Vol. 284, Issue 5, E954-E965, May 2003
Protein intake during hemodialysis maintains a positive whole
body protein balance in chronic hemodialysis patients
Jorden M.
Veeneman1,2,
Hermi A.
Kingma3,
Theo S.
Boer3,
Frans
Stellaard2,3,
Paul E.
De Jong1,2,
Dirk-Jan
Reijngoud2,3, and
Roel M.
Huisman1,2
Division of Nephrology, Departments of 1 Internal
Medicine and 3 Pediatrics, University Hospital
Groningen and 2 Groningen University Institute of
Drug Exploration, 9713 GZ Groningen, The Netherlands
 |
ABSTRACT |
Protein energy
malnutrition is present in 18 to 56% of hemodialysis patients. Because
hemodialysis has been regarded as a catabolic event, we studied whether
consumption of a protein- and energy-enriched meal improves the whole
body protein balance during dialysis in chronic hemodialysis (CHD)
patients. Patients were studied on a single day between dialysis (HD
protocol) in the morning while fasting and in the afternoon while
consuming six small test meals. Patients were also studied during two
separate dialysis sessions (HD+ protocol). Patients were fasted during one and consumed the meals during the other. Whole body protein metabolism was studied by primed constant infusion of
L-[1-13C]valine. During HD, feeding changed
the negative whole body protein balance observed during fasting to a
positive protein balance. Dialysis deepened the negative balance during
fasting, whereas feeding during dialysis induced a positive balance
comparable to the HD protocol while feeding. Plasma valine
concentrations during the studies were correlated with whole body
protein synthesis and inversely correlated with whole body protein
breakdown. We conclude that the consumption of a protein- and
energy-enriched meal by CHD patients while dialyzing can strongly
improve whole body protein balance, probably because of the increased
amino acid concentrations in blood.
protein turnover; stable isotope; valine
| |
INTRODUCTION |
SIGNS OF PROTEIN
ENERGY MALNUTRITION occur frequently in patients with chronic
renal failure (17, 19). Protein energy malnutrition has
been shown to be a major risk factor for increased morbidity and
mortality in the chronic hemodialysis (CHD) patient (2,
17). Multiple factors predispose CHD patients to protein energy
malnutrition, e.g., low caloric intake, low protein intake, and the
hemodialysis procedure itself. Particularly, losses of amino acids or
abnormal protein metabolism during hemodialysis might contribute to the
observed protein energy malnutrition. Studies examining the role of CHD
itself on protein metabolism are limited (22). Several
lines of evidence indicate that the CHD procedure can result in a
negative whole body protein balance. Nitrogen balance has been shown to
be more negative on a dialysis day compared with a nondialysis day
regardless of daily protein intake (5, 23). Lim et al.
(22) studied whole body protein metabolism by applying the
[13C]leucine isotope dilution technique in fasting CHD
patients during hemodialysis. They observed a reduction in whole body
protein synthesis compared with the predialysis period, and this
resulted in a doubling of the negative protein balance already
present in fasting CHD patients. Furthermore, hemodialysis
stimulates muscle protein losses compared with the predialysis period
in fasting CHD patients (18).
In apparently healthy subjects, the consumption of a meal or the
administration of an amino acid mixture reverses the negative protein
balance observed after an overnight fast (for reviews see Refs.
8 and 45). Particularly, amino acids in plasma are
powerful modulators of protein metabolism, as a mixture or in
conjunction with insulin (42). Protein breakdown is
inhibited, while protein synthesis and protein oxidation are stimulated
by amino acid infusion. As a result, whole body protein balance becomes positive (29, 30).
The situation in CHD patients is less well known, and the effects of a
meal during dialysis have not been studied so far. It is common
clinical practice, at least in Europe, that CHD patients are allowed to
eat during a 4-h dialysis session. We adapted this practice for the
purpose of nutritional intervention. A milk-based protein- and
energy-enriched meal was given to the patients during a dialysis
session and on a nondialysis day. The meal was designed with the
assumption that a maximum anabolic response would be elicited in our
patients by a meal enriched in both energy and protein. We studied the
effect of this oral intradialytic nutrition on whole body protein
metabolism during hemodialysis with a biocompatible membrane in CHD
patients. We addressed the following two questions more specifically:
1) to what extent does consumption of a protein- and
energy-enriched meal result in a positive whole body protein balance in
CHD patients, and 2) how effective is such a meal consumed during a dialysis session in the prevention of the negative protein balance in CHD patients during dialysis? The first question was studied
in CHD patients during a nondialysis day, the second question during
two dialysis sessions separated by 1 wk. Whole body protein metabolism
was studied by applying stable isotope infusion techniques using
[1-13C]valine as a tracer (4). The use of
this tracer has certain advantages over [1-13C]leucine
both analytically and metabolically. [1-13C]valine has
been used previously by our laboratory in the study of whole body
protein metabolism and synthesis of several specific proteins
(34) in nephrotic patients. Furthermore, it has been reported that, under certain conditions, high (flooding) doses of
leucine can provoke an insulinomimetic effect on protein metabolism (9, 12), whereas this is not the case for valine. At doses normally applied in the study of whole body protein metabolism, valine
and leucine give similar values of the fluxes of protein breakdown,
synthesis, and oxidation (38).
| |
SUBJECTS AND METHODS |
Study Subjects
All nondiabetic, stable hemodialysis patients aged under 65 yr
in the Dialysis Center Groningen were approached to participate in the
two protocols of the present study, i.e., a nondialysis and a dialysis
protocol. Twelve patients gave their permission, but only three or them
agreed to participate in both protocols. The other patients considered
this too great a demand since they objected to fasting during the
dialysis session. In summary, three patients participated in both
protocols, six patients participated only in the nondialysis protocol,
and three patients participated only in the dialysis protocol (Table
1). The medical ethics committee of the
University of Groningen approved all studies, and written informed consent was obtained from all participants. All participants were clinically stable, without intercurrent acute illness in the 3 mo
before the study protocol and had been in dialysis for 6 mo or more.
The diagnoses were chronic glomerulonephritis in three patients (1 with
hypertension), nephropathy resulting from hypertension in three
patients, quiescent Wegeners disease in one patient, and polycystic
kidney disease in three patients, and the cause of renal failure was
unknown in two cases. Medications included phosphate binders, iron,
multivitamins, antihypertensive drugs, calcitriol, and recombinant
human erythropoietin of which the dose had not been altered for 3 mo
before the study protocol to avoid altered hematopoiesis. No patients
received steroid hormones or immunosuppressive agents in the 6 mo
before the study protocol. The patients were dialyzed with low-flux
biocompatible dialyzers for 4 h three times weekly. Blood flow
ranged from 250 to 350 ml/min, and dialysate flow was 500 ml/min.
