Clinical Science (1988)75, 101-107
101
Elevated protein requirements in cirrhosis of the liver
investigated by whole body protein turnover studies
G. R. SWART, J. W. 0. VAN DEN BERG, J. L. D. WATTIMENA,
T. RIETVELD, J. K. VAN VUuRE AND M. FRENKEL
Department of Internal Medicine 11, University Hospital Dijkzigt, Rotterdam, The Netherlands
(Received 16 July/l4 December 1987; accepted 13 January 1988)
SUMMARY
1. In patients with cirrhosis of the liver and in healthy
control subjects, the rates of nitrogen flux, protein
synthesis and protein breakdown were studied, using a
single oral dose of 200 mg of [lsNN]glycineas a tracer. The
nitrogen flux through the amino acid pool was measured
separately with both urinary ammonia and urinary urea as
end products; the average value was used for further calculations.
2. Subjects were studied in the fed state, both on an
adequate and a protein-restricted diet, and also in the
fasting state.
3. The rates of protein synthesis were markedly
increased in the patients, not only in the fed but also in the
fasting state. Protein breakdown rates were increased in
the patients in the fed state.
4. The nitrogen balance in steady-state conditions in
the fed state was more positive in the patients, while their
nitrogen loss in the fasting state was no higher than that of
control subjects.
5. A hypothesis is put forward that the high protein
requirements of cirrhotic patients could be caused by
small and inadequate liver glycogen stores; due to these
small stores, gluconeogenesis from amino acids will take
place and lead to an extra amino acid loss even during
short-term fasting. This increased amino acid loss could
explain the elevated protein requirements in cirrhotic
patients.
Key words: cirrhosis, liver, protein requirement, protein
turnover, whole body.
([l, 21; G. R. Swart, unpublished work). This would
necessitate administration of protein-enriched diets.
Unfortunately, protein-restricted diets are sometimes
needed to prevent or to treat porto-systemic encephalopathy. Moreover, in decompensated cirrhosis appetite
and food intake can be diminished. Thus, metabolic needs
will not be met and indeed protein-energy malnutrition is
frequently found in cirrhotic patients [3, 41. It is not
known why more protein is required, although both an
increase in protein catabolism and a low rate of protein
synthesis could explain the elevated requirements. Until
recently [2], protein metabolism in cirrhotic patients has
been studied mainly with conventional nitrogen balance
techniques [ 1, 51 that give no information on the rates of
protein synthesis and protein breakdown. Moreover,
these nitrogen balances are calculated over 24 h periods.
Nutritionally, such 24 h periods are discontinuous: the
nitrogen balance will fluctuate from negative, during postabsorptive hours, to positive in the fed state. The overall
24 h nitrogen balance therefore is determined by the rate
of nitrogen loss or gain under the various dietary conditions, and by the time spent in either the fed or fasting
state.
In recent years, whole body protein turnover measurements, applicable under clinical conditions, have become
available [6]. We studied protein turnover in patients with
cirrhosis of the liver and in healthy control subjects, using
['5N]glycine as a tracer. Studies were performed in the fed
state, on both adequate and protein-restricted diets, and
also while the subjects were fasting. The aim of these
studies was to explain the high protein requirements in
cirrhotic patients.
INTRODUCTION
In cirrhosis of the liver, protein requirements to maintain
nitrogen equilibrium have been reported to be elevated
Correspondence: Dr G. R. Swart, Department of Internal
Medicine 11, University Hospital Dijkzigt, 40 Dr. Molewaterplein, 3015 GD Rotterdam, The Netherlands.
MATERIALS AND METHODS
Subjects
Sixteen patients with a biopsy-proven cirrhosis of the
liver were studied (Table 1). Alcoholic cirrhosis was
102
G. R. Swart et al.
present in nine patients, cryptogenic cirrhosis in five,
while in two patients autoimmune chronic active hepatitis
was the cause. The mean age of the patients was 55 years
(range 31-73 years). There were nine male and six female
patients. The patients were studied after they had been in
the hospital for 2 or 3 weeks for treatment of complications of their cirrhosis or for unrelated causes. All were in
a stable clinical condition at the time of the study. There
had been no alcohol intake recently, there was no overt
porto-systemic encephalopathy and no oedema or ascites;
renal function was normal. The Quetelet index was in the
normal range or even elevated, compatible with ample fat
stores, yet most patients had clinical signs of muscle
wasting. The diet in the weeks preceding the turnover
studies had not been standardized; some patients had
taken the regular hospital diet; none had been on a diet
with less than 60 g of protein/day. Some patients were
using diuretics and/or lactulose, none was on prednisone.
