Clinical Science (2001) 100, 101–110 (Printed in Great Britain)
Protein and energy metabolism in chronic
bacterial infection: studies in melioidosis
Nicholas I. PATON*, Brian ANGUS†‡, Wipada CHAOWAGUL§, Andrew J.
SIMPSON†‡, Yupin SUPUTTAMONGKOL†, Marinos ELIAR, Graham CALDER¶,
Eric MILNE¶, Nicholas J. WHITE†‡ and George E. GRIFFIN*†
*St. George’s Hospital Medical School, Cranmer Terrace, London SW17 0RE, U.K., †Faculty of Tropical Medicine,
Mahidol University, Bangkok 10400, Thailand, ‡Nuffield Department of Clinical Medicine, John Radcliffe Hospital, Oxford, U.K.,
§Department of Medicine, Sappasitprasong Hospital, Ubon Ratchatani, Thailand, RDunn Clinical Nutrition Centre,
Cambridge CB2 2DH, U.K., and ¶Rowett Research Institute, Aberdeen AB2 9SB, U.K.
A
B
S
T
R
A
C
T
Chronic infection is often accompanied by a wasting process, the metabolic basis of which is not
fully understood. The aims of the present study were to measure protein and energy metabolism
in patients with melioidosis (a serious and antibiotic-refractory Gram-negative bacterial
infection which is endemic in South-East Asia) in order to define the metabolic abnormalities
that might contribute to wasting. Whole-body protein turnover was measured using the
[13C]leucine technique, both in the fasted state and while consuming a high-energy meal. Resting
energy expenditure was measured by indirect calorimetry, and total energy expenditure by the
bicarbonate/urea method. Results were normalized for fat-free mass, as estimated from skinfold
thickness. Protein turnover was increased in melioidosis patients compared with healthy
controls during fasting (170.9 compared with 124.1 µmol:kg−1:h−1 ; P l 0.04), but the net rate of
catabolism (22.2 compared with 20.5 µmol:kg−1:h−1 ; P l 0.77) and the anabolic response to
feeding were similar in the two groups. Resting energy expenditure was higher in melioidosis
patients compared with controls (191.4 and 157.3 kJ:kg−1:day−1 respectively ; P l 0.04), but
total energy expenditure (measured in a separate group of eight patients with melioidosis) was
low (192.1 kJ:kg−1:day−1). In conclusion, this study found no evidence of metabolic causative
factors, such as accelerated net protein catabolism during fasting, a blunted anabolic response to
feeding or increased daily energy expenditure, and therefore suggests that reduced energy
intake is the prime cause of wasting. The observed normal response to feeding should encourage
nutritional approaches to prevent wasting.
INTRODUCTION
Changes in protein and energy metabolism form an
integral part of the host response to infection. In simple
terms, a hypermetabolic state with negative nitrogen and
energy balance usually arises, which leads to muscle
wasting and weight loss [1]. In acute infections the
changes are short-lived and reverse quickly upon recovery from infection [2,3]. Indeed, the response may
confer some benefit to the host : the breakdown of muscle
may supply amino acids for the synthesis of acute-phase
proteins and for use as an energy source to sustain vital
organ function. However, in chronic infections the
protracted metabolic response can cause such severe
muscle wasting and weight loss that physical function is
compromised. In addition, malnutrition may impair
many aspects of the immune response and recovery
process [4].
Key words: energy expenditure, infection, melioidosis, protein metabolism.
Abbreviations: BMI, body mass index ; CRP, C-reactive protein ; FFM, fat-free mass ; MAC, mid-arm circumference ; REE, resting
energy expenditure ; TEE, total energy expenditure.
Correspondence: Dr N. I. Paton, Department of Infectious Diseases, Communicable Disease Centre, Tan Tock Seng Hospital,
Moulmein Road, Singapore 308433 (e-mail PatonINIJ!notes.ttsh.gov.sg).
# 2001 The Biochemical Society and the Medical Research Society
101
102
N. I. Paton and others
The process of wasting may be conceptualized in
several ways. In terms of protein metabolism, there may
be accelerated net catabolism of body protein in the
fasted state, or there may be a block to the effective
utilization of nutrients upon feeding, such that the
nutrition is stored as fat rather than promoting protein
accretion. In terms of energy metabolism, there may be
increased energy requirements or a decrease in energy
intake. One or all of these processes may be operative in
chronic infection, but the few previous studies that have
been performed have not shown a consistent picture.
Melioidosis is a severe infection caused by the Gramnegative bacillus Burkholderia pseudomallei which is
endemic in South-East Asia and particularly common in
North-East Thailand, where the incidence is estimated at
4.4 per 100 000 population [5]. The acute septicaemic
presentation has a high mortality (40 %), even though the
condition is well recognized, the diagnosis can be
confirmed rapidly [6,7] and optimal antibiotic therapy
can be instituted promptly [8]. Sub-acute or chronic
melioidosis, which may follow the acute episode or
which may be the presenting pattern, is characterized by
chronic abscesses, often in the liver or spleen, which
usually require prolonged antibiotic therapy and surgical
drainage to effect a cure. The response to effective
antibiotic therapy is slow. The median time to fever
clearance is 9 days, and in some cases the fever may
persist for over 1 month [9]. There is a high relapse rate
even when oral antibiotics are continued for more than 3
months [10]. This prolonged infected state is often
associated with profound wasting [11], which contributes
to the morbidity and mortality from the disease and may
increase the relapse rate by impairing the immune
response. Melioidosis therefore represents a good model
in which to study the metabolic effects of a chronic
infection.
The aims of the present study were to use isotope
tracer methods to investigate disturbances of protein and
energy metabolism in chronic melioidosis. In particular,
we set out to determine whether there is increased net
protein catabolism in the fasted phase, a block to the
utilization of nutrients in the fed state, or increased total
energy expenditure (TEE), any or all of which might
contribute to the wasting process.
METHODS
Subjects
The study was performed at Sappasitprasong Hospital,
Ubon Ratchathani, North-East Thailand. This is an area
with a high incidence of melioidosis. Consecutive adult
patients hospitalized with melioidosis who had received
less than 2 weeks of effective antibiotic treatment were
studied. A similar group of subjects who had been fully
treated for melioidosis at least 1 year previously were
# 2001 The Biochemical Society and the Medical Research Society
recruited from an out-patient clinic to serve as controls.
