In their “Against the Grain” Lancet letter, Jim Mann
and co. cited a 2012 Swedish study[1] that correlated a rise in
butter and fall in carbohydrate consumption with a later rise in serum
cholesterol levels in Sweden, after decades of cholesterol-lowering advice was
overturned by a LCHF revolution beginning in 2004.
Recently, the Swedish blog Diet Doctor published this graphic showing the continuing decline in heart disease mortality. The LCHF revolution in 2004 hasn’t exactly slowed the decline. In fact, one could say the decline in MI incidents in men had stalled before 2004, and the decline in MI incidents in women didn’t really start till then.[2]
These data sets raise some questions. Cholesterol is supposed to be raised by meals high in saturated fat; in feeding studies this is an immediate effect, and does not involve any more than a few hours’ time-lag, whereas there seems to be a lag of years in the Swedish correlation.
Does this time lag point to the possibility that MI
incidence will rise in future if the rise in cholesterol is maintained?
For the purpose of the question I will ignore for now
some obvious problems; it is not obvious from Johansson et al that the people
actually eating LCHF are experiencing the rise in cholesterol, nor is it
determined that they are not losing weight (the BMI of the Swedish population
overall continues to rise). In fact, the study of an entire population, in
which only a minority, albeit a large one, is eating an LCHF diet, while the
rest are subject to other contemporary trends, can only provide a relatively
crude and inaccurate critique of LCHF.
A further problem is created by the use of serum
cholesterol as the risk marker, rather than a more reliable measurement such as
LDL:HDL or total cholesterol:HDL, or TG:HDL.
Taking the Swedish data as given, and as if it
represented a homogenous group (which it does not), what grounds do we have for
concern?
In a 1999 paper[3] Law and Wald hypothesised that the “French Paradox” can be explained by the existence of a twenty year (or greater) time lag between high consumption of animal fat, increases in serum cholesterol, and the appearance of increased heart disease.
“For decades up to 1970, France had lower animal fat consumption (about 21% of total energy consumption v 31% in Britain) and serum cholesterol (5.7 v 6.3 mmol/l), and only between 1970 and 1980 did French values increase to those in Britain.”
If this hypothesis, which seemed reasonable in 1999, was correct, heart disease mortality in France would have risen since 1992, the year cited by Law and Wald.
Life expectancy at birth for a woman in France is 85, for a man 78.5.
Life expectancy at birth for a woman in New Zealand is 83.1, for a man 79.4
Age adjusted CHD mortality is 29.25 per 100,000 in France, 76.51 in New Zealand. Even allowing for the French vagary in coding coronary deaths, which according to Law and Wald accounted for 20% of the difference between French and British CHD mortality, France continues to occupy a very low place in the league tables of cardiac mortality[4], 20 years later. France is not unique among European countries; higher intakes of saturated fat correlate with lower incidence of CHD mortality across the continent, including in countries with higher life expectancy and lower rates of alcohol-related mortality.![]()
More evidence against the time-lag hypothesis can be found in the historical ecological narrative from New Zealand. The following graphic comes from Blakely and Woodward’s recent book The Healthy Country? A History of Life and Death in New Zealand.![]()
1950 was the year that rationing ended in New Zealand and sales of cigarettes, sugar, red meat and butter returned to normal (in the case of sugar and cigarettes, after 11 years of significant restriction). The 1967 peak of IHD mortality exactly correlates with the peak of butter consumption (the main source of saturated fat in the New Zealand diet, mostly in the form of shortening in sweet biscuits and cakes and as a spread on white bread).
If atherosclerosis is normally a long-drawn out process, notwithstanding exceptions to this assumption, why would there be an immediate rise in mortality when intake of sugar, cigarette smoke, and fat (in the context of high refined-carbohydrate foods) climbs?
This is explicable if atherosclerosis itself is not a particularly lethal process, and if atherosclerotic plaques rapidly become unstable under conditions of elevated blood pressure, oxidation, inflammation, hyperinsulinaemia, glycation, drug or chemical toxicity, and so on, so that instability in a plaque precipitates a heart attack.
The time-lag hypothesis of heart disease represents the last-ditch stand of opposition to LCHF. It is the “long term safety” quibble that by its nature is difficult to answer (but really, if you care, there is no shortage of LCHF Mediterranean diets to choose from – LCHF with olive oil is still LCHF).
It was answered in useful fashion recently by the latest paper from Jeff Volek’s team[5], and by the latest Harvard epidemiology paper on linoleic acid[6].
If you only read the abstract of this meta-analysis, you’ll think this was the finding from the Harvard team, which included Frank Hu and Walter Willet.
In a 1999 paper[3] Law and Wald hypothesised that the “French Paradox” can be explained by the existence of a twenty year (or greater) time lag between high consumption of animal fat, increases in serum cholesterol, and the appearance of increased heart disease.
