Review of Low-Carb Vs High-Carb Diets

obesity reviews doi: 10.1111/j.1467-789X.2008.00518.
Systematic review of randomized controlled trials of

low-carbohydrate vs. low-fat/low-calorie diets in the
management of obesity and its comorbidities
M. Hession1, C. Rolland1, U. Kulkarni1, A. Wise2 and J. Broom1
1
Centre for Obesity Research and Summary
Epidemiology (CORE), Faculty of Heath and There are few studies comparing the effects of low-carbohydrate/high-protein
Social Care; 2School of Pharmacy and Life diets with low-fat/high-carbohydrate diets for obesity and cardiovascular disease
Sciences, Robert Gordon University, risk. This systematic review focuses on randomized controlled trials of low-
Aberdeen, UK carbohydrate diets compared with low-fat/low-calorie diets. Studies conducted
in adult populations with mean or median body mass index of ⱖ28 kg m-2 were
Received 15 April 2008; revised 25 June included. Thirteen electronic databases were searched and randomized controlled
2008; accepted 15 July 2008 trials from January 2000 to March 2007 were evaluated. Trials were included
if they lasted at least 6 months and assessed the weight-loss effects of low-
Address for correspondence: Dr Catherine carbohydrate diets against low-fat/low-calorie diets. For each study, data were
Rolland, CORE (Centre for Obesity Research abstracted and checked by two researchers prior to electronic data entry. The
and Epidemiology), The Robert Gordon computer program Review Manager 4.2.2 was used for the data analysis. Thirteen
University, Aberdeen AB25 1HG, UK. E-mail: articles met the inclusion criteria. There were significant differences between the
c.rolland@rgu.ac.uk groups for weight, high-density lipoprotein cholesterol, triacylglycerols and sys-
tolic blood pressure, favouring the low-carbohydrate diet. There was a higher
attrition rate in the low-fat compared with the low-carbohydrate groups suggest-
ing a patient preference for a low-carbohydrate/high-protein approach as opposed
to the Public Health preference of a low-fat/high-carbohydrate diet. Evidence
from this systematic review demonstrates that low-carbohydrate/high-protein
diets are more effective at 6 months and are as effective, if not more, as low-fat
diets in reducing weight and cardiovascular disease risk up to 1 year. More
evidence and longer-term studies are needed to assess the long-term cardiovascular
benefits from the weight loss achieved using these diets.
Keywords: Cardiovascular risk, low-carbohydrate, meta-analysis, obesity.
obesity reviews (2008)
Author contributions: Ms Hession, Dr Rolland, Dr Kulkarni, Dr Wise and

Introduction
Professor Broom had full access to all of the data in the study and take
responsibility for the integrity of the data analysis: (i) study concept and The prevalence of overweight and obesity is already high
design: Hession, Rolland, Kulkarni, Wise, Broom; (ii) acquisition of data: and continues to increase in both the developed and devel-
Hession, Rolland; (iii) analysis and interpretation of data: Hession, oping world (1). Obesity has been implicated as the second
Rolland; (iv) drafting of the manuscript: Hession, Rolland, Kulkarni; (v) most preventable cause of death in the United States. After
critical revision of the manuscript for important intellectual content: Wise, remaining reasonably constant in the 1960s and 1970s,
Broom; (vi) statistical analysis: Hession; (vii) administrative, technical or the prevalence of obesity among adults in the United States
material support: Hession, Rolland, Kulkarni, Wise, Broom and (viii) study increased by around 50% per decade throughout the 1980s
supervision: Broom. and 1990s. Two-thirds of adults in the United States today
© 2008 The Authors 1

Journal compilation © 2008 International Association for the Study of Obesity. obesity reviews
2 Review of low CHO vs. low fat diets M. Hession et al. obesity reviews
are obese or overweight. In the United States, 28% of men, review had to be of at least 6-month duration, including the
34% of women and nearly 50% of non-Hispanic black period of active intervention and follow-up.
women are at present obese (2). At any time, approxi-
mately 45% of women and 30% of men in the UK are Types of intervention
trying to lose weight (3). Most adults in England are now
overweight, and nearly one-quarter are obese (http:// The focus of this review was to examine LC/HP diets
www.foresight.gov.uk/obesity/17.pdf). Obesity has been against other types of diets designed to induce weight loss
shown to be associated with increased risk of type 2 dia- and/or prevent weight gain, and induce changes in cardio-
betes mellitus, hypertension, dyslipidemia and consequent vascular risk factors. The types of dietary intervention
cardiovascular disease. Obesity ranks second only to evaluated were:
smoking in the aetiology of cancer and is an important • HP ‘ketogenic’ diet, where the carbohydrate content
factor in osteoarthritis and obstructive sleep apnoea (4). was less than 40 g d-1, irrespective of calorie content.
Recently, low-carbohydrate/high-protein (LC/HP) diets • LC diets (carbohydrate ⱕ 60 g d-1).
have become popular as an aid to weight loss. Significant • ‘Healthy eating’ advice.
weight loss on a LC/HP diet without significant elevations • LF (30% or less daily energy from dietary fat) –
of serum cholesterol has been reported. Studies comparing 600 kcal deficit diet.
the ‘Atkins’ diet with the classical low-fat (LF) diet have
appeared in the literature recently and are the subject of Outcome measures
increasing public interest (5) due to the beneficial improve-
ments in cardiovascular risk and weight loss achieved with Weight loss or prevention of weight gain was the main
this type of dietary approach (6,7). outcomes assessed from the RCTs included in the review.
This systematic review focuses on randomized controlled With regard to cardiovascular disease risk factors, the fol-
trials (RCTs) of LC/HP diets compared with LF/high- lowing outcomes were also included:
carbohydrate (HC) conventional diets. The systematic • Serum lipids, including total cholesterol, low-density
review also examines the outcomes of such trials in relation lipoprotein (LDL) cholesterol, high-density lipoprotein
to effects on cardiovascular disease risk. This systematic (HDL) cholesterol and triacylglycerols.
review focuses on updating the literary evidence from RCTs • Systolic and diastolic blood pressure.
of LC/HP diets compared with LF/HC diets to assess their • Glycemic control.
impact on weight loss and cardiovascular risk. In addition,
it demonstrates lower attrition rates in the LC/HP groups Attrition rates were also analysed for each study to assess
compared with the LF/HC groups suggesting patient pref- patient acceptability.
erence for the former approach.
Search strategy for the identification of
included studies
Methods This systematic review was restricted to RCTs where the
full study report was available. A wide search strategy was
Inclusion criteria applied to identify as many RCTs evaluating dietary inter-
ventions as possible and which were relevant to the man-
The protocol used for this systematic review follows the agement of obesity and cardiovascular disease risk factors.
methods recommended by the Cochrane Collaboration (8). Thirteen electronic databases were searched including
RCTs were included if they assessed the weight-loss effects MEDLINE, Commonwealth Agricultural Bureau (CAB)
of LC/HP diets against LF/HC diets. Only RCTs from abstracts and the Cochrane Central Register of Controlled
January 2000 to March 2007 were evaluated, as this review Trials. The search strategy incorporated weight loss, car-
is intended to assess the current literature in this field and diovascular disease and obesity-related terms and text
update the National Health Service R&D Health Technol- terms, specific to each database. Seven obesity and nutri-
ogy Assessment systematic review of diet and lifestyle on tion journals were hand-searched including the Interna-
weight loss and cardiovascular risk published by Avenell tional Journal of Obesity and Obesity Research. Reference
et al. (8). Only studies conducted in an adult population lists of included studies were searched and authors con-
were included, as defined by minimum age greater than 18 tacted for further details of their trials.
years. RCTs where the participants had a mean or median
body mass index (BMI) of ⱖ28 kg m-2 were included. A
Quality assessment of studies
BMI cut-off of ⱖ28 kg m-2 was used to allow the inclusion
of studies of ethnic groups where the classification of Full copies of studies were assessed by two researchers
obesity is at a lower BMI cut-off (9). RCTs evaluated in this for methodological quality using a standard form. The
© 2008 The Authors

