Protein Metabolism in Obese Patients During Very Low-Calorie Mixed Diets

Containing Different Amounts of Proteins and Carbohydrates

Renato Pasquali,

Francesco

Casimirri,

and Nazario Melchionda

To assess
long-term
nitrogen
sparing
capacity
of very low-calorie
mixed
diets, we administered
two isoenergetic
(2092KJ)
liquid formula
regimens
of different
composition
for 8 weeks
to two matched
groups
of massively
obese
patients
(group
1:
proteins
60 g, carbohydrate
64 g; group
2: proteins
41 g, carbohydrates
81 g). Weight
loss was similar
in both groups.
Daily
nitrogen
balance
(g) during
the second
month
resulted
more negative
in group
2 with respect
to group
1. However,
within
the groups
individual
nitrogen
sparing
capacity
varied
markedly;
only a few patients
in group
1 and one in group
2 were able
to attain
nitrogen
equilibrium
throughout
the study.
Daily urine excretion
of 3-methylhistidine
fell significantly
in group
1
but did not change
in group
2. Unlike
total
proteins,
albumins,
and transferrin.
serum
levels
of retinol-binding
protein,
thyroxin-binding
globulin,
and complement-C3
fell significantly
in both groups
but per cent variations
of complement-C3
were
more
pronounced
in the first group.
Prealbumin
levels
fell persistently
in group
1 and transiently
in group
2. The
results
indicate
that even with this type of diet an adequate
amount
of dietary
protein
represents
the most important
factor
in minimizing
whole
body protein
catabolism
during
long-term
semistarvation
in massively
obese patients.
Moreover,
they
confirm
the possible
role of dietary
carbohydrates
in the regulation
of some visceral
protein
metabolism.
(D 1987 by Grune
& Stratton,
Inc.

INCE very-low-calorie diets (VLCD) have been used in

the treatment of obesity, several studies have been
carried out with the aim of verifying the relative efficiency of
proteins and carbohydrates, taken singly or in various combinations, to spare whole body proteins.6 It is generally
accepted that adequate (ie, 1.5 g/kg ideal body weight
(IBW) per day) high quality protein supplementation as a
unique source of calories may be sufficient to maintain
nitrogen (N) equilibrium and total protein turnover. On the
other hand, several authors are against the use of carbohydrate-free VLCD due to greater stimulation of ketone body
formation, hyperuricemia,5v6and impaired sympathetic nervous system activity. The optimum composition of VLCmixed diets is still under discussion. In fact, several N
balance studies, providing different amounts of proteins and
carbohydrates, have documented satisfactory protein sparing
effects obtained in a short period of time (ie, 3 to 4
weeks),2*8s3
while other studies failed to confirm these
results even with the administration of larger intakes of these
nutrients.*2*8 Moreover, Bistrian et a1,19using the Nglycine infusion technique, found that a 3-week diet containing 0.7 g of carbohydrates added to 0.8 g of proteins per
kg/IBW was unable to maintain adequate protein synthesis
and turnover in moderately obesesubjects. Becauseof these
discrepancies we compared some aspects of the protein
metabolic responseto isoenergentic VLC-mixed diets providing different amounts of proteins and carbohydrates. Particular emphasis was placed on the analysis of the individual N
balance since previous reports have shown that obesepatients
may present an extreme variability in N lossand that some of
them are: unable to obtain N equilibrium even after a long
period of treatment.. Blood protein profiles and urinary
3-methylhistidine excretion, which can be used as a reliable
index of muscle protein breakdown,20*2were also compared
to evaluate the specific effects of these regimens of visceral
and muscle protein metabolism.

MATERIALS
Patients

and

Protocol

AND

METHODS

Study

This study was carried out on 12 massively obese inpatients

admitted to the Institute of Clinical Medicine and Gastroenterology
Metabolism,

Vol36,

No 12 (December),

1987:

