Stratification and selection of therapies to improve survival in severe alcoholic hepatitis

Abstract

Severe alcoholic hepatitis (SAH) is associated with high short-term mortality. The SAH population exhibits extreme heterogeneity in disease severity, clinical presentation, decompensations, and outcomes. Nonetheless, improving outcomes and preventing adverse events is a major challenge when selecting an appropriate treatment for alcoholic hepatitis. Currently, steroids are the standard of care for SAH with Maddrey’s discriminant function > 32 and model for end stage liver disease > 20; however, they have limited usage due to ineligibility in approximately two-third of such patients. Approximately 25% of patients do not respond to steroids and require alternative therapies. An array of evolving therapies, such as granulocyte colony-stimulating factors, plasma exchange, fecal microbiota transplantation, antibiotics, anti-cytokine therapies, and N-acetylcysteine, showing variable success, are emerging. Hence, it is also crucial to select appropriate therapy. The present review discusses the standard of care, the existing therapies, risk stratification for outcomes, and the selection of appropriate therapy to improve survival in SAH patients.

Key Words: Severe alcoholic hepatitis; Steroids; Granulocyte colony stimulating factor; Plasma exchange; Fecal microbiota transplantation

Core Tip: Clinically, severe alcoholic hepatitis (SAH) presents with jaundice, with or without ascites or encephalopathy, and it progresses rapidly. The patient population is heterogeneous and needs to be stratified according to their eligibility for specific therapies. Emerging alternatives or rescue therapies are available for SAH. The decision for the preferred therapy is based on patient profile, clinical experiences, and its side effects profile. This comprehensive review provides insight into the stratification of patients who need specific therapies, with the primary focus on the therapies that improve survival in SAH.

INTRODUCTION

Patients with severe alcoholic hepatitis (SAH) can present with or without underlying cirrhosis. Ascites manifesting with jaundice indicate advanced disease and can develop at varied intervals after an acute insult to the liver[1]. Patients can present with alcoholic hepatitis without decompensation, acute decompensation of preexisting cirrhosis, or acute-on-chronic liver failure (ACLF). Development of jaundice with bilirubin > 5 mg% and ascites within 28 days qualifies for ACLF presentation by the Asian Pacific Association for the Study of Liver criteria indicating a severe course of the disease[2]. The development of ascites after four weeks of jaundice indicates acute decompensation, not mounting to ACLF. The prognosis differs with or without ACLF, and the rate of mortality increasing with ACLF grades. Patients with acute decompensation have a better prognosis, especially in the first episode of presentation, than those with ACLF[3]. In the absence of alcohol abstinence, long-term survival after index presentation is around 30% only, owing to the high rate of recidivism[4]. Survival is low in patients who present second or additional episodes of alcoholic hepatitis with cirrhosis and decompensation[5,6]. Hence, this heterogeneous population can be clinically classified based on prognosis, as shown in Figure 1.

Figure 1 Clinical sub-classification of patients with severe alcoholic hepatitis. Index presentation patient can present either without acute-on-chronic liver failure (ACLF) with varying degrees of fibrosis or ACLF, mostly fulfilling Asian Pacific Association for the Study of Liver criteria. Previously diagnosed cases can decompensate with relapse and can present either as Acute or non-acute decompensation or ACLF.

STRATIFICATION OF PATIENTS

Model for end stage liver disease (MELD) score is a reliable predictor of short-term mortality in patients with SAH[1]. This has been shown that patients with a MELD score > 21 have a significantly higher risk of death within 90 days than those with lower scores. This cutoff value not only aids in identifying high-risk patients but also helps clinicians make informed decisions regarding the urgency of liver transplantation. MELD > 30 should be evaluated for liver transplantation at the time of admission while considering other therapies[2]. In addition, steroid therapy is effective in SAH with MELD up to 39; however, the efficacy decreased with increasing MELD score[7]. Hence, the MELD score can be used for prognostication, identification, and selection of patients for liver transplantation or steroid therapy. Maddrey’s discriminant function (mDF) has been used to define the severity of alcoholic hepatitis. Most centers consider patients with mDF > 32 only for specific therapies, whereas those with mDF < 32 can be managed with nutrition therapy and abstinence[8]. Patients with mDF < 32 have a survival rate of > 90% at 6 months and a survival rate of > 70% at 5 years if they practice abstinence[9]. Although MELD is prognostically better, than mDF, many researchers have avoided including patients with an mDF > 90 in clinical trials. In our experience, an mDF > 90 is not a contraindication for steroids, unless sepsis and organ failures preclude its usage.