obesity reviews Review of low CHO vs. low fat diets M. Hession et al. 3
researchers were not blinded to author, journal or institu- Table 1 Summary of reasons for which papers were not included in the
tion. Differences of opinion were resolved by discussion. systematic review
Trial quality was assessed, including whether or not the Reasons Number of studies
analysis was undertaken on an intention to treat basis.
Not a randomized controlled trial 71
Study was less than 6-month duration 36
Data abstraction
Mean/median body mass index of subjects 5
A data abstraction form was created for this review based was less than 28 kg m-2
Carbohydrate content of the 14
on a standard format (8). For each study, data were
‘low-carbohydrate’ diet was too high
abstracted and checked by different researchers prior to Subjects did not receive an appropriate 50
electronic data entry. treatment
Subjects were not human 5
Subjects were under 18 6
Data analysis
The computer program Review Manager 4.2.2 was used
for the analysis of the data from the reviews. If results from
studies could be quantitatively combined, a statistical duration and six of 12-month. One trial lasted 17 months
meta-analysis of the data was undertaken to determine the and another lasted 36 months. As there was only one study
typical effect size of the intervention. For continuous data, lasting 17 months (11) and one lasting 36 months (12) data
a weighted mean difference (WMD) was calculated. The reported at that time point in that study were not included
chi-square test was used to test for heterogeneity across the in the analysis. All of the studies were designed to reduce
studies. The significance value was set at 0.05. or prevent weight gain and also examined cardiovascular
disease risk factors.
Ten of the studies compared LC/HP diets with LF/HC
Handling of missing data diets and two studies compared medium-protein diets with
Data processing for this review in Review Manager HP diets. Table 2 gives a summary of the diets and carbo-
required the input of the mean and the standard deviation hydrate content for each of the studies.
(SD) of the change between two time points. Where weight
or risk factors were reported as actual values instead of
Participant characteristics
changes, the differences were calculated by subtracting the
end point value from the baseline value. If SD for changes A total of 1222 volunteers were recruited between the 13
in weight and risk factors were missing, the following studies. Fig. 1 shows the percentage attrition rates. Out
assumption was made – a previously published linear of the 1222 participants assigned to the diets, there were
regression of the SD of the mean change in weight on the 441 (36%) attritions during the interventions. There was
absolute mean change for weight (8), derived from weight- a higher attrition rate in the conventional/LF/medium-
loss RCTs, was used to supply missing SD. Similar data protein groups compared with the LC/HP intervention
were used to infer missing SD for the other variables groups. The difference in attrition rates between the two
analysed in this review. groups was significant (P = 0.001) after performing a chi-
squared test.
Results
Quality of trials
Identified studies
For the following variables, the LC/HP refers to the LC/HP
A total of 13 (10–22) out of 1231 articles met the inclusion
intervention groups and the LF/HC refers to the LF/HC
criteria and were included in the systematic review.
comparison/control groups.
Reasons for which they were not included are summarized
in Table 1.
Weight
Study characteristics The WMD in weight change was -4.02 kg in favour of
All the included studies were RCTs ranging from 6- to the LC/HP group at 6 months (Fig. 2a) (P < 0.00001). At
36-month duration. Five of the trials were of 6-month 12 months this difference had fallen to only -1.05 kg
(P < 0.05) (Fig. 2b). There were differences (P < 0.0001)
1
A list of papers that were excluded from the systematic review can be among the studies at 6 months, but agreement shown by
obtained from the corresponding author. lack of heterogeneity at 12 months.
© 2008 The Authors

Table 2 Summary of diets and CHO content
Study Diets compared Amount of CHO consumed per day for Were the diets energy controlled?
LC/HP
Brehm et al., 2002 (10) VLC/HP vs. LF/HC 20 g increased to 40–60 g after 2 weeks LF – kcal restricted
Brinkworth et al., 2004 (11) MP vs. SP 40% CHO Initial 12 weeks energy restriction, followed
by 4 weeks energy balance
Cardillo et al., 2006 (12) LC/HP vs. LF/HC 30 g for 6 months followed by reintroduction LF – reduced by 500 kcal d-1
of CHO
Dansinger et al., 2005 (13) VLC/HP vs. LF/HC 30 g increased to 50 g gradually LF – kcal restricted
Due et al., 2004 (14) MP (12% of energy) vs. 40% CHO No (ad libitum)
HP (25% of energy),
both 30% fat
Foster et al., 2003 (15) VLC/HP vs. LF/HC 20 g then increased gradually LF – women: 1200–1500 kcal d-1; men:
1500–1800 kcal d-1
Gardner et al., 2007 (16) VLC/HP vs. LF/HC 20 g increased to 50 g gradually LF – kcal restricted
Samaha et al., 2003 (17) LC/HP vs. LF/HC 30 g LF – reduced by 500 kcal d-1
Seshadri et al., 2004 (18) LC/HP vs. LF/HC 30 g LF – reduced by 500 kcal d-1
Stern et al., 2004 (19) LC/HP vs. LF/HC 30 g LF – kcal restricted
Truby et al., 2006 (20) Atkins vs. LF/HC 20 g increased to 5 g week-1 when 10 lb LF – kcal restricted
away from target weight
Tsai et al., 2005 (21) LC/HP vs. LF/HC 30 g increased to 50 g gradually LF – kcal restricted
Yancy et al., 2004 (22) VLC/HP vs. LF/HC <20 g increased by 5 g d-1 when reached LF – reduced by 500 kcal d-1
half of target weight
CHO, Carbohydrate; HC, high-carbohydrate; HP, high-protein; LC, low-carbohydrate; LF, low-fat; MP, moderate protein; SP, standard protein; VLC,
very-low-carbohydrate.
60 (Fig. 4b). There were no differences among the studies at

