The impact of serum neuregulin 4 (NRG-4) levels on non-alcoholic fatty liver disease (NAFLD): a systematic review and meta-analysis

Abstract

Purpose

The aim of this systematic review and meta-analysis was to assess the association between serum neuregulin 4 (Nrg4) levels and non-alcoholic fatty liver disease (NAFLD).

Methods

A comprehensive search was performed across Web of Science, EMBASE, Medline (PubMed), and Scopus to find all original published works in NAFLD up to March 2024. Studies were limited to case-control observational studies on humans. After removing duplicated studies, studies were screened based on predefined inclusion and exclusion criteria. The Joanna Briggs Institute checklist (JBI) was used for assessing the quality of case-control studies. The strength of the relationship was assessed by the odds ratios (ORs) and corresponding 95% confidence intervals (CIs). All statistical analyses were performed using Stata 17.0.

Results

Five case-control articles with 323 cases and 308 controls were included in the meta-analysis. The pooled mean ages were 42.6 years for the NAFLD group and 38.0 years for the control group. The pooled estimated OR between Nrg4 and NAFLD was 0.72 (95% CI: 0.67–0.77). The analysis shows that with increasing age (B: 1.381; SE: 0.732; P-value: 0.200; 95% CI: 0.771, 4.534), waist circumference (B: 0.076; SE: 0.041; P-value: 0.207; 95% CI: 0.002, 0.256) and TG levels (B: 0.0426; SE: 0.023; P-value: 0.212; 95% CI: -0.058, 0.143), the association of the Nrg4 index with incident NAFLD increased. Furthermore, the strength of the association between the Nrg4 index and incident NAFLD decreased with increasing levels of ALT (B: -0.124; SE: 0.119; P-value: 0.407; 95% CI: -0.638, 0.389) and HDL (B: -0.221; SE: 0.130; P-value: 0.232; 95% CI: -0.783, 0.340).

Conclusion

In summary, the findings from this meta-analysis strongly suggest that Nrg4 exerts a protective effect against the development of NAFLD. The evidence indicates that measuring Nrg4 levels holds significant promise for facilitating the early diagnosis and prognostication of NAFLD. However, it’s essential to note that further research with larger sample sizes and more accurate study designs is needed to corroborate these findings and establish the clinical utility of Nrg4 as a biomarker in the management of NAFLD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12876-025-04097-2.

Introduction

Non-alcoholic fatty liver disease (NAFLD) spectrum comprises simple steatosis and nonalcoholic steatohepatitis (NASH) that can lead to fibrosis, cirrhosis, and hepatocellular carcinoma [1]. NAFLD pathogenesis is multifactorial and complex. It involves environmental, metabolic, and genetic factors, which are implicated in the excessive accumulation of triglycerides in hepatocytes [2]. The estimated global prevalence of NAFLD is approximately 30.2%. Regionally, the prevalence has been reported as 30.9% in Asia, 16.1% in Australia, 30.2% in Europe, 29% in North America, and 34% in South America [3]. Insulin resistance (IR), metabolic factors, and nutrition contribute to the progression and development of NAFLD, despite this, the underlying pathogenic mechanisms remain unclear [4].

