Sarcopenia is associated with longer hospital stay and multiorgan dysfunction in alcoholic hepatitis

Abstract

Introduction:

Excessive consumption of alcohol has steadily risen to become the third leading cause of preventable death in the United States. One such consequence of heavy alcohol use that has been under considerable investigation of late is alcoholic hepatitis (AH). While many risk factors for developing AH have been documented, our aim in this study was to examine the potential association between sarcopenia and the severity, mortality, 30 days readmission rate, complication, infections and length of hospital stay in patients with AH.

Methods:

A retrospective analysis was performed at a large, academic hospital in one hundred and ninety-four AH patients ages 18-60 who had cross sectional CT imaging and met our clinical definition of AH. The 5^thpercentile of the psoas muscle index was used as a cut off for sarcopenia.

Results:

One hundred and ninety-four patients met the criteria for AH and had cross sectional imaging. Higher MELD score was found in the sarcopenia group when compared to the non-sarcopenia group (mean MELD 21.5 and 24.2, respectively, p = 0.03). Sarcopenia also correlated with significantly longer hospital stay; the average length of stay in the sarcopenia group was 17.2 days while the non-sarcopenia patients had an average of 12.4 days. We found higher risk of developing pneumonia , sepsis and hepatic encephalopathy in sarcopenic patients.

Conclusion:

AH patients with sarcopenia have significantly worse outcomes when compared with the patients without sarcopenia, including a severe form of AH, longer hospital stays, higher risk of developing PNA, sepsis and HE.

Introduction:

Excessive consumption of alcohol has steadily risen to rank as the third leading cause of preventable death in the United States [1, 2]. This statistic encompasses the multitude of effects that alcohol can have on a healthy liver’s architecture and function. This includes both short-term and long-term consequences including permeant liver damage, cancer, and damaged social and family relationships in addition to death [3]. One of the acute manifestations of the alcohol induced liver disease is alcoholic hepatitis. While the term “alcoholic hepatitis” is generally used in the clinical setting, AH refers to both a clinical syndrome and a specific set of histopathological findings. The clinical syndrome generally manifests as jaundice, nausea, fever, and tender hepatomegaly with laboratory testing usually significant for moderately elevated transaminases with an aspartate aminotransferase (AST) to alanine aminotransferase (ALT) ratio greater than two. More severe presentations include upper-quadrant pain and encephalopathy [4]. The pathohistological findings commonly seen include micro- or macrovesicular steatosis, hepatocellular ballooning with cytoplasmic rarefaction, and infiltration by neutrophils (which is characteristic of alcoholic hepatitis) [5]. Previous studies have shown that patients hospitalized for AH result in significant healthcare cost, even higher than that of patients hospitalized in the United States for myocardial infarction, stroke, or acute pancreatitis [6].

Model for End-Stage Liver disease (MELD) score is a scoring system that predicts liver transplant waitlist survival and has successfully decreased waitlist mortality among liver transplant candidates by directing organs to the most ill patients [7]. However, one of the major limitations of the MELD score is that it does not include an assessment of the patients’ nutritional and functional status, which has consistently been shown to negatively affect outcomes following liver transplantation and survival [8]. Some studies went beyond that and proposed to include nutrition into the MELD score [9]. The nutrition status also plays an important role in the long and short outcomes of patient with AH due to the fact that the increased in the energy expenditure requires a good nutrition intake to support the catabolism status of patient with AH [10].

Advances in research over the last few years has helped describe multiple risk factors associated with worse outcomes in patients with AH including genetics (such as polymorphisms in the tumor necrosis factor (TNF)-α promoter region), underlying chronic infectious hepatitis, and presence of nonalcoholic steatohepatitis, and malnutrition [11, 12].

Many previous studies evaluated the role of nutrition in AH and showed that there is a superior survival in those who received extensive nutrition. Sarcopenia considered as one of the major components of malnutrition in cirrhotic patient and was associated with poor outcome in liver transplant patients but the clinical impact of sarcopenia in AH is yet to be studied.