% attrition rate
50
40 6 months (P = 0.65), but there were differences found
30
20
between the studies at 12 months (P < 0.00001).
10
0
03
02
04
05
5
04
06
04
03
04
07
04
00
High-density lipoprotein cholesterol

20
20
20
20
20
20
20
20
20
20
20
,2
,
,
,
,
al
,
,
,
,
,
al
al
al
al
al
al
al
al
al
al
al
et
et
et
et
et
et
et
et
et
et
et
et
er
er
m
ai
n
y
ue
a
ri
er
th
The WMD in HDL cholesterol change was 0.04 mmol L-1

nc
ub
ah
er
ng
ad
Ts
eh
st
dn
or
St
Ya
Tr
Fo
si
sh
kw
Br
ar
an
Sa
Se
G
in
at 6 months (P = 0.03) favouring the LC/HP group

Br
Figure 1 Percentage attrition rate in low-carbohydrate (white) and

(Fig. 5a). There was a slightly greater increase in the WMD
low-fat (black) diets reported in the literature. in HDL cholesterol at 12 months (0.06 mmol L-1) favour-
ing the LC/HP group (P < 0.05). There were no differences
found between the studies at 6 months (P = 0.46) or 12
Total cholesterol
months (P = 0.49).
The WMD in total cholesterol change was 0.19 mmol L-1
at 6 months (P < 0.0001) with the LC/HP group demon-
strating the increased cholesterol (Fig. 3a). This was also Triacylglycerol
the case at 12 months, although the difference between the
The WMD in triacylglycerol was -0.17 mmol L-1 at 6
groups was smaller and not significant (0.10 mmol L-1,
months (P = 0.0001) favouring the LC/HP group (Fig. 6a).
P = 0.31) (Fig. 3b). There were no differences among the
At 12 months the WMD between the groups was
studies at 6 (P = 0.84) and 12 (P = 0.14) months.
-0.19 mmol L-1 favouring the LC/HP group (P = 0.04).
Again, there was evidence of heterogeneity across the
Low-density lipoprotein cholesterol groups (P = 0.01).
The WMD in LDL cholesterol change was 0.14 mmol L-1
at 6 months (P < 0.00001) with the LC/HP group demon-
Systolic blood pressure
strating the increased LDL cholesterol (Fig. 4a). The dif-
ference between the groups was greater at 12 months The WMD drop in systolic blood pressure of -1.35 mmHg
(0.37 mmol L-1) (P < 0.00001) with the LC/HP group at 6 months favouring the LC/HP group was not significant
again demonstrating the increased LDL cholesterol (Fig. 7a). At 12 months the WMD between the groups was
© 2008 The Authors

Review: Systematic Review April 2008
Comparison: 01 Weight change at 6 months
© 2008 The Authors

(a) Outcome: 01 Weight change at 6 months
Study Treatment Control WMD (fixed) W eight W MD (fixed)

or sub category N Mean (SD) N Mean (SD) 95% CI % 95% CI
obesity reviews
Brehm 22 -8.50(1.00) 20 -3.90(1.00) 74.27 -4.60 [-5.21, -3.99]

Brinkworth 21 -8.10(8.20) 22 -8.50(6.10) 1.45 0.40 [-3.94, 4.74]
Dansinger 40 -3.20(4.90) 40 -3.50(5.60) 5.12 0.30 [-2.01, 2.61]
Due 23 -9.40(8.50) 23 -5.90(7.50) 1.27 -3.50 [-8.13, 1.13]
Foster 33 -6.90(6.50) 30 -3.10(5.60) 3.05 -3.80 [-6.79, -0.81]
Samaha 64 -5.80(8.60) 68 -1.90(4.20) 5.01 -3.90 [-6.23, -1.57]
Seshadri 43 -8.50(9.30) 35 -3.50(4.90) 2.63 -5.00 [-8.22, -1.78]
Truby 40 -6.00(6.40) 47 -6.60(5.40) 4.31 0.60 [-1.91, 3.11]
Yancy 59 -12.00(9.30) 60 -6.50(7.70) 2.89 -5.50 [-8.57, -2.43]
Total (95% CI) 345 345 100.00 -4.02 [-4.54, -3.49]

Test for heterogeneity: Chi² = 35.31, df = 8 (P < 0.0001), I² = 77.3%
Test for overall effect: Z = 15.08 (P < 0.00001)
-10 -5 0 5 10
Favours treatment Favours control

Comparison: 02 Weight change at 12 months
(b) Outcome: 01 Weight change at 12 months
Dansinger 40 -2.10(4.80) 40 -3.00(4.90) 24.02 0.90 [-1.23, 3.03]

Due 23 -6.20(7.60) 18 -4.30(7.10) 5.32 -1.90 [-6.42, 2.62]
Foster 33 -4.20(6.76) 30 -2.45(6.31) 10.42 -1.75 [-4.98, 1.48]
Gardner 77 -4.70(7.20) 79 -2.20(6.50) 23.38 -2.50 [-4.65, -0.35]
Stern 44 -5.10(8.70) 43 -3.10(8.40) 8.40 -2.00 [-5.59, 1.59]
Truby 28 -9.00(4.10) 33 -9.10(6.20) 16.00 0.10 [-2.50, 2.70]
Tsai 64 -5.10(8.70) 65 -3.10(8.40) 12.45 -2.00 [-4.95, 0.95]
Total (95% CI) 309 308 100.00 -1.05 [-2.09, -0.01]

Test for heterogeneity: Chi² = 6.71, df = 6 (P = 0.35), I² = 10.5%
Test for overall effect: Z = 1.98 (P = 0.05)
Review of low CHO vs. low fat diets
-10 -5 0 5 10
Control is low-fat/high-carbohydrate and treatment is low-carbohydrate/high-protein

Figure 2 Weight at 6 (a) and 12 (b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
M. Hession et al.
5
Comparison: 03 Total cholesterol change at 6 months
Outcome: 01 Total cholesterol change at 6 months
(a)
Brehm 22 -0.02(1.08) 20 -0.04(1.08) 1.84 0.02 [-0.63, 0.67]

Brinkworth 21 0.00(1.08) 22 -0.20(1.08) 1.88 0.20 [-0.45, 0.85]
Dansinger 40 -0.02(0.46) 40 -0.20(0.54) 16.25 0.18 [-0.04, 0.40]
6 Review of low CHO vs. low fat diets
Due 23 -0.31(1.08) 23 0.03(1.08) 2.02 -0.34 [-0.96, 0.28]

Foster 33 0.12(0.24) 30 -0.11(0.24) 55.76 0.23 [0.11, 0.35]
Samaha 64 0.05(1.08) 68 -0.02(1.08) 5.78 0.07 [-0.30, 0.44]
Seshadri 43 0.07(1.00) 35 -0.07(1.10) 3.53 0.14 [-0.33, 0.61]
Truby 40 -0.29(0.80) 47 -0.55(0.70) 7.73 0.26 [-0.06, 0.58]
Yancy 59 -0.21(1.08) 60 -0.35(1.08) 5.21 0.14 [-0.25, 0.53]
Total (95% CI) 345 345 100.00 0.19 [0.10, 0.28]

