What Is the Current Knowledge?

Alcohol is a driving force of liver fibrosis and cirrhosis in general people.

Due to the continuous destruction of the immune system, PLWH are more prone to develop liver fibrosis and other liver diseases.

What Is New Here?

High alcohol consumption is associated with increased liver fibrosis risk in PLWH, especially HIV/HCV co-infected patients.

Inconsistent definitions of alcohol exposure can lead to different conclusions.

Liver fibrosis is one of the most common liver diseases (1). Liver fibrosis (including cirrhosis) is a leading cause of death, affecting more than 600,000 people each year (2). According to 2015 data, the prevalence of liver fibrosis in America is about 0.27%, which is equivalent to 633,323 adults (3). In China, it is estimated that more than one-fifth of the population is affected by some form of liver disease (4), and more than 7 million (0.5% of the total Chinese population) live with liver cirrhosis, from which stem 460,000 new cases of liver cancer per year in China (5). In chronic liver disease, fibrosis develops as the result of an imbalance between the new deposition of extracellular matrix (ECM) and its resorption, representing the wound healing response of the liver to repeated injury (6). If liver fibrosis cannot be treated and intervened in time, it will progress to liver cirrhosis or even liver cancer, which is a heavy disease burden also in Africa (7). Among people living with HIV (PLWH), the incidence of hepatic fibrosis is increasing (8). The reasons include the use of antiretroviral drugs (9), simultaneous infection with HBV or HCV (10), decreased immune function (11), and drug abuse (12), especially alcohol.

Alcohol is a major risk factor for liver fibrosis in general, and for PLWH in particular (13). Alcohol consumption can complicate HIV infection and lead to comorbidities (14). Alcohol use negatively affects HIV disease progression through multiple mechanisms, including transmission, viral replication (15), immunity system (16), and therapeutic efficacy (17). Among those using PI-based regimens, hazardous alcohol drinkers had significantly higher liver fibrosis (18). HIV can affect the immune system, thereby affecting the function of vital organs including the liver, and then induce fibrosis generation (19). As liver fibrosis is a liver wound healing reaction, which often occurs in alcoholic hepatitis, and is characterized by excessive production and accumulation of fibrotic matrix, mainly type I collagen (20). Previous study (21) reported that whether under acute or chronic alcohol stimulation, the metabolic process of ethanol produces a large amount of reactive oxygen species (ROS), causing mitochondrial damage and fat oxidative degeneration, which is important pathogenesis of the alcoholic liver disease. At the same time, ROS can also inhibit the differentiation ability of cells and the repairability of damaged tissues, increase the synthesis of collagen fibers, and accelerate liver fibrosis formation, leading to liver fibrosis’s progress (22, 23). Researchers believe that cytokines mediate alcohol-induced liver fibrosis, the most important being the pro-fibrotic factor TGF-β ( 24). Also, others verified (25) that drinking leads to intestinal microecological disorders, and the intestinal epithelium’s integrity is destroyed, leading to the influx of immune cells into the liver, forming intestinal endotoxemia and ultimately liver fibrosis. However, there is also literature that indicates alcohol intake does not affect liver disease or liver fibrosis (26, 27).

In the present study, we defined specific liver stiffness cutoffs for alcohol-induced liver fibrosis in PLWH, with particular consideration given to the effects of study design, alcohol determination, and HCV co-infection, and other subgroups where data were available.

Our findings indicate that high-risk alcohol drinking is a risk factor for liver fibrosis for PLWH, especially HIV and HCV co-infected patients. Most importantly, our meta-analysis shows different patterns in subgroups: for different alcohol consumption groups, we consider the daily alcohol consumption is more sensitive for analysis of the effect of alcohol abuse rather than AUDIT-C score; for HCV co-infected or not subgroups, in HIV and HCV co-infected patients, high-risk alcohol consumption has a higher risk of liver fibrosis than HIV mono-infected patients. Although only three included studies used a cohort study approach with a high level of evidence, their results showed that alcohol exposure is a risk factor for liver fibrosis in people infected with HIV, regardless of co-infection with HCV (12, 36, 49).

