Meta-analysis of the clinical efficacy of liraglutide in treating type 2 diabetes mellitus complicated with non-alcoholic fatty liver disease

Abstract

This study aimed to systematically evaluate the efficacy of liraglutide in treating type 2 diabetes mellitus (T2DM) complicated with non-alcoholic fatty liver disease (NAFLD) by comparing liraglutide with placebo or other drugs (mainly insulin). The PubMed, Web of Science, and National Library of Medicine databases were systematically searched from their inception until December 1, 2023. A meta-analysis was performed using Stata 15.1 software. A total of 12 studies with 13 outcome measures were included. The meta-analysis results revealed that liraglutide significantly reduced body mass index (mean difference [MD] = –1.06, 95%CI: –1.41, –0.70, p < 0.001), triglycerides (MD = –0.35, 95%CI: –0.61, –0.09, p = 0.0009), visceral adipose tissue (MD = –21.06, 95%CI: –34.58, –7.55, p = 0.002), and subcutaneous adipose tissue (MD = –20.53, 95%CI: –29.15, –11.90, p < 0.001) levels in patients with T2DM and NAFLD. Of the 11 studies, 2 reported the occurrence of adverse reactions, which were primarily gastrointestinal. Compared with placebo and other drugs (e.g., insulin), liraglutide may improve glucose metabolism, lipid and liver function parameters, and visceral and subcutaneous fat in patients with T2DM and NAFLD, thus constituting an effective treatment for these patients.

Introduction

With continuous improvements in living standards, the incidence and prevalence of diabetes have gradually increased. By 2015, the global number of patients with diabetes had reached 415 million, and it is estimated that by 2040, the global number of patients with diabetes will have risen to 642 million, with type 2 diabetes being the most common type [1]. Type 2 diabetes mellitus (T2DM) is a metabolic disorder characterized by abnormal glucose metabolism and insulin resistance. Patients with diabetes are prone to various liver injuries, including non-alcoholic fatty liver disease (NAFLD), liver cirrhosis, and liver tumors. Non-alcoholic fatty liver disease is characterized by excessive fat deposition in liver cells unrelated to alcohol or other known liver-damaging factors. Acquired metabolic stress liver injury is strongly associated with insulin resistance and genetic predisposition and includes simple fatty liver disease (SFL), non-alcoholic steatohepatitis, and related cirrhosis. Studies report that the prevalence of NAFLD in patients with T2DM is 28%–55% [2, 3], which is significantly higher than the prevalence of SFL in individuals of the same gender, age, and body weight [4]. The prevalence of NAFLD in people hospitalized for diabetes may be even higher. Obesity, dyslipidaemia, impaired glucose tolerance, and insulin resistance often coexist with NAFLD, suggesting a close relationship between NAFLD and T2DM.

Liraglutide is a glucagon-like peptide-1 (GLP-1) receptor agonist (GLP-1RA) that can promote glucose-dependent insulin synthesis and secretion and inhibit glucagon release. After the activation of the GLP-1 receptor in the gastrointestinal tract and brain, liraglutide can reduce gastrointestinal motility, delay gastric emptying, suppress appetite, and reduce food intake. Studies [5] have demonstrated that liraglutide, as a new type of hypoglycaemic drug, can reduce weight and lower blood lipids. One study evaluated the pharmacokinetic characteristics of liraglutide at different doses in 72 healthy males through a double-blind, randomized, placebo-controlled trial. The results revealed that after injecting liraglutide at doses of 2.5–20.0 ug/kg, the maximum blood drug concentration was reached after 9.3–12.0 hours, and the elimination half-life was 11–15 hours. When the liraglutide dose was adjusted to 5 ug/kg, the time to reach the maximum blood drug concentration was 9.3 hours and the half-life was 15 hours [6]. Therefore, the pharmacokinetic curve of liraglutide suggests that it can improve blood glucose levels when administered once daily. Liraglutide and semaglutide are both long-acting GLP-1RAs and have a 24 hour/day pharmacodynamic effect despite having different dosing intervals and dosages. Liraglutide was developed for once-daily administration and is available as an once-daily injection of up to 1.8 mg, whereas semaglutide is available as an once-weekly injection of up to 1.0 mg. Studies have demonstrated that in overweight or obese adults without diabetes, when combined with diet and physical activity, taking semaglutide improves weight loss more effectively than liraglutide [7]. Current studies have determined that liraglutide is not more effective than other GLP-1RA drugs in reducing hypoglycaemia, weight, or lipids [8].

Currently, the results of studies on the efficacy of liraglutide in reducing intrahepatic fat content and liver enzyme levels are inconsistent. Moreover, the antidiabetic drugs commonly used in clinical practice, while lowering blood glucose levels, do not have a definitive therapeutic effect on NAFLD. The present study is based on the above research status and aims to evaluate the clinical efficacy of liraglutide in the treatment of T2DM complicated with NAFLD by comparing liraglutide with placebo or other drugs (mainly insulin), using a systematic review and meta-analysis approach, to provide a reference and evidence for clinical medication.

1.  Data and Methods

1.1  Literature Search Strategy

In accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses, four databases were systematically searched: PubMed, Web of Science, Embase, and the National Library of Medicine. The search period was from database inception until December 1, 2023. A combination of subject headings and free-text terms was used for the search, and a snowballing search was performed. The English keywords used were ‘liraglutide or GLP-1 receptor agonist’ AND ‘T2DM or Type 2 diabetes’ AND ‘non-alcoholic fatty liver disease or nonalcoholic fatty liver or NAFLD or nonalcoholic steatohepatitis’ AND ‘effect or clinical effect.’ The relevant studies were searched by combining medical subject headings and text words in PubMed (see Supplementary Material 1 for details).

