Clinical Science ( 1988) 75,233-242
233
Muscle adenosine 5'-triphosphate and creatine phosphate
concentrations in relation to nutritional status and sepsis in
man
J. C . TRESADERN, C. J. THRELFALL, K. WILFORD AND M. H. IRVING
Department of Surgery, University of Manchester Medical School, and MRC Trauma Unit, Hope Hospital, Salford, U.K.
(Received 27 May 1987/15 January 1988; accepted 11 February 1988)
SUMMARY
INTRODUCTION
1. Intramuscular concentrations of adenosine 5'-triphosphate (ATP) and creatine phosphate were measured
in the vastus lateralis muscle of 28 non-septic malnourished patients and 3 l septic malnourished patients.
Similar measurements were made on the rectus abdomin i s muscle of about one-third of these patients. All results
were compared with those obtained from 15 normally
nourished non-septic control subjects.
2. Objective measurements of nutritional status (both
anthropometric and biochemical) and sepsis were
recorded in all subjects.
3. The vastus lateralis muscle of the non-septic and
septic malnourished patients had intramuscular concentrations of ATP and total adenine nucleotides (TAN)
that were up to 30% lower than control values, depending
on the reference base used.
4. In the rectus abdominis muscle, ATP and TAN concentrations were up to 60% lower than control values,
and creatine phosphate up to 47% lower, again depending
on the reference base used.
5. In both muscles, the changes were more marked in
those patients who were septic as well as malnourished.
Immediately available energy in skeletal muscle is stored
in the terminal high-energy phosphate bonds of adenosine
5'-triphosphate (ATP)and creatine phosphate. Significant
changes have been reported in the intramuscular concentrations of ATP and creatine phosphate in a variety of
conditions,including trauma and sepsis.
In animals subjected to experimental trauma there can
be a marked depletion of intramuscular creatine phosphate, with less severe depletion of ATP [l-61. These
changes are associated with a rise in the blood lactate/
pyruvate ratio, suggesting that the fall in the concentrations of ATP and creatine phosphate is secondary to
agonal tissue anoxia [7,8].Experimental sepsis in rats produced no change in intramuscular ATP or creatine phosphate concentrations [9].
In man, the anoxic pattern of change, with greater relative loss of creatine phosphate than of ATP, has been
reported in normal exercising muscle [lo-141, and cardiac or respiratory failure [15, 161. However, in sepsis and
accidental trauma [16, 171 there is a different pattern of
change, characterized by a relative greater loss of ATP
and smaller loss of creatine phosphate than in the anoxic
conditionsdescribed above.
Many patients with severe trauma or sepsis become
malnourished, losing weight and muscle bulk during their
illness. ATP and creatine phosphate concentrations have
also been reported to be reduced in malnourished patients
with short-bowel syndrome [ 181 and gastric carcinoma
[ 191, with a pattern of change similar to that seen in accidental trauma and sepsis.
The previous studies on changes in intramuscular ATP
and creatine phosphate concentrations have included
only small numbers of septic patients, with no attempt to
quantify the sepsis or associated malnutrition. We present
here the findings in a large group of septic patients, most
of whom were also malnourished. We have compared the
findings with those in a group of normally nourished non-
Key words: adenine nucleotides, adenosine 5'-triphosphate, creatine phosphate, nutritional status, sepsis,
skeletal muscle.
Abbreviations: ADP, adenosine 5'-pyrophosphate; AMP,
adenosine 5'-phosphate; ATP, adenosine 5'-triphosphate;
CP/TC ratio, ratio of creatine phosphate to total creatine;
DNA, deoxyribonucleic acid; FFDM, fat-free dry weight
of muscle; TAN, total adenine nucleotides.
Correspondence: Mr J. C. Tresadern, North Western Injury
Research Centre, Clinical Sciences Building, Hope Hospital,
Eccles Old Road, Salford M6 8HD, U.K.
234
J. C. Tresadern et al.
septic patients, and a group of malnourished non-septic
patients. The aim was to determine the extent to which
the malnutrition associated with sepsis could explain the
low intramuscular concentrations of ATP and creatine
phosphate. We have also measured the degree of sepsis or
malnutrition to determine any correlations with the
observed changes in ATP and creatine phosphate concentrations.
Previous studies have been mainly on the vastus lateralis muscle. We have also studied the rectus abdominis
muscle, to determine if the two muscles respond differently to malnutrition and sepsis. Two reference bases
were used, namely fat-free dry weight of muscle (FFDM)
and deoxyribonucleic acid (DNA)content. The latter has
been advocated as a reference base for muscle metabolites on the grounds that DNA content is relatively constant, even when there is muscle wasting [20,2 1J.
METHODS
This work had the approval of the Salford Area Health
Authority Ethical Committee. Full informed written consent was obtained for every muscle biopsy.
Patients
Three groups of patients were studied.
‘Control’ group. This contained 15 patients undergoing
elective cholecystectomy. None had experienced recent
sepsis or weight loss. Samples of vastus lateralis and
rectus abdominis were obtained from all 15.
‘Malnourished’ group. This consisted of 28 patients,
without sepsis, with recent involuntary loss of at least 10%
of their usual body weight. The conditions responsible for
the weight loss included oesophageal and gastric carcinomas, post-gastrectomy weight loss, and short-bowel syndrome after extensive small-bowel resection.
All 28 patients had a muscle biopsy taken from the
vastus lateralis and 11 of them had a biopsy taken from
the rectus abdominis muscle.
‘Septic’ group. This comprised 31 patients. The source
of the sepsis ranged from simple wound infection to
extensive intraperitoneal or intrathoracic abscess. All 31
patients underwent biopsy of the vastus lateralis and 11
also had a biopsy taken from the rectus abdominis.
