Alcohol consumption is one of the leading risk factors for liver disease with significant morbidity and mortality [1]. Alcohol-related liver disease (ALD) includes a wide disease spectrum, from the simple liver steatosis, to the acute inflammatory condition of alcoholic hepatitis (AH) and ultimately, advanced liver fibrosis and cirrhosis [2]. Alcohol use disorder (AUD) consists in a harmful drinking pattern of more than 3 drinks/day (≥21/week) in men or more than 2 drinks/day (≥14/week) in women and this pattern of alcohol use is the main preventable risk factor for ALD [3]. AH represents the acute phenotype of alcohol-related liver disease, characterized by intense liver inflammation, with increased short and long-term mortality [4], [5]. Over the years, multiple scoring systems have been implemented that enhance disease stratification and ease clinical key decision-making and therapeutic approach [6], [7], [8], [9]. Although much is known about the disease pathogenesis, little progress has been made towards the refinement of therapeutic options. For over 50 years, steroids remain the most important treatment for severe AH, albeit a success rate of less than 50% in reducing mortality [9], [10], [11]. This article details the current-knowledge regarding ALD and specifically AH, highlighting on one hand the unmet theragnostic and prognostic needs, and on the other hand future directions and novel non-invasive biomarkers.
The latest European guidelines recommend a new terminology for the whole alcohol-related liver disease spectrum, in order to prevent patient stigmatization. The term “alcoholic” has been replaced in the medical nomenclature, hence alcoholic liver disease has been renamed to alcohol-related liver disease or alcohol-associated liver disease. Exceptionally, AH has become too standardized and remains unmodified, but it might be subject to change in the future [12]. Additionally, the Delphi panel defined a new nomenclature under the term MetALD, which includes patients with both the criteria for cardiovascular risk and chronic alcohol intake. MetALD has not been studied before and will benefit from inclusion in clinical trials for an integrated therapeutic approach [13].
Alcohol use disorder (AUD) has important medical and socio-economical impact, with multi-organ morbidity, neoplasia risk, decreased professional productivity and increased domestic and social violence. Globally, AUD is responsible of 4% of deaths and over half a million of individuals die yearly from alcohol-related cirrhosis [14]. Advanced ALD varies between 2–5% in the population at risk, with substantial difference between age, gender and pattern of consumption [15]. Harmful alcohol dinking pattern is the main risk factor for the development of ALD. Traditionally, over 2–3 drinks/day was considered the threshold for the development of ALD, although most patients with advanced liver disease consume higher quantities. However, multiple recent studies have shown that there are no health benefits to drinking and even minute amounts of alcohol have deleterious health consequences [16]. The development of advanced forms of ALD is dependent on the dose, duration and type of alcohol [17]. While daily drinking is associated with the development of cirrhosis, binge drinking, defined as at least five drinks for men and four drinks for women in about 2 hours, is particularly related to the risk of AH [18]. However, some patients with ALD never develop AH, although drinking similar amounts of alcohol, suggesting other risk factors. Notably, drinking outside meals and women have a higher risk, probably due to the interaction of alcohol metabolites with the increased body fat component, while also having a lower gastric alcohol dehydrogenase activity [19]. Some studies have shown that high blood alcohol levels also promotes liver fibrosis [20]; however, the exact role of binge drinking pattern as a risk for cirrhosis remains yet to be established. Regarding alcohol type, limited data suggest that liquor has the highest risk for ALD, as opposed to beer or wine, partially due to their antioxidant properties [21]. There is a wide individual variety regarding disease progression attributed to comorbidities, mainly metabolic syndrome and genetic factors. Genetic variants of α-1 antitrypsin, patatin-like phospholipase domain containing protein 3 (PNPLA3), transmembrane 6 superfamily member 2 (TM6SF2), membrane bound O-acyltransferase domain containing 7 (MBOAT7) have been linked to ALD progression and severity [22]. Interestingly, these gene variations have also been linked to cirrhosis progression in diabetic patients [23] and both PNPAL3 and TM6SF2 increased the risk of hepatocellular carcinoma (HCC) in patients with ALD [24]. However, these findings are based on studies that did not include all ethnic groups and the data combined with the low accessibility and high costs is insufficient momentarily to support their inclusion in clinical management algorithms. Metabolic syndrome and obesity increase the risk of hepatic injury in drinkers [18], [19]. Moreover, previously considered protective benefits of moderate alcohol intake actually aggravates metabolic-dysfunction-associated steatotic liver disease (MASLD) [25], [27]. Additionally, diabetes promotes the development of ALD, liver-related admission, mortality and HCC in drinkers [28]. Interestingly, while bariatric surgery can reverse MASLD, it increases the risk of AUD and ALD, particularly in Roux-en-Y surgery, through changes in the profile of alcohol pharmacokinetics, alterations of gut microbiota and neuro-hormonal shifts [29]. Lastly, alcohol in conjunction with other hepatic comorbidities including chronic viral hepatitis B [30], C[31] and hemochromatosis [32], increases the risk of cirrhosis and HCC.