The liver communicates closely with adipose tissue. Dysregulation in lipid metabolism and dysfunction of adipose tissue in an IR status lead to steatosis in the liver [5]. Dysregulated secretion of adipokines, a variety of biologically active substances secreted by adipose tissue, contributes to the development of obesity-related metabolic diseases [6]. Neuregulin 4 (Nrg4) belongs to the epidermal growth factor (EGF) family. It is considered as one of the adipokines [7, 8] that acts as an endocrine, paracrine, or autocrine signal by releasing the EGF-like domain upon proteolytic cleavage [9, 10]; implicated in physiological processes such as angiogenesis, thermogenesis, glucose metabolism, lipid metabolism, and neurogenesis, but also pathological processes, Nrg4 suppresses apoptosis and reduces inflammatory factor levels in inflammatory diseases. Furthermore, Nrg4 could improve obesity, cardiovascular diseases, and insulin resistance [11]. Nrg4 is expressed in multiple organs, with the highest levels found in brown adipose tissue (BAT) [12], additionally, it is upregulated in white adipose tissue (WAT) when exposed to cold and plays a role in the differentiation of brown adipocytes [13]. Neuregulins (Nrg1-4) act as ligands for the receptor tyrosine kinases of the ErbB/HER family, which play key roles in regulating various biological processes [14]. Nrg1 and Nrg2 attach to both ErbB3 (HER3) and ErbB4 (HER4), while Nrg3 and Nrg4 primarily bind to ErbB4 [15, 16]. Investigations have reported that Nrg4 dysregulation is involved in the development and occurrence of several diseases, such as metabolic syndromes [17, 18], chronic inflammation [19], cancers [20], and some other diseases [21, 22]. Moreover, Nrg4 ameliorates NAFLD by improving the metabolism of lipids, increasing autophagy, and suppressing hepatocyte death [11]. Recent studies suggest that Nrg4 suppresses hepatic lipogenesis by downregulating genes such as Srebf1, Fasn, Scd1, and Acaca via inhibition of LXR signaling. It also promotes autophagy through activation of AMPK and inhibition of mTOR signaling, contributing to hepatic lipid clearance. Furthermore, Nrg4 modulates STAT5 and JNK1/2 signaling to reduce hepatocyte apoptosis and inflammation, helping to prevent the progression from simple steatosis to NASH [23]. These mechanisms highlight Nrg4’s potential as both a regulator of lipid metabolism and a protective factor against the progression of NAFLD. When mice overexpress Nrg4, it regulates liver fat production and prevents obesity and NAFLD caused by a high-fat diet. Conversely, Nrg4 deficiency worsens liver fat buildup and insulin resistance. Nevertheless, clinical studies in human adults have challenged this association by demonstrating that patients with NAFLD have lower serum Nrg4 levels compared to non-NAFLD controls [24].

Nrg4, a recently discovered secreted adipokine, has attracted increasing attention due to its close association with NAFLD. Although research on this association has emerged in recent years, a comprehensive understanding of Nrg4’s role in NAFLD pathogenesis is still lacking. The key hypothesis in this present meta-analysis is that changes in Nrg4 levels are significantly associated with NAFLD susceptibility and development. By combining existing study findings, aims to provide a comprehensive assessment of the current evidence, shedding light on Nrg4’s potential diagnostic and therapeutic implications for NAFLD. Furthermore, the results of this meta-analysis may guide future research and clinical interventions, potentially paving the way for new diagnostic biomarkers or therapeutic techniques targeting Nrg4 pathways to mitigate the expanding global impact of NAFLD.

Materials and methods

The present systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) and guidelines for reviewing analytical observational studies [25]. This systematic review is registered at PROSPERO (CRD42023491780).

Eligibility criteria

The research included human individuals with NAFLD or NASH and looked at the relationship between serum Nrg4 levels and NAFLD/NASH. These investigations evaluated Nrg4 levels across patients and healthy controls, as well as between illness severity stages. Furthermore, they reported the effect sizes (odds ratio (OR) or risk ratio (RR) of serum Nrg4 levels on NAFLD/NASH severity and progression. Analytical observational research, such as case-control and cohort studies, were evaluated. Exclusion criteria included studies that did not involve human participants, did not focus on the Nrg4-NAFLD/NASH association, lacked comparison groups or relevant end measures, did not disclose quantitative data on serum Nrg4 levels or their association with NAFLD/NASH, and were in another language.

Information sources, search strategy, and selection process

To acquire relevant studies for this review, authors searched many worldwide databases, including Web of Science, Excerpta Medica Database (EMBASE), Medline (PubMed), and Scopus. The search aimed to find all original published works that investigated and reported on the relationship between Nrg4 and NAFLD up to March 2024. A set of specific keywords such as “non-alcoholic fatty liver disease”, “NAFLD,” “nonalcoholic steatohepatitis”, “neuregulin 4”, “Nrg4”, “neuregulins”, “Nrg4 biomarker NAFLD”, and “NASH” was used to ensure comprehensive coverage (Supplementary File; Table S1).

