Content uploaded by Satoru Yamada
Author content
All content in this area was uploaded by Satoru Yamada on Dec 05, 2014
Content may be subject to copyright.
13
□ORIGINAL ARTICLE □
A Non-calorie-restricted Low-carbohydrate Diet is Effective
as an Alternative Therapy for Patients with Type 2 Diabetes
Yoshifumi Yamada, Junichi Uchida, Hisa Izumi, Yoko Tsukamoto,
Gaku Inoue, Yuichi Watanabe, Junichiro Irie and Satoru Yamada
Abstract
Objective Although caloric restriction is a widely used intervention to reduce body weight and insulin re-
sistance, many patients are unable to comply with such dietary therapy for long periods. The clinical effec-
tiveness of low-carbohydrate diets was recently described in a position statement of Diabetes UK and a sci-
entific review conducted by the American Diabetes Association. However, randomised trials of dietary inter-
ventions in Japanese patients with type 2 diabetes are scarce. Therefore, the aim of this study was to examine
the effects of a non-calorie-restricted, low-carbohydrate diet in Japanese patients unable to adhere to a
calorie-restricted diet.
Methods The enrolled patients were randomly allocated to receive a conventional calorie-restricted diet or
low-carbohydrate diet. The patients received consultations every two months from a registered dietician for
six months. We compared the effects of the two dietary interventions on glycaemic control and metabolic
profiles.
Results The HbA1c levels decreased significantly from baseline to six months in the low-carbohydrate diet
group (baseline 7.6±0.4%, six months 7.0±0.7%, p=0.03) but not in the calorie-restricted group (baseline
7.7±0.6%, six months 7.5±1.0%, n.s.), (between-group comparison, p=0.03). The patients in the former group
also experienced improvements in their triglyceride levels, without experiencing any major adverse effects or
a decline in the quality of life.
Conclusion Our findings suggest that a low-carbohydrate diet is effective in lowering the HbA1c and tri-
glyceride levels in patients with type 2 diabetes who are unable to adhere to a calorie-restricted diet.
Key words: calorie restriction, low-carbohydrate diet, type 2 diabetes
(Intern Med 53: 13-19, 2014)
(DOI: 10.2169/internalmedicine.53.0861)
Introduction
Type 2 diabetes is a worldwide pandemic that is accom-
panied by an obesity pandemic (1, 2). Visceral fat accumula-
tion is a significant risk factor for the onset of metabolic
syndrome and diabetes (3), and weight reduction is a recom-
mended target for the first-line treatment and prevention of
diabetes (4).
The American Diabetes Association (ADA) recommends
that monitoring carbohydrate intake should be a component
of diabetes therapy (4). Moreover, since 2008, the ADA has
also recognised that low-carbohydrate diets, as well as
calorie-restricted diets, are effective interventions for weight
reduction (5). Notably, this recommendation differs from
that reported in 2007, when the ADA did not recommend a
carbohydrate intake of <130 g/day (6). Furthermore, Diabe-
tes UK recently proposed that low-carbohydrate diets may
also provide an effective treatment option (7), and the ADA
reported the effectiveness of such diets for blood glucose
and lipid management in a systematic review (8).
In contrast, the Japan Diabetes Society (JDS) currently
recommends calorie restriction (25-35 kcal/kg of ideal body
weight) for blood glucose management (9). Although calorie
restriction has been confirmed to be effective for treating
metabolic disorders, aging, carcinogenesis and dementia in
Diabetes Center, Kitasato Institute Hospital, Japan
Received for publication April 25, 2013; Accepted for publication July 25, 2013
Correspondence to Dr. Satoru Yamada, yamada-s@insti.kitasato-u.ac.jp
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
14
Rhesus monkeys (10), it may be difficult to adhere to such a
diet for long periods.
Therefore, we examined the effectiveness of a non-
calorie-restricted, low-carbohydrate diet for blood glucose
and body weight management in Japanese patients with type
2 diabetes who were unable to adhere to a calorie-restricted
diet.
