Professor Tom Sanders

Nutritional aspects of chocolate confectionery fats Professor Tom Sanders Nutritional Sciences Division King s College London

Doctors reject chocolate tax Published: Thursday, 12 March 2009, 10:48AM Doctors have narrowly rejected a call for a tax on chocolate to combat obesity.

Confectionery fats Stearic acid rich fats Cocoabutter Cocoabutter substitutes e.g. shea Medium chain fatty acid rich fats Coconut and palm kernel oil Hydrogenated and interesterifed fats e.g. SALATRIM

Diet and Cardiovascular disease (CVD) Age, gender & genes Lipids Smoking CVD Obesity Physical inactivity Haemostatic function Insulin resistance

Hierarchy in Scientific Evidence Hierarchy in Scientific Evidence High Level of Evidence Low Systematic Reviews (Meta-analysis) Randomized Controlled Trials Other Controlled Trials Prospective Cohort studies Case Control studies Prevalence studies Ecological studies Animal studies

Total Cholesterol Distribution: CHD vs Non-CHD Population Framingham Heart Study 26-Year Follow-up 35% of CHD Occurs in People with TC<5.2 mmol/l No CHD CHD 4.0 5.2 Total Cholesterol (mmol/l) Castelli WP. Atherosclerosis. 1996;124(suppl):S1-S9. 1996 Reprinted with permission from Elsevier Science. 6.5 7.8

Metabolic Syndrome Increases Risk for CHD and Type 2 Diabetes High LDL-C Metabolic Syndrome Type 2 Diabetes Coronary Heart Disease NCEP-3. JAMA 2001;285:2486-2497.

Blood lipid metrics of risk of CHD Prospective Studies Collaboration. Lancet 2007;9602:1829-1839

Predicted changes ( ) in the ratio of serum total to HDL cholesterol and in LDL- and HDL-cholesterol concentrations when carbohydrates constituting 1% of energy are replaced isoenergetically with saturated, cis monounsaturated, cis polyunsaturated, or trans monounsaturated fatty acids Mensink, R. P et al. Am J Clin Nutr 2003;77:1146-1155 Copyright 2003 The American Society for Nutrition

Predicted changes in the ratio of serum total to HDL cholesterol and in LDL- and HDL-cholesterol concentrations when carbohydrates constituting 1% of energy are replaced isoenergetically with lauric acid (12:0), myristic acid (14:0), palmitic acid (16:0), or stearic acid (18:0) Mensink, R. P et al. Am J Clin Nutr 2003;77:1146-1155 Copyright 2003 The American Society for Nutrition

Predicted changes ( ) in the ratio of serum total to HDL cholesterol when mixed fat constituting 10% of energy in the "average" US diet is replaced isoenergetically with a particular fat or with carbohydrates Mensink, R. P et al. Am J Clin Nutr 2003;77:1146-1155 Copyright 2003 The American Society for Nutrition

Effect on risk of CHD events of replacing 5% energy saturated fatty acids from pooled analysis of 11 cohort studies Jakobsen M et al. Am J Clin Nutr 2009;89:1-8. 13

14

15 Meta-analysis of trans fatty acids on risk of CHD events

Replacing 2% trans fatty acids with saturated or unsaturated fatty acids reduces risk of CHD WHO/FAO 2009 16

Effects of high stearic acid fats Total cholesterol/hdl cholesterol neutral Postprandial lipaemia Clotting,fibrinolysis and endothelial function Insulin sensitivity

Thrombosis and fibrinolysis

16 THE NORTHWICK PARK HEART STUDY CHD DEATHS BY THIRDS OF DISTRIBUTION Low Medium High 14 12 10 8 6 4 2 0 Cholesterol Fibrinogen Factor VII Meade et al(1986) Lancet i,1052

Effect of different chain length fatty acids on postprandial lipaema Change in plasm a triglycerides after test m eal mmol/l 3 2 1 18:1 cis 18:1trans 18:0 16:0 MCT Low fat 0 0 1 2 3 4 5 6 7 8 hours Sanders et al Atherosclerosis 2000, 149: 413-420

Postprandial lipaemia Sanders et al. Atherosclerosis. 2000; 149:413-20. 22

Postprandial change in FVIIa following meals enriched in specific fatty acids Change in FVIIa ng/ml 1.4 1.2 1 0.8 0.6 0.4 0.2 % change at 3h % change at 7h 0 Low fat MCT Palmitate Stearate Elaidate Oleate Sanders et al Atherosclerosis 2000, 149: 413-420 23

SALATRIM (Short and long triacylglycerol mixture) 87% 11% 2% Decreased energy value 5 kcal/g, stearic acid plus SCFA 24

mmol/l 0.75 0.50 0.25 SALATRIM STUDY Mean change in plasma TAG from fasting with 95%CI Oleate Salatrim Cocoabutter 0.00 3h 6h Sanders et al. Am.J.Clin.Nutr. 2001, 73, 715-721.

