An Evaluation of the Atkins' Diet

Abstract

Low-carbohydrate (LC) weight-reducing diets are popular choices for self-dieters. Eighteen adults (BMI >/= 25 kg/m(2)) were enrolled in this short-term longitudinal study to evaluate dietary intake and weight on their "usual" diets and LC diet. Subjects were instructed to follow the first two phases of the diet described in Dr. Atkins' New Diet Revolution (2 weeks each). Total daily intake of calories and nutrients were calculated from 3-day food diaries. Body weight was measured at the end of each 2-week diet session. All enrolled subjects completed the study (age = 39.8 +/- 8.1 years, BMI = 36.6 +/- 6.6 kg/m(2)). Mean caloric intakes were 1400 +/- 472 kcal/day (Induction diet) and 1558 +/- 490 kcal/day (Ongoing Weight Loss diet) both p </= 0.001 compared to "usual" (Baseline diet) 2481 +/- 723 kcal/day. Body weights were 107.4 +/- 24.2 kg, 103.6 +/- 23.0 kg and 102.1 +/- 22.6 kg at the conclusion of the Baseline, Induction, and Ongoing Weight Loss diets, respectively (both p </= 0.001). Decreases in daily caloric intake correlated with weight loss. Pearson correlation coefficients were, r = 0.64 (p </= 0.001) and r = 0.40 (p </= 0.001) for Induction and Ongoing Weight Loss diets versus "usual" diet. Caloric intake is decreased when otherwise healthy overweight and obese adults self-implement Atkins' Induction and Ongoing Weight Loss diets and significantly altered their dietary micronutrient intake. Weight loss can be explained by the self-selected lower caloric intake on The Atkins' Diet.

Full text

An Evaluation of the Atkins’ Diet
BERNARD V. MILLER III, M.D.,
1
JOSEPH S. BERTINO Jr., Pharm.D.,
1,2,3,4
ROBERTA G. REED, Ph.D.,
4
CHRISTINE M. BURRINGTON, B.S.,
4
LESLIE K. DAVIDSON, B.A.,
4
ALLAN GREEN, Ph.D.,
4
ANNE M. GARTUNG, R.N., M.S., F.N.P.,
2,4
and ANNE N. NAFZIGER, M.D., M.H.S.
1,2,4
ABSTRACT
Low-carbohydrate (LC) weight-reducing diets are popular choices for self-dieters. Eighteen
adults (BMI  25 kg/m
2
) were enrolled in this short-term longitudinal study to evaluate di-
etary intake and weight on their “usual” diets and LC diet. Subjects were instructed to fol-
low the first two phases of the diet described in Dr. Atkins’ New Diet Revolution (2 weeks
each). Total daily intake of calories and nutrients were calculated from 3-day food diaries.
Body weight was measured at the end of each 2-week diet session. All enrolled subjects
completed the study (age = 39.8 ± 8.1 years, BMI = 36.6 ± 6.6 kg/m
2
). Mean caloric intakes
were 1400 ± 472 kcal/day (Induction diet) and 1558 ± 490 kcal/day (Ongoing Weight Loss
diet) both p  0.001 compared to “usual” (Baseline diet) 2481 ± 723 kcal/day. Body weights
were 107.4 ± 24.2 kg, 103.6 ± 23.0 kg and 102.1 ± 22.6 kg at the conclusion of the Baseline, In-
duction, and Ongoing Weight Loss diets, respectively (both p  0.001). Decreases in daily
caloric intake correlated with weight loss. Pearson correlation coefficients were, r = 0.64
(p  0.001) and r = 0.40 (p  0.001) for Induction and Ongoing Weight Loss diets versus
“usual” diet. Caloric intake is decreased when otherwise healthy overweight and obese
adults self-implement Atkins’ Induction and Ongoing Weight Loss diets and significantly
altered their dietary micronutrient intake. Weight loss can be explained by the self-selected
lower caloric intake on The Atkins’ Diet.
299
METABOLIC SYNDROME AND RELATED DISORDERS
Volume 1, Number 4, 2003
© Mary Ann Liebert, Inc.
The
1
Department of Internal Medicine,
2
Clinical Pharmacology Research Center,
3
Department of Pharmacy Ser-
vices, and
4
The Bassett Research Institute, Bassett Healthcare, Cooperstown, New York.
INTRODUCTION
T
HE ATKINS’ DIET and other low carbohydrate
(LC) diets are currently popular in Ameri-
can culture
1
as the prevalence of obesity contin-
ues to increase making it a true epidemic and
public health crisis.
2
The effectiveness of LC diet
has been a topic of debate for over 40 years.
3,4
Cardiovascular risk continues to be an impor-
tant concern due to the restriction of multiple
foods that are low fat and high fiber and inclu-
sion of high fat, high cholesterol foods.
5,6
LC diet studies in the obese have indicated
more favorable metabolic benefit with regard
to serum lipoproteins and insulin sensitivity
than hypocaloric diets without carbohydrate
13644C08.PGS 4/15/04 2:11 PM Page 299

restriction.
7–10
Other evidence refutes a “meta-
bolic advantage” and suggests that the LC diet
temporarily shifts water and electrolyte bal-
ance,
11–15
potentially affecting both lean body
mass and fat mass blood flow.
The Atkins’ Diet proposes weight reduction
without conscious caloric restriction. The LC
diet has been shown in randomized controlled
trials (RCT) to produce greater weight loss
than conventional weight reducing diet with
minimal professional contact.
7
It has been
shown in other RCTs with dietary counseling
that caloric intake is restricted following a LC
diet.
8,9
No studies have shown that caloric in-
take is decreased with minimal professional
assistance (i.e., handing a subject the book).
The aim of this short-term study was to evalu-
ate the dietary intake and body weight of over-
weight or obese adults while on their “usual”
diet compared with the first two stages of The
Atkins’ Diet.
MATERIALS AND METHODS
This study consisted of three consecutive
diets lasting 2 weeks each. The “usual” or
Baseline diet had no carbohydrate restriction
and participants were asked to follow their
usual eating patterns for 2 weeks after screen-
ing and return with completed 3-day diet
records. Subjects were then supplied with a
copy of Dr. Atkins’ New Diet Revolution
18
and
asked to read the book and follow the diets as
described by Dr. Atkins.