Studies were limited to observational studies on humans (Table 1). Additionally, to capture potential grey literature, authors reviewed articles in the first ten pages of Google Scholar. To further minimize the risk of missing relevant studies, authors also performed manual searches of the reference lists and citation tracking of included studies. This process helped identify studies not indexed in major databases. Studies were restricted to observational research involving human participants.

Table 1.

Inclusion, exclusion criteria, and search terms in the present Meta-analysis

Criteria	Description
Population	Human participants diagnosed with NAFLD or NASH.
Exposure	Studies on serum Nrg4 levels and NAFLD/NASH association
Comparison	Studies comparing Nrg4 levels between NAFLD/NASH patients and healthy controls or among different disease severity stages.
Outcome	Studies reporting effect sizes (odds ratio (OR) or risk ratio (RR)) of serum Nrg4 levels concerning NAFLD/NASH severity and progression
Type of studies	Analytical observational studies like case-control or cohort studies.
Exclusion criteria	Studies not involving human participants. Studies not focusing on the association between serum Nrg4 levels and NAFLD/NASH. Studies lacking comparison groups or relevant outcome measures. Studies not reporting quantitative data on serum Nrg4 levels or their association with NAFLD/NASH. Non-English studies.
Search Terms	“Non-alcoholic fatty liver disease” “NAFLD” “Nonalcoholic steatohepatitis” “Neuregulin 4” “Nrg4” “Nrg4 and nonalcoholic steatohepatitis” “Neuregulins” “Nrg4 biomarker NAFLD” “NASH”
Databases	Web of Science Excerpta Medica Database (EMBASE) Medline (PubMed) Scopus

After collecting and organizing articles from each database into a library, duplicates were found and deleted using Endnote 8 software tools as well as manual evaluation. The titles and abstracts of the remaining publications were then screened using preset inclusion and exclusion criteria. Two writers (MAT and SR) evaluated papers independently, with a third author (YM) resolving differences. Following the initial screening, the selected publications’ full texts were reviewed to determine their inclusion in the research.

Data collection process

The following data were extracted using a checklist that included: (1) first author, country of origin, publication year, study type, NGR4 measurement method; (2) age, number of subjects, waist circumference, body mass index (BMI), ALT, AST, LDL, HDL, TG, Cholesterol, FBS in both case and control groups; (3) odds ratios (ORs) and corresponding 95% confidence intervals (Cls). Two investigators (MAT and SR) independently assessed all included articles. They assessed any disagreements, and if no agreement was reached, the study was assessed by a third author (YM).

Quality assessment

The Joanna Briggs Institute (JBI) critical appraisal checklist was used to assess the quality of case-control studies. The goal of this checklist is to evaluate the quality of observational studies. The purpose of this checklist is to establish an article’s methodological quality and the degree of potential bias in its conduct, design, and analysis. JBI and collaborators created the critical appraisal tools, which were approved by the JBI Scientific Committee following comprehensive peer review. The case-control checklist consists of ten questions marked with ‘yes’, ‘no’, ‘not applicable’, and ‘unclear’. Finally, the studies were evaluated based on the number of ‘yes’ responses [26, 27].

Kappa statistics

Kappa statistics were calculated to assess inter-rater agreement between the two independent reviewers during the data screening and selection process. A kappa value of 0.60 or higher was considered acceptable for inter-rater agreement [28]. If the kappa value was below 0.60, the studies were discussed between the two reviewers to reach consensus, and if necessary, the third investigator (YM) was involved in the final decision.

Statistical analysis

The strength of the association between NGR4 and the increase or decrease of NAFLD risk was assessed by odds ratios (ORs) and corresponding 95% confidence intervals (CIs) to estimate its effect on a forest plot. To conduct the meta-analysis, authors first estimate the natural logarithm (log) and log standard errors of each indicator from the selected studies. These transformed values help standardize the data and account for variability among studies. A random-effects model (REM) was applied for the overall pooled analysis due to significant heterogeneity among studies. For subgroup analyses, a fixed-effect model (FEM) was used, as heterogeneity within subgroups was minimal, allowing for more precise effect estimation. The pooled estimate of each indicator, along with its corresponding 95% CI, is then calculated. Funnel plot asymmetry and egger’s test were used to assess and determine the potential publication bias. Heterogeneity was assessed using the Q Cochrane tests and I² index, the heterogeneity rate was categorized into four groups: 0–40% (may not be important), 30–60% (may represent moderate heterogeneity), 50–90% (may represent substantial heterogeneity), and finally 75% and above (considerable heterogeneity) [29–32].