Therapies

Abstinence and nutrition therapy: Abstinence is indispensable for improving survival in patients with SAH. In clinical practice, a subgroup of patients who improved rapidly with nutrition therapy alone post-abstinence were categorized as rapid fallers. Nutrition therapy should be initiated at the time of admission to achieve 35–40 kcal/kg/day in all patients. Small frequent meals can increase palatability and reduce the period of fasting. Nasogastric feeding should be considered in patients with an oral calorie intake of < 21 kcal/kg/day. The improvement in rapid fallers on nutrition and abstinence can be predicted by the trajectory of serum bilirubin, precisely by more than < 20% decrease, on day 7. Such patients can be deferred to corticosteroids due to the lack of additional survival advantage[10]. However, many such patients show partial improvement and may require subsequent therapies. Hence, the trend in liver function in the first week is crucial for therapeutic decisions. Although recommending cutoffs for improving liver function is difficult, a bilirubin reduction of more than > 20% in the first week can be significant[10]. In such patients, the absence of encephalopathy or ascites can predict near-zero mortality in the short term, thus avoiding steroid exposure[1].

Steroids: Indications, patient selection, and outcomes: Different trials have determined variable eligibility of the steroids. Patients with sepsis and organ failure were excluded from almost all clinical trials. Acute kidney injury has been considered a relative contraindication, with different investigators using variable cutoff values for steroid eligibility. Creatinine up to 5.7 mg% has been included in steroid eligibility, in the absence of the need for renal replacement therapy (Table 1)[8]. Traditionally, Gastrointestinal (GI) bleeding is considered a contraindication for steroids; however, recent evidence of steroid use in such patients could alter our future practice. Steroids increase the risk of GI bleeding mildly, which can be effectively mitigated by endo therapy and using steroids after 48 h post-bleeding. Hence, GI bleeding should not be considered an absolute contraindication for steroid use[11]. Patients with severe comorbidities, such as alcohol withdrawal, psychosis, electrolyte disturbances, or hepatic encephalopathy (HE) > grade II, should be stabilized before prescribing steroids. Thus, evidence of sepsis should be searched in this population before starting steroid therapy. Although mDF > 90 is debatable for steroid prescription, evidence suggests that the decision should not be based solely on mDF but rather on the presence of sepsis and organ failure. Also, patients were classified as steroid-eligible or -ineligible. Since the outcomes in the steroid-ineligible population are poor, alternative therapies can be considered to improve survival[6].

Table 1 Inclusion and exclusion of major trials on corticosteroids in severe alcoholic hepatitis.

Trial	Inclusion criteria	Exclusion criteria
STOPAH trial	Clinical diagnosis of alcoholic hepatitis with mDF > 32	Patients with creatinine > 5.7 mg% or on renal support. Patients with gastrointestinal bleed and creatinine up to 5.7 mg% were not excluded. Also, patients with resolved sepsis were not excluded
Louvet et al[13] 2009	Severe alcoholic hepatitis with mDF > 32 and onset of jaundice within 3 months. Patients with treated infections were included	Gastrointestinal bleed within 15 days, peptic ulcers, Hepatitis B and C, neoplasm
GpreAH study	Clinical or biopsy proven alcoholic hepatitis with mDF < 90	Patients with sepsis, creatinine > 1.5 and recent gastrointestinal bleed were excluded
STASH trial	Clinical diagnosis of alcoholic hepatitis	Acute kidney injury, recent gastrointestinal bleed and active infections were excluded

mDF: Maddrey’s discriminant function.

Steroid non-response: Steroid non-response was defined by a Lille score > 0.45 on day 7 (Figure 2). These patients are at increased risk of infections and high short-term mortality. Granulocyte Colony Stimulating factor (GCSF) therapy can be used as rescue therapy, as discussed subsequently.

