TABLE 3.
Summary table of studies of the antidiabetic effects of polyphenolic compounds in humans. Abbreviations: AMPK, AMP-activated protein kinase; BMI, body mass index; GIP, gastric inhibitory polypeptide; GLP-1, glucagon-like peptide 1; GSK-3β, glycogen synthase kinase-3β; HbA1c, hemoglobin A1c; HOMA-β, Homeostatic Model Assessment of β-cell Function; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; PGC-1α, peroxisome proliferator-activated receptor gamma coactivator 1α; RCT, randomized controlled trial; SIRT1, sirtuin 1; T2DM, type 2 diabetes mellitus.
| Compound | Type of study | Dose per day (oral) | Duration | Participant criteria | Outcomes | References |
|---|---|---|---|---|---|---|
| Polyphenols (general) | Meta-analysis | 33–2093 mg | 4 weeks–1 year | Diagnosed T2DM (taking or not taking medication), at high risk of T2DM, or low risk of T2DM | Reduced fasting blood glucose levels and small decrease in HbA1c levels. No effect on insulin levels or HOMA-IR. | Raimundo et al. (2020) |
| Effect was strongest in patients with diagnosed T2DM, especially in conjunction with antidiabetic medication | ||||||
| Green tea catechins | Meta-analysis | 208–1,207 mg | 2 weeks–6 months | Diagnosed or borderline T2DM, overweight to obese subjects, or healthy subjects | Decreased fasting blood glucose and HbA1c levels. No effect on fasting insulin levels or HOMA-IR. | Liu et al. (2013) |
| Effect only seen in studies with catechin intake ≥457 mg (median), and only in subjects with/at risk of metabolic syndrome. No difference between short or long duration studies | ||||||
| Meta-analysis | 235.64–1,206.9 mg | 3–24 weeks | Elevated fasting blood glucose, overweight to obese subjects, or healthy subjects | Lowered fasting blood glucose. No change in HbA1c, insulin levels or HOMA-IR. | Zheng et al. (2013) | |
| Effect only seen in studies with a duration ≥12 weeks. Dose and health status were not effect modifiers | ||||||
| Meta-analysis | 200 mg polyphenols – 4 cups green tea | 4 weeks–18 months | Participants with T2DM or prediabetes | No effect on fasting blood glucose, insulin, HbA1c or HOMA-IR. | Yu et al. (2017) | |
| Epigallocatechin gallate | RCT | 300 mg | 8 weeks | Diagnosed T2DM and not receiving insulin treatment (n = 44) | Decreased fasting blood glucose and marker of inflammation. No effect on insulin levels or HOMA-IR. | Hadi et al. (2020) |
| RCT | 800 mg | 8 weeks | Overweight or obese subjects (n = 88) | No change in fasting glucose, HbA1c, insulin, lipid levels, glucose tolerance, HOMA-IR or HOMA-β. There was a decrease in diastolic blood pressure | Brown et al. (2009) | |
| Resveratrol | RCT | 10 mg | 4 weeks | Diagnosed T2DM and not receiving insulin treatment (n = 19) | Extended time to maximal tissue glucose levels after a meal and decreased HOMA-IR. Did not affect blood glucose, serum insulin, amylin, GLP-1, GIP or lipid levels, or HOMA-β. Decreased marker of oxidative stress and increased marker of insulin signaling | Brasnyo et al. (2011) |
| RCT crossover | 150 mg | 30 days | Participants with obesity (n = 11) | Decreased plasma glucose, insulin, leptin, triglycerides levels, and HOMA-IR. Delayed peak glucose and insulin responses after a meal. Decreased systolic blood pressure. Decreased inflammatory markers. Increased markers of mitochondrial oxidative metabolism (including AMPK, SIRT1 and PGC-1α) and improved liver function | Timmers et al. (2011) | |
| RCT | 40 or 500 mg | 6 months | Diagnosed T2DM and not receiving insulin treatment (n = 179) | No change in fasting serum glucose, HbA1c, insulin or lipid levels, liver function biomarkers, HOMA-IR or blood pressure. Decrease in C-reactive protein (inflammation) | Bo et al. (2016) | |
| RCT crossover | 1,000 mg | 5 weeks | Diagnosed T2DM managed by diet only (n = 14) | No change in fasting or post-prandial blood glucose or GLP-1 levels. No effect on HbA1c levels | Thazhath et al. (2016) | |