The pathogenesis of ALD has been extensively studied over the years and fundamental disease mechanisms have been established regarding alcohol-induced oxidative stress, abnormal lipogenesis, endoplasmic reticulum dysregulation and the role of the immune system [33]. The latest research focuses on the intestinal microbiota, gut-liver axis, epigenetics and the variety of cell death pathways.
The gut-liver axis alterations, with alcohol-induced intestinal permeability and bacterial translocation, are the established mechanisms which contribute to endotoxemia, with consecutive liver and systemic inflammation [34]. However, the exact mechanisms behind the effects of alcohol on the gut microbiota remain partially understood. Alcohol is rapidly absorbed through diffusion primarily in the upper gastrointestinal tract; hence alcohol does not directly alter the gut microbiota. Recent studies have shown that rather alcohol metabolites, such as acetate, induce the intestinal dysbiosis [35]. The microbiome in ALD suffers specific modifications, resulting in a loss of bacterial diversity. An overgrowth of pathogenic Enterobacteriaceae and a decrease in Lactobacillus and Bifidobacterium was seen in most ALD microbiome studies [36]. In addition, Candida albicans, normally present in the healthy gut, has been seen in high, abnormal levels in ALD patients. A decrease in fungal diversity, with an overgrowth in Candida spp and high titers of anti-Saccharomyces cerevisiae antibodies increased the mortality in AH [37]. Porphyromonas gingivalis, a gram-negative bacterium residing in the oral cavity, linked primarily to periodontitis, has been shown to accelerate the disease progression of ALD. P. gingivalis also aggravates alcohol-related gut dysbiosis and promotes hepatic inflammation through overexpression of toll-like receptor 4 (TLR4), interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) [38]. The resulted dysbiosis decreases the production of short-chain fatty acids (SCFAs), particularly butyrate, which is the primary energy source for colonocytes. Consecutively, the shortage in SCFAs further contributes to the alteration of the gut barrier, maintaining liver inflammation [39]. This could be viewed as a cyclical pathomechanism of alcohol-induced alterations in gut-liver axis. Multiple microbiome studies are ongoing and will probably pave the way for updated standards of care in ALD [40].
