Comparison of 5% versus 15% sucrose intakes as part of a eucaloric diet in overweight and obese subjects: Effects on insulin sensitivity, glucose metabolism, vascular compliance, body composition and lipid profile. A Randomised Controlled Trial

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 69 4 – 7 02

Available online at www.sciencedirect.com

Metabolism www.metabolismjournal.com

Comparison of 5% versus 15% sucrose intakes as part of a eucaloric diet in overweight and obese subjects: Effects on insulin sensitivity, glucose metabolism, vascular compliance, body composition and lipid profile. A Randomised Controlled Trial Anthony S. Lewis a,⁎, Hannah J. McCourt b , Cieran N. Ennis c , Patrick M. Bell a , C. Hamish Courtney a , Michelle C. McKinley b , Ian S. Young b , Steven J. Hunter a a b c

Regional Centre for Endocrinology and Diabetes, Royal Victoria Hospital, Grosvenor Road, Belfast, BT12 6BA, Northern Ireland, UK Nutrition and Metabolism Group, Queen's University Belfast, Belfast, UK Regional Endocrine Laboratory, Royal Victoria Hospital, Belfast, UK

A R T I C LE I N FO Article history:

AB S T R A C T Aims. The effect of dietary sucrose on insulin resistance and the pathogenesis of diabetes

Received 16 November 2012

and vascular disease is unclear. We assessed the effect of 5% versus 15% sucrose intakes as

Accepted 23 November 2012

part of a weight maintaining, eucaloric diet in overweight/obese subjects.

Keywords:

median age 46 years, range 37–56 years, BMI 31.7 ± 0.9 kg/m2). Subjects completed two

Insulin resistance

6 week dietary periods separated by 4 week washout. Diets were designed to have identical

Sucrose

macronutrient profile. Insulin action was assessed using a two-step hyperinsulinaemic

Diet

euglycaemic clamp; glucose tolerance, vascular compliance, body composition and lipid

Methods. Thirteen subjects took part in a randomised controlled crossover study (M:F 9:4,

profiles were also assessed. Results. There was no change in weight or body composition between diets. There was no difference in peripheral glucose utilization or suppression of endogenous glucose production. Fasting glucose was significantly lower after the 5% diet. There was no demonstrated effect on lipid profiles, blood pressure or vascular compliance. Conclusion. A low-sucrose diet had no beneficial effect on insulin resistance as measured by the euglycaemic glucose clamp. However, reductions in fasting glucose, one hour insulin and insulin area under the curve with the low sucrose diet on glucose tolerance testing may indicate a beneficial effect and further work is required to determine if this is the case. Clinical Trial Registration number ISRCTN50808730. © 2013 Elsevier Inc. All rights reserved.

Abbreviations: Aix, augmentation index; BMI, body mass index; DBP, diastolic blood pressure; G, glucose; ΔI, insulin increment; OGTT, oral glucose tolerance test; NEFAs, non-esterified fatty acids; Ra, glucose appearance; Rd, glucose disappearance; SBP, systolic blood pressure; SIP, clamp sensitivity index; WISP, weighted intake software programme. ⁎ Corresponding author. Tel.: +44 7793744011; fax: +44 2890310111. E-mail address: [email protected] (A.S. Lewis). 0026-0495/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.metabol.2012.11.008

M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 69 4 – 7 0 2

1.

Introduction

Rising levels of obesity and type 2 diabetes mellitus in westernized countries are widely linked to changing dietary patterns and reduced levels of exercise. In particular increased consumption of refined sugar is often considered to be detrimental [1,2]. A westernized diet is high in refined sugar with increased consumption of fast food, carbonated soda drinks and pre-processed convenience foods. Current dietary guidelines to maintain health and prevent chronic disease recommend restriction of sucrose intake to 10% of total energy requirements [3]. Diets high in sucrose may be potentially diabetogenic through effects on insulin resistance which is a key mechanism in the pathophysiology of diabetes. However, it is not clear whether this may be a direct effect of sucrose per se or alternatively effects related to increased energy consumption and subsequent weight gain or other confounding factor associated with a diet high in sucrose [4]. Recently a large longitudinal study suggested that overall glycaemic load was important in the development of insulin resistance and altered glucose metabolism [5]. It is commonly perceived that diets high in sucrose will also have a high glycaemic index however some starchy foods may have a comparable glycaemic index to high-sucrose foods. Furthermore this was an observational study and conclusions from it must be limited [5]. In a previous study we found no detrimental effect of a high sucrose diet containing 25% of total daily energy from sucrose compared to a 10% sucrose diet on insulin resistance in healthy relatively normal weight subjects [6]. However the effect of sucrose in overweight/obese subjects who may be more metabolically susceptible is unclear. Also the dose response effects of sucrose on insulin resistance have not been characterised and in particular restriction to 5% of total daily energy has not been examined. Despite conflicting evidence there have been calls to restrict dietary sucrose further compared with current recommendations of 10% and actual average intake of 12%–13% in the UK and USA [7,8]. We performed a randomised controlled trial of 5% and 15% sucrose diets (as percent of total energy intake). The diets were isocaloric and weight maintaining with matched macronutrient profiles to enable the assessment of the direct effects of sucrose on insulin resistance in healthy overweight and obese subjects. Comparison of differing sucrose diets is of importance as patients find restriction of dietary sucrose difficult and inconvenient, especially in today's society where fast food and pre-processed food are commonplace. Further investigation of sucrose in the diet is important to differentiate the effect of sucrose itself and the increased total carbohydrate that often coincides.