1.2  Inclusion and Exclusion Criteria

Inclusion criteria: ① Study type: randomized controlled trials (RCTs). ② Study population: patients diagnosed with T2DM complicated with NAFLD, diagnosed according to the T2DM criteria established by the WHO (1999) [9], the T2DM diagnosis and treatment guidelines (2010) [10], and the diagnostic criteria for NAFLD formulated by the Chinese Medical Association Hepatology Society (2010). The detailed diagnostic requirements are as follows: (a) hepatic steatosis determined through imaging or histology; (b) no notable alcohol consumption; (c) no competing aetiologies for hepatic steatosis; and (d) no co-existing causes for chronic liver disease, and a body mass index (BMI) >25 kg/m² at screening. ③ Intervention: The control group received a placebo or other drug treatment, while the experimental group received liraglutide treatment in addition to standard treatment. ④ Outcome measures: liver fat content (LF); glycaemic parameters such as glycated haemoglobin A1c (HbA1c); liver function parameters such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), and alkaline phosphatase (ALP); lipid metabolism parameters such as triglycerides (TG), total cholesterol (TC), high-density lipoprotein (HDL), and low-density lipoprotein (LDL); visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), BMI, and the occurrence of adverse reactions.

Exclusion criteria: ① non-RCT studies; ② duplicate publications; ③ basic research such as animal experiments or molecular mechanisms; ④ patients with fatty liver caused by alcohol consumption, viruses, or other uncertain factors; ⑤ studies without clear diagnostic criteria; and ⑥ studies without extractable data for analysis.

1.3  Literature Screening and Data Extraction

Two researchers conducted an independent literature screening. Initially, the screening was based on titles and abstracts, followed by a second screening of the full texts based on the inclusion and exclusion criteria. In case of disagreements, a third researcher was consulted for consensus through discussion. After completing the literature screening, two researchers independently extracted the data, which included the first author, publication year, study location, sample size, intervention measures, and outcome measures.

1.4  Assessment of Literature Quality

The quality of the literature was evaluated using the Jadad scoring system. Studies with a score ≥3 were considered high quality and met the inclusion criteria, whereas studies with a score <3 were considered low quality [11] and were excluded. The included literature was categorized based on this quality assessment.

1.5  Statistical Analysis

The meta-analysis was performed using Stata 15.1 software. For continuous data, the mean difference (MD) was used as the effect measure, whereas for categorical data, the odds ratio was used. The effect estimates were presented with point estimates and 95% confidence intervals (CI). Heterogeneity was assessed using I² tests: if I² < 50% or p > 0.1, the included studies were considered to have homogeneity, and a fixed-effect model (Mantel–Haenszel) was used for analysis. If I² > 50% or p ≤ 0.1, indicating substantial heterogeneity among the included studies, a random-effects model (DerSimonian–Laird) was used for analysis. The sensitivity analysis was conducted by eliminating the studies one by one, and a funnel plot was employed to identify publication bias. The significance level for the meta-analysis was set at α = 0.05.

2.  Results

2.1  Literature Search Results

A total of 267 articles were obtained through the database search. After removing duplicates (124 articles) and studies such as systematic reviews and case reports (77 articles), the full texts of the remaining articles were assessed based on the inclusion and exclusion criteria. Studies that were not RCTs, did not have clear diagnostic criteria for NAFLD, or included patients with fatty liver caused by alcohol consumption, viruses, or other uncertain factors, were excluded. Finally, 12 articles [12-23] were included in the meta-analysis. Fig. 1 presents the flowchart of the literature screening process.

Fig. 1

Flowchart of the literature screening process

2.2  Characteristics of the Included Studies and Quality Assessment

The 12 included studies were published between 2015 and 2020. A total of 596 patients with T2DM complicated with NAFLD were included, with 289 in the experimental group and 307 in the control group. Of the included studies, six were conducted in China, two in the Netherlands, and one each in the UK, USA, Finland, and Japan. All the included RCTs had quality assessment scores ranging from 5 to 7, indicating medium- to high-quality studies that met the inclusion criteria. The detailed characteristics of the included studies and quality assessment results are presented in Table 1. The BMI at baseline of the included studies was 27.6–40.7 kg/m² in the experimental group and 26.82–41.6 kg/m² in the control group (Supplementary Material 2).