Epigenetic alterations are changes in gene expression that do not involve modification in the underlying DNA gene sequence. Three major epigenetic pathways have been recognized in ALD: DNA methylation, histone modifications and microRNA-induced genetic modulation [41]. DNA methylation usually suppresses gene expression. This process is altered in the context of alcohol exposure, leading to aberrant methylations of genes involved in alcohol and lipid metabolism, oxidative stress and inflammation [42]. Another methylation-related change includes the dysregulation of transcription factors, affecting their binding to DNA. A recent systematic review pointed out specific DNA methylation profiles for ALD, MASLD and MetALD suggesting distinct cellular processes and molecular networks [42]. Specific for ALD, methylated genes in pathways including mitogen-activated protein kinase (MAPK) and apelin signaling have been reported to contribute to inflammation, fibrosis and cell death. MAPK pathways regulate many cellular processes including cell growth, differentiation, apoptosis or inflammation, while apelin pathway has been shown to activate the hepatic stellate cells and promote fibrogenesis [43], [44]. The exact mechanisms through which they influence ALD progression is a subject of ongoing research. Alcohol also alters histone structure by inducing reactions of acetylation/deacetylation and methylation, influencing chromatin structure and impacting gene expression, ultimately leading to hepatic inflammation, fibrogenesis and metabolic dysregulation [45]. MicroRNAs are small non-coding RNA molecules that influence post-transcriptional gene expression. Multiple miRNA pathways have been studied, with particular interest in AH, as to this date, it lacks a curative pharmacological treatment [46]. In AH, there is an impairment of liver regeneration attributed to the differentiation of hepatocytes into cholangiocytes, with increased ductular reaction [47]. A bidirectional correlation has been observed between miR-182 expression in the biliary cells and hepatic inflammation. Moreover, miR-182 expression induces ductular reaction, which correlates with disease severity and mortality [48]. Circulating miRNAs are of particular importance, as they can be detected in bodily fluids and thus may be used as novel non-invasive biomarkers for diagnosis and severity stratification [49]. Interestingly, multiple anti-inflammatory and anti-fibrotic processes are upregulated by miR-122, miR30-a and miR-219b [50], [51]. Whether these processes are causative of AH or simply a feed-back response to severe inflammation is yet to be revealed. However, stimulating the effects of these protective miRNAs might represent important directions for further therapeutical studies. Understanding epigenetic alterations could lead to groundbreaking progress in ALD.
ALD comprises a wide spectrum alcohol-induced hepatic lesions ranging for liver steatosis and subclinical inflammation to advanced fibrosis and cirrhosis. Chronic alcohol consumption leads to liver steatosis in 90% of patients and up to 35% will develop different degrees of liver inflammation [15]. Most of the time, those are subclinical and are not diagnosed in the absence of routine paraclinical assessments. As alcohol drinking persists, liver fibrosis progresses and patients can suffer an acute episode of AH, or they can be directly diagnosed with decompensated cirrhosis [2].
AH is a life-threatening condition in patients with sustained heavy alcohol abuse, characterized by intense hepatic inflammation, leading to liver failure and potential high mortality [52]. In contrast, acute-on-chronic liver failure (ACLF) is defined as an acute deterioration of an underlying chronic liver condition (with or without cirrhosis) leading to multiple organ failure with high mortality risk [53]. ACLF includes multiple risk factors, including AUD, and despite the fact that not all severe AH develop ACLF and not all ACLF are caused by AH, the definitions of these two entities cannot be completely separated and must be understood as overlapping syndromes. The development of new diagnosis biomarkers or disease models could help in better differentiating between these two entities [54].
Although first described in the mid-20th century, up to this date there are no reliable non-invasive biomarkers for the prompt identification of AH, hence this syndrome is diagnosed clinically [55].
Rapidly progressive jaundice, within three to six months of last alcohol intake is the main clinical finding in AH [56]. The other signs and symptoms of liver failure can vary depending on disease severity. Mild cases can present with hepatomegaly or liver tenderness, while palmar erythema and Dupuytren contracture are specific for AUD. As severity increases, patients might exhibit signs of significant liver dysfunction and hepatic decompensation, including ascites, portal-hypertensive bleeding and encephalopathy [52]. Fever usually indicates an infection, but it can also be related to the inflammatory state of the disease [12]. Patients are frequently sarcopenic, due to both impaired liver function and alcoholism, with proximal muscle wasting and decreased muscle functional test such as hand grip strength [57], [58].
