Impacts of Ramadan Fasting on Metabolic and Hepatic Endpoints in Individuals With Metabolic Dysfunction‐Associated Fatty Liver Disease: A Systematic Review and Meta‐Analysis

ABSTRACT

Objective

Previous studies examining the impact of Ramadan fasting on patients with metabolic dysfunction‐associated fatty liver disease (MAFLD) have yielded mixed results. Therefore, assessing the health benefits of such fasting in patients with MAFLD through a systematic review and meta‐analysis (SR/MA) is important.

Methods

A systematic search was conducted on MEDLINE, Scopus, Web of Science, Google Scholar, and ClinicalTrials.gov from their inception to March 5, 2025, to identify relevant studies involving adults with MAFLD fasting during Ramadan. The primary outcome was liver‐related parameters, while additional outcomes included changes in anthropometric and metabolic parameters during the peri‐Ramadan period. Statistical analysis was performed using R software, and the results were presented as mean differences (MD) with 95% confidence intervals (CI).

Results

Eight studies (10 reports), mostly with serious risk of bias, involving 603 subjects revealed that Ramadan fasting was associated with reductions in alanine transaminase (ALT [MD −4.11 U/L]), aspartate transaminase (AST [MD −4.24 U/L]), FIB4 index (MD −0.09), and nonalcoholic fatty liver disease fibrosis score (MD −0.22) from baseline in fasting individuals. However, the changes in ALT and AST were similar in studies comparing fasting and non‐fasting groups. Fasting individuals experienced significantly greater weight loss (MD −1.44 kg), as well as reductions in body mass index (MD −0.66 kg/m²) and waist circumference (MD −0.91 cm), compared to those who did not fast. Individuals who fasted experienced a glycemic benefit characterized by a reduction in glycated hemoglobin (MD −0.4%). However, changes in mean body fat percentage and HOMA‐IR were similar in both the fasting and non‐fasting groups. Individuals who fasted experienced reductions in both systolic and diastolic blood pressure, along with improved lipid parameters.

Conclusion

This SR/MA of small existing data suggests that fasting during Ramadan improves certain MAFLD‐related outcomes. Larger, multinational studies with wider global representation are needed to improve clinical practice decisions.

1. Introduction

Major world religions have employed fasting as a means of expressing dedication to their faith; among them, fasting during Ramadan is rigorously observed by millions of Muslims worldwide [1]. The effects of fasting during Ramadan have been evaluated across various metabolic disorders. A major metabolic disorder today is nonalcoholic fatty liver disease (NAFLD), which has recently been renamed to metabolic dysfunction‐associated fatty liver disease (MAFLD). This condition encompasses a spectrum of increased fat accumulation in the liver, ranging from simple steatosis to steatohepatitis, which can progress to cirrhosis and cancer [2]. The complex interplay between environmental and genetic factors influences the development of MAFLD. The prevalence of MAFLD is rapidly increasing, with estimates suggesting that up to 25% of the adult population worldwide is affected. This condition is more common among men and those who are overweight or obese [3]. Lifestyle modification remains the cornerstone of both preventing and managing MAFLD. On March 14, 2024, the US FDA approved resmetirom for adults with non‐cirrhotic metabolic dysfunction‐associated steatohepatitis (MASH) and moderate to advanced liver fibrosis, as an adjunct to diet and exercise [4]. Intermittent fasting offers benefits similar to those of caloric restriction in terms of body weight management, enhancements in glucose homeostasis and lipid profiles, and anti‐inflammatory effects [5]. Ramadan fasting, a type of intermittent fasting (IF), involves an extended fasting period from sunrise to sunset for a month, followed by a shorter eating period. Ramadan fasting offers several health benefits, including reductions in weight, blood pressure, and blood glucose levels, as well as improvements in lipid parameters [1]. A meta‐analysis of 20 studies involving 601 healthy adults who practiced Ramadan diurnal IF has shown significant, albeit small to medium, positive effects on liver function, suggesting that this approach may offer temporary, short‐term protection against fatty liver disease for these individuals [6].

