Effectiveness of Bariatric Surgeries for Metabolic Dysfunction-Associated Steatotic Liver Disease: A Systematic Review and Meta-Analysis

: Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common liver disease worldwide and simultaneously the most common indication for liver transplants in Western countries This study aims to evaluate the effectiveness of laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB) on MDASLD histologically and biochemically. 14 studies met our criteria with a total population of 1942 who underwent LSG or RYGB 1–14. The Newcastle-Ottawa Scale (NOS) was used for quality evaluation of the included studies. Results: Both surgeries were effective in decreasing laboratory biomarkers like ALP, GGT, AST, and ALT with non-significant superiority of LSG over RYGB which did not significantly improve the AST level after one year. LSG showed more decrease in ALT levels (MD = − 17.56, 95% CI = ( − 23.04, − 12.089), p 0.001) and LSG was associated with increased change in NAS score with slight superiority. Both LSG and RYBG improve NAD and NASH outcomes after one and 10 years of surgery. However, randomized clinical trials with large samples are needed to confirm these results.


Introduction
Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common liver disease worldwide and the most common indication for liver transplants in Western countries [1].The main risk factor for NAFLD is obesity, which affects more than 80% of patients at the time of diagnosis [2].
Bariatric surgery has proven to be effective in improving metabolic dysfunction and related steatohepatitis and fibrosis.Guidelines currently consider bariatric surgery a treatment option in non-cirrhotic MASLD/MASH patients who meet the criteria for metabolic weight loss surgery [3].Many systematic reviews demonstrate that performing bariatric surgery in severely obese patients with compensated cirrhosis is safe and effective [4,5].Therefore, this study aims to evaluate the effectiveness of bariatric surgeries against MASLD histologically and biochemically.
databases PubMed, ScienceDirect, Google Scholar, and Web of Science only for studies published in the ten years from January 2013 to December 2023.
The general search string used in all searches was as follows: (non-alcoholic fatty liver disease) OR (nonalcoholic fatty liver disease) OR (NAFLD) OR (steatosis) OR (steatohepatitis) OR (liver cirrhosis) AND (bariatric surgery) OR (metabolic surgery) OR (weight loss surgery) OR (gastric bypass) OR (sleeve gastrectomy).The search terms were identified in the title, abstract, or medical subject heading.

Inclusion and Exclusion Criteria
This systematic review included case-control studies, cross-sectional studies, retrospective cohort studies, and prospective cohort studies that were published in the English language from January 2013 to December 2023 and included both males and females, over 14 years old, who underwent any bariatric surgery such as gastric bypass or sleeve gastrectomy.The outcome to measure in this study were the histologic improvement of MASLD as reflected by the NAFLD activity score (NAS), presence of steatohepatitis, presence of fibrosis, and fibrosis stage and the biochemical response as measured by laboratory values of aspartate aminotransferase (AST), alanine transaminase (ALT), alkaline phosphatase (ALP), and gamma-glutamyl transferase (GGT).Exclusion criteria included animal studies, studies not available in English, case reports, editorials, systemic reviews, letters, and meta-analyses.Pediatrics, geriatrics populations, and interventions other than bariatric surgeries were excluded.

Selection of Articles and Data Extraction
Four independent authors (R.A., M.A., O.A., and R.A.) screened the articles using Rayyan based on the titles and abstracts [7].Rayyan also was used to eliminate duplicate articles.The full-text screening of all the included studies was carried out by two independent authors (G.A. and A.A.).Disagreements that arose during the screening process were resolved by L.A.All authors manually and independently extracted the data, which included study information (author, country, year of publication, study design, number of patients), demographic data (age, mean ± SD/median [range], gender), and outcomes (NAS score, presence of steatohepatitis, presence of fibrosis, fibrosis stage, biochemical response).

Quality Assessment
The Newcastle-Ottawa Scale (NOS) for non-randomized studies was used for the quality evaluation of the included studies [8].Most of the included primary observational studies in the current review were considered to be of good quality except Elyasinia et al., 2021 which was considered to be of fair quality because they used an ultrasonographic assessment for non-alcoholic steatohepatitis without mentioning the validity assessment of the adopted grading system.In contrast, the other included studies used either a histopathology-based grading system or a validated ultrasonographic evaluation.The quality of the studies was scored using the Newcastle-Ottawa Scale as shown in Table 1.