Patients with Type I (insulin-dependent) diabetes
mellitus were excluded due to potential problems with
the fasting\feeding protocol, but those with Type II
(non-insulin-dependent) diabetes mellitus were permitted to participate. Patients with significant renal impairment (serum creatinine more than twice the upper
limit of normal) were also excluded from the protein
metabolism study.
The study was conducted in accordance with the
Declaration of Helsinki (1989) of the World Medical
Association, and the protocols were approved by the
Ethical Review Subcommittee of the Research Committee, Ministry of Public Health, Thailand. All subjects
gave informed consent to the study.
Anthropometric measurements and clinical
assessment
Subjects were weighed (to the nearest 0.1 kg) on calibrated digital electronic scales and height was measured
(to the nearest 1 mm) using a portable stadiometer. Body
mass index (BMI) was then calculated as weight (kg)
divided by height (m) squared.
Skinfold thickness was measured at four sites (triceps,
biceps, subscapular and supra-iliac), and the measurements were converted into estimates of body density
using the age- and sex-specific prediction equations of
Durnin and Womersley [12]. Percentage body fat was
calculated from body density using the Siri equation [13],
and fat-free mass (FFM) was then calculated from
percentage fat and body weight. Mid-arm circumference
(MAC) was measured (to the nearest 1 mm) using a metal
tape measure on the non-dominant arm, midway between
the tip of the acromion and olecranon processes.
Body temperature was recorded at 4 h intervals during
the study day. Levels of C-reactive protein (CRP),
albumin and total protein were measured in plasma using
standard automated assays.
Protein turnover
Protein-turnover studies were conducted under controlled conditions and at constant ambient temperature in
the Respiratory Care Unit of the hospital. Whole-body
protein turnover was measured using an 8 h protocol
similar that described originally for assessing the protein
metabolic response to oral feeding in normal subjects [14]
and in patients with lung cancer [15]. An infusion of the
stable-isotope tracer amino acid L-[l-"$C]leucine (99
atom % ; MassTrace, Woburn, MA, U.S.A.) was
administered through a forearm intravenous cannula at a
rate of 2.3 µmol:kg−":h−" after a priming dose of
2.88 µmol\kg (equivalent to 1.25 h of infusion). The
infusion began at approximately 08.00 hours on the study
day and was continued for 8 h. Samples of blood (5 ml)
were taken through a second intravenous cannula sited in
the non-infusion arm, before starting the infusion and at
Metabolism in melioidosis
30-min intervals during the plateau phases (hours 2–4 in
the fasted phase and hours 6–8 in the fed phase). Samples
of breath were collected by transfer into evacuated tubes
at the same time points. Total CO production and O
#
#
consumption were measured over periods of 15–20 min
at intervals during the plateau phases using a ventilated
hood and metabolic monitor (Deltatrac ; Datex
Instrumentarium, Helsinki, Finland).
Subjects were studied after an overnight fast and
remained in a fasting state for the first 4 h of the protocol.
In the following 4 h, they were given regular hourly
drinks made by mixing 0.7 g:kg−" body weight of Ensure
powder (Ross Laboratories, Columbus, OH, U.S.A.)
with sterile water to form a volume of approx. 150 ml.
The composition of Ensure is 61.6 g of carbohydrate,
15.8 g of fat and 15.8 g of protein per 100 g of powder.
The energy intake during feeding was 13.4 kJ:kg−" body
weight:h−", in the proportions 14 % protein, 31.5 % fat
and 54.5 % carbohydrate. This regimen was designed to
provide one-third of daily energy requirements (estimated as 160 kJ:kg−" body weight:day−") over the 4-h
feeding period.
Isotopic enrichments of ["$C]leucine and ["$C]αketoisocaproic acid (the deamination product of leucine)
in serum were measured by GC–MS (VG12–250 ; VG
Masslab, Altrincham, Cheshire, U.K.), and isotopic
enrichment of "$CO was measured by gas isotope ratio
#
MS (SIRA 12 ; VG Isogas Ltd, Middlewich, Cheshire,
U.K.). Breath "$CO enrichment was calculated in the
#
fasting state by subtraction of the enrichment before
infusion of leucine, and in the fed state by subtraction of
the mean enrichment seen in a group of four individuals
receiving an identical oral nutrition regimen but not
receiving ["$C]leucine.
Measurement of TEE
TEE was estimated from total daily CO production,
#
measured by the bicarbonate\urea method as described
previously [16]. In brief, approx. 6–7 ml of a solution of
Na"%CO (5 µCi\ml ; Amersham International, Little
$
Chalfont, Bucks., U.K.) was administered by constant
subcutaneous infusion using a mini-pump syringe driver
(Graseby MS26 Syringe Driver ; Graseby Medical Ltd,
Colonial Way, Watford, U.K.) over a 2-day period
starting and ending in the morning. The exact dose of
bicarbonate infused was determined by weighing the
syringe, extension tube (100 cm Lectro-spiral extension
tube ; Vygon UK Ltd, Cirencester, Gloucester, U.K.) and
syringe driver on an electronic balance sensitive to 0.001 g
immediately before connection to the patient and again
after completing the study. The effective whole-body
dose of radiation during the entire study was estimated to
be less than 1 day’s natural background radiation.
A 22-gauge cannula was inserted subcutaneously on
the abdomen and a 0.5 ml priming dose of a solution of
["%C]urea (0.22 µCi\ml ; Sigma Chemical Co., Poole,
Dorset, U.K.) was injected. The extension tube was then
connected to the cannula and the whole was secured to
the skin with transparent plastic adhesive dressing
(Opsite Flexigrid, 10 cmi12 cm ; Smith and Nephew
Medical Ltd, Hull, U.K.). The infusion was begun and
the exact starting time was noted. The studies were
conducted on free-living patients in the general ward
with culture-proven melioidosis. The pump was placed
in a cloth pouch and placed beside the patient on the bed.