Standard dialysate with 140 meq Na+, and 34 meq
bicarbonate, was used for all patients. Glucose content in dialysate
was 5.6 mM in two patients and 11.2 mM in four patients during their
experimental dialysis sessions. Residual renal function was 3 ml/min or
less, which corresponded to a dialysis adequacy (Kt/V) value of 0.45 wk
1 or less.
Materials
L-[1-13C]valine and
NaH13CO3, both with an enrichment of >99 atom
percent excess, were purchased from Cambridge Isotope Laboratories (Andover, MA). Chemical purities were confirmed before use. Pyrogen- and bacteria-free solutions were prepared in sterile saline by the
hospital dispensary the afternoon before the study day. Meal portions
consisted of 150 g yogurt (5.7 g protein, 7.4 g carbohydrate, and 5.4 g fat; Domo), 20 g cream (0.5 g protein, 0.7 g
carbohydrate, and 6.3 g fat; Friesche vlag, Ede, the Netherlands),
and 5 g protein-enriched milk powder (1.5 g protein, 2.4 g
carbohydrate, and 0.8 g fat; Fortify, Nutricia). Consumption of a
meal portion every 30 min for 3 h resulted in a dietary valine
intake of 132 ± 20 µmol · kg1 · h1
(37) and a fluid intake of 350 ml/h. The energy content of a meal portion was 386 kcal/h. Meals were designed to give >50% of
daily protein intake, 0.62 ± 0.09 g/kg protein, and 15 ± 2 kcal/kg in energy content. It was assumed that gastric emptying during
the meal was not disturbed, since our patients had no history of
dyspeptic symptoms during the 3 mo before both protocols and were in a
good nutritional state (41).
Experimental Design
Pilot experiments.
Dialysis by itself was found to gradually increase the
13CO2 enrichment in expired air because of the
entrance of bicarbonate with a high natural enrichment from the
dialysate (4.0 ± 0.3
vs. Pee Dee Belemnite
limestone). Therefore, background enrichment in expired breath
was measured independently in five patients during a dialysis session
before this study. The time course of this change as a percentage of
the initial background enrichment of expired CO2 was used
to correct the value of the 13CO2 excess
enrichment obtained during either the [13C]bicarbonate or
[13C]valine infusion for calculations of whole body
protein metabolism. In a second pilot experiment, the extent to which
the rate of [13C]valine infusion had to be increased
during dialysis was tested. This was deemed necessary since it was
observed that the turnover in the bicarbonate pool was increased during
dialysis, and, consequently, infusion of valine had to be increased to
obtain 13CO2 enrichments in expired air, which
could be measured reliably in excess of the background enrichment that
had already been changed by exchange of plasma and dialysate
bicarbonate. Doubling the [1-13C]valine infusion rate
appeared to be sufficient.
Study protocols.
The present study comprised two protocols. In the nondialysis protocol
(HD), patients were studied on a day between two dialysis days.
Fasting whole body protein metabolism was measured in the morning after
an overnight fast (HD fas). On the same study day, in the afternoon,
this was followed by the measurement of whole body protein metabolism
while patients were consuming the meal (HD fed). The dialysis
protocol (HD+) could not be done on a single day and therefore
consisted of two study days 1 wk apart. Patients were dialyzed normally
on these days, and measurements were made during the dialysis session.
On one occasion, patients were studied while they remained fasting (HD+
fas), and on the other occasion patients consumed a protein-enriched
meal (HD+ fed). The HD+ protocol started after the completion of the
study of whole body protein metabolism during the HD protocol. Before the study (3 wk), all patients visited the Dialysis Center Groningen for a dietary interview and instructions on dietary recording. Patients
consumed a protein intake of 1.0 ± 0.1 g · kg1 · day1,
while caloric intake was not restricted.
Nondialysis protocol.
In the HD protocol, patients had fasted overnight and were studied
during a midweek day without dialysis, having dialyzed the afternoon
before. Patients were admitted to the Hospital Research Unit at 
7:30
AM. A catheter was inserted in the dorsal vein of the hand of the shunt
arm to collect baseline blood samples. Subsequently, breath samples
were taken. A schematic diagram of the study day is shown in Fig.
1A. The
NaH13CO3 infusion was started at 8:00 AM.
During the 1st h, whole body bicarbonate production (details explained
in Evaluation of Primary Data) was measured using a primed
constant infusion of NaH13CO3 (5 µmol/kg
bolus followed by a continuous infusion of 5 µmol · kg1 · h1).
Four breath samples were taken from 30 to 60 min after the start of the
NaH13CO3 infusion at 10-min intervals. The
NaH13CO3 infusion was discontinued immediately
after the last breath sample was taken, and the
L-[1-13C]valine infusion was started with a
bolus of 15 µmol/kg followed by a continuous infusion of 7.5 µmol · kg1 · h1
for the next 4 h. A second catheter was then inserted in the contralateral arm to collect blood samples. Blood and breath samples were taken simultaneously every half hour for 3 h after the start of the [13C]valine infusion. During the 4th h, blood and
breath samples were taken every 15 min. At 1:00 PM, the meal period was
started by consumption of the first portion of the protein-enriched
meal and continued for 3 h by consumption of a portion every 30 min. [13C]valine infusion continued at the same rate
during this study period. Blood and breath samples were taken every 30 min for 2 h after the start of the meal, whereas during the last
hour, samples were taken every 15 min. The study day ended at 4:00 PM.
All catheters were removed, and patients were observed until stable and
then discharged.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1.
A: protocol used to study whole body protein metabolism
in chronic hemodialysis (CHD) patients during a nondialysis day (HD).
After an overnight fast, whole body protein metabolism (HD fas) was
measured in the morning while patients fasted. This was followed in the
afternoon by the measurement of whole body protein metabolism in
patients consuming a protein- and energy-enriched meal in 6 portions
(HD fed). B: protocol used to study whole body protein
metabolism in fasting CHD patients while dialyzing (HD+ fas).
C: protocol used to study whole body protein metabolism in
CHD patients consuming a protein- and energy-enriched meal during
dialysis (HD+ fed). inf, Infusion.
|
|
Dialysis protocol.
In this protocol, patients were studied on two separate dialysis
sessions, separated by 1 wk. On one occasion, patients were studied
while they remained fasting. On the second occasion, patients consumed
six small meals, the first 1 h after the start of dialysis followed by five meals spaced by 30 min. Patients had been dialyzed 44 ± 3 h before they entered the study protocol. Studies
were performed in the afternoon, and patients consumed a late-evening snack the evening before the study day to keep the fasting period comparable to that in the HD protocol. Patients were admitted to the
dialysis center at 11:30 AM. Dialysis needles were inserted in the
arterial-venous shunt to collect baseline blood samples, and breath
samples were collected simultaneously. Dialysis started at 12:00 PM.
