Sunday, September 29, 2013

Thoughts on the Paleolithic Diet Trials

Trial 1
A Palaeolithic diet improves glucose tolerance more than a Mediterranean-like diet in individuals with ischaemic heart disease (2007) [1]
A paleolithic diet is more satiating per calorie than a mediterranean-like diet in individuals with ischemic heart disease (2010) [2]

Trial 2
Beneficial effects of a Paleolithic diet on cardiovascular risk factors in type 2 diabetes: a randomized cross-over pilot study (2009) [3]
Subjective satiety and other experiences of a Paleolithic diet compared to a diabetes diet in patients with type 2 diabetes (2013) [4]

Diet Wars

I’m mainly going to concentrate on the two trials I just blogged about (part 1, part 2) because they are the only Paleo diet trials that have a control group (which is important to minimise the placebo effect and effects from other confounders).  In both trials the ‘control group’ was either the Mediterranean diet or a Diabetes diet, both of which represents fairly conventional dietary advice for weight loss and health.  This design allows one to compare the efficacy of the Paleo diet to the other approaches and see which one is more successful.
First up though, it’s not like any of the diets were unsuccessful.  Many people feel like they have to restrict calories and stay hungry in order to lose weight, and that being on a diet is a pretty miserable existence.  Even without deliberate calorie restriction, the Mediterranean diet, the Diabetes diet and the Paleo diet decreased mean BMI and waist circumference and generally improved other health markers.  What’s more is that the participants mostly had good things to say about the Paleo diet and the Diabetes diet.  If anything, this should be empowering for people who want to lose weight.

The Paleolithic diet was more successful, it:

  • Led to more weight loss [1] [3] 
  • Improved glucose tolerance whereas the Mediterranean diet did not [1] 
  • Was more satiating per calorie [2] [4] 
  • Improved CVD risk factors such as HDL-C, triglycerides, diastolic blood pressure and several markers of glucose tolerance whereas the Diabetes diet did not [3]

The key differences between the diets were essentially that the Paleolithic diet was:

  • Higher in fruit, vegetables and meat, but lower in grains and dairy 
  • Higher in protein, but lower in calories, carbohydrate and saturated fat 
  • More nutrient dense per calorie, except for calcium and sodium

With the implication being that at least one of the key differences between the diets explains why the Paleolithic diet was more successful
 
* I suspect someone (in the CICO is everything camp) will point out that ‘the Paleolithic diet caused greater weight loss and improved XYZ because it had fewer calories’.  That’s obvious, particularly for the fewer calories >> weight loss connection.  But why did the participants eat fewer calories on an ad libitum Paleolithic diet than an ad libitum Mediterranean/Diabetes diet?  This will likely be met with ‘the Paleolithic diet was more satiating because it had more fibre, water and protein, but lower energy density’, and yet the researchers didn’t find any correlations between these and satiety per calorie.  Weight loss must obviously be a result of negative energy balance.  Instead of telling people to simply eat less and go a bit hungry, which is generally unsustainable, how about looking for strategies that promote a spontaneous reduction in calories and support the neuronal control of appetite.
 
Fruit

The Paleolithic groups in both trials had a relatively high fruit intake, averaging 513g in the first trial and 451g in the second.  This was almost twice the amount of fruit as the other group and 6-7x higher than the median intake among Swedish men.  This may sound like a lot of fruit, but it’s only equivalent to about 4 bananas, 2.5 oranges or 3 apples (just a piece of fruit at the end of each meal)
 
Many people in the Paleo community and elsewhere avoid fruit because fruit contains fructose and they hold to the belief that fructose is a poison, is fattening and promotes insulin resistance (see here for some counter-arguments).  These two trials provide evidence that a dietary pattern with a ‘high’ intake of fruit can cause spontaneous weight loss and improve CV risk factors like IR, HDL-C, triglycerides, and blood pressure.  The first study also found a high fruit intake “was not associated with change in AUC Glucose0–120 or AUC Insulin0–120 and “was associated with larger waist loss”.  “Thus, our study lends no support to the notion that fruit intake should be restricted in patients with diabetes or glucose intolerance.”
 
What Conclusions can we Draw?
 
Since there are only two trials with a control group, with a total of 42 participants (39 of whom were men) it’s difficult to draw too many conclusions.  Also, since there were several differences between the diets, you can’t simply ignore the differences except one and draw conclusions along the lines of ‘this proves grains are bad’ (this is kind why of I stressed to not make conclusions from the correlations)*.
 