Table 1

Basic Characteristics of Included Studies and Quality Assessment Grade

Included Studies	Year	Country	Sample Size		Average Age (years)		Intervention		Treatment Duration	Dosage (mg/d)	Outcome Variables	Literature Quality Assessment
			E	C	E	C	E	C
Armstrong et al. [12]	2016	UK	7	7	59.0	56.0	Liraglutide	Placebo	12 weeks	1.8	1, 2, 3, 4, 5, 6, 7, 8, 9, 10	5
Smits et al. (a) [13]	2017	Netherlands	17	17	60.8	65.8	Liraglutide	Placebo	12 weeks	1.8	2, 7, 8, 9, 10, 13	6
Smits et al. (b) [13]	2017	Netherlands	17	17	60.8	61.5	Liraglutide	Sitagliptin	12 weeks	1.8	2, 7, 8, 9, 10, 13	—
Guo et al. (a) [14]	2020	China	31	30	53.1	52.6	Liraglutide	Placebo	26 weeks	1.8	1, 2, 3, 4, 5, 6, 7, 8, 11, 12	5
Bizino et al. [15]	2020	Netherlands	23	26	60	59	Liraglutide	Placebo	26 weeks	1.8	1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12	6
Matikainen et al. [16]	2019	Finland	15	7	62	63	Liraglutide	Placebo	16 weeks	1.8	1, 2, 3, 4, 5, 6, 11, 12, 13	6
Vanderheiden et al. [17]	2016	USA	35	36	52.8	55.5	Liraglutide	Placebo	6 months	1.8	2, 11, 12, 13	5
Zhang et al. [18]	2020	China	30	30	50.2	51.5	Liraglutide	Pioglitazone	24 weeks	1.2	1, 2, 3, 4, 5, 6, 7, 8, 9, 13, 14	6
Feng et al. (a) [19]	2017	China	29	29	46.8	46.3	Liraglutide	Metformin	24 months	1.8	1, 2, 3, 4, 5, 6, 7, 8	6
Feng et al. (b) [19]	2017	China	29	29	46.8	48.1	Liraglutide	Gliclazide	24 months	1.8	1, 2, 3, 4, 5, 6, 7, 8	—
Tian et al. [20]	2018	China	52	75	58.5	56.4	Liraglutide	Metformin	12 weeks	0.6–1.2	1, 2, 3, 4, 5, 6, 7, 8	5
Tang et al. [21]	2015	China	18	17	60.7	60.4	Liraglutide	Insulin	12 weeks	1.8	1, 2, 4, 5, 6, 7, 8	5
Yan et al. (a) [22]	2019	China	24	24	43.1	45.6	Liraglutide	Insulin	26 weeks	1.8	1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 14	7
Yan et al. (b) [22]	2019	China	24	27	43.1	45.7	Liraglutide	Sitagliptin	26 weeks	1.8	1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 14	—
Guo et al. (b) [14]	2020	China	31	30	53.1	52.0	Liraglutide	Insulin	26 weeks	1.8	1, 2, 3, 4, 5, 6, 7, 8, 11, 12	7
Bouchi et al. [23]	2017	Japan	8	9	57	60	Liraglutide	Insulin	36 weeks	0.9	2, 4, 5, 6, 11, 12	6

Note: 1 = body mass index (BMI); 2 = HbA1c (glycated hemoglobin A1c); 3 = TC (total cholesterol); 4 = TG (triglycerides); 5 = HDL (high density lipoprotein); 6 = LDL (low density lipoprotein); 7 = AST (aspartate aminotransferase); 8 = ALT (alanine aminotransferase); 9 = GGT (Gamma-glutamyl transferase); 10 = ALP (alkaline phosphatase); 11 = VAT (visceral adipose tissue); 12 = SAT (subcutaneous adipose tissue); 13 = LF (liver fat); 14 = AEs (adverse events)

E = Experimental group; C = Control group.

2.3  Meta-Analysis Results

A total of 13 outcome measures were included in the meta-analysis. The results indicated that liraglutide had a significant impact on the levels of BMI, HbA1c, TG, VAT, and SAT in patients with T2DM complicated with NAFLD (p < 0.05). Of the 12 studies, 2 [18, 22] reported the occurrence of adverse reactions.

2.3.1  Body Mass Index

In total, 10 RCTs reported BMI levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As illustrated in Fig. 2A, there was substantial heterogeneity among the studies (I² = 78%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results revealed that liraglutide significantly reduced the BMI levels in patients with T2DM complicated with NAFLD (MD = –1.06, 95%CI: –1.41 to –0.70, p < 0.001). The BMI sensitivity analysis was conducted by deleting individual studies one by one. The results demonstrated that the study by Feng et al. (b) [19] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 63%.

Fig. 2

Forest plot comparing BMI and HbA1c between liraglutide and other drugs. A. BMI, B. HbA1c

2.3.2  Glycated Haemoglobin A1c

A total of 12 RCTs reported the HbA1c levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 2B, there was significant heterogeneity among the studies (I² = 92%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results indicated that liraglutide significantly reduced HbA1c levels in patients with T2DM complicated with NAFLD (MD = –0.40, 95%CI: –0.65 to –0.15, p = 0.002). The HbA1c sensitivity analysis was conducted by deleting individual studies one by one. The results demonstrated that the study by Smits (a) [13] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 83%.

2.3.3  Total Cholesterol

A total of eight RCTs reported the TC levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 3A, there was substantial heterogeneity among the studies (I² = 73%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results demonstrated that liraglutide did not have a significant impact on TC levels in patients with T2DM complicated with NAFLD (MD = –0.10, 95%CI: –0.27 to 0.07, p = 0.24). The TC sensitivity analysis was performed by deleting individual studies one by one. The results revealed that the study by Armstrong [12] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 67%.

Fig. 3

Forest plot comparing TC and TG between liraglutide and other drugs. A.TC, B. TG

2.3.4  Triglycerides

A total of nine RCTs reported the TG levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As illustrated in Fig. 3B, there was substantial heterogeneity among the studies (I² = 87%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results demonstrated that liraglutide significantly reduced TG levels in patients with T2DM complicated with NAFLD (MD = –0.35, 95%CI: –0.61 to –0.09, p = 0.0009). By excluding individual studies one by one, the sensitivity analysis revealed that no study had an impact on the overall effect.

