Alcohol is the most widely abused psychoactive substance globally, with high-risk drinking reported in up to 30% of Western populations, contributing substantially to morbidity and mortality [1]. Alcohol use disorder (AUD), as defined by the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-V), encompasses a spectrum of maladaptive drinking behaviors that result in clinically significant physical, psychological, or social dysfunction [2]. Excessive alcohol intake is a well-established cause of liver cirrhosis, which remains a leading indication for liver transplantation (LT), particularly in cases of severe alcoholic hepatitis and end-stage liver disease (ESLD) [3, 4].

Cirrhosis-related neurocognitive decline is commonly attributed to hepatic encephalopathy (HE), a complication of advanced liver dysfunction associated with hyperammonemia and disruption of the liver-brain axis [5, 6]. While blood ammonia levels are a recognized biomarker of HE, they primarily have a high negative predictive value and do not reliably correlate with cognitive impairment (CI) [5, 6]. Although HE is considered reversible after LT, studies indicate that some patients with overt HE pre-transplantation exhibit persistent or even worsening cognitive deficits post-LT, suggesting that CI may result from more complex and multifactorial mechanisms [7, 8].

In patients with AUD, cognitive deficits extend beyond HE. Chronic alcohol consumption leads to widespread and potentially irreversible neurotoxic effects, including oxidative stress, mitochondrial dysfunction, and neuroinflammation. Acetaldehyde, the primary metabolite of ethanol, induces cellular damage by forming protein adducts and generating reactive oxygen species, which impair mitochondrial DNA and neuronal integrity [9]. Concurrently, elevated levels of pro-inflammatory cytokines—such as TNF-α, IL-6, IL-1β, and MCP1—are known to disrupt neuroplasticity and contribute to structural and functional changes in key brain regions, including the hippocampus and prefrontal cortex [10, 11].

As a result, individuals with AUD are particularly susceptible to a broad spectrum of cognitive deficits, including impairments in executive function, attention, abstract reasoning, psychomotor speed, visuospatial skills, language, and both verbal and visual memory [12–14]. These deficits are often linked to alcohol-induced reductions in hippocampal white matter volume and damage to the prefrontal cortex—areas crucial for memory, decision-making, and behavioral regulation. Structural brain abnormalities, minimal and overt HE, chronic alcohol use, diabetes, and gut microbiome dysbiosis may further contribute to pre-transplant CI and could influence its persistence after LT [15, 16].

Despite these known associations, cognitive impairment in patients with alcohol-related liver cirrhosis awaiting LT remains under-investigated. Existing studies on CI in LT recipients with cirrhosis of mixed etiologies report a post-transplant prevalence of CI ranging from 0% to 36% [17], but they are limited by methodological inconsistencies, variable definitions of HE and CI, and heterogeneous patient populations. Additionally, most prior research originates from neurology or psychiatry domains, often employing diverse and non-standardized diagnostic tools, which complicates comparisons and limits clinical applicability [18–20].

To address these limitations, the present pilot study aimed to perform a structured, longitudinal evaluation of cognitive function in a homogeneous cohort of patients with AUD-related ESLD. By assessing cognition both at the time of listing and after liver transplantation in a single-center setting, this study seeks to provide new insight into the nature, evolution, and clinical relevance of cognitive impairment in this vulnerable patient population.

Given the limited availability of long-term, prospective data on the trajectory of cognitive impairment (CI) in patients before and after liver transplantation (LT) - particularly in clinically homogeneous populations - two primary research objectives were formulated:

1. To conduct a quantitative and qualitative assessment of cognitive impairment prior to liver transplantation in patients with alcohol-related liver disease (ALD) qualified for transplantation within a single transplant center.

Between November 2022 and January 2024, a total of 101 consecutive adult patients with alcohol use disorder (AUD) were enrolled in the study (78% male; mean age: 53 ± 11 years; mean MELD score: 16 ± 7), all of whom were evaluated as potential candidates for liver transplantation (LT). Among them, 17% had hepatocellular carcinoma (HCC), while the remaining 83% were assessed for LT due to chronic liver failure.

