HRV and complications of CLD

Reduced HRV has been associated with a higher risk of complications in patients with CLD. Mani et al[35] reported that the values of SDNN and LF decreased in parallel with worsening encephalopathy in patients with CLD. Two explanations have been proposed for this association: (1) The presence of autonomic neuropathy, particularly of the vagal branch, could delay the intestinal transit, leading to bacterial overgrowth and translocation with consecutive endotoxemia, which may predispose to encephalopathy, and (2) A mechanism has been posited that involves circulating inflammatory cytokines. In agreement with the latter, Mani et al[35] found higher levels of plasma proinflammatory cytokines in patients with CLD than in healthy subjects; the results correlated with the severity of HE. Furthermore, interleukin-6 concentrations were significantly associated with HRV parameters, suggesting a common etiology for both variables[35].

In addition to encephalopathy, reduced HRV, as evidenced by a low SDNN, has been associated with the presence of ascites, regardless of the etiology of CLD[33]. Furthermore, a reduced SDNN was related to the presence of esophagogastric varices and ascites, suggesting that HRV is correlated with the degree of PHT[46]. In the same observational study, an HRV parameter termed SDANN showed acceptable sensitivity and specificity in estimating recent variceal bleeding[46]. Also, a prospective study demonstrated that HRV parameters were independent predictors of cardiac dysfunction and mortality in cirrhosis[47]. No data are available regarding the relationship between HRV and variceal bleeding risk, hepatorenal syndrome, spontaneous bacterial peritonitis, or the pulmonary complications of cirrhosis.

Furthermore, complications of cirrhosis develop when clinically significant PHT, defined as a hepatic venous pressure gradient (HVPG) > 10 mmHg, develops[48]. AD worsens as PHT increases[49], and HRV correlates negatively with HVPG; thus, HRV decreases as HVPG increases[50,51]. Although this indirect relation between HRV and HVPG theoretically supports reduced HRV in patients with other complications of cirrhosis, no empirical evidence has as yet been adduced in the medical literature.

Regarding HCC, it is unknown whether cirrhotic patients with reduced HRV have an increased risk of HCC. However, a small study from Taiwan suggested that HRV could be a prognostic factor in patients with HCC[52].

HRV and mortality in CLD

In patients with severe CLD, the presence of ANS dysfunction is associated with a high mortality rate[10]. Follow-up studies of patients with CLD showed that HRV values were lower in non-survivors than in survivors during the same period of time[32,43].

Scientific studies have also established reduction in HRV as a major independent risk factor of mortality in cirrhotic patients by using both linear[16] and nonlinear HRV indices[40]. Recently, Bhogal et al[43] reported an HRV reduction in the time domain measurements, in particular noting a predictive mortality value that is independent of well-established factors, such as age, sex, use of beta-blockers, and etiology of cirrhosis[43]. At the same time, SDNN was found to be a reliable predictor of mortality regardless of cirrhosis severity scales such as MELD and Child–Pugh SD2[35] and approximate entropy, both nonlinear HRV indexes, are also associated with mortality risk[37].

Finally, several authors have proposed that ANS dysfunction evaluated using HRV should be considered for the long-term clinical monitoring of patients with CLD. Because of its prognostic value, it may represent an important parameter for prioritizing patients with advanced disease awaiting LT[12,43]. In fact, in pediatric patients with acute liver failure, reduced HRV was significantly associated with the presence of HE and with poor outcomes (death or listing for LT)[53].

HRV and LT

LT is the best, indeed definitive treatment for advanced decompensated cirrhosis regardless of its etiology. The procedure for LT is closely related to ANS balance because it has a significant impact on the systemic hemodynamics in cirrhotic patients. During surgery, the great abdominal and thoracic vessels are clamped to allow removal of the native liver. After the donor liver is correctly positioned and the connection between the vessels is reestablished, there is a sudden increase in the cardiac preload, which represents a high hemodynamic stress. This hemodynamic instability is amplified in the presence of AD, where the ability to regulate pressure fluctuations during LT is blunted[54].

Sympathovagal imbalance has been proposed as a risk marker for hypotension in patients with post-reperfusion syndrome of the liver[55]. Additionally, significant cardiac and cerebral circulatory complications in cirrhotic patients with AD have been reported to be associated with early death after LT[56]. However, there is little information regarding hemodynamic changes during LT in patients with AD or their long-term implications; therefore, further studies must be conducted in this area.