The control group consisted of 12 healthy subjects, six
male and six female. Their mean age was 48 years (range
29-68 years). The research protocol had been approved
by the ethical committee of the University Hospital.
Diets
Protein metabolism was studied in three metabolic situations: in the fed state, both on an 'adequate' diet and on a
protein-restricted diet, and also in the fasting state. For
practical reasons and to ensure identical hourly meals, the
diet was given as a (commercially available) liquid formula
diet (Nutrison R Nutricia, Zoetermeer, The Netherlands).
This product contains 4.0 g of milk protein, 4.0 g of
vegetable fat and 11.6 g of carbohydrate (mainly maltose)
per 100 ml. Per g of protein, it contains 104 kJ; 2000 ml
of it will supply the recommended daily allowances of
electrolytes and vitamins. An adequate diet was defined
as a diet providing 1 g of natural protein and 104 kJ day-'
kg-l body weight. The protein-restricted diet supplied
0.5 g of protein and 104 kJ day-' kg-l body weight. This
diet was obtained by diluting the liquid formula with a
solution of polymerized glucose ('Ca1400' R Roussel,
Hoevelaken, The Netherlands). In the fasting state, only
tap water was allowed.
Design of the study
During the studies, the patients were in a steady metabolic state. The diet was divided into l l identical hourly
portions; it started at 07.00 h and was continued until
17.00 h. Steady-state conditions in the fed state were
assumed to be reached 2 h after the first liquid meal. The
protein turnover measurements were performed over a
9 h period and took place between 09.00 h and 18.00 h.
During these 9 h the patients were up and about; no extra
physical exercise was performed. The fasting state was
assumed to be reached 6 h after the last meal of the day,
which was taken at 18.00 h. The fasting state studies were
thus performed with the patient asleep in bed, between
24.00 h and 09.00 h.
Whole body protein turnover in liver cirrhosis
Protein turnover measurements
Protein turnover was studied with the two-compartment model of Picou & Taylor-Roberts [7], using ISN as
the isotope to label the amino acid pool while urinary
nitrogen excretion products were used to measure the
nitrogen flux.
We performed protein turnover measurements by
giving a single oral dose of isotope, 200 mg of [‘SN]glycine
99% (KOR Chemical, Boston, U.S.A.) [S]. Two urinary
end products (ammonia and urea) were used to measure
nitrogen flux, always in duplicate, and the average value of
nitrogen flux measured with each of these two end products separately was calculated [9, 101. Using ammonia as
the end product, a collection period of 9 h would be
appropriate in healthy subjects [9].We verified this in control subjects and in four of our cirrhotic patients by
measuring the cumulative ISN excretion in urinary ammonia (Fig. 1) that was found to be completed after 9 h.
However, to ensure complete collection of [ 15N]urea,we
used another approach [ 101. Urinary [L5N]ureaenrichment was determined in a 9 h urine collection. After 9 h
there still was a considerable amount of the [15N]urea
present in the body urea pool, due to be excreted. The
isotope enrichment in body urea after the 9 h of the study
was assumed to be identical with that in an urinary sample
taken immediately after completion of the study. Knowing
this [ISN]ureaenrichment one can calculate, with the aid
of estimated body water [ l l ] , the amount of [lsN]ureastill
present in the body. This calculated quantity of urea
added to the [15N]ureaactually excreted during the test
gives the total quantity of urea produced from the
[lSN]glycinepulse. The sensitivity of our mass spectrometry system precluded direct determination of [ISN]urea
in plasma. Ammonia was isolated from the urine by distil-
0.6
0.5
I-
103
lation at pH 12; in urine, thus made ammonia-free, urea
was converted to ammonia by urease and this ammonia
was then also collected by distillation. Ammonia-nitrogen
was measured in an automatic nitrogen analyser (ANA
1400; Carlo Erba, Milano, Italy) by the Dumas procedure. The nitrogen gas leaving the nitrogen analyser
was directly led into a quadrupole type mass spectrometer
(200 MGA; Centronics, New Addington, Croydon, U.K.).
