Introduction

Type 2 diabetes and non-alcoholic fatty liver disease (NAFLD) represent a significant global public health concern.1 2 It has been estimated that the global prevalence of NAFLD in patients with type 2 diabetes is 55.48%, and 64.36% of patients with obesity with type 2 diabetes have NAFLD.3 Type 2 diabetes and NAFLD frequently coexist, with a bidirectional, interactive relationship between the two disorders.4 Insulin resistance, increased hepatic de-novo lipogenesis, abnormal triglyceride metabolism and gut microbiota dysbiosis are common mechanisms of pathogenesis between type 2 diabetes and NAFLD.5 Excessive hepatic lipid accumulation in NAFLD causes insulin resistance by impairing the insulin signalling and altering hepatokines secretion, which increases the risk of type 2 diabetes.6–8 Insulin resistance in type 2 diabetes increases lipolysis in white adipose tissue, resulting in hepatic fatty acid influx and triglyceride formation and renders some major hepatokines involved in lipid homeostasis dysfunctional, thus aggravating the development of NAFLD.8–10 Patients with type 2 diabetes and NAFLD have poorer metabolic profiles, and the two disorders are mutually exacerbated, accelerating the progression of organ damage and increasing cardiovascular risk. In addition, hepatic decompensation and hepatocellular carcinoma occur more frequently among individuals with type 2 diabetes and NAFLD. As a result, people with type 2 diabetes and NAFLD are more likely to present a higher disability and mortality rate compared with those with only NAFLD or type 2 diabetes alone.11–13 Therefore, the prevention and treatment of type 2 diabetes and NAFLD comorbidities are urgently needed.

Targeting liver fat is a pivotal therapeutic avenue for the treatment of type 2 diabetes and NAFLD comorbidities. NAFLD, whether a comorbidity of type 2 diabetes or a potential high-risk factor for progression to diabetic complications, remains underevaluated, under-recognised and undertreated. Currently, there are no approved pharmacological therapies specifically for the more progressive form of NAFLD.14 Although some glucose-lowering drugs such as metformin, glucagon-like peptide 1 receptor agonists and peroxisome proliferator-activated receptor-gamma agonists, exert glucose-lowering effects that are partially mediated by the remission of NAFLD in patients with type 2 diabetes with NAFLD, the evidence of pharmacological treatment of type 2 diabetes with NAFLD is very limited. In addition, these drugs can cause undesired side effects, and exorbitant prices have restricted the majority of patients.15

Dietary intervention is a safe and effective treatment for NAFLD or type 2 diabetes.16 17 However, very few studies have specifically focused on type 2 diabetes and NAFLD comorbidities. The Mediterranean diet can reverse hepatic steatosis and fibrosis by intrahepatic fat loss in NAFLD.18 19 A calorie-restricted or isocaloric low-carbohydrate diet is known to be able to significantly improve glycaemic control and reduce hepatic fat content in patients with type 2 diabetes, but these studies did not examine diabetes remission, and there was a lack of long-term studies on changes in liver fat.20–22 Recently, a consensus statement from the American College of Lifestyle Medicine noted that dietary intervention as a primary treatment, rather than as an adjunct therapy, is able to achieve diabetes remission.23 The three notable randomised controlled trials (The randomized controlled Diabetes Remission Clinical Trial (DiRECT) among predominantly white Europeans; The Diabetes Intervention Accentuating Diet and Enhancing Metabolism (DIADEM-I) study in Middle Easterners, and South Asian Diabetes Remission Feasibility Study (STANDby) in South Asians) targeting weight loss for different populations with low-energy meal replacements were able to achieve diabetes remission.24–26 Taken together, targeting the liver by dietary intervention is a key factor in diabetes remission in type 2 diabetes and NAFLD comorbidities.