Exclusion criteria included regular use of sedative medications, severe overt hepatic encephalopathy, and psychiatric or neurodegenerative disorders precluding cognitive assessment. No patients with acute alcoholic hepatitis were included in the study.

Plasma ammonia concentration was measured using an enzymatic method with glutamate dehydrogenase (GLDH) on the Dimension EXL analyzer (Siemens Healthineers, Forchheim, Germany), with a reference range of 19–55 μg/dL.

Of the 101 patients who underwent baseline cognitive assessment, 61 successfully received liver transplants. Three patients died before LT and an additional six died postoperatively. Ultimately, 55 individuals (including 14 women, 25%) underwent repeated cognitive evaluation (Figure 1). The mean age in this subgroup was 56 years.

All patients were routinely assessed during hospitalization by a psychiatrist dedicated to the transplant program, as part of the standard pre-transplant evaluation protocol. This included a comprehensive psychiatric consultation and administration of the Addenbrooke’s Cognitive Examination III (ACE-III). In patients who completed both assessments, psychiatric evaluation was conducted at listing for liver transplantation and again at an average of 7.1 months post-transplantation (SD = 1.45; range: 6–11 months; median: 7 months), ensuring methodological consistency and enabling reliable longitudinal analysis.

This study provides a comprehensive evaluation of cognitive function at the time of liver transplant listing and in the early post-transplant period in a homogeneous cohort of patients with alcohol use disorder (AUD)-related liver cirrhosis. Cognitive impairment (CI) is increasingly recognized in patients with end-stage liver disease (ESLD), but available data remain heterogeneous, encompassing mixed etiologies, various diagnostic tools, and differing disease severity. Thus, a critical gap persists in understanding the true prevalence and profile of CI in well-defined liver transplant candidate populations.

To the best of our knowledge, this is the first study to examine cognitive performance both before and after liver transplantation (LT) in a cohort exclusively composed of patients with AUD-related ESLD. It delivers new insights into a crucial yet underexplored dimension of peri-transplant care. A recent review by Siddiqui et al. [17] included 24 studies with a median of 30 patients per study and follow-up ranging from 1 month to 1.8 years post-LT. The prevalence of CI varied from 4% to 36% within the first 8 months post-transplant and from 0% to 16% thereafter. Due to methodological variability, CI was grouped into six cognitive domains: attention, executive functions, working memory, long-term memory, language, and visuospatial abilities. However, no prior study has provided precise pre-LT CI data specifically in AUD candidates.

Our single-center analysis demonstrated a strikingly high prevalence of significant CI suggestive of dementia, with only 14% of patients scoring within the normal range on the ACE-III. The lowest performance was noted in verbal fluency and visuospatial domains, echoing findings by Lee et al. [25] and Sorrell et al. [26], who reported deficits in memory, visuospatial construction, attention, and immediate memory in patients with alcohol-related cirrhosis. Prior work by our group also showed significantly higher ACE-III scores and lower CI prevalence in patients with end-stage kidney disease (ESKD) compared to those with ESLD, with memory and visuospatial function being particularly impaired in the ESLD cohort [27].

Alcohol-induced brain damage is influenced by the quantity, age of onset, and duration of drinking, along with age, education, genetics, and prenatal exposure [14, 20, 28]. Stavro et al. [29] and Crowe et al. [28] described diffuse cognitive deficits in AUD, consistent with our cohort’s mean ACE-III score of 78, with dysfunction across all domains. Verbal fluency and visuospatial impairments were particularly prominent, aligning with findings from Crowe et al. [28] and others [30–34]. Post-transplant studies by Campagna et al. [15] and Siddiqui et al. [17] also reported persistent deficits in attention, executive function, memory, and visuospatial processing.