Sarcopenia which is defined as progressive loss of skeletal muscle mass and strength is thought to progress through a complex multifactorial pathogenesis, which involves not only age-related changes in neuromuscular function but also alcohol-related chronic pro-inflammatory state and oxidative stress [13]. Alcohol has a multi-organ effect that extends to the skeletal muscle in addition to the liver. While many studies focus on liver cirrhosis and liver transplantation, in this study we evaluated the correlation between sarcopenia and alcoholic hepatitis’ outcomes.

Methods:

Patient Selection

After using ICD code of the alcoholic hepatitis, a retrospective analysis was performed at a large, academic hospital included a hundred and ninety-four AH patients ages 18-60 who had cross sectional CT imaging and met our clinical definition of alcohol hepatitis. AH was defined as recent alcohol consumption with elevation in the AST and ALT (ratio >1.5), bilirubin >3 mg/dl, no other causes of liver disease [14]. Data collected from medical charts included gender, age, BMI, albumin, international normalized ratio (INR), sodium, PT, creatinine, gastrointestinal bleeding (GI bleed), pneumonia (PNA), urinary tract infection(UTI), spontaneous bacterial peritonitis(SBP),sepsis, systemic inflammatory response syndrome (SIRS), hepatic encephalopathy(HE), acute kidney injury (AKI), thirty-day readmission, length of stay/hospitalization, BMI, MELD score and the prescience of sarcopenia. We divided the cohort into two groups; sarcopenia and non- sarcopenia.

Sarcopenia was defined by using the cut off values from the published literature after controlling for the sex and the age [15]. The psoas muscle index (PMI) was calculated by measuring the surface area of psoas muscle bilaterally through the cross-sectional imaging at the level of L3-L4 disc line. Total surface area to the height was adjusted to calculate the PMI. The cut-off values (cm²/h²) of sarcopenia using the 5^th percentile of the PMI was used as follow: Male age <50:12.27, female age<50: 10.47, Male age ≥ 50: 10.47 and female age ≥ : 10.33. The institutional review board at our hospital/medical school approved this study, with a waiver of patient informed consent according to HIPAA regulations HHS 45 CFR 46.101.

Definition of Study Events

The primary outcome of our study was to assess the potential association between sarcopenia and the severity of AH. Secondary end points were to assess correlation of sarcopenia with mortality; 30 days readmission rate, length of hospital stay, infections; PNA, UTI, SBP, sepsis, and complications; GI bleed, HE and AKI.

Statistical Analysis

Data were analyzed using the IBM SPSS version 24 statistical software for Mac (IBM Corporation, Armonk, NY, USA). Mean, percentages, and SDs of the mean were used to examine the demographics of the target population. Outcomes were determined using Odds Ratio. The threshold for statistical significance was set at p-values <0.05, with odds ratios reported together with 95% confidence intervals.

Results:

Demographics of Alcoholic Hepatitis

One hundred and ninety-four patients met the criteria for AH and had cross sectional CT imaging; the mean age was 47.6 +/− 10.4 years. Male gender was predominant (70%), and Caucasians represented 93% of the cohort. (Table 1).

Table 1.

Patient Characteristics by Sarcopenia

	No- sarcopenia N=125	sarcopenia N=69	P-value
Age (years), Mean +/− SD	48.3 +/− 10.1	45.1 +/− 10.9	0.04
Sex, N (%)
Female	20 (16.0)	38 (55.1)	<.0001
Male	105 (84.0)	31 (44.9)
White Race, N (%)	116 (93.6)	65 (94.2)	0.56
BMI, Mean +/− SD	29.6 +/− 6.5	24.5 +/− 5.9	<.0001
MELD, Mean +/− SD	21.5 +/− 6.9	24.2 +/− 8.5	< 0.03
Death, N (%)	21 (16.8)	13 (18.8)	0.72
30-Day Readmission, N (%)	24 (19.4)	18 (26.1)	0.25
Duration of Hospitalization (days), Mean +/− SD	12.4 +/− 9.6	17.2 +/− 13.0	0.003

