Sydney diet heart

The recovery and re-evaluation of the old data on Sydney Diet Heart Study uncovers unexpected outcomes. However, interpret with care because things have changed since 1970s.

1. Sydney Diet Heart Study re-evaluation 2013 Ramsden CE, Zamora D, Leelarthaepin B, Majchrzak-Hong SF, Faurot KR, Suchindran CM, et al. Use of dietary linoleic acid for secondary prevention of coronary heart disease and death : evaluation of recovered data from the Sydney Diet Heart Study and updated meta-analysis . BMJ 2013;346 :e8707 . Page 1

2. Sydney Diet Heart Study ended 1973, the same year when Opera House was opened. Now after 40 years, the actual results of the study are presented. Photo: BigStockPhoto

3. Background Sydney Diet Heart Study was one of the key studies in Diet Heart Hypothesis. It was a randomized controlled study (RCT) with disease and mortality end points. For some mysterious reason, the original investigators never reported the cardiovascular outcomes of the study, but total mortality rates only. This is rather puzzling as the purpose of the study was to examine the effects of saturated fat (SFA) and polyunsaturated fat (PUFA) on heart disease Chris Ramsden with coworkers and one original investigator, Dr. Boonseng Leelarthaepin , recovered the original data and analyzed the heart disease mortality Page 3

4. Methods (schematic) Page 4 Randomization of patients (MI survivors/angina, n = 458) Low in SFA, high in PUFA diet. Safflower oil and safflower oil based margarine* n = 221 C ontrol (as at baseline) (normal Aussie diet during 1960s) n= 237 Median follow up : 39 months y. 1966→1973 *) “Increase PUFA intake to about 15% of food energy and to reduce ntake of SFA and dietary cholesterol to less than 10% of food energy and 300 mg per day, respectively”

5. Methods N= 458, males only. All participants had symptomatic coronary heart disease at the baseline (secondary prevention) ~ 70% were smokers at baseline Total cholesterol was ~ 280 mg/DL, ie . 7,2 mmol /l at baseline 8,1 % in PUFA group and 5,5 % in SFA group had diabetes at baseline Blood pressure was around 137/89 mmHg in both groups BMI was around 25 in both groups SFA intake at baseline in both groups was around 16 % of energy and PUFA around 6 % in both groups Intake of carbs was around 40 % of energy at baseline in both groups Page 5

6. Results , changes in diet Page 6

7. Results, changes in cholesterol Page 7 Δ 7,8% Total cholesterol was 7,8 % lower ( ie . 0,59 mmol /l, or 23 mg/DL) in the PUFA group at 12 months. No data at the end of the study is available

8. Results , coronary heart disease mortality Page 8 74 % increased risk in PUFA group PUFA Safflower oil/margarine Control Diet , high in SFA

9. Results , updated meta-analysis (1/2) Page 9 When all PUFA studies are included, there is no difference in cardiovascular disease mortality between the high SFA and the PUFA groups

10. Results , updated meta-analysis (2/2) Page 10 In the PUFA studies where considerable increase in alfa-linolenic (ALA) was also achieved, PUFA+ALA/EPA/DHA interventions delivered 2 1 % decrease in cardiovascular mortality

11. Why did not the reduction in cholesterol bring about any benefit ? My speculations (1/2) There role of trans fat in the margarine ( ie . intervention group)? I don’t think it was of major importance The control group was allowed to use other commercial margarines available at the time (and they used) Margarines high in trans fat still deliver better Total Cholesterol/HDL –ratio than butter ( Mensink et al. 2003, meta-analysis of 63 feeding trials) Role of inflammation (Ω-6 FAs → increase in arachidonic acid levels → inflammation story) Ramsden dismisses the whole concept of inflammation in the discussion even if he has previously been a keen supporter A recent meta-analysis of RCTs in humans: diet very high in linoleic acid do not cause the increase in inflammation markers in humans (Johnson & Fritsche 2012) If anything, high doses of SFA are pro-inflammatory according to many recent meal studies and prospective cohorts Page 11

12. Why did not the reduction in cholesterol bring about any benefit ? My speculations (2/2) Page 12 There role of oxidatized linoleic acid metabolites in ox-LDL (OXLAMs) The number one explanation offered by the authors This theory is pretty new and non-established and represents a major shift in the thinking of cholesterol sceptics The theory contradicts the fact that nuts, high in linoleic acid, do consistently decrease the risk of cardiovascular disease in prospective cohorts studies and improve lipid profile in humans ( M ozaffarian et al. 2011) Perhaps, just a coincidence? Perhaps, the recovery of the old data was not bullet-proof?

13. Meta-analysis. My conclusions Page 13 Vegetable oils devoid of ALA are not protective against cardiovascular deaths Vegetable oils, like canola oil (rapeseed oil in Northern Europe) and flaxseed oil, are higher in ALA and lower in linoleic acid, and therefore likely to improve cardiovascular prognosis Modern margarine are devoid of trans fat, at least in Nordic Countries, and based on canola oil. They are likely to improve health outcomes Nuts are, in spite of their high linoleic acid content, consistently linked to improved health outcomes and should be used regularly In a pooled analysis of modern prospective cohorts, a replacement of SFA with PUFA reduced coronary deaths risk by 26 % ( Jakobsen et al. 2009) Using the currently available and popular vegetable oils and margarines seem to protect from coronary deaths

14. Thank you Chris Ramsden , Boonseng Leelarthaepin et al ! Page 14

15. Welcome aboard! 16.2.2013 15 http://twitter.com/pronutritionist http://www.facebook.com/pronutritionist http://www.pronutritionist.net (Finnish) http://www.pronutritionistblog.com (English) Reijo Laatikainen, Authorized Nutritionist, MBA