pp

1141-l

148

of the University of Bologna. One patient (BT) in group 2 had type

II diabetes with moderate fasting hyperglycemia (115 mg/dL); one
patient in each of the two groups (LV and CL, respectively)
presented combined hyperlipoproteinemia (type IIB); four patients
in group 1 and five in group 2 presented moderate hypertension; none
had clinically evident ischemic heart disease. All subjects gave
informed consent for the study. On admission, all followed a
standardized diet providing approximately 2,000 (8,368 kJ) kcal and
250 kcal of carbohydrates. During this period, baseline data were
collected and clinical, laboratory (which will be reported below), and
diagnostic procedures were performed. None were taking drugs nor
had undergone x-ray examination with iodine compounds. Subjects
were then randomly assigned to one of two experimental dietary
treatments, which lasted for 10 weeks and consisted of an 8-week
semistarvation program followed by 2 weeks of hypocaloric dietary
refeeding. As reported in Table 1, both groups were similar with
regard to sex prevalence, age, body weight and body mass index
(BMI, calculated as weight in kg, divided by height square in m).
During the semistarvation period diets were administered as liquid
formula. The analytical composition of these two regimens containing 500 kcal (2,092 kJ) is reported in Tables 2 and 3. Water was
allowed ad libitum (minimum 2 L/d); allopurinoI(IO0 to 300 mg/d)
and ursodeoxycholic acid (300 mg in the evening) were also administered in order to control serum acid uric levels and reduce cholesterol
concentration in the bile and prevent gallstone formation.* After
semistarvation treatment all patients were gradually transferred to a
normal diet containing 750 kcal (3,138 kJ) (proteins 50 g, lipids 25
g, and carbohydrates 82 g) for a week and, subsequently, 1,000 kcal
(4,184 kJ) (proteins 50 g, lipids 45 g, and carbohydrates 100 g)
during the second week. While in the hospital all patients continued
to follow their normal lifestyle wherever possible. Compliance with
the nutritional program was checked throughout the study period;

From the Institute

of Clinical
Medicine
1, S. Orsola Hospital,
University
of Bologna. Italy.
Supported
in part by a grant from
the Minister0
di Pubblica
Instruzione.
40%, 1984-l 985.
Presented
at the 5th International
Congress on Obesity, Jerusalem, 1986.
Address
reprint requests to Renato Pasquali.
MD, Clinica Medica I, Ospedale S. Orsola, Universith
di Bologna. via Massarenti
9,
40138 Bologna, Italy.
o 1987 by Grune & Stratton,
Inc.

0026-0495/87/3612-0005$03.00/O
1141

1142

PASQUALl,

Table

1.

General

Data

on Obese

Study

CASIMIRRI,

AND

MELCHIONDA

Patients

sex

Age
(Vd

Bodv
Weight
(kgl

Group 1
CC

40

142.0

1.54

59.8

LV
PG

F
M

43
33

115.4
139.1

1.54
1.81

48.6
42.4

8T
AG

F
M

49
32

97.7
141.0

1.47
1.89

45.2
39.4

Subjects

AB
Mean

M
+ SD

Group 2
CL
SE

3M/3F

47
40.6

? 7.1

F
M

23
40

SC

VM
cs

M
M

8A

Mean

r SD

3M/3F

Height
(m)

127.0
127.0

Body Mass
(kg/d

1.65
17.6

1.65

r 0.2

46.7
47.0

f 7.1

128.8
130.8

1.63
1.71

48.4
44.7

26

103.1

1.60

40.2

37
56

202.7
125.1

1.91
1.74

55.5
41.3

40

138.3

1.70

47.8

37.0

k 11.8

frequent checks of daily urine sodium and potassium excretion and

acetoacetate levels (Ketur test, Boehringer-Biochemia-Robin,
Milan, Italy) were performed. Moreover, plasma levels of acetoacetate and @-hydroxybutyrate were measured according to the method
of Williamson et al.23

Biochemical Evaluation
To obtain urine 3-methylhistidine (pmol/d) concentrations during the baseline period, adequate 24-hour urine collection was made
by each subject after three days of meat-free diet, following all
procedures previously described by Lukaski et al.* Daily 3-methylhistidine excretion was then measured weekly starting from the
seventh day of therapy until the end of the semistarvation period.
Assays were performed on urine samples stored at -20C until
analysis. 3-Methylhistidine was measured with an Aminoanalyser
(Kontron, Liquimat III) by an ionic exchange resin technique after
they had been deproteinized with tricloroacetic acid (pH 2) and
filtered on Acrodisc (0.45 &rn) (Gelman, Milano, Italy). According
to Young and Munro19 rates of muscle protein breakdown (g/d)
were calculated from 3-methylhistidine urine values. Serum creatinine levels and daily urine excretions of creatinine determined in the
same urine samples were performed using the enzymatic colorimetric method (Creatinine PAP, Boehringer, Mannheim). N urine
excretion was measured during semistarvation on samples of urine
collected every day, frozen at -20C at the end of collection, and
then analyzed in individual samples, each of which represented a
pool of seven days. The pool was made by adding together a quantity
of urine (I mL/L) from each days sample. Urinary N excretion was
determined by the standard Kjeldahl technique. Keeping fecal,
integumental, and miscellaneous N loss constant (approximately
1.25 g/d),% N balance was calculated according to the formula:
N input - N output (urine N + 1.25 g N)
Baseline blood samples for biochemical analysis were collected on
the day the subjects began the experimental diets and thereafter at
weekly intervals during semistarvation and refeeding. Blood concentration of retinal-binding protein (RBP), thyroxin-binding globulin
(TBG), and prealbumin (PA) were measured by quantitative radial
immunodiffusion using commercially available diffusion plates
(Behringwerke AG, Marburg, 1st. Behring SpA, LAquila, Italy).

