Carbohydrates, glycemic index and diabetes mellitus

Effect of cocoa and green tea on biomarkers of glucose regulation, oxidative stress, inflammation and hemostasis in obese adults at risk for insulin resistance



Flavanols may provide protection against insulin resistance, but little is known about the amounts and types of flavanols that may be efficacious.


This study was designed to determine whether cocoa flavanols, over a range of intakes, improve biomarkers of glucose regulation, inflammation and hemostasis in obese adults at risk for insulin resistance. As an adjunct, green tea and cocoa flavanols were compared for their ability to modulate these biomarkers. In a randomized crossover design, 20 adults consumed a controlled diet for 5 days along with four cocoa beverages containing 30–900 mg flavanol per day, or tea matched to a cocoa beverage for monomeric flavanol content.


Cocoa beverages produced no significant changes in glucose, insulin, total area under the concentration–time curve (AUC) for glucose or total insulin AUC. As the dose of cocoa flavanols increased, total 8-isoprostane concentrations were lowered (linear contrast, P=0.02), as were C-reactive protein (CRP) concentrations (linear contrast, P=0.01). The relationship between cocoa flavanol levels and interleukin-6 (IL-6) concentrations was quadratic, suggesting that a maximum effective dose was achieved (quadratic contrast, P=0.01). There were no significant effects on measured indices of glucose regulation, nor on those of total 8-isoprostane, CRP and IL-6 concentrations, when cocoa and green tea were compared. However, relative to cocoa, green tea lowered fibrinogen concentrations (P=0.0003).


Short-term intake of cocoa and green tea flavanols does not appear to improve glucose metabolism; they do affect selected markers of one or more measures of oxidative stress, inflammation or hemostasis in obese adults at risk for insulin resistance.


Insulin resistance is a risk factor for type 2 diabetes and cardiovascular diseases.1, 2 Insulin resistance and compensatory hyperinsulinemia have been associated with hypertension, dyslipidemia and the metabolic syndrome. Both genetic and environmental factors, including obesity and physical inactivity,1, 3, 4, 5 contribute to the development of insulin resistance.

Lifestyle strategies that include dietary modification, such as consumption of a plant-based diet, are well recognized in disease prevention and may improve insulin resistance.6, 7, 8, 9 Various components of a plant-based diet may contribute to its beneficial health effects, but there has been a keen interest in the possibility that plant flavonoids may have a role. Cocoa and green tea are dietary sources of flavonoids, specifically of the flavanol subclass. Cocoa flavanols are found in the form of monomerics (epicatechin and catechin), dimers (2 units of epicatechin) and polymers. These polymers, or procyanidins, are chains of up to and over 10 monomer units.10, 11 Green tea also contains flavanols; however, flavanols in green tea are primarily in the monomeric form. These monomers are found primarily as gallated flavanol derivatives with epigallocatechin gallate reported to make up 48–55% of the flavanols in green tea.12, 13

Several studies have reported that consumption of cocoa and green tea may beneficially affect early biomarkers of cardiovascular diseases, including endothelial function, blood pressure and insulin sensitivity.12, 14, 15, 16, 17, 18 The prevalence of insulin resistance, a forerunner of diabetes, is increasing in the population. Thus, it is important to determine whether flavanol-rich foods can positively alter early biomarkers of insulin resistance and other biomarkers of cardiovascular diseases. Additionally, it is important to clarify whether flavanol monomers alone, as found in green tea, or in combination with flavanol polymers, as found in cocoa, differentially affect these biomarkers.

This exploratory study was designed with the objective of evaluating the ability of dietary cocoa flavanols to improve health indices in an at-risk group of subjects, obese adults at risk for insulin resistance. The primary outcome variables were those related to glucose metabolism, and secondary, those related to oxidative stress, inflammation, hemostasis, processes that also have prominent roles in diabetes and cardiovascular diseases. As an adjunct, green tea and cocoa flavanols were compared for their ability to modulate these biomarkers.

Materials and methods


Men and women aged 25–55 years were recruited by an advertisement from the greater Washington DC, USA metropolitan area in accord with the experimental protocol, which was approved by the MedStar Research Institute Institutional Review Board. Subjects gave their written informed consent to participate, and were compensated for their participation in the study. Inclusion in the study was based on being at increased risk for insulin resistance. A tree-based classification model was used to determine risk for insulin resistance on the basis of routine clinical measurements, including body mass index (BMI), waist circumference, fasting blood glucose concentration, blood insulin concentration, blood lipids, blood pressure and family history of diabetes mellitus.19 Exclusion criteria included a BMI <27 kg/m2, reported tobacco use, recent pregnancy or lactation, history of cardiovascular diseases, diabetes, kidney diseases, liver diseases and certain cancers. Study entry was approved by a physician on the basis of the subjects’ medical history, blood and urine test results at screening, and a physical examination.