A primed continuous infusion of NaH13CO3 was
administered for 1 h through the venous line of the dialysis machine (5 µmol/kg bolus followed by a continuous infusion of 5 µmol · kg1 · h1),
and four breath samples were taken from 30 to 60 min after the start of
the infusion at 10-min intervals. The NaH13CO3
infusion was discontinued after the last breath sample was taken, and a
primed continuous infusion of
L-[1-13C]valine was
started through the venous line of the dialysis machine for 3 h
(15 µmol/kg bolus followed by a continuous infusion of 15 µmol · kg1 · h1).
Blood samples from the arterial line of the dialysis machine and breath
samples were taken every half hour for the first 2 h after the
start of the [13C]valine infusion. During the third and
last hour, blood and breath were sampled every 15 min (Fig.
1B). When whole body protein metabolism was studied during
the meal period, the same experimental setup was used as described
above, with the exception that, at the start of the
[13C]valine infusion, the first of the six meal portions
was consumed, whereas the remaining five were consumed every 30 min
during the next 3 h (Fig. 1C). Blood pressure was
monitored during all experimental dialysis sessions. Blood flow was
estimated from the flow given on the dialysis machine while dialysate
flow was 500 ml/min plus the ultrafiltration. Approximately 70% of the
ingested fluid was removed during the experimental dialysis session,
while the other 30% was removed during the next dialysis session.
Analytical procedures.
Blood (4 ml) was drawn for each sample in liquid-heparinized vacuum
tubes and centrifuged at 3,000 rpm. Plasma was extracted and stored at
20°C until analysis. Breath samples were collected in gas
collection tubes with a straw, as described earlier (43). Subjects exhaled normally through a straw in the glass container. After
exhalation was completed, tubes were closed immediately and stored at
room temperature until analysis. Dialysate was sampled every half hour
using a syringe to extract 4 ml of dialysate and was stored at 20°C
until analysis.
Amino acid concentrations in plasma and spent dialysate were measured
by the AccQ Tag method using HPLC according to the manufacturer's protocols (Waters, Breda, The Netherlands). Amino acids were grouped according to total amino acids, the sum of the concentration of all
individual amino acids, essential amino acids (the sum of the
concentration of arginine, histidine, lysine, methionine, phenylalanine, threonine, isoleucine, leucine, and valine), and the
nonessential amino acids (the concentration of total amino acids essential amino acids). Insulin in plasma was determined by a double
AB system, as described earlier (24). Glucose and albumin
concentrations were determined by standard clinical chemistry methods.
Measurement of 13CO2 isotopic enrichment was
performed by sampling directly the glass container with a Finnigan
TracerMat (Finnigan MAT, San Jose, CA) continuous-flow isotope
ratio-mass spectrometer as described by Vonk et al.
(43).
The determination of L-[1-13C]valine isotopic
enrichment was done as described earlier (32). In short,
amino acids were isolated from deproteinized plasma using a cation
exchange column (SCX-100, 209800; Alltech, Deerfield, IL). The isolated
amino acids were derivatized to their corresponding
N(O)-methoxycarbonyl methyl ester (MCM) according
to Husek (15). Analysis of isotopic enrichment of plasma
[13C]valine was carried out by GC-MS on a Hewlett Packard
5890 Plus gas chromatograph coupled to a Finnigan SSQ 7000 quadruple
mass spectrometer using methane positive-ion chemical ionization. The gas chromatograph was fitted with a capillary column (AT 1701, length
20 m, ID 0.18 mm, film thickness 0.40 µm; Alltech). The mass
spectrometer was operated in the selected ion-monitoring mode at
fragments with a mass-to-charge ratio (m/z) 190/191 of the
[MH]+ and [MH+1]+ ions of the MCM
derivative of unlabeled valine and
L-[1-13C]valine, respectively.
-[1-13C]ketoisovaleric acid (KIVA) isotopic enrichment
was determined according to Kulik et al. (20). In short,
standards with a tracer mole ratio for [1-13C]KIVA
ranging from 0 to 22% were prepared by enzymatic conversion of
standards of L-[1-13C]valine with the
corresponding tracer mole ratio, as described earlier
(33). Standards of [1-13C]KIVA and patient
plasma samples were processed in the same series. KIVA was converted to
its quinoxalinol-O-t-butyldimethylsilyl derivative. Isotopic enrichment of the derivatized samples was measured
by GC-MS on a Hewlett Packard 5890 Plus gas chromatograph coupled to a
Finnigan SSQ 7000 quadruple mass spectrometer using positive-ion
electron-impact ionization. The gas chromatograph was fitted with a
capillary column (AT 1701; Alltech). The mass spectrometer was operated
in the selected ion-monitoring mode recording fragments at
m/z 245 and 246 of unlabeled KIVA and
[13C]KIVA, respectively. All isotopic enrichments were
measured against standard calibration curves.
Evaluation of Primary Data
Rate of appearance of intracellular valine (Ra) was
calculated at isotopic steady state using the inverted pool model
described by Matthews and colleagues (26-28) for
leucine kinetics. When this isotopic model is applied to
[1-13C]valine, enrichment of plasma KIVA is assumed to
provide an estimate of intracellular enrichment of valine. The rate of
appearance (µmol
valine · kg1 · h1)
was calculated according to
where MPEi(V) is the isotopic enrichment of the valine in the
infusate in mole percent excess, MPE(KIVA) is the isotopic enrichment
of KIVA in plasma in mole percent excess, and i(V) is the infusion rate
of [1-13C]valine
(µmol · kg1 · h1).
The rate of oxidation of valine was calculated following the
approach described by Van Goudoever et al. (40). We did
not use indirect calorimetry in our study to determine CO2
production as a measure of whole body bicarbonate production.
Measurements would be perturbed when the comparison between the HD
and the HD+ protocol is made because bicarbonate from the dialysis
fluid enters the circulation and changes the bicarbonate pool of the patient. As a consequence, an unknown fraction of the whole body bicarbonate flux is derived from the dialysis fluid (22).
