Home
JournalsCollections
For Authors For Reviewers For Editorial Board Members
Article Processing Charges Open Access
Ethics Advertising Policy
Editorial Policy Resource Center
Company Information Contact Us
OPEN ACCESS

Systematic Review and Meta-analysis of Circulating Fetuin-A Levels in Nonalcoholic Fatty Liver Disease

  • Shousheng Liu#,1,2 ,
  • Jianhan Xiao#,3 ,
  • Zhenzhen Zhao1,2 ,
  • Mengke Wang3 ,
  • Yifen Wang3  and
  • Yongning Xin*,2,3 
Journal of Clinical and Translational Hepatology   2021;9(1):3-14

doi: 10.14218/JCTH.2020.00081

Received:

Revised:

Accepted:

Published online:

 Author information

Citation: Liu S, Xiao J, Zhao Z, Wang M, Wang Y, Xin Y. Systematic Review and Meta-analysis of Circulating Fetuin-A Levels in Nonalcoholic Fatty Liver Disease. J Clin Transl Hepatol. 2021;9(1):3-14. doi: 10.14218/JCTH.2020.00081.

Abstract

Background and Aims

Accumulated studies have reported the key role of circulating fetuin-A in the development and progression of nonalcoholic fatty liver disease (NAFLD) but the results have not been consistent. In this study, we performed a systematic review and meta-analysis to explore the relationship between circulating fetuin-A level and the development and classification of NAFLD.

Methods

The PubMed, EMBASE, and Cochrane Library databases were searched to obtain the potentially relevant studies up to May 2020. Standardized mean differences (SMD) and 95% confidence intervals of circulating fetuin-A levels were extracted and summarized. Sensitivity, subgroup analysis and meta-regression analysis were performed to investigate the potential heterogeneity. Association of circulating fetuin-A level with classification of NAFLD was also reviewed.

Results

A total of 17 studies were included, composed of 1,755 NAFLD patients and 2,010 healthy controls. Meta-analysis results showed that NAFLD patients had higher circulating fetuin-A level (SMD=0.43, 95% confidence interval [CI]: 0.22–0.63, p<0.001) than controls. Subgroup analysis indicated that circulating fetuin-A level was markedly increased in adult NAFLD patients (SMD=0.48, 95% CI: 0.24–0.72, p<0.001) and not in pediatric/adolescent patients compared to controls. Circulating fetuin-A level was markedly increased in ultrasound-proven NAFLD pediatric/adolescent patients (SMD=0.42, 95% CI: 0.12–0.72, p=0.007), other than in the liver biopsy-proven NAFLD pediatric/adolescent patients. Body mass index might be the influencing factor to the heterogeneity in adult patients. Circulating fetuin-A level was not associated with the classification of NAFL vs. nonalcoholic steatohepatitis (NASH). Whether the circulating fetuin-A level was associated with the development of fibrosis remains controversial.

Conclusions

Circulating fetuin-A level was significantly higher in NAFLD patients and was not associated with the classification of NAFL vs. NASH. Whether the circulating fetuin-A level was associated with the development of fibrosis remains controversial.

Keywords

Nonalcoholic fatty liver disease, Fetuin-A, Meta-analysis, Fibrosis

Introduction

Nonalcoholic fatty liver disease (NAFLD) has become one of the most common chronic liver diseases in recent years, and the overall prevalence of NAFLD is approximately 25% in the world.1,2 There is a broad spectrum of NAFLD, which ranges from nonalcoholic fatty liver (NAFL) to nonalcoholic steatohepatitis (NASH), fibrosis, cirrhosis, and even hepatocellular carcinoma (HCC).3,4 NAFLD is the hepatic manifestation of metabolic syndrome and is affected by many risk factors, such as obesity, hyperglycemia, type 2 diabetes, and hypertriglyceridemia usually.5 However, under certain conditions, mostly during genetically-determined NAFLD (such as in carriers of the TM6SF2 E167K or PNPLA3 I148M gene polymorphism), NAFLD does not show association with the metabolic syndrome and an increased risk of cardiovascular disease.6

Liver biopsy remains the gold standard for diagnosis and histological assessment of NAFLD, but the obvious defects (e.g., invasiveness, inter-observer differences, sampling error) cannot be ignored.7,8 In clinical practice, imaging methods such as ultrasonography, computed tomography, controlled attenuation parameter and magnetic resonance have been used widely for diagnosing NAFLD.9–13 In addition, many studies have been conducted to explore the valuable serum biomarkers for early diagnosis and progression of NAFLD. Several serum biomarkers, such as alanine aminotransferase, aspartate aminotransferase, gamma-glutamyltranspeptidase, cytokeratin-18 and fibroblast growth factor 21, have been researched in some studies and their potential to serve as the biomarkers in clinical diagnosing of NAFLD have been mentioned.14–18

Fetuin-A, also known as the 2-Heremans-Schmid glycoprotein, is a phosphorylated glycoprotein and a member of the fetuin group of serum binding proteins that are synthesized primarily by hepatocytes.19 As an endogenous inhibitor of tyrosine kinase, fetuin-A can trigger insulin resistance in the target tissues, such as liver and skeletal muscle.20,21 Pal et al.22 reported that fetuin-A acts as an endogenous ligand for toll-like receptor 4 and could enhance both insulin resistance and inflammation response. High serum fetuin-A was also found to strongly interact with the high levels of free fatty acids to induce insulin resistance in rodents, which was then observed in large human studies; moreover, the relationships of fetuin-A with fatty acids, to determine insulin resistance, were particularly strong in patients with NAFLD.23,24 Fetuin-A is also known to inhibit transforming growth factor-b1 signaling, which promotes fibrotic changes in many tissues, including liver and arteries; therefore, fetuin-A could prevent fibrotic changes in organs.25,26 Recently, fetuin-A has been regarded as a potential link molecule between obesity, insulin resistance, and coronary heart disease,27–31 and was found to be strongly related to several parameters such as circulating lipid levels (non-esterified free fatty acids and triglycerides), glucose tolerance, circulating pro-inflammatory and anti-inflammatory factors, and interleukin-6.32 Fetuin-A can also induce low-grade inflammation and repress adiponectin production in animals and humans, playing an important role in the pathogenesis of insulin and possessing a pro-adiposity effect.33,34 Accumulated lines of evidence have reported the significant association between circulating fetuin-A level and the development and progression of NAFLD, but the results have been inconsistent. Additionally, there has been no definite conclusion as to whether circulating fetuin-A can reflect the grading of NAFL vs. NASH and advanced fibrosis.