^*standard deviation (SD); body mass index (BMI); Model for End-Stage Liver disease (MELD)

Demographics of Sarcopenia

Patients with sarcopenia were younger when compared to the non-sarcopenia group, 45.1 years compared to 48.3 years , p = 0.04. A lower BMI was also seen in the group with sarcopenia, 24.5 compared to 29.6, p = <0.001. Finally, sarcopenia was predominantly found in females (p<0.0001) (table 1).

Clinical Outcomes

Higher MELD score was found in the sarcopenia group when compared to the non-sarcopenia group (mean MELD 21.5 and 24.2, respectively, p = 0.03). Sarcopenia also correlated with longer hospital stay. The average length of stay in the sarcopenia group was 17.2 days while the non-sarcopenia patients had an average length of stay of 12.4 days, p = 0.003. As illustrated in table 1, while not statistically significant, a larger percentage of patients had a 30-day readmission in patients with sarcopenia (26% vs. 19%, respectively) and a larger percentage of death occurred with the sarcopenia group. The risk of HE in the sarcopenia group has 2.5 the risk compared to the non-sarcopenia group with percentages of 40% and 21% respectively, P-value <0.05 (Table 2).

Table 2:

The risk of complications and infection

	No-sarcopenia N=125	sarcopenia N=69	OR (95% CI)	P-value
Complications, N (%)
AKI	46 (36.8)	34 (49.3)	1.67 (0.92-3.03)	0.09
HE	27 (21.6)	28 (40.6)	2.48 (1.30-4.71)	0.006
GI bleed	27 (21.6)	17 (24.6)	1.19 (0.59-2.38)	0.63
Infection, N (%)
UTI	14 (11.2)	5 (7.2)	0.62 (0.21-1.80)	0.38
SBP	3 (2.4)	5 (7.2)	3.18 (0.74-13.71)	0.12
PNA	22 (17.6)	23 (33.3)	2.34 (1.19-4.62)	0.01
Sepsis, N (%)	24 (19.2)	25 (36.2)	2.39 (1.23-4.64)	0.01
SIRS, N (%)	35 (28.0)	14 (20.3)	0.66 (0.32-1.32)	0.24

^*acute kidney injury (AKI); hepatic encephalopathy (HE); gastro intestinal (GI); urinary tract infection (UTI); spontaneous bacterial peritonitis (SBP); pneumonia (PNA); systemic inflammatory response syndrome (SIRS)

While no correlation was found between sarcopenia and the risk of developing UTI, SBP, GI bleed and AKI, the risk of sepsis in the sarcopenia group was 2.3 times the risk of sepsis in the non-sarcopenia group with percentages of 36% and 19% respectively, P-Value <0.05. Our data also showed a higher rate of PNA in the sarcopenia group when compared to the non-sarcopenia group with percentages of 33% and 17% respectively, OR of 2.3 and P-value of 0.01 (Table 2).

Obesity, Sarcopenia and AH

Sarcopenia at least partially is related to nutrition status. Thus, we evaluated the BMI in sarcopenic AH patients. Of the 40 patients with sarcopenia, 10 patients had BMI ≥30 and 30 patients had normal BMI (20-24.9) as we excluded the overweight (BMI 25-29.9) and the under-weight patients. . Upon comparing the two groups (BMI>30 and normal BMI) to assess the potential effect of obesity on sarcopenic AH patients, we found that AH patients with sarcopenia obesity and BMI ≥30 had higher MELD score and more risk of developing organ failure including AKI and HE and when compared to the normal weight AH patients with sarcopenia but these differences did not quite reach a statistically significant level (Table 3). We also found that infections, sepsis and death risk was higher in the obese/sarcopenic AH patients when compared with the normal weight AH but was not statistically significant (Table 3).