All subjects provided written informed
consent prior to study participation. Subjects
 18 years of age and body mass index (BMI)
 25 kg/m
2
were recruited from the local area
for study. Medical history, physical exami-
nation and laboratory screening were per-
formed to assure that participants were
healthy and without history of cardiovascular
disease. Screening sessions also included in-
struction for completion of 3-day diet records
if subjects met inclusion and exclusion crite-
ria. Additional exclusion criteria included the
chronic use of any medication and any his-
tory of clinically significant chronic illness
including coronary artery disease, hyperten-
sion, diabetes mellitus, hyperlipidemia, gout,
nephrolithiasis, renal impairment, elevated
liver function tests, and thyroid disorders. The
Institutional Review Board of the Mary Imo-
gene Bassett Hospital approved this study.
The Atkins’ Diet instructs individuals to
begin their weight loss diet with a 2-week pe-
riod of severely reduced intake of dietary car-
bohydrates (<20 g/day) to make subjects
ketotic (Induction diet). Protein consumption
is encouraged, but fat is allowed ad libitum.
Next, individuals are instructed to add 5 g of
carbohydrate/day to the diet while continuing
to lose weight, the Ongoing Weight Loss (OWL)
diet (25 g/day for the first week and 30 g/day
during the second week of this study).
When weight loss slows, the progression stops
and the current carbohydrate intake is main-
tained until the goal weight is achieved. The Pre-
maintenance and Maintenance diets begin when
goal weight is achieved. Carbohydrate is added
to the diet until weight gain begins. In the cur-
rent study this was not investigated.
Participants received two multivitamins
daily for the duration of the study and were
encouraged to maintain sufficient daily so-
dium intake (4–5 g daily) to avoid common
side effects.
19
This was the only instruction
that subjects were given regarding diet from
study investigators. Participants were also in-
structed to maintain their current level of
physical activity throughout the study and to
call a study investigator if they experienced
symptoms for which they would call their pri-
mary care provider. Weight, height, blood
pressure and pulse were measured at the end
of each phase. BMI was calculated.
Urine ketones were assessed with Ketostix
®
(Elkhart, IN) and recorded daily by the study
subjects. Subjects were instructed to contact a
study investigator if ketosis (5–160 mg/dL
urine ketones) did not occur within 48–72 h of
starting the Induction diet or if they did not
maintain ketosis during the diet phases.
Self-reported 3-day diet records were com-
pleted at the end of each study phase. Each
diet record included 2 week days and 1 week-
end day. Diet records were analyzed using the
Minnesota Nutrition Data System for Re-
search
®20
(software version v4.01, developed
by the Nutrition Coordinating Center, [NCC],
University of Minnesota, Minneapolis, MN,
300 MILLER ET AL.
13644C08.PGS 4/15/04 2:11 PM Page 300

Food and Nutrient Database 29; released De-
cember 1998). If an analytic value was not
available for a nutrient in a food, the NCC cal-
culated the value based on the nutrient content
of other nutrients in the same food or on a
product ingredient list, or estimated the value
based on the nutrient content of similar food.
A missing value was allowed only if (1) the
value was believed to be negligible, (2) the
food is usually eaten in very small amounts,
(3) it is unknown if the nutrient exists in the
food at all, or (4) there is no way to estimate
the value because the food is unlike any other.
Diet records were analyzed for total daily
macronutrients, micronutrients and caloric in-
take. Nutrients from multivitamins and drink-
ing water were not included in these analyses.
At the end of each diet phase, a modified
Corso adverse event questionnaire was com-
pleted by each subject.
21
Laboratory
All serum samples were frozen upon col-
lection and stored at 80°C until analyzed.
They were thawed and tested on a single ana-
lytic run. Total cholesterol and glucose con-
centrations were determined with Roche
reagents (Basel, Switzerland). Triglyceride
concentrations were assayed using reagents
from Sigma Diagnostics (Dorset, UK). All
were measured on a Roche COBAS MIRA
Plus instrument that participates and com-
plies with the CDC Lipid Standardization
program. Insulin concentrations were deter-
mined using a DPC Coat-A-Count radioim-
munoassay (RIA) kit. Neuropeptide Y (NPY),
cholecystokinin (CCK) and galanin were
determined using RIA kits from Peninsula
Laboratories (Belmont, CA). C18 Sep-Pak
columns were used to extract the peptides
from the serum sample before testing as rec-
ommended by the manufacturer. Leptin was
determined using the RIA kit from Linco, Inc,
(St. Louis, MO).
Statistical analysis
Comparison of variables was done using
paired ANOVA and Pearson correlations. All
p values of 0.05 were considered statistically
significant. Analyses were performed with SY-
STAT
©
Version 9 (SSPS, Inc.), and SAS version
6.12 (SAS System, SAS Institute, Cary NC).
Data are presented as the mean ± standard de-
viation unless otherwise noted.
RESULTS
Twenty-one subjects were screened for this
study. The first nine male and nine female
adults who met inclusion and exclusion crite-
ria were enrolled and completed the protocol
(n = 18). All subjects were healthy as deter-
mined by medical history, physical examina-
tion and laboratory screening at baseline. All
participants completed the study within the
same 14-week time period February to May
2000. Self-reported urinary ketone analyses in-
dicated all participants were compliant for the
duration of LC diet study period.
Screening data are presented in Table 1.
Mean BMI and weight during the period be-
tween screening and Baseline measurement
were not significantly different, although sub-
jects gained an average of 0.45 ± 2.43 kg. Signif-
icant differences were noted between screening
and Baseline for triglycerides and QUICKI.
Vital signs are listed in Table 2. A value of
3.5 ± 1.3% weight reduction was seen dur-
AN EVALUATION OF THE ATKINS’ DIET 301
TABLE 1. SCREENING DATA
a
Number (men/women) 18 (9/9)
Age (years) 39.8 ± 8.1
Height (m) 1.70 ± 0.09
Weight (kg) 106.9 ± 24.0
BMI (kg/m
2
) 36.4 ± 6.5
Serum Creatinine (mg/dL) 0.9 ± 0.2
AST (U/L) 24 ± 10
ALT (U/L) 34 ± 18
Total bilirubin (mg/dL) 0.7 ± 0.2
Glucose (mg/dL) 97.7 ± 8.1
Fasting insulin (µU/mL) 9.7 ± 6.8
Total cholesterol (mg/dL) 189.6 ± 28.0
Triglycerides (mg/dL)
b
149.9 ± 85.2
HOMA 2.39 ± 1.75
QUICKI
b
0.352 ± 0.042
a
Mean ± standard deviation unless otherwise noted.
b
p  0.02 compared to Baseline.