Materials and Methods
Study design and subjects
This study was a single centre, 6-month, comparative,
two-arm, randomised, open-label trial performed between
April 1, 2011 and January 31, 2012. We recruited patients
with type 2 diabetes who were being treated in our outpa-
tient clinic who had received guidance regarding calorie re-
striction at least once and whose HbA1c level at enrolment
was 6.9-8.4%, suggesting that their blood glucose level was
not adequately controlled. Patients with proteinuria of >1.0
g/day, a serum creatinine level of >132 μmol/L (men) or
106 μmol/L (women), an aspartate aminotransferase (AST)
or alanine aminotransferase (ALT) level of >3 times the up-
per limit of normal, a history of myocardial infarction or
stroke within six months before study entry or an absolute
change in the HbA1c of >1.0% within six months before
study entry were excluded from this study. We avoided re-
cruiting any patients with ketosis because ketoacidosis,
which is a life-threatening complication, are reported during
ketogenic low-carbohydrate diet (11, 12).
The enrolled patients were randomly allocated to receive
either a non-calorie-restricted, low-carbohydrate diet (hereaf-
ter low-carbohydrate diet) or calorie-restricted diet using a
permuted randomised block of four patients per block. The
patients and investigators were not masked to group assign-
ment. This study was undertaken in accordance with the
Declaration of Helsinki and the Guidelines for Good Clini-
cal Practice and was approved by our institutional review
board (study ID 1010-02). Written informed consent was
obtained from all enrolled patients.
Procedures
The enrolled patients were followed up at our outpatient
clinic every two months. They received diet instructions at
every medical consultation. During the study period, we did
not change the medications, unless hypoglycaemia occurred.
The HbA1c level, laboratory blood tests, body weight and
incidence of hypoglycaemic episodes since the previous visit
were recorded every two months. Dietary intake over the
three days prior to each visit was also recorded.
The primary objective was to compare the reductions in
the HbA1c level and body weight from baseline to the end
of the 6-month treatment period between the two groups.
The secondary efficacy variables included the lipid levels
(total cholesterol (TC), triglycerides (TGs), high-density
lipoprotein cholesterol (HDL-C) and low-density lipoprotein
cholesterol (LDL-C) calculated using the Friedewald equa-
tion), blood pressure, markers of atherosclerosis (pulse-wave
velocity (PWV), ankle-brachial pressure index (ABI) and
toe-brachial pressure index (TBI)), the renal function (uri-
nary nitrogen (UN), creatinine (Cr) and estimated glomeru-
lar filtration rate (eGFR), albumin-to-creatinine ratio
(ACR)), the urinary albumin (UA) level and the levels of
liver enzymes (AST, ALT and γ-glutamyl transpeptidase
(γGTP)). To evaluate the effects of the diets on the quality
of life, the patients completed the Diabetes Treatment Satis-
faction Questionnaire (DTSQ) (13) and the Problem Areas
In Diabetes (PAID) scale (14) at enrolment and at the end of
the study.
The safety variables included adverse events reported by
the patients or noted by the investigators, standard hema-
tologic and blood chemistry tests, body weight and hypogly-
caemia. Hypoglycaemia was defined as an event with typical
symptoms (e.g., sweating, palpitations and feelings of hun-
ger) with or without confirmation by a plasma glucose level
of <70 mg/dL.
Meal instruction
Patients who were assigned to the calorie-restricted diet
received face-to-face guidance on how to calculate their
calorie intake by classifying macronutrients. The target calo-
rie intake was defined based on the Japan Diabetes Society
recommendations, as follows: total calorie intake (kcal) =
ideal body weight (kg; =height (m) ×height (m) ×22) ×25.
The target intake of specific macronutrients was as follows:
carbohydrates=50-60%, protein=1.0-1.2 g/kg (<20%) and
fat=<25% (9). For the low-carbohydrate diet, we set the to-
tal carbohydrate intake to be <130 g/day, as proposed by
Accurso et al. (15). To prevent ketosis (11, 12, 16), we set
the lower limit of carbohydrate intake to 70 g/day. To pre-
vent postprandial hyperglycaemia, the target carbohydrate
content in each meal was 20-40 g, and the subjects were al-
lowedtoconsumesweetscontaining5gofcarbohydrates
twice daily, thus resulting in a total carbohydrate intake of
70-130 g/day. To avoid any possible influence of the experi-
ence and consulting skills of the dieticians in this study, four
registered dieticians instructed the patients in both groups.