Test for heterogeneity: Chi² = 4.17, df = 8 (P = 0.84), I² = 0%
M. Hession et al.
-1 -0.5 0 0.5 1

Comparison: 04 Total cholesterol change at 12 months
(b) Outcome: 01 Total cholesterol change at 12 months

Dansinger 40 -0.11(0.59) 40 -0.21(0.62) 55.78 0.10 [-0.17, 0.37]

Due 23 0.10(1.08) 18 0.68(1.08) 8.84 -0.58 [-1.25, 0.09]
Foster 33 0.10(1.08) 30 -0.03(1.08) 13.76 0.13 [-0.40, 0.66]
Stern 44 0.16(1.11) 43 -0.21(0.91) 21.61 0.37 [-0.06, 0.80]
Total (95% CI) 140 131 100.00 0.10 [-0.10, 0.30]

-1 -0.5 0 0.5 1

Figure 3 Total cholesterol at 6(a) and 12 (b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
© 2008 The Authors

obesity reviews
© 2008 The Authors
Comparison: 05 LDL cholesterol change at 6 months
Outcome: 01 LDL cholesterol change at 6 months
(a)
obesity reviews

Brehm 22 -0.02(0.29) 20 -0.15(0.29) 11.48 0.13 [-0.05, 0.31]

Brinkworth 21 -0.20(0.29) 22 -0.30(0.29) 11.77 0.10 [-0.07, 0.27]
Dansinger 40 -0.06(0.36) 40 -0.18(0.62) 7.17 0.12 [-0.10, 0.34]
Foster 33 0.08(0.33) 30 -0.04(0.40) 10.67 0.12 [-0.06, 0.30]
Gardner 70 0.04(0.57) 63 -0.04(0.40) 12.83 0.08 [-0.09, 0.25]
Samaha 64 0.10(0.59) 68 0.07(0.46) 10.78 0.03 [-0.15, 0.21]
Seshadri 43 0.18(0.98) 35 0.15(0.64) 2.71 0.03 [-0.33, 0.39]
Yancy 59 0.04(0.29) 60 -0.19(0.29) 32.59 0.23 [0.13, 0.33]
Total (95% CI) 352 338 100.00 0.14 [0.08, 0.20]

-1 -0.5 0 0.5 1

Comparison: 06 LDL cholesterol change at 12 months
(b) Outcome: 01 LDL cholesterol change at 12 months
Foster 33 0.01(0.42) 30 -0.09(0.31) 22.49 0.10 [-0.08, 0.28]

Gardner 70 0.58(0.02) 63 0.01(0.44) 62.44 0.57 [0.46, 0.68]
Stern 44 0.18(0.91) 43 -0.10(0.75) 6.03 0.28 [-0.07, 0.63]
Tsai 64 -0.10(0.75) 65 0.18(0.90) 9.05 -0.28 [-0.57, 0.01]
Total (95% CI) 211 201 100.00 0.37 [0.28, 0.46]

-1 -0.5 0 0.5 1

Figure 4 LDL cholesterol at 6 (a) and 12 (b) months. LDL, low-density lipoprotein; SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
M. Hession et al.
7
Comparison: 07 HDL cholesterol change at 6 months
Outcome: 01 HDL cholesterol change at 6 months
(a)
Study Control Treatment WMD (fixed) W eight W MD (fixed)
Brehm 22 0.18(0.74) 20 0.10(0.74) 0.60 0.08 [-0.37, 0.53]

Brinkworth 21 0.00(0.74) 22 0.04(0.74) 0.62 -0.04 [-0.48, 0.40]
Dansinger 40 0.09(0.16) 40 0.06(0.23) 16.10 0.03 [-0.06, 0.12]
Due 23 -0.03(0.74) 23 0.23(0.74) 0.66 -0.26 [-0.69, 0.17]
8 Review of low CHO vs. low fat diets
Foster 33 0.17(0.53) 30 0.03(0.31) 2.70 0.14 [-0.07, 0.35]

Gardner 70 0.13(0.24) 63 0.05(0.17) 24.65 0.08 [0.01, 0.15]
Samaha 64 0.00(0.12) 68 -0.02(0.18) 45.04 0.02 [-0.03, 0.07]
Seshadri 43 -0.02(0.20) 35 -0.02(0.33) 7.82 0.00 [-0.12, 0.12]
Yancy 59 0.14(0.74) 66 -0.04(0.74) 1.80 0.18 [-0.08, 0.44]
Total (95% CI) 375 367 100.00 0.04 [0.00, 0.07]

M. Hession et al.
-0.5 -0.25 0 0.25 0.5

Favours control Favours treatment

Comparison: 08 HDL cholesterol change at 12 months
(b) Outcome: 01 HDL cholesterol change at 12 months
Study Control Treatment WMD (fixed) W eight W MD (fixed)

Dansinger 40 0.08(0.18) 40 0.08(0.25) 20.65 0.00 [-0.10, 0.10]

Due 23 0.12(0.74) 18 -0.09(0.74) 0.90 0.21 [-0.25, 0.67]
Foster 33 0.13(0.50) 30 0.02(0.28) 4.81 0.11 [-0.09, 0.31]
Gardner 70 0.12(0.23) 63 0.07(0.19) 36.86 0.05 [-0.02, 0.12]
Stern 44 -0.03(0.18) 43 -0.13(0.16) 36.78 0.10 [0.03, 0.17]
Total (95% CI) 210 194 100.00 0.06 [0.02, 0.11]

-0.5 -0.25 0 0.25 0.5

Favours control Favours treatment

Figure 5 HDL cholesterol 6 (a) and 12 (b) months. HDL, high-density lipoprotein; SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
© 2008 The Authors

obesity reviews
© 2008 The Authors

Comparison: 09 Triacylglycerol change at 6 months
Outcome: 01 Triacylglycerol change at 6 months
(a)
obesity reviews
Brehm 22 -3.93(0.96) 20 0.19(0.96) 2.06 -4.12 [-4.70, -3.54]

Brinkworth 21 -0.50(0.96) 22 -0.10(0.96) 2.12 -0.40 [-0.97, 0.17]
Dansinger 40 -0.11(0.45) 40 -0.11(0.62) 12.37 0.00 [-0.24, 0.24]
Due 23 -0.15(0.96) 23 0.11(0.96) 2.27 -0.26 [-0.81, 0.29]
Foster 33 2.20(3.30) 30 -1.00(2.10) 0.38 3.20 [1.85, 4.55]
Gardner 70 -0.40(0.72) 63 -0.18(0.56) 14.66 -0.22 [-0.44, 0.00]
Samaha 64 -4.20(9.00) 68 -0.70(6.00) 0.10 -3.50 [-6.13, -0.87]
Seshadri 43 -0.01(0.23) 35 0.00(0.25) 60.19 -0.01 [-0.12, 0.10]
Yancy 59 -0.84(0.96) 60 -0.31(0.96) 5.86 -0.53 [-0.87, -0.19]
Total (95% CI) 375 361 100.00 -0.16 [-0.24, -0.08]