2.

Methods

Seventeen volunteers were recruited. Subjects were included in the study if they were over 18 years of age with a body mass index between 25 and 35 kg/m2. Significant cardiac, renal or hepatic disease was excluded on the basis of a clinical history, examination and screening blood tests. Subjects were also

695

excluded if a diagnosis of diabetes was made either through fasting blood glucose (>7.0 mmol/L) or 2-h level during the oral glucose tolerance test (> 11.1 mmol/L). Due to the use of radioisotopes as part of the clamp technique pregnant women and breastfeeding mothers were excluded from participation and female subjects were advised to use effective contraceptive measures for the duration of the study. All patients gave written informed consent and the study was approved by the Northern Ireland Regional Ethics Committee, the Research Committee of the Belfast Trust at the Royal Hospitals and the Administration of Radioactive Substances Advisory Committee of the United Kingdom. Subjects were assigned to a randomised crossover study consisting of 6 weeks of either a high or low sucrose dietary period (containing 15% and 5% of total daily energy respectively) followed by a 4-week washout period before a further 6-week period on the alternative diet. Subjects were randomised in blocks of four using a random number generator to ensure equal numbers of subjects received high- and low-sucrose in the first dietary period.

2.1.

Diet planning and administration

A 4 day food diary was completed by all participants at the study outset in order to provide information about their dietary likes and dislikes and their usual eating routine and this was used in the formulation of their individualised dietary plans. Food intake data from the baseline food diary were analysed using the Weighed Intake Software Program (WISP©) version 3.0 (Tinuviel Software, Warrington, UK) and energy and nutrient intakes were calculated thus providing an assessment of baseline dietary intake for each participant. For the intervention periods, menu plans were formulated on the basis of the initial food diary along with information the study dietitian had gained from conversations with participants before commencing the study. The study dietitian devised 7 day cyclic menu plans for each participant according to their randomisation using WISP© and taking into account dietary likes and dislikes and usual eating patterns and portion size consumption. The macronutrient profile of the diets was designed to be identical: 55% carbohydrate, 30%–35% lipid, 10%–15% protein, 18 g fibre and with differences only in their sucrose content with the lowsucrose diet contributing 5% of total daily energy and the highsucrose diet 15% of total daily energy along with reciprocal changes in starch content. The caloric intake for each dietary period was calculated so that subjects would maintain their initial body weight (eucaloric). This was calculated by multiplying estimated basal metabolic rate by an appropriate activity factor [8]. Body weight was monitored at least once a week to ensure it remained stable and energy intake was modified where appropriate to maintain this while preserving the fixed macronutrient profiles. All diets were constructed using ‘everyday foods’. Participants were asked to refrain from alcohol. Menu plans were frequently adjusted throughout the dietary periods in order to cater for individual food preferences whilst still adhering to the dietary profiles required. Volunteers were free-living but all food and drink for the study periods were provided to participants in a pre-weighed form along with verbal and written instructions regarding the preparation and cooking of food to ensure compliance with the diet. Compliance was checked by close monitoring of the volunteers; through

696

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 69 4 – 7 02

discussion of palatability, monitoring of body weight (biweekly) and occasional telephone discussion between visits, any food that was not eaten was recorded and reported to the study dietitian. Dietary intake for each participant during their last week on each diet was calculated using WISP and this was then used to provide summary information regarding the average profile of the 5% and 15% sucrose diets.

2.2.

Study assessments

Assessments performed at baseline and at the end of each dietary period included fasting blood samples, blood pressure, pulse wave analysis (a measure of vascular compliance), body composition (measured by bioelectrical impedance), anthropometric measurements and a 75 g oral glucose tolerance test. Insulin action was assessed at the end of each dietary period using the hyperinsulinaemic euglycaemic glucose clamp. Blood pressure was measured using an oscillometric device (Omron 705 CP, Omron Healthcare UK Ltd, Milton Keynes, UK).

2.3.

oxidase method using a Beckman Glucose Analyzer 2. Serum non-esterified free fatty acids were measured using a commercial kit (Wako Chemicals, Neuss, Germany).

2.5.

The nonsteady state equations of Steele et al. [12], as modified by De Bodo et al. [13], were used to determine the glucose appearance (Ra) and disappearance (Rd), assuming a pool fraction value of 0.65 and an extracellular volume of 190 mL/ kg. This was measured over three 30-min time periods: before insulin infusion (−30 to 0 min), during the final stages of the lowdose insulin infusion (90–120 min), and during the final stages of the high-dose insulin infusion (210–240 min). The [3-3H]glucose infusion rates were calculated as the sum of the tracer infused continuously and the tracer in the labelled exogenous glucose infusion. Rates of endogenous (hepatic) glucose production were then calculated by subtraction of the exogenous glucose infusion rates required to maintain euglycemia from isotopically determined rates of glucose appearance.

Assessment of insulin action 2.6.