Sample size calculation
We estimated that the change in the HbA1c level over six
months would be 0.0±0.5% in the calorie-restricted group
and 0.7±0.5% in the low-carbohydrate group. We needed 22
patients, with α=0.05 and power=0.90. Therefore, we de-
cided to enrol 24 patients in this study.
Statistical analysis
The results are presented as the mean±standard deviation.
The Mann-Whitney U test was used for within-group and
between-group comparisons. The Spearman’s rank correla-
tion test was used to assess the correlations between the car-
bohydrate or calorie intake and outcomes. Values of p<0.05
were considered to indicate a statistically significant differ-
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
15
Figure
1. Flow diagram of the patients.
Assessed for eligibility
(n=24)
Randomised
(n=24)
Allocated to :
low-carbohydrate diet (n=12)
Received intervention (n=12)
Allocated to :
calorie-restricted diet (n=12)
Received intervention (n=12)
Completed (n=12, 100%) Completed (n=12, 100%)
Table
1. Baseline Characteristics of the Patients Allocated to
Each Diet
Low-carbohydrate diet Calorie-restricted diet
Age (years) 63.3 ± 13.5 63.2 ± 10.2
Sex (male/female) 7/5 5/7
Duration of diabetes (years) 8.9 ± 3.6 9.5 ± 4.8
BMI (kg/m2) 24.5 ± 4.3 27.0 ± 3.0
BW (kg) 67.0 ± 15.9 68.1 ± 7.7
HbA1c (NGSP) (%) 7.6 ± 0.4 7.7 ± 0.6
LDL-C (mg/dL) 99.8 ± 28.2 112.2 ± 20.5
TG (mg/dL) 141.7 ± 76.4 155.2 ± 86.4
HDL-C (mg/dL) 62.8 ± 17.1 59.8 ± 19.1
SBP (mmHg) 124.4 ± 10.8 124.9 ± 10.7
DBP (mmHg) 72.6 ± 6.2 74.8 ± 10.6
DTSQ total score (except
items 2 and 3) 24.0 ± 6.6 21.6 ± 3.3
PAID score 42.1 ± 13.5 57.8 ± 12.6
Glucose-lowering drug ------no. (%)
Insulin 3 (25.0) 4 (33.3)
Metformin 5 (41.7) 1 (8.3)
Sulfonylurea 5 (41.7) 8 (66.7)
Glinide 1 (8.3) 0 (0.0)
Thiazolidinedione 4 (33.3) 6 (50.0)
Į-Glucosidase inhibitor 2 (16.7) 0 (0.0)
DPP-4 inhibitor 2 (16.7) 3 (25.0)
GLP-1 0 (0.0) 0 (0.0)
None 0 (0.0) 0 (0.0)
History of major microvascular disease
Retinopathy
None 10 10
SDR 0 2
PPDR 1 0
PDR 1 0
Nephropathy
Stage 1 7 8
Stage 2 4 3
Stage 3A 1 1
Stage 3B, 4, 5 0 0
Values are means ± standard deviation.
There were no statistically significant differences between the two groups for
any parameter.
BMI: body mass index, BW: body weight, HbA1c: haemoglobin A1c, NGSP:
National Glycohemoglobin Standardization Program, LDL-C: low-density
lipoprotein–cholesterol, TG: triglyceride, HDL-C: high-density
lipoprotein–cholesterol, SBP: systolic blood pressure, DBP: diastolic blood
pressure, DTSQ: Diabetes Treatment Satisfaction Questionnaire, PAID:
Problem Areas In Diabetes scale, DPP: dipeptidyl peptidase, GLP:
glucagon-like peptide, SDR: simple diabetic retinopathy, PPDR: pre
proliferative diabetic retinopathy, PDR: proliferative diabetic retinopathy
ence.
Results
Study population
The methodology of recruitment and screening is summa-
rized in Fig. 1. A total of 24 patients with type 2 diabetes
(mean age, 63.3±11.7 years; 50% women) who were being
followed up at our outpatient clinic were enrolled and ran-
domly allocated to receive either a low-carbohydrate diet or
calorie-restricted diet. The general characteristics of the en-
rolled patients in each group are shown in Table 1. There
were no statistically significant differences in any of the
parameters between the two groups.