Subgroup analyses based on Nrg4 serum level were conducted, and meta-regression analysis was performed based on indices such as age, waist circumference, BMI, ALT, AST, LDL, HDL, TG, Cholesterol, and FBS. All statistical analyses were performed using Stata 17.0. Authors converted FBS, cholesterol, LDL, HDL, and TG units (mmol/l to mg/dl) in Dai et al., Wang et al., and De Munck et al. studies, also authors changed the Nrg4 unit in De Munck et al. study (pg/ml to ng/ml) [33]. To explore potential sources of heterogeneity, a subgroup analysis was performed based on Nrg4 cutoff values. A cutoff of 4 ng/mL was selected as a clinically relevant threshold based on prior studies that examined serum Nrg4 levels in relation to NAFLD severity.

Results

Qualitative results

After completing the search in international databases, 69 articles were retrieved. Twenty-three duplicates were removed, and then forty-six articles were entered into screening stage. After screening the title, abstract and full-text five articles remained for meta-analysis of the association between NAFLD and the Nrg4 factor (Fig. 1). For this meta-analysis all articles were case-control [24, 34–37], of which one included children as participants [37] and others were adults [24, 34–36]. Four articles were conducted in Asia and one study was carried out in Europe. Nrg4 was extracted from serum, and the assay method used was ELISA. All studies were written in English. The five studies included in this article comprised a total of 323 individuals in the NAFLD group and 308 individuals in the control group. Among the cases, 174 were males and 149 were females; in the control group, 189 were males and 119 were females. The mean age of participants in the case group ranged from 10.7 to 54 years, and in the control group from 10.5 to 52.6 years. To provide a general overview of the age distribution, pooled mean ages were calculated using a weighted average based on sample sizes. The resulting pooled mean age was 42.6 years for the NAFLD group and 38 years for the control group. Table 2 shows the baseline characteristics of included articles. Supplementary File; Table S2 contains a list of papers that were excluded during the full text screening stage.

Fig. 1.

The process of screening based on title, abstract, and full text

Table 1.

Characteristics of eligible studies included in the present meta-analysis

First Author	Publication year	Country	Total cases (man/women)	Total controls (man/women)	Average age of Cases/Controls	NAFLD diagnostic test	NRG4 (case/control) (ng/ml)	waist circumference (case/control) (cm)	BMI (case/control) (kg/m2)	ALT (case/control) (U/L)	AST (case/control) (U/L)	LDL (case/control) (mg/dl)	HDL (case/control) (mg/dl)	TG (case/control) (mg/dl)	Cholesterol (case/control) (mg/dl)	FBS (case/control) (mg/dl)	Effect size (OR with a 95% CI) or (RR with a 95% CI)	JBI score
Dai et al	2015	China	87 (53/34)	87 (53/34)	52.3/52.6	Abdominal ultrasonography	0.4/0.5	92.95/81.45	26.13/22.52	22/17	20/19	102.86/99	42.54/48.34	134.63/100.09	179.8/175.56	100.8/95.4	0.251 (0.081-0.779)	8
Wang et al	2018	China	58 (41/17)	65 (43/22)	10.7/10.5	Abdominal ultrasonography	2.24/5.5	91.4/85.6	27.6/25.5	29.5/17	22/21	NR	48.72/49.5	141.72/100.09	175.95/170.92	84.8/83.5	0.129 (0.028-0.587)	9
De Munck et al.	2021	Netherland	65 (31/34)	43 (21/22)	54/40	MRI	0.0892/0.0895	NR	32.37/22.24	40/NR	28/NR	121.04/NR	48.72/NR	131.98/NR	197.6/NR	108/NR	.773 (1.891-17.357)	6
Tutunchi et al.	2021	Iran	50 (28/22)	50 (28/22)	50.2/48.3	Abdominal ultrasonography	2.1/4.3	94.8/88.2	29.3/24.1	27/18	24/19	125.4/122.3	43.9/48.2	169.2/153.2	180.1/178.2	89.9/83.2	0.59 (0.35-0.78)	9