“For decades up to 1970, France had lower animal fat consumption (about 21% of total energy consumption v 31% in Britain) and serum cholesterol (5.7 v 6.3 mmol/l), and only between 1970 and 1980 did French values increase to those in Britain.”
If this hypothesis, which seemed reasonable in 1999, was correct, heart disease mortality in France would have risen since 1992, the year cited by Law and Wald.
Life expectancy at birth for a woman in France is 85, for a man 78.5.
Life expectancy at birth for a woman in New Zealand is 83.1, for a man 79.4
Age adjusted CHD mortality is 29.25 per 100,000 in France, 76.51 in New Zealand. Even allowing for the French vagary in coding coronary deaths, which according to Law and Wald accounted for 20% of the difference between French and British CHD mortality, France continues to occupy a very low place in the league tables of cardiac mortality[4], 20 years later. France is not unique among European countries; higher intakes of saturated fat correlate with lower incidence of CHD mortality across the continent, including in countries with higher life expectancy and lower rates of alcohol-related mortality.
More evidence against the time-lag hypothesis can be found in the historical ecological narrative from New Zealand. The following graphic comes from Blakely and Woodward’s recent book The Healthy Country? A History of Life and Death in New Zealand.
1950 was the year that rationing ended in New Zealand and sales of cigarettes, sugar, red meat and butter returned to normal (in the case of sugar and cigarettes, after 11 years of significant restriction). The 1967 peak of IHD mortality exactly correlates with the peak of butter consumption (the main source of saturated fat in the New Zealand diet, mostly in the form of shortening in sweet biscuits and cakes and as a spread on white bread).
If atherosclerosis is normally a long-drawn out process, notwithstanding exceptions to this assumption, why would there be an immediate rise in mortality when intake of sugar, cigarette smoke, and fat (in the context of high refined-carbohydrate foods) climbs?
This is explicable if atherosclerosis itself is not a particularly lethal process, and if atherosclerotic plaques rapidly become unstable under conditions of elevated blood pressure, oxidation, inflammation, hyperinsulinaemia, glycation, drug or chemical toxicity, and so on, so that instability in a plaque precipitates a heart attack.
The time-lag hypothesis of heart disease represents the last-ditch stand of opposition to LCHF. It is the “long term safety” quibble that by its nature is difficult to answer (but really, if you care, there is no shortage of LCHF Mediterranean diets to choose from – LCHF with olive oil is still LCHF).
It was answered in useful fashion recently by the latest paper from Jeff Volek’s team[5], and by the latest Harvard epidemiology paper on linoleic acid[6].
If you only read the abstract of this meta-analysis, you’ll think this was the finding from the Harvard team, which included Frank Hu and Walter Willet.
“A 5% of energy increment in LA intake replacing
energy from saturated fat intake was associated with a 9% lower risk of CHD
events (RR, 0.91; 95% CI, 0.86-0.96) and a 13% lower risk of CHD deaths (RR,
0.87; 95% CI, 0.82-0.94). These data provide support for current
recommendations to replace saturated fat with polyunsaturated fat for primary
prevention of CHD.”
Well maybe, but according to the paper itself,
“Substituting 5% energy intake from LA for the same amount of energy from carbohydrates was associated with an 13% lower risk of CHD deaths (RR, 0.87; 95% CI, 0.81-0.94) and an 13% lower risk of CHD deaths when substituting for the same amount of energy from SFAs (RR, 0.87; 95% CI, 0.82-0.94). This systematic review and meta-analysis support a significant inverse association between dietary LA intake, when replacing either carbohydrates or saturated fat, and risk of CHD.”
This is the kind of dishonesty that gives abstracts a bad name, though Farvid did give the carbohydrate connection an airing in her press interviews.
If you eat more fat and less carbohydrate, you’re going to eat more LA (whether this is necessary in the context of a low-carbohydrate diet is another question) as a matter of course. And interestingly, the Swedish increase in fats wasn’t just butter – the sale of oil for cooking has also increased, while margarine for cooking has decreased (seriously, who cooks with margarine? The thought of this makes my toes curl).![]()
This is reminiscent of the Richard Lehman quote “Where people eat more saturated fat, they often eat more unsaturated fat. For all I know this may help to explain why nearly everyone everywhere is enjoying their food more and living longer.”[7]
Edit: I had overlooked the implications of this passage in the Law and Wald paper,
"This slow increase in mortality from ischaemic heart disease after an increase in serum cholesterol concentration contrasts with the much more rapid decrease in mortality from ischaemic heart disease after a reduction in serum cholesterol. The randomised controlled trials of reducing serum cholesterol concentration show that the maximal reduction in mortality from heart disease is largely attained after about two years.67 Slow inception and rapid reversal are not inconsistent, and one should not be used to suggest that the other is incorrect. The relative risk of smoking related diseases also increases slowly after starting smoking but falls soon after stopping smoking"
Their reference 67 states that,
"The randomised Trials, based on 45,000 men and 4000 ischaemic heart disease events show that the full effect of the reduction of risk [lowering cholesterol] is achieved by five years"
If this is true for one risk factor, the relatively unreliable one of total cholesterol*, why would it not be true for other, more reliable risk factors, such as total cholesterol:HDL and inflammatory markers?
