Liver Fibrosis and Risk of Incident Dementia in the General Population: Systematic Review With Meta‐Analysis

Abstract

Background and Aims

The relationship between liver fibrosis and the risk of developing dementia remains unclear, with studies yielding inconsistent results. This systematic review and meta‐analysis seek to synthesize the available evidence.

Methods

We systematically searched PubMed, Scopus, Embase, and Web of Science from their respective inception through October 2024 to identify observational studies diagnosing liver fibrosis non‐invasively or via histology. The primary outcome was new‐onset dementia. Risk of bias was evaluated using the Newcastle‐Ottawa Scale (NOS), and pooled estimates of hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated using a random‐effects model.

Results

Eight longitudinal cohorts, including 1,115,759 middle‐aged individuals (31,129 with liver fibrosis at baseline), identified 29,923 new dementia cases over a mean follow‐up of 14 years. Liver fibrosis exhibited a 32% increased risk of developing all‐cause dementia (pooled HR: 1.32, 95% CI: 1.08–1.61; I² = 76.06%). Dementia risk increased with fibrosis severity: HR 1.06 (95% CI: 0.67–1.68) in ≥F2, HR 1.32 (95% CI: 1.06–1.64) in ≥F3, and HR 1.69 (95% CI: 1.01–2.83) in F4. Geographically, the risk appeared higher in Western than Eastern countries. Women had a greater risk, and vascular dementia was more strongly associated with fibrosis than Alzheimer's disease. Sensitivity analyses confirmed the robustness of the findings, and no publication bias was observed.

Conclusion

Liver fibrosis is linked to a 32% increased long‐term dementia risk, independent of common demographic, social, anthropometric, and cardiometabolic factors. Fibrosis severity further increases this risk. Based on our findings, healthcare professionals should recognize the moderately increased risk of developing dementia in individuals with liver fibrosis and perform close surveillance of these patients to enable early detection and timely intervention.

Keywords: chronic liver disease, dementia, liver‐brain axis, MASLD, NAFLD, sex differences

Introduction

Dementia is a progressive decline in cognitive function that impairs independent living and represents one of the most pressing healthcare and public health challenges among ageing populations [1, 2]. Affecting over 55 million people worldwide, dementia is the seventh leading cause of death, significantly contributing to disability, with an estimated global cost of $1.3 trillion in 2019 [3]. Due to population aging, its prevalence is expected to double every 20 years, reaching 78 million in 2030 and 139 million in 2050 [4]. There are two major types of dementia, Alzheimer's disease (AD) and vascular dementia (VD), with women being at higher risk for AD [5] and experiencing greater dementia‐related disability‐adjusted life years (DALYs) and mortality [3]. Given the limited treatment options and the heterogeneity of underlying pathomechanisms, identifying modifiable risk factors for cognitive decline is crucial for effective prevention [6, 7]

Among medical comorbidities contributing to dementia, studies have expanded the scope of the liver‐brain axis to encompass the entire spectrum of chronic liver disease (CLD) [8]. Liver fibrosis, a hallmark of CLD, is characterized by excessive extracellular matrix deposition resulting from repeated inflammation and tissue repair processes, irrespective of the underlying etiology [9]. Advanced fibrosis alters liver architecture and function, leading to portal hypertension, hepatic insufficiency, hepatocellular carcinoma, and, in some cases, the need for liver transplantation [10].

Recent studies have proposed liver fibrosis as a potential risk factor for incident dementia [7, 11], possibly due to increased intracerebral pro‐inflammatory cytokines, insulin resistance, oxidative stress, free fatty acids overflow, impaired mitochondrial homeostasis, and cerebrovascular dysfunction associated with the prothrombotic state in CLD [12, 13]. Achieving sustained virological response (SVR) after antiviral therapy for hepatitis C virus (HCV) has also been associated with a significantly lower risk of developing dementia compared to untreated individuals [14].

However, studies published so far have yielded conflicting findings, with other publications not confirming a nexus between liver fibrosis and dementia [15, 16]. To address these uncertainties, we conducted a systematic review with meta‐analysis focusing on incident dementia, its subtypes, and sex‐specific analysis of data.

Methods

Protocol and Search Strategy

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) framework [17]. Given that the included studies were observational, we also adhered to the Meta‐analysis of Observational Studies in Epidemiology (MOOSE) criteria for conducting and reporting meta‐analyses [18]. A comprehensive search across four major electronic databases: PubMed, Embase, Scopus, and Web of Science, from their inception up to October 22, 2024. The aim was to identify observational studies exploring the risk of new‐onset dementia in adults (≥ 18 years) with and without liver fibrosis. A combination of Medical Subject Headings (MeSH) terms, Emtree terms, and relevant keywords was used, with the complete search strategies outlined in Supporting Information S1: Tables S1 and S2.

Two independent reviewers (M.J. and F.Z.) initially screened the titles and abstracts of all identified studies based on predefined inclusion and exclusion criteria. Full‐text articles were then retrieved and assessed for eligibility by the same two reviewers, with any disagreements resolved through consultation with the third reviewer (A.L.). Additionally, we manually searched the reference lists of selected articles to identify any relevant studies not captured in the electronic search.

The review protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD42024604692. As this study involved secondary analysis of previously published studies, no additional ethical approval was necessary. All original studies included in the review had relevant ethical approvals and, where applicable, obtained informed consent from participants.