At the end of each dietary period insulin sensitivity was assessed by a 2-step hyperinsulinaemic euglycaemic glucose clamp with sequential low and high dose insulin infusion (0.4 mU/kg/min and 2 mU/kg/min respectively) [6,10]. Cannulae were inserted into the left arm for infusions and the right arm for blood sampling. The right hand was placed into a heating box maintained at 55 °C to ensure opening of arteriovenous shunts within the limb. An adjusted primedcontinuous infusion of high performance liquid chromatography purified [3-3H]-glucose was administered during the basal 2-h isotope equilibration period (− 120 min to time zero). The initial primer was adjusted based on the fasting plasma glucose [9,10]. Plasma glucose was measured at 5-min intervals on a bedside analyser using the glucose oxidase method (Beckman 2 Glucose analyser, Beckman Coulter (UK) Limited, High Wycombe, Buckinghamshire, UK). The subject's plasma glucose was kept constant (‘clamped’) during the insulin infusion to the fasting glucose value using an exogenous infusion of [3- 3H]glucose-labelled 20% glucose adjusted to match the predicted basal plasma glucose specific activity. After the prime, the isotope-labelled saline solution was continued at a low-dose which was decreased to 50% at 20 min into the first step and to 25% at 20 min into the second step, thus maintaining glucose specific activity constant.

2.4.

Calculations

Analytical techniques

Arterialised venous blood was used for all analyses in the glucose clamp studies. Samples were spun immediately, separated and stored at −20 °C until analysis. Plasma for measurement of glucose specific activity was deproteinized with barium hydroxide and zinc sulfate by the method of Somogyi [11]. Aliquots of tracer infusate and labelled exogenous glucose infusion were spiked into nonradioactive plasma and processed in parallel to allow calculation of [3-3H]glucose infusion rates. Serum insulin was measured by enzyme-linked immunosorbent assay (Abbot Imx; Abbott Laboratories, Berkshire, U.K.). Glucose was measured using an automated glucose

Glucose tolerance test

A standard 75 g oral glucose tolerance test was performed at the beginning and end of each dietary period. Following the ingestion of reconstituted powdered (75 g Dextrose Anhydrous E.P., Dr Reddys Laboratories (UK) Ltd) glucose blood samples were taken at time 0, 30, 60, 90 and 120 min and venous plasma glucose and serum insulin levels were measured.

2.7.

Vascular compliance

Arterial stiffness was determined using pulse-wave analysis (model SCOR-Px; PWV Medical, Sydney, Australia), as described previously [14,15]. All volunteers rested for 15 min in the supine position, and measurements were taken immediately following determination of brachial artery blood pressure. The right radial artery blood pressure waveform was recorded using a tonometer and calibrated according to the brachial systolic and diastolic pressures. Analysis of the central aortic waveform obtained using the SphygmoCor software identified the outgoing and reflected pressure waves (augmentation), occurring during systole. The augmentation index (expressed as a percentage) was defined as the ratio of augmentation to pulse pressure and was used to estimate overall systemic arterial stiffness [16]. The timing of the reflected waveform (a measure of the transit time between the ascending aorta and the first main reflectance site) was also identified and therefore used to indirectly estimate aortic pulse-wave velocity and, hence, aortic stiffness [17]. Arterial waveforms were also recorded by consecutively applanating the carotid and radial arteries gated to a three-lead electrocardiogram to enable calculation of pulse-wave velocity as previously described [18].

2.8.

Statistical methods

The power calculation for this study, based on previous clamp data [6,10], indicated that 13 subjects were required to give a 90% chance of detecting a 10% difference in insulin action at the 5% level of significance. Results are expressed

697

M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 69 4 – 7 0 2

as mean ± standard error mean for normally distributed variables and median (interquartile range) for non-normally distributed variables. End-of-period results for normally distributed variables were analysed by the method of Hills and Armitage for 2-period crossover studies with testing for interaction [19].

3.

Results

Seventeen subjects were initially enrolled into the study. One subject withdrew due to poor venous access and 3 subjects left citing difficulty with dietary compliance and frequent attendance to the study site as reasons for withdrawal. The clinical characteristics of the thirteen subjects who completed the full protocol are shown in Table 1. Volunteers were on average obese (mean BMI 31.7 kg/m2) but were normotensive with normal fasting lipid profiles and indices of glycaemia. Simple fasting measures of insulin sensitivity (insulin and homeostasis model assessment of insulin resistance) indicated that the subjects were moderately insulin resistant. Mean dietary sucrose intake pre-intervention was 8.8% ± 0.8% of total daily energy intake. The macronutrient content of the subjects' matched intervention diets (matched for total energy, macronutrient content and fibre intake) is displayed in Table 2. Physical activity remained constant throughout the study and although no formal assessment of energy expenditure was used given that weight and energy consumption remained the same it can be assumed that the level of physical activity between the dietary periods was constant.

3.1.