Although the pathophysiological mechanisms linking ESLD and CI remain complex, factors such as neuroinflammation, hippocampal atrophy, and oxidative damage to the prefrontal cortex have been implicated [13, 35, 36]. MRI studies have shown reduced hippocampal volume in AUD patients compared to healthy controls [37]. Our data revealed a significant negative correlation between age and verbal fluency, with a trend toward lower total ACE-III scores in older patients. This supports findings by Campagna et al. [15], who suggested a critical vulnerability window at ages 50–60, possibly reflecting cumulative alcohol-related neurotoxicity.

Educational attainment emerged as a protective factor against CI, with ACE-III scores positively correlated with years of education, in line with Schneeweis et al. [38]. Conversely, a higher Child-Pugh score was associated with poorer cognitive outcomes and an increased likelihood of dementia-range ACE-III results. The MELD score did not correlate with total ACE-III performance but did show associations with attention and language subdomains. This divergence may relate to MELD’s omission of nutritional parameters like serum albumin. Chronic malnutrition and sarcopenia, common in advanced liver disease, are known contributors to cognitive dysfunction [39–41].

Our findings align with reports showing greater post-LT CI prevalence in patients with pre-transplant MELD scores of 22–26 (21%–36%) compared to those with scores of 11–19 (8%–13%) [16, 17]. These trends highlight the multifactorial burden of CI in advanced liver disease.

In our cohort of 55 patients who completed both pre- and post-transplant cognitive evaluations, we observed a significant overall improvement in ACE-III scores, with the most notable gains in memory, verbal fluency, and visuospatial domains. Language function showed only modest improvement and remained one of the most frequently impaired areas. These findings align with prior reports suggesting that LT may reverse cognitive deficits, particularly those associated with hepatic encephalopathy and systemic inflammation [15, 17]. The Wilcoxon signed-rank test confirmed the statistical significance of this improvement (p < 0.0001), further supporting the role of LT in promoting cognitive recovery in the majority of patients (Figure 2).

However, as illustrated in Figure 3, recovery was not universal - approximately 16% of patients experienced cognitive decline, most often in language and memory domains, while 8% remained unchanged. This variation highlights the need for long-term neurocognitive surveillance and for identifying risk factors associated with suboptimal outcomes.

Preliminary findings (Table 7) suggest that younger age and higher educational attainment may offer some protection against cognitive decline following liver transplantation. Given the high prevalence and clinical consequences of cognitive impairment in patients with alcohol-related end-stage liver disease, routine cognitive assessment during pre-transplant evaluation appears warranted. Early detection of individuals at increased risk could enable personalized cognitive support strategies and potentially enhance long-term clinical outcomes. CI has been shown to negatively impact treatment adherence, decision-making capacity, and overall prognosis in transplant recipients [42].

The association between higher educational attainment and cognitive improvement post-transplant is consistent with the cognitive reserve hypothesis, which posits that individuals with greater lifelong cognitive engagement - often reflected by formal education - may be more resilient to the effects of brain injury, including those related to hepatic encephalopathy. Although the current sample size is limited, the statistically significant difference in education between improved and declined patients lends further support to this concept.

Notably, age did not significantly differentiate outcome groups, suggesting that within this cohort, cognitive reserve may have been a more relevant determinant of cognitive recovery than chronological age.

Future studies should aim to incorporate neuroimaging and biomarkers to elucidate the mechanisms underlying CI. Clinical trials of cognitive training and pharmacologic interventions in this population are also needed to guide individualized care strategies.

This study demonstrates a concerning prevalence of severe cognitive impairment, potentially indicative of dementia, in patients with alcohol-related ESLD both before and after liver transplantation. Although most patients showed post-transplant improvement, persistent deficits in memory and language highlight the need for ongoing monitoring. Our findings support routine cognitive screening in this population and underscore the importance of further research into predictive markers and therapeutic interventions aimed at preserving and enhancing cognitive function after LT.