In the approach of Van Goudoever et al., whole body bicarbonate flux is
estimated before the [13C]valine infusion using a primed
continuous infusion of NaH13CO3 of short
duration. In this way, a two-point calibration is obtained with
background 13CO2 enrichment at no infusion of
NaH13CO3 and the measured value of enriched
CO2 at the applied continuous infusion rate of
NaH13CO3. The [13C]bicarbonate
flux originating from the oxidation of [13C]valine was
then calculated by linear interpolation of the measured 13CO2 enrichment in expired air at steady state
during [13C]valine infusion between the two points of the
calibration. In other words, the ratio of enrichments of
13CO2 in expired air during
[13C]valine infusion over that during
NaH13CO3 infusion is a reflection of the ratio
between the rate of [13C]bicarbonate production
originating from the oxidation of [13C]valine over the
rate of continuous infusion of NaH13CO3. From
the KIVA enrichment, which represents the intracellular dilution of
valine, we calculated the amount of valine being oxidized to sustain
this calculated production of [13C]bicarbonate. This
results in the following calculations
![i<SUB>bic</SUB>(V)<IT>=</IT>[IEco<SUB>2</SUB>(V)/IE<SC>co</SC><SUB>2</SUB>(B)]<IT>×</IT>i(b)](/content/vol284/issue5/fulltext/E954/img002.gif)
|
|
in which ibic(V) is the
[13C]bicarbonate production from
[13C]valine during valine infusion,
IECO2(B) is the isotopic enrichment in atom
percent enrichment (APE) of 13CO2 in expired
air at isotopic steady state during the
NaH13CO3 infusion,
IECO2(V) is the isotopic enrichment in APE of
13CO2 in expired air at isotopic steady state
during the [13C]valine infusion, and i(b) is the
NaH13CO3 infusion rate in micromoles per
kilogram per hour. Valine oxidation (Ox) was calculated according to
where 5 is the number of carbon atoms in valine. In this way,
the oxidation rate of [13C]valine could be calculated
without measuring 
CO2. During fasting: Ox = O(fast).
During the meal period, recovery of labeled CO2 will
be increased in comparison with fasting. Estimates from the literature have been used, i.e., 0.74 ± 7 to 0.84 ± 8 during fasting
and meal intake, respectively (11, 14, 21). This
represents an increase of ~13%. Correction of the rate of oxidation
of valine during the meal period is necessary because the two-point
calibration was done while the patient was fasting. O(fed) was thus
calculated according to
Calculation of Whole Body Protein Metabolism
In Fig. 2, the steady-state
isotopic model for whole body valine metabolism is depicted in a
schematic diagram. In this model, influx of valine comes from whole
body protein breakdown (B) and, when appropriate, from meal intake (I).
Valine leaves the plasma amino acid pool by whole body protein
synthesis (S), oxidation (O), and, when applicable, dialysis (D). The
input fluxes in this model result in label dilution of infused
[1-13C]valine in plasma. These fluxes have to be
differentiated from those that result in changes in size of the plasma
amino acid pool. This is of particular importance for the calculation
of the rate of appearance of valine in plasma in the experiments in
which the influence of protein intake has been studied. During protein
intake, plasma amino acid concentrations increased gradually. Therefore, the appearance of dietary valine in the circulation comprised a flux resulting in enlargement of the plasma valine pool and
a flux of dietary valine, adding to whole body protein metabolism. The
appearance of dietary valine was multiplied by 0.8 to correct for
first-pass metabolism (10, 13). The amount of dietary
valine entrapped in the enlarged pool size of valine (
Q) was
calculated by multiplying the increase in valine concentration in
plasma by total body water, defined as 60% of body weight in these
patients (6). The difference of plasma valine
concentration before and at the end of the meal period was used to
calculate the increase in the whole body valine pool. We observed that
the increase of valine during dietary protein intake was continuous during our study period. We assumed this increase to be linear in time
and the associated flux to be constant. Accordingly, the total rate of
appearance of valine comprises the appearance of valine released from
breakdown, infusion of valine (i), and, when appropriate, protein
intake. At steady state, the rate of appearance of valine equals the
rate of disappearance of valine. The total rate of disappearance of
valine comprises protein synthesis, protein oxidation, and, when
appropriate, losses in the dialysate. At steady state
This results in the following calculations.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 2.
Model of whole body protein metabolism. Boldface
indicates the situation during fasting, whereas lightface indicates
that meal intake and pool enlargement (Q) are added to the scheme.
The fluxes shown are whole body protein breakdown (B), synthesis (S),
oxidation (O), dialysate loss (D), infusion of
[13C]valine (i), and the dietary intake of valine (I),
corrected for first-pass absorption effects.
|
|
Without dialysis during fasting (HD fas), I = 0 and
D = 0; so
and
Without dialysis during the meal period (HD fed), I 
0 and D = 0; so
and
While dialyzing during fasting (HD+ fas), I = 0 and D 0; so
and
While dialyzing during the meal period (HD+ fed), I 0 and D 0; so
and
Protein balance was calculated by subtracting protein breakdown
from protein synthesis. Dialysate losses were calculated by measuring
amino acid concentrations in spent dialysate in micromoles per liter
multiplied by the volume flow of dialysate in liters per hour.
Statistics
All values are given as means ± SD. Statistical analysis
was done using SPSS 10.0 (SPSS, Chicago, IL). To compare the changes in
protein metabolism resulting from the meal, the fasting and fed states
were compared using a paired Student's t-test. Differences between the protein metabolism parameters on a nondialysis day and
during dialysis were tested using the unpaired Student's
t-test. Correlations between valine concentrations and
protein metabolism parameters were tested using linear regression
analysis and expressed using the Pearson correlation coefficient.
Statistical significance was assumed at P < 0.05.
| |
RESULTS |
Demographic Data
Table 1 shows the demographic and clinical data of the patients
studied. All patients were well nourished, as can be concluded from the
protein intake and serum albumin concentrations. Dialysis was adequate,
as shown by the equilibrated Kt/V values. There were no statistical
differences between the two study groups with respect to body mass
index, age, or albumin concentration. There were episodes of
hypotension in two out of six patients only during dialysis with
feeding, which could be reversed by discontinuing the ultrafiltration.
The difference between the two dialysate glucose concentrations did not
influence plasma glucose and insulin concentrations. Predialysis
glucose concentrations in plasma were 5.6 mM (range 3.9-7.5 mM)
during fasting and did not change during the dialysis session. During
the meal period, glucose concentrations in plasma were 6.8 mM (range
5.2-8.7 mM). Predialysis insulin values in plasma were 11.6 mIU/l
(range 6.1-15.3 mIU/l) during fasting and did not change during
the dialysis session. During the meal period, insulin concentrations in
plasma were 44.7 mIU/l (range 19.8-84.6 mIU/l). Glucose or insulin
values did not correlate with the other studied variables in our subjects.
