Introduction

Cirrhosis is the final stage of chronic liver disease, marked by extensive hepatic fibrosis with possible consequences for hepatic function. The liver has an essential function in protein, carbohydrate, and fat metabolism [1]. Cirrhosis patients commonly show a reduction in muscle strength and muscle mass [2, 3], which is associated with decreased muscle protein synthesis [4]. As a result, this leads to a decreased ability to perform daily activities, ultimately impacting their quality of life [5,6,7].

The physiological changes resulting from liver cirrhosis encompass impaired hepatic function, electrolyte imbalances, and malnutrition [8]. Due to nutritional and metabolic abnormalities, patients with liver cirrhosis have an imbalance between protein synthesis and degradation [9], which can lead to changes in body composition and further complications such as sarcopenia and sarcopenic obesity [10, 11]. The reported incidence of sarcopenia in patients with liver cirrhosis ranges from 40 to 62% across studies [12]. Furthermore, sarcopenia has been found to be positively associated with increased mortality in liver cirrhosis patients [13, 14], and even increased mortality risk more than doubled in those with sarcopenia [2]. Therefore, measuring body composition is necessary for accurately evaluating the degree of malnutrition in cirrhotic patients [15].

Cirrhosis can lead to symptoms such as fatigue, abdominal discomfort, ascites, and edema [16,17,18]. Fatigue is a state of perception that arises from an abnormal perception of effort [19]. Patients with fatigue often experience persistent tiredness and a feeling of heaviness in their limbs, which can limit their motivation to engage in daily activities and social interactions [20]. The underlying mechanism of fatigue is not fully understood, as it may involve both peripheral and central factors [21]. However, interventions such as behavioral changes, medication, dietary adjustments, and exercise are commonly used to alleviate this symptom [22].

Exercise training can promote physical and mental health, and both aerobic and resistance exercises have been shown to improve liver fat metabolism [23, 24], as well as body composition, strength, and cardiorespiratory fitness in patients with chronic liver disease [25]. Moreover, exercise training can also alleviate fatigue in patients with chronic liver disease [26].

Aamann et al. published a systematic review and meta-analysis on exercise intervention in patients with liver cirrhosis in 2018, with the primary purpose of determining whether exercise intervention reduces mortality and serious adverse events [27]. However, more studies have been conducted in recent years, making it imperative to revise the existing data and conduct additional research on the effects of exercise interventions on fatigue and quality of life in patients with cirrhosis.

This meta-analysis seeks to evaluate the effectiveness of exercise interventions in improving body composition, exercise capacity, fatigue levels, and quality of life to provide a comprehensive analysis and evidence-based recommendations regarding patient care options for individuals with liver cirrhosis.

Methods

This meta-analysis was registered in PROSPERO (registration number: CRD42022376961) prior to initiating the study, and this study followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines. The PRISMA checklist can be found in the Online Resource 1.

Search Strategy

This study searched for relevant studies published before November 2022 in databases including PubMed, Embase, Cochrane, and CINAHL. The search focused on exercise interventions in patients with liver cirrhosis, with outcomes including body composition, exercise capacity, fatigue, and quality of life.

The search terms encompassed all subject terms and free terms of “liver cirrhosis,” “exercise,” “physical fitness,” “fatigue,” and “quality of life.” The literature search was conducted in English without limitations on the earliest publication date. Additional details are provided in the Online Resource 2.

Study Selection

The current study was undertaken by two researchers who individually reviewed all the relevant literature and carried out a verification process. In instances when a disagreement emerged among the researchers and a consensus could not be achieved, the researchers sought the input of a third party to engage in discussion prior to arriving at a decision.

We only included studies that conducted randomized controlled trials (RCTs). The interventions in the RCTs for the experimental group could be exercise alone or exercise in combination with amino acid supplementation. In contrast, the control group got routine care. The outcome measures encompass numerous variables that are assessed to evaluate the effectiveness of the intervention.

Data Extraction

Two independent reviewers utilized a pre-decided data extraction chart to present the data from each study. The table included the author’s name, nationality, year of publication, intervention for the experimental and control groups, sample size, type of exercise, duration, frequency, measurement tools, and outcomes. An outcome parameter was subjected to meta-analysis only if it had been reported in at least two studies. When a study had missing data, we contacted the corresponding author to request additional information. If an outcome measurement evaluates the parameter in a manner contrary to other assessment tools, we reverse the scores.