Study design and interventions

The study had a crossover design with five 5-day treatment periods. The subjects were randomly assigned to one of two balanced Latin squares (William’s design for five treatments and five periods; ten subjects per square, two subjects per sequence within a square). Treatment periods were separated by 10-day washout periods. A treatment consisted of a beverage containing cocoa or green tea.

The subjects consumed two servings of the treatment beverage per day, one at breakfast and one at the evening meal, as a part of a controlled diet. A serving of cocoa powder weighed 28 g and a serving of tea weighed 1.2 g. The cocoa beverages provided flavanols at 30 mg per day (control), 180 mg per day (low-flavanol), 400 mg per day (medium-flavanol) and 900 mg per day (high-flavanol) (Mars Incorporated, Hackettstown, NJ, USA). Tea was provided as commercially available green tea (Lipton Green Tea To Go, Unilever, Englewood Cliffs, NJ, USA). Treatment beverages were prepared from standardized dry powders in individual packets, which is similar to instant cocoa and instant tea. The cocoa and green tea treatments were reconstituted with 150 ml of hot water or 260 ml of room-temperature bottled water per serving, respectively. The subjects did not chill, heat or add additional sweetener to the treatment beverages. Subjects, investigators and staff were blinded to the flavanol content of the three cocoa treatments, but could identify cocoa vs tea treatments.

Treatment beverages were analyzed (Mars Incorporated, and Brunswick Laboratories, Wareham, MA, USA) at the beginning and at the end of the study to determine the composition (Table 1) and verify the stability of the flavanols and other nutrients. The cocoa treatments were formulated to be similar in total kilocalories, macronutrients, micronutrients, theobromine and caffeine. The green tea treatment was chosen to reflect similar monomer content to that of the high-flavanol cocoa dose (Table 2). Daily intake of the green tea provided 36 g of caffeine and 42 kJ of energy (10 kcal). The caffeine content was similar across all the treatments.

Table 1 Daily intake provided by each of the cocoa treatment beverages
Table 2 Monomer content of the high-flavanol cocoa and the green tea treatment beverages consumed per day

Controlled diet

During the treatment periods, subjects consumed a controlled diet. A 5-day menu cycle of typical American foods that were low in polyphenols was formulated using Nutrionist Pro software (version 1.3; Axxya Systems, Stafford, TX, USA). Foods that were limited in the controlled diet included onions, apples, strawberries, oranges, tea, coffee, grape juice, chocolates and cocoa products. All foods and beverages were prepared and supplied by the Human Studies Facility at the Beltsville Human Nutrition Research Center (Beltsville, MD, USA). Food items were weighed and served in proportion to the energy requirements of subjects. Calorie levels were adjusted to compensate for the additional kilocalories provided by the cocoa beverages. Monday through Friday, the subjects consumed breakfast and dinner at the Human Studies Facility. Lunch and weekend meals were provided for carryout. The research dietitian monitored food and treatment beverage selection and consumption at weekday meals. Composites of foods in the 5-day menu cycle were prepared and analyzed for macronutrients, fatty acids, cholesterol and dietary fiber (Covance Laboratories, Madison, WI, USA). Subjects agreed to stop taking vitamin/mineral and herbal supplements 2 weeks before the study, and to avoid food and beverages containing caffeine, except those provided by the Beltsville Human Nutrition Research Center, for 4 days before the start and during the treatment periods.

The subjects were weighed Monday through Friday, before breakfast, when they arrived at the Human Studies Facility. Energy intake was adjusted in 837 kJ (200 kcal) increments so that the subjects maintained a constant body weight during the study. The subjects completed a daily questionnaire regarding their general health, use of any prescription or over-the-counter medications, factors related to dietary compliance, exercise performed, and questions or problems with the diet. The subjects were encouraged to maintain their normal exercise routine throughout the study.

Physiological variables

Blood pressure, anthropometric variables and body composition of the subjects were measured at baseline. Briefly, the subjects were seated in a quiet room for 5 min and blood pressure and heart rate were measured (Dinamap Compact Monitor, Critikon Model TS, Tampa, FL, USA) three times to obtain an average value. Waist circumference of the subjects was measured to the nearest 0.1 cm above the right ilium on the midaxillary line using a fiberglass tape measure. Body composition was measured by air-displacement plethysmography (BOD POD, Life Measurement, Incorporation, Concord, CA, USA). For these measurements, the subjects wore form-fitting spandex (men wore bike shorts and women wore bike shorts and tops) and a swim cap. The measurements were made according to the manufacturer’s guidelines. The subjects were fasted and were refrained from heavy exercise before these measurements.

