20
VOLUME 11 NUMBER 1 • MARCH 2014
REVIEW
SA JOURNAL OF DIABETES & VASCULAR DISEASE
measures of nephropathy, small sample sizes, and varying protein
intakes in the low-protein diets used.
Microalbuminuria
Pijls
et al
. randomised 121 people with type 2 diabetes and
microalbuminuria or high normal albuminuria to either a low-
protein diet (protein restriction of 0.8 g/kg/day) or control group
for one year.
15
After six months, the low-protein group had a 28%
reduction in microalbuminuria compared with the control group,
despite achieving only 1.12 g/kg/day, but this had reduced to
18% by 12 months. The same authors repeated this study with
a follow up of 28 months.
16
No significant differences were found
in glomerular filtration rate (GFR) or microalbuminuria but the
difference in protein intake was not sustained past six months.
Macroalbuminuria
In a small study by Sugimoto
et al
.,
14
hospital in-patients with type
2 diabetes and overt nephropathy (proteinuria > 1 g/day) were
placed either on a conventional diet or a low-protein diet for four
weeks (< 0.8 g/kg/day).
17
The 24-hour urinary protein excretion rate
decreased significantly in the low-protein group but there was no
change in serum creatinine or creatinine clearance rate.
Meloni
et al
. conducted two studies with good compliance
in subjects with either type 1 or type 2 diabetes, with results
not separated for type of diabetes.
18–19
The earlier study (
n
=
69) randomised people with type 1 or 2 diabetes and diabetic
nephropathy to either a low-protein diet (0.6 g/kg/day) or an
ad
libitum
diet.
18
There was no significant difference between the two
groups for GFR after 12 months. The later study compared a group
of 80 patients with diabetic nephropathy (type 1 and 2 diabetes)
with a group of 89 patients with chronic renal failure (CRF) not
due to diabetes.
19
Half of each group were assigned to protein
restriction (0.8 g/kg/day) and the other half to an
ad libitum
diet.
There were no changes in measures of renal function in the group
with diabetes, but in the CRF-only group there was a significant
reduction in the rate of decline of renal function on the protein-
restricted diet, suggesting that a low-protein diet may be more
appropriate for treatment of CRF due to other causes.
Mixed renal function
In a Mexican trial, 60 patients with type 2 diabetes and normo-,
micro- or macro-albuminuria were randomised to either a low-
protein diet (0.6–0.8 g/kg/day) or normal diet for four months.
20
After four months, only the patients with macroalbuminuria on
the low-protein diet experienced a significant decrease in urinary
albumin excretion rate (AER) and increase in GFR. Also in a mixed
diabetic population, Dussol
et al
. followed 63 patients (37 with type
2 diabetes) randomised to either a low-protein or usual protein diet
for two years.
21
In patients with type 2 diabetes, the rate of decline
in GFR was no different between groups over two years. While
both groups consumed 20% of TE as protein at baseline, the usual-
protein group maintained this intake, while the low-protein group
achieved and maintained a reduction to 16% of TE.
Therefore, from the limited literature in patients with type
2 diabetes, while protein restriction may result in a decrease
in urinary AER in the short term (≤ 6 months) in patients with
some renal impairment, in the longer term it appears to make no
difference to renal function. Current available evidence suggests
that long-term protein restriction is difficult to maintain. Moreover,
when dietary protein is restricted, patient satisfaction decreases,
which in turn affects compliance.
22
Concern has also been raised
over the nutritional adequacy of low-protein diets, particularly
when combined with other restrictions, such as of potassium and
phosphorus, as well as their effect on the state of well-being of the
patient.
23
The effect of protein source on renal function in
type 2 diabetes
It has been proposed that red meat intake may have a greater
adverse effect on renal deterioration than that of other protein
sources in those with type 2 diabetes. Two studies in which plant
protein was substituted for animal protein in 50–60% of the daily
protein intake showed no effect on renal function or proteinuria in
those with microalbuminuria.
24–25
When Gross
et al
. replaced red
meat with chicken and compared this with a usual diet and a low-
protein diet in patients with microalbuminuria, only the chicken-
based diet reduced urinary AER,
26
even though the low-protein diet
was significantly lower in protein and energy than both the usual
and chicken-based diets. De Mello
et al
. took this a step further
and compared a chicken-based diet with a lacto-vegetarian diet
(containing dairy products only) and usual diet in a four-week
cross-over study in patients with macroalbuminuria.
27
The urinary
albumin excretion rate was lower after the chicken-based diet than
the usual diet, and still lower after the lacto-vegetarian diet. Both
of these studies were of short duration (four weeks).
It appears that there is some preliminary evidence of an effect of
dietary protein type, regardless of quantity, on urinary AER in type 2
diabetes, favouring diets in which red meat is completely replaced
with chicken and plant proteins. There are several hypotheses
that may explain these results; the protein source may influences
hyperfiltration, with a higher polyunsaturated fat diet having a
beneficial effect on endothelial function, insulin sensitivity and lipid
profile and hence, albuminuria.
27
Reducing red meat in the diet also
appears beneficial for blood pressure, whether due to reduced dietary
sodium intake or net acid load, and this in turn favours optimal renal
function.
28-29
The DASH (Dietary approaches to stop hypertension)
diet has been established as a sodium-controlled diet to prevent
hypertension, but it also contains very little red meat, emphasising
instead vegetables, nuts, legumes and low-fat dairy products.
28
High-protein diets and renal deterioration
Is there evidence that a high-protein diet or an increase in
protein intake accelerates renal impairment? Two cross-sectional
observational studies in patients with type 2 diabetes found
no statistically significant association of protein intake with
microalbuminuria.
30–31
A third cross-sectional observational study,
however, did find a significant association, with patients with
microalbuminuria consuming more total protein and more animal
protein than patients with normal renal function.
32
Observational
studies may not account for differences in glycaemic control and
blood pressure, however, both of which would have significant
impacts on renal function.
Three interventional studies of high-protein diets (30% protein:
40% carbohydrate versus 15% protein: 55% carbohydrate) in
people with type 2 diabetes monitored serum creatinine and
reported no changes for either study group.
6, 33–34
Two of these were
weight-loss trials and one tested the effect on glycaemic control
while weight was stable. In a highly controlled study, Gannon
et al.
conducted a five-week cross-over trial with a 2–5-week washout