AST, aspartate aminotransferase; ALT, alanine
aminotransferase; HOMA, homeostasis Model Assess-
ment, insulin resistance; Quick1, quantitative insulin sen-
sitivity check index.
13644C08.PGS 4/15/04 2:11 PM Page 301

ing the Induction diet and 4.9 ± 1.5% by the
end of the OWL diet (a total of 4 weeks LC
diet) both p  0.001, versus Baseline diet.
There was an average change of 1081 kcal/
day (p  0.001) during the Induction diet.
Table 3 presents caloric and macronutrient in-
take during the study. During OWL diet, caloric
intake remained significantly lower than dur-
ing the Baseline diet and changed an average of
923 kcal/day from baseline (p  0.001). Two
subjects gained weight during the OWL diet
(0.2 kg was the maximal weight gain). Over the
4 weeks of LC diet, the maximal, median and
minimal weight loses were 12.2, 5.1, and 1.7, re-
spectively. Pearson correlation coefficients for
change in caloric intake versus change in BMI
were 0.64 (p  0.001) and 0.40 (p  0.001) for the
Baseline diet compared to the Induction and
OWL diets, respectively. Carbohydrate (CHO)
intake decreased to less than 10% of Baseline in-
take during both LC diet phases. Total daily
calories derived from protein and fat did not
differ significantly between study phases.
Actual mean weight loss and expected
weight loss based on caloric intake reductions
seen during each LC diet are presented in
Table 4. Subjects lost more weight than ex-
pected based on the observed reductions of
caloric intake. The additional weight loss all
occurred during the Induction diet. During the
OWL diet, less weight was lost than expected.
Table 5 shows macronutrients/kg body mass.
Total fiber content of the diet decreased sig-
nificantly during the Induction and OWL diets
(Table 6). All subjects ate higher quantities of
animal protein and lower amounts of plant pro-
tein while on the LC diets. Total saturated and
monounsaturated fat intakes were unchanged.
Total cholesterol intake nearly doubled during
the LC diets and total polyunsaturated fatty
acid intake decreased significantly.
Compared to Dietary Reference Intakes
(DRI) and Recommended Dietary Allowances
(RDA), intake of many micronutrients was sig-
nificantly lower than recommended during
the 4 weeks of LC diet (Table 7). The mean in-
take of calcium met adequate daily allowances
at Baseline, but intake significantly decreased
on the LC diets. Sodium intake exceeded rec-
ommendation at Baseline. Table 7 presents ad-
302 MILLER ET AL.
TABLE 2. VITAL SIGNS (N = 18)
a
Baseline Induction OWL
Weight (kg) 107.4 ± 24.2 103.6 ± 23.0
b
102.1 ± 22.6
c
BMI (kg/m
2
) 36.6 ± 6.6 35.3 ± 6.4
b
34.8 ± 6.2
c
Blood pressure (mm Hg)
Systolic 130 ± 11 128 ± 13 122 ± 11
Diastolic 85 ± 8 82 ± 9
b
77 ± 11
Pulse (bpm) 77 ± 13 77 ± 12 79 ± 8
a
Data are presented as mean ± standard deviation unless otherwise noted.
b
p  0.001, versus Baseline diet.
c
p  0.001, versus Baseline and Induction diets.
TABLE 3. DAILY CALORIC AND MACRONUTRIENT INTAKE (N = 18)
a
Baseline Induction OWL
Daily caloric intake 2481 ± 723 1400 ± 472
b
1558 ± 490
b
(kcal/day)
Carbohydrate (g/day) 265.0 ± 98.8 21.0 ± 9.0
b
23.9 ± 13.7
b
Percentage calories 42.7 ± 8.6 6.3 ± 2.8
b
6.1 ± 2.5
b
as carbohydrate
Protein (g/day) 96.2 ± 35.2 102.4 ± 46.9 110.5 ± 35.7
Percentage calories as protein 15.5 ± 2.6 28.8 ± 4.7
b
28.7 ± 5.5
b
Fat (g/day) 113.2 ± 38.6 99.0 ± 31.7 111.9 ± 38.5
Percentage calories as fat 41.1 ± 6.9 63.8 ± 4.8
b
64.2 ± 5.0
b
a
Data are presented as mean ± standard deviation. Based on 3-day diet records.
b
p  0.001 versus Baseline diet.
13644C08.PGS 4/15/04 2:11 PM Page 302

ditional micronutrients that have the potential
to impact health.
Glucose concentrations remained unchanged
throughout this study, while insulin decreased
significantly during the Induction diet decreas-
ing HOMA and increasing QUICKI for this LC
diet phase only, but not the OWL diet, both
compared to Baseline. Triglycerides decreased
by more than 20% and total cholesterol 7% dur-
ing both LC diet phases compared to Baseline
(Table 8).
NPY and leptin both decreased, 14% and
50% respectively during the LC diets com-
pared to Baseline (Table 9). Screening and
AN EVALUATION OF THE ATKINS’ DIET 303
TABLE 4. ACTUAL WEIGHT LOSS COMPARED TO EXPECTED WEIGHT LOSS (N = 18)
a
Induction OWL Total
Actual mean weight loss (kg) 3.8 1.5 5.3
Mean caloric deficit (kcal) 15,120 12,922 28,042
b
Expected mean weight loss (kg)
c
2.0 1.7 3.7
Actual  expected weight (kg) 1.8 0.2 1.6
a
Based on analyses of the 3-day diet records. Mean caloric deficit = mean daily caloric
deficit {change in kcal/day}  14 days. Mean daily caloric deficit = Baseline mean daily
caloric intake  (Induction or OWL) mean daily caloric intake.
b
Mean caloric deficit of the induction plus ongoing weight loss diet.
c
Based on caloric deficit of 3500 kcal equals approximately 0.45 kg weight loss.