The low-carbohydrate diet improved the blood glu-
cose levels and lipid levels (Table 2)
Six months after starting the diet, the HbA1c levels were
significantly lower than those observed at baseline in the
low-carbohydrate group (baseline 7.6±0.4%, six months
7.0±0.7%, p=0.03), whereas there were no changes in the
HbA1c levels in the calorie-restricted group (baseline 7.7±
0.6%, six months 7.5±1.0%, not significant (n.s.)) (Fig. 2a).
The low-carbohydrate diet significantly improved the HbA1c
levels in comparison with the calorie-restricted diet (p=
0.03). The fasting plasma glucose levels were similar be-
tween baseline and at six months in both groups.
Body weight, BMI and blood pressure did not change sig-
nificantly in either group.
In terms of the lipid levels, the TG levels significantly de-
creased in the low-carbohydrate group (baseline 141.7±76.2
mg/dL, six months 83.5±40.6 mg/dL, p=0.02), whereas no
changes in the TG levels occurred in the calorie-restricted
group (baseline 155.2±86.4 mg/dL, six months 148.4±90.7
mg/dL, n.s.) (Fig. 2b). However, the difference between the
two groups was not statistically significant (p=0.08). The
other markers of lipid profiles, the LDL-C levels and the
HDL-C levels, were not altered in either group. Neither diet
significantly affected markers of atherosclerosis, such as
ABI, TBI and PWV.
Although we were concerned that excess protein intake
may cause deterioration of the renal function in the low-
carbohydrate group, we found no changes in the markers of
the renal function (i.e., UN, Cr, eGFR and ACR) during the
6-month study in either group. A marker of the liver func-
tion, the ALT level, tended to improve in the low-
carbohydrate group, likely due to the decrease in liver fat
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
16
Table
2. Efficacy Outcomes
Variable Low-carbohydrate diet Calorie-restricted diet p value*
Baseline 6 months p value†Baseline 6 months p value†
HbA1c NGSP
(%) 7.6 ± 0.4 7.0 ± 0.7 0.03 7.7 ± 0.6 7.5 ± 1.0 n.s.
(0.45) 0.03
FPG (mg/dL) 138 ± 44 124 ± 22 n.s.
(0.40) 155 ± 46 163 ± 26 n.s.
(0.42)
n.s.
(0.33)
BW (kg) 67.0 ± 15.9 64.4 ± 14.2 n.s.
(0.62) 68.1 ± 7.7 66.7 ±
7.0
n.s.
(0.56)
n.s.
(0.80)
BMI (kg/m2) 24.5 ± 4.3 23.6 ± 3.5 n.s.
(0.39) 27.0 ± 3.0 26.4 ±
2.2
n.s.
(0.42)
n.s.
(0.86)
LDL-C (mg/dL) 99.8 ± 28.2 95.2 ± 21.0 n.s.
(0.77)
112.2 ±
20.5
110.5 ±
21.7
n.s.
(0.77)
n.s.
(0.49)
TG (mg/dL) 141.7 ±
76.2 83.5 ± 40.6 0.02 155.2 ±
86.4
148.4 ±
90.7
n.s.
(0.58) 0.08
HDL-C (mg/dL) 62.8 ± 17.2 68.2 ± 22.1 n.s.
(0.44) 59.8 ± 19.1 55.6 ±
13.9
n.s.
(0.82)
n.s.
(0.13)
SBP (mmHg) 124.4 ±
10.8
122.5 ±
11.9
n.s.
(0.69)
124.9 ±
10.7
121.3 ±
11.6
n.s.
(0.42)
n.s.
(0.54)
DBP (mmHg) 72.6 ± 6.2 66.6 ± 9.4 n.s.
(0.11) 74.8 ± 10.1 73.4 ±
10.1
n.s.
(0.91)
n.s.
(0.30)
PWV (cm/sec) 1,723.0 ±
213.8
1,788.8 ±
230.0
n.s.
(0.60)
1,616.0 ±
162.9
1,626.3 ±
293.7
n.s.
(0.77)
n.s.
(0.69)
ABI 1.104 ±
0.093
1.153 ±
0.101
n.s.
(0.24)
1.172 ±
0.085
1.187 ±
0.080
n.s.
(0.44)
n.s.
(0.36)
TBI 0.635 ±
0.265
0.644 ±
0.216
n.s.
(0.54)
0.727 ±
0.134
0.707 ±
0.168
n.s.
