African Americans May Have to Consume More Than 12 Grams a Day of Resistant Starch to Lower Their Risk for Type 2 Diabetes

Abstract

African Americans have a high prevalence rate of type 2 diabetes mellitus (DM). High-maize™ 260 (National Starch and Chemical Co., Bridgewater, NJ, USA) resistant starch (RS) is a promising food ingredient to reduce risk factors for type 2 DM. A 14-week, double-blind, crossover design study was conducted with African American male (n = 8) and female (n = 7) subjects at risk for type 2 DM. All subjects consumed bread containing 12 g of added RS or control bread (no added RS) for 6 weeks, separated by a 2-week washout period. There were no significant differences in the subjects' fasting plasma glucose levels due to the consumption of the RS bread versus the control bread. Fructosamine levels were significantly lower after consumption of both RS and control bread than at baseline. However, we found no significant difference in fructosamine levels due to treatment effects, i.e., RS bread intake versus the control bread. There were no significant differences in insulin or C-reactive protein levels due to treatment, gender, or sequence effects. Mean homeostasis model assessment of insulin resistance decreased to normal values (>2.5) at the end of the 14-week study, although there were no significant treatment effects. The results of this study suggest that African Americans may need to consume more than 12 g/day of RS to lower their risk for type 2 DM.

Introduction

D iabetes mellitus (DM) is a group of metabolic disorders characterized by hyperglycemia.¹ The prevalence of diabetes in the United States increased by 120% between the years 1980 and 2005.² Risk factors for type 2 DM, the most common form of diabetes, include excess body weight (body mass index [BMI] >25 kg/m²), sedentary lifestyle practices, unhealthy dietary practices, and being of African American, Native American, or Latino American descent.³ The percentage of African American males and females diagnosed with diabetes is expected to increase by 363% and 217%, respectively, between the years 2000 and 2050.⁴

Starch is a primary source of easily digested food carbohydrate.⁵ However, variable portions of food starches are not digested in the small intestine.^6,7 This undigested fraction of starch is referred to as resistant starch (RS). RS as defined by Asp⁸ is “the sum of starch and products of starch degradation not absorbed in the small intestines,” the majority of which is undigested upon entering the large bowel.⁷ RS consumption has been linked to reduced risk of colon cancer^9,10 and ulcerative colitis.^11,12

Because of its extremely low glycemic index RS may also be of benefit in the prevention and treatment of type 2 DM. Results from studies examining the effects of RS intake on postprandial glucose and insulin responses have ranged from no change^13,14 to little change^15,16 to significantly decreased postprandial glucose and insulin levels.^17

–22 Only a few studies have fed RS to human subjects for longer than 3 weeks.^13,18,23,24 Significant reductions in fasting blood glucose levels were reported in one of these studies in subjects who had consumed 30 g of RS for 4 weeks.²⁴ There have been no investigations to date, however, on the effects of RS in African Americans, nor have there been any studies that have attempted to determine the minimum amount of RS that must be ingested on a daily basis to lower fasting blood glucose levels to normal in people at risk for type 2 DM. The major objective of this study, therefore, was to determine if daily consumption of 12 g of High-maize™ 260 (National Starch and Chemical Co., Bridgewater, NJ, USA) RS added to bread could improve glucose homeostasis in African American subjects with risk factors for type 2 DM.

Subjects and Methods

Recruitment and screening of subjects

Eight African American men and nine African American women (mean age, 36.6 ± 1.55 years) who indicated they had a BMI of ≥25 kg/m², according to a provided BMI chart,²⁵ had a first- or second-degree relative with type 2 diabetes, were overweight, exercised ≤30 minutes per week, and answered yes to two or more of the statements listed in the “Simple Pre-Diabetes Risk Assessment Checklist” were invited to participate in the study. Women who were pregnant or individuals who had been previously diagnosed with type 1 or type 2 DM were excluded from the study. In addition, any persons who were currently taking any medication to control blood glucose levels were not permitted to participate. Two of the female participants originally recruited were unable to complete the study. One had to undergo back surgery, and the other was unable to tolerate the required blood samples.

