1270258546633453 1.6

Heppenstall et al. Nutrition Journal 2012, 11:50http://www.nutritionj.com/content/11/1/50 Relationships between glucose, energy intake anddietary composition in obese adults with type 2diabetes receiving the cannabinoid 1 (CB1)receptor antagonist, rimonabant Charlotte Heppenstall, Susan Bunce and Jamie C Smith* Background: Weight loss is often difficult to achieve in individuals with type 2 diabetes and anti-obesity drugs areoften advocated to support dietary intervention. Despite the extensive use of centrally acting anti-obesity drugs,there is little evidence of how they affect dietary composition. We investigated changes in energy intake anddietary composition of macro- and micronutrients following therapy with the endocannabinoid receptor blocker,rimonabant.
Methods: 20 obese patients with type 2 diabetes were studied before and after 6 months dietary intervention withrimonabant. Dietary intervention was supervised by a diabetes dietician. Five-day food diaries were completed atbaseline and at 6 months and dietary analysis was performed using computer software (Dietplan 6).
Results: After 6 months, (compared with baseline) there were reductions in weight (107 ± 21Kg versus 112 ± 21,p < 0.001, 4% body weight reduction), and improvements in HbA1c (7.4 ± 1.7 versus 8.0 ± 1.6%, p < 0.05) and HDLcholesterol. Intake of energy (1589 ± 384 versus 2225 ± 1109 kcal, p < 0.01), carbohydrate (199 ± 74 versus273 ± 194 g, p < 0.05), protein (78 ± 23 versus 98 ± 36 g, p < 0.05), fats (55 ± 18 versus 84 ± 39 g, p < 0.01) andseveral micronutrients were reduced. However, relative macronutrient composition of the diet was unchanged.
Improvement in blood glucose was strongly correlated with a reduction in carbohydrate intake (r = 0.76, p < 0.001).
Conclusions: In obese patients with type 2 diabetes, rimonabant in combination with dietary intervention led toreduced intake of energy and most macronutrients. Despite this, macronutrient composition of the diet wasunaltered. These dietary changes (especially carbohydrate restriction) were associated with weight loss andfavourable metabolic effects.
Keywords: Dietary assessment, Dietary intervention, Drug interventions, Diabetes considered a cornerstone in the management of obese Obesity is extremely common in type 2 diabetes and is a individuals with type 2 diabetes . However, most strat- major contributor to premature morbidity and mortality egies used to combat obesity have not yielded long-term []. Obesity results from an imbalance of energy intake success and so there is increasing interest in the use and and energy expenditure and so any strategy to reduce development of pharmacological agents to tackle obesity body weight must rely on either, a reduction in energy in- []. Centrally-acting anti-obesity drugs such as sibutra- take, an increase in energy expenditure or both. Super- mine or the CB1 receptor antagonist rimonabant are con- vised weight-loss through dietary intervention is therefore sidered to act principally by reducing appetite and/orincreasing satiety, thereby producing reduced energy in-take []. For example, in the RIO-trial programme in- volving obese subjects with type 2 diabetes and other Department of Diabetes & Endocrinology, Torbay Hospital, Lawes Bridge,Torquay, Devon TQ2 7AA, UK cardiovascular risk factors, rimonabant in combination 2012 Heppenstall et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of theCreative Commons Attribution License which permits unrestricted use,distribution, and reproduction in any medium, provided the original work is properly cited.
Heppenstall et al. Nutrition Journal 2012, 11:50 http://www.nutritionj.com/content/11/1/50 with a recommended moderate daily caloric reduction of during the trial with monthly telephone consultations 600 kcal was extensively studied in terms of weight, meta- and with outpatient clinic reviews at 3 and 6 months.
bolic and cardiovascular parameters []. However, despite Patients with contraindications for the use of rimona- these large clinical trials and widespread clinical use there bant (including depressive illness), clinical or echocar- is little evidence of how centrally acting anti-obesity drugs diographic evidence of left ventricular impairment, specifically affect dietary composition in humans.