(0.80)
n.s.
(0.91)
UN (mg/dL) 15.3 ± 3.1 17.1 ± 5.9 n.s.
(0.47) 13.1 ± 3.4 14.0 ±
5.8
n.s.
(0.77)
n.s.
(0.77)
eGFR
(mL/min/1.73m2)69.0 ± 14.5 69.4 ± 15.0 n.s.
(0.71) 69.1 ± 13.2 65.0 ±
12.6
n.s.
(0.33)
n.s.
(0.39)
ACR (mg/gCr) 141.7 ±
322.1
96.8 ±
184.6
n.s.
(0.69) 53.0 ± 93.0 131.5 ±
231.7
n.s.
(0.89)
n.s.
(0.21)
UA (mg/dL) 5.6 ± 1.3 5.7 ± 1.1 n.s.
(0.69) 5.4 ± 1.0 5.4 ± 1.4 n.s.
(0.69)
n.s.
(0.69)
AST (U/L) 26.9 ± 7.6 24.1 ± 7.8 n.s.
(0.36) 31.7 ± 17.1 34.7 ±
27.0
n.s.
(0.77)
n.s.
(0.29)
ALT (U/L) 28.6 ± 12.5 21.4 ± 5.9 0.07 32.4 ± 19.4 32.6 ±
17.5
n.s.
(0.95)
n.s.
(0.11)
ȖGTP (U/L) 38.5 ± 20.2 33.0 ± 16.4 n.s.
(0.53) 35.8 ± 25.4 37.3 ±
33.7
n.s.
(0.95)
n.s.
(0.18)
DTSQ total score 24.0 ± 6.6 27.6 ± 5.7 n.s.
(0.23) 21.6 ± 3.3 24.7 ±
4.5 0.07 n.s.
(0.95)
DTSQ item 2:
High BS 3.50 ± 1.68 2.42 ± 1.83 n.s.
(0.13) 3.83 ± 0.94 3.67 ±
1.37
n.s.
(0.77)
n.s.
(0.21)
DTSQ item 3:
Low BS 1.17 ± 1.90 1.42 ± 1.98 n.s.
(0.95) 1.83 ± 1.53 1.75 ±
1.14
n.s.
(0.98)
n.s.
(0.31)
PAID score 42.1 ± 13.5 37.8 ± 11.3 n.s.
(0.37) 57.8 ± 12.6 57.2 ±
11.9
n.s.
(0.98)
n.s.
(0.64)
*For between-group comparisons; †for within-group comparisons.
Values are means ± standard deviation.
HbA1c: haemoglobin A1c, n.s.: not significant, FBG: fasting blood glucose, BW: body weight, BMI:
body mass index, NGSP: National Glycohemoglobin Standardization Program, LDL-C: low-density
lipoprotein–cholesterol, TG: triglyceride, HDL-C: high-density lipoprotein–cholesterol, SBP: systolic
blood pressure, DBP: diastolic blood pressure, PWV: pulse-wave velocity, ABI: ankle–brachial pressure
index, TBI: toe–brachial pressure index, UN: urinary nitrogen, eGFR: estimated glomerular filtration
index, ACR: albumin-to-creatinine ratio, UA: urinary albumin, AST: aspartate aminotransferase, ALT:
alanine aminotransferase, ȖGTP: Ȗ-glutamyl transpeptidase, DTSQ: Diabetes Treatment Satisfaction
Questionnaire, BS: blood sugar, PAID: Problem Areas In Diabetes scale
content associated with the reduction in body weight. How-
ever, we did not assess the liver fat content using ultrasound
or magnetic resonance spectrometry in this study.
Concerning the quality of life, the DTSQ score and the
PAID score did not change in either group.
Taken together, these results indicate that the low-
carbohydrate diet achieved greater improvements in the
blood glucose and TG levels than the calorie-restricted diet.
Particularly, the low-carbohydrate diet significantly im-
proved the HbA1c levels in comparison with the calorie-
restricted diet.
Three patients treated with a sulfonylurea or insulin in the
low-carbohydrate group experienced symptomatic hypogly-
caemia, although the events did not recur after adjusting the
medications. None of the patients developed ketonuria dur-
ing the study.