Recently published meta-analyses have provided information about alcohol abuse as a risk factor for liver cirrhosis in general people (50). Our meta-analysis confirms previous findings in PLWH and provides additional value with five distinct subgroup analyses. In this study, strict inclusion and exclusion criteria were formulated according to the research purpose, and studies with significant differences in the definition of high-risk alcohol drinking were excluded for analysis (41, 43). In addition, according to the full consideration of the risk of liver fibrosis caused by the drinking of HIV-infected persons in different regions, the population characteristics, and the diagnose methods are different in each research inclusion.

According to our data, the alcohol intake determination pathway may influence the risk of alcohol consumption, which could have impacted our findings. One study noted that a stepwise increased risk of advanced liver fibrosis with greater severity of alcohol use (51). They also demonstrate that all alcohol use categories are strongly associated with advanced hepatic fibrosis in HIV/HCV-coinfected patients. Our study found that daily alcohol intake>50 g was a significant risk factor for F3 or F4 stage liver fibrosis. Although the US AUDIT has adapted the WHO AUDIT to a 14 g standard drink (52), AUDIT-C score ≥ 4 (similarly to ≥ 52 g/day) were not risk factors for liver fibrosis incidence in PLWH in our study. Also, we found there were studies (43) using a structured interview to recall patterns of total lifetime alcohol intake, which may be better-predicted liver injury than current heavy alcohol use. But this method is rarely applied because it is not suitable for large sample research and the accuracy deserves further investigation. Serum phosphatidylethanol (PEth) is a biological marker that reflects alcohol use within an approximate 3–4-week period. Although PEth does not rely on self-report by participants, the period of alcohol determination is too short of reflecting the truth status of the alcohol consumption (48). The results of AUDIT-C show their cut-offs produce a high rate of false positives resulting largely from not measuring alcohol consumption correctly (52). However, the daily alcohol consumption can precise distinct high level of alcohol consumption. We recommend referring more to the results of daily alcohol consumption rather than the AUDIT-C as the measure of alcohol intake, for several reasons that can be summarized as follows. Initially, AUDIT-C is suitable for detecting early drinkers rather than long-term drinkers, and liver fibrosis is a chronic process, so alcohol measurement using AUDIT-C may not be a good guide for liver fibrosis. Secondly, AUDIT-C investigates not only the amount and frequency of alcohol consumption, but also the severity of adverse health conditions that may be caused by alcohol (53), so AUDIT-C may have introduced other non-alcoholic factors in the study of the relationship between alcohol consumption and liver fibrosis, making the study of this association inaccurate. In the general population, both AUDIT-C and daily alcohol consumption are applicable; but in sensitive populations such as HIV infection, using daily alcohol consumption is more accurate.

The result, which was divided according to the different study populations, showed that the combined effect size was similar to the total effect in the three subgroups, and the effect was not statistically significant only in the mixed group that did not distinguish HCV infection status. In the subgroup infected with HIV only, there was an increased risk for alcohol abuse compared to no hazard alcohol consumption. In the subgroup co-infected with HCV, the OR was 2.48. In the mixed subgroup, the pooled OR was 1.97. A cross-sectional study (54) by Kim showed that the risk of liver fibrosis from alcohol exposure was higher in those co-infected with HIV and HCV relative to those infected with HIV only (1.60 vs. 1.00), and that risk was the highest in the mixed group. While another cross-sectional study (55) by Ferguson presented a contrary conclusion, it implied that those infected with HIV only had a higher OR than those co-infected with HIV and HCV (7.68 vs. 3.68). Co-infected with HCV can definitely influence liver health status among PLWH (42), but a more rigorously designed cohort study is needed to confirm the effect of alcohol.