138.1

+ 33.8

1.71

+ 0.1

46.3

L 5.6

The coefficient of variation (CV) was determined three times for

each of these proteins and was between 1% and 8% (average 3.5%).
Total proteins (Biuret-EDTA Method, CV, 2.3%), albumins (cellulose acetate electrophoretic method, CV ~4%) transferrin (immuno-nephelometric method, CV, 3.2%), and complement-C3 (immuno-nephelometric Laser method, CV, 8%) were also measured
using the same blood samples.

Statistics
Statistical evaluation was performed by using the one-way analysis of variance (ANOVA) and the Wilcoxon rank sum test. Results
are reported as mean + SD.
RESULTS

The diets were well-tolerated by all subjects and there

were no significant medical events, postural hypotension, or
abnormalities in routine blood analyses.All subjects considered in the study revealed optimal compliance to dietary
treatment throughout the study. As expected, plasma ketone
bodies rose more in group 1 than in group 2. After the first
week, levels of acetoacetate and @-hydroxybutyrate reached
a level that was subsequently maintained throughout the
semistarvation period. Plateau values were then calculated
according to the mean individual data measured from the
first to the eighth week. Acetoacetate levels rose from basal
0.054 + 0.060 mmol/L to plateau values of 0.164 + 0.060
mmoI/L in group 1 and from 0.043 + 0.015 mmol/L to
0.110 + 0.074 mmol/L in group 2 (P < .05). @-Hydroxybutyrate levels rose from 0.076 + 0.030 mmol/L to 0.449 +
0.157 mmol/L in group 1 and from 0.074 + 0.041 mmol/L
to 0.290 5 0.130 mmol/L in group 2 (P < .05).
Weight Loss
Cumulative weight loss during the IO-week period
resulted 22.4 + 7.0 kg in group 1 and 17.8 +_5.0 kg in group 2
(NS). During semistarvation, weight loss was, respectively,
21.2 2 6.3 kg and 18.1 2 5.0 kg (NS), while during the

PROTEIN

METABOLISM

Table

2.

DURING

Energy

LOW-CALORIE

Content

and Dietary

2 Liquid

Formula

Nutrients

Composition

of the

fg)

Group

1.3
81

and vitamins

Sodium (g)
Potassium (g)
Calcium (g)

0.84
2

Magnesium

fg)

1
0.48

Phosphorus
Clorine (g)

fg)

1.16
1.16
1.01

0.88

Iron (mg)

18
15

Zinc (mg)
Manganese
(mg)
Copper fmg)
Iodine (fig)
Cobalt (pg)
Molybdenum

(pg)

40

Vanadium

45
-

115

107

8oron tpg)
Aluminium (mg)
Vitamin A (IU)
Vitamin

B, (mg)

Vitamin
Vitamin

B, (mg)
B, (mg)

\litamin

B,, (pg)

Vitamin
Vitamin

C tmg)
D, (IU)

Vitamin
Vitamin

E fmg)
PP (mg)

0.4
3,800
0.8
1

2.1
2.6
2.4

1
5

6
97.5
164

43.8
18

Vitamin l-l (fig)

Pantotenic acid (mgl

16.8
21.4

250

38.3
11.2

Folic acid (mg)

0.4

0.5

Coline fmg)

200

The amount
of protein
administered
consisted
of whole protein,
administered
as casein, egg albumin, and milk albumin. Carbohydrates
were

derived

saccarosa;

from

a mixture

in the Nutroclin

of hydrolyzed

preparation

starch,

malt

4 g/d of galattose

2.46
1.47

Cysteine
Phenylalanine

6.00

0.71
2.42

Tvrosine
Threonine

2.77

dextrins
were

1.42

4.62

1.93
4.23

10.68
1.53

6.78
2.31

2.77

2.16

Histidine

1.58
1.98

1.34
0.86

Proline
Serine

7.38
3.96

1.53
2.79

lsoleucine

3.69

1.93

acid
acid

and

analyzed in the first month and second month separately, we

found no difference in N balance during the first month
(F = 0.318, NS) while a statistically highly significant difference was present during the second month (F = 11.83,
P < ,005). Moreover, when the N balance values during the
first and second month were compared in the two groups
separately (N balance, first month v N balance second
month), we observed a highly significant difference (values
of N balance first month more negative then those during
second month) in group 1 (F = 14.77, P < .005) and a much
lesser, though still significant difference in group 2
(F = 5.12, P < .05). Figure 3 reports the N balance results in
each patient studied. The resulting picture shows a notable
individual variation. All subjects showed negative N balance
during the first month, while notably different patterns were
evident during the second month, during which few patients
in group 1 and only 1 in group 2 came into or very near to N
balance equilibrium; on the contrary, the majority of
patients, especially in group 2, presented persistently nega-

present.

