Type 2 diabetes is a prevalent noncommunicable disease that threatens all nations. In 2014, more than 380 million people have diabetes worldwide [
Medicinal plants have always been used as a traditional medicine to treat several diseases [
This single-centre, randomized, two-arm parallel controlled clinical trial was conducted at General Medical Clinic and Endocrine Clinic of Hospital Serdang, a tertiary care government hospital, Malaysia. This trial was approved by the Ethics Committee for Research Involving Human Subjects of Universiti Putra Malaysia (JKEUPM) (FPSK_Ogos (13)05), Herbal Medicine Research Centre, Institute for Medical Research Malaysia (version 1, 8/2014), and Medical Research and Ethics Committee Ministry of Health Malaysia (NMRR-13-1344-18177). This trial was also registered with clinicalTrials.gov, identifier number
Patients were eligible for enrolment if they were aged between 30 and 65 years, have confirmed diagnosis of type 2 diabetes for more than 6 months, have last HbA1C value greater than 7%, were treated with stabilized dose of antidiabetic drugs, and were expected to keep the dose throughout the trial. Exclusion criteria included pregnancy, insulin treatment, acute infection, severe liver disease, kidney disease and gastrointestinal disease, and anticoagulant therapy such as warfarin and aspirin. All subjects gave written informed consent before enrolment.
Subjects were randomly allocated to either diabetic-ulam group or diabetic control group using permuted block randomization in block of 4 and 6. Subjects in the diabetic-ulam group consumed 15 g of fresh
Patients were asked to visit clinical centre every 4 weeks after screening, which includes baseline, week 4 (middle of study), and week 8 (end of study). We recorded subjects’ sociodemographic data at the baseline. At each follow-up, we measured subjects’ weight, height, and waist circumference. Venous blood samples were drawn after an overnight fasting to measure biochemical indices. All biochemical parameters were measured in all 3 visits except HbA1C which was not measured in week 4. Patients were contacted weekly during the study period and all the occurrences of the adverse events such as loose stools, abdominal discomfort, bloating, flatulence, sign of hypoglycemia, and sign of hyperglycemia were recorded.
Fasting blood glucose was measured by the hexokinase method (Architect Ci 8200 analyzer, Abbott Laboratories, USA). HbA1C was assayed by turbidimetric inhibition immunoassay (Cobas Integra 800, Roche Diagnostics, Germany). Serum insulin level was assayed by chemiluminescent microparticle immunoassay (Architect Ci 8200 analyzer, Abbott Laboratories, USA). Homeostatic model assessment-insulin resistance (HOMA-IR) was calculated as fasting insulin (
We need to enroll 38 patients in each group to detect 1% changes in HbA1C [
Statistical analysis was performed using SPSS version 21 for windows (SPSS Inc., Chicago, USA). Data were expressed as mean ± SD for continuous parameters and percentage for categorical parameters. Baseline characteristics between the two groups were compared using independent
A total of 4783 subjects were initially screened in this study. After excluding noneligible subjects and those who refused to participate, a total of 101 patients with type 2 diabetes were recruited and randomly allocated to diabetic-ulam group (50 subjects) and diabetic control group (51 subjects). Patients who did not attend the baseline blood test were excluded. Three subjects from diabetic control group failed to follow up. The final analysis was performed on 77 subjects (38 diabetic-ulam group; 39 diabetic controls) using intention-to-treat analysis. A flowchart of the study trial is presented in Figure
Subjects enrolment and follow-up based on CONSORT statement.
The baseline characteristics of the subjects in both groups are presented in Table
Baseline characteristics of the study participants.