Eligibility Criteria

To be included in this meta‐analysis, studies had to meet the following criteria [1]: prospective or retrospective cohort studies investigating the link between liver fibrosis and the risk of developing dementia [2]; provision of hazard ratios (HRs) with 95% confidence intervals (CIs) as measures of association; and [3] assessment of liver fibrosis using validated biomarkers, fibrosis scoring systems (e.g., FIB‐4, NFS), international classification of diseases (ICD)‐coded diagnoses, imaging techniques, or histological evaluation. The analysis encompassed adult participants across all demographic groups, with no exclusions based on race or ethnicity.

Studies were excluded from our meta‐analysis based on any of the following criteria [1]: non‐eligible study designs, including conference abstracts, case reports, reviews, commentaries, or cross‐sectional studies [2]; cohort studies lacking HRs with 95% CIs for the specified outcome [3]; cohort studies that exclusively examined individuals with liver fibrosis or cirrhosis without a comparison group of those without these conditions; or [4] cohort studies restricted to participants with pre‐existing dementia.

Data Extraction and Quality Assessments

Two reviewers (M.J. and F.Z.) independently extracted pertinent data from eligible studies and recorded it in Microsoft Excel. The extracted data encompassed the author names, publication years, study design, country of origin, population characteristics, methods used to diagnose fibrosis, follow‐up length, key outcomes, and variables controlled for in multivariable analyses, such as matching or confounders. In cases where multiple publications were based on the same cohort, we included the study with the most specific cutoff for liver fibrosis in the overall analysis. Other studies from the same cohort were included in subgroup analyses if they employed different methods for specific subgroups.

The Newcastle–Ottawa Scale (NOS) was used independently by the same two reviewers (M.J. and F.Z.) to evaluate the quality of each included study. Discrepancies were resolved through discussion and, when necessary, consultation with a third reviewer (A.L.). The NOS employs a star‐based rating system to evaluate studies across three key areas: individual selection (maximum of four stars), comparability of groups in the study (up to two stars), and outcome assessment (up to three stars). Studies awarded 9 stars were deemed to have a low risk of bias, 7 or 8 a moderate risk, and 6 or fewer a high risk. Reviewer assessments for these domains were documented using the risk of bias tool available in Review Manager software [19]. Additionally, NOS scores were weighed by the relative significance of each study, as illustrated in Figure 2, to account for the potential bias introduced by studies with a higher risk of bias (lower NOS scores) in the overall outcome.

Figure 2 Forest plot depicting the pooled effect size of the association between liver fibrosis and the risk of developing dementia across eight eligible cohort studies. (Figure not reproduced here; see the original article.)

Data Synthesis and Analysis

Statistical analysis was performed using the “metan” and “metabias” packages in Stata 16 (StataCorp LLC, College Station, TX). The DerSimonian‐Laird random‐effects model was employed to estimate the summary effect size, using the natural logarithm of the hazard ratio (HR) as the effect measure to combine the study outcomes. The results are presented as pooled HRs with 95% confidence intervals (CIs). For studies presenting multiple HRs with different levels of covariate adjustment, we selected and pooled the HRs that accounted for the most extensive set of confounders, applying the inverse variance methods. If studies reported multiple hazard ratios for different ranges of liver fibrosis, we excluded results from patients with inconclusive fibrosis diagnoses, as these correspond to highly heterogeneous populations with significant fibrosis variability. Heterogeneity between studies was assessed using the Q statistic and I². The interpretation of I² values was as follows: 0%–25% indicated minimal heterogeneity, 26%–50% moderate heterogeneity, 51%–75% substantial heterogeneity, and 76%–100% considerable heterogeneity [20]. A funnel plot and Egger's regression test were conducted to evaluate publication bias.

As prespecified in our registered protocol, we performed subgroup analyses based on fibrosis severity, classified according to the METAVIR scoring system into three groups: significant fibrosis (stage F2 or higher), advanced fibrosis (stage F3 or higher), and cirrhosis (stage F4) [21]. Significant fibrosis (≥ F2) was defined via liver biopsy or liver stiffness measurement (LSM) > 8 kPa [22]. Advanced fibrosis (≥ F3) was defined as FIB‐4 > 2.67 or NAFLD fibrosis score (NFS) > 0.676 [23]. Cirrhosis (F4) was identified using liver biopsy findings or ICD codes for cirrhosis.

After the data were gathered, we performed additional analyses to identify possible sources of heterogeneity and evaluate the stability of the observed associations. These included subgroup analyses based on study region, follow‐up length, methods of dementia diagnosis, fibrosis assessment techniques, study quality, and the extent of covariate adjustment. Three sensitivity analyses were also performed [1]: including only studies with the highest level of covariate adjustment [2], limiting to studies with low risk of bias (NOS score of 9), and [3] a leave‐one‐out analysis, systematically excluding each study to assess its influence on the overall estimate. In addition, we performed cause‐specific analyses by pooling effect sizes separately for vascular dementia and Alzheimer's disease, and sex‐specific analyses by pooling effect sizes separately for men and women. All statistical analyses were two‐sided, and p values < 0.05 were considered statistically significant.

Results

Characteristics of Eligible Cohorts

We systematically searched the literature and identified 2913 studies after removing duplicates. Of these, 2854 studies were excluded based on title and abstract screening. We reviewed the full text of the remaining 59 potentially relevant articles and ultimately included 9 studies [8, 11, 14, 15, 16, 24, 25, 26, 27] from 8 cohorts included in our analysis. A detailed flowchart of the study screening and selection process is provided in Figure 1.