Semistarvation

Lipids were mostly polyunsaturated

fats in the form of sunflower
or
peanut oil.
l Liquicl formula diet was administered as Nutroclin VLC, Midy S.p,A.,
Milan,

1.90
0.67

0.79
4.35

(F = 1.58, NS). On the contrary, when the weekly data were

60
300

3.00

Tryptophan

59
23.5

3,000

Methionine

17.6

200

Diets

3.17

Arginine
Glycine

175
18.5

Formula

6.46
5.40

0.8
28

150
20

of the 2 Liquid

Leucine
Lysine

Alanine

3.2
2.3

Content

Group 2

4
2

Acid

Group 1

0.34

60
50

IFluorine (mg)

Amino

Valine
Aspartic
Glutamic

Chromium (pg)
!Selenium (/.rg)
(pg)

3.

Amino acid
w

2t

2,092
41

5.0
54

(g)

Table

Diets

2,092
60

Lipids (g)
Carbohydrates
Minerals

1143

Group 1l

Energy
Kjoule (g)
Proteins

DIETS

Refeeding
L
T

Italy.

a Group

TLiquid formula diet was administered

(Wander Ltd, Milan, Italy).

as Modifast

Precision

4
1

BR

refeeding period a small decrease (- 1.3 k 1.Okg) in group 1

and no changes (0.1 k 1.7 kg) in group 2 were observed, but
these differences were not significant. Weekly mean values
of weight lossin both groups are depicted in Fig 1.

20*pco.o5

24-

Nitrogen

Balance

The weekly urine N excretion (g/d) and the N balance are

reported in Fig 2. The cumulative N balance during the
semistarvation period (0 to 56 days) was not significantly
(ANOVA) more negative in group 2 with respect to group 1

3
Weeks

5
of

Weekly
weight
loss during
Fig 1.
ing in the two groups of obese subjects

treatment
semistarvation
participating

and refeedin the study.

IO

PASQUALI,

t
9
Yii

GROUP
Patieals

GKOUP
Pnlieols

CC

CL

LV

PG

SE

SC

AND

BT

MELCHIONDA

A6

VM

AB

CS

BA

21

-IO-

III-III
0

130

80
Weeks

Individual
weekly
Fig 3.
each obese patient considered

3-

CASIMIRRI,

80
of

80

A0

semistarvation

in

treatment

N balance
during
in the study.

3-Methylhistidine
6-

In basal conditions, values of the urine excretion rate of

3-methylhistidine were similar in both groups. During semistarvation they fell significantly in group 1 while no change
was observed in group 2 and the ANOVA confirmed that a
significantly different (F = 6.89, P < .Ol) behavior of this
parameter existed between the two groups. Similar patterns
were observed when the 3-methylhistidine excretion was
corrected per gram of excreted creatinine (pmol/g creatinine/d) and for the calculated muscle protein breakdown

7.
8
9lo-

(g/d)

ll812345676

(Fig 4).

Serum and Urine Creatinine

Weeks

of

treatment

Fig 2.
Weekly
urine N excretion
(g/d) and N balance
semistarvation
in the two groups
of patients
participating
study.

during
in the

tive N balance values up until the eighth week of treatment,

There was a certain sex-dependent difference in the N
balance, at least in group 1. In fact, the males in this group
showed values of cumulative (0 to 56 days) N balance that
were more negative (-201.7 + 93.1 g) than those observed
in the females (74.0 * 13.9 g), while no difference was
present in the second group (231.9 + 86.8 and 201.2 + 64.6
g, respectively). A significant correlation was found between
the amount of weight lossand the cumulative (0 to 56 days)
N balance in the two groups considered together (r = ,562,
P = .051).