-1 -0.5 0 0.5 1

Comparison: 10 Triacylglycerol change at 12 months
(b) Outcome: 01 Triacylglycerol change at 12 months
Dansinger 40 -0.01(0.94) 40 -0.14(0.68) 23.51 0.13 [-0.23, 0.49]

Due 23 -0.05(0.96) 18 0.33(0.96) 8.67 -0.38 [-0.97, 0.21]
Foster 33 -2.50(2.50) 30 -0.09(4.20) 1.02 -2.41 [-4.14, -0.68]
Gardner 70 -0.33(0.66) 63 -0.16(0.68) 58.34 -0.17 [-0.40, 0.06]
Stern 44 -0.65(1.78) 43 0.05(0.96) 8.47 -0.70 [-1.30, -0.10]
Total (95% CI) 210 194 100.00 -0.19 [-0.36, -0.01]

-1 -0.5 0 0.5 1

Figure 6 Triacylglycerols at 6 (a) and 12 (b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
M. Hession et al.
9
10

Comparison: 11 Systolic blood pressure change at 6 months
Outcome: 01 Systolic blood pressure change at 6 months
(a)
Brehm 22 4.00(12.70) 20 2.00(12.70) 6.12 2.00 [-5.69, 9.69]

Brinkworth 21 -4.90(12.70) 22 -3.60(12.70) 6.28 -1.30 [-8.89, 6.29]
Dansinger 40 -0.70(10.00) 40 -4.80(14.00) 12.73 4.10 [-1.23, 9.43]
Foster 33 -2.70(11.70) 30 1.20(12.20) 10.34 -3.90 [-9.82, 2.02]

Gardner 77 -6.40(9.50) 79 -4.30(7.60) 49.51 -2.10 [-4.80, 0.60]
Truby 40 -7.20(11.60) 47 -4.10(11.70) 15.01 -3.10 [-8.01, 1.81]
Total (95% CI) 233 238 100.00 -1.35 [-3.25, 0.56]

M. Hession et al.
-10 -5 0 5 10

Comparison: 12 Systolic blood pressure change at 12 months
(b) Outcome: 01 Systolic blood pressure change at 12 months

Dansinger 40 0.20(12.00) 40 -2.70(13.00) 15.51 2.90 [-2.58, 8.38]

Foster 33 -0.21(9.40) 30 2.00(11.80) 16.59 -2.21 [-7.51, 3.09]
Gardner 77 -7.60(11.00) 79 -3.10(9.30) 45.53 -4.50 [-7.70, -1.30]
Stern 44 1.00(19.00) 43 2.00(15.00) 9.03 -1.00 [-8.18, 6.18]
Tsai 64 1.00(19.00) 65 2.00(15.00) 13.34 -1.00 [-6.91, 4.91]
Total (95% CI) 258 257 100.00 -2.19 [-4.35, -0.03]

-10 -5 0 5 10

Figure 7 Systolic blood pressure at 6 (a) and 12 (b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
© 2008 The Authors

obesity reviews
a decrease of 2.19 mmHg favouring the LC/HP group weight loss for both groups at either 6 or 12 months, but
(P = 0.05) (Fig. 7b). There was no difference between the this review included studies with dietary approaches that
studies at either time. are not considered LC, which may have affected their
outcomes.
The present review showed that there was a significant
Diastolic blood pressure
improvement in HDL cholesterol and triacylglycerols at 6
The WMD decrease in diastolic blood pressure of and 12 months favouring the LC/HP group, but this was
0.49 mmHg at 6 months favouring the LC/HP group was not significant at 17 months. The lack of significance at 17
not significant (Fig. 8a). At 12 months, the WMD between months may be caused by the reintroduction of carbohy-
the two groups of 0.81 mmHg lowering favouring the drates in the LC/HP group. There was heterogeneity
LC/HP group was greater, but was also not significant between the studies for triacylglycerols, but this may have
(Fig. 8b). There was no evidence of statistical heterogeneity been due to differences in study design.
across the studies at either time. Low HDL cholesterol and raised triacylglycerol levels are
risk factors for cardiovascular disease and impact on the
atherogenicity of the LDL particle and these results indicate
Fasting plasma glucose
that a LC/HP diet may be a better approach to weight loss
The WMD between the groups in fasting plasma glucose and lowering the risk of cardiovascular disease. These
was not significant and there was no evidence of statistical results are consistent with the review carried out by
heterogeneity at either time (Fig. 9). Nordmann et al. (24). However, Bravata et al. (26) did not
show any significant improvement in these parameters,
which again may have been affected by their choice of
Discussion
studies.
The results of the present review show that weight loss was The present review showed a significant improvement in
significantly greater in the LC/HP (treatment) group after total cholesterol and LDL cholesterol favouring the LF/HC
6 and 12 months compared with the LF/HC group. The group at 6 months, at which point total cholesterol and
difference was greater at 6 months and at that time there LDL cholesterol increased more in the LC/HP group but
was significant heterogeneity among the studies, probably not at 12 months or 17 months. Nordmann et al. (24) in
due to the different study designs, but at 12 months the a meta-analysis of LC vs. LF diets found reports on four
heterogeneity was no longer significant. The 36-month groups of patients demonstrating an improvement in total
follow-up by Cardillo et al. (12) reported that mean weight and LDL cholesterol favouring LF diets rather than LC
change between baseline and 36 months was not different diets. This finding is consistent with the studies included in
between the LC/HP and the LF/HC group. However, they the present review. An elevated total cholesterol could in
do report that between 6 and 36 months weight was part be explained by an increase in HDL cholesterol
unchanged for the LF/HC group but that subjects on the observed in the LC/HP group. Also, although an elevated
LC/HP approach regained weight, but this change was not LDL cholesterol increases the risk of acute cardiovascular
significant. events, we have just shown evidence that LC/HP diets
Avenell et al. (23) examined the effects of a protein increase HDL and decrease triacylglycerol which impacts
sparing modified fast (PSMF) compared with a low-calorie on the atherogenicity of the LDL particle. These studies
diet and a very low-calorie diet. A PSMF is a LC diet, which failed to investigate changes in LDL particle size. Further-
allows a maximum of 40 g of carbohydrate per day. The more, evidence from Sharman et al. (27) suggests that on a
review examined weight loss comparing the PSMF with LC/HP LDL particle sizes change from small to large and
low-calorie diets after 12, 18, 24, 36 and 60 months. There therefore resulting in a less atherogenic profile.
was a greater weight loss favouring the PSMF group com- There was a trend towards improvement in diastolic and
pared with the control after 12, 24 and 36 months, but only systolic blood pressure at 6, 12 and 17 months favouring
seven RCTs were included in this analysis, which included the LC/HP group. The difference was significant at 12
a total of 480 participants (23). These results are consistent months favouring the LC/HP group for systolic blood
with the results of the present systematic review. pressure. Bravata et al. (26) reported no change in systolic
A review by Nordmann et al. (24) comparing LC diets blood pressure after the low- and very-low-carbohydrate
with LF diets showed significant weight loss with the LC diets (26). Nordmann et al. (24) showed no significant
group at 6 months, but not at 12 months. The meta- difference in blood pressure at any time point.
regression by Krieger et al. (25) also reports a greater At 6 months there was a trend towards improvement in
weight loss in addition to a greater body fat and percentage fasting plasma glucose only slightly favouring the LF/HC
body fat loss in studies lasting more than 3 months. Bravata group in which there was a greater decrease in fasting
et al. (26), however, showed no significant differences in plasma glucose in the LF/HC group. This was surprising
© 2008 The Authors

