Study participants

One hundred twenty volunteers aged 15 to 60 years, from the outpatient clinic at the Internal Medicine Department at Ramathibodi Hospital, Thailand were recruited by advertisement for the screening of prediabetes between June 2021 and March 2022. Fifty-six eligible participants were enrolled (Fig. 1).

Fig. 1

Flowchart. A total of 120 volunteers were screened for prediabetes and 56 eligible participants were enrolled. Participants were randomized into 2 groups; 28 in the intervention group and 28 in the placebo group and followed for 12 weeks. All participants were completed the study and assessed for clinical and biochemical parameters at week 0 and 12

Inclusion criteria included prediabetic overweight/obese individuals aged 15–60 years who met the American Diabetes Association (ADA) criteria for prediabetes [glycated hemoglobin A1c (HbA1C) 5.7–6.4% and/or fasting plasma glucose (FPG) 100–125 mg/dL] [1] and had a body mass index (BMI) ≥ 23 kg/m2. Participants taking antihypertensive or lipid-lowering medications must not have had any medication adjustments within three months before recruitment. Exclusion criteria included refusal to participate, diabetes diagnosis or current use of antidiabetic agents, current use of medications affecting glucose metabolism (for example, steroid, tacrolimus, atypical antipsychotic drugs), pregnancy, or disability affecting activities of daily living.

The study protocol was approved by the institutional review board of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University (MURA2020/939). The participants were informed about the details of this study, including the purpose, the procedure, risks, compensation for participation, participant rights, contact information, and a confidentiality agreement. After that, written informed consent was obtained from all participants. The clinical trial number is TCTR20210714004 (Thai Clinical Trials Registry).

Study design

Participants were divided into two groups of 28 participants (Fig. 1). The first group received Tremella fuciformis beverage (containing 6.4 g of β-glucan in a 180 mL beverage) and the other received a similar-looking and tasting placebo drink of the same volume. The assignment of participants into treatments was done by computers (simple randomization), where the computer assigns each participant a code number and treatment group. The investigators and participants only knew the code number to avoid bias. Only the study coordinators were allowed to know the treatment received by each participant. Their role was to communicate this information to the company funding the research, specifically regarding the type of beverages administered. The CPF Food Research & Development Center Company Limited directly dispatched the beverages to the participants through mail every 4 weeks. Throughout the study, the list containing the names of participants and their corresponding code numbers remained confidential and undisclosed. This confidentiality measure was maintained until the completion of the study. Participants were requested to drink one bottle of beverage before dinner every day for 12 weeks (Fig. 1).

Participants had to visit the study clinic two times: before the start of the trial and at the end of the 12th week. At both visits, a medical history was taken and body weight, height, waist circumference, and blood pressure were measured. Participants were asked not to modify their lifestyle habits nor their medication unless directed to do so by physician throughout the study period. Participants had to consume a regular diet containing at least 150 g of carbohydrate per day for three days prior to the test and fast for at least ten hours before each visit. 75 oral glucose tolerance test (OGTT) was assessed at baseline and 12th week. Blood samples were taken to measure plasma glucose and insulin at 0, 30, 60, 90 and 120 min. Insulin resistance and sensitivity were assessed using the Matsuda index [18], homeostatic model assessment-insulin resistance (HOMA-IR), the insulinogenic index, and the insulin disposition index. HbA1C and lipid profiles were also measured at both visits. The physical activity and sedentary behavior were assessed by the Global Physical Activity Questionnaire (GPAQ) version 2 [19]. Compliance was confirmed using the participants’ daily self-monitoring logs and monthly telephone follow-ups.

The beverages’ preparation

Tremella fuciformis was obtained from a local supermarket in Thailand. The samples were soaked and boiled with boiling water with a mushroom:water ratio of 15:85, (w/v) at 95°C for 10 min. The aqueous solution containing the extract was mixed with 10% monk fruit solution in water and sterilized by retort sterilization. To prepare a placebo drink, a mixture of 10% monk fruit solution and water was combined to equal the volume of Tremella fuciformis drink (180 mL) and sterilized by retort sterilization.

Nutritional composition of beverages

One bottle (180 mL) of Tremella fuciformis beverage contains 0% protein, 0% fat, 0% carbohydrate, 3% dietary fiber, and 1% sodium of the Thai Recommended Daily Intakes for adults and provides 0 kcal of energy. No vitamins (A, B1 and B2) and minerals (calcium and iron) are found in the content. The placebo beverage contains the same volume and components as the Tremella fuciformis beverage, except for Tremella fuciformis.

The determination of β-glucan and other bioactive compounds in Tremella fuciformis

The β -glucan content of Tremella fuciformis and their fractions were determined by the Thailand Institute of Scientific and Technological Research using a yeast and mushroom beta-glucan assay kit (Megazyme, code K-YBGL). In this study, the Tremella fuciformis beverage (180 mL) contained 15% Tremella fuciformis (6.4 g of β-glucan), 0.08% Luo Han Guo, and 0.04% licorice.

Total phenolic contents and antioxidant activities of Tremella fuciformis were investigated using the Folin-Ciocalteu method and DPPH method, respectively. In this study, the polyphenols content was 132.88 ± 8.95 µg of gallic acid equivalents /ml of sample and the antioxidant capacity was 23.37 ± 12.24 µg Trolox equivalents (TE)/ml of sample.

Outcomes

Primary outcomes were changes in anthropometric, metabolic, and biochemical parameters at the end of the study. Secondary outcomes were adverse effects caused by the treatment and compliance.

Statistical analysis

The intention-to-treat analysis was used to assess treatment efficacy. Continuous variables with normal distribution were reported as mean ± standard deviation (SD), while those with non-normal distribution were reported as median and interquartile range. The Kolmogorov-Smirnov normality test was used to determine whether variables were normally distributed. To compare the difference in clinical characteristics at baseline between the intervention and the placebo groups, we used the Independent-Samples T test, the Mann-Whitney test, and the Chi-square test, as appropriate. Changes in demographic, anthropometric, metabolic, and biochemical parameters after intervention in each group were assessed using paired T-tests. Results were considered significant if the p-value was < 0.05. The Wilcoxon test was used for parameters that were not normally distributed. Cohen’s d (based on differences) was calculated to report the effect size of the differences. Because no previous study has addressed the same issue, the target sample size was to recruit at least 10–12 participants per arm, as previously suggested [20]. Statistical analyses were performed using SPSS software version 28.0 (IBM, USA).