The aim of this study was to investigate the importance of fetuin-A in the development and classification of NAFLD.

Methods

Search strategy

This systematic review and meta-analysis was conducted following a priori established protocol and was reported according to PRISMA guidelines.35 Two independent observers (Shousheng Liu and Jianhan Xiao) performed a systematic search of the PubMed, EMBASE, and the Cochrane Library databases up to May 2020 and with English language restrictions. The first step for information retrieval was to gain the subject term of fetuin-A, NAFLD or NASH in the MeSH database of PubMed; meanwhile, we gained the entry terms of them, respectively. The combined results of fetuin-A, NAFLD or NASH in the three databases were obtained based on the search method of “subject term + entry terms”. In addition, we examined the reference lists in relevant original research and review articles to search additional potentially eligible studies.

Inclusion and exclusion criteria

Studies that investigated the circulating fetuin-A level in patients with NAFLD were eligible for review. Studies were included in this systematic review and meta-analysis if they met the following criteria: (1) original full-text publications; (2) comparison of circulating fetuin-A level between NAFLD patients and healthy controls; (3) investigations of the effect of circulating fetuin-A level on the classification of NAFL vs. NASH or fibrosis. Studies were excluded according to the following criteria: (1) patients with other causes of liver disease (e.g., viral or autoimmune hepatitis, alcoholic fatty liver disease, HCC, coronary artery disease) or in whom NAFLD co-existed with another liver disease; (2) there was overlap of patients who were included in more than one study. Quality of case-control studies were evaluated use the Newcastle-Ottawa scale (NOS) scoring system and the quality of cross-sectional studies were evaluated use the Agency for Healthcare Research and Quality (ARHQ) scoring system.36 (3) Studies of low methodological quality, as defined by a NOS score ≤ 2 or ARHQ score ≤ 3, were excluded. Finally, (4) reviews, editorials, case reports, conference abstracts, letters to the editor, hypotheses, book chapters, and studies on animals or cell lines were excluded.

Data extraction

Available data were extracted from the full text and corresponding supplemental information by two investigators working independently (Shousheng Liu and Jianhan Xiao) and confirmed by a third reviewer (Zhenzhen Zhao). Disagreement was resolved by discussion among all researchers. The following information of each selected publication was extracted: (1) general characteristics, such as first author’s name, year of publication, country where the study was carried out, study design, diagnostic methods of NAFLD, type of samples (e.g., serum, plasma, blood); (2) subjects’ characteristics, such as age, gender, body mass index (BMI); and (3) effect of circulating fetuin-A on the grading of NAFL vs. NASH or fibrosis. When the same population was published in several journals, we retained only the most informative article or complete study, to avoid duplication. If some necessary data were not offered in the article, the corresponding author would be contacted for the data. If no response, the following methods would be carried out: (1) data from the graphical plots were extracted to calculate the circulating fetuin-A levels by using WebPlotDigitizer (version 4.1.0, https://apps.automeris.io/wpd/ );37 (2) circulating fetuin-A levels which were expressed as median (mix-mix) or median (25–75 quartile) were transformed into the standard form of mean, according to the Cochrane book or method (Hozo, Stela Pudar et al.38).

Quality assessment

The included studies in the systematic review and meta-analysis were independently assessed by two investigators (Shousheng Liu and Jianhan Xiao). We assessed the quality of included case-control studies based on the NOS scoring system, and cross-sectional studies based on the AHRQ scoring system. The full NOS score was 9 stars; a study that met 7 or more stars was defined as a high-quality study, less than 3 stars as low-quality and other studies were defined as moderate quality. Article quality by AHRQ was assessed as follows: low quality: 0–3; moderate quality: 4–7; and high quality: 8–11.

Statistical analysis

All statistical analyses were performed using Stata12.0 (StataCorp LP, College Station, TX, USA). The effect sizes were generated by sample sizes, mean circulating fetuin-A levels, and the standard deviation (SD), and presented as standardized mean differences (SMD) and 95% confidence intervals (95% CIs) for circulating fetuin-A levels in comparisons between groups. Given the expected heterogeneity of the outcome, a random-effect inverse-variance model was chosen for this meta-analysis.39 The heterogeneity between the results of different studies was evaluated using the I2 statistic, values of I2 >50% were considered to represent substantial heterogeneity. The potential moderating effects of continuous variables on between-study heterogeneity were evaluated by meta-regression analyses and subgroup-analysis. Subgroup-analysis were first conducted according to age, region, and diagnostic method of NAFLD, then the sex distribution (number of males), mean age, BMI and HOMA-IR of NAFLD patients were regarded as the potential moderators for the adult outcome of the meta-analysis when a high heterogeneity of adult NAFLD was observed. Sensitivity analysis was performed to investigate the influence of each study on the pooled measures by omitting a study each time to assess the stability of our results. A p-value of <0.05 was considered to indicate a statistical difference. Publication bias was assessed by funnel plot and Begg’s and Egger’s tests.

Results

Literature search

A total of 318 studies were retrieved initially from the three databases. After removing duplications (n=97), 221 studies remained for evaluation. In all, 204 studies were excluded for representing reviews, editorials, letters, book chapters or case reports, animal or cell experiments, other liver diseases, patient overlap, conference abstracts, and so on (Fig. 1). The final dataset for the systematic review and meta-analysis comprised 17 full-text studies.25,27,28,40–53 Among them, 16 studies were selected to conduct the meta-analysis and 8 studies were selected to investigate the relationship between circulating fetuin-A level and the classification of NAFL vs. NASH.