2.3.5  High-Density Lipoprotein

A total of 10 RCTs reported the HDL levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 4A, there was moderate heterogeneity among the studies (I² = 55%, p = 0.009), indicating the use of a random-effects model for the meta-analysis. The results demonstrated that liraglutide did not have a significant impact on HDL levels in patients with T2DM complicated with NAFLD (MD = –0.01, 95%CI: –0.05 to 0.03, p = 0.59). The HDL sensitivity analysis was performed by deleting individual studies one by one. The results revealed that the study by Tang [21] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 67%.

Fig. 4

Forest plot comparing HDL and LDL between liraglutide and other drugs. A. HDL, B. LDL

2.3.6  Low-Density Lipoprotein

A total of 10 RCTs reported the LDL levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 4B, there was significant heterogeneity among the studies (I² = 94%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results indicated that liraglutide did not have a significant impact on LDL levels in patients with T2DM complicated with NAFLD (MD = 0.03, 95%CI: –0.17 to 0.23, p = 0.80). The LDL sensitivity analysis was performed by deleting individual studies one by one. The results demonstrated that the study by Tang [21] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 79%.

2.3.7  Aspartate Aminotransferase

A total of nine RCTs reported the AST levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 5A, there was significant heterogeneity among the studies (I² = 92%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results revealed that liraglutide did not have a significant impact on AST levels in patients with T2DM complicated with NAFLD (MD = –1.09, 95%CI: –2.79 to 0.62, p = 0.21). The AST sensitivity analysis was performed by deleting individual studies one by one. The results demonstrated that the study by Armstrong [12] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 85%.

Fig. 5

Forest plot comparing AST and ALT between liraglutide and other drugs. A. AST, B. ALT

2.3.8  Alanine Aminotransferase

A total of nine RCTs reported the ALT levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As illustrated in Fig. 5B, there was significant heterogeneity among the studies (I² = 93%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results demonstrated that liraglutide did not have a significant impact on ALT levels in patients with T2DM complicated with NAFLD (MD = –3.06, 95%CI: –6.17 to 0.05, p = 0.05). The ALT sensitivity analysis was performed by deleting individual studies one by one. The results indicated that the study by Armstrong [12] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 85%.

2.3.9  Gamma-Glutamyl Transferase

A total of four RCTs reported the GGT levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 6A, there was substantial heterogeneity among the studies (I² = 85%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results indicated that liraglutide did not have a significant impact on GGT levels in patients with T2DM complicated with NAFLD (MD = –0.93, 95%CI: –4.24 to 2.38, p = 0.58). By excluding individual studies one by one, the sensitivity analysis showed that no study had an impact on the overall effect.

Fig. 6

Forest plot comparing GGT and ALP between liraglutide and other drugs. A. GGT, B. ALP

2.3.10  Alkaline Phosphatase

A total of three RCTs reported the ALP levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 6B, there was no significant heterogeneity among the studies (I² = 0%, p = 0.98), indicating the use of a fixed-effects model for the meta-analysis. The results demonstrated that liraglutide had no significant impact on ALP levels in patients with T2DM complicated with NAFLD (MD = –0.93, 95%CI: –2.93 to 1.07, p = 0.36). By excluding individual studies one by one, the sensitivity analysis revealed that no study had an impact on the overall effect.

2.3.11  Visceral Adipose Tissue

A total of six RCTs reported the VAT levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 7A, there was substantial heterogeneity among the studies (I² = 78%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results revealed that liraglutide significantly reduced VAT levels in patients with T2DM complicated with NAFLD (MD = –21.06, 95%CI: –34.58 to –7.55, p = 0.002). The VAT sensitivity analysis was performed by deleting individual studies one by one. The results demonstrated that the study by Guo (a) [14] had a significant impact on the overall effect, and after deleting it, the heterogeneity decreased to 58%.

Fig. 7

Forest plot comparing VAT and SAT, LF between liraglutide and other drugs. A. VAT, B. SAT, C. LF

2.3.12  Subcutaneous Adipose Tissue

A total of six RCTs reported the SAT levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As illustrated in Fig. 7B, there was moderate heterogeneity among the studies (I² = 49%, p = 0.06), indicating the use of a fixed-effects model for the meta-analysis. The results demonstrated that liraglutide significantly reduced SAT levels in patients with T2DM complicated with NAFLD (MD = –20.53, 95%CI: –29.15 to –11.90, p < 0.001). By excluding individual studies one by one, the sensitivity analysis revealed that no study had an impact on the overall effect.

2.3.13  Liver Fat Content

A total of four RCTs reported the LF levels before and after treatment in the experimental and control groups. The differences before and after treatment were extracted for analysis. As presented in Fig. 7C, there was substantial heterogeneity among the studies (I² = 81%, p < 0.001), indicating the use of a random-effects model for the meta-analysis. The results revealed that liraglutide had no significant impact on LF levels in patients with T2DM complicated with NAFLD (MD = –1.18, 95%CI: –2.85 to 0.49, p = 0.17). By excluding individual studies one by one, the sensitivity analysis revealed that no study had an impact on the overall effect.

2.3.14  Adverse Events

Of the 12 studies, 2 [18, 22] reported the occurrence of adverse events. One study comparing liraglutide with pioglitazone [18] reported a higher incidence of gastrointestinal adverse events in the liraglutide group. Another study conducted by Yan et al. [22] investigated the efficacy and safety of liraglutide compared with insulin and sitagliptin and found no significant differences in the occurrence of adverse events.