Data Analysis
Data were analyzed with the OpenMetaAnalyst for Windows 8 (64-bit) (built 4 June 2015) [23].We calculated the changes in laboratory and clinical parameters following the formula presented in the Cochrane Handbook for Systematic Reviews and Metaanalysis [24].We used mean difference and 95% confidence interval in the analysis and the results were considered significant when p values were less than 0.05.The heterogeneity was assessed using the Cochrane Q test and the I 2 statistical test, and the results were considered heterogeneous when the p value was less than 0.1 and I 2 was greater than 50%.We used the random-effects model in conducting the meta-analyses because of the presence of a heterogenous population in the included studies, as most of them included both NAFLD and NASH patients without performing subgroup analysis.We tried to resolve the heterogeneity using sensitivity analysis or subgroup analysis; however, we could not perform subgroup analysis because of the limited number of included studies and the fact that the included studies did not present separate data for NAFLD and NASH patients.

Literature Search Results
The database literature search resulted in 318 records; from them, 28 were removed as duplicates.We performed title and abstract screening on 289 records, and only 14 articles were eligible for the full-text screening phase, which resulted in 14 articles that were eligible for qualitative reporting and meta-analysis.Figure 1 shows the full details.

Baseline and Demographic Characteristics
We included 14 studies with a total population of 1942 individuals who underwent laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB).Nine studies were prospective cohort studies, while the rest were retrospective.The full details of age, sample size, country, intervention, gender, and baseline data of studies' outcomes are de-

Baseline and Demographic Characteristics
We included 14 studies with a total population of 1942 individuals who underwent laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB).Nine studies were prospective cohort studies, while the rest were retrospective.The full details of age, sample size, country, intervention, gender, and baseline data of studies' outcomes are described in Tables 2 and 3.
ALP has only been assessed pre-and postoperatively in Esquivel et al., 2018 [19] and they found that its level was significantly decreased after laparoscopic SG from 36 ± 18 (U/L) at baseline to 19 ± 9 after one year, p < 0.0001.
Figure 2 shows the full details of the pooled analysis of GGT change one year after LSG.NR: not reported.
ALP has only been assessed pre-and postoperatively in Esquivel et al., 2018 [19] and they found that its level was significantly decreased after laparoscopic SG from 36 ± 18 (U/l) at baseline to 19 ± 9 after one year, p < 0.0001.

BMI Change
Both surgical procedures were effective in reducing BMI after one year (LSG MD = −11.578,95% CI = (−12.982,−10.173), p < 0.001; RYGB MD = −10.509,95% CI = (−15.516,−5.502), p < 0.001); therefore, the p value for subgroup difference was 0.807.The outcome was heterogeneous, and we could not resolve it by the leave-one-out method or by performing subgroup analysis.Figure 5 shows the full details.The meta-analyses of both AST and ALT changes were heterogeneous; however, we could not resolve the heterogeneity by the leave-one-out method or subgroup analysis.p < 0.001); therefore, the p value for subgroup difference was 0.807.The outcome was heterogeneous, and we could not resolve it by the leave-one-out method or by performing subgroup analysis.Figure 5 shows the full details.
Additionally, Yeo et al., 2019 [18] found no significant difference between (LSG + Laparoscopic adjustable gastric banding) and RYGB in NAFLD fibrosis scores after six months and one year p = 0.205 and 0.119, respectively.Pedersen et al., 2021 [13] found that both surgeries did not significantly improve the Kleiner Fibrosis score and no significant differences were observed between both procedures.Additionally, Netanel et al., 2020 [15] found that LSG did not improve Fibro Test scores after one year (p = 0.061).
Additionally, Yeo et al., 2019 [18] found no significant difference between (LSG + Laparoscopic adjustable gastric banding) and RYGB in NAFLD fibrosis scores after six months and one year p = 0.205 and 0.119, respectively.Pedersen et al., 2021 [13] found that both surgeries did not significantly improve the Kleiner Fibrosis score and no significant differences were observed between both procedures.Additionally, Netanel et al., 2020 [15] found that LSG did not improve Fibro Test scores after one year (p = 0.061).