Clamp studies

Mean plasma glucose levels were comparable during the steady state phase of the clamp (low-dose insulin 5.5 ± 0.2 v 5.5 ± 0.2 mmol/L, p = 0.75; high-dose insulin 5.3 ± 0.2 v 5.2 ± 0.2 mmol/L, p = 0.19). Fasting serum insulin levels were not significantly different between diets (8.2 v 9.0 mU/L, p = 0.05). During the steady state phases of the clamp insulin levels were comparable in the first step but slightly lower after the 15% diet in the second step of the clamp (low-dose insulin 36.7 v

Table 1 – Baseline clinical and anthropomorphic characteristics of study subjects. Characteristics

Mean (SEM)

Sex (M/F) Age (y) Weight (kg) Body mass index (kg/m2) Fasting plasma glucose (mmol/L) Fasting Insulin (mU/L) a HOMA-IR a Total cholesterol (mmol/L) Low density lipoprotein cholesterol (mmol/L) High density lipoprotein cholesterol (mmol/L) Triglycerides (mmol/L) a

9/4 46.1 (1.9) 92.0 (2.9) 31.7 (0.9) 5.2 (0.2) 9.6 (8.3, 12.5) 2.0 (1.6, 3.2) 5.0 (0.2) 3.1 (0.2) 1.2 (0.1) 1.0 (0.9, 1.7)

a

Expressed as median (interquartile range).

Table 2 – Prescribed intervention intakes of energy and macronutrients during the 5% and 15% sucrose diets (expressed as % of total daily intake unless otherwise stated). Nutrient

5% sucrose diet

15% sucrose diet

Energy (kcal/day) Carbohydrate Starch Total sugar of which sucrose Protein Fat Saturated fat Monounsaturated fat Polyunsaturated fat Fibre (g)

2739 (118) 54.8 (0.04) 37.6 (0.3) 17.1 (0.3) 5.2 (0.03) 12.3 (0.1) 32.9 (0.1) 13.7 (0.3) 9.9 (0.1) 4.8 (0.1) 18.3 (0.2)

2715 (98) 55.0 (0.03) 24.7 (0.3) 30.2 (0.3) 14.9 (0.1) 12.1 (0.05) 32.8 (0.1) 14.7 (0.2) 10.0 (0.1) 4.6 (0.2) 17.9 (0.1)

p-value 0.69 <0.01 <0.01 <0.01 <0.01 0.01 0.05 <0.01 0.43 0.35 0.08

37.2 mU/L, p = 0.31; high-dose insulin 179.5 v 179.4 mU/L, p < 0.05). There was no significant difference in glucose infusion rates between the diets (low-dose insulin 11.1 v 11.4 μmol/kg/min, p = 0.74; high-dose insulin 37.6 ± 1.7 v 39.9 ± 2.9 μmol/kg/min, p = 0.25). Clamp insulin sensitivity index (SIP), calculated by the following formula: SIP = ΔRd/(G × ΔI), was calculated to adjust for the difference in insulinaemia at steady state and was statistically significantly higher after the 15% sucrose diet (0.03 ± 0.003 v 0.04 ± 0.005, p = 0.04). Basal endogenous glucose production was not significantly different between the two diets (7.6 ± 0.4 v 8.1 ± 0.3 μmol/kg/min, p = 0.72). Basal glucose disappearance was the same between the two diets (8.3 ± 0.4 v 7.9 ± 0.4 μmol/kg/min, p = 0.43). Basal metabolic clearance rate of glucose was also the same between the two diets (1.5 ± 0.1 v 1.4 ± 0.1, p = 0.09). Percentage suppression of endogenous glucose production was not different between the two diets (low-dose insulin 72.8% ± 5.9% v 63.7% ± 8.7%, p = 0.41; high-dose insulin 74.3% ± 8.1% v 79.5% ± 6.6%, p = 0.49). Basal non-esterified fatty acids were significantly higher after the 15% sucrose diet compared to the 5% sucrose diet (0.81 ± 0.05 v 0.96 ± 0.07 mmol/L, p = 0.03) but were suppressed comparably during the low-dose and high-dose insulin infusions (0.55 ± 0.06 v 0.59 ± 0.07 mmol/L, p = 0.25; 0.26 ± 0.04 v 0.29 ± 0.04 mmol/L, p = 0.51). Percentage suppression from basal to both low-dose and high-dose insulin infusions was comparable (Fig. 1).

3.2.

Oral glucose tolerance test

Fasting plasma glucose was significantly higher after the 15% sucrose diet compared with the 5% sucrose diet (5.0 ± 0.2 v 5.4 ± 0.2 mmol/L, p < 0.01). One and 2-h glucose levels followed a similar trend but did not achieve statistical significance. There was no significant difference in the incremental area under the curve for glucose (Fig. 2A). There was no significant difference in fasting and 2-h serum insulin levels. Fasting serum insulin levels appeared to be lower after the 5% sucrose diet (8.8 v 12.9 mU/L) however this did not reach statistical significance (p = 0.05). The 1-h insulin level was greater after the 15% sucrose diet (59.0 v 109.2 mU/L, p < 0.01) (Fig. 2B). Incremental area under the curve for insulin was significantly

698

Plasma Glucose (mmol/l)

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 69 4 – 7 02

7 6 5 4

Serum Insulin (mU/l)

200 150 100 50 0

Glucose Infusion Rate (µmol/kg/min)

50 40 30 20 10 0 -30

0

30

60

90

120

150

180

210

240

Time (min) 15% sucrose diet

5% sucrose diet

Fig. 1 – Graph showing the plasma glucose levels (mmol/L), serum insulin levels (mU/L) and glucose infusion rates (μmol/kg/min) during the euglycaemic hyperinsulinaemic glucose clamp.

higher after the 15% sucrose diet (109.5 ± 19.4 v 165.3 ± 23.7, p < 0.01) (Table 3).