TABLE 5. MACRONUTRIENT RATIOS
a
Baseline Induction OWL
kcal/kg 23.8 ± 7.1 13.7 ± 4.1
b
15.3 ± 3.6
b
g CHO/kg 2.54 ± 0.90 0.21 ± 0.09
b
0.24 ± 0.12
b
g protein/kg 0.91 ± 0.30 1.00 ± 0.41 1.09 ± 0.27
c
g fat/kg 1.09 ± 0.37 0.97 ± 0.28 1.10 ± 0.29
CHO/protein ratio 2.90 ± 1.09 0.23 ± 0.14
b
0.23 ± 0.1
b
a
Data are presented as mean ± standard deviation.
p  0.001.
c
p  0.05, versus Baseline diet.
TABLE 6. MACRONUTRIENT SUBTYPES (N = 18)
a
Baseline Induction OWL
Carbohydrate
Total fiber (g/day) 16.59 ± 9.90 3.27 ± 1.85
b
4.24 ± 3.03
b
Protein
Animal (g/day) 65.8 ± 28.4 97.9 ± 46.5
||
104.6 ± 33.8
||
Vegetable (g/day) 29.6 ± 15.2 4.0 ± 4.6
b
5.3 ± 6.1
b
Fatty acids (FA)
Cholesterol (mg/day) 390 ± 212 700 ± 279
b
781 ± 296
b
Total saturated FA (g/day) 39.7 ± 14.1 34.8 ± 11.4 42.1 ± 14.4
Total monounsaturated FA (g/day) 43.7 ± 16.3 37.6 ± 13.0 43.1 ± 17.2
Total polyunsaturated FA (g/day) 21.7 ± 8.5 17.8 ± 7.3
c
16.7 ± 8.1
c
Percentage of calories from fat per day
% Calories SFA 14.4 ± 2.6 22.5 ± 3.3
b
24.3 ± 3.5
b
% Calories MFA 15.8 ± 3.4 24.1 ± 2.7
b
24.6 ± 2.9
b
% Calories PFA 7.9 ± 2.4 11.6 ± 2.7
d
9.6 ± 3.1
Miscellaneous
Alcohol (g/day) 4.5 ± 8.6 0.0 ± 0.1
d
0.0 ± 0.0
d
Caffeine (mg/day) 218 ± 175 112 ± 131 109 ± 158
a
Data are presented as mean ± standard deviation.
b
p  0.001,
c
p  0.05,
d
p  0.01 versus Baseline diet.
13644C08.PGS 4/15/04 2:11 PM Page 303

Baseline CCK significantly differed. Galanin
remained unchanged for the duration of the
study.
A number of adverse effects were reported
during the LC diet phases. Gastrointestinal
side effects were the most common, followed
by headaches and changes in vision (Table 10).
CONCLUSION
This longitudinal study of The Atkins’ diet
strongly indicates that the mechanism of
weight loss during LC diet is an unsolicited,
but obligatory decrease in caloric intake. Par-
ticipants lost a significant amount of weight
304 MILLER ET AL.
TABLE 7. PROTEIN AND SELECTED MICRONUTRIENTS COMPARED TO DIETARY REFERENCE INTAKES OR RECOMMENDED DIETARY
ALLOWANCES FOR MEN AND WOMEN AGED 31–50 YEARS
a
Baseline Induction OWL
Men (n = 9)
Protein (g) 63 111 ± 44
b
115 ± 61
b
126 ± 32
b
Calcium (mg) 1000
c
1139 ± 458 420 ± 180
b
503 ± 153
b
Sodium (mg) 920–2300 5251 ± 2205
b
3517 ± 1468
b
3481 ± 588
b
Potassium (mg) 1950–5460 2941 ± 949 1747 ± 598 1962 ± 310
Iron (mg) 7 21 ± 12
b
9 ± 4 12 ± 3
b
Magnesium (mg) 420 350 ± 149 163 ± 61
b
180 ± 41
b
Phosphorus (mg) 700 1741 ± 599
b
1183 ± 686
b
1392 ± 250
b
Zinc (mg) 12 16 ± 7 14 ± 5 17 ± 4
b
Selenium (µg) 55 167 ± 68
b
142 ± 107
b
146 ± 54
b
Vitamin A (µg) 750 1553 ± 1733 513 ± 238
b
622 ± 355
Vitamin E (mg) 15 15 ± 16 29 ± 69 7.1 ± 2.5
b
Vitamin D (µg) 5
c
8.3 ± 9.9 5.5 ± 6.4 4.3 ± 2.9
Vitamin C (mg) 90 90 ± 77 31 ± 27
b
26 ± 22
b
Thiamin (mg) 1.2 2.8 ± 2.1
b
3.9 ± 8.7 0.9 ± 0.3
b
Riboflavin (mg) 1.3 3.1 ± 2.1
b
1.5 ± 0.7 1.8 ± 0.5
b
Niacin (mg) 16 37 ± 25
b
25 ± 22 22 ± 9
b
Folacin (mg) 400 496 ± 446 142 ± 62
b
191 ± 56
b
Vitamin B
6
(mg) 2.4 2.6 ± 2.2 1.5 ± 0.7
b
1.5 ± 0.4
b
Vitamin B
12
(mg) 2.4 8.0 ± 6.3
b
7.6 ± 6.4
b
9.0 ± 3.3
b
Pantothenic acid (mg) 5
c
8.3 ± 6.9 14.8 ± 31.8 4.3 ± 1.0
b
Women (n = 9)
Protein (g) 50 81 ± 13
b
90 ± 25
b
95 ± 33
b
Calcium (mg) 800
c
801 ± 309 513 ± 223
b
517 ± 268
b
Sodium (mg) 920–2300 3751 ± 811
b
3154 ± 914
b
3263 ± 1089
b
Potassium (mg) 1950–5460 2580 ± 632 1478 ± 545 1654 ± 705
Iron (mg) 12.6 13 ± 3 8 ± 2
b
8 ± 4
b
Magnesium (mg) 270 280 ± 72 129 ± 42
b
182 ± 163
Phosphorus (mg) 1000 1283 ± 261
b
1074 ± 299 1192 ± 522
Zinc (mg) 12 10 ± 2
b
11 ±4 12 ±6
Selenium (mmol) 85 128 ± 21
b
108 ± 34
b
112 ± 40
b
Vitamin A (µg) 750 779 ± 387 1023 ± 1094 703 ± 385
Vitamin E (mg) 7 9.4 ± 3.8 10.7 ± 14.9 7.0 ± 5.0
Vitamin D (µg) 5 4.8 ± 2.9 6.1 ± 6.2 3.9 ± 2.0
Vitamin C (mg) 50 84 ± 46
b
48 ± 48 41 ± 33
Thiamin (mg) 0.8 1.5 ± 0.3
b
0.9 ± 1.1 0.8 ± 0.4
Riboflavin (mg) 1.2 1.8 ± 0.4
b
1.9 ± 1.2 1.5 ± 0.4
b
Niacin (mg) 14 19 ± 3
b
19 ± 18 18 ± 7
Folacin (mg) 200 209 ± 65 236 ± 279 177 ± 105
Vitamin B
6
(mg) 1.4 1.5 ± 0.4 1.5 ± 1.7 1.2 ± 0.6
Vitamin B
12
(mg) 2 4.3 ± 1.3
b
7.0 ± 4.7
b
5.6 ± 2.8
Pantothenic acid (mg) 5
c
3.9 ± 1.1
b
6.1 ± 6.9 4.1 ± 1.7
a
Data are presented as mean intake per day ± standard deviation.