| RCT | 1,500 mg | 4 weeks | Participants with obesity (n = 24) | No effect on fasting plasma glucose, HbA1c, insulin, lipid levels, HOMA-IR or blood pressure. No effect on insulin sensitivity by hyperinsulinemic euglycemic clamp. No effect on inflammatory or liver biomarkers, or AMPK and SIRT1 activity | Poulsen et al. (2013) | |
| RCT | 3,000 mg | 12 weeks | Diagnosed T2DM, on oral hypoglycemic treatment (n = 10) | No change in fasting plasma glucose, insulin or lipid levels, or in HOMA-IR. Nonsignificant decrease in HbA1c. Did increased skeletal muscle SIRT1 expression, pAMPK:AMPK ratio, and resting metabolic rate | Goh et al. (2014) | |
| Grape extract/Grape extract + resveratrol | RCT | 302 mg polyphenols ±16.2 mg resveratrol | 12 months | Participants with T2DM, stable coronary heart disease and hypertension taking medication (n = 35) | No effect on serum glucose, HbA1c, lipid levels, or blood pressure. Several inflammation markers were decreased | Tome-Carneiro et al. (2013) |
| Curcumin | RCT | 180 mg | 12 weeks | At risk of developing T2DM and BMI ≥25 (n = 29) | No change in fasting blood glucose but lowered serum insulin levels and HOMA-IR. Decreased serum amylin and GSK-3β levels | Thota et al. (2020) |
| RCT | 300 mg (curcuminoids) | 3 months | Diagnosed T2DM and BMI ≥24, on oral hypoglycemic and/or insulin treatment (n = 100) | Decreased fasting serum glucose, HbA1c, lipid levels, and HOMA-IR. No effect on liver function biomarkers | Na et al. (2013) | |
| RCT phase 2 | 500 mg (≥95% curcuminoids, ≥65% curcumin)±30 mg zinc | 3 months | Overweight or obese subjects with prediabetes (n = 82) | Curcumin with and without zinc decreased fasting plasma glucose, postprandial glucose, serum insulin, HbA1c and insulin resistance. Increased insulin sensitivity. Curcumin + zinc group only decreased BMI. | Karandish et al. (2021) | |
| RCT | 1,500 mg (curcuminoids) | 9 months | Participants with prediabetes not taking medication (n = 240) | Decreased the risk of developing T2DM. Reduced fasting blood glucose, HbA1c, and C-peptide levels. Improved glucose tolerance, HOMA-IR and HOMA-β | Chuengsamarn et al. (2012) | |
| Quercetin | RCT | 250 mg | 8 weeks | Diagnosed T2DM and not receiving insulin treatment (n = 47) | No effect on fasting blood glucose, HbA1c levels, serum insulin, lipid levels, or HOMA-IR. Increased serum antioxidant capacity and decreased markers of oxidative stress | Mazloom et al. (2014) |
| Meta-analysis | 100–1,000 mg | 4–12 weeks | Subjects with (pre)hypertension, polycystic ovary syndrome or T2DM, or were overweight or obese | Overall, no significant effect on fasting plasma glucose, serum insulin, HbA1c or HOMA-IR. Did decrease fasting glucose in studies ≥8 weeks and using ≥500 mg doses. Also decreased insulin levels in studies with participants <45 years and doses ≥500 mg | Ostadmohammadi et al. (2019) | |
| Olive leaf extract/Oleuropein | RCT crossover | 51.1 mg oleuropein, 9.7 mg hydroxytyrosol | 12 weeks | Males with BMI between 25 and 30 (n = 45) | Improved glucose tolerance, insulin sensitivity and β-cell function. Increased IL-6 but no change in other inflammatory markers. No effect on lipid levels or liver function biomarkers | de Bock et al. (2013) |
| RCT crossover | 136.2 mg oleuropein, 6.4 mg hydroxytyrosol | 6 weeks | Males with prehypertension (n = 60) | No change in fasting glucose, insulin or HOMA-IR. Decreased blood pressure and plasma lipid levels. Reduced IL-8 but no effect on other inflammatory markers | Lockyer et al. (2017) | |
| RCT | 320.8 mg oleuropein, 11.9 mg hydroxytyrosol | 12 weeks | Participants with prediabetes and BMI between 23 and 29.9 (n = 56) | Decreased fasting plasma glucose and lipid levels, however increased HbA1c was observed. No change in insulin levels or HOMA-IR. | Araki et al. (2019) | |
| RCT | 500 mg olive leaf extract | 14 weeks | Diagnosed T2DM and not receiving insulin treatment (n = 79) | No change in fasting or postprandial blood glucose but decreased HbA1c and fasting insulin levels | Wainstein et al. (2012) |