Flowchart presenting the literature search process, according to the PRISMA statement.
Fig. 1  Flowchart presenting the literature search process, according to the PRISMA statement.

Characteristics of included studies

A total of 1,755 NAFLD patients and 2,010 healthy controls were included in the 17 studies, the main characteristics of these studies are shown in Table 1. Among these studies, two were conducted with the same participants, but the former investigated the association of plasma fetuin-A level with NAFLD;44 however, the latter study not only investigated the relationship between plasma fetuin-A level and NASH or NAFL but also investigated the association of plasma fetuin-A level and fibrosis.47 As such, 16 studies were selected to perform the meta-analysis.

Table 1

Characteristics of included studies focused on circulating levels of fetuin-A in this meta-analysis and systematic review

Study (Year)CountryRegionStudy designAgeSample sizeTissue typeFetuin-A measurement methodNAFLD diagnosis methodNOS, 0–9 scoreAHRQ, 0–11 scoreAdditional information
Reinehr et al. (2008)GermanyEuropeanCross-sectionalPediatric/Adolescent48SerumELISAUltrasound6
Yilmaz et al. (2010)TurkeyEurasianCase-controlAdult174SerumELISALiver biopsy7
Haukeland et al. (2012)NorwayEuropeanCross-sectionalAdult242PlasmaELISALiver biopsy4
Ou et al. (2012)ChineseAsianCross-sectionalAdult510SerumELISAUltrasound6(1)
Ballestri et al. (2013)ItalyEuropeanCross-sectionalAdult70SerumELISAUltrasound5(2)
Dogru et al. (2013)TurkeyEurasianCross-sectionalAdult189PlasmaELISALiver biopsy5(3)
Kahraman et al. (2013)GermanyEuropeanCross-sectionalAdult118SerumELISALiver biopsy4
Lebensztejn et al. (2014)PolandEuropeanCross-sectionalPaediatric/Adolescent45SerumELISAUltrasound5
Rametta et al. (2014)ItalyEuropeanCross-sectionalAdult397SerumELISALiver biopsy7
Sato et al. (2015)JapaneseAsianCross-sectionalAdult295SerumELISAUltrasound5
Wong et al. (2015)ChineseAsianCase-controlAdult920SerumELISAUltrasound5
Celebi et al. (2015)TurkeyEurasianCross-sectionalAdult/PlasmaELISALiver biopsy5(3)
Cui et al. (2017)ChineseAsianCase-controlAdult158SerumELISAUltrasound5
Siraz et al. (2017)TurkeyEurasianCross-sectionalPaediatric/Adolescent80SerumELISAUltrasound6
Pampanini et al. (2018)ItalyEuropeanCross-sectionalPaediatric/Adolescent183SerumELISAUltrasound6(1)
Mondal et al. (2018)IndiaAsianCross-sectionalAdult188SerumELISAUltrasound6(2)
Nascimbeni et al. (2018)ItalyEuropeanCross-sectionalAdult149SerumELISAUltrasound Liver biopsy6(4)

Studies in this meta-analysis included three case-control and thirteen cross-sectional designs. Enzyme-linked immunosorbent assay was used to test the serum/plasma fetuin-A level in all the studies. Liver biopsy was performed to determine the NAFLD in six studies, and ultrasound was used in other ten. Another study, conducted by Pampanini et al.,27 diagnosed NAFLD with both liver biopsy and ultrasound in different groups; so, we regarded this study as two individual studies. Among these studies, six performed the comparison of circulating fetuin-A level between NAFL and NASH patients and eight determined the relationship of circulating fetuin-A level with liver fibrosis (Table 2).

Table 2

Relationship of circulating fetuin-A levels with the progression of NAFLD

StudiesSample sizesNASH vs. NAFL
Fibrosis vs. No fibrosis
CorrelationDiagnosis methodDataCorrelationDiagnosis methodDataLogistic regression analysis
Yilmaz et al. (2010)99NsLiver biopsyNot shownPositive correlationLiver biopsyGraphic resultsConsistent
Haukeland et al. (2012)111NsLiver biopsyGraphic resultsNsLiver biopsyNot shown/
Kahraman et al. (2013)109NsLiver biopsyDetailed dataNegative correlationLiver biopsyNot shown/
Rametta et al. (2014)137NsLiver biopsyGraphic resultsNsLiver biopsyNot shown/
Celebi et al. (2015)105NsLiver biopsyDetailed dataNsLiver biopsyDetailed data/
Sato et al. (2015)275///Negative correlationserologyDetailed dataConsistent
Pampanini et al. (2018)81NsLiver biopsyDetailed dataNsLiver biopsyGraphic results
Mondal et al. (2018)46///Positive correlationFibroScanDetailed dataConsistent

Quality of included studies

The qualities of included case-control or cohort studies were assessed based on the NOS, and the cross-sectional studies were assessed based on the ARHQ methodology checklist.36 The detailed quality scores of each study are shown in Table 1. All the studies were assessed as moderate quality. No study was eliminated due to low quality (NOS score ≤2 or AHRQ score ≤3).

Effect of circulating fetuin-A level on NAFLD

A random-effect meta-analysis was performed to investigate the effect of circulating fetuin-A level on the development of NAFLD. As the results show in Fig. 2A, the circulating fetuin-A level in NAFLD patients was significantly higher than in healthy controls, with a summarized SMD of 0.43 (95%CI: 0.22–0.63, p<0.001). A striking heterogeneity among included studies was observed in the comparison of circulating fetuin-A level in NAFLD patients and healthy controls; the I2 value was 85.7% (p<0.001).

Meta-analysis and influence of age.
Fig. 2  Meta-analysis and influence of age.

(A) Meta-analysis of the circulating fetuin-A levels in the included NAFLD patients compared to healthy controls. (B) Subgroup analysis on the difference of circulating fetuin-A levels between the included NAFLD patients and healthy controls based on age.