Procedures

Each participant was randomly assigned to a 6-week feeding treatment period of Hi-maize 260 RS added to bread or a 6-week feeding treatment period of control bread, which contained no added RS. Nine hundred sixty grams of Hi-maize 260 donated to us by the National Starch and Chemical Co. was added to the dough mixture, which was divided into 11 loaves. Each loaf contained approximately 87 g of Hi-maize 260 prior to baking. Each bread loaf contained 21 slices, excluding the ends of the bread loaves. Subjects were asked to eat three slices of bread daily, including weekends during each 6-week study period. Any slices of bread that were not eaten on one day were to be eaten the following day. Each slice of Hi-maize 260 bread contained 4.13 g of Hi-maize 260 prior to baking. Therefore each subject was asked to consume 12.39 g of Hi-maize 260 prior to baking, during the 6-week period he or she consumed the RS bread.

A 2-week washout period, during which no RS or control bread was eaten, followed the first 6-week bread feeding period in order to minimize carryover effects. After the 2-week washout period, subjects were crossed over: those who consumed the bread containing RS during the initial 6-week feeding period consumed the control bread during weeks 9–14 of the study, and those who consumed the control bread during the initial 6-week feeding bread consumed the RS bread during weeks 9–14.

Blood samples were drawn at the beginning of the study (baseline), after the first 6-week feeding period, at the end of the 2-week washout period, and at the end of the last 6-week feeding period. Blood was drawn from subjects after they underwent a 10–12-hour overnight fast (only water and necessary medications were allowed). Blood analyses were performed four times during the study period for fasting plasma glucose (FPG), fructosamine, C-reactive protein (CRP), insulin, and short-chain fatty acid (SCFA) levels, including serum acetate, serum propionate, and serum butyrate. Hemoglobin A1c measurements were determined prior to the beginning of the study and at the end of the study.

The RS content of the control and RS bread loaves consumed by the subjects was analyzed according to AOAC Method 2002.2²⁶ using the Megazyme™ (Wicklow, Ireland) RS assay procedure. FPG was determined by the Stanbio (Boerne, TX, USA) glucose Liquicolor^® procedure number 1070. Fructosamine was determined by the Pointe Scientific (Lincoln Park, MI, USA) fructosamine reagent set protocol. Hemoglobin A1c was measured using the Gycohemoglobin Pre-Fil procedure number P350 (Stanbio). Insulin and CRP levels were determined by enzyme-linked immunosorbent assay (ELISA) using the human insulin ELISA (kit number EZH1) from Linco Research, Inc. (St. Charles, MO, USA) and the human CRP ELISA (kit number 1000) from Alpha Diagnostic International (San Antonio, TX, USA). The homeostatic model assessment (HOMA) for assessing beta-cell function (HOMA-Beta) and insulin resistance index (HOMA-IR) were calculated according to the formulas originally described by Matthews et al.²⁷ as modified by Osei et al. ²⁸ Serum acetate, propionate, and butyrate levels were determined by gas chromatography following the procedures of Wolever et al.²⁹

Anthropometric data (weight, height, blood pressure, and waist circumference) were obtained four times during the study (at the beginning of the study [baseline], the end of the first 6-week feeding period, the end of the 2-week washout period, and the end of the last 6-week period). The Gladys Block Dietary Questionnaire was administered at baseline and the end of the 14-week study in order to assess dietary practices and to determine nutrient intakes.³⁰ The questionnaire was completed by each subject following written and verbal instructions. A bowel-symptom questionnaire was provided for each subject in the event they experienced gastrointestinal discomfort. A compliance survey sheet was also provided to gauge compliance. The subjects were asked not to alter their normal physical activity routines during the entire study period. Subjects were asked to substitute the study bread for the bread they would normally eat.