When the patient wished to walk about, the pouch was
carried in the hand or in a sling around the neck. A
normal hospital diet was provided throughout the period
of the study, but details were not recorded.
Urine was collected into a 3-litre bottle to which 5 ml
of chlorhexidine had been added as a preservative. Urine
collection was started after the infusion had been running
for 18–24 h, beginning after the first morning urine had
been passed, and continuing for 24 h to finish with the
inclusion of the first morning specimen of the following
day. The bottle was kept by the patient’s bedside, and
they were instructed to collect all urine passed over the
24 h period. The collection was monitored by the ward
nursing staff, and the investigator also asked the patients
to verify at the end of the time period that they had
achieved a complete collection of urine. Where there was
doubt about the quality of the collection, the samples
were not processed further. Aliquots were taken from the
complete urine collections and were stored at k20 mC
until analysed.
Analysis of the specific radioactivity of "%CO trapped
#
in urinary urea was performed as described previously
[16]. In brief, after removal of acid-labile CO dissolved
#
in the urine, urease was added and the CO released from
#
urea was sequestered by hyamine hydroxide. The specific
radioactivity of the trapped "%CO was determined by
#
scintillation counting.
Calculations
Whole-body protein turnover
Rates of whole-body leucine metabolism were calculated
as described previously [14,17]. Leucine flux was calculated by dividing the rate of tracer infusion by the plateau
serum leucine enrichment. Leucine oxidation was
obtained by multiplying the rate of CO production by
#
the breath "$CO enrichment at plateau (adjusted for
#
incomplete recovery using previously published correction factors of 0.74 and 0.87 for the fasted and fed
phases respectively) [14,17] and then dividing by the
serum enrichment of ["$C]α-ketoisocaproic acid at plateau as the precursor for leucine oxidation. The rate of
leucine infusion was subtracted from the oxidation rate
to give a value for endogenous oxidation. Protein
synthesis was calculated as the difference between the
flux and the rate of oxidation. Protein breakdown was
calculated as the difference between the flux and the
# 2001 The Biochemical Society and the Medical Research Society
103
104
N. I. Paton and others
intake of leucine from the infusion and the diet. Net
protein catabolism or anabolism was calculated as the
difference between protein synthesis and protein breakdown ; the net catabolic rate was equivalent, therefore, to
the rate of oxidation in the fasted phase, and the net
anabolic rate was equivalent to the difference between
oxidation rate and intake in the fed phase. Leucine kinetic
data were normalized using as the denominator FFM
obtained by skinfold thickness measurements.
The sample size was calculated for the primary endpoint of net protein balance in the fed phase, as the
principal hypothesis of interest was whether there was
any evidence of anabolic block. Data obtained previously
using a similar protocol in patients with tuberculosis
showed a mean fed balance of 24.5p9.3 µmol:kg−":h−".
To detect a 50 % difference in mean protein balance
between groups using a two-sample t-test, with type I
and II errors of 5 % and 20 % respectively, a sample size
of nine subjects per group was required.
Substrate utilization
Substrate utilization was calculated from measurements
of O consumption, CO production and protein oxi#
#
dation rate (calculated from leucine kinetics) in the postabsorptive and fed states using a stoichiometric approach
[17,18]. The factor for converting leucine oxidation into
protein oxidation was taken as 0.61 mmol of leucine\g of
protein (leucine content of body protein) during fasting,
and 0.674 mmol of leucine\g of protein (leucine content
of Ensure) during feeding. Results were expressed in
terms of both the amount of each substrate utilized and
the relative contribution to energy supply, assuming
energy values of 15.7, 39.1 and 18.4 kJ\g for carbohydrate
(as monosaccharide), fat and protein respectively.
Energy expenditure
Resting energy expenditure (REE) was calculated from
the measurements of CO production and O con#
#
sumption made during the fasting phase of the protein
turnover protocol, using the following formula [19] :
REE (kJ) l 15.818O j5.176CO
#
#
Predicted REE values were calculated from Schofield’s
equations, which are based on height and weight for
appropriate age categories [20].
Total daily CO production was calculated from the
#
results of the bicarbonate\urea measurements, using the
following equation [16] :
CO production (mol\day) l [0.95i0.85iinfused
#
bicarbonate (d.p.m.\day)]\[specific radioactivity
(d.p.m.\mol)].
TEE was calculated from net CO production using an
#
energy equivalent of 559 kJ\mol of CO [21], assuming a
#
respiratory quotient of 0.80, which is appropriate for
subjects in moderate negative energy balance.
Statistical analysis
Protein turnover rates, substrate utilization rates and
REE were compared between melioidosis patients and
controls using Student’s t-test. The response of protein
metabolism to feeding was assessed within each group by
paired t-test and between groups by unpaired t-test. The
relationships between clinical, laboratory and metabolic
parameters were assessed by calculation of Pearson’s
correlation coefficient.
# 2001 The Biochemical Society and the Medical Research Society
RESULTS
Subject characteristics
Measurements of protein metabolism and REE were
performed on nine patients with melioidosis. The sites of
infection were pneumonia (2), arthritis of the knee (2),
osteomyelitis of the tibia (1), pericarditis (1), liver and
spleen abscess (1), splenic abscess alone (1) and septicaemia (1). B. pseudomallei was grown in bacteriological
cultures for eight of these patients, and melioidosis was
considered the most likely diagnosis on clinical grounds
in the remaining patient. This patient had sterile blood
cultures and a splenic abscess which could not be
aspirated safely, and responded to appropriate antibiotic
therapy for melioidosis. The mean duration of symptoms
before the study was 25 days (range 13–50 days), and the
mean duration of antibiotic therapy (ceftazidime in six,
imipenem in two, and amoxicillin\clavulanic acid in one
patient) before the study was 8 days. All the patients were
febrile on the day of study (mean maximum temperature
on the study day was 38.4 mC).
Nine control subjects were also studied. They had been
treated for melioidosis an average of 4 years previously
(range 1–7 years), and all were well and free from symptoms of infection at the time of study. The underlying
risk factors, sex distribution and age were similar in patient
and control groups (Table 1). Indices of nutritional
status, such as BMI and MAC, were lower in the patients
with current melioidosis than in control subjects, but the
differences were not significant (Table 1). Serum albumin
was significantly lowered and CRP was significantly
increased in the patients with melioidosis (Table 1).