Sample collection and analyses

Oral glucose tolerance tests (OGTTs) were performed the morning following the 5-day treatment periods, after an overnight fast and at least 12 h after the last treatment ingestion. Blood was sampled from intravenous catheters placed in the antecubital vein and kept patent by flushing with normal saline. Two baseline blood samples were taken. The subjects then consumed 75 g of dextrose (Trutol Glucose Tolerance Beverage, NERL Diagnostics, East Providence, Rhode Island, USA) within 10 min, and blood samples were collected at 30, 60, 90 and 120 min after the glucose load. The collected blood samples were used to prepare 0.8–2.0 ml aliquots of plasma (from EDTA and NaFl tubes) and serum that were stored at −80 °C until these were analyzed. Analytes measured at all time points included glucose, insulin, triglycerides, high-sensitivity C-reactive protein (CRP), soluble intercellular adhesion molecule-1 (ICAM) and interleukin-6 (IL-6). Samples for analysis of total 8-isoprostane and fibrinogen were collected at baseline and time points 90 and 60 min, respectively. OGTT values (glucose, insulin and triglyceride concentrations) were used for the homeostasis model assessment of insulin resistance (HOMA-IR), the quantitative insulin sensitivity check index (QUICKI) and the insulin sensitivity index (ISI).20, 21, 22, 23, 24, 25 Serum triglyceride concentrations were determined enzymatically with a commercial kit (Johnson & Johnson/Ortho Clinical Diagnostics, Rochester, NY, USA). Insulin concentrations were measured using an enzyme-multiplied immunoassay method (LINCOplex, LINCO Research, St Charles, MO, USA). Glucose concentrations were measured using standard automated enzymatic procedures on an Olympus automated analyzer (Smith-Kline Beecham Laboratories, Madison, NJ, USA). Analyses of serum triglyceride, insulin and glucose concentrations were performed at Medstar Research Institute, Penn Laboratories, Washington, DC, USA. High-sensitivity CRP concentrations were measured using an enzyme-linked immunosorbent assay (ELISA; Diagnostic Automation Incorporated, Calabasas, CA, USA). ICAM and IL-6 concentrations were measured using an ELISA (R&D Systems, Minneapolis, MN, USA). Fibrinogen was measured in an automated clot assay by using the ACL 1000 coagulation analyzer (Beckman Coulter Incorporated, Fullerton, CA, USA). Analyses of high-sensitivity CRP, ICAM, IL-6 and fibrinogen concentrations were performed at the Food Components and Health Laboratory, Beltsville Human Nutrition Research, US Department of Agriculture, Beltsville, MD, USA. Total (free and esterfied) 8-isoprostane levels were measured at Caymen Chemical (Ann Arbor, MI, USA) using the Caymen Chemical enzyme immunoassay.

Statistical analyses

Data were analyzed by analysis of covariance using a mixed-model procedure appropriate for a crossover study with repeated measures (Version 9; SAS, SAS Institute, Cary, NC, USA). The statistical model included square, treatment, time and the time-by-treatment interaction as fixed effects. Subject nested within square and period nested within square were included as random effects in the model. Time 0 measurements within a period were included as covariates. Within time–treatment effects were evaluated when the time-by-treatment interaction was statistically significant (P<0.05); otherwise, the main effect of treatment was evaluated. If the treatment effect was statistically significant (either within time or when the main effect of treatment was evaluated), three contrasts were evaluated. First, the linear relationship between dose (30, 180, 400 or 900 mg flavanols) and outcome was assessed by evaluating the linear and quadratic contrasts. Second, differences between the green tea and high-flavanol cocoa groups were assessed to evaluate whether flavanol monomers alone or in combination with flavanol polymers altered the outcomes of interest. Glucose and insulin concentrations were log-transformed to yield a normal distribution before analysis. These values are reported, back-transformed, as least-squares means±confidence intervals. Other data, nontransformed, are presented as least-squares means±s.e.m.’s.

Incremental and total area under the concentration–time curve (AUC) data were analyzed by repeated-measures ANCOVA, using a mixed-model procedure appropriate for a crossover study (Version 9; SAS, SAS Institute). The statistical model included square and treatment as fixed effects. Subject nested within square and period nested within square were included as random effects in the model. Period was treated as a repeated measure. Covariates included in the model were BMI, gender, age and fat weight. The assumption of normality was evaluated by inspection of the residuals. Differences between treatment groups were assessed by comparing all pairs of means (Tukey adjusted P-values).