Amino Acid Concentrations
Losses of amino acids in the dialysate during the fasting
period were 74 ± 21 mmol · patient1 · dialysis
session1 (7.7 ± 2.1 g). The valine loss
contributed 21 ± 4 µmol · kg1 · h1
to the total amino acid losses. The amino acid loss was 35% higher during the study day with dialysis while the patients consumed a meal
and was 105 ± 13 mmol · patient1 · session1
(11.7 ± 1.9 g). Valine losses contributed 35 ± 5 µmol
valine · kg1 · h1
to this loss. Plasma valine concentration increased during dietary protein intake by 150 ± 31 µM on a nondialysis day and 126 ± 37 µM on a dialysis day. In Fig. 3,
the arterial plasma amino acid concentrations are shown for each study
group. Total amino acids, essential amino acids, and nonessential amino
acids were all significantly higher during feeding compared with
fasting, both during the HD protocol and during the HD+
protocol. Furthermore, there was a difference in response upon dialysis
between the essential and nonessential amino acids. Although dialysis
has only a minor, not significant, influence on the concentration of
essential amino acids (HD 844 µM to HD+ 732 µM) in plasma, the
concentration of nonessential amino acids (HD 2,006 µM to HD+ 1,269 µM) and thus total amino acids (HD 2,850 µM to HD+ 2,001 µM) in
plasma decreased significantly during the dialysis sessions during
fasting. Consumption of a protein-enriched meal during the HD
protocol also changed the relative composition of plasma amino acids.
Essential amino acids increased relatively more (844 to 1,447 µM)
than the nonessential amino acids (2,006 to 2,626 µM). Consumption of
the meal during dialysis resulted in an increase of essential amino acids (732 to 1,273 µM) and of nonessential amino acids (1,269 to
1,723 µM). It can be seen in Fig. 3 that the concentration of
nonessential amino acids during dialysis and meal intake was lower than
the nonessential amino acid concentration during the HD protocol
while fasted.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 3.
Concentrations of total, essential, and nonessential
amino acids in plasma in the steady-state periods in all study
protocols. +Significant difference in amino acid concentration during
fasting compared with feeding. * Significant difference in amino
acid concentration during the HD+ protocol compared with the same
condition (fasting or feeding) during the HD protocol.
|
|
Protein Metabolism
In Fig. 4, plateau enrichments for
breath CO2 enrichment and plasma KIVA enrichment are shown
to illustrate their steady state in time. Figure 4A shows,
during the HD protocol, plateau 13CO2
enrichment in expired breath before the start of the experiment, during
the NaH13CO3 infusion from 0.5 to 1 h, and
during the [13C]valine infusion while fasting (between 4 and 5 h) or while consuming a protein-enriched meal (between 7 and
8 h). Figure 4B shows plateau enrichments of plasma
KIVA during the HD protocol during fasting (between 4 and 5 h)
and during consumption of a protein-enriched meal (between 7 and 8 h). Figure 4C shows the plateau enrichments of
13CO2 in expired air during the HD+ protocol
before the start of the experiment, during
NaH13CO3 infusion (between 0.5 and 1 h),
and during the [13C]valine infusion while fasted or fed
(between 3 and 4 h). Furthermore, the time course is shown of
13CO2 in expired air during dialysis in the
absence of any infusion of NaH13CO3 or
[13C]valine. Figure 4D shows the plateau
enrichments of plasma KIVA during the HD+ protocol during fasting and
feeding (between 3 and 4 h). The effect of dialysis on total
bicarbonate turnover is already clear from a comparison of steady-state
enrichments of 13CO2 in expired air during
NaH13CO3 isotope infusion in the HD protocol
(
= +20 ± 4) and HD+ protocol ( = +9 ± 3) because of the exchange of plasma bicarbonate for dialysate
bicarbonate. In Table 2, the summary is
given of the evaluation of primary isotopic data and changes in valine concentration for the individual patients. As is clear from this table,
in fasting CHD patients, the total rate of appearance of valine in
plasma is not significantly influenced by dialysis (77 ± 8 vs.
80 ± 15 µmol · kg1 · h1).
A similar observation can be made for the total rate of appearance of
valine in plasma during dietary protein intake (108 ± 13 vs. 117 ± 20 µmol · kg1 · h1).
Furthermore, it can be seen that the increase in plasma valine pool
size is significantly higher in the HD protocol (38 ± 8 µmol · kg1 · h1)
compared with the HD+ protocol (25 ± 7 µmol · kg1 · h1),
most likely because of the loss of valine (35 ± 5 µmol · kg1 · h1)
during dialysis. In Table 3, data on
whole body protein metabolism are given that were derived from the data
given in Table 2. The rates of whole body protein synthesis and
oxidation in the fasting state were significantly lower during dialysis
compared with the rate calculated during the HD protocol. During the
meal period, the rate of whole body protein breakdown was reduced
during both the HD protocol and the HD+ protocol. The rates of
protein synthesis and oxidation were reduced during the HD+ protocol
compared with the HD protocol. Protein balance during fasting,
calculated as the difference between the rates of whole body protein
synthesis and breakdown, was significantly higher in the HD protocol
(8 ± 5 µmol · kg1 · h1)
compared with the balance during the HD+ protocol (15 ± 4 µmol · kg1 · h1).
The consumption of a protein-enriched meal improved protein balance
significantly compared with fasting in both protocols. During the HD
protocol, protein balance increased to 32 ± 9 µmol · kg1 · h1,
while during the HD+ protocol, protein balance increased to 31 ± 5 µmol · kg1 · h1
(not significant). Figure 5 shows whole
body protein synthesis, breakdown, and balance during the fasting and
meal period in the HD and HD+ protocols. In the absence of dialysis,
protein balance increased 39 ± 9 µmol · kg1 · h1
during the meal period, whereas the increase during dialysis was
45 ± 4 µmol · kg1 · h1.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 4.
Breath 13CO2 (A) and plasma
-[1-13C]ketoisovaleric acid
([13C]KIVA, B) enrichments during the
HD protocol. Breath 13CO2 (C) and
plasma [13C]KIVA (D) enrichments during the
HD+ protocol. In A and B, closed squares
represent the respective enrichments during the HD protocol. In
C and D, open circles represent the dialysis
study day while the patients were fasting, and closed circles represent
measurements during consumption of the meal. The open triangles in
C represent the change in background bicarbonate enrichment
resulting from the dialysis treatment, which was tested in a separate
study in 5 patients (see Pilot experiments).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 5.
Summary of whole body protein breakdown (gray bars),
synthesis (open bars), and protein balance (filled bars) under all
experimental conditions. * Whole body protein balance significantly
different from 0. +Whole body protein balance significantly different
between fasting and feeding. #Whole body protein balance significantly
different between the HD+ and HD protocols.
|
|
Correlations
Protein synthesis was positively correlated (Pearson
r = 0.50, P < 0.01) and protein
breakdown negatively correlated (Pearson r = 0.54,
P < 0.01) with plasma valine concentration; this was tested using both parameters as continuous variables. In Fig. 6, this is illustrated for the four
conditions separately.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6.