These trials provide some evidence for the Paleolithic diet being more effective than conventional approaches for weight loss/satiety, IR/T2D and to improve CVD risk factors.  I would like to see further studies and am surprised these results haven’t attracted enough interest from the NIH/etc because they are essentially saying ‘the Paleolithic diet is better than the one which is currently recommended’
 
* Sometimes these trials have been misrepresented to promote an agenda.  This [1] study has been described as a ‘very low carb diet’, but the carbs provided about 40% of total calories.  It’s what I would call a ‘lower, but moderate carb diet’.
 
** No doubt someone is out there thinking: ‘without long term epidemiological studies and further RCTs, how will we know whether a diet of meat, fruit and vegetables is safe?’

Monday, September 23, 2013

Paleolithic Diet Trials: Part 2


13 patients (10 males, 3 females) with type 2 diabetes and without insulin treatment. 

This study was a crossover trial.  For this study it means the subjects were randomised to two groups, then one group goes on the Diabetes diet first and the Paleolithic diet second, and the other group goes on the Paleolithic diet first and then the Diabetes diet second. 

Group A
Group B
Month 0-3
Diabetes diet
Paleolithic diet
Month 3-6
Paleolithic diet
Diabetes diet

Diets: 

Paleolithic diet
“The information on the Paleolithic diet stated that it should be based on lean meat, fish, fruit, leafy and cruciferous vegetables, root vegetables, eggs and nuts, while excluding dairy products, cereal grains, beans, refined fats, sugar, candy, soft drinks, beer and extra addition of salt. The following items were recommended in limited amounts for the Paleolithic diet: eggs (≤2 per day), nuts (preferentially walnuts), dried fruit, potatoes (≤1 medium-sized per day), rapeseed or olive oil (≤1 tablespoon per day), wine (≤1 glass per day). The intake of other foods was not restricted and no advice was given with regard to proportions of food categories (e.g. animal versus plant foods). The evolutionary rationale for a Paleolithic diet and potential benefits were explained” 

Diabetes diet
“The information on the Diabetes diet stated that it should aim at evenly distributed meals with increased intake of vegetables, root vegetables, dietary fiber, whole-grain bread and other whole-grain cereal products, fruits and berries, and decreased intake of total fat with more unsaturated fat. The majority of dietary energy should come from carbohydrates from foods naturally rich in carbohydrate and dietary fiber. The concepts of glycemic index and varied meals through meal planning by the Plate Model were explained. Salt intake was recommended to be kept below 6 g per day.” 

Foods higher in Paleolithic diet
Foods higher in Diabetes diet
Fruits
Potatoes
Vegetables
Beans*
Meat
Cereals without rice
Eggs
Milk/milk products
Bakery*
* Significant, yet small differences 
Macronutrients
Paleolithic diet
Diabetes diet
Energy (kcal)*
1581
1878
Protein (%)*
24
20
Fat (%)*
(SFA*/MUFA/PUFA (g))
39
(19 / 30 / 14)
34
(27 / 26 / 12)
Carbohydrate (%)*
32
42
Alcohol (%)
3
1
Fibre (%)
2.5
2.7
* Significant differences 
Micronutrients
Paleolithic diet
Diabetes diet
Vitamin B6 (mg)
3.2
2.4
Vitamin C (mg)
219
119
Niacin (mg)
27
22
Phosphorous (mg)
1233
1437
Potassium (mg)
3669
3181
Calcium (mg)
356
698
Selenium (µg)
81
55
* Only includes significant differences 
See table 5 

The Paleolithic diet provided more nutrition than the Diabetes diet, even though it contained 16% fewer calories. 

Results: 

Cardiovascular Risk Factors 

The following table lists the CV risk factors where the Paleolithic diet produced a significant difference between groups and baseline, where the results are the averages across both groups after 3 months on the Paleolithic diet and the Diabetes diet.  Italics indicate a non-significant difference compared to baseline 

Paleolithic
Diabetes
HbA1c (%)
5.5
5.9
HDL (mmol/l)
1.34
1.26
Triglycerides (mmol/l)
1.0
1.5
DBP (mmHg)
79
83
BMI (kg/m2)
28
29
Waist (cm)
94
98

Compared to baseline, the Paleolithic diet also improved fasting glucose, fasting insulin, AUC glucose, the insulin sensitivity index and HOMA IR, whereas the Diabetes diet did not 