12

Comparison: 13 Diastolic blood pressure change at 6 months
Outcome: 01 Diastolic blood pressure change at 6 months
(a)
Brehm 22 5.00(8.30) 20 1.00(8.30) 7.24 4.00 [-1.03, 9.03]

Brinkworth 21 -1.70(8.30) 22 -1.60(8.30) 7.43 -0.10 [-5.06, 4.86]
40 -4.00(6.50) 40 -1.80(6.90) 21.19 -2.20 [-5.14, 0.74]
Dansinger
Foster 33 2.00(12.70) 30 -1.10(14.20) 4.10 3.10 [-3.58, 9.78]
Gardner 77 -3.30(6.90) 79 -2.50(5.80) 45.60 -0.80 [-2.80, 1.20]
Truby 40 -4.90(8.30) 47 -4.40(8.60) 14.45 -0.50 [-4.06, 3.06]
Total (95% CI) 233 238 100.00 -0.49 [-1.85, 0.86]

M. Hession et al.
-10 -5 0 5 10

Comparison: 14 Diastolic blood pressure change at 12 months
(b) Outcome: 01 Diastolic blood pressure change at 12 months

Dansinger 40 -1.40(7.50) 40 -1.70(6.40) 29.78 0.30 [-2.76, 3.36]

Foster 33 -2.70(12.40) 30 -2.90(6.70) 11.76 0.20 [-4.66, 5.06]
Gardner 77 -4.40(8.40) 79 -2.20(6.70) 48.74 -2.20 [-4.59, 0.19]
Stern 44 3.00(15.00) 43 1.00(10.00) 9.73 2.00 [-3.35, 7.35]
Total (95% CI) 194 192 100.00 -0.76 [-2.43, 0.90]

-10 -5 0 5 10

Figure 8 Diastolic blood pressure at 6 (a) and 12 (b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
© 2008 The Authors

obesity reviews
© 2008 The Authors
Comparison: 15 Fasting plasma glucose change at 6 months
Outcome: 01 Fasting plasma glucose change at 6 months
obesity reviews
(a)
Brehm 22 0.50(1.35) 20 -0.19(1.35) 3.00 0.69 [-0.13, 1.51]

Brinkworth 21 0.10(1.35) 22 -0.10(1.35) 3.08 0.20 [-0.61, 1.01]
Dansinger 40 -0.43(1.40) 40 -0.21(1.20) 6.14 -0.22 [-0.79, 0.35]
Due 23 0.00(1.35) 18 0.00(1.35) 2.89 0.00 [-0.83, 0.83]
Gardner 70 0.01(0.42) 63 -0.04(0.54) 73.06 0.05 [-0.12, 0.22]
Samaha 64 -0.61(1.30) 68 -0.11(1.10) 11.82 -0.50 [-0.91, -0.09]
Total (95% CI) 240 231 100.00 -0.01 [-0.15, 0.13]

-1 -0.5 0 0.5 1

Comparison: 16 Fasting plasma glucose change at 12 months
(b) Outcome: 01 Fasting plasma glucose change at 12 months
Dansinger 40 0.07(1.66) 40 -0.26(1.05) 6.80 0.33 [-0.28, 0.94]

Due 23 0.10(1.35) 18 0.30(1.35) 3.63 -0.20 [-1.03, 0.63]
Gardner 70 -0.09(0.74) 63 0.02(0.51) 54.87 -0.11 [-0.32, 0.10]
Stern 44 0.17(0.61) 43 0.17(0.67) 34.70 0.00 [-0.27, 0.27]
Total (95% CI) 177 164 100.00 -0.05 [-0.20, 0.11]