Statistical analysis

Statistical analyses were performed using Statistical Analysis for Windows version 9.1 (SAS Institute, Inc., Cary, NC, USA). The SAS Proc Mixed procedure was used for calculations. All results are reported as mean ± SEM values. We used a statistical model that contained gender, sequence (RS or control bread first), treatment (baseline, control, RS, washout), and their interactions. The Bonferroni method was used for post hoc testing. Statistical analysis of dietary data and hemoglobin A1c values was performed using Student's t test. A value of P < .05 was considered statistically significant.

Results

RS content of the bread

We found in preliminary studies that we could not add more than 87 g of Hi-maize 260 RS to bread loaves without detrimental effects to bread quality, including lowered loaf volume and hardening of the bread crumb (data not shown). The RS content of the control loaves of bread ranged from 1.91 g to 3.21 g/100 g, with a mean of 2.64 g/100 g (Table 1), whereas the RS content of the RS loaves of bread for the study ranged from 8.7 g/100 g to 11.0 g/100 g, with a mean RS content of 10.2 g/100 g. The non-resistant, soluble starch content of the control loaves of bread ranged from 65.4 to 70.6 g/100 g, with a mean of 67.2 g/100 g. The soluble starch content of the RS loaves ranged from 58.5 to 69.1 g/100 g with a mean of 64.1 g/100 g.

Table 1.

RS Content of Bread Loaves

	Starch content (g/100 g)
Type of bread	Resistant	Soluble
Control bread	2.64 ± 0.13	67.22 ± 0.63
RS bread	10.17 ± 0.23	64.09 ± 1.18

Data are mean ± SEM values.

Anthropometric data

Anthropometric characteristics for male and female subjects are presented in Table 2. No significant changes occurred in mean weight, BMI, and systolic and diastolic blood pressure of the 15 subjects during the 14-week study period. Waist circumferences were significantly smaller at weeks 6, 8, and 14 than at baseline (P < .05). There were no significant differences in waist circumference due to treatment, i.e., consumption of RS bread versus the control bread.

Table 2.

Anthropometric Data for the Study Subjects

Variable	Baseline	Week 6	Week 8	Week 14
Age (years)	36.6 ± 1.55	36.6 ± 1.55	36.6 ± 1.55	36.6 ± 1.55
Weight (pounds)	256.0 ± 18.0	254.2 ± 15.7	255.7 ± 15.9	253.6 ± 15.4
Height (inches)	69.17 ± 1.2	69.17 ± 1.2	69.17 ± 1.2	69.17 ± 1.2
BMI (kg/m²)	37.7 ± 2.0	37.4 ± 1.9	37.7 ± 1.9	37.3 ± 1.8
WC (inches)	43.9 ± 2.1	43.7 ± 1.7^*	43.3 ± 1.8^*	43.2 ± 1.8^*
SBP (mm Hg)	119.7 ± 6.98	117.9 ± 3.6	117.5 ± 3.2	119.7 ± 3.6
DBP (mm Hg)	79.6 ± 2.3	78.0 ± 2.2	76.1 ± 2.6	81.6 ± 1.7

Data are mean ± SEM values.

Indicates significance at P < .05.

DBP, diastolic blood pressure; SBP, systolic blood pressure; WC, waist circumference.

Dietary data

Dietary analysis demonstrated no significant differences between baseline values (defined as 1 year pre-intervention) and follow-up (defined as weeks 1–14), with one exception: fruit servings were significantly greater during the follow-up period (P = .040) than at baseline (data not shown). The mean total carbohydrate and mean glucose intake of the subjects' diet did not change during the follow-up period despite the addition of the control and RS bread to their diet. Whole grain intake went up during the follow-up period, but not significantly.