From the relative amounts of nitrogen isotopes present at
mass 28 and 29, the enrichment of the sample was calculated. The urinary nitrogen balance during the 9 h of the
study could be calculated from the calculated dietary
nitrogen intake and the measured urinary nitrogen loss.
Due to the short collection periods, faecal nitrogen loss
could not be taken into account.
Presentation of data
All data (rates of nitrogen flux, protein synthesis,
protein breakdown and also the nitrogen balances) are
expressed as mg of N 9 h-l kg-l. Only the average values
of these nitrogen turnover measurements (measured and
calculated separately in duplicate with both urea and
ammonia as the end products) are presented. Data for
groups of patients or controls are presented in Fig. 2 as
median values together with the first and third quartiles.
For statistical analyses, the unpaired rank correlation test
according to Wilcoxon was used, unless stated otherwise.
RESULTS
Control of test procedures
It was not possible to test all three diets in each patient
and control subject; thus there are different subgroups.
However, there were no statistically significant differences
in weight and age within the subgroups of patients or control subjects (Student’s t-test). The precision of the total
analytical procedure (diet, urine collection, chemical
analyses) was tested in five individuals where the whole
procedure was carried out twice. The coefficient of variation for nitrogen flux measurements was 19%, for protein
synthesis 1lo& and for breakdown 14%.The coefficient of
variation of repeated nitrogen determinations in our automatic analyser ( IZ = 9 ) was 1%;the absolute values were
within 3% of values found in Kjeldahl analyses. Between
repeated measurements in our mass spectrometer ( n= 9)
a coefficient of variation of 1.1-1.3% was found.
Control subjects
3
6
9
Urine collection time ( h )
12
Fig. 1. Excretion pattern of ISNH, in healthy controls (0)
and patients with cirrhosis ( 0 ) .Cumulative values of mg
of I5N excreted are presented as a function of the urine
collection time.
From the comparisons in Table 2 it follows that in the
fed state, on an adequate diet, the rate of protein synthesis
was higher than the rate of protein breakdown (213 vs
170 mg of N 9 h-l kg-I). The nitrogen balance was therefore positive ( + 41 mg of N 9 h-l kg-I). On a protein-restricted diet, the rate of protein synthesis fell, while there
was no change in the rate of protein breakdown. The
nitrogen balance almost reached equilibrium ( + 3 mg of
N 9 h-l kg-I). The nitrogen flux diminished on this diet. In
104
G. R. Swart et al.
0
0
0
++
0
0
8
00
a
q2. 8 :
..
0.5
'
..
1
Diet (g of protein/kg body weight)
Fig. 2. Nitrogen flux ( Q ) , protein synthesis (S), and protein breakdown ( B ) in patients with
and healthy controls (.):Bars
denote median values for each group.
cirrhosis (0)
the fasting state, the nitrogen flux was markedly lower
than it was on an adequate diet. The rate of protein
synthesis was lower than on an adequate diet, although it
was statistically not different from the rate of protein
synthesis on a restricted diet (163 vs 191 mg of N 9 h-I
kg-l, NS). In contrast, however, the rate of protein breakdown in the fasting state was higher compared with the
fed state both on an adequate diet and on a protein-restricted diet (218 vs 170 and 183 mg of N 9 h-' kg-l
respectively). Because the rate of protein breakdown now
exceeded that of protein synthesis, the nitrogen balance
was negative in this situation ( - 53 mg of N 9 h-' kg-I).
Thus, when protein intake was restricted, protein synthesis declined without significant change in breakdown.
Only during fasting was the diminished protein synthesis
accompanied by an increase in protein breakdown.
Patients
The comparisons for patients as presented in Table 2
show that in the fed state on an adequate diet, the patients
had a rate of protein synthesis that was higher than the
rate of protein breakdown (294 vs 228 mg of N 9 h-l
kg-I). Their nitrogen balances were positive ( + 76 mg of
N 9 h-l kg-'). On protein restriction no difference could
be detected in nitrogen flux, in the rate of protein synthesis or in the rate of protein breakdown. The nitrogen
balance remained positive, although less positive than on
an adequate diet ( + 24 vs + 76 mg of N 9 h-l kg-I). In the
fasting state, there was no statistically significant decrease
in nitrogen flux and in the rate of protein synthesis and no
significant increase in the rate of protein breakdown.