Figure 2 Algorithm for management of severe alcoholic hepatitis. Steroids are the mainstay of treatment. Response to steroids is assessed at day 7 by Lille score. Steroid non-responder or steroid-ineligible can be considered for rescue therapies on case to case basis. SAH: Severe alcoholic hepatitis; LT: Liver transplant; MELD: Model for end stage liver disease; mDF: Maddrey’s discriminant function; TLC: Total leucocyte count; AKI: Acute kidney injury; PLEX: Plasma exchange; GCSF: Granulocyte colony stimulating factor.

Steroid regimens: The most common regimen used for SAH is 40 mg of prednisolone for four weeks, followed by two weeks of tapering. Since infection risk increases after the second week, several studies have attempted early tapering of steroids. A multicenter trial showed that the risk of infections is reduced with a weekly tapering regimen, i.e., by 10 mg every week starting in the second week, resulting in comparable survival rates[12,13].

Steroid-ineligible SAH: Currently, there are limited options to manage steroid-ineligible alcoholic hepatitis, although a large proportion of ineligible patients due to infections can become steroid-eligible after antibiotics treatment. Steroid therapy can be considered in > 90% of such patients, continuing antibiotic therapy, with resolved infection (13). Patients with controlled GI bleed can be considered for steroids after the bleed is controlled[11], and those with renal failure can be considered for steroid therapy if they are not on renal replacement therapy. Collectively, a large proportion of the ineligible population can be made eligible with supportive treatments. For those with more than two extrahepatic organ failures, candidature for liver transplantation should be evaluated while considering alternative therapies.

GCSF: GCSF improves liver failure and prevents infections in patients with SAH[14]. Randomized trials have shown that GCSF is safe and improves 90-day survival compared to standard medical therapy[15,16]. A recent randomized controlled trial by our group[17] suggests that GCSF improves steroid responsiveness, and combining GCSF with steroids provides better results than monotherapy[17]. However, all trials on GCSF excluded patients with active sepsis[17-19]. Additionally, GCSF can aggravate hemophagocytosis (HLH); hence, patients with cytopenia should be evaluated for HLH before considering GCSF. Also, the outcomes of GCSF alone are similar to steroids with decreased risk of infections and, hence, can be a suitable therapy as a bridge to liver transplantation in such patients[18]. Although European studies have shown opposite results, the heterogeneity of the patient population with organ failure may be one of the reasons for poor outcomes, while the other may be the variable regimens used in trials[20]. Currently, GCSF is the only medical therapy for steroid non-responders that reduces mortality. In a nutshell, GCSF therapy can be utilized alone or in combination with steroids in selected patients with SAH. The inclusion and exclusion criteria of various trials for GCSF therapy have been summarized in Table 2.

Table 2 Patients excluded from granulocyte colony stimulating factor therapy in randomized trial.

Ref.	Intervention/control arm and inclusion criteria	Exclusion criteria	Survival at day 90
Singh et al[19] (2014)	GCSF+SOC, SMT	Upper gastrointestinal bleeding during the previous 10 days, hepatocellular carcinoma or portal vein thrombosis, hepatorenal syndrome, grade 3 or 4 hepatic encephalopathy, uncontrolled bacterial infection, human immunodeficiency virus infection, and other liver disease etiologies	18/23 (78.2%); 05/23 (21.7%)
Singh et al[18] (2018)	GCSF+SMT, SMT+GCSF+NAC, SMT	Same as above	13/19 (68.42%); 16/18 (88.8%); 6/20 (30%)
Shasthry et al[16] (2018)	GCSF, SMT	Lille score < 0.45 at day 7. Multiorgan failure	9/14 (64.2%); 4/14 (28.4%)
Tayek et al[42] (2022)	GCSF + SMT, SMT	Pregnancy, uncontrolled infection, known human immunodeficiency virus infection, recent upper gastrointestinal hemorrhage, white blood cell > 30000/mm3, creatinine > 2 mg/dL, and AH therapy for more than three days before randomization	12/16 (75%); 15/18 (83.2%)
Mishra et al[17] (2024)	GCSF, GCSF+Predni-solone, Prednisolone	Serum creatinine > 1.5 mg/dL, total leukocyte count > 25000/cmm, mDF > 90, hemoglobin of less than 10 g/dL, active infection, recent upper gastrointestinal bleeding, and the existence of any other liver disease	33/42 (78.5%); 37/42 (88.1%); 27/42 (64.2%)

GCSF: Granulocyte colony stimulating factor; SOC: Standard of care; SMT: Standard medical therapy; NAC: N-acetylcysteine.