Intermittent energy restriction can improve hyperglycaemia in type 2 diabetes and alleviate hepatic steatosis in NAFLD.27 28 However, there is still a lack of long-term follow-up and large sample size clinical studies to confirm its long-term safety and clinical efficacy, especially focusing on individuals with type 2 diabetes and NAFLD. In our previous study, we developed a Chinese Medical Nutrition Therapy (CMNT) regimen for management of type 2 diabetes.29 The dietary pattern of CMNT regimen is intermittent energy restriction and CMNT diet is substituted for conventional diet during energy restriction periods. The main ingredient of CMNT diet is whole grain rice flour and supplementary ingredients include medicinal and edible plants such as Ganoderma lucidum, Folium Mori, Poria cocos, Dioscorea opposita Thunb. (Chinese yam), Radix puerariae, Cordyceps militaris and Momordica grosvenori.29 The CMNT diet alone increased the abundance of beneficial bacteria and the production of short-chain fatty acids in the simulator of the human intestinal microbial ecosystem.30 CMNT regimen improved glycaemic control in type 2 diabetes by enhancing islet function and regulating gut microbiota composition.29 Our randomised controlled clinical trial confirmed the efficacy of CMNT regimen in achieving diabetes remission, with some participants with type 2 diabetes being able to discontinue or reduce their glucose-lowering medication use, but the effect on hepatic fat has not been explored.31 The present study will evaluate the effects of CMNT intervention on liver fat and glycaemic control in patients with type 2 diabetes and NAFLD. We will further determine whether CMNT is able to achieve diabetes remission and reduction of long-term liver fat at follow-up periods after the end of intervention.

Methods and analysisMethods/design

The trial is a prospective, multicentre, parallel-design, assessor-blinded randomised controlled trial. The trial is registered on ClinicalTrials.gov (identifier: NCT05439226). The research was initially started in 5 June 2022 and that the estimated study completion date is 5 May 2024. We will use a stratified randomisation method for age and sex. Participants will be randomly assigned to the CMNT group or the usual care group in a 1:1 ratio determined by a computer-generated random assignment sequence. Randomisation data will be kept strictly confidential and accessible only to authorised persons until unblinding occurs. Participants will be stratified based on their age (stratified into three age categories: 18–40 years, 45–60 years and 60–75 years) and sex (stratified by female and male sex). Patients with type 2 diabetes and NAFLD will be recruited from multiple hospital health examination centres in China. Assessments will be performed at baseline, 3 months, 6 months, 1 year and 2 years. This protocol follows guidance from the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 201332 (http://www.spiritstatement.org/). The SPIRIT 2013 checklist is provided in online supplemental additional file 1. A flow diagram of the trial is presented in figure 1.

Figure 1

Study flow diagram. CMNT, Chinese Medical Nutrition Therapy.

Patient and public involvement

The patients and the public are not involved in the conception, design, conduct or dissemination of this trial.

Participant recruitment

We aim to recruit 120 participants for this study. Potential participants will be recruited in the physical examination centre of multiple hospitals in China. One of the following methods will be followed in this recruitment:

Patients will be invited to participate in the study by a physician or nurse at their local outpatient clinic.

The physician will screen the electronic patient database and patient medical notes from the local clinic for recruited potential participants. Patients interested in participation will be urged to contact the study investigator by WeChat and a telephone call.

Potential participants will be recruited through recruitment posters at diabetes outpatient clinics or online advertising from the official websites of clinics.

Diabetes-related health lectures will be held at some community activity centres by the study investigator and the physician. Participants who meet the requirements will be screened in this project.

Inclusion criteria

Patients diagnosed with type 2 diabetes and NAFLD.

Age between 18 and 75 years.

Body mass index (BMI) between 18.0 and 35.0 kg/m2.

Weight stable for at least 3 months prior to the study (gain or loss <4 kg).

Stable treatment for at least 3 months with 1–3 oral antidiabetic medications (with or without insulin therapy), or not yet receiving antidiabetic medication prior to the beginning of the study.

Able to give written informed consent.

Exclusion criteria

Self-reported food allergy.

Alcohol abuse in the last 3 months (alcohol intake greater than 20 g per day for women; alcohol intake for men greater than 40 g per day).

Other liver diseases, such as chronic hepatitis B and C, autoimmune hepatitis, primary biliary cirrhosis, hemochromatosis, nodular regenerative hyperplasia and focal nodular hyperplasia.

Scheduled to be hospitalised for any surgical treatment during screening.