Correlation of valine concentration in plasma with whole
body protein synthesis (positive numbers) and whole body protein
breakdown (negative numbers). Diamond, HD+ protocol while patients were
fasting; triangles, HD protocol while patients were fasting; squares,
HD+ protocol while patients were fed; circles, HD protocol while
patients were fed. There is a significant correlation between valine
concentration and whole body protein synthesis (r = 0.50, P < 0.01) and with whole body protein breakdown
(r = 0.54, P < 0.01).
|
|
| |
DISCUSSION |
The aim of the study was to test the hypothesis that consumption
of a protein- and energy-enriched meal restores whole body protein
balance during dialysis. Therefore, we examined the effects of such a
meal on whole body protein metabolism in CHD patients on a day between
two dialysis days and during dialysis. We used a primed continuous
infusion of [13C]valine and measurement of isotope
dilution of [13C]KIVA in plasma and
13CO2 in expired air. Our study shows that, on
a nondialysis day, protein balance was negative after an overnight
fast. Consumption of a protein-enriched meal resulted in a positive
whole body protein balance. During dialysis, fasting patients were in
an even more negative protein balance than on a nondialysis day.
Consumption of a protein- and energy-enriched meal during dialysis
resulted in a positive protein balance to the same extent as on a
nondialysis day. Dialysis led to considerable losses of plasma amino
acids in dialysate, which could be supplemented by dietary amino acids but with a shift in composition between essential and nonessential amino acids.
Before interpreting our results, we would like to discuss some
methodological issues. In our study, we used high infusion rates of
[13C]valine during dialysis to measure enrichment of
13CO2 in expired air with sufficient precision
above background. Particularly, we anticipated low values of
13CO2 enrichment in expired air in patients
consuming a protein-enriched meal during a dialysis session.
Accordingly, in the HD+ protocol the rate of infusion of
[13C]valine was doubled compared with the HD protocol.
Under conditions of dialysis and dietary protein intake,
13CO2 enrichment in expired air was low because
1) isotope dilution of [13C]valine was
considerable because of the appearance of dietary valine and
2) total bicarbonate production increased because of exchange of plasma bicarbonate with extracorporeal bicarbonate in
dialysate. Enrichment of plasma [13C]KIVA during dialysis
in the absence of protein intake was ~20%. Infusion of
[13C]valine at such high rates is considered a
"flooding" dose of tracer, which could perturb the processes to be
studied (25, 36, 44). However, these perturbations will
most likely be limited in the case of valine. Several studies have
shown that a flooding dose of valine does not elicit an anabolic
response of whole body protein metabolism (7, 35).
Furthermore, infusion of leucine affects plasma valine concentration,
whereas infusion of valine does not affect plasma leucine concentration
(1, 9). During the meal period, plasma
[13C]KIVA enrichment decreased to values of ~15%.
We used an independent infusion of NaH13CO3 of
short duration to estimate whole body bicarbonate production, instead
of indirect calorimetry. During dialysis, plasma bicarbonate exchanges
with extracorporeal bicarbonate in dialysate. With indirect
calorimetry, this effect of dialysis on the whole body bicarbonate
content cannot be estimated, since this method measures the net effect of this exchange. Accordingly, the dilution of
13CO2 derived from [13C]valine
oxidation in the body bicarbonate pool cannot be determined accurately
by applying CO2 measured by indirect
calorimetry while the patient is dialyzing. In one study using indirect
calorimetry to measure whole body bicarbonate production, the
bicarbonate influx from the dialysis machine was estimated by taking
the arterial-venous difference in bicarbonate concentration across the
dialysis machine (22). Influx of bicarbonate from the
dialysis machine was calculated to be negligible compared with whole
body bicarbonate flux. This is true for net bicarbonate gain during the
dialysis procedure. However, arterial-venous differences do not measure
the unidirectional fluxes of bicarbonate exchange across the dialyzing
membrane, and this is what matters in isotope dilution studies. These
unidirectional fluxes contribute to the apparent increase in whole body
bicarbonate production, observed as the increase of isotope dilution of
CO2 resulting from dialysis ( approximately equal to
+10) compared with nondialysis ( approximately equal to +20).
Additionally, bicarbonate dissolved in dialysate was found to be
naturally enriched ( approximately equal to 4) compared with
background enrichment of plasma bicarbonate in our patients ( approximately equal to 25). Exchange of bicarbonate between plasma
and dialysate resulted in a gradual increase in
13CO2 background enrichment that reached steady
state in the last 3 h of dialysis ( approximately equal to
20). We corrected for these changes in
13CO2 background enrichment, otherwise
oxidation rates would have been overestimated by ~20%. This
overestimation of the oxidation rate would have resulted in an
underestimation of protein synthesis. The significant changes in
13CO2 background enrichment observed in our
studies precluded comparison of whole body protein metabolism
immediately preceding the dialysis session with that during dialysis
and immediately after dialysis in a single measurement.
Turning to our results, we found that dialysis mainly decreased whole
body protein synthesis and to a lesser extent whole body protein
oxidation. Whole body protein breakdown was not significantly affected,
or in other words, the rate of appearance of valine in plasma,
corrected for infusion of labeled valine, was not affected by dialysis.
In several studies a similar observation was made (3, 22).
However, Ikizler et al. (18) showed an increase in protein
breakdown upon dialysis. Although the reason for this discrepancy is
not clear, there are differences in the execution of those studies
compared with our study. Leucine oxidation rates were estimated from
the appearance of 13CO2 in expired air and the
total CO2 production measured by indirect calorimetry. Only
Lim et al. (22) corrected for the loss of 13CO2 in dialysate, albeit with a value based
on theoretical considerations. We measured the bicarbonate flux in each
patient studied while dialyzing. We extended the isotopic model of
whole body protein metabolism to accommodate dietary valine influx and
losses of valine in dialysate. When this model is applied to the
results of our measurements in fasted, nondialyzing CHD patients,
interpretation is straightforward. In this case, appearance of valine
in plasma, corrected for infusion of isotope, arises from endogenous
sources, i.e., whole body protein breakdown, and whole body protein
synthesis equals nonoxidative disposal of valine. In cases of protein
intake and/or dialysis, the model becomes more complicated. We reasoned that, during dialysis, loss of valine in dialysate contributed to the
nonoxidative disposal of valine. Accordingly, the associated flux was
subtracted from the rate of nonoxidative disposal of valine. Ikizler et
al. (18) used another modeling approach for the amino acid
losses, which might have influenced their conclusion.