Blood glucose

The mean ± SEM values for FPG, fructosamine, hemoglobin A1c, insulin, and CRP levels and HOMA-IR and HOMA-Beta can all be found in Table 3. The mean FPG for all 15 subjects at baseline was 99.9 mg/dL. Four of the eight male subjects had FPG levels higher than 100 mg/dL at baseline, which is clinically recognized by the American Diabetes Association as prediabetes.¹ In addition, four of the seven female subjects also had FPG levels above the recommended 100 mg/dL by the American Diabetes Association.¹

Table 3.

FPG, Fructosamine, Hemoglobin A1c, Insulin, CRP, HOMA-IR, and HOMA-Beta Levels of the Study Subjects

Time period	FPG (mg/dL)	Fructosamine (mmol/L)	Hemoglobin A1c (%)	Insulin (μU/mL)	CRP (mg/dL)	OMA-IR	HOMA-Beta
Baseline	99.86 ± 2.63	3.46 ± 0.14	6.9 ± 0.11	13.52 ± 2.61	0.62 ± 0.16	3.51 ± 0.76	43.26 ± 8.07
Control period	92.13 ± 2.39	2.58 ± 0.16	NA	9.84 ± 1.42	0.64 ± 0.21	2.23 ± 0.29	35.73 ± 6.69
RS period	98.8 ± 3.11	2.48 ± 0.12	NA	10.99 ± 1.43	0.53 ± 0.18	2.69 ± 0.35	36.77 ± 5.70
End of the study	NA	NA	6.79 ± 0.19	NA	NA	NA	NA

Data are mean ± SEM values.

NA, not applicable.

The mean fasting plasma glucose level for male and female participants combined was 98.8 ± 3.11 mg/dL during the 6-week RS bread periods and 92.1 ± 2.39 mg/dL during the 6-week control bread periods. There were some significant differences in FPG levels due to treatment effects. FPG levels of all 15 subjects were significantly lower (P = .018) after consuming the control bread for 6 weeks than at baseline. FPG levels of all subjects were also significantly lower (P < .0001) after the 2-week washout period than at baseline. FPG levels were significantly higher (P < .0001) after consuming the RS bread for 6 weeks than after the washout period. FPG levels of all 15 subjects due to consumption of the RS versus the control bread approached significance (P = .057).

Fructosamine

The mean fructosamine level for all subjects was 3.46 mmol/L at baseline. Normal adult fructosamine levels range from 1.61 to 2.68 mmol/L.³¹ One female subject was the only participant in the study with a normal fructosamine level at baseline. Laboratory guidelines advise that fructosamine values should be compared to previous values in the same person rather than to reference interval values.³²

Mean fructosamine levels dropped to 2.48 mmol/L after the subjects had consumed RS bread for 6 weeks and to 2.58 ± 0.16 mmol/L after the control bread feeding period. Significant differences in fructosamine levels were observed because of treatment effects. Fructosamine levels of all 15 subjects approached significance (P = .0054) after subjects consumed the control bread for 6 weeks compared to baseline values. Fructosamine levels of all subjects were also significantly lower (P < .0012) after the consumption of RS bread for 6 weeks than at baseline. Fructosamine levels were also significantly higher (P < .0001) after the washout period than at baseline. There were no significant differences in fructosamine levels of all 15 subjects due to consumption of the RS versus the control bread (P = .97).

Hemoglobin A1c

The normal reference range for HbA1c levels according to Burtis and Ashwood³¹ is 4.5–5.7 %. Mean baseline HbA1c levels when male and female subjects were combined was 6.9% (n = 15). This value was lowered to 6.79% (n = 14) at the end of the 14-week study. The paired t test for differences in baseline hemoglobin A1c levels compared to those at the end of the study values revealed no significant difference (P = .71).