TEE was measured in a separate group of eight patients,
all of whom had culture-proven melioidosis. The sites of
infection were pneumonia (3), liver and spleen abscess
(2), kidney (2), septicaemia and splenic abscess (1) and
buttock abscess (1). The duration of antibiotic therapy (9
days), underlying risk factors (diabetes, 5 ; chronic renal
impairment, 1), sex distribution (four males\four
females), age (44 years) and anthropometric characteristics (weight 45.7p10.3 kg ; BMI 18.7p4.2 kg\m# ;
FFM 36.5p7.9 kg) were similar to those of the melioidosis patients participating in the protein metabolism
study (see above and Table 1).
Metabolism in melioidosis
Table 1 Clinical and anthropometric characteristics and
selected laboratory parameters for the study subjects
Values are meanspS.D. (n l 9 in each group).
Parameter
Controls
Melioidosis
patients
P
Type II diabetes (n )
Renal impairment (n )
Thalassaemia (n )
Sex (male/female)
Age (years)
Height (cm)
Weight (kg)
BMI (kg/m2)
FFM (kg)
Fat (kg)
MAC (cm)
Albumin (g/l)
CRP (mg/l)
Total protein (g/l)
7
1
2
4/5
47.9p12.9
155.5p4.2
49.7p8.3
20.6p3.4
37.2p5.8
12.5p5.7
26.0p3.6
34.1p3.9
18.1p15.0
81.1p5.4
5
0
2
4/5
40.2p11.8
159.7p6.7
47.0p8.6
18.4p3.1
35.9p5.7
11.1p5.7
24.5p2.9
22.4p3.6
89.0p74.3
77.6p12.5
–
–
–
–
0.21
0.13
0.51
0.18
0.65
0.60
0.33
0.001
0.01
0.44
Table 2 Protein metabolism parameters in the fasted and
fed states
Data are meanspS.D. (n l 9 in each group). Values are expressed in terms of
leucine flux per unit FFM, as measured by skinfolds.
Value (µmol:kg−1:h−1)
Parameter
Controls
Melioidosis
patients
P
Fasted state
Flux
Synthesis
Breakdown
Oxidation
Net catabolism
124.1p27.2
103.6p20.0
124.1p27.2
20.5p10.9
k20.5p10.9
170.9p58.3
148.7p50.0
170.9p58.3
22.2p13.1
k22.2p13.1
0.044
0.02
0.044
0.77
0.77
Fed state
Flux
Synthesis
Breakdown
Oxidation
Net anabolism
149.0p28.2
86.8p22.0
48.7p34.7
62.1p25.7
38.2p29.7
202.5p54.3
133.5p46.1
100.5p52.6
69.0p19.6
33.0p33.0
0.02
0.015
0.025
0.54
0.73
24.9p11.4
k18.6p22.1
k75.4p16.0
41.7p20.7
58.6p24.1
31.6p18.7
k15.2p20.3
k70.4p21.3
46.8p19.6
55.2p32.3
Fed–fasted change
Flux
Synthesis
Breakdown
Oxidation
Net anabolic response
0.37
0.87
0.58
0.59
0.80
Protein metabolism
The results of protein metabolism measurements are
shown in Table 2. In the fasted state, the rate of protein
turnover, as indicated by leucine flux, was increased by
Table 3
Substrate utilization in the fasted and fed states
Values are meanspS.D. (n l 9 in each group). Values are expressed as the
amount of each substrate used per kg FFM, as measured by skinfolds, and in terms
of the percentage contribution to REE at rest. Balance is the net utilization in the
fed phase after subtraction of intake. Negative values indicate storage.
Parameter
Fasted state
Utilization (g:kg−1:day−1)
Carbohydrate
Fat
Protein
Contribution to REE ( %)
Carbohydrate
Fat
Protein
Controls
Melioidosis
patients
P
3.34p1.01
2.32p0.84
0.80p0.43
1.92p2.12
3.64p1.22
0.87p0.52
0.09
0.016
0.77
34.3p12.9
56.2p14.8
9.5p4.8
16.7p17.3
75.1p17.2
8.2p3.6
0.027
0.024
0.53
4.99p1.52
1.84p0.97
2.21p0.92
3.53p2.62
3.08p1.31
2.46p0.70
0.17
0.036
0.53
41.8p13.4
37.1p16.1
21.1p6.9
25.6p18.3
54.1p18.4
20.4p5.0
0.048
0.053
0.79
Fed state (balance)
Utilization (g:kg−1:day−1)
Carbohydrate
k8.79p2.21
Fat
k1.70p1.1
Protein
k1.32p1.06
k10.48p4.00
k0.51p1.32
k1.14p1.17
Fed state (total)
Utilization (g:kg−1:day−1)
Carbohydrate
Fat
Protein
Contribution to REE ( %)
Carbohydrate
Fat
Protein
0.28
0.05
0.73
38 % in the melioidosis patients compared with the
controls (170.9 and 124.1 µmol:kg−":h−" respectively ; P
l 0.044). The rates of protein synthesis and breakdown
were increased to an equal degree, so that the rate of
protein oxidation, and hence the net protein catabolic
rate, did not differ between the melioidosis and control
groups.
In the fed phase, protein turnover was increased by
36 % in the melioidosis patients compared with the
controls (202.5 and 149.0 µmol:kg−":h−" respectively ; P
l 0.02). Synthesis and breakdown were increased in
parallel, so that the rate of protein oxidation and the rate
of net protein anabolism did not differ between the
melioidosis and control patients.
Upon feeding, both groups showed a significant
decrease in the rate of protein breakdown (P 0.01) and
a significant increase in the rate of protein oxidation (P
0.01), but the rate of protein synthesis was unchanged (P
0.05). The magnitude of these changes, as well as the
switch from net catabolism to net anabolism, was the
same in both groups.