See table 3 

You can see the results from the first table in the graphs below, which came from figure 1.  Closed circles depicts the group starting the Diabetes diet first and open circles depicts group starting the Paleolithic diet first.  I added ‘D’ and ‘P’ to make it easier to read


“In exploratory analyses of primary endpoints, within-subject differences (Δ) in HbA1c (ΔHbA1c) correlated with Δwaist circumference, which correlated with Δweight, which correlated with ΔCRP. Furthermore, ΔHDL correlated with Δcholesterol and ΔDBP with ΔHOMA2 IR. In exploratory analyses of estimated intake of nutrients, ΔHbA1c correlated with Δpotassium, ΔHDL with Δfatty acid C20:5 n-3, ΔTG with Δthiamin, ΔDBP with Δdietary cholesterol, Δweight with Δenergy density per meal, and Δwaist circumference with Δbakery, Δenergy density per meal, Δsauce and Δvitamin E”.  See table 6 

Satiety and Other Experiences 
Like the other trial, there was no significant difference between the groups in the time between meals (~3 hours), the meals per day (~5) and reported satiety before and 30 minutes after the meal.  The difference was in the number of calories it took to reach a given level of satiety. 

Paleolithic
Diabetes
Energy from good and drink per day (MJ)
1581
1878
Satiety quotient for energy per meal (RS/MJ)*
1.8
1.5

* The change in satiety before and after the meal divided by the energy content of the meal 

See table 1 

“In post hoc analysis of within-subject differences (value during Paleolithic diet minus value during diabetes diet), satiety quotients for energy per meal correlated with triglyceride levels and vitamin B6 intake, satiety quotients for energy density per meal correlated with water from food.”  See table 2

The patients were given a survey to comment how they found the diets.  You can find the results of the survey here.  They found mostly positive comments for both diets and that the Paleolithic diet was better for weight loss, but harder to stick to.  See table 3 
The Discussion and Some Thoughts

Cardiovascular Risk Factors

“The lower DBP after the Paleolithic diet compared to the Diabetes diet did not correlate with sodium intake, which did not differ significantly and was rather low in both diets (2.5 g and 3.0 g per day respectively for the Paleolithic and Diabetes diet).”  Instead, reductions in DBP were correlated with improvements in insulin sensitivity (HOMA2 IR), which is exactly what you would expect.  That being said, don’t place too much importance on the associations.  For example, potassium is probably only associated with HbA1c because it was higher in the Paleolithic groups and is a marker of fruit and vegetable intake and compliance 

Satiety and Other Experiences

The survey results are interesting.  What stuck out for me was: 


  • 3 people said their weight or glucose got worse after the diabetes diet
  • One person said they were always hungry on the Paleolithic diet, even though both diets were ad libitum*
  • A few people missed bread and dairy (mostly cheese) on the Paleolithic diet
  • A few people mentioned the lack of salt on the Paleolithic diet, which is surprising because there wasn’t a large difference in sodium intake (2530 vs. 2963 mg) and if you adjusted for calorie intake the numbers are almost the same (I don’t know why, perhaps because of higher potassium?  At least this is an unnecessary restriction in my opinion) 

* I think this is because people overestimate the number of calories of ‘Paleo’ foods, or simply have no idea.  They might put together a piece of lean meat, some non-starchy vegetables and a piece of fruit and call that a meal.  That’s a really nutrient dense meal, but it’s only about 300 calories.  If you’re eating those sorts of meals you’re going to get hungry.  Another example is snacks.  A common snack of a donut/muffin/chocolate/whatever and soft drink/coffee with cream and sugar is easily a few hundred calories, if you go Paleo and your snack is an apple and a glass of water that’s only about 100 calories.  Unless you go down the ‘stuff swimming in butter’, ‘bag of nuts a day’ or ‘Paleo junk food’ path, not eating enough calories is probably a more likely problem than eating too many (ironically).

Sunday, September 15, 2013

Paleolithic Diet Trials: Part 1


29 male patients with ischaemic heart disease and either glucose intolerance or type 2 diabetes and a waist circumference >94cm were randomised to 2 groups (a Paleolithic group and a Mediterranean/Consensus group).  Patient characteristics at baseline were not significantly different, except the Paleolithic group was significantly older 65 vs. 57. 