Pregnancy or planning of pregnancy during the study.

Inability, physically or mentally, to adhere to the procedures required by the study protocol.

Level 3 hypoglycaemic events (at least three times) occurring within the 90 days prior to screening.

Hospitalisation or emergency department visit for hyperglycaemia, diabetic ketoacidosis, lactic acidosis, hyperosmolar non-ketotic coma or diabetes.

History of cancer within the past 5 years.

Acute coronary or cerebrovascular event in the past 90 days or heart failure.

Haemorrhagic or ischaemic stroke within the last 6 months.

Hypothesis

The key hypotheses of the trial are that a CMNT diet accompanied by intermittent energy restriction in patients with type 2 diabetes and NAFLD, compared with usual care, will result in the following:

Reduction of liver fat.

Improvement in glycaemic control.

Improvements in blood glucose control, liver function and glucolipid metabolism indicators.

Potential diabetes remission (diabetes remission was defined as glycated hemoglobin (HbA1c) less than 6.5% without use of any diabetes medications for more than 3 months).

Primary objective

To study the effect of CMNT intervention on liver fat and HbA1c in patients with type 2 diabetes and NAFLD compared with usual care.

Secondary objectives

To study the effect of CMNT intervention on body weight, waist circumference, liver function, glucolipid metabolism indicators, diabetes remission and multiomics of patients with type 2 diabetes and NAFLD to explore the possible mechanism of CMNT intervention in regulating liver lipid metabolism and glucose homeostasis.

OutcomesCoprimary outcomes

The coprimary outcomes of the study are controlled attenuation parameter (CAP) values by transient elastography and HbA1c from baseline to 1, 2, 3, 6, 12 and 24 months.

Secondary outcomes

Anthropometric parameters: bodyweight, waist circumference, hip circumference, waist-to-hip ratio (WHR), BMI, systolic blood pressure and diastolic blood pressure.

Hepatic steatosis: Hepatic Steatosis index (HSI) and fatty liver index (FLI). Hepatic fibrosis: liver stiffness measurement (LSM) value by transient elastography, Fibrosis-4 index (FIB-4) and NAFLD fibrosis score (NFS).

Serum liver-related markers: γ-glutamyl transferase (γ-GGT), cytokeratin 18 (CK18) M65ED, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), albumin, globulin and platelet count.

Glucose metabolism parameters: fasting plasma glucose (FPG), fasting insulin (FINS), C-peptide, Homeostasis Model Assessment (HOMA)-IR and insulin-like growth factor 1 (IGF-1).

Blood lipid: triglyceride, total cholesterol, low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein-cholesterol (HDL-C).

Questionnaires: Food Frequency Questionnaire (FFQ), Pittsburgh Sleep Quality Index (PSQI), International Physical Activity Questionnaire (IPAQ), Patient Health Questionnaire 9 (PHQ-9), Generalized Anxiety Disorder scale (GAD-7), Social Support Rate Scale (SSRS) and Cost Consequences Analysis (CCA).

Glucose-lowering medication dosage.

Gut microbiota.

Metabolomics.

Intervention

The eligible participants were screened according to the inclusion and exclusion criteria. All participants will be informed of the study and signed informed consent prior to participation. A 90-day intervention period will be preceded by a 2-week run-in phase. Participants who enrol in the study will undergo baseline assessment over this period. Baseline outcome measures, including anthropometric measurements, haematological indexes, imaging indicators and questionnaire responses, are listed in table 1. Following the baseline assessments, participants will be randomly assigned to the CMNT group or the usual care group in a 1:1 ratio determined by a computer-generated random assignment sequence.