Consumption of a protein-enriched meal by CHD patients on a nondialysis
day resulted in a positive whole body protein balance. Whole body
protein breakdown was reduced to about two-thirds the rate observed
during fasting in these patients. Synthesis was slightly increased to
125%, and oxidation was strongly increased to 205%. In view of the
absolute values of the rates of whole body protein breakdown, the
positive whole body protein balance at the end of a meal was mainly the
consequence of the strong inhibition of whole body protein breakdown.
Similar observations have been made in apparently healthy individuals
(29, 30). A difference with earlier studies is that we
corrected for the enlarged valine pool. During the consumption of the
protein-enriched meal, valine concentration in plasma of CHD patients
increased continuously. Accordingly, dietary valine influx was
calculated as the difference between the enteral release of valine
appearing in the circulation and the flux of valine associated with the enlargement of the plasma valine pool (see Fig. 2). This correction of
the enteral release of valine for pool enlargement makes the calculation of whole body protein breakdown sensitive to changes in the
size of the plasma valine pool. It does not influence the calculation
of whole body protein synthesis. Furthermore, we assumed that enteral
release of valine was the same as the amount of ingested valine
hydrolyzed in 0.5 h and that first-pass absorption was 20%
(10, 13). This represents, most likely, an
oversimplification, but it will not change the conclusions drawn in
this study. When different values for the first-pass effects are
brought into the calculations, whole body protein breakdown will
increase proportionally in both HD and HD+ protocols.
Protein intake by CHD patients during dialysis restored the whole
protein balance completely compared with a nondialysis day. The effects
of dietary protein intake on whole body protein synthesis and oxidation
measured in the HD+ protocol were comparable to those in the HD
protocol, i.e., an increase to 128 and 200% of fasting values,
respectively, as shown in Table 3. Furthermore, the effect of protein
intake was comparable between the HD+ and HD protocol with respect to
the rate of appearance of valine, corrected for the infusion of labeled
valine. Protein breakdown was reduced to about one-half the rate
observed during fasting in these patients. It might well be that the
high effectiveness of dietary protein in inhibiting whole body protein
breakdown during dialysis might be overestimated because of the
corrections used to account for the increase of the valine pool. The
valine concentration in plasma during dialysis increased less than on a
nondialysis day. The associated flux of dietary valine to enlarge the
plasma valine pool is thus smaller, and the calculated value of whole
body protein breakdown becomes larger. The values of whole body protein
balance thus represent a minimal estimate under the condition of a
patient during dialysis while consuming a protein-enriched meal.
Recently, Pupim et al. (31) published their study on the
effect of parenteral nutrition during dialysis on whole body and forearm protein metabolism in CHD patients. Infusion of an amino acid
solution, containing dextrose and lipids as well, during dialysis
resulted in an inhibition of whole body protein breakdown and
stimulation of protein synthesis by ~50% each. Although
qualitatively the same, quantitatively there are discrepancies with our
study. As yet, we do not have an explanation. It might be the
consequence of differences in experimental setup or in the model used
in the calculations. Pupim et al. applied an intravenous infusion of amino acids together with dextrose and lipids, whereas we used a
protein- and fat-enriched meal. Similar to our observations during
consumption of a protein-enriched meal, Pupim et al. observed an
increase in the plasma amino acid concentrations in CHD patients during
dialysis as a consequence of parenteral nutrition. It is not clear from
their description of the isotopic model how they corrected for this
increase in pool size.
Substantial amounts of plasma amino acids were lost during dialysis.
Losses of amino acids in dialysate amounted to 7.7 ± 2.1 g
of amino acids during dialysis of fasting patients, similar to
published figures (16, 39, 46). Losses of amino acids were
11.7 ± 1.9 g in patients while consuming a protein-enriched meal. Similar losses were observed by Wolfson et al. (46)
during their infusion of 39.5 g of amino acids with 200 g of
glucose. Essential amino acid concentrations responded differently
during dialysis than nonessential amino acid concentrations. When
fasted patients were dialyzed, plasma essential amino acid
concentration decreased 13% compared with the concentration on a
nondialysis day. The decrease in concentration of plasma nonessential
amino acids was more pronounced (37%). Because body protein is
enriched in essential amino acids compared with plasma amino acids,
breakdown of body protein will result in an increase of essential amino acids relative to nonessential amino acids in plasma, as was shown in
Fig. 3. Consumption of a protein-enriched meal on a nondialysis day
also changed the relative composition of plasma amino acids. The
increase in concentration of plasma essential amino acids (71%) was
more pronounced than the increase of plasma nonessential amino acids
(31%). In view of the amino acid composition of milk proteins, enteral
protein hydrolysis will release essential amino acids in relative
excess to nonessential amino acids. Combining the effects of dialysis
and a protein-enriched meal resulted in a 57% increase in plasma
essential amino acid concentration and in a small increase of plasma
nonessential amino acid (26%) concentrations. Thus, at the end of the
dialysis session during which the patients consumed a protein-enriched
meal, nonessential amino acids were in shortage relative to essential
amino acids. This effect has not been described before. It is tempting
to speculate that a misbalance in plasma free amino acid composition
after dialysis prevents whole body protein metabolism to revert quickly
to its normal, predialysis, condition. Hypothetically, this relatively small derangement in protein metabolism could contribute to
malnutrition over longer periods of time.
Oral intradialytic nutrition by means of a protein-enriched meal
appeared to be an effective treatment for dialysis-induced protein loss
resulting from clearance of plasma amino acids by the dialysis machine.
In the study protocol, we used a protein intake of 0.6 g
protein/kg, comparable to 50% of daily protein intake in this group of
patients. We think that this amount of protein might be too much for
the average dialysis patient during a 4-h dialysis session. Pupim et
al. (31) infused 15 g of amino acids, whereas we
estimated a dietary amino acid influx in the circulation of 39 g,
assuming that 80% of all protein taken was digested during the
dialysis session. The effects of smaller doses of oral protein on whole
body protein metabolism in CHD patients during dialysis are unknown,
but our results show that an oral protein load during dialysis has a
positive effect that is not less than that of the same load given
without dialysis.
In conclusion, we found that consumption of a protein- and
energy-enriched meal abolished the negative effect of dialysis on whole
body protein balance. This offers a possibility for nutritional intervention in preventing protein energy malnutrition. It also shows
that, even though a meal during dialysis may increase the occurrence of
hypotension, it is metabolically useful and should therefore be
standard practice.
| |
ACKNOWLEDGEMENTS |
We appreciate the time from the patients and nursing staff of the
Dialysis Center Groningen.
| |
FOOTNOTES |
Part of this work was presented at the 34th Annual Meeting of the
American Society of Nephrology, San Francisco, CA, 2001.