Plasma exchange: Plasma exchange (PE) lessens the onset of multiorgan failures in individuals with ACLF and improves systemic inflammation[21]. Compared to medical treatment, PE is linked to a greater resolution of systemic inflammatory response syndrome, a lower and delayed onset of multiorgan failure, and fewer deaths from liver failure in a large multinational cohort study[22]. It removes inflammatory cytokines, damage-associated molecular patterns, and endotoxins from the blood. Moreover, PE improves monocyte function and mitochondrial respiration in responders as well as removes large non-dialyzable and pathogenic macromolecules in the progression of liver failure. These findings suggest that while PE is beneficial across different etiologies of ACLF, 65% of ACLF patients treated with PE were alcoholics[22]. Notably, these studies excluded patients with multiorgan failure, sepsis, GI bleeding, and coagulopathy; these patients should be optimized before considering PE. Low-dose steroids combined with PE have been used to treat SAH in a study in India, wherein it improved one-year survival[23,24]. The authors suggested that centrifugal PE with a slow removal rate provides better results than membrane-based PE. Although patients with sepsis and GI bleeding were excluded from this study, the authors showed a positive impact of PE on renal dysfunction. Hence, patients with alcoholic hepatitis presenting as ACLF with or without renal dysfunction in the absence of sepsis or multiorgan failure can be favorable candidates for PE therapy (Table 3). The selection criteria for various therapies based on current evidence is summarized in Table 4.

Table 3 Plasma exchange in Severe alcoholic hepatitis.

Ref.	Inclusion	Exclusion criteria	Survival at day 90 in intervention group
Kumar et al[24]	APASL defined alcohol ACLF	Liver illness due to other causes (such as drug- or virus-induced hepatitis, Wilson's disease, autoimmune hepatitis, or nonalcoholic fatty liver disease), sepsis, bleeding that has occurred recently (within the last three months), cancer, acute pancreatitis, or human immunodeficiency virus infection	35.6%
Kumar et al[24]	EASL defined ACLF	Hepatocellular cancer, patients with PLEX contraindications, such as infection, recent gastrointestinal bleeding, and hypotension	70% with centrifugal plasma exchange
Ramakrishnan et al[23]	ACLF defined by APASL with AARC score 8-10 and not responding to medical therapy	Nonalcoholic etiologies, coexisting hepatocellular carcinoma or extra-hepatic malignancy, bleeding, pregnancy, multiorgan failure (≥ 3 organ failures), severe pre-existing cardiopulmonary disease, APASL-AARC score > 10 or < 8, mechanically ventilated and requiring inotropic support at the time of enrolment were excluded from the study.	64%

APASL: Asian pacific association for study of liver; AARC: Asian pacific association for study of liver acute on chronic liver failure research consortium; EASL: European association for study of the liver; ACLF: Acute on chronic liver failure.

Table 4 Selection criterias for various therapies in severe alcoholic hepatitis.

Therapy	Selection criteria	Remarks
Corticosteroids	Severe alcoholic hepatitis with mDF > 32 or MELD > 20 in absence of active sepsis and multiorgan dysfunctions. MELD > 39 should be avoided to exposed to corticosteroids	GI bleed and renal failure are relative contraindications. Corticosteroids can be considered on case to case basis in absence of other contraindications. Patients with resolved infections can be considered for steroid therapy
GCSF	Severe alcoholic hepatitis with mDF > 32 or MELD > 20 in absence of active sepsis and multiorgan dysfunctions. Patients with severe cytopenias, hemoglobin < 8 g/dL or leukocytosis with TLC > 30 k should be avoided for GCSF therapy. Patient with hypersplenism (spleen size > 15 cm) should be avoided	Limited literature exist regarding use of GCSF in resolved infection, GI bleed or Acute kidney injury. Hence it should be avoided till further evidence supports its use
Plasma exchange	Severe alcoholic hepatitis with mDF > 32 or MELD > 20 in absence of active sepsis and multiorgan dysfunctions. ACLF patients with progressive liver failure without extrahepatic failures are best candidates	Patients with acute renal injury in absence of other extrahepatic organ failure may benefit from therapy. Steroid non responders can be considered on case to case basis. Patients with GI bleed or resolved infection can be considered for plasma exchange
Fecal microbiota transplantation	Severe alcoholic hepatitis with mDF > 32 or MELD > 20 in absence of active sepsis and multiorgan dysfunctions	Patient with intestine paralysis should be avoided for FMT