Insulin

Mean insulin levels for all subjects at baseline was 13.52 μU/mL. Normal insulin levels are expected to range from 2.0 to 25 μU/mL for adults <g60 years of age following a 12-hour fast.³¹ The mean ± SEM insulin levels were 10.99 ± 1.43 μU/mL after the subjects ate RS bread for 6 weeks and 9.84 ± 1.42 μU/mL after eating the control bread. No significant differences were found for treatment, gender, or sequence effects for insulin levels during the 14-week study (P > .05).

HOMA-IR and HOMA-Beta function

Mean HOMA-IR for the subjects at baseline was 3.51 ± 0.76. Values greater than 2.5 for HOMA-IR indicate insulin resistance in adults. Mean HOMA-IR dropped to 2.69 ± 0.35 after the subjects consumed the RS bread for 6 weeks and to 2.23 ± 0.29 after consuming the control bread. No significant differences were found for sequence (P = .67), treatment (P = .14), or gender (P = .78) effects for HOMA-IR during the 14-week study.

Mean HOMA-Beta was 43.26 ± 8.07 at baseline and 36.99 ± 5.70 and 35.73 ± 6.69 after eating the RS and control bread, respectively. There were no significant differences in HOMA-Beta due to sequence (P = .53), treatment (P = .59), or gender (P = .86) effects.

CRP

The mean CRP level at baseline was 0.62 mg/dL. Median normal concentrations of CRP levels are 0.08 mg/dL. Ninety percent of presumable healthy persons have CRP values ≤0.3 mg/dL, and 99% of seemingly healthy persons have CRP values of ≤1.0 mg/dL.³³ CRP levels are elevated in type 2 diabetes and in those at high risk for the disease.^34,35 Two male subjects and two female subjects had CRP levels >1.2 mg/dL at baseline. The mean CRP level of all 15 subjects was 0.53 mg/dL after they ate RS bread for 6 weeks and 0.64 mg/dL after eating the control bread. There were no significant differences in mean CRP levels due to treatment, gender, or sequence effects during the 14-week study (P > .05).

SCFAs

The mean serum acetate level of the subjects was 105.45 μmol/L at baseline, 95.81 μmol/L after consuming the RS bread for 6 weeks, and 102.68 ± 11.13 μmol/L after the control bread feeding period (data not shown). No significant differences were found for treatment or gender effects for serum acetate levels during the 14-week study. Sequence effects were statistically significant (P = .013).

Mean serum propionate level of the subjects was 6.76 μmol/L at baseline and 6.99 μmol/L and 6.17 μmol/L after eating the RS and the control bread, respectively. Mean serum butyrate level was 0.82 μmol/L at baseline and 0.71 and 0.74 μmol/L after 6 weeks of eating the RS and the control bread, respectively. There were no significant differences in serum propionate or butyrate levels due to treatment, gender, or sequence effects.

Male subject 1

Unlike the 14 other African American adults who participated in our study, one of the male study subjects showed definite signs of improvements in glucose homeostasis. Male subject 1 was the only participant who was previously diagnosed as having prediabetes. His FPG was 122 mg/dL at baseline. An FPG of ≥126 mg/dL is presently considered by the American Diabetes Association as the cutoff range for diabetes.¹ Male subject 1's FPG dropped to 105 mg/dL after eating the RS bread and to 98 mg/dL after the washout period and then increased to 106 mg/dL after 6 weeks of the control bread (data not shown). In addition, his fructosamine level went from 3.4 mmol/L at baseline to 2.1 mmol/L at the end of the RS bread period to 3.1 mmol/L after the washout period to 4.0 mmol/L after eating the control bread. His baseline insulin level was 29.5 μU/mL, 6.8 μU/mL after eating the RS bread, 16.1 μU/mL at the end of the washout period, and 9.9 μU/mL after the control bread. Male subject 1's hemoglobin A1c level also improved during the study from 7.4% at baseline to 6.2% at the end of the study.