Figure 1 PRISMA flow diagram outlining the study screening and selection process. (Figure not reproduced here; see the original article.)

The main characteristics of the eight eligible cohorts are summarized in Table 1. Most cohorts recruited participants from general populations or large health examination programs. Geographically, four cohorts were conducted in European countries (Italy, Sweden, the Netherlands, and the United Kingdom), two in the United States, and two in Asia (Taiwan and Israel).

Regarding fibrosis diagnosis methods, one cohort utilized liver biopsy, one used imaging modalities, and three used blood biomarkers/scores (FIB‐4 and NFS). Two cohorts relied on ICD codes for fibrosis diagnosis. Additionally, studies from the UK Biobank employed different methods, with one using FIB‐4 and another using ICD codes. Dementia diagnoses were also based on ICD codes in six studies, while the remaining two used validated neuropsychological assessments and medical records.

The cohorts assessed different fibrosis levels. Two cohorts provided data on significant liver fibrosis (≥ F2), four on advanced liver fibrosis (≥ F3), and four on cirrhosis (F4). Among the included studies, five cohorts accounted for the most comprehensive set of confounders, including age, sex, body composition metrics, smoking habit, alcohol consumption, blood pressure or hypertension, lipid levels or dyslipidemia, and diabetes status or fasting glucose levels. The remaining three studies applied adjustments for a subset of these factors.

Dementia was defined as all‐cause dementia in seven cohorts. The UK Biobank cohorts additionally reported separate outcomes for AD and VD. One study reported outcomes for both all‐cause dementia and AD, while another study reported only AD as the outcome of interest. Lastly, two cohorts provided sex‐specific outcomes for both males and females.

Publication Bias and Quality Assessments

The Egger regression test yielded a p value of 0.79, indicating publication bias was unlikely in our analysis. Similarly, the funnel plot analysis (Supporting Information S1: Figure S1) showed no apparent asymmetry, further supporting the absence of substantial publication bias. Therefore, publication bias is unlikely to have a meaningful impact on the findings of this systematic review.

As summarized in Table 1 and Supporting Information S1: Table S3, five studies received nine stars on the NOS, indicating a low risk of bias, while three studies received eight or fewer stars, signifying an overall moderate to low risk of bias. Risk of bias in our meta‐analysis was further visualized using the Cochrane Collaboration tool, as shown in Supporting Information S1: Figures S3 and S4. The pooled outcome had a mean weighted NOS score of 8.28, indicating a generally moderate to low risk of bias across the included studies.

Liver Fibrosis and Risk of Incident Dementia

Figure 2 illustrates the distribution of included studies evaluating the relationship between liver fibrosis and the risk of incident dementia. A meta‐analysis of eight eligible cohorts, comprising aggregate data on 1,115,759 middle‐aged individuals (56.5% of them were women, average age 55.8 years [SD: 8.4]), with 31,129 participants having liver fibrosis at baseline, identified 29,923 cases of dementia over a mean follow‐up period of 13.97 years (SD: 4.02). The presence of liver fibrosis was associated with a moderately increased risk of incident dementia (pooled HR: 1.32, 95% CI: 1.08–1.61, p = 0.006; I² = 76.06%).

Since fully adjusted HR were consistently extracted and pooled, the pooled HR remained independent of age, sex, alcohol consumption, measures of adiposity, smoking status, diabetes, hypertension, dyslipidemia, and other prevalent cardiometabolic risk factors. This independence was further validated through a sensitivity analysis restricted to studies with the maximum level of covariate adjustment (n = 5), which demonstrated a similar association (pooled HR: 1.29, 95% CI: 1.07–1.56, I² = 45.28%).

The risk of dementia increased incrementally with the liver fibrosis severity, as shown in Figure 3. The pooled HRs were as follows: significant fibrosis (≥ F2): 1.06 (95% CI: 0.67–1.68, I² = 0.00%), advanced fibrosis (≥ F3): 1.32 (95% CI: 1.06–1.64, I² = 58.07%), and cirrhosis (F4): 1.69 (95% CI: 1.01–2.83, I² = 94.08%).

Figure 3 Subgroup analyses were conducted to evaluate the associations between liver fibrosis and the risk of incident dementia, categorized by fibrosis severity, fibrosis diagnosis method, dementia diagnosis method, mean follow‐up duration, level of covariate adjustment, risk of bias (evaluated using the Newcastle–Ottawa Scale), and study location. FIB‐4, Fibrosis‐4 Index; ICD, International Classification of Diseases; NFS, NAFLD fibrosis score; USA, United States of America. (Figure not reproduced here; see the original article.)

Subgroup Analysis

Subgroup analyses were conducted to explore potential sources of heterogeneity among the included studies (Figure 3). The link between liver fibrosis and incident dementia remained stable across stratifications based on fibrosis severity, diagnostic methods for fibrosis and dementia, extent of covariate adjustment, risk of bias (NOS), and follow‐up duration. However, regional subgroup analysis revealed significant heterogeneity at the continental level (p for subgroup = 0.01). The pooled HRs were 1.12 (95% CI: 0.96–1.29, I² = 60.91%) in Asia, 1.34 (95% CI: 1.07–1.67, I² = 8.39%) in Europe, and 2.04 (95% CI: 1.62–2.56, I² = 0.00%) in the USA. This subgroup analysis substantially explained the overall heterogeneity across the included cohorts; however, significant heterogeneity persisted in the Asian subgroup. This may be attributed to the inclusion of one study in Asia that was not primarily designed to assess the relationship between liver fibrosis and incident dementia, instead examining liver fibrosis univariately as a covariate and focusing on Alzheimer's disease rather than all‐cause dementia as the outcome.