Sample size was estimated using data from previous pilot studies of similar designs conducted at the Beltsville Human Nutrition Research Center. In these studies, the standard deviation of treatment differences for glucose AUC was 440 mg/h per 100 ml. To detect a 30% change in glucose AUC with α=0.05 and power=80%, the estimated sample size was 19. SigmaStat for Windows (Version 3.5; Aspire Software International, Ashburn, VA, USA) was used for sample size estimations.26


Subject characteristics

Out of the 201 persons who attended the study information meetings, 156 signed an informed consent and 130 completed the screening process. Thirty-nine persons were determined to have a probability of at least 84−94% of being insulin resistant on the basis of Stern et al.’s19 classification tree models for insulin resistance. Ultimately, 20 subjects, that is, 10 women and 10 men, were randomly assigned to the treatments and completed all the treatment periods. The OGTT results were normal in 65% (n=13) of the subjects, and 35% (n=6) were classified as having impaired glucose tolerance. Five percent (n=1) were classified as having type 2 diabetes on the basis of the classification of diabetes mellitus from the American Diabetes Association.27 The data for the subject with type 2 diabetes were excluded from the analysis. Thus, complete data were analyzed and are presented for 19 subjects. Forty-two percent of the subjects reported a family history of diabetes mellitus. The BMI of the subjects (mean 36.8 kg/m2; range 30−45 kg/m2) indicated that they were obese. The baseline physical characteristics of the 19 subjects that are included in the final data are presented in Table 3.

Table 3 Baseline subject characteristics a

Diets and compliance

The experimental diets provided on average 54% of calories from carbohydrates, 32% from fat and 14% from protein. Total dietary fat had a ratio of polyunsaturated:monounsaturated:saturated fatty acids of 0.7:0.8:1.1. At an average daily kilocalorie level of 2700, the diet provided 297 mg of cholesterol and 24 g of dietary fiber daily. Flavanols and other nutrients in the treatment beverages were found to be stable at the end of the study when compared with analyses at the beginning of the study (data not shown). Body weight was measured each weekday morning, and once the subjects’ estimated energy requirement was determined, constancy of body weight without further changes in energy was considered to be an indicator of compliance. Body weight did not change significantly throughout the study (pre- and post-treatment means 108.8±4.2 and 109.1±4.2 kg, respectively, P>0.05).

Glucose regulation biomarkers

Cocoa treatments did not significantly alter concentrations of glucose, insulin or triglyceride, whether measured after an overnight fast or in response to the OGTT (Figure 1). Similarly, values for AUCs for glucose and insulin were not different among the treatments. Surrogate markers of insulin resistance (HOMA-IR) and insulin sensitivity (ISI, QUICKI) were not significantly affected by the cocoa treatments (Table 4). There were also no significant effects on concentrations of glucose, insulin and triglyceride; values for AUCs for glucose and insulin; and surrogate markers of insulin resistance and insulin sensitivity when the high-flavanol cocoa was compared with the green tea.

Figure 1

Plasma glucose, insulin and serum triglyceride concentrations during the OGTT. Data are presented as least-squares means±confidence intervals, n=19 (10 women, 9 men), from repeated-measures ANOVA (analysis of variance). There were no significant treatment effects between the control (), low (), medium (), and high () flavanol cocoas.

Table 4 Effect of the 5-day cocoa-dose treatments and green tea on biomarkers of glucose regulation a

Oxidative stress, inflammatory response and hemostatic biomarkers

The cocoa flavanols significantly affected the biomarkers of oxidative stress and inflammation (total 8-isoprostane, CRP and IL-6 concentrations) (Table 5). As the cocoa flavanol dose increased, the total 8-isoprostane concentrations were lowered (linear contrast significant, P=0.02). Similarly, as the cocoa flavanol dose increased, the CRP concentrations were lowered (linear contrast significant, P=0.01). A quadratic relationship between cocoa flavanol levels and IL-6 concentrations was observed (quadratic contrast significant, P=0.01), suggesting that a maximum effective dose was achieved. No significant differences were observed among the cocoa treatments on ICAM and fibrinogen concentrations. There were no significant effects on total 8-isoprostane, CRP, IL-6 and ICAM concentrations when the high-flavanol cocoa and green tea treatments were compared (Table 6). However, when the green tea treatment was compared with the high-flavanol cocoa, the green tea was found to lower fibrinogen concentrations (P=0.0003) by 7%.

Table 5 Effect of the 5-day cocoa-dose treatments on biomarkers of oxidative stress, inflammation and hemostasis a
Table 6 Effect of the 5-day high-dose cocoa and the green tea treatment on biomarkers of oxidative stress, inflammation and hemostasis a