No significant variation in serum creatinine was found. On

the contrary, urine levels fell significantly with nadir values
occurring at the seventh week in both groups: group 1,1.53 k
0.46 v basal 2.17 2 0.68 g/d (P < .OS)and group 2; 1.57 k
0.50 v basal 2.01 + 0.45 g/d (P < .05). Creatinine clearance
fell significantly in both groups, with no differences between
them; in spite of this, nadir values (12 1.1 + 3 1.5 mL/min in
group 1 and 101.3 + 11.6 ml/min in group 2, NS) remained
within normal range values. Refeeding increased values of
urine creatinine modestly but not significantly.
Serum Proteins

No significant variation occurred in serum concentrations

of total proteins, albumins or transferrin, although the values
of the latter protein seemed to decreaseprogressively, especially during the second month (Table 4). RPB, TBG, and

PROTEIN

METABOLISM

DURING

LOW-CALORIE

DIETS

complement-C3 fell significantly in both groups during

semistarvation, without further increaseduring the refeeding
period. On the contrary, PA (prealbumin) levels fell significantly only in group 1 while no changes occurred throughout
the study in group 2, except for the values at the secondweek,
which were significantly lower with respect to basal values.
In terms of per cent variations transferrin, RBP, TBG, and
PA showed similar variations while in complement-C3 a
lesser variation was seen in group 2 with respect to group 1.
Refeeding did not induce any increment in blood concentrations of all proteins in either group, which persisted
unchanged with respect to the values observed during the
second month of semistarvation.
DISCUSSION

Numerous studies carried out recently have reported that

N equilibrium could be achieved within a few weeks in the
majority of obese patients treated with VLCD of different
types and formulations, irregardless of their initial body
weight, fatty mass, and lean body mass.24*6-8*10*3,27
However,
it should be considered that there are at least two factors
against the possibility of N lossesbeing homogenous in obese
subjects. First, basal protein requirements differ substantially between individuals; second, the majority of obese
subjects have variably increased lean body mass, which is
also related to energy intake.*Our data confirm that there is
in fact a notable variability between individual subjects in N
sparing adaptive mechanisms, and also, they show that only
few patients were able to attain N equilibrium throughout
the two-month diet period. Similar results were also obtained
by Yang et al who studied a group of obese subjects very
similar to ours and who were administered a diet containing
58 g of carbohydrates and 66 g of proteins for 64 days.
Moreover, Fisler et al* failed to obtain N equilibrium in
most of the obese subjects who were fed a carbohydrate-free
VLCD with a protein content ranging from 50 to 120 g/d for
two months. We found a certain relation between the amount
of weight lossand the cumulative (0 to 56 days) N balance;
the more weight was lost the more persistently negative was
the N balance. Since the loss of weight is generally proportional to the initial weight as well as to the energy deficit
induced by the nutritional treatment, the possible role of
these factors must be taken into account in explaining, at
least in part, the variability of the adaptive responsesin the
protein sparing mechanisms. On the other hand, they cannot
explain the differences found between the two groups in our
study. It would seem rather that the inadequate protein
intake was probably the most important factor leading to the
greater N loss observed in group 2. However, these differences with respect to group 1 became evident only after 4
weeks of treatment, thus suggesting that the adaptive phenomena occurring during the early phase of the treatment
with this type of diet are relatively independent of the
specific:nutrient intake and may possibly involve the dissipation of the so-called labile N po01.~These results, therefore, confirm previous reports which suggest that the early
cumulative N loss is not very useful in predicting long-term
lossduring semistarvation in obesity.3 Other factors should,
however, also be considered, which could justify the differ-

1145

.-5

Group

600~

150,

loo-

o61

2345676

Weeks
Fig 4.
3-Methylhistidine
muscle
protein
breakdown
participating
in the study.
P < .05.

of

12345676-

treatment

daily urine excretion

and calculated
in the two groups
of obese patients
*Significantly
different
Y basal values,

ence in N balance between the two groups. It is well known

that women maintain N balance more readily than men.**
Our experimental groups were closely matched for sex
distribution; however, while in group 1 the subjects who
showed a better N sparing capacity were women, this
behavior was not confirmed in group 2, due in particular to
the patient CL, who was on the other hand characterized by
an android type body fat distribution and by modest hirsutism without increased androgen plasma levels. Despite
these differences, it is probable that the N balance results in
the two groups depend only to a small extent on sex-linked
factors. The adequacy of potassium intake could also affect
N 10~s.~~
Potassium intake in group 2 was significantly lower
than in group 1, but no patient had a significant reduction in
serum potassium values (data not shown). It therefore seems
unlikely that inadequate potassium supplementation may
have played some role in determining the excessivelossof N
observed in group 2. Finally, it is improbable that the
differences observed may be dependent on the evaluation of
fecal, integumentary, and miscellaneous lossesof N, which
we considered constant (1.25 g/d) in calculating the N
balance,*j also on the basis of what is reported in several
balanced studies, which failed to observe significantly dif-

ComplementCB

ImslcLl

as rmB

l P -5 .05 Y basal VdIuos.