Sensitivity Analysis and Publication Bias

To further validate our findings, we conducted several sensitivity analyses. First, we restricted the analysis to studies with the maximum level of covariate adjustment (n = 5 studies), yielding a pooled HR of 1.29 (95% CI: 1.07–1.56, I² = 45.28%). Second, we included only studies with the lowest risk of bias (NOS = 9), which further supported our results (n = 4 studies, pooled HR: 1.33, 95% CI: 1.08–1.63, I² = 55.31%). Additionally, a leave‐one‐out analysis demonstrated that excluding any single study had minimal impact on the overall pooled hazard ratio and its corresponding 95% CI (Supporting Information S1: Figure S2), further confirming the robustness and reliability of our findings.

Sex‐ and Cause‐Specific Association of Liver Fibrosis and Dementia

The risk of incident dementia appears higher in women with liver fibrosis (n = 2 studies, pooled HR: 1.94, 95% CI: 0.75–5.03, I² = 0.00%) compared to men with liver fibrosis (n = 2 studies, pooled HR: 1.18, 95% CI: 1.07–1.30, I² = 0.00%). However, given the small number of studies eligible for meta‐analysis, these results should be considered within the context of a systematic review rather than a meta‐analysis. Notably, only two cohort studies (Weinstein et al. [11] from Israel and Yuan et al. [27] from the United Kingdom) investigated the relationship between liver fibrosis and incident dementia with sex‐specific manner. While both studies reported a significant association in their individual analyses (Table 1), the pooled HR in women was nonsignificant due to the lack of overlapping confidence intervals and substantial variability between the studies, leading to a wide confidence interval.

Furthermore, liver fibrosis appears to increase the risk of vascular dementia (VD) more than Alzheimer's disease (AD), although cause‐specific analysis can only be discussed systematically. Among the available studies, only those from the UK Biobank analyzed vascular dementia specifically. Parikh et al. [8] reported a significant association with advanced fibrosis (≥ F3, HR: 1.64, 95% CI: 1.03–2.61), while Yuan et al. [27] reported an even higher association with cirrhosis (F4, HR: 3.30, 95% CI: 1.55–3.13). For AD, studies from the UK Biobank, Taiwan, and Sweden cohorts also assessed this outcome. Parikh et al. [8] and Shang et al. [15] examined the risk of AD in patients with significant or advanced fibrosis (≥ F2) and found no significant association (≥ F2, n = 2 studies, pooled HR: 1.25, 95% CI: 0.88–1.78, I² = 0.00%). Shang et al. [15] also examined AD in patients with cirrhosis (F4), and together with Wen et al. [25] and Yuan et al. [27], the pooled HR for AD remained nonsignificant (F4, n = 3 studies, pooled HR: 1.44, 95% CI: 0.82–2.53, I² = 86.49%).

Discussion

Main Findings

In this meta‐analysis of 8 diverse cohorts, including 1,115,759 middle‐aged individuals (56% were women with an average age of 55 years), we identified 31,129 participants with liver fibrosis at baseline and 29,923 new cases of dementia over an average follow‐up of 14 years. Our review led to four novel findings. Firstly, liver fibrosis is associated with a modest though significantly increased risk of developing all‐cause dementia over a long‐term follow‐up (pooled HR: 1.32, 95% CI: 1.08–1.61, I² = 76.06%). It is important to pinpoint that the observed association between liver fibrosis and dementia risk is independent of confounding factors, including age, sex, race, ethnicity, education level, measurement of adiposity, smoking habit, alcohol consumption, type 2 diabetes, hypertension, dyslipidemia, and other cardiometabolic risk factors. Sensitivity analyses confirmed the robustness of the findings, with the association persisting after restricting the analysis to studies with low risk of bias (NOS 9) and studies with maximal covariate adjustment. Secondly, the risk of dementia increases incrementally with the severity of liver fibrosis, indicating a dose–response relationship (Figure 3). The pooled HR for significant fibrosis (≥ F2), advanced fibrosis (≥ F3), and cirrhosis (F4) were 1.06 (95% CI: 0.67–1.68), 1.32 (95% CI: 1.06–1.64), and 1.69 (95% CI: 1.01–2.83), respectively. This finding is consistent with previous research showing that the burden of extrahepatic complications, such as cardiovascular disease, tends to rise in parallel with the increasing severity of liver fibrosis [28]. The observed gradient further supports the biological plausibility of a causal relationship, reinforcing the need for early identification and management of liver fibrosis to potentially mitigate downstream neurological outcomes.

Additionally, subgroup analyses revealed notable regional variations at the continental level, with the strongest associations observed in studies conducted in the USA, followed by Europe and Asia (Figure 3). These differences highlight the potential influence of population characteristics, healthcare practices, or environmental factors on the association of liver fibrosis and dementia. Finally, sex‐ and cause‐specific analyses showed that women with fibrosing CLD were at higher risk of incident dementia than men. This finding is supported by accumulating evidence showing that women with CLD, particularly postmenopausal women, are at greater risk of fibrosis progression [29], which may lead to a higher burden of extrahepatic complications, including cerebrovascular events and stroke [9, 30]. These sex‐specific findings highlight the need for precision medicine approaches in liver fibrosis and dementia research and clinical care. Moreover, our data suggest that liver fibrosis is more strongly associated with the risk of VD than AD, likely due to shared metabolic and vascular mechanisms. This is consistent with a recent Mendelian randomization study that found a potential causal relationship between metabolic dysfunction‐associated steatotic liver disease (MASLD)‐related fibrosis and VD, but not AD [31].