Rates of substrate utilization are shown in Table 3. In
# 2001 The Biochemical Society and the Medical Research Society
105
106
N. I. Paton and others
Table 4
Energy expenditure parameters
Values are meanspS.D. (n l 9 in each group). Predicted REE was obtained using
Schofield’s equations [20]. The predicted REE for the nine patients who underwent
REE measurements was 5.31p0.56 MJ/day, and that for the eight patients who
underwent TEE measurements was 5.37p0.80 MJ/day.
Melioidosis
patients
P
REE absolute (MJ/day)
5.83p1.09
REE/body weight (kJ:kg−1:day−1) 119.3p25.8
REE/FFM (kJ:kg−1:day−1)
157.3p21.8
REE absolute/predicted REE ( %)
108.4p19.2
6.77p1.20
148.1p35.6
191.4p38.7
128.7p26.0
0.10
0.067
0.04
0.08
TEE absolute (MJ/day)
TEE/body weight (kJ:kg−1:day−1)
TEE/FFM (kJ:kg−1:day−1)
TEE absolute/predicted REE
6.89p1.41
153.3p24.9
192.1p30.7
1.29p0.20
–
–
–
–
Energy parameter
Controls
–
–
–
–
Figure 1 Protein turnover (in the fasted state) against
temperature in melioidosis patients
the fasted phase, patients with melioidosis derived more
energy from fat and less from carbohydrate than did the
controls. There was no difference in the contribution of
protein oxidation to energy requirements. A similar
pattern was seen in the fed phase. There was net storage
of all three substrates in the fed phase (indicated by
negative values for utilization), and the pattern was
similar in the two groups, although the melioidosis
patients appeared to store less fat than the controls (P l
0.05).
Energy metabolism
The results of energy expenditure measurements are
shown in Table 4. In the melioidosis patients, REE was
elevated by a mean of 28 % above predicted levels (P l
0.009) and by 22 % above the control values normalized
for FFM (P l 0.04).
The mean value of TEE in melioidosis patients was
6.89p1.41 MJ\day (range 4.84–9.58 MJ\day). For this
group of subjects, the mean predicted REE was
5.37p0.80 MJ\day and the ratio of TEE to predicted
REE was 1.29p0.20 (range 1.12–1.69).
# 2001 The Biochemical Society and the Medical Research Society
Figure 2 Protein oxidation (in the fasted state) against BMI
in melioidosis patients
Relationships between metabolic, clinical
and laboratory parameters
There was no significant correlation between any metabolic variable and the nutritional and clinical variables in
the control group. In the melioidosis patients, flux in the
fasted phase was significantly correlated with CRP (r l
0.70, P l 0.04) and fever (r l 0.825, P l 0.006), as shown
in Figure 1, but not with BMI (r lk0.493, P l 0.18) or
MAC (r lk0.26, P l 0.49). Oxidation in the fasted
phase showed a significant negative correlation with
indices of nutritional status (BMI, r lk0.71, P 0.05 ;
MAC, r lk0.64, P l 0.06), as shown in Figure 2, but
not with indices of inflammation (CRP, r l 0.34, P l
0.36 ; fever, r l 0.38, P l 0.31). A similar pattern was
seen in the fed phase, with protein turnover showing
relationships with parameters of inflammation (fever, r l
0.92, P l 0.001 ; CRP, r l 0.67, P l 0.047), but not with
nutritional parameters. Fed-phase protein oxidation had
a weak relationship with some nutritional parameters
(BMI, r lk0.60, P l 0.09 ; MAC, r lk0.58, P l 0.10)
and a significant relationship with fever (r l 0.68, P l
0.04).
In the melioidosis patients, normalized REE was
significantly correlated with protein flux (r l 0.70, P l
0.02) and oxidation (r l 0.73, P l 0.02) in the fasted
phase.
DISCUSSION
We have shown that patients with melioidosis, a serious
systemic Gram-negative bacterial infection characterized
by abscess formation and a marked inflammatory response, had increased rates of protein turnover (the
continuous cycling between breakdown and re-synthesis
of proteins) in comparison with a group of control
subjects without active infection. The increased turnover
was found in both the fasted and fed states. Accelerated
whole-body protein turnover is a well recognized feature
of acute infection, and has been described in patients with
Metabolism in melioidosis
surgical sepsis due to heterogeneous bacterial pathogens
[22,23], as well as in patients with specific acute infections
such as measles [24] and malaria [25]. The findings of the
present study, together with several previous studies that
have demonstrated increased turnover in HIV infection
[26,27], suggest that increased turnover is also a feature of
sub-acute and chronic infection. However, one study of
HIV infection [28] and another of tuberculosis [29]
found reduced or normal protein turnover. The reason
for the variability in chronic infection is unknown,
although severe malnutrition may attenuate the response
to infection [24], and variations in cytokine secretion may
also play a role, as discussed below.
The cause of this increase in protein turnover is unclear.
We found a close relationship (high correlation) between
the maximum body temperature on the day of the study
and the magnitude of the elevation of protein turnover, as
noted previously in some studies of acute infection
[24,25]. As well as a direct effect of temperature on
protein metabolism, which has been documented in
animal studies [30], this relationship probably also
represents a parallel response to one or more inflammatory mediators, such as the pro-inflammatory cytokines tumour necrosis factor, interleukin-1 and
interleukin-6, which are thought to affect protein turnover [3]. Although we did not measure cytokines in the
present study, previous studies have shown that levels of
tumour necrosis factor [31], interleukin-6, interleukin-8
(the latter possibly reflecting interleukin-1 secretion)
[32], interferon-γ and soluble interleukin-2 receptors [33]
are elevated considerably in patients with melioidosis. A
large part of the increased turnover may be accounted for
by increased synthesis of acute-phase reactants. This is
supported by our finding of a close relationship between
the CRP level and the rate of protein synthesis (r l 0.72)
in both the fasted and the fed phases.
The accelerated turnover in the fasted state comprised
almost equal increases in the rates of synthesis and
breakdown, so that the rate of oxidation and net protein
catabolism were no different from those in the controls.