,P < .05 between
Qoups.

are given

(mQ/b)

wws

PA

hQ/ct)

TBG

lmg/bl

lmslal

RBP

Trsndmin

353.4

+ SD

28.8

27.9

2.6

2.6

6.5

5.6

96.6

107.0

347.2

+ 89.9

* 7.5

+ 4.8

* 0.3

* 0.3

+ 1.8

+ 1.4

e 12.6

+ 17.3

+ 63.4

346.8

26.2

22.8

2.3

2.5

6.4

4.9

92.8

95.8

350.6
12.1.
9.6

1.5

+ 6.6

f 5.1.

+_ 0.3

* 0.2

+ 1.2

k 48.3

+ 79.8

Table

4.

21.4

20.3

2.2

2.4

4.6

4.2

92.4

89.0

333.2

373.8

Protein

+ 9.8

+ 3.6.

t 0.5

+ 0.2

+ 1.7.

+ 0.7

L 7.6

+ 12.7

+ 69.8

2 41.2

Serum

26.3

22.0

2.2

2.1

5.4

4.0

95.2

87.9

328.6

324.2

19.3.

1.6

+ 8.1

e 6.3.

+ 0.6

+ 0.2.

+ 0.7.

+ 6.6

+ 46.4

t 68.8

Variation

During

26.0

20.7

2.1

2.3

6.2

4.0

88.2

90.0

354.0

346.6

1.3.

+ 7.1

+ 4.1.

+ 0.4

+ 0.2.

* 0.8.

+ 10.7

+ 26.2.

+ 92.2

5 75.2

Semistarvation

314.6

26.7

23.1

2.3

2.2

4.9

4.6

88.6

90.8

260.8

64.2.

e 7.0

+ 5.0.

e 0.6

f 0.3.

+ 1.1.

+ 0.8.

+ 15.4

+ 31.6.

* 47.6

(Week

l-8)

24.7

23.1

2.0

2.3

4.7

3.6

83.2

81.2

314.6

325.2

1.8

+ 6.3

+ 7.6.

+ 0.4

f 0.4

+ 1.2.

f 9.7.

2 24.4.

e 88.1

+ 81.3

and Refeeding

26.6

24.1

2.2

2.3

4.7

3.9

79.6

76.4

304.4

310.6

in the

,.I

7.1

6.2.

* 0.4.

f 0.4.

f 0.9.

2 9.9.

+ 21.3.

+ 98.8

? 47.4

2 Groups

289.2

25.4

22.9

2.2

2.2

4.9

4.1

60.6

73.0

333.2

10.1.

,.3*+

7.2

+ 5.4.

* 0.4.

* 0.1.

t 0.7.

5 20.4.

97.5

* 36.6
f

328.4

25.7

26.2

2.2

2.2

5.2

4.2

80.4

76.4

315.0

e 108.8

k 7.0

+ 4.0

f 0.3.

2 0.4

+ 1.0.

+ 0.8.

+ 7.9.

+ 21.2.

+ 47.1

290.0

27.6

23.7

2.3

2.3

5.1

4.4

81.0

77.0

297.2

+ 73.5

+ 4.8

+ 5.0.

k 0.2.

e 0.1

+ 1.4.

e 0.6.

e 10.9.

+ 21.1*

f 66.1

PROTEIN

METABOLISM

DURING

LOW-CALORIE

1147

DIETS

ferent effects dependent on the specific composition

of
different VLCDS.*~~
The N balance which is lost may have various origins that
are difficult to define. Although this problem remains controversial,34 several arguments seem to indicate that the
urinary excretion rate of 3-methylhistidine
seems to be a
valid index of the rate of skeletal muscle protein breakdown.*.* Fasting decreases 3-methylhistidine
excretion,5,6
and similar findings may be observed with zero-carbohydrate
VLCDS.~, The results with both diets in the present study
suggest that, unlike group 1, nonadaptive changes probably
occurred throughout the study in group 2. Previously, Garlick et aP5 observed no reduction in 3-methylhistidine
excretion in obese subjects fed with a 500 kcal, 50 g protein, 75 g
carbohydrate diet, or one providing 500 kcal glucose only. On
the contrary, Hoffer et al33 found a similar fall (25% to 30%)
of 3-lmethylhistidine
in two groups of moderately
obese
subjects treated with diets providing 1.5 g of proteins or 0.8 g
proteins + 0.7 g carbohydrates/kg
IBW. Garlick et al35 also
measured whole body protein breakdown
with different
tracer techniques;
despite no changes in the rate of
3-methylhistidine
excretion, they found the breakdown rate
markedly reduced during zero-protein diet and minimally
affected during the mixed diet. Therefore,
considered
together, these data seem to support the hypothesis that
urinary 3-methylhistidine
excretion rate during hypocaloric
feeding may also be determined by the carbohydrate intake
other than by the protein and absolute energy intake.
Blood protein concentrations showed some differences in
behavior between the two groups. We found a more or less
identical reduction in the concentrations
of RBP and of
TBG, while PA fell persistently in group 1 and transiently in
group 2; complement-C3 seemed to undergo a greater reduction in group 1 with respect to group 2 and transferrin failed
to fall significantly
in either group. Recent studies have