peripheral vascular disease or significant renal impair- In the case of rimonabant, animal data suggest that ment (estimated GFR < 30 ml/min) were excluded. All weight loss occurs not only because of reduced energy subjects had been receiving a stable regime of blood intake, but also due to increased energy expenditure glucose-lowering treatment for at least 3 months prior through increased fat oxidation in adipose tissue In to study entry. All subjects were studied at baseline and addition, animal data also suggest that CB1 receptor an- then at 6 months following the weight loss intervention.
tagonism in the rat hypothalamus leads to a preferential Study measurements and blood samples were under- reduction in the intake of palatable fatty and sugary food taken in the morning after a 12 hour fast and medica- To date, these findings have not been demon- tions were omitted on the morning of study. Dietary strated in human studies. Because of concerns relating follow up consisted of monthly telephone calls and re- to depression and suicidal risk, rimonabant’s license was view and analysis of the food diary at 6 months. The withdrawn by the European regulatory authorities in telephone calls involved informal discussion with the in- 2008 but there is still interest in this therapeutic class dividual patient about general well-being, tolerability of the study medication, appetite and food intake. Patients In recent times there has been substantial interest in were given encouragement to adhere to their individua- the way macronutrient intake affects both weight and lised diet plan. The 5 day food diaries at baseline and at glycemia in type 2 diabetes [especially the effects of 6 months were analysed using the dietary analysis soft- restricting carbohydrate intake Several studies ware package, Diet Plan 6 (Forestfield Software Ltd, have reported benefits in terms of improved glucose UK). The 5 day food diaries were discussed with the in- control when a low carbohydrate diet is compared with dividual patients during the baseline and 6 monthly vis- conventional intake of carbohydrate although its. A food portion Atlas [] was used with the patient evidence is conflicting with not all studies demonstrating so that the patient could identify their portion size. This was coded by the dietitian and the amounts entered into The aim of this study was as follows:- [1] to assess weight and metabolic changes following rimonabant Body composition analysis was performed using the therapy in obese patients with type 2 diabetes; [2] to in- method of bioelectrical impedance (Tanita).
vestigate in detail, changes in energy, macro- and micro- Fasting blood samples (10 ml venous blood for each nutrient intake following rimonabant therapy; and [3] to subject) were drawn on the morning of study. The study investigate how changes in macronutrient intake (eg.
had approval from the local research ethics committee, carbohydrate, fat) might influence changes in weight and together with clinical trial authorisation. All subjects gave informed consent to participate in the study.
All statistical analyses were performed using SPSS (ver- Twenty subjects (age range 30-70 yrs) (11 male, 9 sion 14) for Windows. Data are expressed as mean females) with type 2 diabetes (11 insulin-treated) were values ± SD. Paired t-tests were used to evaluate differ- recruited from the multidisciplinary diabetes clinic at ences between group means in the same subjects over Torbay Hospital, a UK district general hospital in 2008.
time for normally distributed data. Correlation between All subjects were obese with a body mass index of variables was evaluated using Spearman’s and Pearson’s greater than 30 Kg/m2 and had expressed the desire to correlation coefficients. A p value of less than 0.05 was lose weight. In an open design, all subjects were studied before, during and after 6 months dietary and lifestyleintervention and rimonabant therapy, 20 milligrams once daily (the standard licensed dose).
The study group consisted of 11 males and 9 females.
At baseline, all subjects received dietary and lifestyle The mean age of participants was 58 ± 11 years and the advice from a specialist diabetes dietician, having com- mean duration of diabetes was 7 ± 7 years. Eleven out of pleted a 5 day food diary, and were prescribed an indivi- 20 were receiving insulin therapy. Mean Body mass dualised 600-800 kcal deficit diet based on healthy index (Kg/m2) was 38 ± 5 Kg/m2. There was 1 active eating and portion control. Patients were followed-up Heppenstall et al. Nutrition Journal 2012, 11:50 http://www.nutritionj.com/content/11/1/50 The effects of rimonabant on physical and biochemical Table 2 Effects of Rimonabant with dietary intervention measurements are shown in Table . Weight was reduced on daily energy and macronutrient intake before and from 112 ± 21 Kg at baseline to 107 ± 21 Kg after 6 months (p < 0.001) equating to a mean 4% weight reduction and this was accompanied by a reduction in waist circumfer- ence from 124 ± 13 cm to 121 ± 13 cm (p < 0.05). Nine out of 20 patients lost more than 5% of body weight over the 6 month treatment period and 2 out of 20 lost more than 10% of body weight. Weight changes were associatedwith a reduction in HbA1c from 8.0 ± 1.6% at baseline to 7.4 ± 1.7% at 6 months (p < 0.05). For the 11 subjects re- ceiving insulin therapy, mean insulin dose fell from 116 ± 59 units/day at baseline to 102 ± 71 units/day after 6 months (p < 0.05). There was a rise in HDL cholesterol from 1.2 ± 0.2 mmol/L at baseline to 1.3 ± 0.2 mmol/L at *p < 0.05 in comparison to baseline.