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
17
Figure
2. Changes in the HbA1c (a) and triglyceride (b) levels during the 6-month intervention.
Solid line: caloric restriction diet, dotted line: low-carbohydrate diet
Table
3. Nutrition Intake at 6 Months
Low-carbohydrate diet Calorie-restricted diet p value*
Intake Energy
ratio (%) Intake Energy
ratio (%)
All patients
Calorie intake (kcal) 1,634 ± 531 100 1,610 ± 387 100 n.s. (0.84)
Males
Calorie intake (kcal) 1,891 ± 400 1,684 ± 526 n.s. (0.42)
Calorie intake/IBW 29.9 ± 6.7 27.2 ± 8.5 n.s. (0.68)
Females
Calorie intake (kcal) 1,274± 507 1,557 ± 286 n.s. (0.22)
Calorie intake/IBW 24.1±10.7 30.3 ± 5.2 n.s. (0.17)
Intake of specific nutrients
Protein (g) 100.4 ± 36.6 25.3 ± 7.3 67.6 ± 21.2 16.6 ± 2.8 0.021
Protein/BW 1.592 ± 0.573 1.022 ± 0.329 0.009
Fat (g) 82.1 ± 33.0 45.4 ± 8.9 58.5 ± 20.7 32.3 ± 5.2 0.028
Carbohydrate (g) 125.7 ± 71.9 29.8 ± 12.5 202.9 ± 42.0 51.0 ± 4.6 0.008
Salt (g) 10.2 ± 2.5 10.4 ± 2.4 n.s. (0.30)
*For between-group comparisons.
Values are means ± standard deviation.
n.s.: not significant, IBW: ideal body weight, BW: body weight
Nutrient intake at six months
Although we did not prescribe calorie restriction to the
patients assigned to the low-carbohydrate group, the calorie
intake at six months was actually similar in both groups (Ta-
ble 3). Stratified by sex, the calorie intake in the low-
carbohydrate group tended to be higher in men and lower in
women compared with that observed in the calorie-restricted
group. The relative nutrient intake of carbohydrates, protein
and fat was 29.8±12.5%, 25.3±7.3% and 45.4±8.9% in the
low-carbohydrate group, compared with 51.0±4.6%, 16.6±
2.8% and 32.3±5.2%, respectively, in the calorie-restricted
group. The carbohydrate intake was significantly lower in
the low-carbohydrate group than in the calorie-restricted
group. The mean carbohydrate intake in the low-
carbohydrate group was <130 g/day, suggesting that most
patients were able to adhere to the meal instructions.
Correlations between carbohydrate or calorie intake
and outcomes
We next performed correlation analyses in 10 patients
with nutrient intake data both at baseline and at the end of
the study (five patients in each group). In these patients, we
found that the change in body weight was significantly cor-
related with the change in carbohydrate intake (r=0.764, p=
0.0078) and the change in calorie intake (r=0.769, p=
0.0071). The change in the HbA1c level was significantly
correlated with the change in carbohydrate intake (r=0.670,
p=0.0321) but not with the change in calorie intake (r=
0.439, n.s.).
Discussion
In this randomised study, we examined whether a non-
calorie-restricted, low-carbohydrate diet is effective and safe
in Japanese patients with type 2 diabetes and inadequate
glycaemic control while on a calorie-restricted diet. Signifi-
cant improvements in the HbA1c level were observed in the
low-carbohydrate group but not in the calorie-restricted
group. Overall, these findings are compatible with the rec-
ommendations of the ADA (4, 5, 8) and Diabetes UK (7).
We also found improvements in the TG levels in the low-
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
18
carbohydrate group. Two previous meta-analyses, which in-
cluded five trials (17) and 13 trials (18), respectively,
showed worsening of the TC and LDL-C levels and im-
provements in the TG and HDL-C levels with a low-
carbohydrate diet. However, another meta-analysis of 22 tri-
alsshowednodeteriorationintheTCorLDL-Clevelsand
significant improvements in the TG and HDL-C levels with
a low-carbohydrate diet (19). Consistent with that report, we
found improvements in the TG levels without deterioration
in the TC or LDL-C levels. Therefore, a low-carbohydrate
diet appears to have beneficial effects on lipid profiles as
well as glycaemic control.