The laboratory profile in AH includes: hyperbilirubinemia (>3mg/dL), elevated liver transaminases up to 400 UI/mL with the characteristic De Ritis ration of AST:ALT=2:1 and neutrophilia. Macrocytosis is also commonly found as a consequence of alcohol-induced folate and vitamin B12 deficiency [59]. Thrombocytopenia can be seen mainly as a result of splenic sequestration, but also due to the toxic effect of alcohol on the bone marrow [60]. Severe forms of AH will associate coagulation disturbances, hypoalbuminemia and renal dysfunction with increased serum creatinine levels [11], [61].
The National Institute for Alcohol Abuse and Alcoholism (NIAAA) criteria for the diagnosis of AH were first proposed in 2016 as follows: (1) definite AH: clinical criteria met and confirmed by liver biopsy; (2) probable AH: clinical criteria met without confounding factors; (3) possible AH: clinical diagnosis with confounding factors, including drug-induced liver injury (DILI), non-specific laboratory modifications or uncertainty regarding alcohol intake. For this subset, liver biopsy is recommended [62].
On the same note, the European Association for the Study of the Liver (EASL) guidelines support the clinical diagnosis of AH in the typical laboratory setting and alcohol abuse history. However, they state that 10–20% of clinically diagnosed AH are actually other diagnoses, enforcing the importance of liver biopsy. The transjugular approach is preferred in order to reduce the risk of bleeding, a procedure that is hard to implement in clinical practice [12]. Liver biopsy is of particular importance, not only for the diagnosis of uncertain AH, but also for disease prognostication, as severity of fibrosis is one of the most important morality predictors [63].
Regarding imaging studies, abdominal ultrasound is the first choice for liver assessment. Ultrasound findings vary from hepatic steatosis to profound liver architectural alteration in advanced fibrosis. Typically, we can find an enlarged liver, with increased echogenicity, attenuation of ultrasounds and smooth borders [64]. Ascites can be located either perihepatic or in the hepatorenal fossa (Morrison’s pouch), while more advanced liver disease will present with overt ascites [65]. Transient elastography techniques do not reliably reflect liver stiffness in AH, since results are influenced by hepatic inflammation [66]. Therefore, non-invasive hepatic fibrosis assessment should be evaluated after the resolution of AH. On the other hand, certain muscle abnormalities including stiffness, myosteatosis and muscle diameter are promising sarcopenia surrogate ultrasound parameters that could readily assess disease severity in clinical practice [58].
Prognosis stratification has been intensely studied over the years. Disease severity varies from mild to severe cases, with high short- and long-term mortality. Different parameters have been included into prognostic scores in order to better stratify these patients.
Maddrey discriminant fraction (MDF), which includes serum bilirubin and prothrombin time (PT), is the first prognostic score that stratifies high risk patients that benefit from steroid treatment [10]. However, it has some downsides; notably the PT which is not standardized between laboratories and the absence of other organ-failure parameters, particularly renal function, as kidney failure is an important predictor of mortality. Subsequently, other parameters of hepatic or extrahepatic disfunction have been included in more complex scores such as Lille, ABIC, Glasgow alcoholic hepatitis score for a more precise evaluation of AH [67], [68]. Lille score has shifted the paradigm of steroid treatment duration, limiting their administration in non-responsive patients. Model of end-stage liver disease (MELD), initially proposed for liver transplant list stratification and subsequently for cirrhosis severity, has been shown to accurately predict mortality in AH as well [69]. A MELD score >20 defines severe AH with a 90-day mortality of 20% [9], [70]. Advantages of using the MELD score in AH severity stratification are its higher accuracy, use of International Normalized Ratio over PT, the inclusion of serum creatinine and its globally accepted use in organ allocation stratification. A histologic score has also been proposed and values between 6 and 9 defined severe AH with a 50% mortality at 90 days. However it did not attain popularity in the clinical setting due to difficult implementation [63].