Table 3:

Characteristics, Complications, and Comorbidities by BMI Categories in Patients with Sarcopenia

	Sarcopenia+ BMI >30 N=10	Sarcopenia+ BMI 20-24.9 N=30	OR, >30 vs. 20- 24.9 (95% CI)	P value
MELD, Mean +/− SD	23.5 +/− 6.9	20.6 +/− 5.4	1.09 (0.96-1.23)	0.19
BMI, Mean +/− SD	35.0 +/− 7.5	21.9 +/− 1.5	-	0.50
Complications, N (%)
AKI	4 (40.0)	11 (36.7)	1.15 (0.27-4.99)	0.85
HE	5 (50.0)	12 (40.0)	1.50 (0.36-6.32)	0.58
GI bleed	1 (10.0)	8 (26.7)	0.31 (0.03-2.81)	0.30
Infection, N (%)
UTI	0 (0.0)	1 (3.3)	-	0.98
SBP	1 (10.0)	1 (3.3)	3.22 (0.18-56.88)	0.42
PNA	4 (40.0)	9 (30.0)	1.56 (0.35-6.88)	0.56
Sepsis, N (%)	5 (50.0)	7 (23.3)	3.29 (0.73-14.74)	0.12
SIRS, N (%)	2 (20.0)	4 (13.3)	1.62 (0.25-10.58)	0.61
Death, N (%)	2 (20.0)	3 (10.0)	2.25 (0.32-15.90)	0.42
30-Day Readmission, N (%)	3 (30.0)	9 (30.0)	1.00 (0.21-4.77)	0.99
Duration of Hospitalization (days), Mean +/− SD	15.6 +/− 12.1	16.1 +/− 15.5	1.00 (0.95-1.05)	0.92

^*standard deviation (SD); body mass index (BMI); Model for End-Stage Liver disease (MELD); acute kidney injury (AKI); hepatic encephalopathy (HE); gastro intestinal (GI); urinary tract infection (UTI); spontaneous bacterial peritonitis (SBP); pneumonia (PNA); systemic inflammatory response syndrome (SIRS)

Discussion:

The prevalence of alcoholic hepatitis has been increasing in the last several years with studies hoping to improve patient outcomes and offering liver transplantation to a highly selected group of patients with AH [16]. The MELD score has been historically used as the sole predictor to help determine three-month survival in patients awaiting transplantation, candidacy for liver transplantation or the severity of alcoholic hepatitis. A major flaw of this score is the underrepresentation of the patient’s nutritional status. Sarcopenia is one of the major components of the nutritional status in cirrhosis and has been associated with serious health consequences in terms of frailty, disability, morbidity and mortality [17]. In patients following liver transplantation, DiMartini et. al. demonstrated the negative impact of malnutrition on longer hospital and intensive care unit (ICU) stay, infections, graft impairment, morbidity and mortality [18]. Tandon et. al, showed that the mortality rate in sarcopenia patients is two times the rate in non-sarcopenia patients whom they are undergoing liver transplantation [19].

Sarcopenia is often related to malnutrition in liver diseases and numerous studies showed the importance of nutrition in chronic liver diseases. The nutritional status is also important in patients with alcoholic hepatitis and nutrition affects survival in AH [20]. We found that sarcopenia was more frequent in alcoholic patients with patients with low BMI. Motano-lorza et. al found that the median survival in sarcopenia patients undergoing liver transplant evaluations is almost half the survival duration of the non-sarcopenia patients with a median survival of 19 ± 6 and 34 ± 11 months, respectively [21].