Clinical evidence for IF in MAFLD is limited; however, short‐term clinical trials have reported improvements in insulin sensitivity, glucose homeostasis, lipid metabolism, hepatic steatosis, autophagy, oxidative stress, and inflammation, as well as modulation of the gut microbiota [7, 8]. Several observational studies have reported modest but variable impacts of IF, including Ramadan fasting, on anthropometric, hepatic, and metabolic parameters in patients with MAFLD [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21]. Lin et al. recently published a systematic review of six studies involving 397 patients with NAFLD, noting that Ramadan fasting may benefit those with NAFLD by improving body weight, body composition, cardiometabolic risk factors, glucose profiles, liver parameters, and inflammation markers [22]. However, their systematic review has not encompassed all available studies. Furthermore, they have not performed meta‐analyses to determine the pooled effect size of changes in various outcomes. Given this background, we conducted a systematic review and meta‐analysis (SR/MA) to thoroughly evaluate the effects of Ramadan fasting on anthropometric and metabolic indicators, liver function tests, and various scores for assessing MAFLD. This study examines existing evidence to fill the knowledge gap in this field and provides evidence‐based insights that could aid patients with MAFLD who engage in fasting.

2. Methods

2.1. Ethical Compliance

This SR/MA was conducted following the procedures outlined in the Cochrane Handbook for Systematic Reviews of Interventions and is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) checklist [23, 24]. It has been registered with PROSPERO (CRD420251019498), and the protocol summary is accessible online.

2.2. Search Strategy

A systematic search was performed across several databases and registries, including MEDLINE (via PubMed), Scopus, Web of Science, Google Scholar, and ClinicalTrials.gov. This search spanned from the inception of each database to March 5, 2025. Using the Boolean operators “AND” and “OR,” the following terms were searched: “nonalcoholic fatty liver disease,” “nonalcoholic steatohepatitis,” “fatty liver,” “metabolic dysfunction‐associated steatotic liver disease,” “metabolic dysfunction‐associated fatty liver disease,” “Ramadan fasting,” “religious fasting,” “Ramadan intermittent fasting.” The search terms were applied to all fields. The aim was to identify both recently published and unpublished studies in English. Additionally, the search included reviewing references within the published articles retrieved for this study and relevant journals.

2.3. Study Selection

The Population, Intervention, Comparison, Outcomes, and Study (PICOS) design served as a framework for developing eligibility criteria for studies in this SR/MA. The patient population (P) consisted of adults of either sex with MAFLD; the intervention (I) involved fasting during Ramadan; the comparison or control (C) group (if applicable) included individuals with MAFLD who were not fasting during Ramadan; the outcomes (O) centered on metabolic and hepatic endpoints; and prospective observational/intervention studies or randomized controlled trials (RCTs) were deemed the study type (S) for inclusion. The single‐arm studies that included patients with MAFLD who fasted during Ramadan, along with comparative studies involving these patients who fasted and those who did not, were considered. We excluded editorials, commentaries, conference abstracts, case reports, animal studies, studies involving healthy individuals, and studies that do not report the outcomes of interest. The study selection process involved four independent review authors who, after removing duplicates, evaluated the titles and abstracts of all identified records based on the predefined inclusion and exclusion criteria. This initial screening eliminated studies that were clearly irrelevant, allowing only those that potentially met the criteria to proceed to the next stage. Once all the review authors reached a consensus, the full texts of the remaining articles were retrieved and reviewed independently by the authors to determine eligibility in greater detail. During the full‐text screening, the reasons for exclusion were recorded for each article, as reported in the PRISMA flow diagram. Any disagreements during the screening or full‐text review were resolved through consensus.