**p < 0.01 in comparison to baseline.
Macronutrient intake and dietary compositionCompleted and detailed dietary data using Dietplan was reductions in saturated fatty acids (30 ± 16 versus available for 18 out of 20 subjects. Food diaries were 20 ± 6 g, p < 0.01) (33% reduction), monounsaturated deemed incomplete and not suitable for analysis in 2 fatty acids (29 ± 12 versus 19 ± 7 g, p < 0.01) (34% reduc- subjects. The changes in daily dietary intake of energy tion) and polyunsaturated fatty acids (16 ± 8 versus and macronutrients during the study are shown in 10 ± 4 g, p < 0.01) (38% reduction).
Table Six months following rimonabant treatment The dietary composition of macronutrients for the daily energy intake was reduced from 2225 ± 1109 to group, expressed as a % of total energy intake at baseline 1589 ± 384 kcal (p < 0.01). Total carbohydrate intake was was carbohydrate 46%, fat 34% and protein 17%. After reduced from 273 ± 194 to 199 ± 74 g (p < 0.05) with sta- 6 months this composition had not changed significantly tistically significant reductions in starch (148 ± 48 versus with carbohydrate 47%, fat 31% and protein 18%.
114 ± 36 g, p < 0.05) and non-starch polysaccharide(18 ± 5 versus 14 ± 3 g, p < 0.01) but a non-significant re- duction in intake of sugar (121 ± 171 versus 74 ± 64 g Details of daily micronutrient intake before and after (p = 0.10). Protein intake was reduced from 98 ± 36 to rimonabant treatment are shown in Table With re- 78 ± 23 g (p < 0.05). Total fat intake was reduced from spect to micronutrient intake, after 6 months (in com- 84 ± 39 to 55 ± 18 g (p < 0.01) (35% reduction) with parison with baseline), levels of intake of sodium(2544 ± 866 versus 3793 ± 1487 mg, p < 0.01), potassium Table 1 Effects of Rimonabant with dietary interventionon physical and biochemical parameters over the study Table 3 Effects of Rimonabant with dietary intervention on daily micronutrient intake before and after *p < 0.05 in comparison to baseline.
**p < 0.01 in comparison to baseline.
*p < 0.05 in comparison to baseline.
***p < 0.001 in comparison to baseline.
**p < 0.01 in comparison to baseline.
Heppenstall et al. Nutrition Journal 2012, 11:50 http://www.nutritionj.com/content/11/1/50 (2874 ± 944 versus 4307 ± 3299 mg, p < 0.01), vitamin E investigated the effects of a dietary intervention involving (5.9 ± 2.6 versus 8.8 ± 5.1 mg, p < 0.05) and folate the use of the anti-obesity agent, rimonabant on blood (292 ± 103 versus 431 ± 324 mg, p < 0.05) were reduced.
glucose, energy intake and dietary composition in obese There was no evidence of a statistically significant increased intake of any micronutrient studied following The main findings of the present study were that an the intervention. In terms of Reference Nutrient Intake individualised diet based on healthy eating and portion (RNI) for micronutrients, 9 of the 20 subjects did not control, in combination with rimonabant therapy led to meet the RNI for potassium at baseline and this significant weight loss over a 6 month period with a ma- increased to 15 at 6 months. No subjects over the age of jority of patients losing in excess of 5% of their body 65 met the RNI for Vitamin D. With regards to calcium weight. The reduction in body weight was accompanied intake, 7 of the 20 subjects did not meet the RNI at by a reduction in weight circumference and favourable baseline and 9 did not meet the RNI at 6 months. With changes to the lipid profile and blood glucose control.