This work was supported by Grant no. C 97-1694 from the Dutch Kidney Foundation.
Address for reprint requests and other
correspondence: R. Huisman, University Hospital Groningen,
Internal Medicine, Section Nephrology, Hanzeplein 1, 9713 GZ Groningen,
The Netherlands (E-mail: R.M.huisman{at}int.azg.nl).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published January 21, 2003;10.1152/ajpendo.00264.2002
Received 13 June 2002; accepted in final form 15 January 2003.
| |
REFERENCES |
1.
Adibi, SA.
Roles of branched-chain amino acids in metabolic regulation.
J Lab Clin Med
95:
475-484,
1980[ISI][Medline].
2.
Bergstrom, J.
Nutrition and mortality in hemodialysis.
J Am Soc Nephrol
6:
1329-1341,
1995[Abstract].
3.
Berkelhammer, CH,
Baker JP,
Leiter LA,
Uldall PR,
Whittall R,
Slater A,
and
Wolman SL.
Whole-body protein turnover in adult hemodialysis patients as measured by 13C-leucine.
Am J Clin Nutr
46:
778-783,
1987[Abstract/Free Full Text].
4.
Bier, DM.
Intrinsically difficult problems: the kinetics of body proteins and amino acids in man.
Diabetes Metab Rev
5:
111-132,
1989[ISI][Medline].
5.
Borah, MF,
Schoenfeld PY,
Gotch FA,
Sargent JA,
Wolfsen M,
and
Humphreys MH.
Nitrogen balance during intermittent dialysis therapy of uremia.
Kidney Int
14:
491-500,
1978[ISI][Medline].
6.
Brodie, D,
Moscrip V,
and
Hutcheon R.
Body composition measurement: a review of hydrodensitometry, anthropometry, and impedance methods.
Nutrition
14:
296-310,
1998[ISI][Medline].
7.
Buse, MG,
and
Reid SS.
Leucine. A possible regulator of protein turnover in muscle.
J Clin Invest
56:
1250-1261,
1975[ISI][Medline].
8.
Darmaun, D.
Role of nutrients in the regulation of in vivo protein metabolism in humans.
Acta Paediatr Suppl
88:
92-94,
1999[Medline].
9.
Eriksson, S,
Hagenfeldt L,
and
Wahren J.
A comparison of the effects of intravenous infusion of individual branched-chain amino acids on blood amino acid levels in man.
Clin Sci (Colch)
60:
95-100,
1981[Medline].
10.
Forslund, AH,
Hambraeus L,
Olsson RM,
el Khoury AE,
Yu YM,
and
Young VR.
The 24-h whole body leucine and urea kinetics at normal and high protein intakes with exercise in healthy adults.
Am J Physiol Endocrinol Metab
275:
E310-E320,
1998[Abstract/Free Full Text].
11.
Goodship, TH,
Mitch WE,
Hoerr RA,
Wagner DA,
Steinman TI,
and
Young VR.
Adaptation to low-protein diets in renal failure: leucine turnover and nitrogen balance.
J Am Soc Nephrol
1:
66-75,
1990[Abstract/Free Full Text].
12.
Hagenfeldt, L,
Eriksson S,
and
Wahren J.
Influence of leucine on arterial concentrations and regional exchange of amino acids in healthy subjects.
Clin Sci (Colch)
59:
173-181,
1980[Medline].
13.
Hoerr, RA,
Matthews DE,
Bier DM,
and
Young VR.
Leucine kinetics from [2H3]- and [13C]leucine infused simultaneously by gut and vein.
Am J Physiol Endocrinol Metab
260:
E111-E117,
1991[Abstract/Free Full Text].
14.
Hoerr, RA,
Yu YM,
Wagner DA,
Burke JF,
and
Young VR.
Recovery of 13C in breath from NaH13CO3 infused by gut and vein: effect of feeding.
Am J Physiol Endocrinol Metab
257:
E426-E438,
1989[Abstract/Free Full Text].
15.
Husek, P.
Rapid derivatization and gas chromatographic determination of amino acids.
J Chromatogr A
552:
289-299,
1991.
16.
Ikizler, TA,
Flakoll PJ,
Parker RA,
and
Hakim RM.
Amino acid and albumin losses during hemodialysis.
Kidney Int
46:
830-837,
1994[ISI][Medline].
17.
Ikizler, TA,
and
Hakim RM.
Nutrition in end-stage renal disease.
Kidney Int
50:
343-357,
1996[ISI][Medline].
18.
Ikizler, TA,
Pupim LB,
Brouillette JR,
Levenhagen DK,
Farmer K,
Hakim RM,
and
Flakoll PJ.
Hemodialysis stimulates muscle and whole body protein loss and alters substrate oxidation.
Am J Physiol Endocrinol Metab
282:
E107-E116,
2002[Abstract/Free Full Text].
19.
Kopple, JD.
McCollum Award Lecture, 1996: protein-energy malnutrition in maintenance dialysis patients.
Am J Clin Nutr
65:
1544-1557,
1997[Abstract/Free Full Text].
20.
Kulik, W,
Jakobs C,
and
de Meer K.
Determination of extracellular and intracellular enrichments of [1-(13)C]-alpha-ketoisovalerate using enzymatically converted [1-(13)C]-valine standardization curves and gas chromatography-mass spectrometry.
J Chromatogr B Biomed Sci Appl
729:
211-216,
1999[Medline].
21.
Leijssen, DP,
and
Elia M.
Recovery of 13CO2 and 14CO2 in human bicarbonate studies: a critical review with original data.
Clin Sci (Colch)
91:
665-677,
1996[Medline].
22.
Lim, VS,
Bier DM,
Flanigan MJ,
and
Sum PS.
The effect of hemodialysis on protein metabolism. A leucine kinetic study.
J Clin Invest
91:
2429-2436,
1993[ISI][Medline].
23.
Lim, VS,
and
Flanigan MJ.
The effect of interdialytic interval on protein metabolism: evidence suggesting dialysis-induced catabolism.
Am J Kidney Dis
14:
96-100,
1989
TOP
ABSTRACT
INTRODUCTION
![R<SUB>a</SUB><IT>=</IT>[MPEi(V)/MPE(KIVA)<IT>−</IT>1]<IT>×</IT>i(V)](/content/vol284/issue5/fulltext/E954/img001.gif)
![Ox = i<SUB>bic</SUB>(V) × 5 × [100/MPE(KIVA)] (&mgr;mol · kg<SUP>−1</SUP> · h<SUP>−1</SUP>)](/content/vol284/issue5/fulltext/E954/img003.gif)