suggested that dietary carbohydrates apparently modulate

serum concentrations of some of these proteins irregardless
of caloric intake.36.37 These conclusions were not confirmed
by Hoffer et a13 who observed similar declines in RBP and
PA in obese subjects following VLCD providing proteins
alone (1.5 g/kg IBW) or protein (0.8 g) plus carbohydrates
(0.7 g/kg IBW). Our data show that not even 81 g of
carbohydrates would prevent the fall of RBP and TBG levels
while this amount was probably responsible for the minor
reduction observed in the levels of complement-C3
and PA.
The brief period of refeeding was unable to induce a significant variation in any of the proteins with respect to the values
observed at the end of the semistarvation period. This was
probably due to the fact that during refeeding, we made only
a slight increase in energy intake (500 kcal/d) almost
entirely in the form of lipids. Our study would therefore seem
to confirm that carbohydrates,
as well as proteins and
calories, are related to serum protein homeostasis; the mechanisms by which this comes about are unknown, but a
decrease in the synthetic rate of these proteins could be
suggested.36 Fisler et al* were able to observe a significant
relationship between complement-C3 levels and total N loss
in obese subjects fed approximately
400 kcal/d as sole
proteins. Their conclusions do not seem to be applicable to
VLC-mixed diets, since we found that a greater N loss was
associated with a lower reduction of the levels of this protein,
as occurred in group 2 with respect to group 1. We therefore
suggest that it is improbable that changes in serum complementC3 may represent a reliable index of protein depletion
in obese subjects consuming VLC-mixed diets.
ACKNOWLEDGMENT

We thank all the nursing staff for their cooperation

during the
study
(Anna Bacialli, Franca Tassi) and Sandro Piazzi for
3-methylhistidine

measurements.

REFERENCES

Vertes V: Weight reduction in obesity

JAMA 230:987-991, 1974
2. McLean Baird I, Parsons RL, Howard AN: Clinical and

I. Genuth
by outpatient

SM. Castro JH,

semistarvation.

metabolic
studies of chemically
defined diets in the management
of
obesity. Metabolism
23:645-647,
1974
3. Bistrian
BR, Winterer
J, Blackburn
JL, et al: Effect of a
protein-sparing
diet and brief fast on nitrogen metabolism
in mildly
obese subjects. J Lab Clin Med 89:1030-1035,
1977
4. .4pfelbaum
M, Baigts F, Giachetti
I, et al: Effects of a high
protein very-low-energy
diet on ambulatory
subjects with a special
reference
to nitrogen balance. Int J Obes 5: 117- 130, 198 1
5. Marliss EB, Murray
FT, Nakooda
AF: The metabolic
response
to hypocaloric
protein diets in obese man. J Clin Invest 62:468-479,
1978
6. Howard
AN: The historical
development,
efficacy and safety
of very-low-calorie
diets. Int J Obes 5:195-208,
1981
7. Wilson JH, Lamberts
SWJ: Nitrogen
balance in obese patients
receiving
a very low calorie liquid formula
diet. Am J Clin Nutr
32:1612-1616,
1979
8. Ditschuneit
H, Ditschuneit
HH, Wechsler
JG: Adipositasbehandlung-NulldiLt
oder kalorienreduzierte
Diat. Internist
20: 15 l158.1979
9. Contaldo
F, Di Biase G, Fischetti
A, et al: Evaluation
of the

safety of very-low-calorie
diets in the treatment
of severely obese
patients in a metabolic
ward. Int J Obes 5221-226,
1981
10. DeHaven
J, Sherwin
R, Hendler
R, et al: Nitrogen
and
sodium balance and sympathetic-nervous-system
activity
in obese
subjects treated with a low-calorie
protein or mixed diet. N Engl J
Med 302:477-482,
1980
11. Yang MU, Barbosa-Saldivar
JL, Pi-Sunger
FX, et al: Metabolic effects of substituting
carbohydrate
for protein in a low-calorie
diet: A prolonged
study in obese patients.
Int J Obes 5:231-236,
1981
12. Fisler JS, Drenick EJ, Blumfield DE, et al: Nitrogen
economy
during very low calorie reducing diet: quality and quantity
of dietary
protein. Am J Clin Nutr 35:471-486,
1982
13. Davie M, Abraham
RR, Godsland
I, et al: Effect of high and
low-carbohydrate
diets on nitrogen balance during calorie restriction
in obese subjects. Int J Obes 6:457-462,
1982
14. Felig PH: Very-low-calorie
protein
diet. N Engl J Med
310:589-591,
1984
15. Van Gaal LF, Snyders D, De Leeuw IH, et al: Anthropometric and calorimetric
evidence for the protein sparing effects of a new
protein
supplemented
low calorie preparation.
Am J Clin Nutr
41:540-544,
1985
16. Hoffer LJ, Bistrian BR, Young VR, et al: Metabolic
effects of