Nitrogen balances now were negative ( - 42 mg of N 9 h-l
kg-I ). It was considered that this lack of significance in the
observed changes might be due to inhomogeneity of the
patient group. Thus, a paired Wilcoxon test was per-
formed in a subgroup ( n= 11) of these patients in whom
both the adequate diet and the fasting.state had been
studied. In this subgroup, in the fasting state, there was a
decrease in nitrogen flux [244 (202/353) vs 371 (325/
416) mg of N 9 h-l kg-l, P<O.O5] and a decrease in the
rate of protein synthesis [213 (166/291) vs 295 (277/385)
mg of N 9 h-l kg-', P<O.Ol]. Again, no difference was
found in the rate of protein breakdown [244 (202/353) vs
240 (194/289) mg of N 9 h-l kg-l, NS].
Comparison of patients with control subjects
Table 3 shows that on an adequate diet, the cirrhotic
patients had a higher nitrogen flux than the control subjects. The rate of protein synthesis was higher than in control subjects, while the rate of protein breakdown was also
increased. The nitrogen balance was more positive in the
cirrhotic patients than it was in the control subjects [ + 76
( + 5 3 / + 100) vs + 4 1 ( + 3 5 / + 5 0 ) mg of N 9 h-I kg-l,
P<O.Ol]. On protein restriction the nitrogen flux in the
patient group still exceeded that in the control subjects.
Their rate of protein synthesis was higher than it was in
the control subjects and the rate of protein breakdown
was above that in the control group. The nitrogen balance
on such a restricted diet was still markedly positive in the
cirrhotic patients, in contrast with the nitrogen balance in
the control group, which was almost in equilibrium [ + 24
( + 1 4 / + 2 8 ) vs + 3 ( - 1 1 / + 6 ) mg of N 9 h-l kg-l,
P < 0.011. In the fasting state, no significant differences
were found between the patients and the control subjects
with regard to the nitrogen flux or the rate of protein breakdown. However, in the patients, the rate of protein
synthesis still exceeded that in control subjects, yet there
was no difference in nitrogen balance under these circumstances[-42(-54/-31)vs -53 ( - 6 1 / - 4 5 ) m g o f N
9 h-I kg-l, NS]. In the patients, there was a greater varia-
218
(183/231)
163
(137/170)
218
(183/231)
- 53
(-61/-45)
None
**
*
**
**
~
~
300
(265/315)
213
(172/241)
170
(130/182)
+41
( + 35/ + 50)
1g
*
NS
*
**
248
(208/259)
191
(117/196)
183
(143/193)
+3
(-11/+6)
0.5 g
- 42
(176/370)
(-54/-31)
262
(176/370)
23 1
(163/316)
262
None
**
NS
NS
NS
360
(310/414)
294
(267/378)
228
(178/287)
+ 76
(+53/100)
1g
Patients
**
NS
NS
NS
Nitrogen balance (mg of N 9 h-' kg-I)
Protein breakdown (rng of N 9 h-l kg-I)
Protein synthesis (mg of N 9 h'' kg-l)
Nitrogen flux (mg of N 9 h-' kg-I)
360
(310/414)
294
(267/378)
228
(178/287)
+ 76
( + 5 3 / + 100)
Patients
**
*
**
*
300
(265/315)
213
(172/241)
170
(130/182)
+41
(+35/+50)
Controls
1 g of proteinlday
334
(263/359)
311
(218/331)
278
(197/303)
+ 24
(+14/+28)
Patients
**
**
**
**
248
(208/259)
191
(117/196)
183
(143/193)
+3
(-11/+6)
Controls
0.5 g of proteinlday
262
(176/370)
23 1
(163/316)
262
(176/370)
- 42
(-54/-31)
Patients
NS
NS
*
NS
0.5 g
218
(183/231)
163
( 137/170)
218
(183/231)
- 53
(-61/-45)
Controls
334
(263/359)
31 1
(218/331)
278
(197/303)
+ 24
(+14/+28)
No protein
Table 3. Comparison of protein synthesis and breakdown and nitrogen flux and balance in patients and controls under three dietary conditions
Statistical significance:*P< 0.05, **P< 0.0 1. Values in parentheses are first and third percentiles. Abbreviation: NS, not significant.
Nitrogen balance (mg of N 9 h-l kg-l)
Protein breakdown (mg of N 9 h-' kg-I)
Protein synthesis (mg of N 9 h-l kg-l)
Nitrogen flux (mg of N 9 h-' kg-l)
~~~~~~
Protein (g/day)...