b
p  0.05 compared to dietary reference intake.
c
Adequate intakes.
Data from previous research.
64
13644C08.PGS 4/15/04 2:11 PM Page 304

without prescribed reduction in caloric intake,
professional dietary counseling or behavioral
modification therapy. This finding is consis-
tent with other studies that show short-term
adherence to LC diet yields weight reduc-
tion.
15,20–23
Previous studies have found that
more rapid weight loss occurs during the ini-
tial 1–2 weeks of a dieter-implemented LC diet
than a conventional weight-reducing diet.
After this initial period, weight loss has been
similar in diets with comparable caloric in-
take.
24,25
Most recently, in longer-term RCTs,
LC diet induced greater weight loss than con-
ventional weight loss diet.
7,8,9,10
The reduced caloric intake seen during this
study of LC diet provides additional support
for the belief that self-imposed carbohydrate re-
striction results in decreased caloric intake.
3,26
Men achieved a 5.3% weight loss and women
4.5% in only 1 month. As little as 5–10% weight
loss has been shown to have beneficial effect on
hypertension, hyperlipidemia, hyperglycemia,
and insulin sensitivity.
27–30
Weight change
correlated with caloric intake; however, weight
loss was greater than expected based on the
3-day food diary analysis (Table 4). Participants
were asked not to change their exercise habits,
AN EVALUATION OF THE ATKINS’ DIET 305
TABLE 8. METABOLIC PARAMETERS (N = 18)
a
Baseline Induction OWL
Glucose (mg/dL) 98.8 ± 16.0 98.9 ± 11.0 102.3 ± 9.4
Fasting insulin (µU/mL) 15.4 ± 12.5 8.2 ± 7.8
b
10.2 ± 5.7
Total cholesterol (mg/dL) 194.7 ± 31.0 177.9 ± 28.0
b
178.3 ± 26.3
b
Triglycerides (mg/dL) 196.4 ± 110.9 118.4 ± 50.5
b
130.4 ± 76.5
b
HOMA 3.81 ± 3.13 2.09 ± 2.23
b
2.64 ± 1.62
QUICKI 0.326 ± 0.029 0.362 ± 0.040
b
0.342 ± 0.033
a
Data are presented as mean intake per day ± standard deviation.
b
p  0.01 versus Baseline diet.
TABLE 9. SAFETY FACTORS
a
Baseline Induction OWL
NPY (pg/mL) (n = 13) 184.1 ± 34.6 163.0 ± 25.4
b
159.0 ± 16.6
b
Leptin (mg/dL) (n = 15) 22.6 ± 8.2 11.4 ± 4.9
c
11.7 ± 5.2
c
Galanin (pg/mL) (n = 13) 27.8 ± 8.2 25.2 ± 5.0 25.8 ± 4.8
CCK (pg/mL) (n = 13) 35.2 ± 8.5 40.3 ± 11.2 38.0 ± 11.1
a
Data are presented as mean intake per day ± standard deviation.
b
p  0.05, versus Baseline.
c
p  0.001, versus Baseline diet.
TABLE 10. ADVERSE EVENTS DURING LOW-CARBOHYDRATE
DIET (N = 18)
Organ system Number of subjects (%)
Constitutional
Fatigue 1 (6)
Gastrointestinal
Diarrhea 6 (33)
Constipation 4 (22)
Nausea 4 (22)
Bad taste in mouth 4 (22)
Halitosis 3 (17)
Dyspepsia 3 (17)
Anorexia 2 (11)
Abdominal pain 1 (6)
Flatulence 1 (6)
Decreased dyspepsia 1 (6)
Rectal irritation 1 (6)
Dry mouth 1 (6)
Genitourinary
Polyuria or nocturia 2 (11)
Odiferous urine 1 (6)
Neurological
Headache 4 (22)
Change in vision 4 (22)
Altered mood 3 (17)
Dermatologic
Erythema of skin 1 (6)
Dandruff 1 (6)
Musculoskeletal
Leg cramps 3 (17)
Myalgias 2 (11)
Flank pain 1 (6)
Low back pain 1 (6)
13644C08.PGS 4/15/04 2:11 PM Page 305

although increased physical activity could have
influenced weight loss, as it was not monitored.
There may be other confounding factors, as
RCTs of weight reducing diet alone versus
weight reducing diet with exercise have not
shown increased weight loss.
31–33
For example, the unanticipated additional
weight loss is consistent with the 1–2-kg
weight loss seen with glycogen mobilization
(in liver and muscle) and ketosis-induced di-
uresis (increased delivery of sodium to distal
lumen of the kidney by non-reabsorbable ke-
tones inducing water loss) that occurs with LC
diets.
34,35
Some studies suggest that weight
loss is solely related to caloric reduction and
not to diet composition.
15,20,24,36
This contrasts
with the minority of authors who claim that
weight loss is possible without caloric restric-
tion.
37,38
Others have reported no change in
urinary excretion of nitrogen or creatinine
with modest weight loss during ketogenic
diets, suggesting that LC diets achieve weight
loss with preservation of lean body mass.
39–41
Body composition was not measured due to
the short nature of the current study.
Theoretical and data-supported mecha-
nisms of weight loss with LC diets differ.
20,21
Another theory of weight loss is that the LC
diet becomes a high-protein diet (and high-fat)
leading to early satiety and decreased caloric
intake.
42–44
Advocates of LC diets claim that
ketosis also induces anorexia.