Subgroup analysis based on age implied that the circulating fetuin-A level of NAFLD patients was significantly elevated in adults (SMD=0.48, 95% CI: 0.24–0.72, p<0.001) and no obvious difference was observed in pediatric/adolescent patients (SMD=0.25, 95% CI: −0.18–0.67, p=0.256) (Fig. 2B). Based on age, we also performed subgroup analysis according to the region of subjects and diagnostic method of NAFLD, respectively. As shown in Fig. 3A, the circulating fetuin-A level in adult NAFLD patients was increased among Europeans (SMD=0.71, 95% CI: 0.35–1.07, p<0.001), and no significant differences were observed in the Eurasians (SMD=0.73, 95% CI: −0.04–1.50, p=0.062) nor Asians (SMD=0.18, 95% CI: −0.08–0.43, p=0.174). In the pediatric/adolescent group, there was no significant difference of circulating fetuin-A level between NAFLD patients and controls of European populations (SMD=0.27, 95% CI: −0.25–0.79, p=0.303) and Eurasian populations (SMD=0.15, 95% CI: −0.58–0.88, p=0.688). Subgroup analysis results according to NAFLD diagnosis method (ultrasound vs. liver biopsy) are shown in Fig. 3B. In adults, the level of circulating fetuin-A was higher in both ultrasound-proven NAFLD patients and liver biopsy-proven NAFLD patients than in healthy controls (SMD=0.21, 95% CI: 0.01–0.42, p<0.001; SMD=0.86, 95% CI: 0.51–1.21, p<0.001, respectively). Interestingly, the circulating fetuin-A level in ultrasound-proven NAFLD pediatric/adolescent patients was significantly increased compared to pediatric/adolescent controls (SMD=0.42, 95% CI: 0.12–0.72, p=0.007), but no difference was observed between the liver biopsy-proven NAFLD pediatric/adolescent patients and healthy controls (SMD=−0.37, 95% CI: −0.84–0.09, p=0.116). In addition, heterogeneity in the pediatric/adolescent patients with ultrasound diagnosis was markedly lower (I2=7.4%, p=0.356) than the overall heterogeneity in the remaining pediatric/adolescent patients (I2=64.4%, p=0.024) (Figs. 2B and 3B).

Influence of region and diagnostic method.
Fig. 3  Influence of region and diagnostic method.

(A) Subgroup analysis on the difference of circulating fetuin-A levels between the included NAFLD patients and healthy controls based on region. (B) Subgroup analysis on the difference of circulating fetuin-A levels between the included NAFLD patients and healthy controls based on diagnostic method.

To further investigate the cause of heterogeneity in adult NAFLD patients, we performed univariate, random-effects meta-regression analysis to test whether the continuous variables, including sex distribution (percentage of males), mean age, BMI and HOMA-IR of NAFLD patients, could explain the high heterogeneity among studies. As the results show in Table 3 and Fig. 4, BMI was the significant influencing factor of the meta-analysis (R2=41.60, β=1.058, p=0.023), and the other tested variables did not show moderating effects.

Table 3

Demographic and clinical data of patients with NAFLD and healthy controls among the adults

StudiesGroupSize, nMales, %Age in years, mean (SD)BMI in kg/m2, mean (SD)HOMA-IR
Yilmaz et al. (2010)NAFLD
Control
99
75
46.0
51.0
47.0 (9.0)
47.0 (8.0)
30.7 (4.9)
27.5 (4.3)
3.80 (0.40)
1.40 (0.30)
Haukeland et al. (2012)NAFLD
Control
111
131
60.0
44.0
46.5 (11.6)
43.3 (3.0)
30.5 (4.3)
23.9 (3.0)
2.21 (1.14)
1.40 (0.77)
Ou et al. (2012)NAFLD
Control
255
255
56.0
60.0
61.1 (10.3)
62.1 (11.3)
26.7 (3.1)
23.3 (2.8)
1.21 (0.12)
0.58 (0.16)
Ballestri et al. (2013)NAFLD
Control
29
41
69.0
68.3
64.5 (10.5)
70.6 (12.7)
29.2 (5.0)
25.8 (3.1)
1.50 (0.325)
1.40 (0.350)
Dogru et al. (2013)NAFLD
Control
115
74
100.0
100.0
31.0 (5.2)
28.0 (5.2)
28.4 (2.97)
24.0 (2.65)
3.35 (2.18)
1.22 (0.62)
Kahraman et al. (2013)NAFLD
Control
108
10
23.0
50.0
41.9 (0.9)
32.5 (5.5)
53.3 (1.1)
23.9 (1.2)
Na
Na
Rametta et al. (2014)NAFLD
Control
137
260
77.4
80.0
49.7 (12.1)
47.7 (12.1)
26.9 (3.4)
25.1 (2.8)
2.50 (2.80)
1.30 (0.20)
Sato et al. (2015)NAFLD
Control
275
20
55.0
65.0
56.4 (6.9)
61.0 (7.0)
26.5 (3.6)
22.2 (2.6)
Na
Na
Wong et al. (2015)NAFLD
Control
263
657
54.0
37.4
51.0 (9.0)
47.0 (11.0)
25.3 (4.0)
21.3 (3.1)
2.50 (0.37)
1.10 (0.15)
Cui, Xuan, and Yang (2017)NAFLD
Control
79
79
73.0
73.0
42.0 (10.8)
40.0 (12.0)
26.0 (3.0)
22.0 (2.0)
3.27 (2.18)
1.81 (1.80)
Mondal et al. (2018)NAFLD
Control
46
142
57.0
66.0
49.5 (12.2)
46.2 (12.7)
27.5 (6.2)
25.7 (4.8)
1.10 (0.26)
1.10 (0.10)
Nascimbeni et al. (2018)NAFLD
Control
80
69
79.0
77.0
70.0 (7.0)
73.0 (8.2)
28.0 (3.8)
25.0 (2.5)
1.80 (3.05)
1.40 (1.87)
R2 (%)15.95−10.9041.60−13.12
β0.2970.9951.0580.041
p value0.0970.6930.0230.812
Meta-regression analysis for the effect of BMI on the NAFLD patients and healthy controls.
Fig. 4  Meta-regression analysis for the effect of BMI on the NAFLD patients and healthy controls.