2.3.15  Publication Bias Analysis

A publication bias analysis was performed for all outcomes. As illustrated in Fig. 8, the funnel plots for HbA1c (A), TC (B), AST (C), and GGT (D) are clearly asymmetrical, suggesting the possibility of publication bias.

Fig. 8

The funnel plots of HbA1c (A), TC (B), AST (C) and GGT (D) for publication bias. A. HbA1c, B. TC, C. AST, D. GGT

3.  Discussion

This study evaluated the clinical efficacy of liraglutide in treating patients with T2DM combined with NAFLD and used a meta-analysis to analyze the combined outcome measures. The results of the study revealed that liraglutide had a significant impact on the levels of BMI, HbA1c, TG, VAT, and SAT in patients with T2DM combined with NAFLD (p < 0.05). Two studies [18, 22] reported the occurrence of adverse reactions.

Non-alcoholic fatty liver disease is a chronic disease with a prevalence of 62% among patients with T2DM. Oestrogen has protective effect in women, which may result in a slightly higher occurrence rate in men. The most common factors associated with clinical NAFLD are dyslipidaemia, obesity, and T2DM. These factors are collectively known as metabolic syndrome, which is also associated with coronary heart disease, hypertension, and atherosclerosis. The common underlying mechanism for these conditions is insulin resistance [24]. Studies have suggested that patients with NAFLD have relatively high plasma resistin levels associated with insulin resistance, and resistin levels are positively correlated with IR. Insulin resistance can disrupt lipid metabolism, and elevated fasting insulin levels can decrease the plasma HDL cholesterol concentration while increasing TG levels [25]. In healthy individuals, insulin-sensitive lipase inhibition leads to the effective storage of fatty acids without their release into the bloodstream. However, in patients with diabetes, insulin resistance in the liver leads to increased levels of circulating free fatty acids and cholesterol, decreased levels of HDL cholesterol, and the occurrence of dysregulated lipid metabolism. At this stage, hepatocytes are infiltrated by a large amount of fat, leading to hepatic steatosis, hepatocyte degeneration, necrosis, and the subsequent development of fatty liver. Hepatic insulin resistance is a key factor in the development of NAFLD [26]. It can therefore be inferred that T2DM and NAFLD have a common mechanism, insulin resistance. In addition, liraglutide has a role in improving insulin resistance, which may be the main mechanism for treating patients with T2DM combined with NAFLD.

The results of the meta-analysis in this study indicated that liraglutide can significantly reduce BMI, HbA1c, and TG levels in patients with T2DM combined with NAFLD, as well as decreasing VAT and SAT levels. These results suggest that liraglutide may improve glucose metabolism, lipid profiles, and liver function and reduce visceral and subcutaneous fat in patients with T2DM combined with NAFLD, thereby effectively treating this condition. The results of this study are consistent with those of previously published studies on the treatment of patients with T2DM combined with NAFLD using GLP-1RAs [27-30]. For example, Zhu et al. [29] included a total of 468 patients in 8 studies, concluding that GLP-1RAs can significantly reduce SAT, VAT, BMI, HbA1c, and other levels. Thus, GLP-1RAs provide an effective treatment for improving SAT and some metabolic indices in patients with T2DM and NAFLD. However, in the above study, GLP-1RA drugs, including liraglutide, dulaglutide, and exenatide, were administered to the experimental group, and their efficacy was evaluated. By contrast, the present study was highly targeted and only included the effect of liraglutide in the treatment of patients with T2DM and NAFLD, demonstrating the efficacy of a single drug. Moreover, more studies and patients were included in this study, and more studies were included under a single analysis indicator, which further improved the estimation accuracy of effect size.

However, the exact mechanism of liraglutide in treating NAFLD is not yet clear. The potential mechanisms of action are the following. (1) The CCAAT/enhancer binding protein homologous protein (CHOP) signaling pathway: CHOP plays a critical role in cellular lipid metabolism transcription and is involved in endoplasmic reticulum stress-mediated apoptosis and hepatocyte apoptosis in NAFLD [31]. (2) Endoplasmic reticulum protein 46 (ERp46) signaling pathway: Hepatocyte apoptosis in NAFLD is mainly induced by the accumulation of free fatty acids in the liver, leading to endoplasmic reticulum stress [32]. Also known as endoplasmic reticulum protein disulfide isomerase or protein disulfide isomerase domain-containing protein 5 precursor, ERp46 is highly expressed in liver/plasma and endothelial tissues [33]. Liraglutide may exert its effects through the ERp46 pathway to limit endoplasmic reticulum stress-induced apoptosis, prevent the development of NAFLD, and inhibit its progression [34, 35]. (3) Insulin-like growth factor 2 mRNA binding protein (IGF2BP3): IGF2BP3 is a secretory protein involved in the transport, stabilization, cell proliferation, and migration of RNA. It is a highly specific member of the insulin-like growth factor binding protein family that can specifically bind to the mRNA encoding insulin-like growth factor II (IGF-II) [36, 37].