regards to iron intake 3 subjects did not meet the RNI at These beneficial effects on body weight and metabolic baseline and this increased to 5 at 6 months. For vitamin parameters occurred in association with a significant re- A, 4 subjects did not meet the RNI at baseline and this duction in energy intake, together with reductions in the intake of the principal macronutrients, carbohydrate, fatand protein. The levels of these macronutrients were Relationship between glycaemia and dietary composition reduced to a similar degree such that the overall dietary A strong correlation was observed between the reduc- tion in carbohydrate intake and improvement in gly- As well as conventional lifestyle intervention, pharmaco- caemic control (as indicated by change in HbA1c logical therapies are increasingly used to combat obesity.
concentration) over the study period (r = 0.76, p Rimonabant is a cannabinoid-1 receptor blocker that < 0.0001). This correlation persisted after controlling for induces weight loss and improves the cardiovascular risk weight loss and energy intake. Multiple stepwise regres- profile, glycemia and insulin sensitivity in diabetic subjects sion analysis revealed carbohydrate intake to be an inde- []. The endocannabinoid system, consisting of cannabin- pendent predictor of change in HbA1c (Table It was oid type 1 (CB1) receptors and endogenous ligands is noted that 1 individual recorded an excessively large expressed widely, not only in the central nervous system carbohydrate intake at baseline, which reduced dramat- but also in peripheral organs including visceral adipose ically following the intervention. This led to a very high tissue []. The effects of CB1 receptor blockers on change in carbohydrate intake for this individual of weight, energy expenditure and calorific intake have been around 550 g. On further enquiry, the carbohydrate in- studied in detail in rodent models. In these studies CB1- take was verified. Results were re-analysed excluding this receptor antagonism at the hypothalamic level resulted individual but the statistical relationship persisted not only in reduced food intake but also a change in the (r = 0.54, p < 0.05) (Figure Reduction in fat intake composition of the diet [. In particular, Mathes correlated with degree of weight loss (change in weight et al. using a novel dessert protocol in female rats demon- from baseline to end of study) (r = 0.48, p < 0.05). In strated a lowered calorific consumption with reduced in- contrast, reduction in carbohydrate intake did not cor- take of palatable food (sugar fat whip) following relate with the degree of weight loss.
rimonabant treatment []. In addition, rimonabant hasbeen shown in a rodent model to enhance lipolysis in adi- pose tissue leading to increased energy expenditure Dietary intervention is fundamental to the control of glu- through oxidation of fatty acids and this effect was consid- cose and weight in obese individuals with type 2 diabetes.
ered an important determinant of weight loss independent Understanding how specific dietary interventions and anti-obesity agents affect an individual’s dietary compos- Inferences from these animal studies include the con- ition and energy intake should enable us to target these tention that the endocannabinoid system is an important interventions more effectively. In the present study we modulator of the rewarding properties of foods by actingthrough specific mesolimbic areas in the brain and that Table 4 Stepwise multiple regression analysis with CB1 receptor antagonists may have the potential in ΔHbA1c as dependent variable (R2 = 0.58) humans to reduce the intake of hedonistic type foods in favour of less energy-dense and healthier alternatives]. To date these observations remain unproven in human studies and indeed, results from our present study are at variance with these animal studies in that our patients’ dietary composition following rimonabant Heppenstall et al. Nutrition Journal 2012, 11:50 http://www.nutritionj.com/content/11/1/50 Figure 1 Relationship between reduction in carbohydrate intake and improvement in glycaemic control (as indicated by change inHbA1c concentration) in rimonabant-treated subjects (n = 17) (r = 0.54, p < 0.05).
therapy was unchanged, with no evidence of a shift to- following this type of dietary intervention and is an area wards a healthier, less palatable diet. Instead, individuals appeared to simply reduce calorie intake by globally re-ducing intake of most macronutrients and several micro- nutrients in addition. ‘Surrogate markers’ of a healthier This study was intended to mimic that of a clinical situ- diet such as an increased intake of potassium or dietary ation as much as possible to give an insight into how fibre were not observed and in fact levels of both these CB1 receptor antagonists impact patient’s diets and to nutrients fell following rimonabant therapy.