16
It is also hy-
pothesized that a high-protein diet may also
increase energy expenditure even in the ab-
sence of reduced caloric intake by increasing
post-prandial thermogenesis.
45
None of these
theories have been proven, and only one study
has suggested that protein and fat intake do
not change during LC diet.
3
The average caloric reduction was approxi-
mately 1000 kcal/day, or approximately 40% re-
duction in caloric intake (Table 3). Of the two
study subjects who gained weight during the
OWL diet, one had a caloric intake greater than
during the Baseline diet. There is no ready ex-
planation for the weight gain of the other subject
except the known limitations of food diaries.
46
A
National Institutes of Health panel considered
the optimal composition of the diet for treat-
ment of obesity and recommended a low-calorie
diet of approximately 1000–1200 kcal/day to re-
duce body weight by approximately 8% over
3–12 months (500–1000 kcal/day). This reduc-
tion from estimated daily requirements is rec-
ommended for a weight loss of 1–2 lb. per
week.
47
Subjects in this study restricted caloric
intake to this degree and achieved weight re-
duction greater than expected.
During the LC diet, subjects did not com-
pensate for decreased carbohydrate intake by
increasing protein or fat intake (g/day). This
concurs with the early findings of Yudkin and
Cary,
3
in which subjects on diets containing
less than 50 g of daily carbohydrate reduced
their caloric intake by 13–55% and did not in-
crease the absolute amount of dietary protein
or fat intake. Our study participants main-
tained the recommended protein intake for di-
eters of at least 1 g/kg body weight per day
with self-selection of foods (Table 5),
48
but car-
bohydrate content was markedly decreased
and the percentage of both dietary protein and
fat were higher than recommended by the
NIH Expert Panel on the treatment of obesity.
47
A small study by Evans et al.
49
reported no
untoward effects of low-carbohydrate diets
on “nutrient” intake. We observed dietary
deficiencies in vitamin A, vitamin B
6
, folate, vi-
tamin C, magnesium, iron, calcium, and potas-
sium on the LC diets, although it should be
noted that 7-day food records more accurately
assess calcium intake.
50
We did not observe re-
duced intake of vitamin E or thiamin, as has
been previously reported.
9,51
At Baseline, daily
fiber intake was lower than the federal recom-
mendations of 20–30 g/day
47
and then de-
creased markedly during the LC diets.
These findings suggest that LC diets should
be supplemented with dietary vitamin, min-
eral and fiber substitutes. It is possible to
compose a LC diet that meets RDA/DRI rec-
ommendations. This was not the case with our
subjects when self-selecting foods. It is uncer-
tain whether supplements can adequately
replace the range of micronutrients and phyto-
chemicals consumed in a balanced diet con-
taining a variety of whole grains, fruits and
vegetables.
In the current study, significant decreases in
serum total cholesterol and triglycerides were
seen, despite almost double total cholesterol
intake during the LC diets (Tables 6 and 8).
306 MILLER ET AL.
13644C08.PGS 4/15/04 2:11 PM Page 306

Other investigators have reported improve-
ment, worsening, or no change in total choles-
terol and triglycerides after following LC
diets.
20,24,26,52,54,55
Of note, Raeini-Sarjaz et al.
found that caloric restriction was the cause of
lower lipids on diets and that this finding was
present even with high-fat, hypocaloric diets.
56
It has also been shown that weight loss alone re-
sults in decreases in serum cholesterol and
triglycerides independent of carbohydrate con-
tent of the diet.
36
HDL was not measured; thus,
LDL cannot be calculated from the data col-
lected in our study.
This study provides evidence that decreases
in fasting insulin and increases in insulin sen-
sitivity occur with as little as 2 weeks of LC
diet (Table 8). This finding did not persist dur-
ing the OWL diet as carbohydrate intake was
liberalized. Other evidence of LC diet on in-
sulin sensitivity is mixed, but most studies in
obese subjects without type 2 diabetes mellitus
have shown improvement but no difference
when compared to a more conventional
weight-reducing diet.
7–9,20,21,57,58
Serum insulin concentration at screening
and Baseline were not significantly different,
however, both serum triglycerides and QUICKI
significantly worsened from screening to Base-
line measurement (Tables 1 and 8). This is
suggestive of a caloric binge, particularly of
carbohydrates, prior to commencement of this
study (a “last-supper” pre-study effect).
Weight loss studies that report changes in
blood glucose, hormones, lipids, and other
cardiovascular risk factors should be inter-
preted cautiously for this reason. Conven-
tional diet may not be the best “control” diet
during a RCT for measuring these parameters.
In general, improving metabolic parameters
and cardiovascular risk factors with self-im-
plemented diet probably occurs because of
weight loss and decreasing caloric intake, irre-
spective of dietary composition.
21,57,59–63
The observed decrease in NPY and increase
in CCK may reflect pathways that contribute
to a sense of satiety, leading to the consump-
tion of fewer calories on a diet regimen that al-
lows unrestricted food intake. The observed
decrease in NPY is unexpected since the de-
crease in leptin would be expected to result in
increased NPY and increased food intake. This
finding suggests that the LC diet or weight
loss induces changes in NPY that are indepen-
dent of the leptin pathway.
Short-term adherence to The Dr. Atkins’
New Diet Revolution diet with minimal pro-
fessional instruction resulted in weight loss in
healthy adults who were overweight or obese,
but without proper dietary training it also
achieved several micronutrient deficiencies
that could lead to serious heath consequences
if continued long-term. The Atkins’ Diet and
similar diets may be appropriate for short-
term use and induction of weight loss in the
self-implemented dieter, followed by a more
conventional, nutritious diet.
ACKNOWLEDGMENTS
This work was supported by a grant from
The E. Donnall Thomas Resident Research Pro-
gram and the Stephen C. Clark Research Fund.
REFERENCES
1. Westman EC, Mavropoulos J, Yancy WS, et al. A re-
view of low-carbohydrate ketogenic diets. Curr Ath-
eroscler Rep 2003;5:476–483.
2. Wyatt HR. The prevalence of obesity. Prim Care
2003;30:267–279.
3. Yudkin J, Carey M. The treatment of obesity by the
“high-fat” diet: the inevitability of calories. Lancet
1960;2:939–941.
4. Rabast U, Kasper H, Schönborn J. Obesity and low-
carbohydrate diets—comparative studies. Nutr Metab
1977;21:56–59.
