This study was a randomized, controlled, open clinical trial. The study was approved by the Ethics Committee of Shanghai Yangpu Shidong Hospital and completed registration with the China Clinical Trial Registry (ChiCTR1900027663, 11/23/2019).

Patients with type 2 diabetes mellitus (T2DM) and MASLD were recruited from January 2024 to March 2024 in our hospital using the Interactive Web Response System (IWRS) in a 1:1:1 ratio, and were randomly divided into three groups. Figure 1 provides a flowchart of the recruitment and allocation of study participants.

The inclusion criteria were as follows: (a) aged between 40 and 70 years, (b) presence of T2DM lasting between 1 and 5 years, (c) Body mass index (BMI)≥18kg/m2, (d) T2DM (HbA1c≥6.5% or fasting glucose≥7.0 mmol/L) (32), (e)The diagnosis of MASLD complies with EASL-EASD-EASO Clinical Practice Guidelines (33). MASLD was assessed according to the degree of echo weakening in the liver by color ultrasound examination (34). Liver ultrasound was performed using the same ultrasonographic method as liver disease ultrasonography, aiming to obtain uniform MASLD stratification and to avoid interobserver variations in findings. A liver ultrasound was performed on each study participant by a Toshiba Xario SSA-660A (Toshiba, Japan) device, equipped with a 5 MHz convex probe.

Exclusion Criteria: For patients who drink a lot of alcohol, the amount of alcohol consumed should equate to the amount of alcohol consumed, which is more than 140g/week for men and more than 70g/week for women; T1DM, special type diabetes mellitus, acute complications of diabetes mellitus; pregnant and lactating women; ethanol-induced or drug-induced liver disease, autoimmune disease, viral hepatitis, cholestasis, and metabolic liver disease, hepatomegaly; total parenteral nutrition; autoimmune liver disease; thyroid disease; rheumatoid immune disease.

Changes in glycated hemoglobin in the two intervention groups supplemented with dietary fiber were (-1.1 ± 0.8) and (-1.6 ± 0.6), respectively (35). According to the PASS 15.0 software calculations, a minimum of 30 participants per group is required, considering the 15% dropout rate. With a power of 80% and a two-sided significance level of 0.05, statistical variance tests were conducted on 98 participants.

To address missing data in the study, this study employed multiple imputation based on the random forest algorithm (Multiple Imputation by Chained Equations, MICE), incorporating age, gender, and HbA1c levels as predictor variables, and generated a total of 50 complete data sets to reflect imputation uncertainty.

After the baseline assessment, patients were randomized to any of the three groups: the control group, the 12g intervention group, or the 24g intervention group, with the help of stand-alone randomization software. To ensure high confidentiality in the randomization process, all randomization operations were performed by a centralized, web-based automated randomization system. To provide a high degree of privacy in the randomization process, all assignments are performed by a centralized, networked, automated randomization system that operates around the clock (24 hours a day). For the specific implementation of random assignment, a static random assignment scheme was used and programmed to operate using the PROC PLAN program in SAS version 9.4 software (provided by SAS Institute Inc). The system relies on a predefined algorithm that generates random sequences with unpredictability, thus ensuring that the distribution of baseline characteristics between the two groups of patients is equalized.

All staff will be divided into three groups based on the needs of the study, and each group will be tasked with guiding the control group, the 12g intervention group, and the 24g intervention group. Strict confidentiality was maintained between the groups, with no disclosure of the specific tasks to each other. In addition, the designers of the study protocol and the staff responsible for the covert grouping and data processing were explicitly excluded from direct participation in the intervention to maintain the objectivity and impartiality of the study.

The control patients received traditional diabetes education. The content covered the basic knowledge of diabetes, dietary management, exercise therapy, drug treatment, and blood glucose monitoring. Each session lasted about one hour, and the teaching team consisted of endocrinologists, dietitians, and nurses.

The intervention group took dietary fiber (DF) supplements (Shanghai Jiuben Technology Co., Ltd.) daily for 12 weeks based on health education. Each packet of the supplement is 50 grams and contains approximately 12 grams of dietary fiber. The nutritional information and composition are shown in Tables 1 , 2 .

The intervention group was randomly assigned to two groups. One group received 24 grams of fiber by consuming two packets, each containing 12 grams of fiber, and the other received 12 grams of fiber by consuming one packet. Take the dietary fiber supplement daily. Add it to 150 ml of warm water, stir well, and drink. Then drink an additional 200 ml of water and eat between breakfast and lunch (10 a.m.).