Risk of Bias Assessment

The Cochrane Collaboration's tool for assessing risk of bias (RoB) 2.0 was used in this study to evaluate the trial's quality. It is comprised of five domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result, for quality assessment. We conducted a publication bias analysis using Egger's test within the Comprehensive Meta-Analysis Version 3.0 (Biostat Inc, Englewood, NJ, USA). If the p-value of Egger's test is greater than 0.1 and the confidence interval includes 0, it indicates no publication bias [28].

Statistical Methods

In this study, we used the Cochrane Review Manager software (RevMan 5.4) to perform statistical analysis. Mean difference (MD) was used for pooling studies with identical variables, and when the variables were not directly comparable, we employed the standardized mean difference (SMD).

The degree of heterogeneity was assessed using the I2 statistic, with values of 25%, 50%, and 75% representing low, moderate, and high heterogeneity, respectively [29]. Since the true effects of each study may differ and the weight of each study cannot be determined by the sample size alone, we used the random-effects model for all analyses.

Subgroup analysis was performed under the following circumstances: when different intervention methods were employed, or when similar parameters were measured using different assessment tools. This study is an analysis of data that have already been published, and thus it does not need approval from an ethics committee.

Results

Initially, a total of 739 relevant articles were identified. After removing 29 duplicated articles using EndNote X9 and eliminating four more duplicates manually, we screened the titles and abstracts of 706 articles, excluding 684 articles. Twenty-two articles met the inclusion criteria and were subjected to full-text assessment. Of these, eight articles were excluded because they were either trial register entries, posters, or abstracts lacking complete articles. Among the fourteen remaining articles, six were excluded due to the following reasons: two articles had both the experimental and control groups receiving exercise intervention; one article was not a randomized controlled trial, one article was not published in English, and two were review articles. As a result, eight randomized controlled trials were included in this meta-analysis, and the literature search process is presented in Fig. 1.

Fig. 1
figure 1

The flow diagram of study selection

Full size image

The total sample size throughout these eight randomized controlled trials consisted of 220 participants, with each individual study including a range of 17 to 40 participants. The average age of the participants was 58.02 years old with the Child–Pugh stages A, B, or C. The basic characteristics of the eight studies are presented in Table 1. The duration of exercise intervention was 8 weeks in two studies [30, 31], 12 weeks in five studies [32,33,34,35,36], and 14 weeks in one study [37]. Two studies included in this analysis used a combination of exercise and amino acid supplementation as the intervention [33, 34]. As for types of exercise, one study used resistance training [36]; five studies conducted aerobic exercises [30, 31, 33, 34, 37]; two studies had a combination of aerobic and resistance exercise [32, 35].

Table 1 The characteristics of the included studies
Full size table

Risk of Bias and Quality

Using RoB 2.0, the literature's risk of bias was categorized into three levels: low risk, moderate risk, and high risk. Overall, the inadequate blinding of participants and assessors was the main source of risk of bias. Figure 2 displays the bias risk assessment results, while detailed information about bias across studies is available in Online Resource 3.

Fig. 2
figure 2

Risk of bias assessment

Full size image

The p-values derived from all Egger’s tests were greater than 0.1, indicating no statistically significant asymmetry in the funnel plot.

Body Composition

Body composition parameters, such as body mass index, thigh circumference, thigh muscle thickness, fat mass, and fat-free mass reported in more than two studies, were incorporated into a meta-analysis.

Six studies provided BMI data [30, 33,34,35,36,37]. The effect of exercise intervention on BMI showed no statistically significant difference (MD = 0.76, 95% CI − 0.16 to 1.69, Z = 1.62, p = 0.11, I2 = 0%) (Fig. 3).

Fig. 3
figure 3

Forest plot of exercise versus control group and exercise combined with amino acid versus control group, outcome: BMI (kg/m2)

Full size image

Further subgroup analysis was conducted according to the interventions. The subgroup analysis revealed that there was no significant difference in BMI in either the sole exercise group (MD = 0.86, 95% CI − 0.16 to 1.89, Z = 1.65, p = 0.10, I2 = 8%) or the exercise combined with amino acid supplementation group (MD = − 0.73, 95% CI − 4.22 to 2.76, Z = 0.41, p = 0.68, I2 = 0%).