2.4. Outcomes Analyzed

The primary outcomes of interest were the changes from baseline (CFB) to the end of the study in hepatic parameters, which included serum alanine transaminase (ALT), aspartate transaminase (AST), the fibrosis 4 (FIB4) index, and the NAFLD fibrosis score (NFS). The secondary outcomes included changes in anthropometric and metabolic parameters such as body weight, body mass index (BMI), waist circumference (WC), body fat percentage, systolic and diastolic blood pressure (BP), fasting plasma glucose (FPG), glycated hemoglobin (HbA1c), homeostasis model assessment‐estimated insulin resistance (HOMA‐IR), and lipid profile. For studies lacking a non‐fasting control group, the pooled means of the CFB for the variables were calculated. The mean differences (MDs) of the CFB for the variables between the fasting and non‐fasting groups were computed for the comparative studies.

2.5. Data Extraction

Three review authors independently conducted data extraction using standardized forms. The results were gathered by identifying various publications from one study group, with pertinent data from each article incorporated into the analyses. The data retrieved for all eligible studies included in the review are: first author, publication year, country of the study, study design, main inclusion criteria for subjects, sample size, number of male and female participants, mean age, and the hepatic, anthropometric, and metabolic outcomes mentioned above. Disagreements were settled through consensus.

2.6. Dealing With Missing Data

The necessary supplementary files for the articles were obtained from the journals' websites. Furthermore, additional information was collected from the corresponding authors of the relevant articles via email communication. All relevant data collected in this manner were carefully integrated into the meta‐analysis. Furthermore, attrition rates—including dropouts, losses to follow‐up, and withdrawals—were thoroughly examined.

2.7. Statistical Analysis

Statistical analyses were conducted using the RStudio software environment (R version 4.4.2, released on October 31, 2024), specifically utilizing the “meta” and “metafor” packages, which are designed for performing meta‐analyses in R [25]. For continuous variables, outcomes were presented as mean differences (MD) with 95% confidence intervals (CI). For the single‐arm studies involving only fasting individuals, the mean changes from baseline values of the continuous outcomes were pooled to generate the effect sizes. For studies with a non‐fasting control group, the MDs for the changes from baseline values of the outcomes in the fasting versus non‐fasting groups were calculated. The random‐effects model was selected to address the expected heterogeneity arising from population characteristics in the included studies. The inverse variance statistical method was utilized for all instances. The details of the meta‐analysis included the use of the restricted maximum‐likelihood estimator for tau², the Q‐profile method for the confidence interval of tau² and tau, and the calculation of I ² based on Q statistics using untransformed (raw) means. A significance level of p < 0.05 was used.

2.8. Assessment of the Risk of Bias

The risk of bias (RoB) in the included studies was independently assessed by two authors using the Risk Of Bias In Non‐randomized Studies—of Interventions, version 2 (ROBINS‐I V2) assessment tool [26]. The details of the RoB assessment using the ROBINS‐I V2 have been discussed in our previously published article [27]. The Risk‐Of‐Bias VISualization (robvis) web application was utilized to create risk‐of‐bias plots [28]. Publication bias, when appropriate (at least 10 studies in a forest plot), was assessed using funnel plots generated in R software [29].

2.9. Assessment of Heterogeneity

The assessment of heterogeneity was initially conducted by studying forest plots. Subsequently, a Chi² test was performed using N−1 degrees of freedom and a significance level of 0.05 to determine statistical significance. The I ² test was also employed in the subsequent analysis. Thresholds for I ² values were set at 25% for low heterogeneity, 50% for moderate heterogeneity, and 75% for high heterogeneity [30].

3. Results

3.1. Search Results

Figure 1 presents the PRISMA flow diagram outlining the steps in the study selection process. The initial search identified 4903 articles, which were narrowed down to 10 after screening titles and abstracts and conducting subsequent full‐text reviews. Ultimately, eight studies comprising 10 reports involving 603 subjects who met all the prespecified criteria were included in this SR/MA [11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. Two studies were excluded: one involved animals, while the other included patients with MAFLD and chronic liver disease from other causes [31, 32].