This study determined that adverse reactions were reported in only two articles, which also indicates the safety of liraglutide. Because of liraglutide’s glucose-dependent mechanism of action, the risk of hypoglycaemia is low. The most common adverse events associated with its use are gastrointestinal-related events, such as nausea, abdominal pain, decreased appetite, diarrhoea, and vomiting. Delayed gastric emptying is the main cause of gastrointestinal reactions to liraglutide. Delayed gastric emptying can inhibit the entry of nutrients into the small intestine, leading to gastric distension, a feeling of fullness, and nausea. Since liraglutide is a long-acting agonist of the GLP-1 receptor, which can keep the GLP-1 receptor activated for a long time, the effect of delayed gastric emptying quickly desensitizes. Therefore, gastrointestinal reactions generally occur within the first 4 weeks of treatment and are transient, gradually disappearing over time. Additionally, the incidence of gastrointestinal adverse reactions to liraglutide is dose dependent, with a higher proportion of patients experiencing adverse reactions at higher doses. This suggests that the occurrence of gastrointestinal adverse reactions is dose dependent and mostly transient during the process of increasing the drug dose. Few patients discontinue the medication because of gastrointestinal reactions, and the incidence of gastrointestinal adverse reactions decreases rapidly to below 10% within 1 month, maintaining a very low level thereafter. Some studies have demonstrated that a small percentage (0.2%) of patients with T2DM treated with liraglutide may develop acute pancreatitis. However, compared with the general population, patients with T2DM are at a relatively high risk of developing acute pancreatitis. The FDA and EMA have expressed concerns about the increased risk of hospitalization for acute pancreatitis associated with incretin-based therapies. Further research and review revealed that incretin-based drugs were not associated with pancreatitis or pancreatic cancer [38]. Liraglutide, as a highly homologous GLP-1 analogue, has low immunogenicity and induces minimal production of anti-liraglutide antibodies. Only 4%–13% of patients in the studies developed anti-liraglutide antibodies, and therefore, its clinical safety and efficacy were not affected.

At present, there is a meta-analysis showing that liraglutide can reduce all-cause mortality and cardiovascular mortality in type 2 diabetes [39]. Glucagon-like peptide -1 (GLP-1) receptor agonists such as liraglutide have beneficial effects on cardiovascular, mortality and renal prognosis in patients with type 2 diabetes [40], but its effect on reducing body weight and blood pressure is weaker than that of Semaglutide and Sodium-glucose cotransporter -2 [41]. In the further study, we will explore the combined effect of liraglutide and other hypoglycemic drugs.

This study has some limitations. 1) Liraglutide was approved later than other antidiabetic drugs (sulfonylureas, metformin, etc.), and there have been fewer clinical RCTs focusing on its use in the treatment of NAFLD. 2) The sample size of the RCTs included in this study was small, which may lead to false-positive and false-negative results. This indicates the need for large-sample, multicenter studies to further verify the findings. 3) No search was conducted for ‘grey literature,’ such as conference data. Additionally, because of language and permission limitations, the literature search was restricted to articles in English, and some databases could not be searched, which may introduce some selection bias and missing literature. 4) The results of the meta-analysis in this study revealed different degrees of heterogeneity in each indicator. Although the source of heterogeneity was explored through a sensitivity analysis, heterogeneity remained. Therefore, future studies should further limit the criteria of the included studies in terms of study design and data collection to reduce heterogeneity.

4.  Conclusion

This study systematically evaluated the efficacy of liraglutide in the treatment of T2DM combined with NAFLD by comparing liraglutide with placebo or other agents, mainly insulin. The results of the meta-analysis revealed that liraglutide reduced BMI, VAT, and SAT levels in patients with T2DM combined with NAFLD. In terms of improvements in other indicators, more high-quality studies should be included in the future for further exploration. These results suggest that liraglutide may improve not only glucose metabolism in patients with T2DM and NAFLD but also blood lipid and liver function indexes, as well as visceral fat and subcutaneous fat, to effectively treat patients with low levels of T2DM and NAFLD. Although this study has some limitations, it may provide a reference for clinical treatment or drug use.

Declarations

Ethics approval and consent to participate

An ethics statement is not applicable because this study is based exclusively on published literature.

Competing Interest

The authors declare that they have no competing interests.

Funding

This research is supported by research project of Jiangsu Health Vocational College (JKC2022051) and Science and technology development project for social undertakings of Pukou District (S2023-7).

Author Contributions

Xu YY conceived of the study, and Wang X and She YQ participated in its design and data analysis and statistics, Liu J and Zhang Q helped to draft the manuscript. All authors read and approved the final manuscript.

Consent for publication

Not applicable.