Six studies [30,31,32, 34,35,36] reported thigh circumference values (Fig. 4). The findings of the meta-analysis indicated that the implementation of exercise intervention resulted in a statistically significant rise in thigh circumference (MD = 1.26, 95% CI 0.12 to 2.39, Z = 2.16, p = 0.03, I2 = 0%). Interestingly, further subgroup analysis showed that thigh circumference did not significantly increase in either the sole exercise group or the exercise combined with amino acid supplementation group.

Fig. 4
figure 4

Forest plot of exercise versus control group, outcome: thigh circumference

Full size image

Three studies reported data on thigh muscle thickness through the average compression index measured by ultrasonography [30, 31, 35]. The combined findings of the meta-analysis revealed a statistically significant increase in average compression index (MD = 0.07, 95% CI 0.00 to 0.14, p = 0.04, I2 = 0%) (Fig. 5).

Fig. 5
figure 5

Forest plot of exercise versus control group, outcome: average compression index

Full size image

Two studies measured fat mass [33, 36]. The pooled results of the meta-analysis indicated that the implementation of exercise intervention did not yield a statistically significant impact on fat mass (SMD = 0.30, 95% CI − 0.29 to 0.90, Z = 1.01, p = 0.31, I2 = 0%) (Fig. 6). Both studies also provided data on fat-free mass [33, 36]. The meta-analysis results showed that the introduction of exercise intervention did not result in a statistically significant effect on fat-free mass (SMD = 0.34, 95% CI − 0.25 to 0.94, p = 0.26, I2 = 0%) (Fig. 7).

Fig. 6
figure 6

Forest plot of exercise versus control group, outcome: fat mass

Full size image
Fig. 7
figure 7

Forest plot of exercise versus control group, outcome: fat-free mass

Full size image

Exercise Capacity

Six studies presented the outcomes of six-minute walk distance (6MWD), allowing data aggregation for meta-analysis [30, 31, 33,34,35,36]. The experimental group receiving exercise intervention showed a significant increase of 68.93 m in 6MWD (95% CI 14.29 to 123.57, Z = 2.47, p = 0.01), but with moderate heterogeneity (I2 = 73%) (Fig. 8). Subgroup analysis was conducted based on the interventions of the experimental group, which involved either exercise alone or exercise combined with amino acid supplementation. The results showed that the improvement was not significant in the four studies that solely used exercise as the intervention [30, 31, 35, 36] (MD = 37.41, 95% CI − 15.20 to 90.01, Z = 1.39, p = 0.16, I2 = 61%). On the other hand, the analysis of the two studies [33, 34] that combined exercise with amino acid supplementation showed a significant increase in the 6MWD, with a mean difference of 144.72 m (95% CI 87.44 to 202.01, Z = 4.95, p < 0.00001, I2 = 0%).

Fig. 8
figure 8

Forest plot of exercise versus control group and exercise combined with amino acid versus control group, outcome: exercise capacity

Full size image

Quality of Life

Seven studies reported numerical values regarding the impact of exercise intervention on quality of life [30, 31, 33,34,35,36,37] (Fig. 9). Among the seven studies, four used the Chronic Liver Disease Questionnaire (CLDQ) [30, 31, 35, 37]; two used the 36-item Short Form (SF-36) [34, 36], and one used the Sickness Impact Profile [33]. The pooled analysis revealed a tendency toward better scores in the exercise intervention group; however, this finding did not reach statistical significance (MD = 0.29, 95% CI − 0.01 to 0.58, Z = 1.91, p = 0.06, I2 = 0%). Subgroup analysis indicated no significant difference in the results across different subgroups, suggesting that the exercise had no significant impact on quality of life regardless of the assessment tool used.

Fig. 9
figure 9

Forest plot of exercise versus control group, outcome: quality of life

Full size image

Fatigue

Four research studies used CLDQ and reported the levels of fatigue in people diagnosed with cirrhosis [30, 31, 35, 37]. Results indicated that the intervention group had significantly improved fatigue scores in comparison to the control group, demonstrating a mean increase of 0.70 points (95% CI 0.38 to 1.03, Z = 4.23, p < 0.0001). There is a statistically significant difference and no heterogeneity was observed (I2 = 0%), as shown in Fig. 10.