Potential Pathomechanisms Involved

Emerging evidence suggests that liver fibrosis may contribute to cognitive decline through multiple interconnected mechanisms, beyond the well‐established hepatic encephalopathy seen in advanced liver disease [26, 32]. Fibrosing CLD is increasingly recognized as a state of persistent low‐grade inflammation, marked by elevated cytokines and immune mediators that impair vascular homeostasis [33]. This subclinical inflammation promotes endothelial dysfunction, small vessel disease, and disruption of the blood–brain barrier, all of which are central to the development of vascular dementia [34]. Additionally, immune activation and oxidative stress further compromise cerebrovascular health [35]. Liver fibrosis is also associated with subclinical atherosclerosis [36] and increased stroke severity [37], which may further contribute to cognitive impairment via impaired cerebral perfusion.

Gut dysbiosis and increased intestinal permeability, common in MASLD and CLD owing to other etiologies, may exacerbate neuroinflammation through the translocation of bacterial endotoxins into the systemic circulation [38, 39, 40, 41]. These gut‐liver‐brain interactions are increasingly recognized as strong contributors to either vascular or neurodegenerative processes [42]. Moreover, CLD‐associated impaired hepatic clearance may lead to intracerebral deposition and accumulation of circulating amyloid‐β—a hallmark of Alzheimer's disease [43, 44]. Supporting this notion, neuroimaging studies have shown reduced volumes of cortical gray matter in patients with MASLD and liver fibrosis [45].

These liver‐related changes in the size of the cerebral cortex may also serve as a sex‐specific mediator of dementia risk in MASLD. Women generally have lower brain volumes than men, a difference that increases with age [46]. Additionally, the postmenopausal vulnerability to liver fibrosis progression [29] may further accelerate brain atrophy and the risk of dementia in women [47]. Moreover, women with liver fibrosis face a higher stroke risk [48], and post‐stroke brain atrophy can further amplify the susceptibility to develop dementia [49]. This sex‐specific liver–brain axis is supported by our pooled analysis, which showed a stronger association between liver fibrosis and dementia in women. However, the potential role of brain volume as a sex‐specific mediator warrants further investigation to quantify its contribution and clarify its significance within the liver–brain axis. Additionally, severe reduction and failure of hepatic detoxification capacity, often associated with portal hypertension, via raised concentrations of circulating neurotoxins, among which ammonemia is the best characterized, may further contribute to cognitive dysfunction even in the absence of overt hepatic encephalopathy [13].

Metabolic dysregulation—featuring insulin resistance, type 2 diabetes, arterial hypertension, and dyslipidemia—is highly prevalent in MASLD and contributes to both liver fibrosis and cognitive decline [50, 51]. These comorbidities may amplify systemic inflammation and vascular injury, thereby accelerating neurodegenerative changes [50]. In this regard, metabolic syndrome has been independently associated with brain structural abnormalities, such as hippocampal atrophy and cortical thinning, even in the absence of overt liver disease [52]. Therefore, metabolic dysfunction probably plays the dual role of shared risk factor and mediator in the liver–brain axis.

Alcohol (ab)use may further modulate the liver‐brain cross‐talks. As a known hepatotoxin, alcohol promotes steatosis, inflammation, and fibrosis [53]. However, alcohol is also directly neurotoxic, and chronic ethanol consumption has been linked to cerebral atrophy, white matter injury, and global cognitive decline [54]. Of concern, alcohol may synergize with metabolic comorbidities in exacerbating both hepatic and neurologic injury [55]. Therefore, given that alcohol consumption represents both a confounder and an effect modifier in this context, future studies should carefully assess pattern, quantity, type, and duration of drinking when evaluating the independent role of liver fibrosis in dementia risk.

Strengths and Limitations

Our meta‐analysis has notable strengths. First, to the best of our knowledge, this is the first systematic review with meta‐analysis documenting the association of liver fibrosis with incident dementia. Second, the substantial sample size (~ 1.1 million) and the large number of incident dementia cases (~ 30,000) that occurred over the mean follow‐up of 14 years provide high statistical power. Additionally, the overall quality of the included studies was acceptable, indicating a medium‐to‐low risk of bias, as reflected by a weighted mean NOS score of 8.3. Moreover, Sensitivity analyses restricted to studies with maximal covariate adjustment and low risk of bias (NOS = 9) further validate the findings. Iterative exclusion of each study from the meta‐analysis also showed that the findings were not disproportionately dependent on any individual study. Finally, no evidence of publication bias was detected based on Egger's regression test and visual inspection of the funnel plot.