Diets: 

Paleolithic diet
“They were advised to increase their intake of lean meat, fish, fruit and vegetables and to avoid all kinds of dairy products, cereals (including rice), beans, sugar, bakery products, soft drinks and beer. The following items were accepted in limited amounts for the Paleolithic group: eggs (one or fewer per day), nuts (preferentially walnuts), potatoes (two or fewer medium-sized per day), rapeseed or olive oil (one or fewer tablespoons per day). The intake of other foods was not restricted and no advice was given with regard to proportions of food categories (e.g. animal vs. plant foods).” 

Mediterranean diet
“Based on whole-grain cereals, low-fat dairy products, potatoes, legumes, vegetables, fruits, fatty fish and refined fats rich in monounsaturated fatty acids and alpha-linolenic acid” 

“Advice about regular physical activity was given equally to the two groups. Both groups were advised not to consume more than one glass of wine per day” 

Significant differences in food
Foods higher in the Paleolithic diet
Foods higher in the Mediterranean diet
Fruits
Cereals without rice
Nuts*
Rice*
Meat
Milk/milk products
* Significant, yet small differences 

Macronutrients
Paleolithic diet
Mediterranean diet
Energy (kcal)*
1388
1823
Weight (g)
1493
1649
Energy Density (kcal/g)
1.07
1.29
Protein (%)*
27
20
Fat (%)
(SFA*/MUFA/PUFA (g))
28
(13 / 18 / 10)
28
(19 / 22 / 12)
Carbohydrate (%)*
39
47
Alcohol (%)
3
2
Fibre (%)
3
3
* Significant differences 

You may realise that the average energy density doesn’t equal the average energy/weight.  That’s because they would have calculated each individual’s energy density and averaged that instead 

There were only a few significant differences in micronutrients.  The Mediterranean diet provided significantly more calcium and sodium.  However, the Paleolithic diet contained 24% fewer calories 

See table 2 

Results: 

Glucose Tolerance Related 

The table below has the results where the Paleolithic group had a significant difference within group compared to baseline and a significant difference at 12 weeks between groups and a significant difference in the change 0-12 weeks between groups.  Italics indicate a non-significant difference within group compared to baseline 

Paleolithic
(n=14)
Mediterranean
(n=15)
Waist circumference (cm)
Baseline
105.8
106.6
12 Weeks
100.2
103.6
Change 0–12 weeks
-5.6
-2.9
AUCb Glucose0–120 (mmol/l×min)
Baseline
1104
1145
12 Weeks
807
1065
Change 0–12 weeks
-290
-80
30 min plasma glucose (mmol/l)
Baseline
10.0
10.7
12 Weeks
8.0
10.3
Change 0–12 weeks
-2.0
-0.4
120 min plasma glucose (mmol/l)
Baseline
8.9
8.8
12 Weeks
5.6
7.9
Change 0–12 weeks
-3.3
-0.9
Normal Glucose Levels
Baseline
2
2
6 Weeks
10
10
12 Weeks
14
7
Diabetic Glucose Levels
Baseline
10
9
6 Weeks
1
3
12 Weeks
0
5

Insulin levels didn’t change that much.  Italics indicate a non-significant difference within group compared to baseline 

Paleolithic
(n=14)
Mediterranean
(n=15)
Fasting plasma insulin (pmol/l)
Baseline
102
123
12 Weeks
86
101
Change 0–12 weeks
-16
-22
30 min plasma insulin (pmol/l)
Baseline
575
625
12 Weeks*
453
507
Change 0–12 weeks
-121
-118
120 min plasma insulin (pmol/l)
Baseline
988
674
12 Weeks
615
631
Change 0–12 weeks*
-374
-42
AUCb Insulin0–120 (nmol/l×min)
Baseline
80.5
69.7
12 Weeks
56.1
60.4
Change 0–12 weeks*
-24.3
-9.3
* Significant differences between groups 

Both groups also lost weight, but the differences between groups was not significant 

Paleolithic
(n=14)
Mediterranean
(n=15)
Weight (Kg)
Baseline
91.7
96.1
12 Weeks
86.7
92.2
Change 0–12 weeks
-5.0
-3.8

See table 2, table 3 and table 4 of the first study 

Satiety 

There was no significant difference between the groups in the time between meals (~3.5 hours), the meals per day (~4.5) and reported satiety before and 30 minutes after the meal.  The difference was in the number of calories it took to reach a given level of satiety. 