Structured stearic rich TAG (SALATRIM) meal decreased FVIIc and FVIIa change in FVIIc activity % 30 20 10 FVII c Oleate SALATRIM Cocoabutter * * 0 3h 6h Hours following test meal change in FVII a concentration ng/ml 2 1 0 FVIIa * * 3h 6h Hours following test meal Sanders, Oakley & Miller. Am J Clin Nutr. 2001; 73:715-21

G L Y C E R O L Triacylglycerol structure fatty acid fatty acid fatty acid sn-1 sn-2 sn-3 L I P A S E fatty acid fatty acid TAG FFA Conflicting results have been obtained of the effect of stearic rich TAG on postprandial lipaemia, which may be a consequence of different TAG structure Saturated fatty acids are better absorbed as a 2-MAG than as free fatty acids sn-1 sn-3 G L Y C E R O L fatty acid 2-MAG sn-2 27

Stearate in Cocoabutter G L Y C E R O L Stearate Oleate Stearate/Palmitate

Randomized vs Unrandomized cocoabutter Randomized cocoabutter (asymmetrical TAG) versus unrandomized cocoabutter (symmetrical TAG) 2.5 Unrandomized CCB 300 P<0.001 Plasma TAG (mmol/l) 2.0 1.5 1.0 Randomized CCB Diet x time effect P=0.003 200 100 0 Unrandomized CCB Randomized CCB Incremental area under curve 0.5 0 1 2 3 4 5 6 Time (h) Sanders et al. (2003)Am J Clin Nutr 77: 777-782

Postprandial change in FVIIa Change in Factor VII:a from fasting units/l 15 10 5 0-5 -10-15 P<0.05 3h P<0.05 6h Unrandomized CCB Randomized CCB Sanders et al. (2003)Am J Clin Nutr 77: 777-782

Randomized stearic acid rich TAG* (asymmetrical TAG) versus unrandomized cocoabutter (symmetrical TAG) Plasma TAG (mmol/l) 2.5 Unrandomized CCB Randomized HOS 2.0 1.5 1.0 0.5 Diet x time effect P=0.033 0 0 1 2 3 4 5 6 Time (h) 300 200 100 P=0.006 Unrandomized Randomized CCB HOS Incremental area under curve * Same as that used in Sanders et al (2000)

Differential scanning calorimetry Cocoabutter Randomised SART

Berry et al (2007) AJCN Berry et al. Am J Clin Nutr 2007;85:1486 94.

Berry et al (2007) Lack of effect of randomization of a stearic rich TAG on LDL 3 0.4% NS -3.2% NS Low stearate Treatment LDL cholesterol mmol/l 2 1 0 Shea butter Randomized Shea

Berry et al (2007) Lack of effect of randomization on HDL cholesterol HDL cholesterol mmol/l 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Shea butter 5.7% NS Randomized Shea 0% NS Low stearate Treatment

Unrandomized Shea butter vs high oleic sunflower oil Plasma TAG (mmol/l) 2.0 High oleic sunflower 1.5 1.0 Unrandomized shea Diet x time effect 0.5 P<0.001 0 1 2 3 4 5 6 Time (h) 250 200 150 100 50 0 P=0.017 Unrandomized shea High oleic sunflower oil Incremental area under curve

Shea butter does not activate FVII postprandially HOS Change in FVIIa from fasting (%) 50 25 0 Unrandomised shea 3h * 6h -25 Berry et al. Am J Clin Nutr 2007;85:1486 94.

Endothelial Dysfunction in Atherosclerosis 38

Endothelial function measurement Pre-occlusion Post occlusion Post GTN 39

Postprandial lipaemia and vascular function HOS = high oleic sunflower oil Shea = stearic acid rich fat (SOS) Berry et al. J. Nutr. 2008;138: 1910 1914, 40

Sundram et al (2007) comparisons with palm olein 1.75-7.6% P<0.001-9.1% P<0.001 1.50 1.25 1.00 0.75 0.50 0.25 HDL cholesterol mmol/l Palm Olein PHSO Interesteried oil 0.00

Changes in fasting glucose reported by Sundram et al (2007) Sundram et al. (2007)

Sundram et al. 2007

Sundram et al. (2007)

Plasma insulin before and after native and randomized shea butter Plasma insulin concentrations (mmol/l) 60 Randomised pre Randomised post 50 Unrandomised pre Unrandomised post 40 30 20 10 0 0 15 30 60 90 120 Time (min) Berry et al. Am J Clin Nutr 2007;85:1486 94.

Plasma glucose before and after native and randomized shea butter Plasma Glucose in mmol/l 8 7 6 5 4 Randomised pre Randomised post Unrandomised pre Unrandomised post 0 15 30 45 60 75 90 105 120 Time (minutes)

Conclusions 1. Chocolate confectionery fats do not have an adverse effect on the lipid profile 2. Chocolate confectionery fats are potentially less thrombogenic than high monounsaturated oils such as olive oil and high oleic sunflower oil 3. Postprandial lipaemia and consequently changes in haemostatic function may be favourably reduced if the solid fat index at 37C is high 4. There is no consistent effect of stearic acid rich fats on insulin and glucose but more research is needed 5. Chocolate confectionery is high in calories and excessive intakes contribute to obesity

Acknowledgements Professor George Miller Jackie Cooper Dr Sarah Berry Dr Fiona Lewis Robbie Gray Dr Tamara de Grassi Dr Francesca Oakley Prof Phil Chowienczyk