FIGURE 1.

Flowchart on study retrieval and inclusion in the meta‐analysis.

3.2. Characteristics of Included and Excluded Studies

Tables 1 and S1 provide details on the included and excluded studies, respectively. Among the eight studies included, seven were prospective observational [11, 12, 13, 14, 15, 16, 17, 18, 20], while one was a retrospective case–control study [19]. Four studies included non‐fasting control groups of patients with MAFLD who opted not to fast during Ramadan [11, 16, 17, 19, 20]. The two groups in these comparative studies shared the same baseline characteristics. Arabi et al. reported outcome results for males and females separately [12, 13], with analysis labeled as Arabi 2015 (m) for males and Arabi 2015 (f) for females. Aliasghari et al. presented body fat percentage data for males and females separately, denoted as Aliasghari 2017 (m) and Aliasghari 2017 (f) [11].

TABLE 1.

Baseline characteristics of individual studies and study participants included in the meta‐analysis.

Study ID [Reference], Study place	Study design	Major inclusion criteria	Groups	N	Sex (n)	Age (years)	Body weight (kg); BMI (kg/m2)	ALT (U/L)	AST (U/L)	FPG (mg/dL); HbA1c (%)	Study duration
Aliasghari et al. [11], Iran	Prospective observational	Age 20–50 years, documented NAFLD, Fasting > 20 days. Excluded: DM, HTN	Fasting	42	M 25, F 17	37.6 ± 7.1	83.6 ± 13; 30.1 ± 4.5	NR	NR	94.0 ± 8.0; NR	3 days before to 3 days after Ramadan
			Nonfasting	41	M 32, F 9	35.8 ± 7.3	80.8 ± 10; 28.2 ± 2.5	NR	NR	94.7 ± 7.4; NR
Arabi et al. [12], Arabi et al. [13], Iran	Prospective observational	Age 18–65 years, diagnosed NAFLD by USG. Excluded: BMI < 25, Fasted < 10 days	Fasting males and females	50	M: 33, F: 17	40.5 ± 10.9	M: 29.4 ± 3.7; NR F: 33.6 ± 7.1; NR	M: 17.5 ± 9.7, F: 18.7 ± 17.8	M: 26.2 ± 8.9, F: 24.5 ± 14.1	M: 86.0 ± 9.7; NR F: 96.7 ± 12.1; NR	7 days before to 6 days after Ramadan
Badran et al. [14], Iran	Prospective observational	Age 18–70 years, diagnosed NAFLD by USG	Fasting	98	M: 23, F: 75	18–70 years	97.4 ± 16.4; 37.0 ± 6.6	36.5 ± 20.2	38.5 ± 19.6	130.2 ± 64.1; 6.1 ± 1.1	2 days before to 1 month after Ramadan
Dündar et al. [15], Turkey	Prospective observational	Age ≥ 18 years, Diagnosed hepatic steatosis by MRI, BMI ≥ 25 kg/m2	Fasting	34	M: 28, F: 6	44.5 ± 10.8	86.8 ± 12.8; 30.3 ± 3.4	26.1 ± 13.5	23.0 ± 10.3	NR	30–34 days, peri‐Ramadan
Ebrahimi et al. [16], Ebrahimi et al. [17], Iran	Prospective observational	Age 20–50 years, documented NAFLD by liver biopsy, fasting > 20 days. Excluded: DM.	Fasting	42	M: 25, F: 17	37.6 ± 7.1	83.6 ± 13.0; 30.1 ± 4.5	NR	NR	NR	3 days before to 3 days after Ramadan
			Nonfasting	41	M: 32, F: 9	35.8 ± 7.3	80.8 ± 10.0; 28.2 ± 2.5	NR	NR	NR
Gad et al. [18], Egypt	Prospective observational	Age 18–65 years, diagnosed NAFLD by USG and FibroScan. Excluded: Fasting < 20 days, BMI < 25 kg/m2, uncontrolled DM.	Fasting	40	M: 27, F: 13	46 ± 9	NR; 30.9 ± 2.4	45.2 ± 7.7	35.9 ± 4.3	108.3 ± 11.7; 7.3 ± 0.6	7 days before to 7 days after Ramadan
Mari et al. [19], Israel	Retrospective, case–control	Adults with NAFLD diagnosed by USG	Fasting	74	M: 39, F: 35	51.8 ± 20.9	NR; 36.7 ± 7.1	51.4 ± 9.4	44.2 ± 12.8	NR; 5.9 ± 0.6	Before Ramadan to after Ramadan
			Nonfasting	81	M: 42, F: 39	52.6 ± 19.3	NR; 34.3 ± 6.3	53.2 ± 11.8	43.8 ± 10.6	NR; 5.8 ± 0.8
Rahimi et al. [20], Iran	Prospective observational	Adults with NAFLD diagnosed by a gastroenterologist	Fasting	34	M: 25, F: 9	46.0 ± 111.7	88.3 ± 19.8; 29.5 ± 4.5	34.6 ± 13.5	NR	NR	Before Ramadan to after Ramadan
			Nonfasting	26	M: 14, F: 12	49.6 ± 11.0	83.9 ± 10.4; 30.1 ± 4.1	34.9 ± 12.3	NR	NR