MELD: Model for end stage liver disease; mDF: Maddrey’s discriminant function FMT: Fecal microbiota transplantation GI: Gastrointestinal; ACLF: Acute on chronic liver failure; GCSF: Granulocyte colony stimulating factor; TLC: Total leucocyte count.

Fecal microbiota transplantations: Fecal microbiota transplantations (FMTs) show promising results in the treatment of SAH. Several studies have demonstrated improved survival rates and clinical outcomes in patients with SAH receiving FMT compared with those receiving standard therapies. A study comparing FMT to corticosteroids, pentoxifylline, and nutritional support showed that SAH patients receiving FMT had the highest three-month survival rate of 75% compared to 38%, 30%, and 29% in the other groups, respectively[25]. Another trial found that 90-day survival was significantly higher with FMT (75%) than prednisolone (56.6%)[26]. Interestingly, FMT modulated the gut microbiome in beneficial ways. Some studies have observed an increase in beneficial bacteria, such as Bifidobacterium, and a decrease in pathogenic species following FMT[26,27]. FMT was associated with improvements in clinical parameters, such as the resolution of HE and ascites[27,28]. Pande et al[26] and Philips et al[27] enrolled patients with mDF < 90 and without sepsis and organ failure, i.e., a steroid-eligible population. The rate of infection was lower with FMT than with steroids, although not statistically significant. Hence, FMT can be used as an alternative if steroids are contraindicated. However, the role of FMT in steroid-ineligible SAH remains unclear. The major trials evaluating FMT in SAH are summarized in Table 5.

Table 5 Fecal microbiota transplantation in severe alcoholic hepatitis.

Ref.	Study design and inclusion criteria	Exclusion criteria	Infection rate
Pande et al[26] 2022	RCT, Adults with SAH	Creatinine > 1.5, MELD > 35; mDF > 90, GI bleed within 1 months	18.2%
Sharma et al[28] 2022	Prospective cohort study. Adults with SAH	Other causes of liver disease, uncontrolled infections, uncontrolled upper gastrointestinal bleeding, grade 3 or 4 hepatic encephalopathy, more than two organ failures, malignancy, intestinal paralysis, or perforation	30%
Philips et al[27] 2022	Retrospective. Adults with Steroids ineligible SAH	Hemodialysis at admission, disseminated intravascular coagulation, multi-organ failure, uncontrolled sepsis on inotropes, and other liver disease etiologies were excluded, as was gastrointestinal bleeding within a month	NA

RCT: Randomized controlled trial; SAH: Severe alcoholic hepatitis; MELD: Model for end stage liver disease; mDF: Maddrey’s discriminant function.

Pentoxifylline: Pentoxifylline has shown mixed results in the treatment of SAH. Some studies suggested potential benefits, while others indicated limited efficacy compared to other treatments. Several studies have shown that pentoxifylline reduces the incidence of fatal hepatorenal syndrome (HRS) compared to placebo[29] and also improves short-term survival rates[30]. However, the overall effect on mortality remains controversial. The STOPAH study showed no benefit of pentoxifylline for medium- and long-term mortality[8]. Interestingly, a study by De et al[31] found pentoxifylline to be superior, with lower mortality rates and renoprotective effects compared to prednisolone[31]. Another trial compared a combination of corticosteroids and pentoxifylline with corticosteroids alone and found no additional survival advantage with combination therapy[32]. In conclusion, although pentoxifylline may offer some benefits in treating SAH, particularly in reducing the risk of HRS, its overall efficacy remains uncertain. Therefore, further investigation is essential to determine the role of pentoxifylline in SAH management.