This contrasts with the classical description of changes
occurring in the stressed state [1] and in many studies of
acute infections [23], where protein breakdown is
increased more than protein synthesis, so that there is an
increase in the net loss of nitrogen. However, our findings
are consistent with several previous studies of chronic
infections, such as HIV infection [26], chronically
infected cystic fibrosis patients [34] and tuberculosis [29],
which have shown normal oxidation and net protein
catabolism in the fasted phase. This suggests that subacute or chronic infection may be distinguished from
other stress states by an adaptive response that serves to
limit oxidation and thereby attenuate nitrogen loss. This
phenomenon is well illustrated by an experiment (conducted over half a century ago) in which repeated
episodes of malaria were induced in a group of patients as
treatment for meningovascular syphilis. There was a
progressive decrease in the magnitude of negative nitrogen balance with each febrile episode [35]. The trigger
for this adaptive response might be the prolonged
reduction in energy and\or protein intake caused by the
anorexia associated with chronic infection, or perhaps an
alteration in the pattern of cytokine secretion during the
course of chronic infection. The significant negative
correlation which we found between the protein oxidation rate and BMI is of interest, suggesting that the
adaptive response is more successful in better-nourished
patients, perhaps because such patients are more readily
able to switch to utilization of fat for energy supply in
times of need. This is also consistent with the observation
that fatter people lose a lower proportion of lean tissue
(and, by inference, have lower protein oxidation rates)
than do lean people during starvation [36]. This present
study is limited, as are all previous stable isotope studies
of protein metabolism in chronic infections, by its crosssectional design. Longitudinal studies, with repeated
measurements at various stages of chronic infection,
might prove valuable in defining further the adaptive
changes that may occur during chronic infection.
The protocol used in the present study permitted us to
examine specifically the effects of feeding on protein
metabolism. We found a similar response to feeding in
patients and controls, with the primary change being a
decrease in the rate of protein breakdown, as has been
shown previously in normal individuals [14] and patients
with lung cancer [15]. Although fat storage was significantly reduced (reflecting the increased reliance on
oxidization of fat for meeting energy requirements during
sepsis), the net storage of protein was quantitatively
normal in patients with melioidosis. This is in keeping
with studies in which parenteral nutrition led to a normal
protein anabolic response in patients with HIV infection
[26,27], but contrasts with a study of enteral nutrition in
tuberculosis, which appeared to show an ‘ anabolic block ’
to the utilization of nutrients [29]. It may be that
tuberculosis has unique properties not shared by other
chronic infections : indeed, there is some evidence from
body composition studies in patients with HIV infection
suggesting that concomitant tuberculosis may result in a
‘ hypermetabolic ’ pattern of wasting (excessive lean tissue
loss), as opposed to the ‘ starvation ’ pattern (lean and fat
tissue loss seen in balanced proportions) observed with
other opportunistic infections [37,38]. In addition, a
small longitudinal study of body composition found that
HIV patients with secondary infections (including mycobacterial infections) fail to accrue lean tissue when fed
with parenteral nutrition [39]. Nevertheless, in patients
with melioidosis (and probably in most other chronic
infection states), the normal anabolic response to nutrition provides a rationale for the provision of additional
feeding to ensure that adequate energy and protein intake
is maintained.
# 2001 The Biochemical Society and the Medical Research Society
107
108
N. I. Paton and others
We also found that REE was significantly elevated in
the patients with melioidosis. The magnitude of the
increase (22 % above control values) is similar to that seen
in patients with other chronic infections, such as tuberculosis (21 % increase) [29] and stable patients with
HIV infection (10 % increase) [40]. There was a
significant correlation (r l 0.7) between fasting protein
turnover and REE, suggesting that the increased REE
may be attributable in part to the energy cost of increased
protein turnover. In contrast with the raised REE, TEE
(which includes the energy produced by feeding and the
energy expenditure due to physical activity) appeared
relatively low : the ratio of TEE to predicted REE was
1.29, in comparison with the ratios of approx. 1.5–1.8 that
are characteristic of lightly to moderately active healthy
men and women [41]. If an REE of 28 % above predicted
values is used for the denominator, then the ratio would
be even lower (close to 1.0). These low ratios indicate that
patients with melioidosis have a dramatic decrease
(indeed, a virtual abolition) of physical activity. This is in
keeping with the clinical observation that the patients in
the present study, although free to move about as desired,
spent the majority of their time completely immobile in
bed. A similar pattern of raised REE and low TEE has
been observed in patients with HIV infection [42,43], and
this may be characteristic of chronic infection in general.
Energy requirements are traditionally considered to be
increased in the presence of fever, but the present study
suggests that the converse may be true in patients with
chronic infection. Studies of TEE during infection have
been restricted, until recently, by the availability of only
one technique for measurement : the double-labelled
water method. This is expensive, requires access to
sophisticated laboratory equipment for analysis, and is of
limited accuracy in very hot climates due to the rapid
turnover of body water. In contrast, the newer
bicarbonate\urea method is cheap, relatively simple to
analyse and is unlikely to be affected by climate. The
present study is the first to apply the method in a tropical
setting, and we found the technique to be feasible and to
give plausible results. This should encourage further
studies using this novel method.
A large number of the patients and control subjects
had Type II diabetes mellitus, which was expected, as
diabetes is known to be the major risk factor for
melioidosis [5,44]. Although Type I diabetes mellitus is
associated with increased protein oxidation when diabetic
control is inadequate [45], it has been shown that patients
with Type II diabetes have normal protein metabolism, at
least in the fed state [46]. As (1) we excluded patients with
Type I diabetes, (2) there were equal numbers of patients
with Type II diabetes in each group, and (3) there was no
difference between diabetics and non-diabetics in any
protein turnover parameter, it is unlikely that this factor
had any major effect on the results. Although all the
nutritional parameters were lower in the patients with
# 2001 The Biochemical Society and the Medical Research Society
melioidosis compared with the controls, only the decrease in serum albumin was statistically significant. This
reflects both the limitations of the methods used for the
measurement of nutritional status (an expansion of
extracellular water would minimize the observed decrease in FFM) [37], and perhaps the fact that patients
were studied relatively early in the course of infection.
Wasting is likely to become marked only after many
weeks of persistent inflammation.