-1 -0.5 0 0.5 1

Figure 9 Fasting plasma glucose at 6(a) and 12(b) months. SD, standard deviation; WMD, weighted mean difference; CI, confidence interval; df, degree of freedom.
M. Hession et al.
13
when compared with the review by Layman et al. where including weight, cholesterol, blood pressure and glycemic
there is clear evidence of improvements in fasting glucose, control were evaluated, as these are important in weight
postprandial glucose and insulin responses and glycosy- loss and cardiovascular disease risk.
lated haemoglobin (HbA1c) for individuals on an LC/HP Evidence from this systematic review demonstrates that
diet (6). At 12 months, the opposite occurred in which LC/HP diets are more effective at 6 months and are as
there was a greater decrease in fasting plasma glucose, effective, if not more, as LF diets in reducing weight and
favouring the LC/HP group. The difference was not signifi- cardiovascular disease risk up to 1 year. As there were only
cant at 6, 12 and 17 months. Bravata et al. (26) reported 13 studies included and several of them allowed the
no change in fasting serum glucose among recipients of reintroduction of carbohydrates in the LC/HP diet, the
the low- and very-low-carbohydrate diets. Nordmann et al. evidence of the long-term efficacy of these diets is not
(24) showed a greater improvement in fasting plasma complete. Certainly at 6 months, the evidence is in favour
glucose favouring the LC group at 6 months, but this was of the use of LC/HP diet. It may not be appropriate to
no longer significant at 12 months. return to a HC intake for weight maintenance (29,30). A
Furthermore, fasting glucose provides a limited assess- gradual reintroduction while still limiting the intake of
ment of overall glycaemic status; therefore, future studies carbohydrate may be more appropriate.
should use HbA1c values or more direct measurements of With the prevalence of obesity increasing there is a need
insulin sensitivity. for larger and long-term RCTs of low- or very-low-
There was a higher attrition rate in the LF/HC compared carbohydrate diets compared with the LF/HC diets to be
with the LC/HP groups (Fig. 1). Reasons for attrition carried out. The influence of behavioural therapy and exer-
included difficulty in complying with the diet or disliking cise interventions needs to be evaluated, as well as lifestyle,
the diet, difficulty in maintaining the scheduled visits and appetite and mood questionnaires.
significant events such as pregnancy and surgery. It is not known with certainty which aspect of LC diets
causes the weight loss and cardiovascular disease risk
factor changes. Whether it is the LC, the HP or calorie
Limitations
restriction needs to be examined. In addition, there is a
It is important to take account of attrition rates in the need to assess if the greater weight loss achieved at 6
interpretation of outcomes as high attrition rates lead to a months on a LC/HP diet results in more important long-
smaller statistical power. An intention to treat approach is term improvements of cardiovascular disease.
commonly used to overcome attrition rates and possible There is a need for trials to include a follow-up period,
bias in the outcomes. There are, however, limitations when to examine adherence to the LC diets and whether partici-
using this approach in lifestyle trials as the intention to pants maintain their weight loss and CVD risk factor
treat approach has been derived from drug trials and may change when there is minimum contact with the study
not yield robust outcomes. This results in the need for investigators. Finally, taking account of high attrition rates
higher retention rates to assess for real changes in response when using RCTs for dietary and lifestyle interventions,
to the dietary interventions. perhaps we will witness a move towards a continuous im-
In addition, the use of a RCT design in dietary interven- provement methodology in the future.
tions may not be appropriate. In general, any weight-loss
strategy has a maximum weight loss at 6 months followed
by a return to initial weight. It is clear that patients are Conflict of Interest Statement
changing their treatment by their own accord, perhaps No conflict of interest was declared.
subconsciously or perhaps due to a metabolic response of
the body aiming to return to its initial weight. The current
thinking within the field of obesity suggests that the use Acknowledgement
of continuous improvement methodology may be more
MH was supported by a commercial grant from
appropriate for weight-loss management (28).
LighterLife.
Also there was some evidence of heterogeneity between
the studies included in this analysis. This calls for the use of
more consistent and robust study designs for which we References
have to establish a clear definition of a LC/HP diet.
1. Caballero B. The global epidemic of obesity: an overview.
Epidemiol Rev 2007; 29: 1–5.
Conclusion 2. Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA,
Brody J, Hayflick L, Butler RN, Allison DB, Ludwig DS. A poten-
This systematic review included all known RCTs of LC tial decline in life expectancy in the U.S. in the 21st century. N Engl
diets vs. the LF/HC diet from 2000 to 2007. Factors J Med 2005; 352: 1138–1145.
© 2008 The Authors

3. Ware JH. Interpreting incomplete data in studies of diet and 18. Seshadri P, Iqbal N, Stern L, Williams M, Chicano KL, Daily
weight loss. N Engl J Med 2003; 348: 21. DA, McGrory J, Gracely EJ, Rader EJ, Samaha FF. A randomized
4. Wilding J. Science medicine and the future. BMJ 1997; 315: study comparing the effects of a low CHO diet and a conventional
997–1000. diet on lipoprotein subfractions and C-reactive protein levels in
5. Veech LR. The therapeutic implications of ketene bodies in patients with severe obesity. Am J Med 2004; 117: 398–405.
pathological conditions: ketosis, ketogenic diet, redox states, 19. Stern L, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory
insulin resistance, and mitochondrial metabolism. Prostaglandins J, Williams M, Gracely EJ, Samaha FF. The effects of a low car-
Leukot Essent Fatty Acids 2004; 70: 309–319. bohydrate diet versus conventional weight loss diets in severely
6. Layman DK, Clifton P, Gannon MC, Krauss RM, Nuttall FQ. obese adults: a 1-year follow-up of a randomized trial. Ann Intern
Protein in optimal health: heart disease and type 2 diabetes. Am J Med 2004; 140: 778–785.
Clin Nutr 2008; 87: 1571S–1575S. 20. Truby H, Baic S, deLooy A, Fox K, Livingstone MBE, Logan
7. Volek JS, Fernandez ML, Feinman RD, Phinney SD. Dietary LM, MacDonald IA, Morgan LM, Taylor MA, Millward DJ.
carbohydrate restriction induces a unique metabolic state posi- Randomized controlled trial of four commercial weight loss pro-
tively affecting atherogenic dyslipidemia, fatty acid partitioning, grammes in the UK: initial findings from the BBC diet trials. BMJ
and metabolic syndrome. Prog Lipid Res 2008; 47: 307–318. 2006; 332: 1309–1314.
8. Avenell A, Broom J, Brown TJ, Poobalan A, Aucott L, Stearns 21. Tsai AG, Glick HA, Shera D, Stern L, Samaha FF. Cost-
SC, Smith WCS, Jung RT, Campbell LK, Grant AM. Systematic effectiveness of a low-carbohydrate diet and a standard diet in
review of the long term effects and economic consequences of severe obesity. Obes Res 2005; 13: 1834–1840.
treatments for obesity and implications for health improvement. 22. Yancy SW, Olsen MK, Guyton JR, Bakst RP, Westman EC. A
HTA 2004; 8: 1–458. low carbohydrate ketogenic diet versus a low fat diet to treat
9. Caterson I, Broom J. 2. Definition, assessment and epidemiol- obesity and hyperlipidemia. Ann Intern Med 2004; 140: 769–777.
ogy. Obesity. Mosby International Ltd.: London, 2001, pp. 10–19. 23. Avenell A, Brown TJ, McGee MA, Campbell MK, Grant MK,
10. Brehm BJ, Seeley R, Daniels SR, D’Alessio DA. A randomized Broom J, Jung RT, Smith WCS. What are the long term benefits of
control trial comparing a very low carbohydrate diet and a calorie weight reducing diets in adults? A systematic review of randomized
restricted low fat diet on body weight and cardiovascular risk control trials. J Hum Nutr Diet 2004; 17: 317–335.
factors in healthy women. J Clin Endocrinol Metab 2002; 88: 24. Nordmann AJ, Nordmann A, Briel M, Keller U, Yancy WS,
1617–1623. Brehm BJ, Bucher HC. Effects of low-carbohydrate vs. low-fat
11. Brinkworth GD, Noakes M, Keogh JB, Luscombe ND, Wittert diets on weight loss and cardiovascular risk factors. A meta-
GA, Clifton PM. Long term effects of a high protein, low carbo- analysis of randomized controlled trials. Arch Intern Med 2006;
hydrate diet in weight control and cardiovascular risk factors in 166: 285–293.
obese, hyperinsulinemic subjects. Int J Obes 2004; 28: 661–670. 25. Krieger JW, Sitren HS, Daniels MJ, Langkamp-Henken B.
12. Cardillo S, Seshadri P, Iqbal N. The effects of a low- Effects of variation in protein and carbohydrate intake on body
carbohydrate versus low-fat diet on adipocytokines in severely mass and composition during energy restriction: a meta-regression.
obese adults: 3-year follow-up of a randomized trial. Eur Rev Med Am J Clin Nutr 2006; 83: 260–274.
Pharmacol Sci 2006; 10: 99–106. 26. Bravata DM, Sanders L, Huang J, Krumholz HM, Olkin I,
13. Dansinger ML, Gleason JA, Griffith JL, Selker HP, Schaefer Gardner D. Efficacy and safety of low carbohydrate diets – a
EJ. Comparison of the Atkins, Ornish, WeightWatchers, and Zone systematic review. JAMA 2003; 289: 1837–1850.
Diets for weight loss and heart disease risk reduction. JAMA 2005; 27. Sharman MJ, Kraemar WJ, Love DM, Avery NG, Gomez AL,
293: 43–53. Scheett TP, Volek JS. A ketogenic diet favourably affects serum
14. Due A, Toubro S, Skov AR, Astrup A. Effects of normal fat biomarkers for cardiovascular disease in normal weight men.
diets, either medium or high in protein, on body weight in over- J Nutr 2002; 132: 1879–1885.
weight subjects: a randomized control 1-year trial. Int J Obes 28. McQuigg M, Broom J, Laws RA, Reckless JPD, Noble PA,
2004; 28: 1283–1290. Kumar S, McCombie EL, Lean MEJ, Lyons GF, Frost GS, Quinn
15. Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, MF, Barth JH, Haynes SM, Finer N, Haslam DW, Ross HM.
Mohammad BS, Szapary PO, Rader DJ, Edman JS, Klein S. A Tricks and tools for the primary care provider. The Counterweight
randomized control trial of a low carbohydrate diet for obesity. N Programme: a continuous improvement methodology model of
Engl J Med 2003; 348: 2082–2090. weight management in UK primary care. Int J Obesity 2008; 32:
16. Gardner CD, Kiazand A, Alhassan S, Kim S, Stafford RS, S6.
Balise RR. Comparison of the Atkins, Zone, Ornish, and LEARN 29. Mazlan N. Effects of fat and carbohydrate on energy intake
diets for change in weight and related risk factors among over- and macronutrient selection in humans. 2001. PhD thesis,
weight premenopausal women. JAMA 2007; 297: 969–977. Aberdeen University, Aberdeen, Scotland.
17. Samaha FF, Iqbal N, Seshadri P, Chicano KL, Daily DA, 30. Stubbs RJ, Johnstome AM, Harbron CG, Reid C. Covert
McGrory J, Williams T, Williams M, Gracely EJ, Stern L. A low manipulation of energy density of high carbohydrate diets in
carbohydrate diet as compared with a low fat diet in severe ‘pseudo free-living’ humans. Int J Obes Relat Metb Disord 1998;
obesity. N Engl J Med 2003; 348: 2074–2081. 22: 885–892.
© 2008 The Authors