Sensitivity and publication bias analyses

A leave-one-out sensitivity analysis was conducted to evaluate the stability of this meta-analysis (Fig. 5A). Each study included in our meta-analysis was evaluated one-by-one, to reflect the effect of pooled SMDs. The overall statistical significance did not change when any single study was omitted at one time. Therefore, the data presented in this meta-analysis is relatively stable and credible. Publication bias in this meta-analysis was confirmed by Egger’s test, the results showed no significant publication bias (p=0.152). In addition, no significant publication biases were observed in the adult population Egger’s test (p=0.275) and in the pediatric/adolescent population Egger’s test (p=0.450). Funnel plots of effect size vs. standard error were symmetrical (p>0.05) (Fig. 5B).

Sensitivity analysis and forest plotting.
Fig. 5  Sensitivity analysis and forest plotting.

(A) Sensitivity analysis of the summary odds ratio coefficients on the difference of circulating fetuin-A levels between NAFLD patients and healthy controls. (B) Forest plots on the difference of circulating fetuin-A levels between NAFLD patients and healthy controls.

Effect of fetuin-A levels on grading of NAFLD

In order to investigate whether circulating fetuin-A level can be used as a potential diagnostic biomarker for the classification of NAFLD, all the available information of circulating fetuin-A level on the classification of NAFLD were collected. As shown in Table 2, there were no significant differences of circulating fetuin-A level in the classification of NAFL vs. NASH in liver biopsy-proven NAFLD patients. In the liver biopsy-proven fibrosis patients, results from four studies suggested there was not association between circulating fetuin-A level and the development of fibrosis, two studies showed that circulating fetuin-A levels were negatively correlated with fibrosis, and one study showed a positive correlation. In addition, one study found that circulating fetuin-A level was negatively correlated with serology-proven fibrosis, and another one study showed positive correlation in FibroScan-proven fibrosis (Table 2).

Discussion

NAFLD has become one of the most prevalent chronic liver diseases in the world and the major cause of liver-related morbidity and mortality.1 Up to now, liver biopsy remains the recognized gold standard for the diagnosis of NAFLD. In consideration of the defects of biopsy, several imaging and serological diagnosis methods have been developed. Fetuin-A is a member of the fetuin group of serum binding proteins that are primarily synthesized by the hepatocytes.54 Some studies have investigated the circulating fetuin-A level between NAFLD patients and healthy controls in adult or pediatric/adolescent populations, but the results have been inconsistent. In this study, we conducted a meta-analysis to summarize the circulating fetuin-A levels in NAFLD patients and healthy controls, and a systematic review to determine the correlation of circulating fetuin-A level with the classification of NAFLD. As the results show, the circulating levels of fetuin-A in patients with NAFLD were significantly higher than in healthy controls for the included subjects. There was no difference of circulating fetuin-A level between NAFL and NASH patients. In addition, the relationship of circulating fetuin-A level with fibrosis remains unclear. Sensitivity analysis suggested that our meta-analysis was stable, and no significant publication bias was observed.

In this meta-analysis, the included studies were carried out from a different region in adult and pediatric/adolescent patients, and the diagnostic methods of NAFLD were also different. A marked heterogeneity was observed (I2=85.7, p<0.001) when the circulating fetuin-A levels were analyzed for the patients with NAFLD and healthy controls. In order to explore the potential factors which contribute to the heterogeneity, subgroup analysis based on the different variables was conducted. In the adults, circulating fetuin-A level was significantly increased in patients with NAFLD compared to healthy controls, with a high heterogeneity (I2=88.8, p<0.001). Subsequently, the region of subjects and diagnostic methods of NAFLD were considered for analysis of the origin of heterogeneity. In the adults, circulating fetuin-A levels in Europeans were increased in patients with NAFLD but not in Eurasians and Asians, which suggested that region may be one of the influencing factors that contributes to the heterogeneity.

The circulating level of fetuin-A in patients with NAFLD did not vary according to the different diagnostic methods. Furthermore, meta-regression analysis suggested that BMI was the significant influencing factor to the heterogeneity. The reason why BMI contributes to the heterogeneity may be due to the incomplete correlation between BMI and the risk of NAFLD. Usually, subjects with metabolically healthy obesity, which is predominantly characterized by low liver fat content, may possess low risk of developing NAFLD, and the body fat distribution (more strongly than BMI) determines NAFLD in the general population; even patients with newly developed lipodystrophy can strongly develop NASH.55–57

The data of circulating fetuin-A level in NAFLD in pediatric/adolescent patients were relatively insufficient in that only four studies were included in this meta-analysis. We analyzed the pediatric/adolescent data using five individual studies, due to the study which was conducted by Pampanini et al.27 that recruited two independent cohorts in which NAFLD was diagnosed with ultrasound and liver biopsy, respectively. There was no significance difference of circulating fetuin-A levels between patients with NAFLD and healthy controls. It is noteworthy that the high circulating fetuin-A level was observed in the ultrasound-diagnostic NAFLD in pediatric/adolescent patients. Notably, the heterogeneity of this subgroup analysis was very low (I2=7.4%, p<0.001). The circulating fetuin-A levels in biopsy-proven NAFLD and healthy controls in pediatric/adolescent patients was tested for the first time and no obvious difference of circulating fetuin-A levels was found between the two groups. The different fetuin-A levels in ultrasound-diagnostic NAFLD compared to biopsy-proven NAFLD in pediatric/adolescent patients may due to the difference of diagnostic results of NAFLD. In other words, some ultrasound-diagnostic NAFLD may belong to healthy controls when diagnosed by biopsy. In order to illuminate this query, more studies should be conducted to investigate the circulating fetuin-A level in liver biopsy-proven NAFLD in pediatric/adolescent patients.