ascertain whether further research on CB1 receptor In the present study we observed a strong relationship antagonists and indeed other anti-obesity medication on between carbohydrate intake and blood glucose levels, dietary composition is worthwhile. There are certainly with reduction in carbohydrate intake during the study limitations to this study, namely the relatively small sam- being closely correlated with reduction in glycated ple size, open design and lack of a control group. Ideally haemoglobin independent of body weight change or a randomised, placebo-controlled trial would have been total energy intake. In contrast, changes in body weight the preferred study design but this was not feasible appeared to be more closely related to fat intake rather within the resources available to us. Another alternative than carbohydrate intake. These data are in keeping with would have been to study a parallel group of similar the growing body of evidence in favour of carbohydrate patients who lost weight through dietary intervention restriction as a key intervention to optimise glucose con- alone, without anti-obesity medication. This latter type trol in individuals with type 2 diabetes of study design would have significantly enhanced the Micronutrient intake was also assessed in the present quality of this study and indeed was originally intended.
study. Of note, levels of several micronutrients fell fol- Unfortunately, it was not possible to achieve meaningful lowing dietary and rimonabant therapy, paralleling the weight loss through our dietary intervention alone in changes observed in macronutrients. Indeed, there our intended control group of type 2 diabetic patients.
appeared to be a trend for an increasing number of sub- Since no valid comparisons in terms of dietary changes jects failing to meet the Reference Nutrient Intake (RNI) between rimonabant-treated patients and controls are for several important micronutrients following the inter- possible without comparable weight loss achieved in vention. This observation appears to suggest that some each group, it was decided that the rimonabant-treated patients could be at risk of adopting a poorer quality diet patients alone would be included in the final analysis.
Heppenstall et al. Nutrition Journal 2012, 11:50 http://www.nutritionj.com/content/11/1/50 Detailed dietary analysis can be a difficult method to Buyken AE, Mitchell P, Ceriello A, Brand-Miller J: Optimal dietary employ in clinical research but it is hoped that future re- approaches for prevention of type 2 diabetes: a life-course perspective.
Diabetologia 2010, 53:406–418.
search including larger controlled trials will be possible Pagotto U, Pasquali R: Fighting obesity and associated risk factors by to examine dietary changes induced by both centrally antagonising cannabinoid type 1 receptors. Lancet 2005, 365:1363–1364.
acting anti-obesity drugs and other potential anti-obesity Akbas F, Gasteyger C, Sjödin A, Astrup A, Larsen TM: A critical review of thecannabinoid receptor as a drug target for obesity management. Obes agents such as glucagon-like-peptide-1 (GLP-1) agonists.
This is certainly an area worthy of further study given Scheen AJ, Finer N, Hollander P, Jensen MD, Van Gaal LF: RIO-Diabetes the potential widespread application of such agents and Study Group: Efficacy and tolerability of rimonabant in overweight orobese patients with type 2 diabetes: a randomised controlled study.
the knowledge gleaned from the present study suggest- ing significant change in dietary composition associated Herling AW, Kilp S, Elvert R, Haschke G, Kramer W: Increased energy with CB-1 receptor antagonism. In particular, our obser- expenditure contributes more to the body weight-reducing effect ofrimonabant than reduced food intake in candy-fed wistar rats.
vations that levels of some important micronutrients Endocrinology 2008, 149:2557–2566.
may fall below recommended levels of intake following Simiand J, Keane M, Keane PE, Soubrié P: SR 141716, a CB1 cannabinoid anti-obesity medication is an area of special concern.
receptor antagonist, selectively reduces sweet food intake in marmoset.
Behav Pharmacol 1998, 9:179–181.
We would suggest that detailed dietetic input is war- Gessa GL, Orrù A, Lai P, Maccioni P, Lecca R, Lobina C, Carai MA, Colombo ranted in this situation to ensure that dietary intake of G: Lack of tolerance to the suppressing effect of rimonabant on essential micronutrients is not compromised during this chocolate intake in rats. Psychopharmacology (Berl) 2006, 185:248–254.