Review of Low-Carb Vs High-Carb Diets

Uploaded by

Copyright:

Available Formats

Review of Low-Carb Vs High-Carb Diets

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Review of Low-Carb Vs High-Carb Diets

Uploaded by

Copyright:

Available Formats

What was the purpose of this systematic review?

What was the purpose of this systematic review?

What diets were compared in the studies included in the review?

What diets were compared in the studies included in the review?

obesity reviews doi: 10.1111/j.1467-789X.2008.00518.

Systematic review of randomized controlled trials of

M. Hession1, C. Rolland1, U. Kulkarni1, A. Wise2 and J. Broom1

Keywords: Cardiovascular risk, low-carbohydrate, meta-analysis, obesity.

obesity reviews (2008)

Author contributions: Ms Hession, Dr Rolland, Dr Kulkarni, Dr Wise and

© 2008 The Authors 1

© 2008 The Authors

© 2008 The Authors

Table 2 Summary of diets and CHO content

60 (Fig. 4b). There were no differences among the studies at

High-density lipoprotein cholesterol

The WMD in HDL cholesterol change was 0.04 mmol L-1

at 6 months (P = 0.03) favouring the LC/HP group

Figure 1 Percentage attrition rate in low-carbohydrate (white) and

© 2008 The Authors

© 2008 The Authors

Study Treatment Control WMD (fixed) W eight W MD (fixed)

Brehm 22 -8.50(1.00) 20 -3.90(1.00) 74.27 -4.60 [-5.21, -3.99]

Total (95% CI) 345 345 100.00 -4.02 [-4.54, -3.49]

Review: Systematic Review April 2008

Dansinger 40 -2.10(4.80) 40 -3.00(4.90) 24.02 0.90 [-1.23, 3.03]

Total (95% CI) 309 308 100.00 -1.05 [-2.09, -0.01]

Control is low-fat/high-carbohydrate and treatment is low-carbohydrate/high-protein

Brehm 22 -0.02(1.08) 20 -0.04(1.08) 1.84 0.02 [-0.63, 0.67]

Due 23 -0.31(1.08) 23 0.03(1.08) 2.02 -0.34 [-0.96, 0.28]

Total (95% CI) 345 345 100.00 0.19 [0.10, 0.28]

Test for overall effect: Z = 4.23 (P < 0.0001)

Review: Systematic Review April 2008

Study Treatment Control WMD (fixed) W eight W MD (fixed)

Dansinger 40 -0.11(0.59) 40 -0.21(0.62) 55.78 0.10 [-0.17, 0.37]

Total (95% CI) 140 131 100.00 0.10 [-0.10, 0.30]

Control is low-fat/high-carbohydrate and treatment is low-carbohydrate/high-protein

© 2008 The Authors

Study Treatment Control WMD (fixed) W eight W MD (fixed)

Brehm 22 -0.02(0.29) 20 -0.15(0.29) 11.48 0.13 [-0.05, 0.31]

Total (95% CI) 352 338 100.00 0.14 [0.08, 0.20]

Review: Systematic Review April 2008

Foster 33 0.01(0.42) 30 -0.09(0.31) 22.49 0.10 [-0.08, 0.28]

Total (95% CI) 211 201 100.00 0.37 [0.28, 0.46]

Favours treatment Favours control

Control is low-fat/high-carbohydrate and treatment is low-carbohydrate/high-protein

Brehm 22 0.18(0.74) 20 0.10(0.74) 0.60 0.08 [-0.37, 0.53]

Foster 33 0.17(0.53) 30 0.03(0.31) 2.70 0.14 [-0.07, 0.35]

Total (95% CI) 375 367 100.00 0.04 [0.00, 0.07]

Test for overall effect: Z = 2.20 (P = 0.03)

-0.5 -0.25 0 0.25 0.5

Review: Systematic Review April 2008

Study Control Treatment WMD (fixed) W eight W MD (fixed)

Dansinger 40 0.08(0.18) 40 0.08(0.25) 20.65 0.00 [-0.10, 0.10]

Total (95% CI) 210 194 100.00 0.06 [0.02, 0.11]

-0.5 -0.25 0 0.25 0.5

Control is low-fat/high-carbohydrate and treatment is low-carbohydrate/high-protein

© 2008 The Authors

© 2008 The Authors