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; DM, diabetes mellitus; FPG, fasting plasma glucose; HbA1c, glycated hemoglobin; HTN, hypertension; MRI, magnetic resonance imaging; NAFLD, non‐alcoholic fatty liver disease; NR, not reported; SD, standard deviation; USG, ultrasonography.

3.3. Risk of Bias in the Included Studies

Figure S1 depicts RoB across the eight studies assessed using ROBINS‐I V2 included in the meta‐analysis. The overall RoB was serious in all but one study; this bias stemmed from confounding factors.

3.4. Hepatic Outcomes

In fasting individuals, the pooled mean change in ALT (seven studies, n = 372) was −4.11 U/L (95% CI [−8.03, −0.19], I ² = 94.6% [high heterogeneity]) (Figure 2A‐1). In comparative studies (three studies, n = 298), the fasting and non‐fasting groups achieved comparable changes in ALT (MD −3.08 U/L, 95% CI [−13.75, 7.59], I ² = 94.7% [high heterogeneity], p = 0.5721) (Figure 2A‐2). The pooled mean change in AST among fasting individuals (six studies, n = 338) was −4.24 U/L (95% CI [−6.81, −1.67], I ² = 84.1% [high heterogeneity]) (Figure 2B‐1). The fasting and non‐fasting groups achieved comparable changes in AST (MD −5.58 U/L, 95% CI [−12.28, 1.13], I ² = 91.3% [high heterogeneity], p = 0.1031) in comparative studies (two studies, n = 238) (Figure 2A‐2). The pooled mean change among fasting individuals in the FIB4 index (two studies, n = 138) and NFS (two studies, n = 114) were −0.09 (95% CI [−0.17, −0.02], I ² = 9.5% [low heterogeneity]) (Figure 2C) and −0.22 (95% CI [−0.27, −0.16], I ² = 0% [low heterogeneity]) (Figure 2D), respectively. Badran et al. reported a significant reduction in the aspartate aminotransferase to platelet ratio index (APRI) score among fasters [14]. Gad et al. reported that fasting during Ramadan could improve liver steatosis in patients with MAFLD, as evidenced by a significant reduction in the controlled attenuation parameter (CAP) and liver stiffness measurement (LSM) [18]. Mari et al. reported an improvement in the BARD score for the fasting subjects [19]. Gad et al. reported no change in the ultrasonographic grading of fatty liver [18]; however, such changes were significant in Badran et al.'s study [14]. Badran et al. found no substantial reduction in the MAFLD fibrosis stage measured by LSM [14].

FIGURE 2.