Nonetheless, several limitations should be noted. First, the observational nature of the included studies precludes any strong conclusions regarding causality. Second, the primary analysis showed significant heterogeneity, which dictates cautious interpretation. This heterogeneity likely reflects variations among various geographic areas, methods used for capturing liver fibrosis and dementia, gauging fibrosis severity, duration of follow‐up, study quality, and the degree of covariate adjustment. Additionally, although most included studies adjusted their findings for relevant confounding factors, residual confounding from unmeasured factors and reverse association cannot be ruled out with certainty. Therefore, Mendelian randomization studies are warranted to prove cause‐and‐effect relationships. Finally, the power of sex‐ and cause‐specific analyses was limited by the number of available studies and should be interpreted with caution.

Conclusion

Given disease complex causes and limited treatment options, the prevention of dementia requires a clear understanding of its modifiable risk factors. Our meta‐analysis supports the role of liver fibrosis as a true barometer of systemic health [10] and reveals a modest but statistically significant association with incident dementia over a long‐term follow‐up. These findings underscore the importance of additional well‐designed and well‐conducted studies to clarify the subtype‐specific and sex‐specific risks of dementia, particularly Alzheimer's disease and vascular dementia. Based on our findings, healthcare professionals should recognize the moderately increased risk of developing dementia in individuals with liver fibrosis and perform close surveillance of these patients to enable early detection and timely intervention.

Author Contributions

Mohamad Jamalinia: conceptualization, methodology, data curation, formal analysis, writing – original draft, writing – review and editing. Fatemeh Zare: writing – review and editing, data curation. Amedeo Lonardo: supervision, writing – original draft, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Transparency Statement

The lead author, Mohamad Jamalinia, affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

Acknowledgments

All authors have read and approved the final version of the manuscript. Mohamad Jamalinia had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis. All authors reviewed and approved the final version of the manuscript. The authors received no specific funding for this study.

Jamalinia M., Zare F., and Lonardo A., “Liver Fibrosis and Risk of Incident Dementia in the General Population: Systematic Review With Meta‐Analysis,” Health Science Reports 8 (2025): 1‐12. 10.1002/hsr2.71530.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data used in this article are available in the full text or the supporting material. For further inquiries, readers may contact the corresponding author.

References

Associated Data

Supplementary Materials

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data used in this article are available in the full text or the supporting material. For further inquiries, readers may contact the corresponding author.