Controls
Table 2. Nitrogen balance, nitrogen flux, protein synthesis and protein breakdown in patients and controls on diets containing no protein, 0.5 g or 1 g of proteidday
Statistical significance:*P< 0.05, **P< 0.0 1.Values in parentheses are first and third percentiles. Abbreviation: NS, not significant.
$
v1
g.
4
a.
1
2
<
Y
!
i
2
G. R. Swart et al.
106
tion in the results than in the control group, especially on
the protein-restricted diet and in the fasting state.
DISCUSSION
The aim of our study was to investigate why protein
requirements are elevated in cirrhotic patients. This was
done by performing protein turnover measurements
under different dietary conditions. Protein turnover was
measured with a nitrogen isotope, using urinary ammonia
and urea isotope enrichment as the measured variable.
The validity of this technique in cirrhotic patients could
be questioned; however we believe that it is valid. Indeed,
urea formation is a function of the liver and in cirrhotic
patients, the rate of urea production can be diminished.
However, for protein turnover measurements, the amount
of nitrogen converted to urea and excreted during the 9 h
of the test is not essential as long as the assumption basic
to the method holds, i.e. that there is no preferential use of
I5N for anv route of nitrogen excretion. There is no
reason to believe that this would be the case in cirrhotic
patients.
In this study, marked differences in protein metabolism
were found between cirrhotic patients and control subjects. Whole body protein synthesis was higher in the
patients under the different dietary circumstances. Moreover, it did not change when protein intake was reduced.
In our control group, the rate of protein synthesis fell on a
restricted diet and even more so during fasting, as would
be expected [12, 131. In the cirrhotic patients, this effect
did not occur, neither on the proteinirestricted diet nor
during fasting. O’Keefe et al. [2] also described increased
rates of protein synthesis in the fasting state in cirrhotic
patients. Mullen et at. [14], however, found that rates of
protein synthesis were not significantly different in fasting
cirrhotic patients, while Millikan et at. [ 151 found that
protein synthesis in cirrhotic patients and control subjects
did not differ between those groups whether fasting or in
the fed state. Different mechanisms may be postulated to
explain the increased rates in our study. In conventional
nitrogen balances, an identical diet will induce a greater
nitrogen retention in a protein-malnourished patient than
in a healthy subject [ 16, 171. This phenomenon has also
been shown by determining whole body protein turnover
[ 171. Thus the increased rates of protein turnover in the
cirrhotic patients might be a consequence of giving a
nutritious diet to patients with protein malnutrition.
Another mechanism might be that elevated protein breakdown rates in cirrhotic patients will supply more free
amino acids that may stimulate protein synthesis. Thus,
the increased rates of protein synthesis could be the
consequence of primarily elevated rates of protein breakdown. Increased rates of protein breakdown concomitant
with elevated rates of protein synthesis have been
described in post-operative septic patients [ 181.
In the control group we observed constant rates of
protein breakdown on either diet in the fed state. However, we found slightly increased rates of protein breakdown in the fasting state. In healthy subjects [19] and in
cirrhotic patients [ 141 constant rates of protein break-
-
down over the 24 h of the day have been described. In our
patient group we did not detect any significant difference
in protein breakdown between feeding and fasting. In our
study, in cirrhotic patients, the rates of protein breakdown
were increased in the fed state. In the fasting state, however, no differences between patients and control subjects
were found. The cause of this increase in protein breakdown in the patients is not known, but a number of possibilities should be considered. Amino acids might be
needed and used for purposes other than protein synthesis, i.e. for gluconeogenesis. Increased breakdown might
also be caused by the high plasma glucagon levels that are
present in cirrhotic patients [20]. However, plasma insulin
levels are also elevated in cirrhotic patients [21]. These
high insulin levels should result in anabolism, but because
of insulin resistance [22] they might not be fully effective.
Since we did not study plasma hormone levels, we can
give no further interpretation of the observed phenomena.
Thus, our data indicate that in cirrhotic patients the
rate of protein turnover is increased: we think that this is
the result of giving an adequate diet to patients suffering
from protein-energy malnutrition. Alternatively, this
phenomenon could be caused by the abnormal hormonal
status in cirrhotic patients.
In our study the 9 h nitrogen balance was positive
during the fed state and negative in the fasting period both
in control subjects and in cirrhotic patients. This was to
be expected. However, there were two remarkable
observations. The nitrogen balance in the fed state was
more positive in the cirrhotic patients than in the controls.