Forest plot demonstrating the (A‐1) pooled mean change in ALT levels in the fasting individuals; (A‐2) mean difference in changes from baseline in ALT levels in the fasting versus non‐fasting group; (B‐1) pooled mean change in AST levels in the fasting individuals; (B‐2) mean difference in changes from baseline in AST levels in the fasting versus non‐fasting group; (C) pooled mean change in FIB4 index in the fasting individuals; (D) pooled mean change in NFS in the fasting individuals.

3.5. Anthropometric Outcomes

The pooled mean change in body weight among fasting individuals (four studies, n = 216) was −1.67 kg (95% CI [−2.48, −0.86], I ² = 64.1% [moderate heterogeneity]) (Figure 3A‐1). The fasting group achieved greater weight loss than the non‐fasting group (MD −1.44 kg, 95% CI [−2.76, −0.12], I ² = 78.4% [high heterogeneity], p = 0.0327) in comparative studies (three studies, n = 226) (Figure 3A‐2). In fasting individuals (seven studies, n = 380), the pooled mean change in BMI was −0.80 kg/m² (95% CI [−1.18, −0.43], I ² = 67.3% [moderate heterogeneity]) (Figure 3B‐1). In comparative studies (four studies, n = 381), the reduction in BMI was greater in the fasting than the non‐fasting group (MD −0.66 kg/m² (95% CI [−1.32, −0.00], I ² = 83.6% [high heterogeneity], p = 0.0496)) (Figure 3B‐2). The pooled mean change in WC among fasting individuals (five studies, n = 232) was −0.95 cm (95% CI [−1.10, −0.79], I ² = 0% [low heterogeneity]) (Figure 3C‐1). The fasting group achieved a greater reduction in WC than the non‐fasting group (MD −0.91 cm, 95% CI [−1.10, −0.72], I ² = 0% [low heterogeneity], p < 0.0001) in comparative studies (two studies, n = 166) (Figure 3C‐2). In fasting individuals (three studies, n = 134), the pooled mean change in body fat percentage was −0.68% (95% CI [−0.86, −0.49], I ² = 0% [low heterogeneity]) (Figure 3D‐1). In comparative studies (two studies, n = 164), the reduction in body fat percentage was comparable in the fasting and the non‐fasting groups (MD −0.39% (95% CI [−0.81, 0.02], I ² = 0% [low heterogeneity], p = 0.0638)) (Figure 3D‐2).

FIGURE 3.

Forest plot demonstrating the (A‐1) pooled mean change in body weight in the fasting individuals; (A‐2) mean difference in changes from baseline in body weight in the fasting versus non‐fasting group; (B‐1) pooled mean change in BMI in the fasting individuals; (B‐2) mean difference in changes from baseline in BMI in the fasting versus non‐fasting group; (C‐1) pooled mean change in waist circumference in the fasting individuals; (C‐2) mean difference in changes from baseline in waist circumference in the fasting versus non‐fasting group; (D‐1) pooled mean change in body fat percentage in the fasting individuals; (D‐2) mean difference in changes from baseline in body fat percentage in the fasting versus non‐fasting group.

3.6. Glycemic Outcomes

The pooled mean change among fasting individuals in FPG (four studies, n = 230) and HbA1c (three studies, n = 212) was 6.71 mg/dL (95% CI [−5.95, 19.97], I² = 97.7% [high heterogeneity]) (Figure 4A) and −0.40% (95% CI [−0.69, −0.12], I ² = 78.9% [high heterogeneity]) (Figure 4B), respectively. In fasting individuals (three studies, n = 214), the pooled mean change in HOMA‐IR was −0.31 (95% CI [−0.76, 0.14], I ² = 90.1% [high heterogeneity]) (Figure 4C‐1). In comparative studies (two studies, n = 238), the reduction in HOMA‐IR was comparable in the fasting and the non‐fasting groups (MD −0.39 (95% CI [−1.08, 0.31], I ² = 88.1% [high heterogeneity], p = 0.2755)) (Figure 4C‐2).