Quantitative results

One article was excluded from this study due to a poor effect size [35]. In this article the smallest effect size belonged to Wang et al. with 0.13 (95% CI; 0.10–0.17) and the highest effect size belonged to Gado et al. with 0.87 (95% CI; 0.81–0.94). After combining the results, the pooled estimate OR was 0.36 (95% CI; 0.16–0.84). The heterogeneity in this study was high and equal to 98.50% (Fig. 2). To assess potential publication bias, Egger’s regression-based test was conducted and revealed statistically significant small-study effects (β = − 11.18, SE = 5.99, p = 0.032) (Supplementary File; Figure S1). Additionally, the nonparametric Trim-and-Fill method was applied, which suggested that no studies needed to be imputed, indicating minimal evidence of publication bias despite the small-study effect signal.

Fig. 2.

Forest plot of the Impact of Serum Neuregulin 4 (NRG‐4) Levels on Non-Alcoholic Fatty Liver Disease (NAFLD)

Subgroup analysis

The result of subgroup analysis by Nrg4 cutoff of 4 ng/dl showed that pooled OR was higher in studies with Nrg4 more than 4 ng/dl (OR = 0.80; 95% CI: 0.74–0.85) compared to those with Nrg4 lower than 4 ng/dl (OR = 0.13; 95% CI: 0.10–0.17). Heterogeneity was significant among studies with Nrg4 levels above 4 ng/dl (I² = 96.20%) (Table 3).

Table 3.

Subgroup analysis regarding the relationship between Nrg4 serum level and NAFLD

Subgroup	Number of studies	Summary OR (95% CI)	I2	Q	P value
NRG4 cutoff (ng/ml)
Lower than 4	1	0.13 (0.10–0.17)	-	-	-
Higher than 4	3	0.80 (0.74–0.85)	96.20%	52.65	< 0.0001
Overall	4	0.72 (0.67–0.77)	98.50%	66.70	< 0.0001

Meta-regression analysis

Meta-regression analyses were conducted to explore potential sources of heterogeneity, including age, sample size, and publication year. However, none of the covariates demonstrated a statistically significant association (all p > 0.05), and these results should therefore be interpreted as exploratory and hypothesis-generating. The results indicated that with an increase in patients’ age, the association of the Nrg4 index with NAFLD occurrence also increased (B: 1.381; SE: 0.732; P value: 0.200; 95% CI: 0.771, 4.534). This means that with advancing age, Nrg4 became a better indicator for determining the severity and occurrence of NAFLD. Additionally, for waist circumference and BMI, the results showed that with an increase in waist circumference (B: 0.076; SE: 0.041; P value: 0.207; 95% CI: 0.002, 0.256) and BMI (B: 0.156; SE: 0.086; P value: 0.071; 95% CI: −0.013, 0.326), the association of the Nrg4 index with NAFLD occurrence also increased. Also, these results indicated that with an increase in the mean ALT level, the association and strength of the relationship between the Nrg4 index and NAFLD occurrence decreased (B: −0.124; SE: 0.119; P value: 0.407; 95% CI: −0.638, 0.389), although this was not statistically significant. Conversely, with an increase in mean AST (B: 0.023; SE: 0.331; P value: 0.07; 95% CI: −1.404, 1.450), FBS (B: 0.022; SE: 0.090; P value: 0.844; 95% CI: −1.132, 1.178), total cholesterol (B: 0.017; SE: 0.011; P value: 0.280; 95% CI: −0.033, 0.067), and LDL (B: 0.326; SE: 0.004; P value: 0.08; 95% CI: −0.024, 0.090), no significant associations were observed between the Nrg4 index and NAFLD occurrence. With an increase in HDL, the strength and association of the Nrg4 index with NAFLD occurrence decreased (B: −0.221; SE: 0.130; P value: 0.232; 95% CI: −0.783, 0.340), whereas with an increase in TG, this association increased (B: 0.0426; SE: 0.023; P value: 0.212; 95% CI: −0.058, 0.143). However, none of these associations were statistically significant (Figs. 3 and 4).