This study showed no significant changes in glucose, insulin and triglyceride responses to the OGTT challenge after consuming cocoa beverages for 5 days along with a highly-controlled diet. Additionally, values for the AUCs for glucose and insulin were not different among the cocoa treatments. There was also no dose–response effect on surrogate markers of insulin resistance, HOMA-IR and insulin sensitivity: QUICKI and ISI. Intervention studies of the effect of cocoa or chocolate consumption on glucose regulation have shown equivocal results. Some investigators who used different techniques and populations have shown no improvement in glucose metabolism. Muniyappa et al.18 found that daily consumption of cocoa containing 900 mg flavanols for 14 days did not improve insulin resistance, as determined by the hyperinsulinemic glucose clamp technique in adults with essential hypertension; surrogate markers for insulin sensitivity were not affected. In a randomized trial, adults with type 2 diabetes who consumed cocoa containing 963 mg or 75 mg of flavanols for 30 days had no significant changes in glucose, hemoglobin A1c or triglycerides.28 However, intervention with dark chocolate as a confectionary (providing 500 mg flavanols for 14 days), as opposed to cocoa as a beverage, have shown positive results on glucose metabolism and insulin sensitivity. Dark chocolate improved surrogate markers of insulin sensitivity in both healthy and hypertensive individuals.29, 30 These studies did not control for the habitual diet or for other bioactive components, such as caffeine and theobromine, found in chocolate and cocoa. Our study controlled for these components across the cocoa treatments. Further research is required to differentiate the flavanol component of cocoa and chocolate from other naturally occurring bioactive compounds with regard to glucose homeostasis and related markers.

Obesity and insulin resistance may be significant determinants of oxidative stress.31, 32 Isoprostanes, a family of eicosanoids produced mainly through the non-enzymatic oxidation of arachidonic acid by reactive oxygen species, are recognized biomarkers of in vivo lipid peroxidation,33 and the production of isoprostanes is increased in the presence of oxidative stress. In the current study, we observed a linear relationship between consumption of the medium-flavanol cocoa and total 8-isoprostane concentrations. We found that the medium-flavanol cocoa (400 mg/day) consumed for 5 days decreased the plasma levels of total 8-isoprostanes, which was consistent with the reduced oxidative stress in our subject population. Effects of cocoa consumption on oxidative stress have not previously been reported for obese individuals at risk for insulin resistance. Wiswedel et al.,34 however, conducted an acute randomized crossover trial in healthy men consuming a cocoa beverage containing either 187 mg or 14 mg of flavanols with isoprostane concentrations determined at 2, 4 and 6 h after consumption. The high-flavanol cocoa, combined with physical exercise, was found to lower isoprostane concentrations. Given these findings and the paucity of data on flavanols and oxidative status, research is warranted to better understand how chronic daily cocoa consumption affects biomarkers of oxidative stress and related health outcomes.

Circulating markers of inflammation, such as IL-6, a proinflammatory cytokine, ICAM and high-sensitivity CRP, an early response marker, have been identified as potential predictors of present and future risk of cardiovascular disease in individuals.35 A quadratic relationship was observed with consumption of the cocoa flavanols and IL-6 concentrations, suggesting that a maximum dose was achieved with the medium-flavanol cocoa (400 mg). Additionally, a linear relationship was observed with cocoa flavanols and high-sensitivity CRP concentrations, which shows that as the dose of cocoa flavanols increased, high-sensitivity CRP concentrations were lowered. There was no effect of the cocoa flavanols on ICAM. There are limited data on the dose–response effects of cocoa flavanol consumption on biomarkers of inflammation in obese adults at risk for insulin resistance. However, Muniyappa et al.18 showed no effect of high-flavanol cocoa (900 mg) consumed for 14 days on ICAM and IL-6 in men and women with essential hypertension. Similarly, in the same subject population, Grassi et al.30 found that flavanol-rich dark chocolate (88 mg) consumed for 15 days did not affect high-sensitivity CRP or ICAM. These results may reflect differences in dose of cocoa or chocolate flavanols or subjects’ health status. Flavanols in cocoa vary among cocoa and chocolate products; milk chocolate or dark chocolate, and natural cocoa powder contain 3, 14 and 40 mg/g of flavanols, respectively.36 Longer-term clinical trials are needed to confirm the protective effects of cocoa consumption in those at risk for insulin resistance.

In an attempt to compare flavanol-containing beverages, the current study evaluated the effects of green tea and the high-flavanol cocoa (900 mg/day) on biomarkers of glucose regulation, oxidative stress, inflammation and hemostasis. The green tea treatment was chosen to reflect a monomer content similar to that of the high-flavanol cocoa treatment. Cocoa contains similar flavanols, predominantly in the form of monomers and oligomers (that is, chains of catechins). Our comparison of high-flavanol cocoa with green tea identified no differences for biomarkers of glucose regulation, oxidative stress or inflammation, even though the tea beverage contained a higher monomer and total flavanol content. However, green tea significantly lowered fibrinogen, an acute-phase protein involved in blood clot formation. Increased fibrinogen concentration is an independent risk factor for cardiovascular diseases, including ischemic heart disease and stroke.37, 38 Additionally, a meta-analysis showed that daily green tea consumption may prevent the onset of ischemic stroke.39 Few human studies have evaluated the effects of consumption of green tea on hemostasis. de Maat et al.40 reported that consuming green tea for 28 days had no effect on hemostasis in normal-weight healthy smokers. Further long-term experimental studies are required to confirm the potential role of green tea consumption on hemostasis in individuals with obesity, insulin resistance and diabetes.