These protein and energy metabolism studies have
effectively refuted all three of the metabolic abnormalities
which we hypothesized might be present in melioidosis
(increased net protein catabolism in the fasted phase, a
blunted anabolic response to feeding, and increased
TEE). The logical conclusion from this pattern of normal
protein metabolism and reduced energy requirements is
that reduced energy and\or protein intake is the principal
cause of wasting associated with this chronic infection.
Although we did not measure food intake in the patients
and controls (this would have been very difficult to
perform with worthwhile accuracy in this situation), this
conclusion is supported by the simple clinical observation
that patients with melioidosis, as with many other
infections, are often anorectic and therefore reduce their
food intake. Taken together with the observed normal
anabolic response to feeding, this provides strong support
for the use of nutritional approaches to prevent or reverse
the wasting associated with chronic infections such as
melioidosis. The increased protein turnover may represent futile cycling or possibly destruction of muscle
protein to fuel synthesis of acute-phase proteins. Therefore attempts to attenuate this response, with cytokine
antagonists for example, may also be of benefit in
reducing energy requirements and preserving muscle
mass. The dramatic decrease in physical activity during
chronic infection may also contribute to muscle wasting,
and represents another potential target for therapy.
ACKNOWLEDGMENTS
We thank the staff of Sappasitprasong Hospital for their
support, Graham Jennings for performing the analysis
for the bicarbonate\urea studies, Vanaporn Wuthiekanun, Paul Howe and Philippa Newton for microbiological and logistical support, Derek Macallan for
providing equations for the calculation of substrate
utilization, and Sanjeev Krishna for practical advice and
for providing the initial impetus for the study. This work
was supported by The Wellcome Trust of Great Britain.
REFERENCES
1 Cuthbertson, D. P. (1932) Observations on the disturbance
of metabolism produced by injury to the limbs. Q. J. Med.
25, 233–244
Metabolism in melioidosis
2 Beisel, W. R., Sawyer, W. D., Ryll, E. D. and Crozier, D.
(1967) Metabolic effects of intracellular infections in man.
Ann. Int. Med. 67, 744–779
3 Chang, H. R. and Bistrian, B. (1998) The role of cytokines
in the catabolic consequences of infection and injury.
J. Parenter. Enteral Nutr. 22, 156–166
4 Chandra, R. K. (1983) Nutrition, immunity and infection :
present knowledge and future directions. Lancet i, 688–691
5 Suputtamongkol, Y., Hall, A. J., Dance, D. A. B. et al.
(1994) The epidemiology of melioidosis in Ubon
Ratchatani, northeast Thailand. Int. J. Epidemiol. 23,
1082–1090
6 Dance, D. A., Wuthiekanun, V., Naigowit, P. and White,
N. J. (1989) Identification of Pseudomonas pseudomallei in
clinical practice : use of simple screening tests and API
20NE. J. Clin. Pathol. 42, 645–648
7 Wuthiekanun, V., Dance, D., Chaowagul, W.,
Suputtamongkol, Y., Wattanagoon, Y. and White, N.
(1990) Blood culture technique for the diagnosis of
melioidosis. Eur. J. Clin. Microbiol. Infect. Dis. 9, 654–658
8 White, N. J., Dancem, D. A. B., Chaowagul, W.,
Wattanagoon, Y., Wuthiekanun, V. and Pitakwatchara, N.
(1989) Halving of mortality of severe melioidosis by
ceftazidime. Lancet ii, 697–701
9 Chaowagul, W., White, N. J., Dance, D. A. B. et al. (1989)
Melioidosis : a major cause of community acquired
septicaemia in Northeastern Thailand. J. Infect. Dis. 159,
890–899
10 Chaowagul, W., Suputtamongkol, Y., Dance, D. A.,
Rajchanuvong, A., Pattara-arechachai, J. and White, N. J.
(1993) Relapse in melioidosis : incidence and risk factors.
J. Infect. Dis. 168, 1181–1185
11 White, N. J. (1994) Melioidosis. Zentralbl. Bakteriol. 280,
439–443
12 Durnin, J. V. G. A. and Womersley, J. (1974) Body fat
assessed from total body density and its estimation from
skinfold thickness : measurements on 481 men and women
aged from 16 to 72 years. Br. J. Nutr. 32, 77–97
13 Siri, W. E. (1956) The gross composition of the body.
Adv. Biol. Med. Phys. 4, 239–280
14 Melville, S., McNurlan, M. A., McHardy, K. C. et al.
(1989) The role of degradation in the acute control of
protein balance in adult man ; failure of feeding to
stimulate protein synthesis as assessed by [1–"$C]leucine
infusion. Metab. Clin. Exp. 38, 248–255
15 Melville, S., McNurlan, M. A., Calder, A. G. and Garlick,
P. J. (1990) Increased protein turnover despite normal
energy metabolism and responses to feeding in patients
with lung cancer. Cancer Res. 50, 1125–1131
16 Elia, M., Jones, G., Jennings, G. et al. (1995) Measurement
of energy expenditure by a new tracer method. Am. J.
Physiol. 269, E172–E182
17 Garlick, P. J., McNurlan, M. A., McHardy, K. C. et al.
(1987) Rates of nutrient utilisation in man measured by
combined respiratory gas analysis and stable isotopic
labelling : effect of food intake. Hum. Nutr. Clin. Nutr.
41C, 177–191
18 Garlick, P. J. (1987) Evaluation of the formulae for
calculating nutrient utilisation rates from respiratory gas
measurements in fed subjects. Hum. Nutr. Clin. Nutr.
41C, 165–176
19 Elia, M. and Livesey, G. (1992) Energy expenditure and
fuel selection in biological systems : the theory and practice
of calculations based on indirect calorimetry and tracer
methods. World Rev. Nutr. Diet. 70, 68–131
20 Schofield, W. N. (1985) Predicting basal metabolic rate :
new standards and review of previous work. Hum. Nutr.