5. Yang EJ, Chung HK, Kim WY, et al. Carbohydrate in-
take is associated with diet quality and risk factors
for cardiovascular disease in U.S. adults: NHANES
III. J Am Coll Nutr 2003;22:71–79.
6. Bravata DM, Sanders L, Huang J, et al. Efficacy and
safety of low-carbohydrate diets: a systematic re-
view. JAMA 2003;289:1837–1850.
7. Foster GD, Wyatt HR, Hill JO, et al. A randomized
trial of a low-carbohydrate diet for obesity. N Engl J
Med 2003;348:2082–2090.
8. Samaha FF, Iqbal N, Seshadri P, et al. A low-carbo-
hydrate as compared with a low-fat diet in severe
obesity. N Engl J Med 2003;348:2074–2081.
9. Brehm BJ, Seeley RJ, Daniels SR, et al. A randomized
trial comparing a very low carbohydrate diet and a
calorie-restricted low fat diet on body weight and
cardiovascular risk factors in healthy women. J Clin
Endocrinol Metab 2003;88:1617–1623.
AN EVALUATION OF THE ATKINS’ DIET 307
13644C08.PGS 4/15/04 2:11 PM Page 307

10. Sondike SB, Copperman N, Jacobson MS. Effects of a
low-carbohydrate diet on weight loss and cardiovas-
cular risk factor in overweight adolescents. J Pediatr
2003;142:253–258.
11. Atkinson RL. Role of nutrition planning in the treat-
ment for obesity. Endocrinol Metab Clin 1996;25:
955–964.
12. Sigler MH. The mechanism of natriuresis of fasting. J
Clin Invest 1975;55:377–387.
13. DeHaven J, Sherwin R, Hendler R, et al. Nitrogen
and sodium balance and sympathetic nervous sys-
tem activity in obese subjects treated with a low calo-
rie protein or mixed diet. N Engl J Med 1980;302:
477–482.
14. Yang M-U, Van Itallie TB. Composition of weight
lost during short-term weight reduction. J Clin Invest
1976;58:722–730.
15. Phinney SD, Bistrian BR, Wolfe RR, et al. The human
metabolic response to chronic ketosis without caloric
restriction: physical and biochemical adaptation. Me-
tabolism 1983;32:757–768.
16. Atkins RC. Dr. Atkins’ New Diet Revolution. New
York: Avon Books, 1992.
17. Weber JL, Reid PM, Greaves KA, et al. Validity of
self-reported energy intake in lean and obese young
women, using two nutrient databases, compared
with total energy expenditure assessed by doubly la-
beled water. Eur J Clin Nutr 2001;55:940–950.
18. Nafziger AN, Bertino Jr JS, Bley AG, et al. Incidence
of sexual dysfunction in normal volunteers on flu-
voxamine therapy. J Clin Psychiatry 1999;60:
187–190.
19. Matthews DR, Hosker JP, Rudenski AS, et al. Ho-
meostasis model assessment: insulin resistance and
beta-cell function from fasting plasma glucose and
insulin concentrations in man. Diabetologia 1985;28:
412–419.
20. Golay A, Eigenheer C, Morel Y, et al. Weight-loss
with low or high carbohydrate diet? Int J Obes
1996;20:1067–1072.
21. Baba NH, Sawaya S, Torbay N, et al. High protein vs
high carbohydrate hypoenergetic diet for the treat-
ment of obese hyperinsulinemic subjects. Int J Obes
1999;23:1202–1206.
22. Rabast U, Vornberger KH, Ehl M. Loss of weight, so-
dium and water in obese persons consuming a high-
or low-carbohydrate diet. Ann Nutr Metab 1981;
25:341–349.
23. Genuth SM, Castro JH, Vertes V. Weight reduction in
obesity by outpatient semistarvation. JAMA 1974;
230:987–991.
24. Golay A, Allaz A-F, Morel Y, et al. Similar weight loss
with low- or high-carbohydrate diets. Am J Clin Nutr
1996;64:174–178.
25. Bainbridge JL, Gidal BE, Ryan M. The ketogenic diet.
Pharmacotherapy 1999;19:782–786.
26. Stein K. High-protein, low-carbohydrate diets: do
they work? J ADA 2000;100:760–761.
27. Muscelli E, Camastra S, Catalano C, et al. Metabolic
and cardiovascular assessment in moderate obesity:
effect of weight loss. J Clin Endocrinol Metab 1997;82:
2937–2943.
28. Wadden TA, Anderson DA, Foster GD. Two-year
changes in lipids and lipoproteins associated with
the maintenance of a 5% to 10% reduction in initial
weight: some findings and some questions. Obes Res
1999;7:170–178.
29. Klein S, Wadden T, Sugerman HJ. AGA technical re-
view on obesity. Gastroeneterology 2002;123:882–932.
30. Goldstein DJ. Beneficial health effects of modest
weight loss. Int J Obes Relat Metab Disord 1992;16:
397–415.
31. Ross R, Dagnone D, Jones PJ, et al. Reduction in obe-
sity and related comorbid conditions after diet-
induced weight loss or exercise-induced weight loss
in men. A randomized, controlled trial. Ann Intern
Med 2000;133:92–103.
32. Miller WC, Koceja DM, Hamilton EJ. A meta-
analysis of the past 25 years of weight loss research
using diet, exercise or diet plus exercise intervention.
Int J Obes Relat Metab Disord 1997;21:941–947.
33. Garrow JS, Summerbell CD. Meta-analysis: effect of
exercise, with or without dieting, on the body com-
position of overweight subjects. Eur J Clin Nutr
1995;49:1–10.
34. Denke MA. Metabolic effects of high-protein, low-
carbohydrate diets. Am J Cardiol 2001;88:59–61.
35. Council on Foods and Nutrition. A critique of low-
carbohydrate ketogenic weight reduction regimens:
a review of Dr. Atkins’ Diet Revolution. JAMA
1973;224:1415–1419.
36. Alford BB, Blankenship AC, Hagen RD. The effects
of variations in carbohydrate, protein, and fat con-
tent of the diet upon weight loss, blood values, and
nutrient intake of adult obese women. J Am Dietetic
Assoc 1990;90:534–540.
37. Kerwick A, Pawan GLS. Calorie intake in relation to
body-weight changes in the obese. Lancet 1956;ii:
155–161.