Malnutrition is common in ALD, with a considerate impact on mortality. Recently, biological and ultrasound sarcopenia surrogate parameters have been proposed for disease prognosis [7], [58]. Ultrasound muscle alterations including decreased muscle thickness and increased muscle stiffness are corelated with severe outcomes [58].
There is a growing interest in discovering novel biomarkers, given the limitations of the classical laboratory-based models. Circulating levels of different miRNAs reflecting disease severity are being studied, however their utility and addressability in clinical practice is yet to be established [49]. Serum keratin fragments are established biomarkers of hepatocellular apoptosis. Particularly, keratin 18 (K18-M30 and K18-M65) and 19 (CYFRA 21-1) have been demonstrated to predict 90-day mortality and response to corticosteroids, independent of the MELD score. Additionally, there is a strong proportional correlation between AH histological severity and K18 and CYFRA 21-1levels [71], [72], [73]. Hence, K18 has been proposed for inclusion into clinical practice as a theragnostic and prognostic biomarker in AH.
Alcohol abstinence is crucial for managing AH, as it may facilitate liver damage reversal. Studies indicate significantly higher five-year survival rates for patients with compensated cirrhosis who abstain from alcohol compared to those who continue drinking. In severe AH, alcohol abstinence is the main determinant factor for long-term prognosis [74]. Medications like acamprosate, naltrexone, nalmefene, and baclofen have been investigated in maintaining abstinence and while they do improve long-term survival, adherence rates are less than 40% [75], [76].
Over 90% of patients with AH experience significant malnutrition, primarily due to chronic alcohol consumption leading to appetite suppression and inadequate nutrient intake. The severity of malnutrition is closely linked to mortality rates in these patients, particularly those presenting with jaundice and hepatic encephalopathy [75], [77]. Despite concerns regarding the potential exacerbation of hepatic encephalopathy, increased protein intake is generally beneficial given the extent of malnutrition [78], [79]. The European Society for Clinical Nutrition and Metabolism (ESPEN) proposes a daily intake of 35–40 kcal/kg of body weight and a daily protein intake of 1.2–1.5 g/kg of body weight. This should be attained preferably through oral intake, frequently necessitating the placement of a naso-gastric tube. Parenteral nutrition can be used in patients who do not tolerate enteral feeding, although for a limited period, as it can increase the risk line sepsis and intestinal bacterial translocation [12], [56]. B-complex vitamin supplementation is recommended for the prevention of Wernicke’s encephalopathy [80] and considering that 85% of patients with AH are deficient in zinc, ongoing clinical trials study its therapeutic effect [81], [82].
Although patients with severe AH are at an increased risk of infections, prophylactic antibiotherapy has not shown an improvement in survivability [83].
The use of glucocorticoids in treating alcoholic hepatitis has been extensively studied since the 1960s, with inconsistent results largely influenced by trial variability in design, patient selection, and treatment parameters. The rationale for the use of glucocorticoids is centered upon blocking the cytotoxic and inflammatory pathways in AH. Glucocorticoids decrease circulating inflammatory cytokines such as TNF-α, ICAM-1 expression, and have demonstrated short term histologic improvement [84]. Prednisolone is the primary treatment for severe AH, without ongoing infections or renal impairment, typically administered at a dosage of 40 mg daily for four weeks, followed by a tapering period [85]. While there is short-term survival benefit, particularly in patients with a Maddrey discriminant function (DF) score ≥ 32, the long-term mortality remains high, often due to complications like gastrointestinal bleeding or infection [85]. Therefore, Lille score at day 4 or 7 is used to assess futility of steroids. In patients receiving steroids, Lille score also has excellent sensitivity to predict 30-day survival, but poor specificity [8].
Pentoxifylline (PTX), a tumor necrosis alpha suppressor, has been shown to have no effect on over-all survival in severe AH compared to steroids. The only benefit seems to be in reducing the risk of hepato-renal syndrome in this category of patients [61].