Two large studies, reported by Mendenhall et al, found that there is a correlation between alcoholic hepatitis and malnutrition, these studies found there is a linear correlation between the severity of malnutrition and severity of alcoholic hepatitis [22]. While the direct effects of alcohol on the muscle are poorly characterized, alcohol and its metabolites disturb protein homeostasis that is critical for normal muscle function [23]. Other mechanisms leading to sarcopenia are likely related to the effects of pro-inflammatory cytokines on the muscle as pro-inflammatory cascade activation is a major characteristic of alcoholic hepatitis. Our study supports what Mendenhall et al concluded, in that malnutrition in the form of sarcopenia is associated with severe disease of alcoholic hepatitis. Our results showed that the higher MELD score in sarcopenia patients when compared to the non- sarcopenia patients with mean of 24.2 +/− 8.5 and 21.5 +/− 6.9, respectively.

The liver is the site of production of almost all coagulation factors including fibrinogen, thrombin, and factors V, VII, IX, X, and XI [24]. Defects in the liver’s synthetic function seen in alcoholic hepatitis may lead to derangements in these proteins which can be observed as an elevated INR. For this process, the liver is not only responsible for the creation of polypeptide chains and glycosylation but also other appropriate post-translational modifications, all of which can be disrupted in acute and/or chronic liver damage seen in alcoholic hepatitis [25, 26]. Similarly, as discussed above, we postulate that these biochemical functions are further disrupted in patients who have protein, lipid, and carbohydrate depletion as seen in sarcopenia.

Complication in the forms of features of portal hypertension like hepatorenal syndrome, ascites or hepatic encephalopathy also reported to have a strong correlation with the degree of malnutrition [27]. Lattanzi et al in his prospective study reported that muscle depletion was associated with a higher risk of developing HE [28]. Bhanji et al stated that the risk of HE in sarcopenia is 2.25 times the risk in non-sarcopenia cirrhotic patients while Chang et al in their met analysis of 1795 patients also found a similar result, sarcopenia was positively associated with HE with an OR of 2.74 [29, 30]. While Merli, Bhanji and Chang studies were focusing on cirrhotic patients, our study focused only on AH. Our results showed a similar finding as the sarcopenia patients have a higher risk of developing HE with OR of 2.48, p-value of 0.006 when compared with the non-sarcopenia patients.

A meta-analysis of 12 randomized trials by Hmoud et. al. found a cumulative incidence of infection of 20% in patients with AH [31]. Similarly, Takagi et. al. where surgical patients with sarcopenia had a significantly higher incidence of in-hospital mortality and infectious complications [32].

Gariballa et. al. demonstrated that elderly patients with sarcopenia had longer hospital length of stay, higher mortality, more depressive symptoms, and higher rate of six-month readmission [33]. Kim et al in their systematic review found that the length of stay in cirrhotic patients with sarcopenia is twice the duration in the non-sarcopenia patients and the risk of sepsis in sarcopenia is 2.8 the risk in non-sarcopenia patients [34]. Our data support the same conclusion as our results showed that the risk of sepsis in sarcopenia is 2.39 the risk in non-sarcopenic patients with alcoholic hepatitis.

In the geriatric population, studies found that there is higher risk of developing pneumonia in sarcopenia patients when compared to the non-sarcopenia patients according to Altuna-venegas et al. [35]. Our data also showed that pneumonia is higher in the sarcopenia group as its carries two times the risk of developing pneumonia when compared to non-sarcopenia group with an OR of 2.34, P-value of 0.001.

When compared to the non-sarcopenia patients, our study showed that the sarcopenia patients have longer length of hospital stay 12 Vs 17 days (P- value<0.05). Thus, the longer length of stay in the sarcopenia group can be explained by the higher rate of hepatic encephalopathy, pneumonia, sepsis or the severity of the disease.

Sarcopenia seems to be the result of complex interactions involving inadequate nutritional intake, impaired synthesis of glycogen (of which a large majority is produced in the liver), concomitant hypermetabolism, and protein synthesis derangements [36].The pathogenesis of sarcopenia in combination with the complex interplay found in AH between ethanol metabolism, inflammation, and alteration in innate immunity could partially explain the difference in hospital outcomes between those with sarcopenia and those without. Mortality in severe alcoholic hepatitis correlates with development of dysfunction or failure of other organs. Our observations of the AH patients with sarcopenia had increased frequency of acute kidney injury, hepatic encephalopathy, pneumonia and sepsis indicates that sarcopenia correlated with multi-organ dysfunction in AH.