There were no specific changes in diet (excluding high-fiber packages) or physical activity during the study period. Participants record daily physical activities, including type, duration, and intensity. At the same time, the researchers also telephoned the participants at a set frequency. Throughout the study period, the researchers will also use a diary card to record the participants’ weight data, dietary patterns, bowel movement patterns, medication use, and adverse reactions at regular intervals. It is important to note that all participants were prohibited from using any medication to treat hypoglycemic symptoms or promote weight loss during the intervention.

In the follow-up questionnaire, researchers surveyed participants at 4, 8, and 12 weeks, asking them to report the number of days per week they had supplemented their diet with dietary fiber over the past 12 weeks. Respondents were also asked to keep a daily food diary.

Height, weight, and BMI were measured using a body composition scale (OMRON, HNH-318). Waist circumference (WC) and hip circumference (HC) were measured using a circumference tape. During measurement, the subject stood upright with arms hanging naturally at their sides, legs together, abdomen relaxed and held their breath at the end of exhalation. Waist circumference (WC) was measured at the midpoint of the line connecting the lower edge of the ribs and the upper edge of the iliac crest (typically corresponding to the intersection of the umbilical level line and the mid-axillary line). Hip circumference (HC) was measured by wrapping a soft tape measure horizontally around the most prominent point of the hips, snug against the skin but without compression. The waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) were calculated using the formulas WHR = waist circumference/hip circumference and WHtR = waist circumference/height, respectively.

Various vital indicators were carefully observed and recorded before and after the 12-week treatment interval. The primary outcome indicator was HbA1c levels at 12 weeks post-intervention. Secondary indicators include BMI, WC, HC, WHR, WHtR, FBG, HOMA-IR, Ferretin, the Fibrosis-4 score (FIB-4) level, Liver stiffness, and blood lipid profiles (TG, TC, LDL-C, HDL-C) at 8 and 12 weeks of follow-up.

The blood samples were sent to our laboratory within 30 minutes for uniform testing and analysis using a fully automated biochemistry analyzer (BECKMAN COULTER AU5800).

All blood samples were collected after 12 hours of fasting and allowed to clot for 1–2 hours at room temperature. 1 to 1.5 mL of venous blood specimens are collected in vacuum tubes containing sodium fluoride for measuring plasma glucose, lipids, and HbA1c. The insulin resistance (HOMA-IR) index was calculated as HOMA-IR = fasting insulin × fasting plasma glucose/22.5.

The Fibrosis-4 (FIB-4) index is a noninvasive tool for assessing liver fibrosis (i.e., FIB-4 index = age × AST/[platelet count × (ALT)1/2]). It is easy to use in clinical practice, and its diagnostic ability for advanced liver fibrosis is comparable to magnetic resonance elastography. Transient elastography was used to assess liver stiffness (FibroScan 502 Touch; Echosens). To ensure the accuracy and consistency of MASLD stratification and minimize differences in examination results between observers, a specialist physician with extensive experience in ultrasound diagnosis of liver diseases was specifically assigned to perform the liver ultrasound examination. Specialized probes (such as the M or XL probes from FibroScan) were used, with the appropriate probe selected based on the patient’s body type. The equipment is regularly calibrated to ensure that the measurement error for shear wave propagation velocity is less than 5%.

Data were analyzed using a per-protocol approach. Data were analyzed using SPSS 25.0. Continuous data are expressed as x ̄± s. Multiple group comparisons were analyzed using Analysis of Covariance (ANCOVA). We performed the Shapiro-Wilk test before ANOVA, and the residuals obeyed a normal distribution. Changes in baseline were treated as dependent variables, while group, visit time, and the interaction of group and visit time (group × visit) were treated as independent effects. The Tukey test tested the differences between the groups. Differences were statistically significant at P < 0.05 and significant at P < 0.01.

Figure 1 shows the participant’s flowchart. 114 eligible patients with T2DM combined with MASLD were randomly assigned to the treatment regimen. Of these participants, 98 (86%) completed the 12-week intervention. For the 12-week trial, 5 in the control group, 6 in the Intervention with 12 g, and 5 in the Intervention with 24 g withdrew from the trial, all due to refusal to cooperate. Ultimately, 31 participants in the control group, 33 participants in the Intervention with 12g group, and 34 participants in the Intervention with 24g group completed the trial and were included in the analysis. No significant differences were observed between the three groups ( Table 3 ).