Fetuin-A possesses a pro-inflammatory role and is down-regulated during inflammation; expression of fetuin-A is also inversely correlated with the level of C-reactive protein, which is a well-known marker of systemic inflammation.33,47 Sato et al.25 reported that fetuin-A might promote insulin resistance and inhibit NAFLD progression, but whether circulating fetuin-A level is positively or negatively correlated with the classification of NAFLD remains controversial. Our results have suggested that circulating fetuin-A level is not related with the classification of NAFL vs. NASH, as no significant change of fetuin-A level was found between the two groups. In consideration of the previous controversial results, more studies should be conducted to investigate the relationship of fetuin-A with the markers of inflammation. Besides, there was not a definite conclusion as to whether circulating fetuin-A level was associated with fibrosis due to the positive, negative, and unrelated correlations that have been reported. In order to clarify the significant role of circulating fetuin-A level on the diagnosis of fibrosis, more attention should be paid to the relationship of circulating fetuin-A level with fibrosis in different regions and ethnicity groups.

Our study had several limitations. First, considerable heterogeneity among studies limits the reliability of the results. Although we performed subgroup and meta-regression analyses to investigate some potential sources of heterogeneity, the high levels of heterogeneity cannot be reasonably explained in adult NAFLD. Second, due to the lack of detailed data on fetuin-A in the classification of NAFLD in some studies, we just reviewed the change of fetuin-A rather than making a quantitative analysis. Finally, all these studies are the cross-sectional or case-control epidemiological design, and the dynamic changes of circulating fetuin-A level in NAFL or NASH patients were unclear.

Conclusions

In summary, the circulating fetuin-A level was significantly higher in NAFLD patients than in healthy controls in adults and no difference was observed in pediatric/adolescent patients. BMI might be the risk factor that affects the stability of meta-analysis in adults. In pediatric/adolescent patients, ultrasound-proven NAFLD patients possess a markedly higher circulating fetuin-A level than healthy controls and there were no differences in biopsy-proven NAFLD patients. Our results suggest that circulating fetuin-A level could be regarded as a potential serum biomarker for the early diagnosis of NAFLD. Diagnostic values and differences of ultrasound and biopsy in pediatric/adolescent should be further studied to illuminate the significant effect of circulating fetuin-A level on pediatric/adolescent NAFLD. In addition, association of circulating fetuin-A level with fibrosis should be studied further.

Abbreviations

ARHQ: 

Agency for Healthcare Research and Quality

BMI: 

body mass index

CI: 

confidence interval

HCC: 

hepatocellular carcinoma

NAFL: 

nonalcoholic fatty liver

NAFLD: 

nonalcoholic fatty liver disease

NASH: 

nonalcoholic steatohepatitis

NOS: 

Newcastle-Ottawa scale

SD: 

standard deviation

SMD: 

standardized mean differences

Declarations

Funding

This study was supported by grants from the National Natural Science Foundation of China (No. 31770837).

Conflict of interest

The authors have no conflict of interests related to this publication.

Authors’ contributions

Study concept and design (SL, YX), data collection (SL, JX, ZZ, MW, and YW), analysis of data (SL, JX, and ZZ), drafting and writing of the manuscript (SL and JX), revision of the manuscript (YX).

