VOLUME 10 NUMBER 3 • SEPTEMBER 2013
89
SA JOURNAL OF DIABETES & VASCULAR DISEASE
REVIEW
diet for blood glucose control and had the added benefit of
reducing total cholesterol with the fat manipulation. By the 1980s,
a high carbohydrate high fibre diet had become the standard
recommendation for T2DM.
8
The comparison HC control diets in
later studies examining the effect of the HP and low carbohydrate
diets have aimed to be high in complex or unrefined carbohydrates,
high fibre, and in some studies, also low glycaemic index.
The western population however continued to gain weight and
the incidence of T2DM continued to rise. Therefore it was assumed
that either a high carbohydrate–low fat approach did not work, or
patients could not comply. Attention was turned to other dietary
approaches that might be more favourable to patients. Studies then
aimed to manipulate dietary fat intake
9,10
or carbohydrate intake
11,12
with varying degrees of success regarding both compliance and
maintenance of weight loss. A more recent approach is to increase
protein intake in both absolute amount and relative to fat and
carbohydrate.
High-protein diets
High protein diets (> 25% total energy) have attained popularity
recently, as a less extreme dietary manipulation to aid weight
loss, while maintaining glycaemic control and optimal cholesterol
profiles, and appealing to patients for better compliance. Most
high protein diets consist of around 40% carbohydrate, 30%
protein and 30% fat. A high protein (HP) diet is hypothesised to be
beneficial in facilitating weight loss due to increased satiety,
13–19
diet
induced thermogenesis, and in maintaining lean body mass relative
to fat mass during weight loss.
20–23
There is also a proposed benefit
of HP diets reducing post-prandial glucose excursions compared
with high carbohydrate diets.
24,25
Weight loss
To date seven studies have considered the effect of a high protein
diet on weight loss in those with T2DM.
26–32
Of these, only one
study demonstrated a significantly greater weight loss for the HP
group.
31
The remaining studies observed no difference between diet
groups, suggesting that total calorie restriction is more important
than macronutrient manipulation. One study did observe a greater
weight loss for a HP diet plus resistance training vs a conventional
low fat diet plus resistance training, but in the non-exercising arms
there was no difference in weight loss.
32
Glycaemic control
A study of subjects with hyperinsulinaemia, but without diabetes,
suggested that protein may blunt the post-prandial glucose rise
which in a person with diabetes may improve glucose handling and
glycaemic control.
33
The specific impact of a HP diet on glycaemic
control in subjectswith T2DMhas generally beendisappointing.
26,27,29–
32,34
Most studies report no difference between HP and control diets
in measures of glycaemia, despite greater weight loss in the HP
group in one of these studies.
26,27,29,31,32
In a very small and short study, Gannon et al used a cross over
design to experiment with a 30% protein, 20% carbohydrate and
50% fat diet (
n
= 8). After five weeks on the HP diet, fasting glucose
and glycosylated haemoglobin were significantly lower than after
a HC diet.
34
However the carbohydrate level was also lower and
fat much higher than other studies making interpretation of which
macronutrient manipulation was most important very difficult. In
contrast, Sargrad
et al.
randomised 12 subjects with diabetes to an
energy restricted HP or HC diet for eight weeks.
30
No differences
were observed in weight loss or energy intake between diets but
significant decreases in HbA
1c
and fasting glucose concentrations
occurred in the HC group only. The recently published DEWL
trial is the largest and longest (
n
= 419 individuals over 2 years)
randomised controlled trial published to date to explore this,
and did not demonstrate any difference between a HP or a HC
diets on either weight or glycaemic control.
26
The strength of this
study was that it was conducted in a “real world” setting with
the large numbers and long duration, using resources that could
be translated to typical clinical practice. However this was also a
limitation, in that participants struggled to achieve protein intake
targets and tended to drift back to habitual diets over time. This
highlights the difficulty with any dietary approach over the long
term in real clinical practice.
CVD risk factors
Despite concerns, none of the HP studies in type 2 diabetes showed an
overall detrimental effect of a high protein diet on lipids. Evangelista
et al favoured the HP diet for better HDL results, dependant on the
type of protein and fat encouraged by the study with a differential
benefit for plant derived protein and fats over animal sources.
31
This study also recorded a greater decrease in LDL on the HP diet.
31
The most consistent results for the HP diet (in populations with
and without diabetes) appear to be in the effect on triglycerides,
with two of the studies in diabetes demonstrating a significantly
greater decrease on the HP diet compared to the HC diet.
31,35
High
carbohydrate diets have been shown to increase plasma triglycerides
however, so this effect is likely due to the reciprocal reduction in
carbohydrate, rather than the increase in protein.
36,37
There is no consistent effect of a HP diet on blood pressure in
those with diabetes. Neither Evangelista, Krebs nor Larsen
et al
.
reported an effect of diet on blood pressure.
26,27,31
A larger decrease
was noted for the HP group by Sargrad
et al.
with systolic BP
decreasing by 10.5 mmHg and diastolic by 18 mmHg.
30
This effect
was independent of weight loss but seems remarkable for the
8-week duration of the study, in contrast to the long-term DEWL
study with similar weight loss where blood pressure change was
minimal and no different from a HC diet.
26
The only data on long-term cardiovascular outcomes are from
observational studies rather than randomised controlled trials, and
in general populations rather than specifically those with type 2
diabetes. In a large Greek cohort study (subgroup of the European
Prospective Investigation into Cancer and nutrition study EPIC)
higher protein intake particularly if combined with low carbohydrate
intakewas associatedwith increasedmortality.
38
A similar association
was observed in the Women’s Lifestyle and Health study, a 12-year
observational follow up in Swedish women.
39
This association is
supported by evidence of worsening of regional blood flow on
myocardial perfusion imaging in individuals after 12 months on a
high protein diet.
40
In contrast, in the Nurses’ Health Study, which controlled for
age, smoking, total energy intake, type of fat and other coronary
risk factors, higher protein intake was associated with lower
relative risk of cardiovascular mortality.
41
This is also supported in
the Iowa Women’s Health Study where higher vegetable protein
consumption was associated with reduced mortality compared with
an energy equivalent amount of carbohydrate.
42
Therefore whilst
some observational studies raise concern about either high protein
or low carbohydrate diets, they do not answer the question whether