Table 1

Chinese Medical Nutrition Therapy schedule of assessments

CMNT group

The CMNT regimen contains a specific meal replacement food and its application is accompanied by intermittent energy restriction. An overview of the study design is provided in figure 2. Participants in the CMNT group will consume the provided CMNT diet for 5 consecutive days, followed by 10 days of regular food intake for 6 cycles (3 months). Participants will be instructed in a 600 kcal/day CMNT diet in continuous energy restriction 5 days per cycle. The CMNT diet includes four ready-to-consume prepared foods: composite nutritional rice, solid beverages, meal replacement biscuits, and fruit and vegetable gruel, which will be assigned to three meals a day. Multivitamin and mineral tablets will be used during the study. Detailed descriptions of all methods used, CMNT components and energy ratios are provided in online supplemental additional file 2. After the 3-month CMNT intervention period, the follow-up period will be scheduled to assess the long-term effect on diabetes remission and changes in liver fat. Participants will be asked to continue following their regular diets at 6, 12 and 24 months. In addition, regular on-site home visits, including screening visits, intervention visits and follow-up visits, will be performed by multiperson teams consisting of research investigators and nurses to complete the questionnaires by face-to-face conversations and collect blood, urine and faecal samples.

Figure 2

Overview of the study design. Participants with type 2 diabetes and non-alcoholic fatty liver disease (NAFLD) will be randomised to either Chinese Medical Nutrition Therapy (CMNT) group or a control group. Participants in CMNT group consumed the low-calorie CMNT diet for 5 consecutive days, followed by 10 days of regular diet for 6 cycles (3 months). Measurements will be performed prior to CMNT diet (baseline) and the end of intervention after the sixth cycle. The follow-up period will be 24 months, and clinical data will be collected at 6, 12 and 24 months. Participants in control group continued their regular diet and evaluated within the same time frame.

The CMNT diabetes management mobile application (app) will be used to document adherence to the CMNT intervention. We developed the CMNT app which is available in Android App store or Apple App store. Through the CMNT app, participants can real-time interaction with health managers, dietitians and physicians. The CMNT app allows participants to enter self-measured blood glucose readings, weight, food entries and physical activity. Participants will be trained on how to use the CMNT app and enter data correctly for the daily surveys before the start of the intervention. In this study, all participants will be instructed to complete real-time meal-logging throughout the entire CMNT intervention periods. This will allow research investigator to assess adherence to the CMNT intervention. Participants who fail to adhere to study procedures will be withdrawn from the study.

Usual care group

Participants in the control group will be provided dietary recommendations based on the guidelines for the Prevention and Treatment of Type 2 diabetes mellitus in China (2020 edition). Participants in the control group will be exposed to the same conditions and undergo the same tests as those in the CMNT intervention group except for the CMNT diet for 5 consecutive days every cycle.

Outcome measures

Blood, urine and faecal samples will be collected. Anthropometric measures, biochemical measurements, liver fat and liver fibrosis assessment and questionnaires will be performed. The gut microbiota of faecal samples will be assessed, and the metabolomics of plasma, urine and faecal samples will be determined. Detailed information and measurements at each stage of the study are summarised in table 1. To evaluate the primary and secondary outcomes, all relevant data will be collected by research investigators.

Anthropometric measures

Height (m) and body weight (kg) will be measured by using an anthropometric measurement instrument to the nearest 0.1 cm and the nearest 0.1 kg, respectively, and BMI will be calculated as weight (kg)/height2 (m2). Waist circumference and hip circumference will be assessed by using a standard tape measure, and WHR will be calculated. Blood pressure will be measured after the participant rests for at least 5 min to avoid measurement error.

Biochemical measurements

Peripheral venous blood of all participants will be collected by venipuncture after an overnight fast of at least 10 hours and will be used to measure HbA1c, serum liver-related markers including γ-GGT, CK18 M65ED, ALT, AST, ALP, albumin, globulin and platelet count, glucose metabolism parameters including FPG, FINS, C-peptide and IGF-1, and blood lipid profiles including TG, TC, LDL-C and HDL-C in the hospital health examination centre. Insulin resistance indexes were calculated as follows: HOMA-IR=FPG (mmol/L) × FINS (mU/L)/22.5.33 In addition, serum and plasma will be stored for further metabolomic analysis.