Fig. 10
figure 10

Forest plot of exercise versus control group, outcome: fatigue

Full size image

Discussion

The findings of this study indicate that exercise interventions have a notable positive effect on exercise capacity, thigh muscle thickness and thigh circumference, as well as a reduction in fatigue, among individuals with cirrhosis. However, there was no statistically significant disparity observed between the groups that received exercise intervention and those that did not, when assessing body composition using BMI, fat mass, and fat-free mass. Subgroup analysis further showed that regardless of whether supplementing amino acids or not, short-term exercise intervention had no significant effect on BMI and fat mass in patients with liver cirrhosis.

This meta-analysis revealed that patients diagnosed with cirrhosis experienced a notable improvement in exercise capacity after engaging in exercise intervention, contrasting with the findings of the 2018 meta-analysis by Aamann et al. [27]. However, our subgroup analysis indicated that exercise intervention alone was not significantly effective in increasing the 6MWD among these patients. Notably, the combination of exercise with amino acid supplementation led to a marked enhancement in 6MWD. This is particularly relevant given the common occurrence of protein malnutrition in cirrhosis due to reduced liver function. Leucine, a branched-chain amino acid (BCAA), is emphasized for its role in stimulating muscle protein synthesis. Clinical trials and guidelines from the European Association for the Study of the Liver [8] recommend BCAA supplements, especially for advanced liver cirrhosis, to improve nutritional status, delay liver failure, and enhance overall health [38,39,40]. Our meta-analysis, however, included only two studies on exercise combined with amino acid supplementation, one with leucine [34] and the other with unspecified essential amino acids [33]. Furthermore, the potential impact of baseline differences in albumin levels, as observed in the study by Chen et al. [33], on the outcomes remains unaddressed, raising questions about the influence of such factors on our meta-analysis. While there is extensive research on the protein needs of cirrhosis patients, a clear consensus on estimating their nutritional requirements is still lacking, underscoring the need for further research in this area.

Assessing body composition in cirrhosis patients precisely is challenging due to fluid distribution imbalances across compartments. Anthropometry and bioelectrical impedance results may be affected by edema [41]. Using magnetic resonance imaging (MRI) or computed tomography to measure muscle mass would be ideal; however, only one study included in our analysis used MRI to measure lean mass, and their results showed a significant increase of 3% in whole-body lean mass (fat-free mass) [36].

The type and intensity of exercise among different studies may directly affect the results of muscle mass. Resistance exercise has demonstrated a better effect on muscle protein synthesis. However, in the studies included in this meta-analysis, five studies only used aerobic exercise as a therapeutic intervention. The intensity of aerobic exercise interventions was mostly moderate, such as 60–80% baseline peak VO2 [30] or 60 to 80% of heart rate reserve [31, 37]. Aamann et al. employed resistance exercise, starting with 15 repetitions per session and gradually increasing the resistance while reducing the repetitions to 8 per set in the final two weeks; the exercise intensity was moderate [27]. Ideally, the inclusion of detailed measurement data on the intensity and duration of exercise interventions in each study would significantly enhance the comprehensiveness of the research. However, progress in this research is currently hampered by a lack of basic patient data, including body weight and oxygen consumption. In addition, detailed information on the specific exercise interventions utilized in the various studies is limited, which is also a significant challenge. The lack of data impedes accurate assessment of exercise interventions, thus constraining our ability to draw more nuanced conclusions from these studies.

Observations revealed a noticeable increase in thigh circumference. While thigh circumference may not serve as a direct indicator of thigh muscle mass, Zenith et al. employed ultrasound as a method to evaluate thigh muscle thickness alongside thigh circumference in the studies under consideration [30]. The findings of their investigation suggest that the observed increase in thigh circumference can be attributed to the growth of muscular tissue. Likewise, in the study by Aamann et al., the experimental group showed an increase in both thigh circumference and fat-free mass. This indirectly confirms that the rise in thigh circumference is associated with muscle growth [36]. Furthermore, the pooled data on average compression index indicated a significant increase in thigh muscle thickness.

