Efficacy and Safety of Variable Treatment Options in the Prevention of Hepatic Encephalopathy: A Systematic Review and Network Meta-Analysis

There are no guidelines for the most effective medication to reduce hepatic encephalopathy (HE) or the associated mortality. The purpose of this study is to determine the most effective possible treatment among the single treatment options or the combined treatment options for decreasing the morbidity and mortality of HE. We evaluated the outcomes by various parameters such as the quality of life, reduction in ammonia, all causes of mortality, adverse events, reversal of minimal HE, and development of overt HE. We systematically searched PubMed, Cochrane, Web of Science, and Scopus till the 19th of January 2023 for studies that assess various treatment options for HE. Data were extracted from eligible studies and pooled in a frequentist network meta-analysis as standardized mean difference (SMD) and their 95% confidence interval (CI) using the MetaInsight web-based tool. The Cochrane Tool was used to assess the randomized controlled trials' quality (RCT), while the NIH tool was used to assess the quality of the included cohort studies. Utilizing the R software, the network meta-analysis was conducted. In addition to a significant variation in cases of (Lactulose and Rifaximin) compared with Rifaximin (RR= 0.39, 95% CI [0.17; 0.89]), the results demonstrated a significantly lower incidence of overt HE in (Lactulose and Rifaximin) compared with placebo (RR=0.19, 95% CI [0.09; 0.40]). Most arms demonstrated a statistically significant reduction in the incidence of overt HE compared to albumin and placebo. The results also demonstrated a significant reduction in ammonia between L-ornithine-L-aspartate (LOLA) and probiotics (MD= -19.17, 95% CI [-38.01; -0.32]), as well as a significant difference in the incidence of LOLA compared to placebo (MD= -22.62, 95% CI [-39.16; -6.07]). This network meta-analysis has significant data for managing subclinical HE in people without a history of overt HE. Our analysis showed that (Lactulose and Rifaximin), followed by (Rifaximin and L-carnitine), followed by (Lactulose and Rifaximin with zinc) were the best combinations regarding overt HE. LOLA reduced ammonia best, followed by Nitazoxanide and finally Lactulose. (Lactulose and Nitazoxanide) have the least adverse effects, followed by (Rifaximin and L-carnitine), then Probiotics. Yet, all mortality outcomes and quality of life changes yielded no useful findings. Future studies like RCTs must be done to compare our therapies directly.


Introduction And Background
Cirrhosis is the 14th most common cause of death worldwide [1], and the 12th most common cause of death in the United States [2], which represents a challenge for healthcare providers to extend the patient's life without going through liver transplantation.Under physical stress, patients with liver cirrhosis have a lower ventricular ejection fraction than non-cirrhotic subjects.Hyperdynamic circulation, cirrhotic cardiomyopathy, and pulmonary vascular abnormalities are just some of the cardiovascular abnormalities linked to liver cirrhosis.Moreover, cirrhosis can occur with hepatic encephalopathy (HE) due to the brain accumulation of ammonia and manganese accompanied by inflammation [3].HE refers to brain dysfunction caused by liver insufficiency and/or portal-systemic blood shunting [4].In both chronic liver disease and acute liver failure, HE is a common and potentially fatal complication [5].The mildest form of HE is called minimal HE, and it's linked to forgetfulness and an inability to focus.The mild HE that can develop from the minimal HE causes alterations in mood and sleep patterns.Mild HE has the potential to progress to moderate HE, which can have an impact on the patient's personality and behavior in addition to their slurred speech and mathematical difficulties.Disorientation, prolonged periods of sleep, coma, and even death can result from untreated severe HE if the case becomes more complex.Patients with chronic liver disease but no obvious HE have a reported prevalence of 30%-84% for minimal HE.The main way to treat HE is the reduction in the ammonia level in the blood [6].Minimal HE is the highest incidence as it ranges from 30-84% in patients with chronic liver disease without overt HE [7,8].Minimal HE is easily underestimated due to the word minimal as a prefix, although previous studies have reported high rates of deaths due to incorrect diagnoses of minimal HE as well as the damages that minimal HE may cause to society [8], as patients with minimal HE were reported to have a higher crash rate due to accidents than the normal population due to the effects of minimal HE on impairing driving skills and reducing conscious awareness [9].There are a lot of treatment options to reduce the high level of ammonia in blood as Lactulose, Rifaximin, branched-chain amino acids (BCAA), L-ornithine-L-aspartate (LOLA), and Probiotics.In our network meta-analysis we are investigating the multiple interventions to treat the different types of HE.

Review Methods
We conducted our study in substantial accordance with the Cochrane handbook guidelines for systematic reviews of interventions, then we reported according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) [10,11].