Antioxidants have also been studied, notably N-acetylcysteine (NAC) showed conflicting results regarding short-term survival and no effect on long-term survival [11]. However, given its high safety profile and the results of a meta-analysis showing short-term survival benefits in combination with prednisolone, ACG guidelines recommends NAC as adjuvant therapy to steroids [56], [86].
Anti-inflammatory therapeutic options have centered around cytokine suppression through interleukin (IL) antagonists. Particularly, IL-1 antagonist (Anakinra) in association with zinc have been studied for the treatment of severe AH. The initial DASH trial, which also included PTX, showed a non-significant reduction in mortality when compared to Prednisone (p=0.3) [87]. Later phase II trials excluded PTX, as it was demonstrated to have no effect on survival, and the results were unsatisfactory. Only Prednisone significantly reduced mortality and the incidence of acute kidney injury [88]. Further studies are needed in order to implement IL antagonists as treatment for severe AH.
Immune dysregulation and liver regeneration impairment are common pathological mechanisms in AH. Granulocyte colony-stimulating factor (G-CSF), a glycoprotein that stimulates bone marrow stem cells and neutrophil production, is thought to improve survival through stem cells mobilization and enhanced liver regeneration. However, up to this date, there is a high discrepancy between European and Asian study yields. While Asian studies report lower rates of infection and mortality, the European studies showed a non-significant tendency of increased mortality. Moreover, G-CSF was compared with either PTX or placebo and not corticosteroids [89]. The latest EASL and AGA guidelines do not recommend G-CSF for the treatment of severe AH [12], [56].
Patients unresponsive to steroid treatment have low survival rates, and liver transplantation (LT) is often the only viable option. A recent prospective multicenter study (QUICK-TRANS) compared the timing of LT and survivability of patients with severe AH and ALD cirrhosis. After 2 years follow-up, survival rates were higher, but similar in both early LT group and standard 6-month abstinence LT group when compared to non-LT group [90]. The rate of relapse was not significantly different between the two group (34% early LT and 25% late LT, p=0.45), although higher alcohol intake was reported in early LT group. Early relapse (<1 year post-LT) and heavy alcohol intake are associated with low post-LT survival [91]. While early transplantation may significantly improve survival rates for select patients, the 6-month abstinence regulation renders this option inaccessible in most countries [92]. Extracorporeal recycling systems have been studied, primarily as a bridge-therapy to LT. The molecular adsorbent recycling system (MARS) improved cholestatic pruritus and renal function without a significant impact on survival [93]. A preliminary report showed that low volume plasma exchange in combination with low dose of Prednisolone (20 mg daily) improved 1 year survival in patients with alcohol-related ACLF compared to standard therapy [94]. However, larger studies are needed to validate these results.
Recent therapeutic advancements target the gut microbiota regulation, primarily through lowering lipopolysaccharide (LPS) production. Elevated blood LPS levels have been correlated with worse outcomes and immune dysregulation in AH, characterized by increased expression of immunoinhibitory factors and impaired antibacterial immunity. Fecal microbiota transplantation (FMT) has shown promising results in a recent trail when compared to prednisolone, regarding 90-day survival rates [95]. However, given the immunocompromised status of patients with severe AH, caution is advised as donor-stool infections have been reported [96]. Strategies that modulate the gut microbiome and immune response may offer promising avenues for therapy [97].
In summary, ALD global health-burden is recognized and ongoing research is being conducted in order to improve the theragnosis and prognosis stratification of this highly prevalent liver disease. There are still many unmet needs regarding severity stratification and efficient therapeutic options in AH. While the disease pathogenesis knowledge has drastically advanced, a non-invasive diagnosis biomarker is yet to be discovered and standard therapy has not improved over the years. For more than 50 years, steroids are still the cornerstone treatment of severe AH, albeit with less than 50% response rate. There are several ongoing studies targeting hepatic inflammation and oxidative stress, epigenetic alterations, gut-liver axis and microbiome modulation that will aid the therapeutic advancement in patients with AH.