References

1. Sheehan B., “Assessment Scales in Dementia,” Therapeutic Advances in Neurological Disorders 5, no. 6 (2012): 349–358.
2. Mokdad A. H., Ballestros K., Echko M., et al., “The State of US Health, 1990–2016: Burden of Diseases, Injuries, and Risk Factors Among US States,” Journal of the American Medical Association 319, no. 14 (2018): 1444–1472.
3. “Dementia Fact Sheet,” World Health Organization, 2024, https://www.who.int/news-room/fact-sheets/detail/dementia#:~:text=Currently%20more%20than%2055%20million,injuries%20that%20affect%20the%20brain.
4. “Dementia Statistics,” International AsD, 2024, https://www.alzint.org/about/dementia-facts-figures/dementia-statistics/.
5. Hogervorst E., Temple S., and O'Donnell E., “Sex Differences in Dementia,” Current Topics in Behavioral Neuroscience 62 (2023): 309–331.
6. Scheltens P., De Strooper B., Kivipelto M., et al., “Alzheimer's Disease,” Lancet 397, no. 10284 (2021): 1577–1590.
7. Gao P. Y., Ou Y. N., Wang H. F., et al., “Associations of Liver Dysfunction With Incident Dementia, Cognition, and Brain Structure: A Prospective Cohort Study of 431 699 Adults,” Journal of Neurochemistry 168, no. 1 (2024): 26–38.
8. Parikh N. S., Kamel H., Zhang C., et al., “Association Between Liver Fibrosis and Incident Dementia in the UK Biobank Study,” European Journal of Neurology 29, no. 9 (2022): 2622–2630.
9. Jamalinia M., Lonardo A., and Weiskirchen R., “Sex and Gender Differences in Liver Fibrosis: Pathomechanisms and Clinical Outcomes,” Fibrosis 2, no. 4 (2024): 10006.
10. Lonardo A., Ballestri S., Baffy G., and Weiskirchen R., “Liver Fibrosis as a Barometer of Systemic Health by Gauging the Risk of Extrahepatic Disease,” Metabolism and Target Organ Damage 4, no. 4 (2024): 41.
11. Weinstein G., Schonmann Y., Yeshua H., and Zelber‐Sagi S., “The Association Between Liver Fibrosis Score and Incident Dementia: A Nationwide Retrospective Cohort Study,” Alzheimer's & Dementia 20, no. 8 (2024): 5385–5397.
12. Lecube A., Hernández C., Genescà J., and Simó R., “Proinflammatory Cytokines, Insulin Resistance, and Insulin Secretion in Chronic Hepatitis C Patients,” Diabetes Care 29, no. 5 (2006): 1096–1101.
13. Cheon S. Y. and Song J., “Novel Insights into Non‐Alcoholic Fatty Liver Disease and Dementia: Insulin Resistance, Hyperammonemia, Gut Dysbiosis, Vascular Impairment, and Inflammation,” Cell & Bioscience 12, no. 1 (2022): 99.
14. Tao M. H., Gordon S. C., Wu T., et al., “Antiviral Treatment and Response Are Associated With Lower Risk of Dementia Among Hepatitis C Virus‐Infected Patients,” American Journal of Geriatric Psychiatry 32, no. 5 (2024): 611–621.
15. Shang Y., Nasr P., Ekstedt M., et al., “Non‐Alcoholic Fatty Liver Disease Does Not Increase Dementia Risk Although Histology Data Might Improve Risk Prediction,” JHEP Reports 3, no. 2 (2021): 100218.
16. Xiao T., van Kleef L. A., Ikram M. K., de Knegt R. J., and Ikram M. A., “Association of Nonalcoholic Fatty Liver Disease and Fibrosis With Incident Dementia and Cognition: The Rotterdam Study,” Neurology 99, no. 6 (2022): e565–e573.
17. Page M. J., McKenzie J. E., Bossuyt P. M., et al., “The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews,” BMJ 372 (2021): n71.
18. Stroup D. F., “Meta‐Analysis of Observational Studies in Epidemiologya Proposal for Reporting,” Journal of the American Medical Association 283, no. 15 (2000): 2008–2012.
19. Cumpston M., Li T., Page M. J., et al., “Updated Guidance for Trusted Systematic Reviews: A New Edition of the Cochrane Handbook for Systematic Reviews of Interventions,” Cochrane Database of Systematic Reviews 10, no. 10 (2019): Ed000142.
20. Higgins J. P. T. T. J., Chandler J., Cumpston M., Li T., Page M. J., and Welch V. A., ed., Cochrane Handbook for Systematic Reviews of Interventions, 2nd ed. (John Wiley & Sons, 2019).
21. Bedossa P., “Intraobserver and Interobserver Variations in Liver Biopsy Interpretation in Patients With Chronic Hepatitis C,” Hepatology 20, no. 1 (1994): 15–20.
22. Hsu C., Caussy C., Imajo K., et al., “Magnetic Resonance vs Transient Elastography Analysis of Patients With Nonalcoholic Fatty Liver Disease: A Systematic Review and Pooled Analysis of Individual Participants,” Clinical Gastroenterology and Hepatology 17, no. 4 (2019): 630–637.e8.
23. Sterling R. K., Patel K., Duarte‐Rojo A., et al., “AASLD Practice Guideline on Blood‐Based Noninvasive Liver Disease Assessment of Hepatic Fibrosis and Steatosis,” Hepatology 81, no. 1 (2025): 321–357, 10.1097/HEP.0000000000000845.
24. Solfrizzi V., Scafato E., Custodero C., et al., “Liver Fibrosis Score, Physical Frailty, and the Risk of Dementia in Older Adults: The Italian Longitudinal Study on Aging,” Alzheimer's & Dementia: Translational Research & Clinical Interventions 6, no. 1 (2020): e12065.
25. Wen L. Y., Wan L., Lai J. N., et al., “Increased Risk of Alzheimer's Disease Among Patients With Age‐Related Macular Degeneration: A Nationwide Population‐Based Study,” PLoS One 16, no. 5 (2021): e0250440.
26. Lu Y., Pike J. R., Hoogeveen R. C., et al., “Liver Integrity and the Risk of Alzheimer's Disease and Related Dementias,” Alzheimer's & Dementia 20, no. 3 (2024): 1913–1922.
27. Yuan S., Dan L., Zhang Y., et al., “Digestive System Diseases, Genetic Risk, and Incident Dementia: A Prospective Cohort Study,” American Journal of Preventive Medicine 66, no. 3 (2024): 516–525.
28. Mantovani A., Csermely A., Petracca G., et al., “Non‐Alcoholic Fatty Liver Disease and Risk of Fatal and Non‐Fatal Cardiovascular Events: An Updated Systematic Review and Meta‐Analysis,” Lancet Gastroenterology & Hepatology 6, no. 11 (2021): 903–913.