1148

very low calorie weight reduction diets. J Clin Invest 73:750-758,

1974
17. Winterer J, Bistrian BR, Blimazes C, et al: Whole body
protein turnover studied with Nglycine, and muscle protein breakdown in mildly obese subjects during a protein-sparing diet and a
brief total fast. Metabolism 29:575-581, 1980
18. Wynn V, Abraham RR, Demsen JW: Method for estimating
rate of fat loss during treatment of obesity by caloric restriction.
Lancet 2:482-486, 1985
19. Bistrian BR, Sherman M, Young V: The mechanisms of
nitrogen sparing in fasting supplemented by protein and carbohydrate. J Clin Endocrinol Metab 53:874-878, 1981
20. Young VR, Munro HN: N-methylhistidine (3-Methylhistidine) and muscle protein turnover: a review. Fed Proc 37:2291-2300,
1978
21. Ballard FJ, Tomas FM: 3-Methylhistidine as a measure of
skeletal muscle protein breakdown in human subjects: The case for
its continued use. Clin Sci 65:209-215, 1983
22. Mazzella G, Pasquali R, Aldini R, et al: EITects of caloric
intake restriction and ursodeoxycholic acid administration on bile
lipids in obese patients, in Berchtold P, Cairella M, Jacobelli A, et al
(eds): Regulators of Intestinal Absorption in Obesity, Diabetes and
Nutrition. Rome, Sot Ed Universo, 1981, pp 61-65
23. Williamson DL, Mellanby I, Krebs HA: Enzymatic determination of D( -)-B
hydroxybutyric acid and acetoacetic acid in
blood. Biochem J 82:90-96, 1962
24. Lukaski H, Mendez J: Relationship between fat-free weight
and urinary 3-Methylhistidine excretion in man. Metabolism
29:758-761, 1980
25. Munro HN, Fleck A: Analysis of tissues and body fluid for
nitrogenous constituents, in Munro HN (ed): Mammalian Protein
Metabolism, ~013. Orlando, FL, Academic, 1960, pp 424-525
26. Food and Agriculture Organization/World Health Organization: Report of a joint FAO/WHO ad hoc expert committee.
Geneve, WHO Tech Rep Series No 522,1973

PASQUALI,

CASIMIRRI,

AND

MELCHIONDA

27. Sheen AJ, Luyckx AS, Scheen-Lavigne MC, et al: Hormonal

and metabolic adaptation to protein-supplemented fasting in obese
subjects. Int J Obes 6:165-174, 1982
28. Committee on Dietary Allowances: Food and Nutrition
Board. Recommended Dietary Allowances. National Academy of
Sciences, Washington, DC, 1980
29. Forbes GB, Welle SL: Lean body mass in obesity. Int J Obes
7:99-107, 1983
30. Allison JB, Wannemacher RW Jr: The concept and significance of labile and overall protein reserves of the body. Am J Clin
Nutr 16:445-452, 1965
31. Yang MU, Van Itallie T: Variability in body protein loss
during protracted severe caloric restriction: Role of triiodothyronine
and other possible determinants. Am J Clin Nutr 40:61 l-622, 1984
32. Sapir DG, Chambers NE, Ryan JW: The role of potassium in
the control of ammonium excretion during starvation. Metabolism
26:211-220, 1976
33. Hoffer LJ, Bistrian BR, Young VR, et al: Metabolic effects of
carbohydrate in low-calorie diets. Metabolism 33:820-825, 1984
34. Rennie MJ, Millward DJ: 3-methylhistidine excretion and
the urinary 3-methylhistidine/creatinine ratio are poor indicators of
skeletal muscle protein breakdown. Clin Sci 65:217-225, 1983
35. Garlick PJ, Glugston GA, Waterlow JC: Influence of lowenergy diets on whole-body protein turnover in obese subjects. Am J
Physiol238:E235-E244,1980
36. Kelleher PC, Phinney SD, Sims EAH, et al: Etfects of
carbohydrate-containing and carbohydrate-restricted hypocaloric
and eucaloric diets on serum concentrations of retinol-binding
protein, thyroxin-binding prealbumin and transferrin. Metabolism
32:95-101, 1983
37. Bogardus C, Lagrange BM, Horton ES: Metabolic fuels and
the capacity for exercise during hypocaloric diets. Int J Obes
5:295-296, 1981