References

  1. Friedman SL, Neuschwander-Tetri BA, Rinella M, Sanyal AJ. Mechanisms of NAFLD development and therapeutic strategies. Nat Med 2018;24:908-922 View Article PubMed/NCBI
  2. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology 2016;64:73-84 View Article PubMed/NCBI
  3. Starley BQ, Calcagno CJ, Harrison SA. Nonalcoholic fatty liver disease and hepatocellular carcinoma: a weighty connection. Hepatology 2010;51:1820-1832 View Article PubMed/NCBI
  4. Hashimoto E, Taniai M, Tokushige K. Characteristics and diagnosis of NAFLD/NASH. J Gastroenterol Hepatol 2013;28(Suppl 4):64-70 View Article PubMed/NCBI
  5. Kim CH, Younossi ZM. Nonalcoholic fatty liver disease: a manifestation of the metabolic syndrome. Cleve Clin J Med 2008;75:721-728 View Article PubMed/NCBI
  6. Stefan N, Häring HU, Cusi K. Non-alcoholic fatty liver disease: causes, diagnosis, cardiometabolic consequences, and treatment strategies. Lancet Diabetes Endocrinol 2019;7:313-324 View Article PubMed/NCBI
  7. Krishan S, Jain D, Bathina Y, Kale A, Saraf N, Saigal S, et al. Non-invasive quantification of hepatic steatosis in living, related liver donors using dual-echo Dixon imaging and single-voxel proton spectroscopy. Clin Radiol 2016;71:58-63 View Article PubMed/NCBI
  8. Zhang Q, Zhang HM, Qi WQ, Zhang YG, Zhao P, Jiao J, et al. 3.0T 1H magnetic resonance spectroscopy for assessment of steatosis in patients with chronic hepatitis C. World J Gastroenterol 2015;21:6736-6744 View Article PubMed/NCBI
  9. Deng J, Fishbein MH, Rigsby CK, Zhang G, Schoeneman SE, Donaldson JS. Quantitative MRI for hepatic fat fraction and T2* measurement in pediatric patients with non-alcoholic fatty liver disease. Pediatr Radiol 2014;44:1379-1387 View Article PubMed/NCBI
  10. Keese D, Korkusuz H, Huebner F, Namgaladze D, Raschidi B, Vogl TJ. In vivo and ex vivo measurements: noninvasive assessment of alcoholic fatty liver using 1H-MR spectroscopy. Diagn Interv Radiol 2016;22:13-21 View Article PubMed/NCBI
  11. de Lédinghen V, Vergniol J, Capdepont M, Chermak F, Hiriart JB, Cassinotto C, et al. Controlled attenuation parameter (CAP) for the diagnosis of steatosis: a prospective study of 5323 examinations. J Hepatol 2014;60:1026-1031 View Article PubMed/NCBI
  12. Runge JH, Bakker PJ, Gaemers IC, Verheij J, Hakvoort TB, Ottenhoff R, et al. Measuring liver triglyceride content in mice: non-invasive magnetic resonance methods as an alternative to histopathology. MAGMA 2014;27:317-327 View Article PubMed/NCBI
  13. Chabanova E, Bille DS, Thisted E, Holm JC, Thomsen HS. 1H MRS assessment of hepatic steatosis in overweight children and adolescents: comparison between 3T and open 1T MR-systems. Abdom Imaging 2013;38:315-319 View Article PubMed/NCBI
  14. Yki-Järvinen H. Diagnosis of non-alcoholic fatty liver disease (NAFLD). Diabetologia 2016;59:1104-1111 View Article PubMed/NCBI
  15. Katsagoni CN, Georgoulis M, Papatheodoridis GV, Panagiotakos DB, Kontogianni MD. Effects of lifestyle interventions on clinical characteristics of patients with non-alcoholic fatty liver disease: A meta-analysis. Metabolism 2017;68:119-132 View Article PubMed/NCBI
  16. Franzini M, Fornaciari I, Fierabracci V, Elawadi HA, Bolognesi V, Maltinti S, et al. Accuracy of b-GGT fraction for the diagnosis of non-alcoholic fatty liver disease. Liver Int 2012;32:629-634 View Article PubMed/NCBI
  17. Kosasih S, Zhi Qin W, Abdul Rani R, Abd Hamid N, Chai Soon N, Azhar Shah S, et al. Relationship between serum cytokeratin-18, control attenuation parameter, NAFLD fibrosis score, and liver steatosis in nonalcoholic fatty liver disease. Int J Hepatol 2018;2018:9252536 View Article PubMed/NCBI
  18. He L, Deng L, Zhang Q, Guo J, Zhou J, Song W, et al. Diagnostic value of CK-18, FGF-21, and related biomarker panel in nonalcoholic fatty liver disease: A systematic review and meta-analysis. Biomed Res Int 2017;2017:9729107 View Article PubMed/NCBI
  19. Jirak P, Stechemesser L, Moré E, Franzen M, Topf A, Mirna M, et al. Clinical implications of fetuin-A. Adv Clin Chem 2019;89:79-130 View Article PubMed/NCBI
  20. Mathews ST, Chellam N, Srinivas PR, Cintron VJ, Leon MA, Goustin AS, et al. Alpha2-HSG, a specific inhibitor of insulin receptor autophosphorylation, interacts with the insulin receptor. Mol Cell Endocrinol 2000;164:87-98 View Article PubMed/NCBI
  21. Mathews ST, Singh GP, Ranalletta M, Cintron VJ, Qiang X, Goustin AS, et al. Improved insulin sensitivity and resistance to weight gain in mice null for the Ahsg gene. Diabetes 2002;51:2450-2458 View Article PubMed/NCBI
  22. Pal D, Dasgupta S, Kundu R, Maitra S, Das G, Mukhopadhyay S, et al. Fetuin-A acts as an endogenous ligand of TLR4 to promote lipid-induced insulin resistance. Nat Med 2012;18:1279-1285 View Article PubMed/NCBI
  23. Stefan N, Häring HU. Circulating fetuin-A and free fatty acids interact to predict insulin resistance in humans. Nat Med 2013;19:394-395 View Article PubMed/NCBI
  24. Stefan N, Schick F, Häring HU. Ectopic fat in insulin resistance, dyslipidemia, and cardiometabolic disease. N Engl J Med 2014;371:2236-2237 View Article PubMed/NCBI
  25. Sato M, Kamada Y, Takeda Y, Kida S, Ohara Y, Fujii H, et al. Fetuin-A negatively correlates with liver and vascular fibrosis in nonalcoholic fatty liver disease subjects. Liver Int 2015;35:925-935 View Article PubMed/NCBI
  26. Gerst F, Fritz AK, Lorza Gil E, Kaiser G, Wolf E, Haering HU, et al. Fetuin-A impairs islet differentiation and function via inhibition of TGFbeta-1 signalling. Diabetologia 2018;61:S205-S206 View Article PubMed/NCBI
  27. Pampanini V, Inzaghi E, Germani D, Alterio A, Puglianiello A, Alisi A, et al. Serum Fetuin-A levels in obese children with biopsy proven nonalcoholic fatty liver disease. Nutr Metab Cardiovasc Dis 2018;28:71-76 View Article PubMed/NCBI
  28. Ballestri S, Meschiari E, Baldelli E, Musumeci FE, Romagnoli D, Trenti T, et al. Relationship of serum fetuin-A levels with coronary atherosclerotic burden and NAFLD in patients undergoing elective coronary angiography. Metab Syndr Relat Disord 2013;11:289-295 View Article PubMed/NCBI