Both interventions showed a more significant decrease in HbA1c levels compared to the control group. Still, only the difference between the control group and the 24g intervention group was statistically significant(-0.6 ± 0.5 vs -1.6 ± 0.6, P<.001) ( Table 4 ).

The 24g intervention group showed a significant decrease in FPG compared to the control group and the 12g intervention group at both week 8 and week 12 (p < 0.05) ( Figure 2B ).

The Group-Time interaction effect of BMI was significantly different. After 8 and 12 weeks, a significant difference in BMI was observed between the control group and the 24g intervention group. No significant difference in BMI was found between the 12g intervention group and the control group ( Figure 2A ).

After 12 weeks, WC (P = 0.034) was lower than baseline in the 24g intervention group ( Table 4 ).

The Group-Time interaction effect (P = 0.014) for HOMA-IR was significant. HOMA-IR levels in the intervention with the 24g group (P<.001) were significantly lower than in the control group after 12 weeks. The intervention with the 12g group showed a slight but not statistically significant improvement in HOMA-IR ( Table 4 ).

The intervention with the 24g group experienced significant reductions in ferritin levels both at week 8 and week 12 (p<0.05). There was no significant difference in ferritin reduction between the control and 12 g intervention groups. ( Figure 2C ).

Significant liver stiffness reductions were observed in the intervention group with 24g (p<0.05), while there were no substantial changes in the other groups. ( Figure 2D ).

Both intervention groups experienced significant reductions in FIB-4 levels (p<0.05), with the intervention group with 24g showing the most significant decrease ( Figure 2E ).

In the intervention with the 24g group, we observed a significant reduction in AST, ALT, and GGT (P<.001) ( Table 4 ).

The Group-Time interaction effect (P = 0.011) for TG was significant. After 12 weeks, there was a significant decrease in TG in the 12g intervention group (P = 0.026) and the 24g intervention group (P<.001) compared to the control group. Both intervention groups showed decreases in TC levels compared to the control group, but the difference was significant only for the 24g intervention group when compared separately (P = 0.012) ( Table 4 ).

Among the participants, the proportions who responded to the follow-up questionnaire and achieved the recommended intake of dietary fiber five or more times per week at each follow-up were 92.7%, 90.4%, and 89.1%, respectively.

During the intervention implementation phase, three participants in the 24g intervention group reported abdominal bloating, which mainly occurred within 1–3 hours after taking dietary fiber. The severity was described as “mild discomfort that was tolerable” and did not affect daily activities. No other severe adverse reactions were recorded in the control group and the 12g intervention group.

Dietary fiber can be classified into two main categories based on its solubility in water: soluble dietary fiber and insoluble dietary fiber (36). Insoluble dietary fiber is predominant in whole grains, such as wheat bran, and constitutes the coarse fiber structure of the outer layer of grains. When mixed with gastric juice, insoluble fiber absorbs water and expands to 3–5 times its original volume, forming a high-viscosity gel-like substance that significantly slows gastric emptying, thereby reducing postprandial glucose (PPG) and insulin levels (37).

Soluble fiber, primarily consisting of β-glucan, pectin, and resistant dextrin, forms a high-viscosity solution in the stomach and small intestine. By coating starch particles and inhibiting the activity of α-glucosidase, it directly slows down the rate of glucose production (38).

The results of our study suggest that increasing dietary fiber intake is beneficial for glycemic control in T2DM. The 24g intervention group achieved superior results in glycemic management compared to the 12g intervention group. Further research suggests that increasing daily dietary fiber intake from 15 grams to 35 grams may be beneficial for HbA1c levels in people with diabetes (39). In an RCT study of MASLD, there was a trend toward less hepatic steatosis in the group consuming 24 g of fiber per day compared to the group consuming 12 g of fiber per day (p = 0.07) (40).

The mechanism may be that a high-fiber diet helps promote intestinal peristalsis, maintains intestinal ecological balance, lowers postprandial blood glucose levels, improves insulin sensitivity, and reduces blood glucose fluctuations (38). Zhao and other researchers have revealed that the gut microbiota can provide additional health benefits to patients with T2DM through carbohydrate fermentation and the production of short-chain fatty acids (SCFAs) (41). In another systematic review, foods rich in soluble fiber, such as β-glucan, improved blood glucose levels in diabetic patients (42). Moreover, a high intake of dietary fiber promotes satiety. It reduces the intake of high-energy foods, thereby reducing the risk of overweight/obesity, an estimated risk factor for T2DM (30).