Fig. 3.

Meta-regression analysis for Assessing the Impact of NRG‐4 Levels on NAFLD based on Age, Waist Circumference, BMI, and Cholesterol

Fig. 4.

 Meta-regression analysis for Assessing the Impact of NRG‐4 Levels on NAFLD based on AST, HDL, TG, FBS, and ALT

Discussion

This is the first meta-analysis to assess the association between Nrg4 and NAFLD in humans. Nrg4 has emerged as a potential predictive factor in NAFLD patients. The present paper provides evidence supporting this hypothesis. This study, based on statistical data, found that Nrg4 has a 28% protective role in NAFLD development, and lower serum Nrg4 levels increase the chance of developing NAFLD. However, the reported protective impact is based simply on statistical associations, and causality cannot be shown. This is in good agreement with Wang et al. [37] study in which increasing serum Nrg4 levels in children with obesity was correlated with reduced risk of NAFLD. These result is similar to Dai et al. [24] who designed an investigation with 87 NAFLD and 87 controls. Moreover, they demonstrated that a lower serum level of Nrg4 was an independent risk factor for NAFLD. This is consistent with the findings of Tutunchi et al. [36] and confirms these findings that NAFLD patients had lower levels of Nrg4 than the control group. NAFLD prevalence was considerably lower in subjects in the highest quartile of serum Nrg4 levels in comparison to individuals in the lowest quartile.

The pathogenesis of NAFLD is strongly associated with IR and dyslipidemia. Although the exact mechanisms underlying NAFLD remain unclear, no effective and specific treatments are available for NAFLD [24]. BAT has been indicated to be active metabolically in adult humans where Nrg4 is highly enriched [38, 39]. Increased levels of Nrg4, also known as neuregulin 4, have been demonstrated to improve metabolic balance. Nrg4 regulates a variety of metabolic activities that contribute to overall metabolic balance. One important impact of increased Nrg4 is to promote fuel oxidation, which improves the consumption of glucose and fatty acids for energy production. This helps maintain energy balance and prevent the buildup of extra fuel sources in the body. Furthermore, Nrg4 has been linked to a healthier adipokine profile, which includes increased secretion of beneficial adipokines such as adiponectin and lower secretion of pro-inflammatory adipokines like TNF-alpha and IL-6. This alteration in adipokine secretion may increase insulin sensitivity and reduce inflammation. Nrg4 also regulates hepatic lipid metabolism by blocking lipogenesis and boosting lipid oxidation in the liver, hence preventing the accumulation of excess lipids and lowering the risk of fatty liver disease. Moreover, Nrg4 has been linked to angiogenesis, particularly in adipose tissue, which contributes to normal tissue function and nutrient supply. In addition, it has been shown to promote nerve growth and branching, which may have ramifications for metabolic function and control [40–45]. Nrg4 has a protective role against metabolic diseases like IR, diabetes, NAFLD, and obesity in animal models [46]. Human research has attempted to assess the potential association between serum Nrg4 levels and clinical and metabolic indices of NAFLD, metabolic syndromes, obesity, and diabetes after that the protective effect of Nrg4 metabolic homeostasis was discovered in animal models [47–49]. In this regard, patients with type 2 diabetes, obesity, metabolic syndrome, coronary artery diseases, and gestational diabetes had lower serum levels of Nrg4 in comparison to healthy controls [50–53]. The role of Nrg4 in metabolic illnesses is clear from these studies. A lack of Nrg4 may exacerbate the severity of NAFLD and other metabolic diseases. Monitoring Nrg4 levels in patients could be a good prognostic sign for illness progression. Measuring Nrg4 levels in humans allows for earlier detection of metabolic illnesses, allowing for timely intervention and prevention of disease consequences. The discovery of Nrg4 as a possible biomarker offers up new avenues for better diagnosis and therapy of metabolic disorders. Early identification of Nrg4 deficiency could lead to tailored therapies to increase Nrg4 levels and slow the course of certain disorders.