Obesity and sarcopenia are frequently encountered in patients with cirrhosis and recent studies showed that both of them are associated with poor outcomes. Hara et al reported that sarcopenia and sarcopenic obesity were associated with poor outcome in cirrhotic patients [37]. Kamo et al showed that sarcopenic obesity patients have worse survival after liver transplant when compared to non-sarcopenic/non-obese patients [38]. In contrast to what Hara et al and Kamo concluded in their studies, our data suggested that obesity does not have an impact on the severity, hospital stay, infections or complications when compared to non-sarcopenic/non-obese patients. Our result might be related to the small size of the cohort that underwent the sub group analysis, (40 patients) and larger studies are needed to confirm our findings.

The average age of the patients in our study with sarcopenia was 45.1 years old which is significantly different than the majority of studies conducted on sarcopenia. A recent meta-analysis conducted by Yang et. al. had a mean age of 83.7 years old [39]. This result clearly demonstrates a gap in knowledge about alcoholic hepatitis patients and about the potential implications sarcopenia has on their morbidity and mortality at a significantly younger age than generally studied. Although, sarcopenia is usually associated with elderly or chronic ill patients, the age of our cohort was relatively young when compared to the published literatures, this indicates that alcohol consumption when accompanied with poor nutrition and sarcopenia is highly associated with poor outcome. Indeed, this group of patients has a nutrition status equivalent to chronically ill patients with an age of 65 and older [40].

Our paper has limitations. First as with any retrospective review, information collection is incomplete, particularly regarding follow-up evaluation for patients. We had an uneven distribution on the number of patients in the sarcopenia and non-sarcopenia groups. Finally, due to the small sample size used for our study several of our secondary outcomes were not statistically significant, 30-day readmission rate, and in-hospital mortality, likely secondary to a lack of power. Again, as any retrospective study it is important to note that this type of study is unable to define exact causality. Further larger sample and prospective studies are needed to help confirm the observation seen in our study.

Conclusion:

Alcoholic hepatitis is a one of the most severe consequences of heavy alcohol intake with a high mortality rate in its severe form. Many risk factors for increased morbidity and mortality have been characterized and studied. Sarcopenia, generally defined as a loss of significant muscle mass, is found to have a poor prognosis and outcomes in liver transplanted patients. Here we show that AH patients with concurrent sarcopenia have significantly worse outcomes when compared with the patients without sarcopenia, including a severe form of AH and longer hospital stay. Higher risk of developing pneumonia, sepsis, hepatic encephalopathy was also found in AH patients with sarcopenia. We suggest that malnutrition in the form of sarcopenia should be assessed in the patients with excessive alcohol consumption admitted for alcohol hepatitis especially in the beginning of their hospital course as it may help to trigger focus on the nutritional need of these patients. Larger and prospective studies are needed to confirm our finding.

Key Points:

1. Sarcopenia is associated with severe form of AH.

2. Patients with sarcopenia are more likely to develop pneumonia and sepsis.

3. Patients with sarcopenia are more likely to develop hepatic encephalopathy and have longer hospital stay.

4. Obesity does not affect patients’ outcome in AH patients with sarcopenia.

Abbreviations:

(AH)

alcoholic hepatitis

(ALT)

aspartate aminotransferase

(AST)

to alanine aminotransferase

(MELD)

Model for End-Stage Liver disease

(TNF)

tumor necrosis factor

(INR)

international normalized ratio

(GI bleed)

gastrointestinal bleeding

(PNA)

pneumonia

(UTI)

urinary tract infection

(SBP)

spontaneous bacterial peritonitis

(SIRS)

systemic inflammatory response syndrome

(HE)

hepatic encephalopathy

(AKI)

acute kidney injury

(PMI)

psoas muscle index

(ICU)

intensive care unit