Liver fat and liver fibrosis assessment

Non-invasive estimation of liver fat will be measured using vibration-controlled ultrasound transient elastography. The CAP score, which will be measured by FibroScan, will be used to quantify and detect liver fat. CAP <238 dB/m indicated no hepatic steatosis, 238 ≤CAP ≤259 dB/m denoted mild steatosis, 260 ≤CAP ≤291 dB/m indicated moderate steatosis, and CAP>291 dB/m denoted severe steatosis.34 35

The HIS and FLI, as surrogate markers of hepatic steatosis, will also be applied to quantify the degree of steatosis in patients with type 2 diabetes and NAFLD.36 HSI was calculated as follows: HSI=8 × (ALT∕AST ratio) + BMI (+ 2, if female; +2, if diabetes mellitus). HSI values <30 exclude NAFLD with a sensitivity of 93.1%, while values >36 detect NAFLD with a specificity of 92.4%.37 The FLI was calculated as follows: FLI= [e (0.953×loge (triglycerides)+0.139 × BMI + 0.718 × loge (γGGT) + 0.053 × waist circumference − 15.745)]/[1+e (0.953×loge (triglycerides)+0.139 × BMI + 0.718 × loge (GGT) + 0.053 × waist circumference − 15.745)] × 100. FLI values <30 rule out liver steatosis, whereas FLI ≥60 confirmed NAFLD with a specificity of 61%.38

In addition, serum γ-GGT is mainly derived from the liver. Reduction in serum γ-GGT levels is correlated with reductions in excess liver fat.39 CK18 M65ED is used to monitor the different stages of NAFLD.40 We will also measure serum γ-GGT and CK18 M65ED levels, which indicate changes in liver fat.

Liver fibrosis will be assessed by using non-invasive vibration-controlled transient elastography to estimate LSM, which is highly correlated with the gold standard measures of liver biopsy.41 A median LSM ≥8.2 kPa was considered indicative of significant fibrosis (≥ F2), whereas LSM ≥9.7 kPa was considered indicative of advanced (≥ F3) fibrosis, and LSM ≥13.6 kPa was considered indicative of cirrhosis (F4).42

The degree of liver fibrosis will also be assessed by the FIB-4 index and NFS. The FIB-4 index, a non-invasive measurement, will be used to evaluate the degree of liver fibrosis. FIB-4 index=age × AST/[platelet count × (ALT)1/2].43 An FIB-4 index <1.45 had a negative predictive value of 94.7% to exclude progressive fibrosis, whereas an FIB-4 index >3.25 was considered advanced fibrosis or cirrhosis.44

NFS=−1.675 + (0.037 ×age [years]) + (0.094×BMI [kg/m2]) + (1.13×type 2 diabetes [yes=1, no=0]) + (0.99×AST/ALT ratio) – (0.013×platelet count [×109/L]) – (0.66×serum ALB [g/dL]).45 The probability of advanced liver fibrosis was classified into three groups: high NFS (NFS >0.676), intermediate NFS (NFS: −1.455 to 0.676, or 0.120 to 0.676 if age >65), or low NFS (NFS < −1.455, or < 0.120 if age >65).46

Questionnaires

We will use multiple questionnaires to assess the acceptability of the programme and process evaluation. Nutrition surveys will use a FFQ, which takes into account the frequency, portion size, variety, and other foods and beverages.47 Daily intake of total energy, macronutrients, certain food components and food groups will be calculated by dietary software based on the China Food Composition Database. The PSQI will be used to assess sleep duration and self-rated health.48 We assessed leisure time physical activity, walking for exercise and transportation over the previous week by the IPAQ.49 However, we did not intervene regarding normal daily exercise. All participants will be instructed to engage in daily moderate exercise. The PHQ-9, as a self-rated depression scale, and the GAD-7 will be suggested to assess informant-rated symptoms, including depression, apathy and anxiety questions.50 51 Social support will be measured using the SSRS.52 The primary economic analysis will consist of a within-trial CCA from healthcare and societal perspectives.53 The set of questionnaires can be found in online supplemental additional file 3.