Clin. Nutr. 39C (Suppl. 1), 5–41
21 Elia, M. (1991) Energy equivalents of CO2 and their
importance in assessing energy expenditure when using
tracer techniques. Am. J. Physiol. 260, E75–E88
22 Long, C. L., Jeevanandam, M., Kim, B. M. and Kinney,
J. M. (1977) Whole body protein synthesis and catabolism
in septic man. Am. J. Clin. Nutr. 30, 1340–1344
23 Shaw, J. H. F., Wildbore, M. and Wolfe, R. R. (1987)
Whole body protein kinetics in severely septic patients :
the response to glucose infusion and total parenteral
nutrition. Ann. Surg. 205, 288–294
24 Tomkins, A. M., Garlick, P. J., Schofield, W. N. and
Waterlow, J. C. (1983) The combined effects of infection
and malnutrition on protein metabolism in children.
Clin. Sci. 65, 313–324
25 Berclaz, P.-Y., Benedek, C., Jequier, E. and Schutz, Y.
(1996) Changes in protein turnover and resting energy
expenditure after treatment of malaria in Gambian
children. Pediatr. Res. 39, 401–409
26 Macallan, D. C., McNurlan, M. A., Milne, E., Calder,
A. G., Garlick, P. J. and Griffin, G. E. (1995) Whole-body
protein turnover from leucine kinetics and the response to
nutrition in human immunodeficiency virus infection.
Am. J. Clin. Nutr. 61, 818–826
27 Selberg, O., Suttmann, U., Melzer, A. et al. (1995) Effect
of increased protein intake and nutritional status on
whole-body protein metabolism of AIDS patients with
weight loss. Metab. Clin. Exp. 44, 1159–1165
28 Stein, T. P., Nutinsky, C., Condoluci, D., Schluter, M. D.
and Leskiw, M. J. (1990) Protein and energy substrate
metabolism in AIDS patients. Metab. Clin. Exp. 39,
876–881
29 Macallan, D. C., McNurlan, M. A., Kurpad, A. V. et al.
(1998) Whole body protein metabolism in human
pulmonary tuberculosis and undernutrition : evidence for
anabolic block in tuberculosis. Clin. Sci. 94, 321–331
30 Baracos, V. E., Wilson, E. J. and Goldberg, A. L. (1984)
Effects of temperature on protein turnover in isolated rat
skeletal muscle. Am. J. Physiol. 246, C125–C130
31 Suputtamongkol, Y., Kwaitowski, D., Dance, D. A. B.,
Chaowagul, W. and White, N. J. (1992) Tumour necrosis
factor in septicaemic melioidosis. J. Infect. Dis. 165,
561–564
32 Friedland, J. S., Suputtamongkol, Y., Remick, D. G. et al.
(1992) Prolonged elevation of interleukin-8 and
interleukin-6 concentrations in plasma and leukocyte
interleukin-8 mRNA levels during septicaemic and
localised Pseudomonas pseudomallei infection. Infect.
Immun. 60, 2402–2408
33 Brown, A. E., Dance, D. A., Suputtamongkol, Y. et al.
(1991) Immune cell activation in melioidosis : increased
serum levels of interferon-gamma and soluble interleukin-2
receptors without change in soluble CD8 protein. J. Infect.
Dis. 163, 1145–1148
34 Morton, R. E., Hutchings, J., Halliday, D., Rennie, M. J.
and Wolman, S. L. (1988) Protein metabolism during
treatment of chest infection in patients with cystic fibrosis.
Am. J. Clin. Nutr. 47, 214–219
35 Howard, J. E., Bigham, R. S. J. and Mason, R. E. (1946)
Studies on convalescence. V. Observations on the altered
protein metabolism during induced malarial infections.
Trans. Assoc. Am. Physicians 59, 242–256
36 Forbes, G. B. (1987) Lean body mass-body fat
interrelationships in humans. Nutr. Rev. 45, 225–231
37 Paton, N. I., Castello-Branco, L. R. R., Jennings, G. et al.
(1999) Impact of tuberculosis on the body composition of
HIV-infected men in Brazil. J. Acquired Immune Defic.
Syndr. 20, 265–271
38 Paton, N., Macallan, D., Jebb, S. et al. (1997) Longitudinal
changes in body composition measured with a variety of
methods in patients with AIDS. J. Acquired Immune
Defic. Syndr. 14, 119–127
39 Kotler, D. P., Tierney, A. R., Culpepper-Morgan, J. A.,
Wang, J. and Pierson, Jr., R. N. (1990) Effect of home total
parenteral nutrition on body composition in patients with
acquired immunodeficiency syndrome. J. Parenter. Enteral
Nutr. 14, 454–458
40 Melchior, J. C., Salmon, D., Rigaud, D. et al. (1991)
Resting energy expenditure is increased in stable,
malnourished HIV-infected patients. Am. J. Clin. Nutr.
53, 437–441
41 Food and Agriculture Organisation\World Health
Organisation\United Nations University (1985)
Energy and protein requirements. W.H.O. Tech. Rep. Ser.
724
42 Paton, N. I. J., Elia, M., Jebb, S. A., Jennings, G., Macallan,
D. C. and Griffin, G. E. (1996) Total energy expenditure
and physical activity measured with the bicarbonate-urea
method in patients with human immunodeficiency virus
infection. Clin. Sci. 91, 241–245
# 2001 The Biochemical Society and the Medical Research Society
109
110
N. I. Paton and others
43 Macallan, D. C., Noble, C., Baldwin, C. et al. (1995)
Energy expenditure and wasting in human immunodeficiency virus infection. N. Engl. J. Med. 333, 83–88
44 Suputtamongkol, Y., Chaowagul, W., Chetchotisakd, P. et
al. (1999) Risk factors for melioidosis and bacteremic
melioidosis. Clin. Infect. Dis. 29, 408–413
45 Lariviere, F., Kupranycz, D. B., Chiasson, J. L. and
Hoffer, L. J. (1992) Plasma leucine kinetics and urinary
nitrogen excretion in intensively treated diabetes mellitus.
Am. J. Physiol. 263, E173–E179
46 De Feo, P. (1998) Fed state protein metabolism in diabetes
mellitus. J. Nutr. 128, 328S–332S
Received 4 February 2000/17 August 2000; accepted 29 September 2000
# 2001 The Biochemical Society and the Medical Research Society