3.3.

Haemodynamic studies

There was no difference in any haemodynamic variables, i.e. systolic blood pressure (SBP), diastolic blood pressure (DBP), central augmentation pressure, augmentation index (AIx), time to reflectance and pulse wave velocity (Table 4).

3.4.

Body composition

There was no significant difference in weight between the 5% and 15% sucrose diets at the start (91.9 ± 2.9 v 91.5 ± 3.0 kg, p = 0.45) or at the end of each dietary period (91.1 ± 2.9 v 91.8 ± 3.0 kg, p = 0.19). There was also no significant difference in total body fat after the dietary periods (27.7 ± 2.1 v 27.6 ± 2.4 kg, p = 0.89). Lean body mass was also comparable at the end of the study periods (63.4 ± 3.2 v 63.3 ± 3.2 kg, p = 0.22) as was total body water (46.9 ± 2.3 v 47.2 ± 2.2 l, p = 0.19).

3.5.

erides (1.5 ± 0.2 v 1.5 ± 0.1 mmol/L, p = 0.92) at the end of the two dietary periods.

Lipid profile

There was no significant difference between the 5% and 15% diets in total cholesterol (4.4 ± 0.2 v 4.6 ± 0.2 mmol/L, p = 0.20), LDL-cholesterol (2.8 ± 0.1 v 2.9 ± 0.2 mmol/L, p = 0.12), HDLcholesterol (1.0 ± 0.1 v 1.1 ± 0.05 mmol/L, p = 0.37) and triglyc-

4.

Discussion

This study demonstrates no beneficial effect on insulin resistance with lower dietary sucrose intakes of 5% total daily energy intake compared to a diet with high sucrose intake of 15% total daily energy in otherwise healthy overweight/obese individuals. This study further adds to the literature on this topic by studying a population who is already moderately insulin resistant and may potentially be metabolically susceptible to effects of high sucrose intake. The elevation of fasting glucose with the higher sucrose diet is a potentially important finding to consider when compiling dietary guidelines for those at risk of developing type 2 diabetes. Previous studies examining the effect of dietary sucrose on insulin sensitivity have provided conflicting findings. Animal studies have, in general, shown a detrimental effect of high sucrose and fructose diets on insulin action but at dietary intakes exceeding 60% of total daily energy [20–25]. However, these levels of dietary sucrose and fructose are not palatable to humans. Human studies provide less convincing evidence with no clear consensus on the effect of high dietary sucrose

699

M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 69 4 – 7 0 2

Plasma glucose (mmol/L)

12

Table 3 – Glycaemic parameters from the 75 g oral glucose tolerance test expressed as mean (SEM).

10

5% sucrose diet

†

8 6 4 2 0 0

30

60

90

120

Time (mins)

A

15% sucrose diet

5% sucrose diet 120

‡

Serum Insulin (mU/L)

100

‡

80 60

Fasting plasma 5.0 (0.2) 5.4 (0.2) glucose (mmol/L) 30 min plasma 8.0 (0.5) 9.1 (0.7) glucose (mmol/L) 60 min plasma 8.2 (0.7) 9.7 (1.0) glucose (mmol/L) 90 min plasma 7.2 (0.6) 8.5 (0.8) glucose (mmol/L) 120 min plasma 6.3 (0.5) 7.4 (0.7) glucose (mmol/L) Fasting serum 8.8 (7.0, 11.8) 12.9 (10.6, 15.0) insulin (mU/L) a 30 min serum 62.9 (44.0, 67.3) 74.8 (62.1, 119.4) insulin (mU/L) a 59.0 (43.8, 100.6) 109.2 (88.5-182.9) 60 min serum insulin (mU/L) a 90 min serum 55.0 (38.6, 113.8) 91.9 (88.3, 134.9) insulin (mU/L) a 50.0 (31.5, 74.4) 64.4 (45.2-87.0) 120 min serum insulin (mU/L) a AUC glucose 14.5 (1.0) 16.9 (1.3) iAUC glucose 4.6 (0.8) 6.1 (1.1) 103.2 (78.3, 211.1) 159.9 (130.0, 252.2) AUC insulin a 85.6 (62.8, 145.3) 144.2 (102.2, 220.8) iAUC insulin a a

40

15% sucrose diet

pvalue <0.01 0.03 0.06 0.07 0.11 0.05 <0.01 <0.01 0.05 0.07 0.03 0.10 <0.01 <0.01

Expressed as median (IQR).