Medication use

A recent history of medication use of all participants will be obtained from the available electronic medical records or dictation of each participant as a baseline medication use. There will be no restriction on related to the type of diabetes treatment. Dosages of glucose-lowering medications during the trial will be adjusted by physicians depending on changes in blood glucose values to avoid the risk of hypoglycaemia. The CMNT app will be applied to glycaemic management. Prior to initiation of the intervention diet, all participants will download the software in their cell phones. Fasting blood glucose, and bedtime glucose will be determined using a glucometer, which can automatically upload blood glucose values to the management software. Physicians can answer participants’ questions and provide real-time feedback on their blood glucose levels and medication taking for healthcare consultations and to increase healthcare accessibility when the remote system of the project suggests that the participant has hyperglycaemia, hypoglycaemia or other abnormal medical indicators. Hypoglycaemic medications will be prescribed strictly by their physician in accordance with the diabetes medication management standards.54 The principles of adjustment of glucose-lowering agents can be found in online supplemental additional file 4.

Participant withdrawal

Participants who withdraw from the trial will not be replaced. To perform intention-to-treat analysis, participants will be asked to return at 3, 6, 12 and 24 months for measurements of the primary and secondary endpoints. Participants could choose to withdraw from the study at any time. If the female participant is pregnant, she will be withdrawn from the study, and pregnancy events will be followed up until pregnancy outcomes are available. If the participant voluntarily withdrew from the experiment and withdrew their informed consent, we will analyse all data generated up to the point at which the study was discontinued and document the reason for discontinuation (if known). If the participant was lost to follow-up (the participant could not be contacted for visit assessment or examination by phone or email 3 consecutive times), details of all unsuccessful attempts should be recorded in the research document); if termination of trial by sponsor (eg, funding reasons, administrative reasons, change in R&D plan, etc), or the ethics committee requested that the trial be stopped for some reason, or for any reason the sponsor, regulatory authority or ethics committee requests that the study be terminated. An individual participant will be considered to have completed the trial when he or she completed his or her last planned visit. The trial will be closed when the last participant completed the last planned visit.

Adverse events

No side effects or serious side effects occurred during a previous study with CMNT intervention.31 Despite this, adverse events (AEs) and serious adverse events will be monitored, assessed and documented by physicians who are asked to report these events. Whether in the intervention or control arm, even if the intervention has been implemented, AEs will be managed by a site visit or site initiation aimed at proactively monitoring and treating promptly.

Monitoring

The data monitoring committee will be composed of at least two members from the Biomedical Research Ethics Committee of Hunan Agricultural University who have no conflicts of interest with our study. To ensure that the rights and interests of participants are protected, the data recorded and reported in the study are accurate and complete, the study follows the approved protocol and relevant laws and regulations, the contents of the study protocol will be strictly observed, and the study data will be filled out correctly. A safety report will be submitted to the Medical Research Ethics Committee annually.

Incidental findings

A restricted-calorie diet might entail certain risks, such as hunger, anxiety, fatigue, dizziness, headaches, muscle pain, low blood pressure, and, in rare cases, fainting. These dietary interventions may also result in arrhythmias, short-term nutrient deficiencies and weakened immune responses. A prolonged low-calorie diet is particularly dangerous for individuals who are already malnourished, such as those with acute gastroenteritis or chronic diarrhoea. During the study, participants could stop the diet and resume regular eating at any time. If participants experience any significant discomfort, immediate medical attention should be sought. Participants should drink plenty of water to prevent dehydration and avoid vigorous exercise during the trial. Participants should refrain from operating vehicles or heavy machinery. They should avoid exposure to high temperatures, such as hot showers or baths, and abstain from alcohol. If participants have any concerns or feel unwell, they should contact the physicians, consult with healthcare professionals or seek immediate medical care. After the diet period ends, participants should avoid binge eating and gradually reintroduce regular foods, starting with liquids such as soups and fruit juices, and then progressing to a light diet. Participants may experience dizziness during blood drawing. In rare cases, participants may experience minor risks such as abrasions, excessive bleeding, infection (a slight risk when the skin is broken), dizziness and fainting. Participants could discontinue the blood drawing procedure at any time. If participants experience bleeding or infection related to the blood drawing, they should contact healthcare professionals or seek immediate medical care.

Confidentiality

The data and materials will be housed in individual rooms and stored on a drawer with a lock. Electronic data will be kept as password-protected files in a secure computer database. Scientific reports generated from the study will be anonymised. Randomisation data accessible only to authorised persons are able to access the dedicated information needed to perform this work. The study results will be disseminated to the participant via email or telephone.