HCV infection is an independent risk factor for liver fibrosis (56). HIV and HCV share the same routes of virus transmission. According to the previous study, the co-infection ratio of HIV and HCV was about 10% of patients with HIV in the western world (57). Previous reported that in China, the highest HIV/HCV prevalence was in the Central region (28.2% of PLWH) (58). HCV has the same infection route as HIV, especially injecting drugs, so the HIV-HCV co-infection rate of injecting drug users (IDUs) is much higher than that of other populations. In a study of 359 IDUs with HIV infection by A-Mei Zhang et al., the HCV co-infection rate was as high as 45.68% (2). Over two-thirds of patients with HIV-HCV co-infection had moderate-to-significant liver fibrosis as measured by either APRI or FIB4 in China (36). In other countries, the liver fibrosis rate in HIV-HCV co-infected patients is 30%-50% (59, 60) detected by transient elastography. HIV/HCV co-infected patients were more likely to have liver fibrosis compared with HIV or HCV mono-infected (61). Patients infected with HIV-1 or HCV or who drink alcohol exhibit metabolic abnormalities that cause changes in the levels of n-acetylaspartate, choline, and creatine in different regions of many organs, including the brain and liver (62). Treatment of HIV/HCV coinfection in alcohol users is complicated by drug-drug interactions and the effects of alcohol on drug metabolism (63). But other researchers found liver fibrosis progresses faster in HIV mono-infection, other than in HIV/HCV co-infection (64). Direct-acting antivirals (DAA) lead to a decreased risk of HCV progression. After DAA treatment, HIV/HCV co-infected patients had similar risk of liver-related deaths and events but had a higher risk of all-cause and non-liver-related deaths and non-liver-related cancers compared with HCV mono-infected individuals (65). Although DAA treatment can successfully increase HCV cure rates with minimal toxicity in HIV/HCV patients (66), there are few cohort studies on liver fibrosis and other diseases in HIV/HCV co-infected patients receiving DAA for several years. The delayed effect of liver fibrosis and disease should be continually followed in the future study.

There are some limitations of this review. First, the number of the included articles and the contributions’ quality were limited. Only three studies in our study are cohort studies and cannot be analyzed. Also, we were unable to investigate in detail the role of many study design characteristics, such as adjustment for potential confounders, the length of alcohol abuse, and others that may play a role in the development of liver fibrosis, because of the small number of studies published. Second, although self-reported alcohol consumption is generally reliable, it may result an underestimate with the actual consumption of alcohol. Third, different studies adopting different diagnose criteria will cause differences between studies. Although it is generally accepted that the diagnosis of liver fibrosis is 4≥F≥3, the application of different detection methods will have certain errors in the diagnosis of liver fibrosis. Fourth, the heterogeneity within studies is very high, so we used a random effects model to analysis the data.

As for clinical use, it is necessary to elucidate the role of other risk factors, such as genetic vulnerability, gender, metabolic risk factors, and drinking patterns over the life course, in high-quality research. Also, alcohol abuse should be avoided in PLWH, especially in HIV and HCV co-infected patients. Patients drinking at high levels should regularly receive inspections of liver fibrosis-related indicators and related interventions to reduce their intake. If necessary, patients should be treated with hepatoprotective drugs. For doctors, when prescribing antiviral treatments for HIV-infected patients with alcohol abuse, hepatotoxicity and metabolic toxicity of the drugs should be fully considered, and therapeutic drugs with less hepatotoxicity should be used as much as possible.

This study reported that high-risk alcohol consumption is a risk factor of liver fibrosis for PLWH, especially for HIV/HCV co-infected patients. For different alcohol consumption subgroups, we considered the daily alcohol consumption is more sensitive for analysis of the effect of alcohol abuse rather than AUDIT-C score. Inconsistent definitions of alcohol exposure can lead to different conclusions. As for whether HCV co-infected subgroups, high-risk alcohol consumption has a higher risk of liver fibrosis in HIV and HCV co-infected patients rather than HIV mono-infected patients. Hepatotoxicity drugs should be considered by doctors when prescribing antiviral treatments for HIV-infected patients with alcohol abuse, or co-infected with HCV.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

Conceptualization: HL and YZ. Methodology: HL and YZ. Software: HL and HT. Validation: HL and YL. Formal analysis: SH, YW, and YZ. Data curation: YL. Writing—original draft preparation: HL and HT. Writing—review and editing: HL and YZ. Visualization: HL. Supervision: YZ. Project administration: HL and YZ. Funding acquisition: YZ. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The handling editor WT has declared a past co-authorship with one of the authors YZ.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.