Can the NZ mortality data be explained by the hypothesis that rapid reversal of risk, after slow inception, can itself be rapidly reversed? The parallel is with alcoholic liver disease, which takes years to develop, will be reversed relatively quickly if one stops drinking, but progresses more rapidly if one starts drinking again.
Thus, removal of sugar and cigarette smoke (for example) led to a rapid reversal of a slowly acquired risk (assuming mortality rates were growing before rationing), but when these factors became prevalent again, the risk returned quickly.
* "In the seven countries study, at a cholesterol value of 5.2 mmol/l, the CHD mortality rates were five times higher in northern Europe than in Mediterranean southern Europe." [8]
“Substituting 5% energy intake from LA for the same amount of energy from carbohydrates was associated with an 13% lower risk of CHD deaths (RR, 0.87; 95% CI, 0.81-0.94) and an 13% lower risk of CHD deaths when substituting for the same amount of energy from SFAs (RR, 0.87; 95% CI, 0.82-0.94). This systematic review and meta-analysis support a significant inverse association between dietary LA intake, when replacing either carbohydrates or saturated fat, and risk of CHD.”
This is the kind of dishonesty that gives abstracts a bad name, though Farvid did give the carbohydrate connection an airing in her press interviews.
If you eat more fat and less carbohydrate, you’re going to eat more LA (whether this is necessary in the context of a low-carbohydrate diet is another question) as a matter of course. And interestingly, the Swedish increase in fats wasn’t just butter – the sale of oil for cooking has also increased, while margarine for cooking has decreased (seriously, who cooks with margarine? The thought of this makes my toes curl).
This is reminiscent of the Richard Lehman quote “Where people eat more saturated fat, they often eat more unsaturated fat. For all I know this may help to explain why nearly everyone everywhere is enjoying their food more and living longer.”[7]
Edit: I had overlooked the implications of this passage in the Law and Wald paper,
"This slow increase in mortality from ischaemic heart disease after an increase in serum cholesterol concentration contrasts with the much more rapid decrease in mortality from ischaemic heart disease after a reduction in serum cholesterol. The randomised controlled trials of reducing serum cholesterol concentration show that the maximal reduction in mortality from heart disease is largely attained after about two years.67 Slow inception and rapid reversal are not inconsistent, and one should not be used to suggest that the other is incorrect. The relative risk of smoking related diseases also increases slowly after starting smoking but falls soon after stopping smoking"
Their reference 67 states that,
"The randomised Trials, based on 45,000 men and 4000 ischaemic heart disease events show that the full effect of the reduction of risk [lowering cholesterol] is achieved by five years"
If this is true for one risk factor, the relatively unreliable one of total cholesterol*, why would it not be true for other, more reliable risk factors, such as total cholesterol:HDL and inflammatory markers?
Can the NZ mortality data be explained by the hypothesis that rapid reversal of risk, after slow inception, can itself be rapidly reversed? The parallel is with alcoholic liver disease, which takes years to develop, will be reversed relatively quickly if one stops drinking, but progresses more rapidly if one starts drinking again.
Thus, removal of sugar and cigarette smoke (for example) led to a rapid reversal of a slowly acquired risk (assuming mortality rates were growing before rationing), but when these factors became prevalent again, the risk returned quickly.
* "In the seven countries study, at a cholesterol value of 5.2 mmol/l, the CHD mortality rates were five times higher in northern Europe than in Mediterranean southern Europe." [8]
[1] Associations
among 25-year trends in diet, cholesterol and BMI from 140,000 observations in
men and women in Northern Sweden Johansson et al. Nutrition Journal 2012,
11:40 doi:10.1186/1475-2891-11-40
[2] Heart
attacks 1990-2013 - Myocardial infarctions in Sweden 1990-2013 ISBN:
978-91-7555-237-8
[3]
Why heart disease mortality is low in France: the time lag explanation. Law, M.
and Wald, N. BMJ. May 29, 1999; 318(7196): 1471–1480. PMCID: PMC1115846
[4] http://www.worldlifeexpectancy.com/world-health-review/france-vs-new-zealand
[5] Effects
of Step-Wise Increases in Dietary Carbohydrate on Circulating Saturated Fatty
Acids and Palmitoleic Acid in Adults with Metabolic Syndrome, Volk B.M. et al. PLoS
ONE 9(11): e113605. doi:10.1371/journal.pone.0113605
[6] Dietary
Linoleic Acid and Risk of Coronary Heart Disease: A Systematic Review and
Meta-Analysis of Prospective Cohort Studies. Farvid, M.S. et al. doi: 10.1161/CIRCULATIONAHA.114.010236
[7] http://blogs.bmj.com/bmj/2014/04/22/richard-lehmans-journal-review-22-april-2014/
[8]
[8]