The strengths and limitations of our study warrant consideration. The choice of study design, a randomized crossover design, is one of the most powerful designs for evaluating the efficacy of dietary treatments.41 Additionally, the study was conducted with a high degree of dietary control. All foods were provided to the subjects to control for confounding dietary factors, a research dietitian monitored the subjects’ treatment intake to ensure adherence to the study protocol and all subjects’ weights were kept constant throughout the study. The treatments provided a broad range of flavanol intakes to enhance the relevance of the data, given that optimal intake levels are unknown, but vary widely among countries and among various dietary patterns. Results for cocoa treatments were directly due to flavanol content; confounding bioactive compounds such as theobromine were held constant. There were, however, several limitations to the study. This was an exploratory study with a small sample size and the study may not have been adequately powered to detect existing differences for all outcome variables. Intervention duration is another consideration. Our intervention of 5 days was relatively short, as different results are possible with longer exposure. Several previous studies have shown that short-term interventions, ranging from an acute dose to 7 days of consumption, with cocoa or tea components may produce significant changes in glucose metabolism and other biomarkers of disease.42, 43, 44 However, long-term studies on this topic are desirable to ensure that metabolic changes have been stabilized. Further, our results may have been different if the cocoa and tea treatments had been consumed along with the glucose tolerance challenge rather than 12 h or more before the challenge. Dietary flavonoids are rapidly metabolized; for example, after consumption of green tea, catechin concentrations in human plasma reach their peak within 1.5−2 h and decline to undetectable levels after 24 h.45

The results of this exploratory study indicate that short-term consumption of flavanols from cocoa and green tea do not improve glucose metabolism in obese adults at risk for insulin resistance. However, consumption of cocoa and green tea improved certain biomarkers of oxidative stress, inflammation and hemostasis in this population. Given that these processes likely promote diabetes, cardiovascular diseases and other chronic diseases, long-term studies of flavanols are warranted particularly in at-risk populations.


  1. 1

    Meigs JB, Rutter MK, Sullivan LM, Fox CS, D’Agostino RB, Wilson PW . Impact of insulin resistance on risk of type 2 diabetes and cardiovascular disease in people with metabolic syndrome. Diabetes Care 2007; 30: 1219–1225.

    CAS  Article  Google Scholar 

  2. 2

    Lorenzo C, Wagenknecht LE, D’Agostino RB, Rewers MJ, Karter AJ, Haffner SM . Insulin resistance, beta-cell dysfunction, and conversion to type 2 diabetes in a multiethnic population: the Insulin Resistance Atherosclerosis Study. Diabetes Care 2010; 33: 67–72.

    CAS  Article  Google Scholar 

  3. 3

    Reaven G . Insulin resistance, type 2 diabetes mellitus, and cardiovascular disease: the end of the beginning. Circulation 2005; 112: 3030–3032.

    Article  Google Scholar 

  4. 4

    Reaven GM . Pathophysiology of insulin resistance in human disease. Physiol Rev 1995; 75: 473–486.

    CAS  Article  Google Scholar 

  5. 5

    Eckel RH, Kahn SE, Ferrannini E, Goldfine AB, Nathan DM, Schwartz MW et al. Obesity and type 2 diabetes: what can be unified and what needs to be individualized? Diabetes Care 2011; 34: 1424–1430.

    Article  Google Scholar 

  6. 6

    Esposito K, Maiorino MI, Ceriello A, Giugliano D . Prevention and control of type 2 diabetes by Mediterranean diet: a systematic review. Diabetes Res Clin Pract 2010; 89: 97–102.

    CAS  Article  Google Scholar 

  7. 7

    Salas-Salvado J, Bullo M, Babio N, Martinez-Gonzalez MA, Ibarrola-Jurado N, Basora J et al. Reduction in the incidence of type 2 diabetes with the Mediterranean diet: results of the PREDIMED-Reus nutrition intervention randomized trial. Diabetes Care 2011; 34: 14–19.

    Article  Google Scholar 

  8. 8

    Serra-Majem L, Roman B, Estruch R . Scientific evidence of interventions using the Mediterranean diet: a systematic review. Nutr Rev 2006; 64 (Part 2), S27–S47.

    Article  Google Scholar 

  9. 9

    Shai I, Schwarzfuchs D, Henkin Y, Shahar DR, Witkow S, Greenberg I et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med 2008; 359: 229–241.

    CAS  Article  Google Scholar 

  10. 10

    Cooper KA, Donovan JL, Waterhouse AL, Williamson G . Cocoa and health: a decade of research. Br J Nutr 2007; 99: 1–11.

    Article  Google Scholar 

  11. 11

    Engler MB, Engler MM . The emerging role of flavonoid-rich cocoa and chocolate in cardiovascular health and disease. Nutr Rev 2006; 64: 109–118.