This analysis showed a high level of heterogeneity (98.50%), which may be attributed to variations in Nrg4 levels across the enrolled studies. To address this issue, authors conducted a subgroup analysis, which resulted in a slight reduction in heterogeneity (96.20%). The subgroup analysis was based on a cutoff value for Nrg4 levels, specifically set at 4 ng/dl. Recognizing the significance of Nrg4 and its possible role as a prognostic indicator allows healthcare practitioners to take preventive actions to address metabolic disorders.

It seems the Nrg4 cutoff value has a diagnostic value for predicting NAFLD. Dai et al. [24] reported patients with Nrg4 levels greater than 0.6800ng/mL had a considerably lower prevalence of NAFLD. Furthermore, based on ultrasound classification, no association was found between the level of serum Nrg4 and severity of NAFLD. Gado et al. [34] provide an Nrg4 level of 40.55 ng/ml for distinguishing controls from NAFLD patients. Wang et al. [37] claimed that the Nrg4 level was the best predictor for NAFLD patients and concluded that the Nrg4 cutoff value is 3.39 ng/ml. Consequently, an important issue is setting a standard cutoff value for Nrg4. Further work needs to be done to establish whether the Nrg4 level is related to the severity of NAFLD or not.

Meta-regression analysis was used to identify potential causes of heterogeneity as an alternative solution and to quantify the correlation between Nrg4 levels and the risk of NAFLD based on age, waist circumference, BMI, ALT, AST, LDL, HDL, TG, Cholesterol, and FBS. Although meta-regression was conducted to explore sources of heterogeneity, the findings were not statistically significant and should be considered hypothesis-generating. These exploratory results may guide future studies in identifying meaningful modifiers of effect estimates. Analysis shows that with an increase in age, waist circumference, and TG association of the Nrg4 index with NAFLD occurrence also increases. Further analysis indicates that an increase in ALT, and HDL strength of the relationship between the Nrg4 index and NAFLD occurrence decreases. There was no correlation between Nrg4 and NAFLD occurrence with increasing AST, FBS, cholesterol, and LDL. Wang et al. [37] mention that there is a negative correlation between Nrg4 and obesity indices, including BMI, waist-to-hip ratio, and waist circumference, indicating that Nrg4 may have a role in controlling lipid metabolism and body weight. Also, Nrg4 levels were negatively correlated with biomedical parameters such as fasting blood insulin, and serum TG but positively associated with serum HDL-C (p ≤ 0.05). Gado et al. [34] state that there is a significant association between serum Nrg4 and LDL-C levels (p = 0.030). Furthermore, there wasn’t any correlation between Nrg4 and age, BMI, waist circumference, ALT, AST, cholesterol, HDL-C, and serum TG levels. Previous investigations by Chen et al. and Kang et al. showed a positive association between serum Nrg4 levels and adiposity parameters and a negative association with serum HDL-C levels [47, 54]. Tutunchi et al. [36] reported that serum Nrg4 has a negative association with, waist circumference, BMI, and serum TG level. However, serum Nrg4 has a positive correlation with HDL-C. The discrepancy between the results may be due to the difference in methodology or variations in metabolic situations of the study patients of the studies, the results had some differences.

Impaired Nrg4 signaling leads to Nrg4 insufficiency. Nrg4 activates signaling pathways involving HER3 and HER4 in hepatocytes, which subsequently suppress lipogenesis through the regulation of liver X receptor (LXR) and sterol-regulatory element binding proteins 1c (SREBP-1c) [46]. Excessive liver fat is linked to metabolic risk factors like insulin resistance, waist circumference, BMI, WC, dyslipidemia, and hyperglycemia [55]. Insulin resistance is a vital factor in NAFLD development and other metabolic disease [56, 57]. In animal studies, deficiency of Nrg4 exacerbates insulin resistance and hepatic steatosis. Conversely, Nrg4 transgenic mice had improved insulin sensitivity and reduced serum TG [46, 58]. Furthermore, Nrg4 exerts a positive impact on healthy adipokine profile and energy balance in obesity [40]. Gado et al. [34] claim that reduced serum Nrg4 increases NAFLD risk independent of obesity indices and Nrg4 protection role is independent of obesity and insulin resistance. Altogether, it seems that serum Nrg4 has a protective role against NAFLD through pathways that are not directly related to these factors.