Discussion

There are several possible reasons why there were no significant decreases in this study in mean FPG and fructosamine levels due to the consumption of the RS bread. FPG is considered to be a short-term indicator of blood glucose control, and fructosamine is an indicator of blood glucose concentrations over the previous 1–3 weeks.³⁶ Twelve grams per day of RS may have been an insufficient amount of RS to positively affect the glucose homeostasis of our study subjects. This may be true in spite of the fact that a recent study reported that the addition of as little as 2.6 g of RS to a meal resulted in significant reductions in postprandial plasma glucose concentrations in both normal weight and overweight women.³⁷ Zhang et al.²⁴ recently conducted a crossover design study similar to ours, where they fed type 2 diabetic subjects a control diet and a diet supplemented with RS for separate 4-week periods. They found that the subjects had significantly lower fasting blood glucose and fructosamine levels after consuming the RS-supplemented diet. Their subjects, however, consumed much higher amounts of RS (30 g/day) than in our study.

The results of this study could also have been affected by our choice of subjects. The fact that male subject 1's FPG and fructosamine fell to near normal levels after eating the RS bread suggests that we might have seen significant improvements in glucose homeostasis in the present study if we had selected only subjects that could be classified by the American Diabetes Association as prediabetic with FPG between 101 and 125 mg/dL. Seven of our 15 subjects had baseline FPGs below 100 mg/dL. All of our subjects did, however, have several risk factors for type 2 diabetes, including being overweight and having a sedentary lifestyle, and some had family members with the disease. In addition, 14 out of the 15 subjects had above normal fructosamine, and all 15 subjects had above normal hemoglobin A1c levels at baseline.

Lack of compliance could also have been a confounding factor in our study. We attempted to ascertain if our subjects had actually eaten the RS bread by analysis of serum SCFAs. Fermentation of RS by colonic bacteria produces SCFAs, some of which are then absorbed into the bloodstream.^7,38 It is very unlikely that all our subjects were noncompliant. Yet, following 6 weeks of consumption of the RS bread, all three SCFAs were elevated by greater than 10% in only one male subject (data not shown). Serum acetate concentrations were significantly elevated in a group of healthy subjects by ingestion of 59 g/day RS, but not by a much lower intake of another RS at 5 g/day.³⁹ It is possible that 12 g/day is too low of an intake of RS to raise serum levels of SCFAs in the majority of adults. Results from one study indicate that it may take longer than 6 weeks for human subjects on high RS diets to begin producing elevated levels of SCFAs.³⁸

The results of this and other research studies are inconclusive with regard to how much RS should be eaten to help prevent chronic disease in humans. More research is needed to determine the minimal daily intake of RS needed to lower the risk of type 2 DM in humans, particularly in those at high risk for the disease.

Footnotes

Acknowledgments

The authors would like to thank Dr. Ruby Cox for helping to fund this research study with her Virginia Tech Aspires Grant. We thank Dr. Anne Birkett and the National Starch and Chemical Company for kindly providing the Hi-maize 260 resistant starch used in this study. We also thank Dr. Thomas M.S. Wolever and Kervan Riveria of the University of Toronto for performing laboratory analysis of short-chain fatty acids reported in this study. Appreciation is also extended to the staff and personnel at the Montgomery Regional Hospital and Our Daily Bread Bakery in Blacksburg, VA. Finally, a warm thank you goes to each of the African American subjects who participated in this study.

Author Disclosure Statement

The authors have no competing financial interests with regard to the execution of this research study, reporting of the data, or writing of the manuscript.

References

American Diabetes Association: Diagnosis and classification of diabetes mellitus. Diabetes Care, 2008; 31,Suppl 1:S55–S60.

Centers for Disease and Control and Prevention: Data & Trends. Crude and Age-Adjusted Prevalence of Diagnosed Diabetes per 100 Population, United States, 1980–2005. www.cdc.gov/diabetes/statistics/prev/national/figage.htm. 2008 July 3.

WebMD: Prevent Type 2 Diabetes by Knowing the Risks Factors. http://diabetes.webmd.com/guide/risk-diabetes. 2008 July 3.