Diabetic-ulam group ( |
Diabetic controls ( |
|
|
---|---|---|---|
Age (years) | 48.4 ± 9.1 | 50.9 ± 9.1 | 0.213 |
Gender: male (%) | 19 (50) | 24 (61.5) | 0.308 |
Gender: female | 19 (50) | 15 (38.5) | 0.308 |
Duration of diabetes (years) | 6.2 ± 4.6 | 7.7 ± 6.6 | 0.248 |
T2DM treatment | 0.350 | ||
Metformin | 26 (68.4) | 23 (59) | |
Metformin + SU | 11 (28.9) | 16 (41) | |
Metformin + acarbose | 1 (2.6) | 0 (0) | |
History of hypertension | 23 (60.5) | 30 (76.9) | 0.120 |
History of dyslipidemia | 23 (60.5) | 24 (61.5) | 0.927 |
Weight (kg) | 77.79 ± 13.58 | 82.52 ± 10.72 | 0.144 |
BMI (kg/m2) | 29.14 ± 4.94 | 30.41 ± 4.37 | 0.233 |
Waist circumference (cm) | 98.12 ± 12.82 | 102.45 ± 11.14 | 0.117 |
FBG (mmol/L) | 9.66 ± 3.20 | 8.44 ± 2.68 | 0.073 |
HbA1C (%) | 8.81 ± 1.71 | 8.78 ± 1.40 | 0.942 |
Serum insulin ( |
10.67 ± 7.42 | 12.73 ± 6.77 | 0.208 |
HOMA-IR | 4.46 ± 3.39 | 3.94 ± 3.67 | 0.722 |
QUICKI | 0.53 ± 0.08 | 0.52 ± 0.07 | 0.459 |
hs-CRP (mg/L) | 4.48 ± 3.16 | 4.03 ± 3.14 | 0.591 |
TC (mmol/L) | 5.04 ± 1.23 | 4.91 ± 1.39 | 0.680 |
TG (mmol/L) | 2.26 ± 1.97 | 1.94 ± 0.91 | 0.370 |
HDL-c (mmol/L) | 1.40 ± 0.31 | 1.45 ± 0.30 | 0.437 |
LDL-c (mmol/L) | 2.82 ± 0.86 | 2.62 ± 1.19 | 0.421 |
T2DM: type 2 diabetes mellitus, SU: sulphonylurea, BMI: body mass index, FBG: fasting blood glucose, HbA1C: glycated hemoglobin, HOMA-IR: homeostatic model assessment-insulin resistance, QUICKI: quantitative insulin sensitivity check index, hs-CRP: high-sensitivity C-reactive protein, TC: total cholesterol, TG: triglycerides, HDL-c: high-density lipoprotein cholesterol, and LDL-c: low density lipoprotein cholesterol.
The means of biochemical parameters on glycemic status comparing the two groups are presented in Table
Mean changes in biochemical parameters.
Parameters | Group |
|
||
---|---|---|---|---|
Diabetic-ulam ( |
Diabetic controls ( |
|||
Mean ± SD | Mean ± SD | |||
HbA1C (%) | Baseline | 8.81 ± 1.71 | 8.78 ± 1.40 | 0.942 |
Week 8 | 8.05 ± 1.67 | 8.41 ± 1.40 | 0.302 | |
|
||||
Serum insulin ( |
Baseline | 10.67 ± 7.42 | 12.73 ± 6.77 | 0.208 |
Week 4 | 9.37 ± 6.26 | 14.66 ± 7.36 | 0.002 |
|
Week 8 | 9.52 ± 5.73 | 16.64 ± 8.04 | 0.001 |
|
|
||||
HOMA | Baseline | 4.46 ± 3.39 | 3.94 ± 3.67 | 0.722 |
Week 4 | 3.58 ± 2.90 | 5.44 ± 3.23 | 0.014 |
|
Week 8 | 3.56 ± 1.94 | 6.28 ± 3.71 | 0.001 |
|
|
||||
QUICKI | Baseline | 0.53 ± 0.08 | 0.52 ± 0.07 | 0.459 |
Week 4 | 0.56 ± 0.09 | 0.50 ± 0.05 | 0.001 |
|
Week 8 | 0.58 ± 0.11 | 0.49 ± 0.07 | 0.001 |
|
|
||||
hs-CRP (mg/L) | Baseline | 4.48 ± 3.16 | 4.03 ± 3.14 | 0.591 |
Week 4 | 3.09 ± 3.04 | 3.65 ± 3.21 | 0.431 | |
Week 8 | 2.95 ± 2.68 | 3.79 ± 2.80 | 0.182 | |
|
||||
Total cholesterol (mmol/L) | Baseline | 5.04 ± 1.23 | 4.91 ± 1.39 | 0.680 |
Week 4 | 4.78 ± 1.15 | 4.59 ± 1.15 | 0.479 | |
Week 8 | 5.04 ± 1.18 | 4.69 ± 1.30 | 0.228 | |
|
||||
Triglycerides (mmol/L) | Baseline | 2.26 ± 1.97 | 1.94 ± 0.91 | 0.370 |
Week 4 | 2.01 ± 1.27 | 1.78 ± 0.90 | 0.371 | |
Week 8 | 1.97 ± 1.41 | 1.72 ± 0.64 | 0.303 | |
|
||||
HDL-c (mmol/L) | Baseline | 1.40 ± 0.31 | 1.45 ± 0.30 | 0.437 |