Protocol delivery fidelity

To minimise the potential for observational error across all practices, a standard training protocol was developed. Intervention will be assessed in the study by retraining and team supervision, who will use a digital lifestyle intervention system to ensure that all participants in the CMNT group and usual care group receive a uniform training programme. The delivery of the intervention will be assessed to confirm the effect of interventions.

Sample size calculations

The sample size calculation is based on the reduction in HbA1c. According to preliminary studies, the mean and SD values of HbA1c after intervention were 7.173% ±1.919 in the CMNT group and 8.124% ±1.144 in the control group.29 Forty-four participants in each group will be recruited, and the following formula will be used to calculate the required sample size:

With consideration of stratification analysis on type 2 diabetes and NAFLD comorbidity, we presume a dropout rate of 20%. In addition, our trial adds a 12-month follow-up time point compared with our primary care, and we will amplify the sample size to 60 in each group.

Statistical analysis

All statistical analyses will be conducted using IBM SPSS V.25. We will use a t-test or χ2 test to determine the differences for normal data and Mann-Whitney U test for non-normal data between the CMNT group and usual care group. Data will be analysed on an intention-to-treat basis, comparing the CMNT group with the usual care group. CAP values and HbA1C values at baseline and after treatment will be evaluated by repeated measure analysis of variance. To control for confounders, the primary and secondary outcomes will be compared via analysis of covariance with age, sex, IPAQ score and adherence to the diet as covariates. Multivariate linear regression analysis will be performed to assess the association between changes in CAP values and changes in HbA1c values between two groups. P values <0.05 are considered statistically significant and the data are analysed using SPSS software.

Discussion

NAFLD is highly prevalent among individuals with type 2 diabetes, highlighting the urgent need for a coordinated management of steatotic liver disease and glycaemic control in type 2 diabetes and NAFLD.55 A strength of the present study is the investigation of the long-term effects of CMNT intervention on liver fat and glycaemic control compared with usual care in patients with type 2 diabetes and NAFLD, which enabled assessment of the progress of NAFLD and diabetes remission simultaneously. As a result, these data help establish relationships between reduction of liver fat and diabetes remission, and better characterise the benefit–risk profile of CMNT intervention.

The coexistence of type 2 diabetes and NAFLD results in worse metabolic status and higher risk of cardiometabolic diseases.4 Hepatic steatosis, particularly accompanied by impaired metabolic health, might precede or promote the development of cardiometabolic diseases via insulin resistance, atherogenic dyslipidaemia and hepatic production of a variety of proatherogenic, procoagulant and proinflammatory mediators, and dysregulation of hepatokines.56 In addition, chronic hyperglycaemia worsens already existing cardiometabolic risk factors through endothelial dysfunction, oxidative stress and chronic inflammation.57 Another strength of this study is the investigation of the potential mechanisms involved in gut microbiota and metabolomics. We will test whether CMNT intervention improves cardiometabolic risk in type 2 diabetes and NAFLD in future studies. The metabolic subphenotyping strategies developed by using anthropometrics, metabolic characteristics, gut microbiota and metabolomics in the setting of cluster analyses among these populations may be available regarding the prediction, prevention and treatment of cardiometabolic diseases.58 59

A limitation of the trial is that the lack of liver biopsy in patients with type 2 diabetes and NAFLD will not allow us to evaluate whether there are beneficial the liver histological changes, which range from steatosis alone to steatohepatitis (NASH) with or without fibrosis to cirrhosis. In this study, we use CAP value by transient elastography to estimate the change of liver fat content, which is suitable to quantify hepatic steatosis.34 Another limitation of the trial is that except for the provided CMNT diet for 5 consecutive days, the participants’ regular diet is varied though dietary guidelines are offered. The effect of diet as a potential confounding factor may be directly controlled for in future studies using a standard identical diet.

Trial status

Recruitment for this trial was initiated in July 2022 and is progressing according to the time plan. Baseline measurement started in December 2022 for the first enrolled group. CMNT interventions started in December 2022. The recruitment of participants will last until May 2024.