Boyle

, Honeycutt

, Narayan

KMV

et al. Projection of diabetes burden through 2050. Diabetes Care, 2001; 24:1936–1940.

Langkilde

, Ekwall

, Bjorck

, Asp

N-G

, Andersson

. Retrograded high-amylose corn starch reduces cholic acid excretion from the small bowel in ileostomy subjects. Eur J Clin Nutr, 1998; 52:790–795.

Anderson

, Levine

, Levitt

. Incomplete absorption of the carbohydrate in all-purpose flour. N Engl J Med, 181; 304:89–-892.

Topping

, Clifton

. Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides. Physiol Rev, 2001; 26:1031–1064.

Asp

N-G

. Resistart starch. Proceedings from the second plenary meeting of EURESTA: European FLAIR concerted action No. 11 on physiological consumption of resistant starch in man. Eur J Clin Nutr, 1992; Suppl 2:S1.

Toden

, Bird

, Topping

, Conlon

. Resistant starch prevents colonic DNA damage induced by high dietary cooked red meat or casein in rats. Cancer Biol Ther, 2006; 5:267–272.

10.

Fassler

, Gill

CIR

, Arrigoni

, Rowland

, Amado

. Fermentation of resistant starches: influence of in vitro models on colon carcinogenesis. Nutr Cancer, 2007; 58:85–92.

11.

Moreau

, Martin

, Toquet

et al. Restoration of the integrity of rat caeco-colonic mucosa by resistant starch, but not by fructo-oligosaccharides, in dextran sulfate sodium-induced experimental colitis. Br J Nutr, 2003; 90:75–85.

12.

Morita

, Tanabe

, Sugiyama

, Kasaoka

, Kiriyama

. Dietary resistant starch alters the characteristics of colonic mucosa and exerts a protective effect on trinitrobenzene sulfonic acid-induced colitis in rats. Biosci Biotechnol Biochem, 2004; 68:2155–2164.

13.

de Roos

, Heijnen

M-L

, de Graaf

, Woestenenk

, Hobbel

. Resistant starch has little effect on appetite, food intake and insulin secretion of healthy young men. Eur J Clin Nutr, 1995; 49:532–541.

14.

Jenkins

DJA

, Vuskan

, Kendall

CWC

et al. Physiological effects of resistant starches on fecal bulk, short chain fatty acids, blood lipids and glycemic index. J Am Coll Nutr, 1998; 17:609–616.

15.

van Amelsvoort

JMM

, Weststrate

. Amylose-amylopectin ratio in a meal affects postprandial variables in male volunteers. Am J Clin Nutr, 1992; 55:712–718.

16.

Weststrate

, van Amelsvoort

JMM

. Effects of the amylose content of breakfast and lunch on postprandial variables in male volunteers. Am J Clin Nutr, 1993; 58:180–186.

17.

Raben

, Tagliabue

, Christensen

, Madsen

, Holst

, Astrup

. Resistant starch: the effect on postprandial glycemia, hormonal response, and saiety. Am J Clin Nutr, 1994; 60:544–551.

18.

Behall

, Howe

. Effect of long-term consumption of amylose vs amylopectin starch on metabolic variables in human subjects. Am J Clin Nutr, 1995; 61:334–340.

19.

Granfeldt

, Drews

, Bjorck

. Arepas made from high amylose corn flour produce favorable low glucose and insulin responses in healthy humans. J Nutr, 1995; 125:459–465.

20.

Achour

, Flourie

, Briet

et al. Metabolic effects of digestible and partially indigestible cornstarch: a study in the absorptive and postabsorptive periods in healthy humans. Am J Clin Nutr, 1997; 66:1151–1159.

21.

Behall

, Halfrisch

. Plasma glucose and insulin reduction after consumption of breads varying in amylose content. Eur J Clin Nutr, 2002; 56:913–920.