The ADA recommends monitoring the renal function and
protein intake, particularly in patients with nephropathy, dur-
ing low-carbohydrate dietary interventions in addition to
monitoring lipid profiles (4, 5). Although the subjects in the
low-carbohydrate group consumed less carbohydrates, more
protein and more fat than those in the calorie-restricted
group, the renal function did not deteriorate in the former
group. The protein intake in the low-carbohydrate group in
this study was 1.6±0.6 g/kg. Although the Japanese Refer-
ence Diet Intake report 2010 recommended an upper limit
of protein intake of 2.0 g/kg (20) based on data from criti-
cally ill patients, critically ill patients should be allowed to
receive undernutrition (21) and the Dietary Reference Intake
in the US has no upper limit for protein intake (22).
Although a previous report claimed that atherosclerosis
may be caused by a low-carbohydrate diet (23), we did not
observe cerebrovascular end points due to the short observa-
tion period. There was no deterioration in the levels of
markers of atherosclerosis (i.e., ABI, TBI and PWV) in the
low-carbohydrate group. It is necessary to evaluate the long-
term safety of low-carbohydrate diets for atherosclerosis
carefully in the future.
Although we did not limit calorie intake in the low-
carbohydrate diet group, the calorie intake at six months af-
ter the intervention was almost equal in both groups. In the
low-carbohydrate diet group, the patients ate protein and fat,
such as meat or fish, in substitution of carbohydrates. As a
result, the quantity of the meal increased and the patients
may therefore more easily feel full. In addition, lightly sea-
soned staple foods (such as rice, bread, noodles and so on),
which are rich in carbohydrates, may promote overeating of
strongly seasoned side dishes (such as meat, fish and so on),
and it may be difficult to frequently eat only side dishes. On
the other hand, because all enrolled patients had previously
received guidance on calorie restriction at least once, we
cannot deny the possibility that the patients allocated to the
low-carbohydrate diet group limited not only the carbohy-
drate volume, but also the calorie intake under the influence
of former meal instruction.
There are several limitations to this study that should be
discussed. First, the number of subjects enrolled was too
small to detect significant differences in the between-group
and within-group comparisons. Santos et al. reviewed 23 re-
ports, consisting of 1,141 patients in total, concerning the
effects of low-carbohydrate diets and reported that low-
carbohydrate diets are clearly associated with significant de-
creases in body weight, BMI, abdominal circumference, sys-
tolic blood pressure, diastolic blood pressure, plasma TG,
fasting plasma glucose, glycated haemoglobin, plasma insu-
lin and plasma CRP as well as increases in HDL-C (24). We
did not observe any significant improvements in body
weight, BMI, blood pressure or HDL-C in this study; how-
ever, increasing the number of subjects may prove the sig-
nificant effectiveness of low-carbohydrate diets in improving
these parameters. Second, this was a 6-month study. Several
clinical trials of low-carbohydrate diets have shown small or
moderate rebounds in clinical parameters between six
months and 12 months after starting the interven-
tion (25, 26). Therefore, the improvements in the HbA1c
levels and body weight observed at six months in our study
may be attenuated over a longer observation period.
In conclusion, we found that a non-calorie-restricted, low-
carbohydrate diet is effective in lowering the HbA1c and TG
levels and is safe as an alternative therapy for patients with
type 2 diabetes. Because the number of subjects enrolled
was small and the study duration was short, longer multi-
centre studies with a larger sample size are needed to con-
firm our findings.
The authors state that they have no Conflict of Interest (COI).
References
1. The diabetes pandemic. Lancet 378: 99, 2011.
2. Ramachandran A, Ma RC, Sneharatha C. Diabetes in Asia. Lancet
375: 408-418, 2010.
3. Alberti KG, Zimmet P, Shaw J; IDF Epidemiology Task Force
Consensus Group. The metabolic syndrome: a new worldwide
definition. Lancet 366: 1059-1062, 2005.
4. American Diabetes Association. Standards of medical care in dia-
betes 2012. Diabetes Care 35: S11-S63, 2012.
5. American Diabetes Association. Nutrition recommendations and
interventions for diabetes. Diabetes Care 31: S61-S78, 2008.
6. American Diabetes Association. Nutrition recommendations and
interventions for diabetes. Diabetes Care 30: S48-S65, 2007.
7. Dyson PA, Kelly T, Deakin A, et al. Diabetes UK evidence-based
nutrition guidelines for the prevention and management of diabe-
tes. Diabet Med 28: 1282-1288, 2011.