The hyperdynamic circulatory state can resolve after LT, along with the restoration of portal pressure and ANS balance. In fact, some post-LT studies reported regression of AD in 63% to 70% of patients[57,58], supporting the hypothesis that AD is mainly due to the circulatory and neurohumoral changes in cirrhosis rather than the result of structural nerve damage. The potential reversibility of AD after transplantation points toward HRV as a potential evaluation tool. Although cirrhotic patients awaiting LT have lower HRV than healthy subjects, some HRV parameters nevertheless normalize after LT surgery[59]. Also, deceleration capacity, another HRV index, has been shown to be one of the best predictors of 1-year mortality after LT in combination with the MELD score and heart rate complexity[60].

Prolonged QT interval (QT), which is negatively correlated with HRV[50], is a common electrocardiographic finding in cirrhotic patients. A QT segment reduction has been associated with improved outcomes in cirrhotic patients undergoing LT surgery[61]. Remarkably, it was observed that patients with prolonged QT before LT surgery showed normal QTc values after transplantation[5,62]. On the other hand, the cardiotoxic immunosuppressant tacrolimus is commonly used in patients after LT. Compared to patients undergoing treatment with a non-cardiotoxic drug, tacrolimus leads to alterations in HRV and increased levels of N-terminal pro B-type natriuretic peptide[63], confirming the relationship between HRV and cardiovascular function after LT.

As most studies are retrospective and have intermediate-term follow-up data, it is not yet possible to state conclusively that AD constitutes a significant risk factor for increased mortality and morbidity during and after LT. Further studies analyzing the changes in HRV before and after transplantation in patients with CLD are required to support the existing evidence and validate HRV as a useful tool in assessing CLD as well as for stratifying patients awaiting LT.

HRV measurement, a noninvasive test that lasts less than 5 minutes and that can even be performed with wearable devices, could be a cost-effective and fast way to measure ANS balance in the setting of LT[64,65]. In contrast, the MELD score requires raking blood samples and waiting for the results, which is costlier and longer, presents interlaboratory parameter variability, and yields inaccurate mortality predictions in some situations[66]. Despite the growing literature on HRV, it must be analyzed carefully because standardized protocols, validity criteria, and widely validated normal values are still lacking[65]. Nevertheless, HRV could become a complement to MELD scores in LT, thus aiding in the development and clinical evolution of patients with CLD, including LT.

HRV and therapy for CLD

In patients with CLD, the use of drugs modifying the autonomic balance is a common practice. The prime example is the administration of beta-blockers for the prophylaxis of variceal bleeding[67]. In this regard, HRV could be useful in fine-tuning the drug dose and reducing adverse drug reactions.

The use of beta-blockers increases HRV values in patients with coronary disease[68] and cardiac failure[69], as well as in healthy people[70]. Cirrhotic patients treated with propranolol showed a significant reduction in the QTc, which is associated with reduced AD and mortality[71]. This is consistent with improvement in the HRV parameters related to vagal function: RMSSD, HF, and sample entropy[72]. Contrariwise, it has been reported that patients without esophageal varices but with clinically significant PHT could benefit from treatment with beta-blockers[73]. As PHT correlates with HRV, it may be a useful noninvasive tool that should be assessed in future studies as a marker of clinically significant PHT.

There is lack of evidence regarding the effects of other drugs commonly used in CLD on HRV. However, a study involving patients with compensated cardiac failure linked furosemide use to decreased RMSSD[74]. In the same manner, the use of spironolactone in patients with cardiac failure has been linked to a decrease in the LF/HF ratio[75]. There are no data available on the effect of either drug on HRV in cirrhotic patients.

Limitations of HRV

Although HRV is a good clinical indicator of ANS function, it has many limitations with respect to its measurement and interpretation. For example, it is influenced by the position of patients at the time of measurement and their respiratory pattern, requiring cooperation from the patient and making the test less reproducible and harder to perform in patients with HE[15,19]. Also, it is important to note that most evidence on HRV in CLD comes from observational studies, and therefore the presence of confounding by medications and comorbidities that influence HRV is possible. Since type 2 diabetes mellitus is associated with CLD and also causes autonomic cardiomyopathy that in turns reduces HRV, further studies that specifically analyze HRV measurement in these conditions are needed[76,77]. Also, future studies on HRV and CLD should consider a prospective design and proper control of bias and confounding to yield better-quality evidence in this field.