22.

Yamada

, Hosoya

, Nishimura

et al. Effect of bread containing resistant starch on postprandial blood glucose levels in humans. Biosci Biotechnol Biochem, 2005; 69:559–566.

23.

Robertson

, Bickerton

, Dennis

, Vidal

, Frayn

. Insulin-sensitizing effects of dietary starch and effects on skeletal muscle and adipose tissue metabolism. Am J Clin Nutr, 2005; 82:559–567.

24.

Zhang

, Wang

, Zhang

, Yang

. Effects of resistant starch on insulin resistance of type 2 diabetes mellitus patients [in Chinese] Zhonghua Yu Fang Yi Xue Za Zhi, 2007; 41:101–104.

25.

National Institutes of Health. Clinical Guidelines on the Identification, Evaluation and Treatment of Overweight and Obesity in Adults: The Evidence ReportNIH Publication number 98-4083National Insitutes of Health: Bethesda, MD, 1998.

26.

McCleary

, Monaghan

. Measurement of resistant starch. J AOAC Int, 2002; 85:665–675.

27.

Matthews

, Hosker

, Rudenski

, Naylor

, Treacher

, Turner

. Homeostasis model assessment: insulin resistance and β-cell function from plasma glucose and insulin concentrations in man. Diabetologia, 1985; 28:412–419.

28.

Osei

, Rhinesmith

, Gaillard

, Schuster

. Is glycosylated hemoglobin A1c a surrogate for metabolic syndrome in nondiabetic, first-degree relatives of African-American patients with type 2 diabetes? J Clin Endocrinol Metab, 2003; 88:4596–4601.

29.

Wolever

TMS

, Fernandes

, Rao

. Serum acetate:propionate ratio is related to serum cholesterol in men but not women. J Nutr, 1996; 126:2790–2797.

30.

American Diabetes Association: Follow-up report on the diagnosis of diabetes mellitus: The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care, 2003; 26:3160–3167.

31.

Burtis

, Ashwood

. Tietz Fundamentals of Clinical Chemistry, 4 ^th. W.B. Saunders Co.: Philadelphia, 1996.

32.

Wallach

. Interpretation of Diagnostic Tests, 7 ^th. Lippincott, Williams and Wilkins Publishers: Philadelphia, 2000.

33.

Pepys

. The acute phase response and C-reactive protein. Oxford Textbook of Medicine, 3 ^rd. Wetherall

, Ledingham

JGG

, Warrell

. Oxford University Press: Oxford, United Kingdom, 1996; 1527–1533.

34.

Ford

. Body mass index, diabetes, and C-reactive protein among US adults. Diabetes Care, 1999; 22:1971–1977.

35.

Pradhan

, Manson

, Rifai

, Buring

, Ridker

. C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA, 2001; 286:327–334.

36.

van Eijk

, Peters

MJL

, Nurmohamed

, van Deutekom

, Dijkmans

BAC

, Simsek

. Decrease of fructosamine levels during treatment with adalimumab in patients with both diabetes and rheumatoid arthritis. Eur J Endocrinol, 2007; 156:291–293.

37.

Behall

, Scholfield

, Hallfrisch

, Liljeberg-Elmstahl

HGM

. Consumption of both resistant starch and β-glucan improves postprandial glucose and insulin in women. Diabetes Care, 2006; 29:976–981.

38.

Wolever

TMS

, Schrade

, Vogt

, Tsihlias

, McBurney

. Do colonic short-chain fatty acids contribute to the long-term adaptation of blood lipids in subjects with type 2 diabetes consuming a high-fiber diet? Am J Clin Nutr, 2002; 75:1023–1030.

39.

Muir

, Lu

, Young

, Cameron-Smith

, Collier

, O'Dea

. Resistant starch in the diet increases breath hydrogen and serum acetate in human subjects. Am J Clin Nutr, 1995; 61:792–799.