Search Strategy and Data Collection
(albumin OR probiotics OR prebiotics OR lactobacillus OR bifidobacterium OR symbiotics OR lactulose OR lactitol OR disaccharides OR rifaximin OR rifagut* OR xifaxan* OR rcifax* OR "branched chain amino acid" OR "branched chain amino acids" OR "L-ornithine L-aspartate" OR LOLA OR BCAA OR BCAAs OR "branched chain" OR "amino acids")AND (encephalopathy OR cirrhosis OR "minimal hepatic encephalopathy" OR "subclinical hepatic encephalopathy" OR "latent hepatic encephalopathy" OR "covert hepatic encephalopathy" OR "liver cirrhosis")

Data Extraction
We extracted the data related to the following: Summary of the included studies, including inclusion criteria, study design, follow-up, study groups, etiology of cirrhosis, and conclusion.Baseline characteristics of the enrolled population, including sample size, age, study ID, gender, site, ammonia, MELD score, and outcomes of change in health-related quality of life, reduction in ammonia, adverse events, all causes of mortality, development of overt HE, reversal of minimal HE.

Quality Assessment
We used Cochrane's risk of bias tool (version 1) to assess the included interventional studies' quality.The tool is reported in chapter 8.5 of the Cochrane Handbook for Systematic Reviews of Interventions 5.1.0.The tool consists of the following assessment items: sequence generation, allocation sequence concealment, blinding of participants and personnel, blinding of outcome assessors, incomplete outcome data, selective outcome reporting, and any other bias; author judgments fall into three categories; low, unclear or high risk of bias for each item.We used the quality assessment table in part 2, Chapter 8.5 of the same book [12].We used the tool of observational cohort studies, which is composed of questions assessing the risk of bias and confounders.Each question was answered by "yes", "no", "not applicable", "not reported", and "cannot determine", then each study was given a score to guide the overall quality either "poor", "fair", or "good".

Statistical Analysis
Using frequentist network meta-analysis with random-effects models, the continuous data were pooled as mean difference (MD) and their 95% confidence interval (CI), while the binary data was extracted as risks ratios (RRs).MetaInsight Version 3.14, an interactive web-based tool for network meta-analyses based on the R-shiny and netmeta statistical packages, was used for all statistical analyses.Heterogeneity was assessed under clinical, methodological, and statistical domains.Statistical heterogeneity was assessed using the I2 statistics [13].

Literature Search
The initial results after searching our main four databases were 21017 results, with the help of the EndNote program we removed 6467 duplicates then we did a title and abstract screening for 14550 results.The final result was to include 43 studies .The full PRISMA is presented in Figure 1.

Included Population's Baseline Characteristics & Included Studies' Summary
Our study included 32 randomized clinical trials (RCTs) and 10 cohort studies, both retrospective and prospective, as well as one case-control study.Numerous international and/or multicenter studies contributed to the global scope of the included studies.We included patients from various nations, including Germany, the United States, Japan, Egypt, China, and Taiwan.The full baseline and summary characteristics are presented in Table 1.

Risk of Bias Assessment
RCTs assessed by the Cochrane ROB tool showed a high risk of bias, mostly due to the high risk of bias regarding blinding status. Figure 2 shows the summary of the risk of bias in RCTs.
2024 Most of our observational studies assessed by the NIH tool showed fair quality; this could be attributed to insufficient sample size, not reporting the number of exposure assessments, and the patients were not blinded.The summary of cohort and case-control studies is shown in Table 2.The results showed a significant reduction in the incidence of development of overt HE between (Lactulose and Rifaximin) compared with Lactulose (RR=0.57,95% CI [0.36; -0.92]), in addition to a significant variation in cases of (Lactulose and Rifaximin) compared with Rifaximin (RR=0.39,95% CI [0.17;0.89])with another significantly lower incidence of development of overt HE in (Lactulose and Rifaximin) compared with placebo (RR=0.19,95% CI [0.09; 0.40]).Most of the arms showed a significant reduction in the incidence of development of overt HE compared with albumin and placebo.The order of drugs in terms of incidence of development of overt HE according to P-score is (Lactulose and Rifaximin), (Rifaximin and L-carnitine), (Lactulose and Rifaximin with zinc), Nitazoxanide, Probiotics, Lactulose, L-ornithine L-aspartate, Rifaximin, placebo, and Albumin (Figure 3).

FIGURE 4: Reduction in ammonia
Reversal of Minimal HE In the outcome of reversal of minimal HE all the significant results were with placebo which is expected, so we find the only non-significant results compared to Placebo were with branched-chain amino acids (RR=0.30,95% CI [0.06; 1.59]), on the other hand, Placebo showed a significant result compared with Rifaximin (RR=0.44,95% CI [0.27; 0.73]), and with Lactulose (RR=0.49,95% CI [0.35; 0.69]) respectively.The order of drugs in terms of reversal of minimal HE according to P-score from low to high is Placebo, Probiotics, Probiotics and Lactulose, Lactulose, L-ornithine L-aspartate, Rifaximin, and branched-chain amino acids (Figure 5).