Moreover, nitrogen loss in fasting cirrhotic patients was
no greater than in control subjects. These data suggest,
that cirrhotic patients have a better nitrogen economy
than healthy subjects. This is in contradiction with their
increased protein needs. It should be realized that a conventional nitrogen balance is a composite measurement,
being the overall result of a process fluctuating between
nitrogen incorporation after a meal and nitrogen loss
in the postabsorptive, fasting state. Protein turnover
measurements, however, are performed over a short time
span, in our study 9 h, during which the subject is kept in
steady-state metabolic conditions. A calculation of a 24 h
nitrogen balance from nitrogen turnover data could only
be made if one also knew the amount of time spent
in the fed and in the fasting state and the speed of transition from either state into the other. In cirrhosis, the 9 h
nitrogen balances suggest a better nitrogen economy,
while the 24 h balances used to measure protein requirements indicate the opposite. This discrepancy can be
explained if one assumes that cirrhotic patients spend
more hours in the fasting state and less in the fed state in
comparison with healthy subjects. With a comparable
eating pattern, this would indicate a faster transition from
fed into fasting state conditions in cirrhotic patients.
Indeed, an accelerated starvation reaction in cirrhotic
patients was recently demonstrated for which reduced
liver glycogen stores were held responsible [23,24].
The following hypothesis is postulated. The high
protein requirements in cirrhosis of the liver can be
explained as a consequence of an accelerated rate of
Whole body protein turnover in liver cirrhosis
transition from fed into fasting state conditions possibly
due to diminished liver glycogen stores. When glycogen
stores are insufficient to ensure normal blood glucose
levels even on short-term fasting, early onset of gluconeogenesis from amino acids will result. This will lead to an
additional ainino acid loss and will deplete tissue protein
stores. Extra dietary protein intake above the normal
requirements for protein renewal is then needed for repletion of these drained stores. The repletion of drained
protein stores after an adequate meal can be compared
with refeeding a malnourished person; this could explain
the higher rates of protein turnover.
The practical consequence of our hypothesis is that
nitrogen balance and ultimately the nutritional state
should improve and nitrogen requirements should be
lowered in cirrhotic patients if this early reaction of
starvation could be prevented. This could be done by supplying protein and calories in frequent small meals during
the day and the evening hours.
REFERENCES
I. Gabuzda, G.J. & Davidson, C.S. (1954) Protein metabolism
in patients with cirrhosis of the liver. Airna1.s offlieNew York
Academy of Sciences, 57,776-785.
2. OKeefe, S.J.D., Abraham, R., El-Zayadi, A., Marshall, W.,
Davis, M. & Williams, R. ( 1981) Increased plasma tyrosine
concentrations in patients with cirrhosis and fulminant
hepatic failure associated with increased plasma tyrosine
flux and reduced hepatic oxidation capacity. Gnstroenfer010gy, 81,1017-1024.
3. Morgan, M.Y. (1981) Enteral nutrition in chronic liver
disease. Acta Cliiriirgica Scandinavica, 5 0 7 (Suppl.), 8 1-90.
4. Mendenhall, C.L., Anderson, S., Weesner, R.E., Feldberg,
S.J. & Crelic, K.A. (1984)Protein-calorie malnutrition associated with alcoholic hepatitis. Veterans administration
cooperative study group on alcoholic hepatitis. American
Joiirnal ofMedicine, 26,211-222.
5. Horst, D., Grace, N.D., Conn, H.O., Schiff, E., Schenker, S.,
Viteri, A., Law, D. & Atterbury, C.E. ( 1 984) Comparison of
dietary protein with an oral branched chain-enriched amino
acid supplement in chronic portal-systemic encephalopathy.
A randomized controlled trial. Hepatology, 4,279-287.
6. Waterlow, J.C., Golden, M.H.N. & Garlick, P.J. (1978)
Protein turnover in man measured with I5N: comparison of
end products and dose regimes. American Journal of
I’liysiology, 225, E 165-E 174.
7. Picou, D. & Taylor-Roberts, T. (1969)T h e measurement of
total protein synthesis and catabolism and nitrogen turnover
in infants in different nutritional states and receiving different amounts of dietary protein. Clinical Science, 36,
283-296.
8. Waterlow, J.C. (1981) “N-end-product methods for the
study of whole body protein turnover. Proceedings of fhe
NutrifionalSociety,40,317-320.