38. Dwyer JT, Lu D. Popular diets for weight loss: From
nutritionally hazardous to healthful. In: Stunkard AJ,
Wadden TA (eds). Obesity: Theory and Therapy, 2nd
ed. New York: Raven Press, 1993:231.
39. Rabast U, Vornberger KH, Ehl M. Loss of weight, so-
dium and water in obese persons consuming a high-
or low-carbohydrate diet. Ann Nutr Metab 1981;25:
341–349.
40. Sigler MH. The mechanism of natriuresis of fasting. J
Clin Invest 1975;55:377–387.
41. Reid D, Rasio E. Glucose utilization and nitrogen
balance in the obese on a protein-lipid hypocaloric
diet. Can J Physiol Pharmacol 1982;60:319–323.
42. Raben A, Agerholm-Larsen L, Flint A, et al. Meals
with similar energy densities but rich in protein, fat,
carbohydrate, or alcohol have different effects on en-
ergy expenditure and substrate metabolism but not
on appetite and energy intake. Am J Clin Nutr
2003;77:91–100.
43. Vozzo R, Wittert G, Cocchiaro C, et al. Similar effects
of foods high in protein, carbohydrate and fat on
308 MILLER ET AL.
13644C08.PGS 4/15/04 2:11 PM Page 308

subsequent spontaneous food intake in healthy indi-
viduals. Appetite 2003;40:101–107.
44. Westerterp-Plantenga MS. The significance of pro-
tein in food intake and body weight regulation. Curr
Opin Clin Nutr Metab Care 2003;6:635–638.
45. Johnston CS, Day CS, Swan PD. Postprandial ther-
mogenesis is increased 100% on a high-protein, low-
fat diet versus a high-carbohydrate, low-fat diet in
healthy, young women. J Am Coll Nutr 2002;21:
55–61.
46. Mahalko JR, Johnson LK, Gallagher SK, et al. Com-
parison of dietary histories and seven-day food
records in a nutritional assessment of older adults.
Am J Clin Nutr 1985;42:542–553.
47. National Institutes of Health, National Heart, Lung,
and Blood Institute: Clinical Guidelines on the Iden-
tification, Evaluation, and Treatment of Overweight
and Obesity in Adults: Evidence Report. U.S. De-
partment of Health and Human Services. NIH publi-
cation number 00-4083, June 1998.
48. Marliss EB. Protein diets for obesity: metabolic and
clinical aspects. CMA J1978;119:1413–1421.
49. Evans E, Stock A, Yudkin J. The absence of undesir-
able changes during the consumption of the low car-
bohydrate diet. Nutr Metabol 1974;17:360–367.
50. Bingham SA, Gill C, Welch A, et al. Validation of di-
etary assessment methods in the UK arm of EPIC
using weighed records, and 24-hour urinary nitro-
gen and potassium and serum vitamin C and
carotenoids as biomarkers. Int J Epidemiol 1997;26:
S137–S151.
51. Freedman MR, King J, Kennedy E. Popular diets: a
scientific review. Obesity Res 2001;9:1S–40S.
52. Kasper H, Thiel H, Ehl M. Response of body weight
to a low carbohydrate, high fat diet in normal and
obese subjects. Am J Clin Nutr 1973;26:197–204.
53. Allara L. The return of the high-protein, low-
carbohydrate diet: weighing the risks. Nutr Clin
Pract 2000;15:26–29.
54. Larosa JC, Fry AG, Muesing R, et al. Effects of high-
protein, low-carbohydrate dieting on plasma lipo-
proteins and body weight. J Am Diet Assoc 1980;
77:264–70.
55. Lewis SB, Wallin JD, Kane JP, et al. Effect of diet
composition on metabolic adaptations to hypo-
caloric nutrition: comparison of high carbohydrate
and high fat isocaloric diets. Am J Clin Nutr 1977;30:
160–170.
56. Raeini-Sarjaz M, Vanstone CA, Papamandjaris AA,
et al. Comparison of the effect of dietary fat restric-
tion with that of energy restriction on human lipid
metabolism. Am J Clin Nutr 2001;73:262–267.
57. Layman DK, Shiue H, Sather C, et al. Increased di-
etary protein modifies glucose and insulin homeo-
stasis in adult women during weight loss. J Nutr
2003;133:405–410.
58. Layman DK, Boileau RA, Erickson DJ, et al. A reduced
ratio of dietary carbohydrate to protein improves
body composition and blood lipid profiles during
weight loss in adult women. J Nutr 2003;133:411–417.
59. Golay A, Allaz AF, Ybarra J, et al. Similar weight loss
with low-energy food combining or balanced diets.
Int J Obes Relat Metab Disord 2000;24:492–496.
60. Lean ME, Han TS, Prvan T, et al. Weight loss with
high and low carbohydrate 1200 kcal diets in free liv-
ing women. Eur J Clin Nutr 1997;51:243–248.
61. Piatti PM, Pontiroli AE, Saibene A, et al. Insulin sen-
sitivity and lipid levels in obese subjects after slim-
ming diets with different complex and simple
carbohydrate content. Int J Obes Relat Metab Disord
1993;17:375–381.
62. Piatti PM, Monti F, Fermo I, et al. Hypocaloric high-
protein diet improves glucose oxidation and spares
lean body mass: comparison to hypocaloric high-
carbohydrate diet. Metabolism 1994;43:1481–1487.
63. Poppitt SD, Keogh GF, Prentice AM, et al. Long-term
effects of ad libitum low-fat, high-carbohydrate diets
on body weight and serum lipids in overweight sub-
jects with metabolic syndrome. Am J Clin Nutr 2002;
75:11–20.
64. Recommended Dietary Allowances, 10
th
ed. National
Academy Press, 1989. Dietary Reference Intakes, Di-
etary Reference Intakes for Calcium, Phosphorus, Magne-
sium, Vitamin D, and Fluoride. National Academy
Press, 1997.
Address reprint requests to:
Bernard V. Miller III, M.D.
Department of Internal Medicine
Division of Geriatrics and Nutritional Sciences
Center for Human Nutrition
Washington University School of Medicine
660 South Euclid Ave.
Campus Box 8031
Saint Louis, MO 63110
E-mail: bvmiller@im.wustl.edu
or diaitaleptos@sbcglobal.net
AN EVALUATION OF THE ATKINS’ DIET 309
13644C08.PGS 4/15/04 2:11 PM Page 309