20 0 0

30

60

90

120

Time (mins)

B

5% sucrose diet

15% sucrose diet

Fig. 2 – (A) Graph showing plasma glucose levels during 75 g oral glucose tolerance test. *p < 0.01, † p = 0.03. (B) Figure showing the serum insulin levels during the 75 g oral glucose tolerance test. ‡ p < 0.01.

intake and varying methods of assessing insulin sensitivity employed [4,6,26–31]. A strength of this study was the randomised controlled design and rigorous dietetic supervision. The 6-week dietary periods separated by a 4-week washout phase compares to other human dietary studies in this area although it is not possible to predict the effects of the diets over a more prolonged period. Every possible effort was made to ensure compliance with diet and subjects were closely followed up every 2–3 days throughout the dietary periods by the research dietician. This rigorous dietary supervision along with regular weight measurements ensured as best as possible that compliance with the diet was maintained. Many other dietary studies rely on dietary advice or the addition of a supplement to an ad libitum diet which is unlikely to be as accurate. The primary outcome measure of insulin resistance was measured by the gold standard method of the hyperinsuli-

naemic euglycaemic glucose clamp. In this study we found no significant difference in peripheral glucose utilisation between the high- and low-sucrose diet. Fasting plasma glucose was higher after the 15% sucrose diet compared to the 5% sucrose diet. Fasting plasma glucose is largely determined by hepatic glucose production and this finding of higher fasting glucose would suggest increased hepatic insulin resistance after the higher sucrose diet. There was no significant difference in basal endogenous glucose production or its suppression during the low-dose insulin infusion between the high- and low-sucrose diets. Basal metabolic clearance rate of glucose was also the same between diets. However De Fronzo et al. demonstrated that

Table 4 – Haemodynamic variables at the end of each dietary period expressed as mean (SEM).

Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Augmentation (mm/Hg) Aortic augmentation index (%) Time to wave reflection (ms) Brachial pulse-wave velocity (ms)

5% sucrose diet

15% sucrose diet

123.7 (13.1)

128.0 (13.5)

0.09

78.4 (8.5)

82.5 (11.4)

0.20

7.9 (5.6) 21.8 (10.7)

6.8 (6.8) 19.2 (15.6)

0.47 0.31

143.5 (32.1)

145.2 (10.0)

0.36

8.6 (1.5)

7.9 (2.0)

p-value

0.56

700

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 69 4 – 7 02

in non-diabetic subjects basal endogenous glucose production contributed less to fasting glycaemia than basal peripheral glucose uptake [32]. In this non-diabetic population this higher fasting plasma glucose could be explained by a combination of both subtle alteration of basal endogenous glucose production and reduced basal metabolic clearance rate of glucose. Basal non-esterified fatty acids (NEFAs) were significantly higher after the high sucrose diet (0.96 v 0.81, p = 0.03) and this may also explain the changes in fasting plasma glucose. Elevated NEFAs can result in hyperglycaemia through direct effects on GLUT-4 in peripheral skeletal muscle and through stimulation of gluconeogenesis [20,24,29]. It must also be noted that the although every effort was made to match the intervention diets in their composition there were differences in saturated fat composition with the lower sucrose diet containing 13.7% saturated fat compared to 14.7% in the higher sucrose diet (Table 2). This may also have had a bearing on the results. The oral glucose tolerance test (OGTT) data indicated higher insulin levels after the high sucrose diet at comparable levels of glycaemia suggesting a degree of insulin resistance associated with the high sucrose diet. Fasting serum insulin levels appeared higher with the higher sucrose diet however this did not reach statistical significance. There is the possibility for a type 2 error here due to small sample size as the study was powered to the primary outcome, the euglycaemic glucose clamp. The OGTT however as a formal measurement of insulin sensitivity is fraught with problems, not least poor reproducibility and multiple confounding influences including gastric emptying, absorption and the incretin effect. The OGTT is more an indicator of endogenous insulin response as opposed to a measure of insulin sensitivity and so the gold standard euglcaemic glucose clamp is the more accurate method of assessment of insulin resistance [33– 36]. Nevertheless the increased insulin secretion could perhaps contribute to beta cell failure if maintained in the long term. The finding of higher fasting glucose after the 15% sucrose diet is however of particular interest as it has been demonstrated that higher levels of fasting glycaemia, even within the normoglycaemic range are associated with an increased risk of the development of type 2 diabetes. Tirosh et al. demonstrated a progressively increased risk of the development of type 2 diabetes in a cohort of 13,163 young men whose fasting plasma glucose levels measured 4.8–5.0, 5.1–5.2 and 5.3–5.5 mmol/L. The risk increased further in those who had higher BMI and triglyceride levels. Although this current study has a very different study population it is notable that the higher sucrose diet increased fasting plasma glucose from 5.0 to 5.4 mmol/L which potentially increases future risk of the development of diabetes especially in an overweight/obese population who is already at increased risk [37]. The premise of this study was that a high-sucrose diet may alter the risk of developing type 2 diabetes mellitus and cardiovascular disease compared to a low-sucrose diet. Although the primary outcome of this study was the assessment of insulin sensitivity, a feature of both conditions, it is possible that other markers may link sucrose intake and