A meta-analysis reported a significant dose-response relationship between body mass index and MASLD risk (43). In our study, we found a dose-dependent effect of dietary fiber on improving BMI. Although the mechanisms underlying the association between obesity and MASLD are not fully understood, the adipose tissue (AT) distensibility hypothesis could link obesity to the development of MASLD (44). Obesity, a growing health problem, is emerging as a significant challenge worldwide. Unhealthy eating habits are an essential factor in the development of obesity. With the accelerated pace of life and changes in dietary patterns, people are increasingly inclined to choose foods high in calories, fat, and sugar. Dietary fiber can have a positive effect on the prevention and treatment of obesity through a variety of mechanisms, including increasing satiety, lowering the glycemic index, promoting intestinal health, regulating fat metabolism, and lowering cholesterol (45). Studies have shown that increased intake of vegetables, fruits, legumes, and diets high in soluble fiber can reduce body weight and body fat percentage (46). In our study, both groups in the dietary fiber intervention had a significant BMI reduction compared to the control group. Still, the difference in BMI between the 12g intervention group and the 24g intervention group was insignificant.

This study did not systematically record participants’ energy and nutrient intake during the intervention period, making it impossible to directly assess the contribution of dietary structure changes to the intervention’s effectiveness. Participants may have spontaneously reduced their intake of high-calorie foods due to increased health awareness resulting from the intervention, and this behavioral change was not captured in this study, leading to an overestimation of the independent effect of the high dietary fiber cereal meal intervention.

One study showed that 23.76 grams of combined dietary fiber content per day resulted in a significant decrease in total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) compared to a conventional diet (47). This study is similar to our findings. This result may be due to the key role played by the viscous properties of DF, which delay gastric emptying, meaning that food stays in the stomach longer, thus slowing down digestion and absorption. In addition, DF inhibits the transit of triglycerides and cholesterol through the intestines. This inhibition helps reduce the amount of these lipids absorbed into the bloodstream by the small intestine, lowering blood lipid levels (48).

A cross-sectional study of 265 healthy adults and animal experiments found that individuals in the highest quartile of vegetable intake had lower rates of ALT abnormalities. Additionally, supplementation with insoluble/soluble dietary fiber significantly reduced serum ALT/AST levels, and increased serum albumin and total protein levels (49). Another study further confirmed that dietary fiber intake was positively correlated with improvements in MASLD biochemical indicators (50). These are similar to our research findings.

The mechanism by which dietary fiber affects MASLD may be that its well-known role in modulating the gut microbiota is an integral part of the gut barrier (51). When the gut barrier is compromised, the liver is the first parenchymal organ to be exposed to bacteria and bacterial products, which can lead to inflammation and liver damage (52). Second, Short-Chain Fatty Acids (SCFAs) produced by the gut microbiota fermenting dietary fiber prevent MASLD (53). In addition, by modulating the gut microbiota and metabolite production, dietary fiber may inhibit the expression of inflammatory factors in the liver and reduce the inflammatory response (54). This improvement may be associated with fiber, oligofructose, inulin, and oligosaccharides. Oligofructose, inulin, and oligo-xylulose selectively promote the growth of bifidobacteria (55). The meta-analysis findings revealed statistically significant correlations between MASLD and C-reactive protein (CRP), interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and intercellular adhesion molecule-1 (ICAM-1), with corresponding odds ratios (ORs) of 1.41, 1.08, 1.50, 1.15, and 2.17, respectively (56). In this study, participants in the 24g intervention group demonstrated a statistically significant decrease in ferritin levels at both week 8 and week 12 of the intervention period (p<0.05).

Compared with traditional dietary approaches, bagged fiber foods have significant advantages in diabetes-combined MASLD populations and effectively boost daily fiber intake. First, the convenience of fiber-in-bag foods is one of their major highlights. In addition, the fiber in fiber-in-a-bag foods is derived from various high-quality grains, making them more nutritious and diverse.

In real life, consuming 24 grams of dietary fiber daily is feasible; however, individual differences and nutritional adjustments should be taken into account. Maintaining this intake level over the long term can provide health benefits, but excessive intake should be avoided to prevent potential risks. Excessive fiber intake (>50 grams/day) may reduce the absorption of minerals such as iron, zinc, and calcium, but the risk is low at 24 grams. Sudden increases in fiber intake may cause bloating or diarrhea, and adequate hydration can help alleviate these symptoms (57).

In summary, long-term dietary fiber intake should be adjusted according to individual health conditions, age, and special needs in terms of intake methods and food choices.