FIGURE 4.

Forest plot demonstrating the (A) pooled mean change in FPG levels in the fasting individuals; (B) pooled mean change in HbA1c levels in the fasting individuals; (C‐1) pooled mean change in HOMA‐IR in the fasting individuals; (C‐2) mean difference in changes from baseline in HOMA‐IR in the fasting versus non‐fasting group; (D) pooled mean change in SBP in the fasting individuals; (E) pooled mean change in DBP in the fasting individuals; (F) pooled mean change in TC in the fasting individuals; (G) pooled mean change in TG in the fasting individuals; (H) pooled mean change in LDL‐C in the fasting individuals; (I) pooled mean change in HDL‐C in the fasting individuals.

3.7. Cardiovascular Outcomes

The pooled mean changes among fasting individuals for the cardiovascular outcomes were as follows: SBP (two studies, n = 92): −2.28 mmHg (95% CI [−9.72, 5.17], I ² = 73.0% [moderate heterogeneity]) (Figure 4D), DBP (two studies, n = 92): −1.82 mmHg (95% CI [−7.77, 4.13], I ² = 90.4% [high heterogeneity]) (Figure 4E), TC (five studies, n = 264): 0.50 mg/dL (95% CI [−18.69, 19.69], I ² = 93.0% [high heterogeneity]) (Figure 4F), TG (five studies, n = 264): −6.20 mg/dL (95% CI [−42.87, 30.47], I ² = 88.4% [high heterogeneity]) (Figure 4G), LDL‐C (five studies, n = 264): −5.65 mg/dL (95% CI [−25.69, 14.39], I ² = 95.8% [high heterogeneity]) (Figure 4H), and HDL‐C (five studies, n = 264): 1.88 mg/dL (95% CI [−1.01, 4.77], I ² = 91.9% [high heterogeneity]) (Figure 4I).

3.8. Other Outcomes

In Aliasghari et al.'s study, the post‐Ramadan level of high‐sensitivity C‐reactive protein (hs‐CRP) among fasting individuals was significantly lower than the pre‐Ramadan level; the reduction of hs‐CRP was greater in the fasting group than in the non‐fasting group [11]. Furthermore, studies by Badran et al. [13] and Mari et al. [19] found that Ramadan fasting significantly reduced CRP levels. Aliasghari et al. reported a greater reduction of IL‐6 in the fasting group compared to the non‐fasting group [11]. Arabi et al. observed no change in fat‐free mass among fasting individuals [12]. In fasting individuals, the pre‐ and post‐Ramadan plasma free fatty acid levels remained identical, except for a decrease in the elaidic acid levels after Ramadan [13]. Ebrahimi et al. reported that fasting during Ramadan decreased adipokine levels, including vaspin and omentin‐1 [16]. In fasting individuals, liver volume and fat fractions did not change after Ramadan, as reported by Dündar et al. [15].

4. Discussion

This SR/MA included eight studies comprising 10 reports, mostly with serious RoB, that involved 603 subjects. This provides an update on the effect of Ramadan fasting on hepatic and other cardiometabolic outcomes in individuals with MAFLD who fast during Ramadan. This study revealed that fasting was associated with reductions in ALT and AST levels among those who fasted during Ramadan. However, the changes in ALT and AST levels were not significantly different between the fasting and non‐fasting groups. The FIB4 index and NFS also decreased among the fasting individuals. In comparison to non‐fasting individuals, the reductions in body weight, BMI, and WC were more pronounced among those who fasted. However, the decrease in body fat percentage was similar between the two groups. Fasting individuals experienced reduced HbA1c and improved lipid levels.