    Article  Google Scholar 

  12. 12

    Basu A, Lucas EA . Mechanisms and effects of green tea on cardiovascular health. Nutr Rev 2007; 65 (Part 1), 361–375.

    Article  Google Scholar 

  13. 13

    Cabrera C, Artacho R, Gimenez R . Beneficial effects of green tea--a review. J Am Coll Nutr 2006; 25: 79–99.

    CAS  Article  Google Scholar 

  14. 14

    Corti R, Flammer AJ, Hollenberg NK, Luscher TF . Cocoa and cardiovascular health. Circulation 2009; 119: 1433–1441.

    Article  Google Scholar 

  15. 15

    Fukino Y, Ikeda A, Maruyama K, Aoki N, Okubo T, Iso H . Randomized controlled trial for an effect of green tea-extract powder supplementation on glucose abnormalities. Eur J Clin Nutr 2007; 62: 953–960.

    Article  Google Scholar 

  16. 16

    Grassi D, Necozione S, Lippi C, Croce G, Valeri L, Pasqualetti P et al. Cocoa reduces blood pressure and insulin resistance and improves endothelium-dependent vasodilation in hypertensives. Hypertension 2005; 46: 398–405.

    CAS  Article  Google Scholar 

  17. 17

    Shrime MG, Bauer SR, McDonald AC, Chowdhury NH, Coltart CE, Ding EL . Flavonoid-rich cocoa consumption affects multiple cardiovascular risk factors in a meta-analysis of short-term studies. J Nutr 2011; 141: 1982–1988.

    CAS  Article  Google Scholar 

  18. 18

    Muniyappa R, Hall G, Kolodziej TL, Karne RJ, Crandon SK, Quon MJ . Cocoa consumption for 2 wk enhances insulin-mediated vasodilatation without improving blood pressure or insulin resistance in essential hypertension. Am J Clin Nutr 2008; 88: 1685–1696.

    CAS  Article  Google Scholar 

  19. 19

    Stern SE, Williams K, Ferrannini E, DeFronzo RA, Bogardus C, Stern MP . Identification of individuals with insulin resistance using routine clinical measurements. Diabetes 2005; 54: 333–339.

    CAS  Article  Google Scholar 

  20. 20

    Bonora E, Kiechl S, Willeit J, Oberhollenzer F, Egger G, Targher G et al. Prevalence of insulin resistance in metabolic disorders: the Bruneck Study. Diabetes 1998; 47: 1643–1649.

    CAS  Article  Google Scholar 

  21. 21

    Katz A, Nambi SS, Mather K, Baron AD, Follmann DA, Sullivan G et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000; 85: 2402–2410.

    CAS  Article  Google Scholar 

  22. 22

    Mannucci E, Bardini G, Ognibene A, Rotella CM . Comparison between 2 insulin sensitivity indexes in obese patients. Diabetes Care 2000; 23: 1042–1043.

    CAS  Article  Google Scholar 

  23. 23

    Matsuda M, DeFronzo RA . Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999; 22: 1462–1470.

    CAS  Article  Google Scholar 

  24. 24

    Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC . Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419.

    CAS  Article  Google Scholar 

  25. 25

    Antuna-Puente B, Disse E, Rabasa-Lhoret R, Laville M, Capeau J, Bastard JP . How can we measure insulin sensitivity/resistance? Diabetes Metab 2011; 37: 179–188.

    CAS  Article  Google Scholar 

  26. 26

    Baer DJ, Novotny JA, Harris GK, Stote K, Clevidence B, Rumpler WV . Oolong tea does not improve glucose metabolism in non-diabetic adults. Eur J Clin Nutr 2011; 65: 87–93.

    CAS  Article  Google Scholar 

  27. 27

    American Diabetes Association Diagnosis and classification of diabetes mellitus. Diabetes Care 2011; 34: S62–S69.

    Article  Google Scholar 

  28. 28

    Balzer J, Rassaf T, Heiss C, Kleinbongard P, Lauer T, Merx M et al. Sustained benefits in vascular function through flavanol-containing cocoa in medicated diabetic patients a double-masked, randomized, controlled trial. J Am Coll Cardiol 2008; 51: 2141–2149.

    CAS  Article  Google Scholar 

  29. 29

    Grassi D, Desideri G, Necozione S, Lippi C, Casale R, Properzi G et al. Blood pressure is reduced and insulin sensitivity increased in glucose-intolerant, hypertensive subjects after 15 days of consuming high-polyphenol dark chocolate. J Nutr 2008; 138: 1671–1676.

    CAS  Article  Google Scholar 

  30. 30

    Grassi D, Lippi C, Necozione S, Desideri G, Ferri C . Short-term administration of dark chocolate is followed by a significant increase in insulin sensitivity and a decrease in blood pressure in healthy persons. Am J Clin Nutr 2005; 81: 611–614.