Elhayany A, Lustman A, Abel R, Attal-Singer J, Vinker S: A low carbohydrate Mediterranean diet improves cardiovascular risk factors and diabetescontrol among overweight patients with type 2 diabetes mellitus: a 1-year prospective randomized intervention study. Diabetes Obes Metab2010, 12:204–209.
Weight loss achieved following dietary intervention in Acheson KJ: Carbohydrate for weight and metabolic control: where do combination with rimonabant therapy occurred because we stand? Nutrition 2010, 26:141–145.
of reduced energy intake. This was associated with an Sargrad KR, Homko C, Mozzoli M, Boden G: Effect of high protein vs highcarbohydrate intake on insulin sensitivity, body weight, hemoglobin unselective reduction in intake of all the major macro- A1c, and blood pressure in patients with type 2 diabetes mellitus. J Am nutrients without any change to the overall composition of the diet. Thus, although pre-clinical studies support Gannon MC, Nuttall FQ: Effect of a high protein, low carbohydrate dieton blood glucose control in people with type 2 diabetes. Diabetes 2004, the hypothesis that CB1 receptor antagonists modulate feeding behaviour by altering the perception of palatabil- Evangelista LS, Heber D, Li Z, Bowerman S, Hamilton MA, Fonarow GC: ity, our present study does not support this hypothesis Reduced body weight and adiposity with a high-protein diet improvesfunctional status, lipid profiles, glycemic control, and quality of life in in human subjects with type 2 diabetes.
patients with heart failure: a feasibility study. J Cardiovasc Nurs 2009,24:207–215.
Kirk JK, Graves DE, Craven TE, Lipkin EW, Austin M, Margolis KL: Restricted- CH and SB declare that they have no competing interests. JS has received carbohydrate diets in patients with type 2 diabetes: a meta-analysis.
honoraria for lecturing and an unrestricted educational grant from Sanofi Krebs JD, Elley CR, Parry-Strong A, Lunt H, Drury PL, Bell DA, Robinson E,Moyes SA, Mann JI: The diabetes excess weight loss (DEWL) trial: a randomised controlled trial of high-protein versus high-carbohydrate CH provided dietary input to patients in the study, performed the dietary diets over 2 years in type 2 diabetes. Diabetologia 2012, 55:905–914.
analysis and contributed to the writing of the manuscript. SB contributed to Nelson M, Atkinson M, Meyer J: A Photographic Atlas of Food Portion Sizes.
the planning of the study and was responsible for the day-to-day running of the study. JS was principal investigator, responsible for study design and Howlett AC, Breivogel CS, Childers SR, Deadwyler SA, Hampson RE, Porrino wrote the manuscript. All authors read and approved the final manuscript.
LJ: Cannabinoid physiology and pharmacology: 30 years of progress.
Neuropharmacology 2004, 47(Suppl 1):345–358.
Mathes CM, Ferrara M, Rowland NE: Cannabinoid-1 receptor antagonists Design of study (JS,SB,CH), data collection (SB,CH), data analysis (JS,CH), reduce caloric intake by decreasing palatable diet selection in a novel manuscript preparation and review (JS,CH).
dessert protocol in female rats. Am J Physiol Regul Integr Comp Physiol2008, 295:R67–R75.
Kirkham TC: Endocannabinoids in the regulation of appetite and body Unrestricted educational grant from Sanofi Aventis.
weight. Behav Pharmacol 2005, 16:297–313.
Received: 11 November 2011 Accepted: 5 July 2012 Cite this article as: Heppenstall et al.: Relationships between glucose,energy intake and dietary composition in obese adults with type 2 diabetes receiving the cannabinoid 1 (CB1) receptor antagonist, Alexander CM, Landsman PB, Teutsch SM, Haffner SM: NCEP-defined rimonabant. Nutrition Journal 2012 11:50.
metabolic syndrome, diabetes, and prevalence of coronary heart diseaseamong NHANES III participants age 50 years and older. Diabetes 2003,52:1210–1214.
Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA, Brody J,Hayflick L, Butler RN, Allison DB, Ludwig DS: A potential decline in lifeexpectancy in the United States in the 21st century. N Engl J Med 2005,352:1138–1145.

Source: http://www.nutritionj.com/content/pdf/1475-2891-11-50.pdf