risk of cardiovascular disease. The secondary endpoints assessed here included the assessment of arterial stiffness using the Sphygmocor device to assess pulse wave analysis and pulse wave velocity. Pulse wave is a non-invasive method for measuring arterial stiffness and has been shown to be associated with cardiovascular risk [38]. In the present study, no difference in markers of arterial stiffness, such as augmentation pressure/index, time to reflectance or brachial pulse wave velocity, was observed, which is in agreement with previous studies conducted within this centre [10]. We also observed no difference between the diets on lipid profiles or blood pressure. This study has limitations which should be recognised. The subjects were all overweight/obese individuals with mean BMI 31.7 kg/m2. None were diabetic and only one subject had impaired fasting glucose at the beginning of the study. All were white apart from one Asian subject. These factors may influence the response to the diets and our conclusions are restricted to this described group. Our subjects were classified as overweight or obese by measurement of body mass index which ignores fat distribution. An obvious limitation with all dietary studies is the reliance on the subjects to comply with the diet provided. With no measurable biochemical marker of compliance we must rely on subject interviews and markers such as regular weight measurements to ensure compliance. This study also employed a rigid diet that may not be comparable to the ad libitum diet that can occur under normal circumstances. High sucrose consumption may be accompanied by high dietary fat intake, caloric excess and subsequent weight gain which could then have further impact on insulin sensitivity. Another limitation was the small sample size however this was based on previous studies within our unit and has been shown to provide adequate power. Although a reduction in sucrose content in the diet has not demonstrated an effect on the primary outcome of insulin resistance as measured by the euglycaemic clamp, the reductions in fasting glucose and 1 h insulin levels during the oral glucose tolerance test indicate a potential benefit in dietary sucrose restriction in this overweight/obese study population. This is a potentially important finding that may influence dietary recommendations in the overweight /obese population to reduce the risk of or delay the onset of the development of diabetes. In conclusion restriction of sucrose from 15% to 5% of total energy intake as part of a eucaloric, weight maintaining diet has no beneficial effect on insulin resistance, as measured by the euglycaemic hyperinsulinaemic glucose clamp, in overweight/obese non-diabetic subjects. This study does not exclude indirect effects of sucrose related to excess calorie intake and weight gain or the fact that diets high in sucrose may also be high in fat which is recognised to have adverse effects on insulin action. Furthermore the finding of the reduction in fasting glucose with the 5% diet although unexplained and requiring confirmation is potentially important. Although the primary endpoint was negative, measurement of secondary endpoints on glucose tolerance testing indicates a potentially beneficial effect of lowering sucrose further with significant reductions in fasting glucose, one hour insulin

M ET ABOL I SM CL IN I CA L A N D EX PE RI ME N TA L 6 2 ( 2 0 13 ) 69 4 – 7 0 2

levels and insulin area under the curve. Further similar dietary studies would be of interest in a population with impaired glucose tolerance to assess if this tight reduction in dietary sucrose affects the development of overt type 2 diabetes mellitus.

Author contributions AL, HMcC and CE performed the study. AL and HMcC analysed the data. CE performed the laboratory work. AL wrote the paper. SH designed the study. HMcC, PB, HC, MMcK, IY and SH edited the paper and assisted with data analysis and interpretation.

Funding Funded by Sugar Nutrition UK.

Conflict of interest This study was funded by an unrestricted grant from Sugar Nutrition UK. They had no role in study design, data analysis or submission of this paper.

REFERENCES

[1] Hofmann SM, Tschöp MH. Dietary sugars: a fat difference. J Clin Invest 2009;119:1089–92. [2] Arola L, Bonet ML, Delzenne N, et al. Summary and general conclusions/outcomes on the role and fate of sugars in human nutrition and health. Obes Rev 2009;10(Suppl 1):55–8. [3] World Health Organisation. Diet, nutrition and the prevention of chronic diseases: report of a joint WHO/FAO expert consultation. Geneva: World Health Org; 2002 (Tech. Rep. Ser., no. 916). [4] Spence M, McKinley M, Hunter SJ. Session 4: CVD, diabetes and cancer: diet, insulin resistance and diabetes: the right (pro)portions. Pro Nutr Soc 2010;69:61–9. [5] O'Sullivan TA, Bremner AP, O'Neill S, Lyons-Wall P. Glycaemic load is associated with insulin resistance in older Australian women. Eur J Clinl Nutr 2010;64:80–7. [6] Black RNA, Spence M, McMahon R, et al. Effect of eucaloric high- and low-sucrose diets with identical macronutrient profile on insulin resistance and vascular risk. Diabetes 2006;55:3566–72. [7] Frayn FN, Kingman SN. Dietary sugar and lipid metabolism in humans. Am J Clin Nutr 1995;62:250s–63s. [8] Marriott BP, Olsho L, Haddem L, Connor P. Intake of added sugars and selected nutrients in the United States, National Health and Nutrition Examination Survey (NHANES) 2003–2006. Crit Rev Food Sci Nutr 2006;50(3):228–58. [9] Schofield WN. Predicting basal metabolic rate, new standards and review of previous work. Hum Nutr Clin Nutr 1985;39(Suppl):1–96. [10] Hunter SJ, Boyd AC, O'Harte FP, et al. Demonstration of glycated insulin in human diabetic plasma and decreased biological activity assessed by euglycemic-hyperinsulinemic clamp technique in humans. Diabetes 2003;52:492–8.