8. Wheeler ML, Dunbar SA, Jaacks LM, et al. Macronutrients, food
groups, and eating patterns in the management of diabetes: a sys-
tematic review of the literature, 2010. Diabetes Care 35: 434-445,
2012.
9. Japan Diabetes Society. Diet therapy. In: Treatment Guide for Dia-
betes 2012-2013. Bunkodo, Tokyo, 2012: 39-42 (in Japanese).
10. Colman RJ, Anderson RM, Johnson SC, et al. Caloric restriction
delays disease onset and mortality in rhesus monkeys. Science
325: 201-204, 2009.
11. Chen TY, Smith W, Rosenstock JL, Lessnau KD. A life-
threatening complication of Atkins diet. Lancet 367: 958, 2006.
12. Shah P, Isley WL. Ketoacidosis during a low-carbohydrate diet. N
Engl J Med 354: 97-98, 2006.
13. Bradley C. Diabetes Treatment Satisfaction Questionnaire. Change
version for use alongside status version provides appropriate solu-
tion where ceiling effects occur. Diabetes Care 22: 530-532, 1999.
14. Welch GW, Jacobson AM, Polonsky WH. The Problem Areas in
Intern Med 53: 13-19, 2014 DOI: 10.2169/internalmedicine.53.0861
19
Diabetes Scale. An evaluation of its clinical utility. Diabetes Care
20: 760-766, 1997.
15. Accurso A, Bernstein RK, Dahlgvist A, et al. Dietary carbohydrate
restriction in type 2 diabetes mellitus and metabolic syndrome:
time for a critical appraisal. Nutr Metab (Lond) 5: 9, 2008.
16. Westman EC, Feinman RD, Mavropoulos JC, et al. Low-
carbohydrate nutrition and metabolism. Am J Clin Nutr 86: 276-
284, 2007.
17. Nordmann AJ, Nordmann A, Briel M, et al. Effects of low-
carbohydrate vs low-fat diets on weight loss and cardiovascular
risk factors. Arch Intern Med 166: 285-293, 2006.
18. Hession M, Rolland C, Kulkarni U, Wise A, Broom J. Systematic
review of randomized controlled trials of low-carbohydrate vs.
low-fat/low-calorie diets in the management of obesity and its
comorbidities. Obes Rev 10: 36-50, 2009.
19. Kodama S, Saito K, Tanaka S, et al. Influence of fat and carbohy-
drate proportions on the metabolic profile in patients with type 2
diabetes: a meta-analysis. Diabetes Care 32: 959-965, 2009.
20. Ministry of Health, Labour, and Welfare, Japan. Protein, in Die-
tary Reference Intakes for Japanese 2010. Daiichi Shuppan Pub-
lishing Co., Tokyo, 2010: 62-76 (in Japanese).
21. Casaer MP, Mesotten D, Hermans G, et al. Early versus late par-
enteral nutrition in critically ill adults. N Engl J Med 365: 506-
517, 2011.
22. Institute of Medicine of the National Academies. Protein and
Amino Acids, in Dietary Reference Intakes for Energy, Carbohy-
drate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Ac-
ids. The National Academies Press, Washington DC, 2005: 589-
768.
23. Smith SR. A look at the low-carbohydrate diet. N Engl J Med
361: 2286-2288, 2009.
24. Santos FL, Esteves SS, da Costa Pereira A, Yancy WS Jr, Nunes
JP. Systematic review and meta-analysis of clinical trials of the ef-
fects of low carbohydrate diets on cardiovascular risk factors.
Obes Rev 13: 1048-1066, 2012.
25. Shai I, Schwarzfuchs D, Henkin Y, et al. Weight loss with a low-
carbohydrate, Mediterranean, or low-fat diet. N Engl J Med 359:
229-241, 2008.
26. Gardner CD, Kiazand A, Alhassan S, et al. Comparison of the
Atkins, Zone, Ornish, and LEARN diets for change in weight and
related risk factors among overweight premenopausal women: the
A TO Z Weight Loss Study: a randomized trial. JAMA 297: 969-
977, 2007.
Ⓒ2014 The Japanese Society of Internal Medicine
http://www.naika.or.jp/imonline/index.html