All Causes of Mortality
The results showed no significant decrease in all causes of death in the comparison among all the tested medications compared with the control.The order of drugs in terms of all causes of mortality incidence according to P-score is Albumin followed by (Lactulose and BCAA), PEG, (Lactulose and Rifaximin), Lornithine L-aspartate, (Lactulose and Rifaximin with zinc), (Lactulose and Nitazoxanide), Probiotics, Lactulose, Placebo, and finally Rifaximin (Figure 7).

FIGURE 7: All causes of mortality
Change in Health-Related Quality of Life (HRQOL) No available treatment was found significant in the improvement of health-related quality of life, as lactulose showed no significant result compared to placebo (MD=-3.29,95% CI [-7.54; 0.95]), and the order of drugs in terms of change in HRQOL according to P-score is Lactulose, Rifaximin, Nitazoxanide, Probiotics, and placebo (Figure 8).

Discussion
In our network meta-analysis, 32 RCTs and 11 cohorts were included with a total population of 19622.Our results showed that lactulose and rifaximin were significant compared to most of the arms in the outcome of the development of overt HE, in addition to the reversal of minimal HE that showed significance in the placebo group.Moreover, the reduction in ammonia showed significance in the L-ornithine L-aspartate drug compared to other groups.No significant results were found in HRQOL as all arms had the same effect on the quality of life.Hepatic encephalopathy was found to impair daily function and driving skills due to a decrease in cognition [9,57].HE increases the burden on medical care providers, especially with the poor prognosis for HE in all its forms [58].Some papers found that a combination of Rifaximin and Lactulose would benefit in decreasing mortality [59], although our study showed no significance among the available treatment options as they all would have decreased mortality, no results are significant.In our study results, Lactulose was a common factor in most of our significant results and showed the highest rate of improvement in the quality of life [60], which matched previous studies in addition to the benefit of cost reduction due to the cheap cost of lactulose usage compared to other drugs [61].
L-ornithine L-aspartate showed the highest reduction in serum ammonia in our results and the results were significant compared to the other arms, the serum ammonia is one of the core factors for hepatic encephalopathy pathogenesis.A previous study proved the effectiveness of L-ornithine L-aspartate in decreasing the ammonia levels in HE [62].Although the combination between Lactulose and nitazoxanide is showing promising results in adverse event outcomes, more studies investigating this arm are still required with higher populations [62,63].All our included drugs were well tolerable when it comes to adverse events even the lactulose except for nausea, bloating, and diarrhea but all these side effects could be overcome with dose adjustments or the addition of nitazoxanide [63].This analysis allowed us to compare available treatment options for achieving clinically relevant endpoints and provided a ranking of each treatment's efficacy.This method was the best alternative to a direct comparison study for comparing interventions for HE.
The strength points of this study are that it contains a large number of RCTs in addition to cohort studies.
Our study has a high total population as 19622 patients were included, which would provide stronger evidence as long as the RCTs would provide more solid data with low bias, as RCTs are the gold standard for strong evidence.Our study also contained variable arms with different comparisons between these arms as we almost included every available way of treatment for hepatic encephalopathy in the market, every single part of our meta-analysis was reviewed at least twice, moreover all PRISMA recommendations were followed while making this network meta-analysis.Our study also had some limitations as not all inclusion criteria for patients with hepatic encephalopathy were the same.The duration of the follow-up period wasn't equal in all included studies.In addition, some of the doses were different and we made no subgroups for the different doses.

Conclusions
We provided an extensive body of evidence for the management of subclinical HE in patients without a history of overt HE which is compiled in this network meta-analysis (NMA).We can show evidence of the development of overt HE (Lactulose and Rifaximin followed by Rifaximin and L-carnitine, and followed by Lactulose and Rifaximin with zinc).According to the reduction of ammonia, LOLA was ranked first, followed by Nitazoxanide, and finally Lactulose.The order of incidence of adverse events is lowest in (Lactulose and Nitazoxanide) followed by (Rifaximin and L-carnitine) and finally Probiotics.On the other hand, no valuable results were obtained from all causes of mortality outcomes and changes in the quality of life.

FIGURE 1 :
FIGURE 1: PRISMA Flow diagram PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and registers only.

FIGURE 8 :
FIGURE 8: Change in HRQOL HRQOL: Health-related quality of life

TABLE 2 : NIH Quality Assessment tool
OutcomesDevelopment of Overt HE 2024 Hammd et al.Cureus 16(1): e53341.DOI 10.7759/cureus.53341 Further studies in the future are required to investigate our interventions with direct comparisons among all of our interventions.