Recent research has documented notable enhancements in both muscle strength and muscle mass among individuals diagnosed with cirrhosis and sarcopenia following a 24-week regimen of branched-chain amino acid (BCAA) supplementation [42]. Additionally, Hernández-Conde et al. observed enhanced muscle mass in patients with cirrhosis after a 12-week intervention involving exercise and BCAA supplementation, significantly higher than the control group receiving only exercise [43]. However, it is crucial to acknowledge that the existing literature presents divergent findings. In Mohta et al.'s research, it was observed that the addition of BCAA supplementation to standard treatment and exercise did not result in a significant increase in muscle mass in cirrhotic patients with sarcopenia [44]. Unfortunately, studies integrating amino acids with exercise interventions did not include measurements of muscle mass. Consequently, we are unable to conduct an analysis comparing the effectiveness of exercise alone versus exercise combined with amino acid supplementation on muscle mass.

Patients with liver cirrhosis often experience a substantial decline in their quality of life across various daily living domains [45, 46]. Symptoms like muscle cramps, while not life-threatening, can profoundly distress patients and deteriorate their quality of life [47]. In our study, although there was a positive trend observed in the experimental group, the improvement in quality of life did not reach statistical significance, aligning with the findings of Aamann et al. [27]. Notably, within the seven studies included in our meta-analysis that assessed quality of life, the study by Kruger et al. reported lower baseline quality of life scores in the experimental group [31], which could have influenced the overall outcome of our analysis. Upon re-evaluating this aspect and excluding the data from Kruger et al., the combined results from the remaining six studies indicated a significant enhancement in the quality of life. This finding suggests that exercise interventions are likely to positively impact the quality of life in cirrhosis patients.

Fatigue in individuals diagnosed with liver cirrhosis can be attributed to various factors, such as diminished hepatic functionality, metabolic abnormalities, insufficient nutritional intake, and the presence of other associated complications. Studies by Ahl et al. and Iwasa et al. have documented a reduction in cerebral blood flow in individuals with liver cirrhosis [48, 49], a phenomenon associated with the presence of fatigue [50]. We observed significantly reduced fatigue levels in the exercise intervention groups. Exercise promotes better blood circulation by enhancing cardiovascular health. This enhanced circulation ensures that vital organs, including the brain, receive an adequate supply of oxygen and nutrients, thereby potentially reducing fatigue and lethargy [51, 52]. Furthermore, engagement in exercise elicits the secretion of endorphins, an endogenous compound possessing analgesic attributes that enhance an individual's emotional state. Endorphins possess the capacity to mitigate the depressive symptoms commonly linked to chronic illnesses like cirrhosis, thereby indirectly ameliorating fatigue resulting from psychological distress [53, 54]. Patients with liver cirrhosis can indeed benefit from exercise as a means of reducing fatigue and enhancing their overall health if they approach it correctly. The type, duration, and intensity of exercise should be tailored to the health status and physical capabilities of the individual. However, investigating fatigue is a challenging task, given its subjective nature and absence of a clear and precise definition [55].

There are several limitations to this meta-analysis, including a limited number of studies and small sample sizes. The inclusion criteria or potential baseline imbalances in a single trial could influence the results of this meta-analysis. Additionally, the included studies varied in terms of the stage of liver cirrhosis, baseline nutritional status, type of exercise intervention, exercise intensity, and exercise duration, which may increase the heterogeneity of the results. Moreover, out of 220 participants, 15 (6.8%) were lost to follow-up.

In this meta-analysis, five studies included subjects with Child–Pugh class A and B; two studies involved subjects with Child–Pugh class A only; and only one study included a population with class B or C. Further research is needed to investigate the effects of exercise on decompensated liver cirrhosis patients.

Conclusion

Exercise intervention significantly improved exercise capacity and thigh muscle thickness in patients with liver cirrhosis. However, as a standalone intervention, exercise did not result in a significant enhancement of exercise capacity. To achieve better improvements in exercise capacity, it is recommended to combine exercise with amino acid supplementation. Furthermore, exercise significantly reduced fatigue levels, but improvements in quality of life, BMI, fat mass, or fat-free mass were not significant.