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

References

• 1   Li  Y (2016) Prospective cohort study on the association between non-alcoholic fatty liver disease, liver enzyme levels, and the risk of type 2 diabetes. Huazhong University of Science and Technology (In Chinese).
• 2   Roden  M (2006) Mechanisms of disease: hepatic steatosis in type 2 diabetes-pathogenesis and clinical relevance. Nat Clin Pract Endocrinol Metab 2: 335–348.
• 3   Yasutake  K,  Kohjima  M,  Kotoh  K,  Nakashima  M,  Nakamuta  M, et al. (2014) Dietary habits and behaviors associated with nonalcoholic fatty liver disease. World J Gastroenterol 20: 1756–1767.
• 4   Miele  L,  Forgione  A,  Hernandez  AP,  Gabrieli  ML,  Vero  V, et al. (2005) The natural history and risk factors for progression of non-alcoholic fatty liver disease and steatohepotitis. Eur Rew Med Phamacal Sci 9: 273–277.
• 5   Liu  K (2017) The effect of liraglutide on hepatic steatosis in patients with type 2 diabetes complicated by non-alcoholic fatty liver disease. Chinese Journal of Integrated Traditional and Western Medicine on Digestion 25: 195–198 (In Chinese).
• 6   Yang  W (2010) Latest research progress on human GLP-1 analog liraglutide. Chinese Journal of Endocrinology and Metabolism 26: 841–844 (In Chinese).
• 7   Eslam  M,  Sarin  SK,  Wong  VW,  Fan  JG,  Kawaguchi  T, et al. (2020) The asian pacific association for the study of the liver clinical practice guidelines for the diagnosis and management of metabolic associated fatty liver disease. Hepatol Int 14: 889–919.
• 8   Rubino  DM,  Greenway  FL,  Khalid  U,  O’Neil  PM,  Rosenstock  J, et al. (2022) Effect of weekly subcutaneous semaglutide vs. daily liraglutide on body weight in adults with overweight or obesity without diabetes: the STEP 8 randomized clinical trial. JAMA 327: 138–150.
• 9   Alberti  KG,  Zimmet  PZ (1998) Definition, diagnosis and classifications of diabetes mellitus and its complications. Diabet Med 15: 539–553.
• 10   Chinese Diabetes  Society (2012) Guidelines for the prevention and treatment of type 2 diabetes in China (2010 edition). Chinese Journal of Diabetes 20: 54–109 (In Chinese).
• 11   Jadad  AR,  Moore  RA,  Carroll  D,  Jenkinson  C,  Reynolds  DJ, et al. (1996) Assessing the quality of reports of randomized clinical trials: is blinding necessary. Control Clin Trials 17: 1–12.
• 12   Armstrong  MJ,  Hull  D,  Guo  K,  Barton  D,  Hazlehurst  JM, et al. (2016) Glucagon-like peptide 1 decreases lipotoxicity in non-alcoholic steatohepatitis. J Hepatol 64: 399–408.
• 13   Smits  MM,  Tonneijck  L,  Muskiet  MH,  Kramer  MH,  Pieters-van den Bos  IC, et al. (2017) Pancreatic effects of liraglutide or sitagliptin in overweight patients with type 2 diabetes: a 12-week randomized, placebo-controlled trial. Diabetes Care 40: 301–308.
• 14   Guo  W,  Tian  W,  Lin  L,  Xu  X (2020) Liraglutide or insulin glargine treatments improves hepatic fat in obese patients with type 2 diabetes and nonalcoholic fatty liver disease in twenty-six weeks: a randomized placebo-controlled trial. Diabetes Res Clin Pract 170: 108487.
• 15   Bizino  MB,  Jazet  IM,  de Heer  P,  van Eyk  HJ,  Dekkers  IA, et al. (2020) Placebo-controlled randomised trial with liraglutide on magnetic resonance endpoints in individuals with type 2 diabetes: a pre-specified secondary study on ectopic fat accumulation. Diabetologia 63: 65–74.
• 16   Matikainen  N,  Söderlund  S,  Björnson  E,  Pietiläinen  K,  Hakkarainen  A, et al. (2019) Liraglutide treatment improves postprandial lipid metabolism and cardiometabolic risk factors in humans with adequately controlled type 2 diabetes: a single-centre randomized controlled study. Diabetes Obes Metab 21: 84–94.
• 17   Vanderheiden  A,  Harrison  LB,  Warshauer  JT,  Adams-Huet  B,  Li  X, et al. (2016) Mechanisms of action of liraglutide in patients with type 2 diabetes treated with high-dose insulin. J Clin Endocrinol Metab 101: 1798–1806.
• 18   Zhang  L,  Qu  X,  Sun  Z,  Zhang  Y (2020) Effect of liraglutide therapy on serum fetuin A in patients with type 2 diabetes and non-alcoholic fatty liver disease. Clin Res Hepatol Gastroenterol 44: 674–680.
• 19   Feng  W,  Gao  C,  Bi  Y,  Wu  M,  Li  P, et al. (2017) Randomized trial comparing the effects of gliclazide, liraglutide, and metformin on diabetes with non-alcoholic fatty liver disease. J Diabetes 9: 800–809.
• 20   Tian  F,  Zheng  Z,  Zhang  D,  He  S,  Shen  J (2018) Efficacy of liraglutide in treating type 2 diabetes mellitus complicated with non-alcoholic fatty liver disease. Biosci Rep 38: BSR20181304.
• 21   Tang  A,  Rabasa-Lhoret  R,  Castel  H,  Wartelle-Bladou  C,  Gilbert  G, et al. (2015) Effects of insulin glargine and liraglutide therapy on liver fat as measured by magnetic resonance in patients with type 2 diabetes: a randomized trial. Diabetes Care 38: 1339–1346.