701

[11] Somogyi M. Determination of blood sugar. J Biol Chem 1945;160:69–73. [12] Steele R, Bishop JS, Dunn A, Atszuler N, Rathegeb I, DeBodo RC. Inhibition by insulin of hepatic glucose production in normal dog. Am J Physiol 1965;208:301–6. [13] De Bodo RC, Steele R, Altszuler N, Dunn A, Bishop JS. On the hormonal regulation of carbohydrate metabolism; studies with C14 glucose. Rec Prog Horm Res 1963;19:445–88. [14] Wilkinson IB, Fuchs SA, Jansen IM, et al. Reproducibility of pulse wave velocity and augmentation index measured by pulse wave analysis. J Hypertens 1998;16:2079–84. [15] Filipovsky J, Svobodova V, Pecen L. Reproducibility of radial pulse wave analysis in healthy subjects. J Hypertens 2000;18: 1033–40. [16] Wilkinson IB, Cockcroft JR, Webb DJ. Pulse wave analysis and arterial stiffness. J Cardiovasc Pharmacol 1998;32(Suppl. 3): S33–7. [17] Marchais SJ, Guerin AP, Pannier BM, Levy BI, Safar ME, London GM. Wave reflections and cardiac hypertrophy in chronic uremia: influence of body size. Hypertension 1993;22:876–83. [18] Mullan BA, Ennis CN, Fee HJ, Young IS, McCance DR. Protective effects of ascorbic acid on arterial hemodynamics during acute hyperglycemia. Am J Physiol Heart Circ Physiol 2004;287:H1262–8. [19] Hills M, Armitage P. The two-period cross-over clinical trial. Br J Clin Pharmacol 1979;8:7–20. [20] Storlien LH, Oakes ND, Pan DA, Kusunoki M, Junkins AB. Syndromes of insulin resistance in the rat. Inducement by diet and amelioration with benfluorex. Diabetes 1993;42: 457–62. [21] Thorburn AW, Storlien LH, Jenkins AB, Khouri S, Kraegen EW. Fructose-induced in vivo insulin resistance and elevated plasma triglyceride levels in rats. Am J Clin Nutr 1989;49: 1155–63. [22] Zavaroni I, Sander S, Scott S, Reaven GM. Effect of fructose feeding on insulin secretion and insulin action in the rat. Metabolism 1980;29:970–3. [23] Pamies-Andreu E, Fiksen-Olsen M, Rizza RA, Romero JC. High-fructose feeding elicits insulin resistance without hypertension in normal mongrel dogs. Am J Hypertens 1995;8:732–8. [24] Pagliassotti MJ, Shahrokhi KA, Moscarello M. Involvement of liver and skeletal muscle in sucrose-induced insulin resistance: dose–response studies. Am J Physiol 1994;266: R1637–44. [25] Martinez F, Rizza RA, Romero RC. High-fructose feeding elicits insulin resistance, hyperinsulinism and hypertension in normal dogs. Hypertension 1994;23:456–63. [26] Kiens B, Richter EA. Types of carbohydrate in an ordinary diet affect insulin action and muscle substrates in humans. Am J Clin Nutr 1996;63:47–53. [27] Lê KA, Ith M, Kreis R, et al. Fructose overconsumption in healthy subjects with and without a family history of type 2 diabetes. Am J Clin Nutr 2009;89:1760–5. [28] Koivisto V, Yki-Jarvinen H. Fructose and insulin sensitivity in patients with type 2 diabetes. J Intern Med 1993;233:145–53. [29] Thorburn AW, Crapo PA, Griver K, Wallace P, Henry RR. Long-term effects of dietary fructose on carbohydrate metabolism in non-insulin dependent diabetes mellitus. Metabolism 1990;39:58–63. [30] Byrnes AE, Frost GS. Increased sucrose intake is not associated with a change in glucose or insulin sensitivity in people with type 2 diabetes. Int J Food Sci Nutr 2007;58(8): 644–51. [31] Daly ME, Townsley M, Vale CP, et al. Dietary carbohydrate source and insulin sensitivity. Diabet Med 1997;14(Suppl.):S25 (abstr). [32] DeFronzo RA, Ferrannini E, Simonson DC. Fasting hyperglycaemia in non-insulin-dependent diabetes

702

M ET ABOL I SM CL IN I CA L A N D E XP E RI ME N TAL 6 2 ( 2 0 13 ) 69 4 – 7 02

mellitus: contributions of excessive hepatic glucose production and impaired tissue glucose uptake. Metabolism 1989;34: 387–95. [33] Albareda M, Rodrίguez-Espinosa J, Murugo M, de Leiva A, Corcoy R. Assessment of insulin sensitivity and beta-cell function from measurements in the fasting state and during an oral glucose tolerance test. Diabetologia 2000;43: 1507–11. [34] Horowitz M, Edelbroek MA, Wishart JM, Straathof JW. Relationship between oral glucose tolerance and gastric emptying in normal healthy subjects. Diabetologia 1993;36:857–62. [35] Ko GT, Chan JC, Woo J, et al. The reproducibility and usefulness of the oral glucose tolerance test in screening for

diabetes and other cardiovascular risk factors. Ann Clin Biochem 1998;35(pt1):62–7. [36] Bergman RN, Hϋckling K, Watanabe M. Chapter 16 measuring insulin action in vivo. In: DeFronzo RA, Ferrannini E, Keen H, Zimmet P, editors. International textbook of diabetes mellitus. 3rd ed. Wiley; 2004. [37] Tirosh A, Shai I, Tekes-Manova D, et al. Normal fasting plasma glucose levels and type 2 diabetes in young men. N Engl J Med 2005;353:1454–62. [38] Graham MR, Evans P, Davies B, Baker JS. Arterial pulse wave velocity, inflammatory markers, pathological GH and IGF states, cardiovascular and cerebrovascular disease. Vasc Health Risk Manag 2008;4:1361–71.