    CAS  Article  Google Scholar 

  31. 31

    Keaney JF, Larson MG, Vasan RS, Wilson PW, Lipinska I, Corey D et al. Obesity and systemic oxidative stress: clinical correlates of oxidative stress in the Framingham Study. Arterioscler Thromb Vasc Biol 2003; 23: 434–439.

    CAS  Article  Google Scholar 

  32. 32

    Meigs JB, Larson MG, Fox CS, Keaney JF, Vasan RS, Benjamin EJ . Association of oxidative stress, insulin resistance, and diabetes risk phenotypes: the Framingham Offspring Study. Diabetes Care 2007; 30: 2529–2535.

    CAS  Article  Google Scholar 

  33. 33

    Roberts CK, Sindhu KK . Oxidative stress and metabolic syndrome. Life Sci 2009; 84: 705–712.

    CAS  Article  Google Scholar 

  34. 34

    Wiswedel I, Hirsch D, Kropf S, Gruening M, Pfister E, Schewe T et al. Flavanol-rich cocoa drink lowers plasma F(2)-isoprostane concentrations in humans. Free Radic Biol Med 2004; 37: 411–421.

    CAS  Article  Google Scholar 

  35. 35

    Ridker PM, Hennekens CH, Buring JE, Rifai N . C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 2000; 342: 836–843.

    CAS  Article  Google Scholar 

  36. 36

    Gu L, House SE, Wu X, Ou B, Prior RL . Procyanidin and catechin contents and antioxidant capacity of cocoa and chocolate products. J Agric Food Chem 2006; 54: 4057–4061.

    CAS  Article  Google Scholar 

  37. 37

    Danesh J, Lewington S, Thompson SG, Lowe GD, Collins R, Kostis JB et al. Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: an individual participant meta-analysis. JAMA 2005; 294: 1799–1809.

    CAS  Google Scholar 

  38. 38

    Kannel WB . Overview of hemostatic factors involved in atherosclerotic cardiovascular disease. Lipids 2005; 40: 1215–1220.

    CAS  Article  Google Scholar 

  39. 39

    Arab L, Liu W, Elashoff D . Green and black tea consumption and risk of stroke: a meta-analysis. Stroke 2009; 40: 1786–1792.

    CAS  Article  Google Scholar 

  40. 40

    de Maat MP, Pijl H, Kluft C, Princen HM . Consumption of black and green tea had no effect on inflammation, haemostasis and endothelial markers in smoking healthy individuals. Eur J Clin Nutr 2000; 54: 757–763.

    CAS  Article  Google Scholar 

  41. 41

    Monsen E Research: Successful Approaches 3rd edn, Monsen E, (ed), American Dietetic Association: Chicago, Illinois, USA, 2008.

    Google Scholar 

  42. 42

    Grassi D, Mulder TP, Draijer R, Desideri G, Molhuizen HO, Ferri C . Black tea consumption dose-dependently improves flow-mediated dilation in healthy males. J Hypertens 2009; 27: 774–781.

    CAS  Article  Google Scholar 

  43. 43

    Vazquez-Agell M, Urpi-Sarda M, Sacanella E, Camino-Lopez S, Chiva-Blanch G, Llorente-Cortes V et al. Cocoa consumption reduces NF-kappaB activation in peripheral blood mononuclear cells in humans. Nutr Metab Cardiovasc Dis 2011 ; e-pub ahead of print 6 August 2011 doi:10.1016/j.numecd.2011.03.015.

  44. 44

    Venables MC, Hulston CJ, Cox HR, Jeukendrup AE . Green tea extract ingestion, fat oxidation, and glucose tolerance in healthy humans. Am J Clin Nutr 2008; 87: 778–784.

    CAS  Article  Google Scholar 

  45. 45

    Henning SM, Niu Y, Lee NH, Thames GD, Minutti RR, Wang H et al. Bioavailability and antioxidant activity of tea flavanols after consumption of green tea, black tea, or a green tea extract supplement. Am J Clin Nutr 2004; 80: 1558–1564.

    CAS  Article  Google Scholar 

Download references


We thank Mars Incorporated for providing the cocoa treatment beverages. We also thank Dr Catherine Kwik-Uribe for providing flavanol analysis of the cocoa treatment beverages and thoughtful manuscript review. This study was supported by the US Department of Agriculture.

Author information



Corresponding author

Correspondence to K S Stote.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Stote, K., Clevidence, B., Novotny, J. et al. Effect of cocoa and green tea on biomarkers of glucose regulation, oxidative stress, inflammation and hemostasis in obese adults at risk for insulin resistance. Eur J Clin Nutr 66, 1153–1159 (2012).

Download citation


  • cocoa
  • green tea
  • polyphenols
  • insulin resistance
  • obesity

Further reading