Regarding HRV interpretation, there is no clear consensus on the value limits to distinguish normal and abnormal values, making the measurement less objective and limiting its use in clinical practice[19]. In addition, the evaluation of HRV in each patient requires a post-measurement edition of artifacts and signal processing, implying the need for more time and specialized personnel. However, many wearable devices show HRV in real-time, reducing the barriers to implementing HRV in CLD patient evaluation.

NEEDS AND FUTURE PERSPECTIVES

Despite the importance of ANS balance in the natural history of CLD (Table 2), to date, there are neither any recommendations regarding the evaluation of the ANS in these patients nor any clinical guidance on AD in CLD. Symptoms secondary to AD should be detected and managed because it is necessary to establish the criteria for the diagnosis, management, and treatment of this condition, which could improve the prognosis and quality of life of these patients. Thus, the physician must not overlook clinical manifestations of AD, especially in patients with advanced CLD. The study of HRV could offer some advantages in furthering our understanding of this clinical entity.

Table 2 Summary findings of heart rate variability regarding the natural history and clinical management of chronic liver disease.

Potential clinical utility in CLD	Main HRV findings
Diagnosis	↓ HRV in CLD patients compared with healthy subjects. HRV is a risk factor for MASLD development
Disease staging	↓ HRV values are associated with greater severity when evaluated using the Child-Pugh score. Prothrombin activity and plasma albumin are correlated with SDNN
Assessment of associated complications	↓ HRV is associated with encephalopathy, ascites, esophageal varices, and cardiac dysfunction, and it could be a prognostic marker in hepatocellular carcinoma
Mortality risk	↓ HRV increases mortality risk in cirrhotic patients, independently of MELD, Child-Pugh score, and other risk factors
Management in LT	Some HRV parameters normalize after LT. In combination with MELD score, HRV predicts 1-year survival after LT
Follow-up of medical treatment	The use of beta-blockers ↑ HRV in patients with CLD

CLD: Chronic liver disease; LT: Liver transplant; HRV: Heart rate variability; SDNN: Standard deviation of “NN” intervals; MELD: Model for end-stage liver disease.

However, as HRV is a widely known and validated tool, serial HRV measurements should be tested in more clinical scenarios. The utility of HRV in the screening, diagnosis, and clinical evaluation of CLD complications must be evaluated in larger prospective studies (Figure 2). Given the lack of consensus on the diagnosis of AD in CLD, the validation and study of specific cut-off points for the HRV parameters should be encouraged. Also, larger-scale cohort and randomized clinical trials, with better control of bias and confounding, are needed to better understand HRV in CLD natural history and clinical management. On the other hand, consensus guidelines and reference values for HRV measurement are also needed, given the variety of methods for measuring HRV (from 24-hour registers to ultra-short-term, i.e., lower than 5 min)[65]. Such improvements in HRV assessment protocols would allow better recognition of its advantages and make HRV results more reproducible across operators according to the clinical context.

Figure 2 Proposed utility of heart rate variability in chronic liver disease. Heart rate variability can be assessed across different stages of chronic liver disease progression with different aims at each stage.

For clinical follow-up, HRV could be useful in the future in the continuous monitoring of patients with CLD from early stages (Figure 2). With respect to the therapeutic management of patients with CLD, HRV could be useful in dose titration and the therapeutic adjustment of beta-blockers and other drugs; however, prospective studies are necessary to support this strategy. Conversely, while the use of angiotensin-converting enzyme inhibitors[5] and beta-blockers has been studied in cirrhotic patients with AD, further evidence is needed. Finally, further large-scale studies with a higher level of scientific evidence are needed to establish the usefulness of HRV in risk assessment, monitoring, and therapeutic management of LT.

CONCLUSION

HRV is a well-validated tool in the assessment of ANS. In CLD, this noninvasive test could be useful in the diagnosis and assessment of the severity, complications, therapy, and prognosis of the disease (Figure 2 and Table 2); however, further studies are needed to validate its clinical utility. Thus, despite its limitations, HRV could become an important tool in the integral management of patients with CLD.