• 22   Yan  J,  Yao  B,  Kuang  H,  Yang  X,  Huang  Q, et al. (2019) Liraglutide, sitagliptin, and insulin glargine added to metformin: the effect on body weight and intrahepatic lipid in patients with type 2 diabetes mellitus and nonalcoholic fatty liver disease. Hepatology 69: 2414–2426.
• 23   Bouchi  R,  Nakano  Y,  Fukuda  T,  Takeuchi  T,  Murakami  M, et al. (2017) Reduction of visceral fat by liraglutide is associated with ameliorations of hepatic steatosis, albuminuria, and micro-inflammation in type 2 diabetic patients with insulin treatment: a randomized control trial. Endocr J 64: 269–281.
• 24   Liu  X,  Zhang  N,  Wu  F (2012) Relationship between type 2 diabetes complicated by non-alcoholic fatty liver disease and chronic kidney disease. Chinese General Practice 15: 3348–3350 (In Chinese).
• 25   Zheng  X,  Zhao  W,  Zhang  X,  Feng  L,  Zhang  H (2014) Study on the correlation between non-high-density lipoprotein cholesterol levels and non-alcoholic fatty liver disease in patients with type 2 diabetes. Journal of Practical Medicine 30: 437–439 (In Chinese).
• 26   Olmos  G,  Llado  J (2014) Tumor necrosis factor alpha: a link between neuroinflammation and excitotoxicity. Mediators Inflamm 2014: 1–12.
• 27   Dong  Y,  Lv  Q,  Li  S,  Wu  Y,  Li  L, et al. (2017) Efficacy and safety of glucagon-like peptide-1 receptor agonists in nonalcoholic fatty liver disease: a systematic review and meta-analysis. Clin Res Hepatol Gastroenterol 41: 284–295.
• 28   Luo  Q,  Wei  R,  Cai  Y,  Zhao  Q,  Liu  Y, et al. (2022) Efficacy of off-label therapy for non-alcoholic fatty liver disease in improving non-invasive and invasive biomarkers: a systematic review and network meta-analysis of randomized controlled trials. Front Med (Lausanne) 9: 793203.
• 29   Zhu  Y,  Xu  J,  Zhang  D,  Mu  X,  Shi  Y, et al. (2021) Efficacy and safety of GLP-1 receptor agonists in patients with type 2 diabetes mellitus and non-alcoholic fatty liver disease: a systematic review and meta-analysis. Front Endocrinol (Lausanne) 12: 769069.
• 30   Tang  W,  Xu  Q,  Hong  T,  Tong  G,  Feng  W, et al. (2016) Comparative efficacy of anti-diabetic agents on nonalcoholic fatty liver disease in patients with type 2 diabetes mellitus: a systematic review and meta-analysis of randomized and non-randomized studies. Diabetes Metab Res Rev 32: 200–216.
• 31   Rahman  K,  Liu  Y,  Kumar  P,  Smith  T,  Thorn  NE, et al. (2016) C/EBP homologous protein modulates liraglutide-mediated attenuation of non-alcoholic steatohepatitis. Lab Invest 96: 895–908.
• 32   Knoblach  B,  Keller  BO,  Groenendyk  J,  Aldred  S,  Zheng  J, et al. (2003) ERp19 and ERp46, rew members of the thioredoxin family of endoplasmic reticulum proteins. Mol Cell Proteomics 2: 1104–1119.
• 33   Alberti  A,  Karamessinis  P,  Peroulis  M,  Kypreou  K,  Kavvadas  P, et al. (2009) ERp46 is reduced by high glucose and regulates insulin content in pancreatic beta-cells. Am J Physiol Endocrinol Metab 297: e812–821.
• 34   Ao  N,  Yang  J,  Wang  X,  Du  J (2016) Glucagon-like peptide-I preserves non- alcoholic fatty liver disease through inhibition of the endoplasmic reticulum stress-associated pathway. Hepatol Res 46: 343–353.
• 35   Li  C,  Zota  V,  Woda  BA,  Rock  KL,  Fraire  AE, et al. (2007) Expression of a novel oncofetal mRNA-binding protcin IMP3 in endometrial carcinomas: diagnostic significance and clinicopathologic correlations. Mod Pathol 20: 1263–1268.
• 36   Song  Y,  Liu  Z,  Fan  Y,  Zhang  Y,  Li  X, et al. (2016) Effects of liraglutide on insulin-like growth factor in the liver of rats with non-alcoholic fatty liver disease. Chinese Journal of Hepatology 24: 614–616 (In Chinese).
• 37   Rodriguez  S,  Eiriksdottir  G,  Gaunt  TR,  Harris  TB,  Launer  LJ, et al. (2010) IGF2BP1,IGF2BP2 and IGF2BP3 genotype, haplotype and genetic model studies in metabolic syndrome traits and diabetes. Growth Horm IGF Res 20: 310–318.
• 38   Luo  H,  Tan  W,  Zhang  Y,  Hong  H (2017) Research progress on glucagon-like peptide-1 and its analogs in the treatment of type 2 diabetes. China Pharmacy 28: 3290–3294 (In Chinese).
• 39   Tsapas  A,  Avgerinos  I,  Karagiannis  T,  Malandris  K,  Manolopoulos  A, et al. (2020) Comparative effectiveness of glucose-lowering drugs for type 2 diabetes: a systematic review and network meta-analysis. Ann Intern Med 173: 278–286.
• 40   Kristensen  SL,  Rørth  R,  Jhund  PS,  Docherty  KF,  Sattar  N, et al. (2019). Cardiovascular, mortality, and kidney outcomes with GLP-1 receptor agonists in patients with type 2 diabetes: a systematic review and meta-analysis of cardiovascular outcome trials. Lancet Diabetes Endocrinol 7: 776–785.
• 41   Tsapas  A,  Karagiannis  T,  Kakotrichi  P,  Avgerinos  I,  Mantsiou  C, et al. (2021). Comparative efficacy of glucose-lowering medications on body weight and blood pressure in patients with type 2 diabetes: a systematic review and network meta-analysis. Diabetes Obes Metab 23: 2116–2124.