29. Balakrishnan M., Patel P., Dunn‐Valadez S., et al., “Women Have a Lower Risk of Nonalcoholic Fatty Liver Disease but a Higher Risk of Progression vs Men: A Systematic Review and Meta‐Analysis,” Clinical Gastroenterology and Hepatology 19, no. 1 (2021): 61–71.e15.
30. Jamalinia M., Zare F., Mantovani A., Targher G., and Lonardo A., “Metabolic Dysfunction‐Associated Steatotic Liver Disease and Sex‐Specific Risk of Fatal and Non‐Fatal Cardiovascular Events: A Meta‐Analysis,” Diabetes, Obesity and Metabolism 27, no. 9 (2025): 5171–5181.
31. Li Y. S., Xia Y. G., Liu Y. L., et al., “Metabolic‐Dysfunction Associated Steatotic Liver Disease‐Related Diseases, Cognition and Dementia: A Two‐Sample Mendelian Randomization Study,” PLoS One 19, no. 2 (2024): e0297883.
32. Lu H., Zhang H., Wu Z., and Li L., “Microbiota‐Gut‐Liver‐Brain Axis and Hepatic Encephalopathy,” Microbiome Research Reports 3, no. 2 (2024): 17.
33. Hammerich L. and Tacke F., “Hepatic Inflammatory Responses in Liver Fibrosis,” Nature Reviews Gastroenterology & Hepatology 20, no. 10 (2023): 633–646.
34. Tian Z., Ji X., and Liu J., “Neuroinflammation in Vascular Cognitive Impairment and Dementia: Current Evidence, Advances, and Prospects,” International Journal of Molecular Sciences 23, no. 11 (2022): 6224.
35. Kumar V., Bishayee K., Park S., Lee U., and Kim J., “Oxidative Stress in Cerebrovascular Disease and Associated Diseases,” Frontiers in Endocrinology 14 (2023): 1124419.
36. Jamalinia M., Zare F., and Lankarani K. B., “Systematic Review and Meta‐Analysis: Association Between Liver Fibrosis and Subclinical Atherosclerosis in Nonalcoholic Fatty Liver Disease,” Alimentary Pharmacology & Therapeutics 58, no. 4 (2023): 384–394.
37. Toh E., Joseph Ravi P., Ming C., et al., “Risk of Liver Fibrosis Is Associated With More Severe Strokes, Increased Complications With Thrombolysis, and Mortality,” Journal of Clinical Medicine 12, no. 1 (2023): 356.
38. Boursier J., Mueller O., Barret M., et al., “The Severity of Nonalcoholic Fatty Liver Disease Is Associated With Gut Dysbiosis and Shift in the Metabolic Function of the Gut Microbiota,” Hepatology 63, no. 3 (2016): 764–775.
39. De Minicis S., Rychlicki C., Agostinelli L., et al., “Dysbiosis Contributes to Fibrogenesis in the Course of Chronic Liver Injury in Mice,” Hepatology 59, no. 5 (2014): 1738–1749.
40. Nicoletti A., Ponziani F. R., Biolato M., et al., “Intestinal Permeability in the Pathogenesis of Liver Damage: From Non‐Alcoholic Fatty Liver Disease to Liver Transplantation,” World Journal of Gastroenterology 25, no. 33 (2019): 4814–4834.
41. Ferreiro A. L., Choi J., Ryou J., et al., “Gut Microbiome Composition may be an Indicator of Preclinical Alzheimer's Disease,” Science Translational Medicine 15, no. 700 (2023): eabo2984.
42. Yan M., Man S., Sun B., et al., “Gut Liver Brain Axis in Diseases: The Implications for Therapeutic Interventions,” Signal Transduction and Targeted Therapy 8, no. 1 (2023): 443.
43. Wang Y. R., Wang Q. H., Zhang T., et al., “Associations Between Hepatic Functions and Plasma Amyloid‐Beta Levels‐Implications for the Capacity of Liver in Peripheral Amyloid‐Beta Clearance,” Molecular Neurobiology 54, no. 3 (2017): 2338–2344.
44. Weinstein G., O'Donnell A., Davis‐Plourde K., et al., “Non‐Alcoholic Fatty Liver Disease, Liver Fibrosis, and Regional Amyloid‐β and Tau Pathology in Middle‐Aged Adults: The Framingham Study,” Journal of Alzheimer's Disease 86, no. 3 (2022): 1371–1383.
45. Weinstein G., O'Donnell A., Frenzel S., et al., “Nonalcoholic Fatty Liver Disease, Liver Fibrosis, and Structural Brain Imaging: The Cross‐Cohort Collaboration,” European Journal of Neurology 31, no. 1 (2024): e16048.
46. Guo J. Y., Isohanni M., Miettunen J., et al., “Brain Structural Changes in Women and Men During Midlife,” Neuroscience Letters 615 (2016): 107–112.
47. Briceno Silva G., Arvelaez Pascucci J., Karim H., et al., “Influence of the Onset of Menopause on the Risk of Developing Alzheimer's Disease,” Cureus 16, no. 9 (2024): e69124.
48. Parikh N. S., Koh I., VanWagner L. B., Elkind M. S. V., Zakai N. A., and Cushman M., “Liver Fibrosis Is Associated With Ischemic Stroke Risk in Women but Not Men: The REGARDS Study,” Journal of Stroke and Cerebrovascular Diseases 30, no. 7 (2021): 105788.
49. Rost N. S., Brodtmann A., Pase M. P., et al., “Post‐Stroke Cognitive Impairment and Dementia,” Circulation Research 130, no. 8 (2022): 1252–1271.
50. Karway G. K., Krzyzanowski B., Killion J. A., et al., “Regional Variability of the Impact of Cardiometabolic Diseases on Incident Dementia in United States Medicare Beneficiaries,” Alzheimer's & Dementia 21, no. 5 (2025): e70199.
51. Huang D. Q., Wilson L. A., Behling C., et al., “Fibrosis Progression Rate in Biopsy‐Proven Nonalcoholic Fatty Liver Disease Among People With Diabetes Versus People Without Diabetes: A Multicenter Study,” Gastroenterology 165, no. 2 (2023): 463–472.e5.
52. Kordestani‐Moghadam P., Assari S., Nouriyengejeh S., Mohammadipour F., and Pourabbasi A., “Cognitive Impairments and Associated Structural Brain Changes in Metabolic Syndrome and Implications of Neurocognitive Intervention,” Journal of Obesity & Metabolic Syndrome 29, no. 3 (2020): 174–179.
53. Lackner C. and Tiniakos D., “Fibrosis and Alcohol‐Related Liver Disease,” Journal of Hepatology 70, no. 2 (2019): 294–304.
54. Sachdeva A., Chandra M., Choudhary M., Dayal P., and Anand K. S., “Alcohol‐Related Dementia and Neurocognitive Impairment: A Review Study,” International Journal of High Risk Behaviors and Addiction 5, no. 3 (2016): e27976.
55. Shin W. Y., Kang E. S., Oh Y. H., et al., “Metabolic Dysfunction‐Associated Steatotic Liver Disease, Metabolic Alcohol‐Related Liver Disease, and Incident Dementia: A Nationwide Cohort Study: Masld, Metald, and Dementia Risk,” BMC Gastroenterology 25, no. 1 (2025): 308.