N-acetylcysteine: N-acetylcysteine (NAC) is promising in the treatment of SAH. In a randomized controlled trial, combination therapy with prednisolone and NAC significantly improved one-month survival compared with prednisolone alone (8% vs 24% mortality) in patients with SAH. However, the primary outcome of six-month survival did not improve significantly (27% vs 38%) mortality[33,34]. Nonetheless, the improvement was attributed to the renoprotective effects, thereby reducing the incidence of kidney injury in the study group and a decline in the infection rate. NAC combined with GCSF did not provide any additional benefits for survival, suggesting that a reduction in infection rates could be the mechanism for improved survival, which is significant when combined with other steroids[18].

Combination therapies: SAH is associated with defective hepatocyte regeneration, progressive inflammation, bacterial translocation from the gut, and dysbiosis, leading to progressive liver and extrahepatic organ failures. A single therapy may not be optimal for combating all pathogenic mechanisms and rebalancing hepatocyte homeostasis. Hence, a combination of anti-inflammatory and regenerative therapies of those that modulate gut microbiota can improve survival. Combination therapies with survival advantages over monotherapy include GCSF with NAC, GCSF plus prednisolone, and PE with low-dose steroids.

Other therapies: Various anti-cytokine therapies, such as tumor necrosis factor (TNF) and interleukin-1 antagonists, have failed to show survival advantages[34,35].

Nonetheless, TNF antagonists increase the risk of infections and mortality, suggesting that some of these cytokines are essential for hepatocyte regeneration[34]. Also, S-adenosyl methionine, antibiotics, and zinc did not provide survival benefits[36-38].

Liver transplantation: Liver transplantation for SAH has been debatable due to ethical concerns and the traditional six-month abstinence rule. However, recent studies have shown promising results for early liver transplantation in carefully selected SAH patients. Mathurin et al. reported improved survival rates for patients with SAH who received early transplants compared to those who did not receive timely treatment[39]. The study observed a six-month survival rate of 77% for early transplant recipients vs 23% for non-recipients with steroid non-response defined by Lille score > 0.45. Interestingly, alcohol relapse rates in SAH patients who received early transplants were similar to or even lower than those in patients with alcoholic cirrhosis who met the six-month abstinence criterion[40]. However, there are concerns about harmful drinking patterns post-transplantation in patients with SAH, with approximately 15%–20% of the population having early relapse and poor graft function, emphasizing careful patient selection and assessment[41,42]. In conclusion, although early liver transplantation for SAH remains controversial, emerging evidence suggests that it is a lifesaving option for carefully selected patients who do not respond to medical therapy. Also, the traditional six-month abstinence rule may need to be reconsidered in light of these findings. Nevertheless, in-depth studies are required to determine the long-term outcomes and develop validated models for predicting alcohol relapse risk in this population. The expected survival on various therapies and liver transplantation is summarized in Table 6.

Table 6 Expected survival on various therapies and liver transplantation in severe alcoholic hepatitis.

Therapy	Expected 28 days survival	Expected 90 days survival	Expected 6 months survival
Corticosteroids[8,17]	80% to 85%	60% to 70%	50% to 55%
GCSF[17,18]	80% to 90%	75% to 80%	60% to 65%
Plasma exchange[23,24]	60% to 70%	35% to 65%	20% to 30%
Fecal microbiota transplantation[26-28]	75% to 80%	70% to 75%	Variable
Liver transplantation	80% to 85%	80% to 85%	69% to 85%

CONCLUSION

Not all patients with SAH need specific therapies. Specific therapies should be given to patients not improving on supportive treatment and abstinence. The decision to start specific therapies can be based on improvement on serum bilirubin at day 7. Steroids are the mainstay of treatment for SAH and have been shown to improve survival. The GCSF and PE are potential alternatives and can be life-saving in carefully selected patients. Fecal microbiota transplantation is also emerging as disease-modifying therapy, but data are limited. The combination of therapies may improve survival compared to mono-therapies, but more research is needed before using such combinations in practice.