In this meta-analysis, 5 studies were initially selected, but only 4 studies were included in the final analysis. The study by De Munck et al. [35] was excluded due to methodological differences and the use of different measurements to calculate and report Nrg4. This discrepancy led to the reporting of a significantly different effect size or potentially exaggerated results in this study. Therefore, this study was not included in the analysis to ensure a more accurate estimation of the overall effect size.

This research has some limitations and strengths. First, the Nrg4 level reported in the Gado et al. [34] study was significantly different from others which can be explained by a difference in measurement methods. Second is insufficient article related to topic. Third, there was only one study from Europe, and the majority of the included studies were conducted in Asian nations, which may have limited the generalizability of these findings.

Clinical implication

This meta-analysis underscores the promise of Nrg4 as a non-invasive biomarker in the clinical evaluation of NAFLD. Owing to its inverse association with NAFLD risk, assessing Nrg4 levels may help in the early detection of individuals at increased risk, even before clinical symptoms or imaging reveal the condition. Incorporating Nrg4 into standard clinical procedures could serve as a valuable tool for assessing risk, enabling timely interventions such as lifestyle modifications or targeted therapies. In addition, tracking Nrg4 levels over time could offer an additional approach for assessing disease progression or treatment effectiveness, thereby supporting a more tailored and proactive strategy in NAFLD management.

Future directions

Even though the current meta-analysis supports the protective association between Nrg4 and NAFLD, further studies are needed to confirm these findings in larger, diverse populations. Future studies should focus on determining standardized cutoff levels for serum Nrg4, assessing its relationship with disease severity, and exploring its potential function in predicting treatment outcomes. Longitudinal cohort studies and randomized controlled trials are essential for understanding causality and evaluating whether therapeutic modulation of Nrg4 levels can change the trajectory of NAFLD progression. Additionally, mechanistic studies of the molecular pathways regulated by Nrg4 may reveal new therapeutic strategies for NAFLD and related metabolic disorders.

Conclusion

The findings of the meta-analysis suggest that Nrg4 has an independent protective effect on the development of NAFLD. However, the underlying mechanisms by which Nrg4 exerts its effects in NAFLD are not yet fully understood, and further studies are needed to elucidate these mechanisms. The measurement of Nrg4 levels holds promise in the early diagnosis and prognosis of NAFLD. By assessing Nrg4 levels, healthcare professionals may be able to identify individuals at higher risk of developing NAFLD at an early stage, enabling timely interventions and preventive measures. Furthermore, monitoring Nrg4 levels over time could aid in predicting disease progression and assessing treatment response in NAFLD patients.

Abbreviations

NAFLD

Non-alcoholic fatty liver disease

NASH

Nonalcoholic steatohepatitis

Nrg4

Neuregulin 4

BAT

Brown adipose tissue

WAT

White adipose tissue

Nrg1-4

Neuregulins

PRISMA

The Preferred Reporting Items for Systematic Reviews and Meta-Analysis

OR

Odds ratio

RR

Risk ratio

Cls

Confidence intervals

JBI

The Joanna Briggs Institute

FEM

Fixed-effect model

REM

Random-effects model

BMI

Body mass index

LXR

Liver X receptor

SREBP-1c

Sterol-regulatory element binding proteins 1c

Authors’ contributions

Y.M. contributed to the study design, while M.A.T. and S.R.G contributed to the study screening and data extraction. M.A.T. drafted the manuscript and Y.M. edited the language. Y.M. contributed to the complement and refinement of the manuscript. All the authors participated in the critical revision and approved the final version of the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

None.

Data availability

The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval was not required for this study, as it is a meta-analysis and did not involve the collection of primary data.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.