Discussion
Our systematic review and meta-analysis revealed that both surgeries effectively decreased laboratory biomarkers such as ALP, GGT, AST, and ALT, with nonsignificant superiority of LSG over RYGB, which did not significantly improve the AST level after one year.Additionally, this review found no significant difference between BMI and fibrosis between both procedures; however, LSG significantly improved the NAS score while RYGB did not.
This systematic review has compared the levels of laboratory biomarkers such as ALT, AST, GGT, and ALP in both groups as an improvement of MASLD after surgeries, as another study has proven that bariatric surgeries are effective in decreasing ALT and AST levels after two and ten years of surgery compared to only medical treatment [25].Despite these laboratory tests having low specificity and sensitivity, they can be used beside imaging and scoring systems in evaluating NAFLD activity instead of biopsy, which is considered an invasive and expensive procedure with a risk of bleeding or bile leaking [26].
By comparing both procedures, LSG showed nonsignificant superiority in improving liver enzymes compared to RYGB [9,19].Additionally, the ALP enzyme was reported only in Esquivel et al., 2018 [19], who found a significant reduction in its level one year of LSG, p < 0.001.In AST, there was a significant reduction one year after LSG in comparison to RYGB.Furthermore, in ALT, LSG showed more decrease than RYGB, which was also nonsignificant, which was supported by another systematic review that compared both procedures and found that LSG significantly decreased ALT and ALP compared to RYGB; however, no significant differences were observed between both procedures regarding GGT and AST [27].These results were also supported by another systematic review that found nonsignificant improvement of ALT and AST favoring LSG over RYGB [28].
Improvement in NAS score was reported in only two studies [18,21].Pedersen et al., 2021 [13] found that both procedures had the same effect on the NAS score without worsening fibrosis 10.In addition, Salman et al. [17] found that LSG significantly decreased the NAS score after one year of follow-up.Our net analysis result showed that LSG significantly improved the NAS score while RYGB did not; however, the analysis included only two studies with a small number of patients; therefore, other studies are needed to confirm the findings.On the contrary, another systematic review and meta-analysis showed nonsignificant differences between both procedures favoring RYGB, which did not align with our findings [28].
The MASLD fibrosis score was reported in only two studies [11,14] and the net result was a slight nonsignificant superiority of LSG over RYGB.On the other hand, Baldwin et al., 2019 [28] found that both surgeries decreased MASLD fibrosis scores without significant differences between them.Despite many studies reporting improvement in fibrosis [12,14,21,23].Pedersen et al., 2021 andNetanel et al., 2020 found that bariatric surgeries were ineffective in improving fibrosis [13,15].In future studies, we recommend further comparing LSG and RYGB and its effectiveness in evaluating all liver enzymes.
The mechanism behind improving NAFLD parameters after LSG and RYGB is not clearly studied in the literature.After these surgeries, nutrient entry through the digestive tract decreases.Additionally, the absorption of fat is decreased, especially after RYBG in comparison to LSG, which results in improving NAFLD clinical outcomes and laboratory markers [29]; however, we found no difference between the two procedures.
The preference for a specific type of bariatric surgery is still unclear, as previous systematic reviews and meta-analyses showed controversial results; however, most of them corroborated our findings.Fakhry et al., 2018 showed that RYGB had greater improvement in the histological outcomes compared to other procedures [30], while other systematic reviews showed no difference between both procedures in improving clinical outcomes in MASLD patients [27,28].These findings could help in choosing the most appropriate type of surgery with the mildest side effects given the similar efficacy results of both procedures.However, the side effects were also comparable in both procedures, except that LSG was associated with shorter operative time than RYGB [31,32]; other side effects were comparable between the two procedures [31,33,34].

Conclusions
Both LSG and RYGB improve MASLDoutcomes one year after surgery, with a slight superiority of LSG over RYGB; however, randomized clinical trials with large samples are needed to confirm these results.

Strength and Limitations
Limited studies are comparing both surgical procedures in NAFLD or NASH patients after one year, which is considered the main strength of the study.However, there are some limitations: all of the included studies are observational studies, which increases the risk of selection bias; there were high levels of heterogeneity in many outcomes that could not be resolved, which also may be due to the presence of both NAFLD and NASH patients in the same population in the included studies.Additionally, some outcomes were reported in only a small number of studies; further studies are required to confirm our findings.Finally, advanced evaluation methods such as MRI fat fraction, biopsy, and elastography were not commonly used in the assessment of NAFLD patients to accurately compare disease activity after follow-up; therefore, we recommend further studies using these methods to monitor the improvement of disease.

Table 2 .
Baseline characteristics of the included studies.
SummaryFourteen studies were included with a total population of 1942 individual who underwent laparoscopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass (RYGB).Most of the studies were prospective cohort studies (nine), while the rest were retrospective.

Table 3 .
Baseline characteristics of study outcomes.
NR: not reported.