9. Fern, E.B., Garlick, P.J., McNarlan, M.A. & Waterlow, J.C.
(1981) T h e excretion of isotope in urea and ammonia for
107
estimating protein turnover in man with [‘sN]glycine. CliriicolScierice, 61,217-228.
10. Fern, E.B., Garlick, P.J. & Waterlow, J.C. (1985) Apparent
compartmentalisation of body nitrogen in one human subject: its consequences in measuring the rate of whole body
protein synthesis with “N. ClinicalScience, 1 8 , 2 7 1-282.
11. Watson, P.E., Watson, I.D. & Batt, R.D. (1980) Total body
water volumes for adult males and females estimated from
simple anthropometric measurements. American Joirrrinl of
Clinical Nufrifion,33, 27-39.
12. Clague, M.B., Keir, M.J., Wright, P.D. & Johnston, I.D.A.
(1983) T h e effects of nutrition and trauma on whole-body
protein metabolism in man. ClinicalScience, 65, 165-1 75.
13. Rennie, M.J.. Edwards, R.H.T., Halliday, D., Matthews,
D.E., Wolman, S.L. & Millward, D.J. (1982) Muscle protein
synthesis measured by stable isotope techniques in man: the
effects of feeding and fasting. CIitiicalScience, 6 3 , 5 19-523.
14. Mullen, K.D., Denne, C.D., McCullough, A.J., Savin, S.M.,
Bruno, D., Twill, A S . & Kalhan, S.C. (1986) Leucine
metabolism in stable cirrhosis. Hepatology, 6,622-630.
15. Millikan, W.J., Henderson, J.H., Galloway, J.R., Warren,
W.D., Matthew, D.E., McGhee, A. & Kutner, M.H. (1985)
I n vivo measurement of leucine metabolism with stable
isotopes in normal subjects and in those with cirrhosis fed
conventional and branched-chain amino acid-enriched
diets. Surgery, 98,405-413.
16. Golden, M.H.N., Waterlow, J.C. & Picou, D. (1977) Protein
turnover, synthesis and breakdown before and after
recovery from protein-energy malnutrition. Clinical Science
arid Moleciilar Medicine, 53,473-477.
17. Pencharz, P., Hill, R. & Archibald, E.J. (1986) Effect of
energy repletion on dynamic aspects of protein metabolism
of malnourished adolescent and young adult patients with
cystic fibrosis during the first 12 days. I’ediatric Gastroenterologynnd Nutrition, 5,388-392.
18. Yamamori, H., Tashiro, T., Mashima, Y. & Okui, K. (1987)
Effects of severity of surgical trauma on whole body protein
turnover in patients receiving total parenteral nutrition.
Journal of Enteral arid Parentera1 Nutrition, 11,454-457.
19. Garlick, P.J., Clugston, G.A., Swick, R.W. & Waterlow, J.C.
( 1 980) Diurnal pattern of protein and energy metabolism
in man. American Joiirnal of Clinical Nutrition, 33,
1983-1986.
20. Sherwin, R.S., Prakash, J., Hendler, R., Felig, P. & Conn,
H.O. ( 1974) Hyperglucagonemia in Leannec’s cirrhosis: the
role of portal-systemic shunting. New England Joitrncil of
Medicine, 290,239-242.
2 1. Megyesi, C., Samuels, E. & Marks, V. ( 1967) Glucose tolerance and diabetes in chronic liver disease. Lancef, ii,
1051-1056.
22. Marchesini, G., Forlani, G., Zoli, M., Dondi, C., Bianchi, G.,
Bua, V., Vannini, P. & Pisi, E. (1983) Effect of euglycemic
insulin infusion on plasma levels of branched-chain amino
acids in cirrhosis. Hepatology, 3, 184-187.
23. Owen, O.E., Reichle, F.A., Mozzoli, M.A. ef 01. (1981)
Hepatic, gut and renal substrate flux rates in patients with
hepatic cirrhosis. Joiirnal of Clinical Investigation, 6 8 ,
240-252.
24. Owen, O.E., Trapp, V.E., Reichard, A,, Mozzoli, M.A.,
Moctezuma, J., Paul, P., Skutches, C.L. & Boden, G. (1983)
Nature and quantity of fuels consumed in patients with alcoholic cirrhosis of the liver. Joitrncil of Clinical Investigation,
72,1821-1832.