Week 4 | 1.37 ± 0.29 | 1.44 ± 0.29 | 0.374 | |
Week 8 | 1.46 ± 0.31 | 1.40 ± 0.26 | 0.397 | |
|
||||
LDL-c (mmol/L) | Baseline | 2.82 ± 0.86 | 2.62 ± 1.19 | 0.421 |
Week 4 | 2.54 ± 0.87 | 2.51 ± 1.03 | 0.875 | |
Week 8 | 2.49 ± 0.99 | 2.56 ± 1.10 | 0.771 | |
|
||||
AST (U/L) | Baseline | 25.71 ± 13.99 | 25.44 ± 12.34 | 0.927 |
Week 4 | 23.66 ± 9.90 | 26.72 ± 13.72 | 0.266 | |
Week 8 | 23.63 ± 10.92 | 25.62 ± 11.81 | 0.447 | |
|
||||
ALT (U/L) | Baseline | 33.32 ± 21.18 | 31.82 ± 19.81 | 0.773 |
Week 4 | 29.84 ± 18.63 | 34.41 ± 22.48 | 0.335 | |
Week 8 | 30.97 ± 19.91 | 33.51 ± 18.39 | 0.563 | |
|
||||
Urea (mmol/L) | Baseline | 4.95 ± 1.49 | 4.71 ± 2.09 | 0.562 |
Week 4 | 4.50 ± 1.42 | 4.52 ± 2.02 | 0.975 | |
Week 8 | 4.90 ± 1.61 | 4.38 ± 1.68 | 0.168 | |
|
||||
Creatinine ( |
Baseline | 85.42 ± 21.08 | 91.18 ± 22.62 | 0.252 |
Week 4 | 84.53 ± 20.92 | 88.20 ± 23.28 | 0.469 | |
Week 8 | 84.53 ± 22.34 | 87.99 ± 21.86 | 0.494 |
SD: standard deviation, HDL-c: high-density lipoprotein cholesterol, LDL-c: low density lipoprotein cholesterol, AST: aspartate aminotransferase, and ALT: alanine aminotransferase.
Mean difference from baseline, week 4, and week 8. (a) HbA1C. (b) Insulin IR,
A HOMA-IR level represents insulin resistance [
Levels of hs-CRP, a measurement of inflammatory marker, in the diabetic-ulam group were reduced at all follow-up visits compared to diabetic controls (Table
To determine possible adverse effect of
HbA1C is an important parameter to evaluate glycemic control over two to three months [
Insulin resistance and insulin deficiency are two well-known key factors in the pathogenesis of type 2 diabetes [
In addition to its beneficial effect on glycemic control,
Furthermore, it has been shown that
Additionally, dietary intakes and physical activity level of the patients did not differ between two groups. There was a decrease in the energy and macronutrients intake in both groups throughout the study. However, the reduction was comparable between two groups (data not shown). Similarly, physical activity level increased in both groups. Nevertheless, the increment was comparable between groups and there were no statistically significant changes over the duration of study (data not shown). Hence, it is unlikely that dietary intake and physical activity confounded this study.
Possible mechanisms are being studied in order to explain the role of
Overall, we found that an eight-week supplementation of
However, there are limitations in this trial. Dose-dependent response evaluation is not carried out in this study. Secondly, lack of placebo used in this trial could contribute to bias in the outcomes. Besides, other inflammatory parameters such as interleukin 6 (IL-6) and tumor necrosis factor (TNF-
Our results indicate that short-term
The authors declare that there is no conflict of interests regarding the publication of this paper.
This trial was funded by grant from Agro-Biotechnology Institute, Ministry of Science, Technology and Innovation, Malaysia (no. 6370700).