Ramadan fasting often resembles a form of time‐restricted eating (TRE) behavior, as in many countries, the fasting period extends beyond 12 h due to the shorter daylight hours in those regions [33]. After 12 h of fasting, the body's glycogen stores are mostly depleted, prompting the mobilization of body fat to supply a continuous glucose source to tissues, including the nervous system. Ketogenesis also occurs, representing crucial mechanisms through which TRE delivers health benefits [34]. The improvements in ALT, AST, FIB4 index, and NFS score that we observed may be explained by the mechanism described above. The absence of statistically significant changes in these parameters between fasting and non‐fasting comparators in our review could be attributable to the small sample sizes of the studies included.

Notable improvements in anthropometric outcomes, including reductions in body weight, BMI, and WC, are observed in the fasting group overall and when compared to the non‐fasting group. These parameters have significant clinical implications for metabolic and liver health. Various lifestyle interventions, such as TRE, IF, and dietary modifications, have been shown in previous studies to enhance metabolic profiles, reduce inflammation, and improve liver outcomes in cases of MAFLD [35, 36]. Consequently, the decrease in body fat percentage observed in our review is expected to have a significant clinical impact on patients with MAFLD, as indicated by the positive changes in liver parameters.

Previous studies have shown that various fasting modalities, including IF and TRE, are linked to improvements in metabolic outcomes such as HbA1c, FPG, and HOMA‐IR [37, 38]. Even with the short duration of Ramadan fasting, our review showed a statistically significant reduction in HbA1c (−0.4%); however, other metabolic outcomes, such as changes in FPG and HOMA‐IR, did not reach statistical significance, presumably due to the small sample sizes of the studies included. Similarly, the cardiometabolic benefits, including improvements in BP and lipid profile, noted in previous studies associated with fasting, were not evident in our review, likely due to the small sample size of the study [39, 40, 41].

We acknowledge the following limitations to our study. (1) For the evaluation of a common metabolic disorder like MAFLD, the sample sizes of the included studies are quite small, which may have negatively influenced the outcomes we observed. (2) Ramadan fasting is a holy ritual practiced by religious Muslims around the globe, with varying daily fasting durations due to significant differences in daylight hours across different time zones. As a result, the applicability of our findings may be limited for global clinical practice recommendations. (3) The serious RoB in the included studies and the high heterogeneity of several analyzed parameters may have diminished the quality of the outcomes we reported. (4) All studies included in this SR/MA were observational studies. A major limitation of a meta‐analysis that uses only observational studies is the increased risk of bias and confounding, which can distort effect estimates. This may lead to unreliable conclusions about the relationship between an intervention and an outcome. Despite these limitations, this meta‐analysis is the first to comprehensively assess the impact of Ramadan fasting on the clinical and biochemical outcomes of MAFLD. Additionally, we could also assess the impact of Ramadan fasting on anthropometric and cardiometabolic outcomes, which are closely intertwined with the clinical and prognostic courses of patients suffering from this serious disease that has negatively impacted global healthcare budgets in recent years.

5. Conclusion

Fasting during Ramadan improves liver parameters, including ALT, AST, the FIB4 index, and the NFS, in patients with MAFLD. Modest improvements in anthropometric outcomes, such as body weight, BMI, WC, and body fat percentage, represent additional potential health benefits associated with fasting during Ramadan. Fasting may also enhance metabolic profile. Larger multinational RCTs and observational studies with broader global representation are needed to better inform clinical practice recommendations regarding the role of Ramadan fasting among patients with MAFLD.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Table S1:. Characteristics of the excluded cross‐sectional studies and study participants.

Figure S1:. (A) Risk of bias summary: review authors' judgments about each risk of bias item for each included study; (B) Risk of bias graph: review authors' judgments about each risk of bias item presented as percentages across all included studies.

Kamrul‐Hasan A. B. M., Ashraf H., Nagendra L., et al., “Impacts of Ramadan Fasting on Metabolic and Hepatic Endpoints in Individuals With Metabolic Dysfunction‐Associated Fatty Liver Disease: A Systematic Review and Meta‐Analysis,” JGH Open 9, no. 8 (2025): e70264, 10.1002/jgh3.70264.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.