  29. Weikert C, Stefan N, Schulze MB, Pischon T, Berger K, Joost HG, et al. Plasma fetuin-a levels and the risk of myocardial infarction and ischemic stroke. Circulation 2008;118:2555-2562 View Article PubMed/NCBI
  30. Fisher E, Stefan N, Saar K, Drogan D, Schulze MB, Fritsche A, et al. Association of AHSG gene polymorphisms with fetuin-A plasma levels and cardiovascular diseases in the EPIC-Potsdam study. Circ Cardiovasc Genet 2009;2:607-613 View Article PubMed/NCBI
  31. Stefan N, Häring HU. The role of hepatokines in metabolism. Nat Rev Endocrinol 2013;9:144-152 View Article PubMed/NCBI
  32. Bourebaba L, Marycz K. Pathophysiological implication of Fetuin-A glycoprotein in the development of metabolic disorders: A concise review. J Clin Med 2019;8:2033 View Article PubMed/NCBI
  33. Hennige AM, Staiger H, Wicke C, Machicao F, Fritsche A, Häring HU, et al. Fetuin-A induces cytokine expression and suppresses adiponectin production. PLoS One 2008;3:e1765 View Article PubMed/NCBI
  34. Lanthier N, Lebrun V, Berghmans MP, Molendi-Coste O, Leclercq IA. Impact of kupffer cells on high fat induced insulin resistance and liver fetuin-A expression. J Hepatol 2015;62:S702 View Article PubMed/NCBI
  35. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol 2009;62:e1-e34 View Article PubMed/NCBI
  36. Zeng X, Zhang Y, Kwong JS, Zhang C, Li S, Sun F, et al. The methodological quality assessment tools for preclinical and clinical studies, systematic review and meta-analysis, and clinical practice guideline: a systematic review. J Evid Based Med 2015;8:2-10 View Article PubMed/NCBI
  37. Liu CH, Ampuero J, Gil-Gómez A, Montero-Vallejo R, Rojas Á, Muñoz-Hernández R, et al. miRNAs in patients with non-alcoholic fatty liver disease: A systematic review and meta-analysis. J Hepatol 2018;69:1335-1348 View Article PubMed/NCBI
  38. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005;5:13 View Article PubMed/NCBI
  39. Seo JA, Kim NH, Park SY, Kim HY, Ryu OH, Lee KW, et al. Serum retinol-binding protein 4 levels are elevated in non-alcoholic fatty liver disease. Clin Endocrinol (Oxf) 2008;68:555-560 View Article PubMed/NCBI
  40. Reinehr T, Roth CL. Fetuin-A and its relation to metabolic syndrome and fatty liver disease in obese children before and after weight loss. J Clin Endocrinol Metab 2008;93:4479-4485 View Article PubMed/NCBI
  41. Yilmaz Y, Yonal O, Kurt R, Ari F, Oral AY, Celikel CA, et al. Serum fetuin A/α2HS-glycoprotein levels in patients with non-alcoholic fatty liver disease: relation with liver fibrosis. Ann Clin Biochem 2010;47:549-553 View Article PubMed/NCBI
  42. Haukeland JW, Dahl TB, Yndestad A, Gladhaug IP, Løberg EM, Haaland T, et al. Fetuin A in nonalcoholic fatty liver disease: in vivo and in vitro studies. Eur J Endocrinol 2012;166:503-510 View Article PubMed/NCBI
  43. Ou HY, Yang YC, Wu HT, Wu JS, Lu FH, Chang CJ. Increased fetuin-A concentrations in impaired glucose tolerance with or without nonalcoholic fatty liver disease, but not impaired fasting glucose. J Clin Endocrinol Metab 2012;97:4717-4723 View Article PubMed/NCBI
  44. Dogru T, Genc H, Tapan S, Aslan F, Ercin CN, Ors F, et al. Plasma fetuin-A is associated with endothelial dysfunction and subclinical atherosclerosis in subjects with nonalcoholic fatty liver disease. Clin Endocrinol (Oxf) 2013;78:712-717 View Article PubMed/NCBI
  45. Lebensztejn DM, Białokoz-Kalinowska I, Kłusek-Oksiuta M, Tarasów E, Wojtkowska M, Kaczmarski M. Serum fetuin A concentration is elevated in children with non-alcoholic fatty liver disease. Adv Med Sci 2014;59:81-84 View Article PubMed/NCBI
  46. Rametta R, Ruscica M, Dongiovanni P, Macchi C, Fracanzani AL, Steffani L, et al. Hepatic steatosis and PNPLA3 I148M variant are associated with serum Fetuin-A independently of insulin resistance. Eur J Clin Invest 2014;44:627-633 View Article PubMed/NCBI
  47. Celebi G, Genc H, Gurel H, Sertoglu E, Kara M, Tapan S, et al. The relationship of circulating fetuin-a with liver histology and biomarkers of systemic inflammation in nondiabetic subjects with nonalcoholic fatty liver disease. Saudi J Gastroenterol 2015;21:139-145 View Article PubMed/NCBI
  48. Wong VW, Wong GL, Chan HY, Yeung DK, Chan RS, Chim AM, et al. Bacterial endotoxin and non-alcoholic fatty liver disease in the general population: a prospective cohort study. Aliment Pharmacol Ther 2015;42:731-740 View Article PubMed/NCBI
  49. Cui Z, Xuan R, Yang Y. Serum fetuin A level is associated with nonalcoholic fatty liver disease in Chinese population. Oncotarget 2017;8:107149-107156 View Article PubMed/NCBI
  50. Şiraz ÜG, Doğan M, Hatipoğlu N, Muhtaroğlu S, Kurtoğlu S. Can fetuin-A be a marker for insulin resistance and poor glycemic control in children with type 1 diabetes mellitus?. J Clin Res Pediatr Endocrinol 2017;9:293-299 View Article PubMed/NCBI
  51. Nascimbeni F, Romagnoli D, Ballestri S, Baldelli E, Lugari S, Sirotti V, et al. Do nonalcoholic fatty liver disease and fetuin-A play different roles in symptomatic coronary artery disease and peripheral arterial disease?. Diseases 2018;6:17 View Article PubMed/NCBI
  52. Mondal SA, Dutta D, Kumar M, Singh P, Basu M, Selvan C, et al. Neck circumference to height ratio is a reliable predictor of liver stiffness and nonalcoholic fatty liver disease in prediabetes. Indian J Endocrinol Metab 2018;22:347-354 View Article PubMed/NCBI
  53. Kahraman A, Sowa JP, Schlattjan M, Sydor S, Pronadl M, Wree A, et al. Fetuin-A mRNA expression is elevated in NASH compared with NAFL patients. Clin Sci (Lond) 2013;125:391-400 View Article PubMed/NCBI
  54. Meex RCR, Watt MJ. Hepatokines: linking nonalcoholic fatty liver disease and insulin resistance. Nat Rev Endocrinol 2017;13:509-520 View Article PubMed/NCBI
  55. Stefan N, Häring HU, Schulze MB. Metabolically healthy obesity: the low-hanging fruit in obesity treatment?. Lancet Diabetes Endocrinol 2018;6:249-258 View Article PubMed/NCBI
  56. Stefan N. Causes, consequences, and treatment of metabolically unhealthy fat distribution. Lancet Diabetes Endocrinol 2020;8:616-627 View Article PubMed/NCBI
  57. Eigentler T, Lomberg D, Machann J, Stefan N. Lipodystrophic nonalcoholic fatty liver disease induced by immune checkpoint blockade. Ann Intern Med 2020;172:836-837 View Article PubMed/NCBI