Paleolithic
(n=13)
Mediterranean
 (n=14)
Energy from good and drink per day (MJ)
5.8
7.6
Quotient of mean change in satiety during meal and mean energy from food and drink per meal (RS/MJ)*
2.5
1.6

* The change in satiety before and after the meal divided by the energy content of the meal 

See table 1 

There weren’t many significant differences related to leptin and the only notable significant difference between groups was the relative change in fasting plasma leptin over 0-12 weeks when an outlier in the Paleolithic was removed (too many cereal grains). 

Paleolithic
(n=14)
Mediterranean
(n=15)
Fasting plasma leptin (ng/ml)
Baseline
10.7
13.5
12 Weeks
7.1
11.0
Change 0–12 weeks
-35%
-18%

See table 3 

“In post hoc analysis, the strongest correlation between relative change in leptin after 12 weeks and dietary variables was with intake of cereals excluding rice”.  See figure 3.  This correlation was dependent on changes with waist circumference. 

“In post hoc analysis, quotients of mean change in satiety during meal and mean consumed energy from food and drink did not correlate with any of the group dietary differences (intake of energy, protein, carbohydrates, GL, saturated fatty acid, fatty acid C14:0, vitamin A, calcium, sodium, fruits, nuts, meat, cereals without rice, rice, milk/milk products) except for fatty acid C4:0-10.0 and fatty acid C12, and also did not correlate with fiber, energy density, water or beverages.” 

Among the group dietary differences there was a correlation between Satiety Quotient for energy and intake of energy, absolute intake of carbohydrates, GL, saturated fatty acids and sodium.” 

The Discussion and Some Thoughts 

Insulin Resistance Related 

“The more pronounced improvement of glucose tolerance in the Palaeolithic group was unrelated to weight loss or decrease in waist circumference. In contrast, the insulin response changed more as a result of change in waist circumference than of dietary assignment or food choice.” 

To make my life easier, and this blog post easier to read, I didn’t include the data for 6 weeks, the change between 0-6 weeks and the change between 6-12 weeks.  In the results section you see that the Mediterranean group was more competitive in weeks 0-6 than in weeks 6-12.  At 6 weeks there was no significant difference between groups except for waist circumference in favour of the Paleolithic group, although a fair bit of this is probably due to the low number of subjects.  In several measurements the Mediterranean group had non-significant increases, which is reflected in the number of ‘diabetic glucose levels’ increasing 

So you would expect there to be more dropouts in the Mediterranean group, but that actually wasn’t the case.  3 people dropped out of the Paleolithic group, whereas none dropped out of the Mediterranean group.  Perhaps a decrease in compliance outside of dropping out was responsible, but if anything I would expect the Paleolithic diet to be harder to stick to (more expensive, more restrictive and more difficult to eat out).  Perhaps the Mediterranean diet had yielded the majority of its benefits by the 6 week stage, I don’t know? 

By the way, this is the study that people are referring to if they say the Paleo diet reversed type 2 diabetes 

Satiety Related 

“Our previously reported marked improvement of glucose tolerance from this study was not correlated to new data reported here on satiety, leptin, leptin receptor or free leptin index.” 

The leptin results are consistent with both groups achieving a significant reduction in weight and the satiety results are consistent with both groups being on an ad libitum diet.  See figure 4 

I wouldn’t put much weight on the correlations they found.  When you compare two fairly different diets there are many quantitative differences.  It’s difficult to control for the confounding variables and use correlations to find the actual cause.  You can see this especially in the two last paragraphs I quoted.  The researchers put forward the following mechanisms for sodium and carbohydrate:  ↓ sodium >> ↓ palatability >> ↑ satiety*; and replacing carbohydrate with protein >> ↑ satiety.  Beside those, why would intake of fatty acids 4:0-10:0 and fatty acid 12:0 (in such small quantities), other SFAs and total energy* consumed be associated with reduced satiety per calorie for any other reason than because those nutrients were significantly lower in the Paleolithic diet and the Paleolithic diet had greater satiety per calorie? 

* ‘↓ Sodium >> ↓ palatability >> ↑ satiety’ may be true at high sodium intakes but I suspect within the context of comparing two low sodium diets this would apply: ‘↓ sodium >> ↑ nutrient hunger >> ↓ satiety’ 

** This correlation is simply a consequence of measuring satiety per calorie on ad libitum diets 

I’ll save any general thoughts about both the Paleo RCTs for a later blog post