Twenty-six of the patients were also ‘malnourished’, having lost more than 10%of their normal body weight during the period of sepsis before muscle biopsy.
Some patients had more than ope biopsy, when indicated by a change in their clinical condition, such as a
change from the non-septic to the septic state. All
measurements of sepsis and nutritional status were
repeated at the time of the second biopsy, and the results
analysed as if obtained from a new patient. In all cases
where such repeat biopsies were taken, the time interval
was never less than 1 week (median 6 weeks, range 1-40
weeks).
It was not possible, or necessary, to match the three
groups for sex and age. Other authors [22] have shown
that there is no difference in high-energy phosphate con-
centrations between males and females, and age has only
a minimal effect. By chance, the three patient groups were
well-matched for age, the median ages being 5 1 (range
23-69) years for the ‘control’ group, 54 (range 25-80)
years for the ‘malnourished group and 52 (range 23-73)
years for the ‘septic’ group. There was no statistically significant difference between these ages.
Assessment of malnutrition
The criterion for inclusion of a non-septic patient in the
‘malnourished’ group was the involuntary loss of more
than 10% of his usual body weight. All patients in the
study were given a ‘malnutrition score’ derived from
simple anthropometric and biochemical indices of nutrition (see the Appendix). The weight and anthropometry
‘standards’ were those given by Blackburn et al. [23]. As
these standards are different for men and women, their
use eliminated the need to ensure that the groups contained similar proportions of each sex. The ‘standard’ for
plasma albumin was the mid-point of the normal range for
plasma albumin in our clinical laboratories.
The ‘standards’ used by Blackburn et al. [23] are
derived from those of Jelliffe [24], and may not be appropriate to current normal Western populations [25, 261.
This was not a drawback in our study, as the ‘standard’
data were used only to compare groups, rather than to
make a definitive evaluation of nutritional status.
Assessment of sepsis
Each patient was scored for severity of sepsis using the
system described by Elebute & Stoner [27],which utilizes
simple clinical and laboratory indices of sepsis. This
system has been used successfully to demonstrate that in
septic patients oxidation rates for glucose and fat [28] and
plasma cortisol and catecholamine concentrations [29]
are related to severity of sepsis. A factor not included in
this system, which could be important in the present
context, is the duration of sepsis before biopsy. This was
therefore recorded separately.
Muscle sampling
Biopsies were usually taken after an overnight fast.
Some septic patients receiving parenteral nutrition had
this replaced by 0.9% (w/v) NaCl at least 8 h before
biopsy, except for nine patients requiring emergency
surgery who received full nutrition until shortly before
biopsy.
Vastus lateralis. The biopsy technique for this muscle
was essentially that described by Bergstrom [30], using a
punch biopsy needle at a standard site and limb [311, but
freezing the sample in liquid nitrogen immediately it had
been taken. In a few of the most wasted subjects the thin
muscle layer would not prolapse into the cutting window
of the needle and so open biopsies were obtained through
a small skin incision. Local anaesthesia was used for this
unless the patient was undergoing therapeutic surgery
under general anaesthesia.
Muscle adenosine 5’-triphosphateand creatine phosphate
Rectus abdominis. Samples of this muscle (approximately 500 mg wet weight) were obtained from anaesthetized patients undergoing laparotomy. The rectus sample
was usually taken from the upper half of the muscle and
immediately dropped into liquid nitrogen.
Effect of fat and connective tissue
The concentrations of ATP, creatine phosphate,
adenine nucleotides and glycogen in the anterior rectus
sheath (effectivelypure connective tissue) were known to
be very low [32],so contamination of the samples with fat
and connective tissue could affect the apparent concentrations of intramuscular metabolites. The fat was therefore extracted from each sample, after initial freezedrying, by immersion in light petroleum ether [20]. The
muscle results could then be corrected for connective
tissue contamination by estimating the hydroxyproline
content [20,21]. This correction rarely exceeded 10%. No
correction was required for the DNA concentration, as
this is similar in muscle and connective tissue [20].
Acid extraction of metabolites
After initial freeze-drying and fat extraction, muscle
samples of approximately 10 mg fat-free dry weight were
extracted with perchloric acid (500 mmol/l) [31]. The
supernatant from the neutralized extract was used to
estimate adenine nucleotides, creatine, creatine phosphate, pyruvate and lactate.
The residue from the initial perchloric acid extraction
was immediately washed with ice-cold water and stored at
- 50°C for subsequent analysis for DNA and hydroxyproline. We found that if the washing was not done immediately there was an apparent loss of DNA from the
residues.
Analytical techniques
Pyruvate, adenine nucleotides, creatine and creatine
phosphate were measured on the day of extraction, and
lactate, DNA and hydroxyproline when convenient as it is
known they are stable in storage [33, 201. Pyruvate and
lactate were measured by the methods of Passonneau &
Lowry [34] and Gutmann & Wahlefeld [35], respectively.
Adenosine 5’-pyrophosphate (ADP) and adenosine 5’phosphate (AMP) were determined sequentially by the
method of Jaworek et al. [36].It was necessary to include
exogenous ATP (0.042 mmol/l) to facilitate the conversion of AMP to ADP with myokinase. ATP and creatine
phosphate were estimated sequentially by the method of
Harris et al. [31]. However, we used a glucose concentration of 0.25 mmol/l instead of the 2.25 mmol/l recommended because our glucose-6-phophate dehydrogenase
had an allowable hexokinase activity of < 0.02’/0, which
was sufficient to produce a continuing side-reaction
between endogenous ATP and added glucose at the
higher glucose concentration. Creatine was measured
separately [311. DNA and hydroxyproline were estimated
235
as described by Milewski et al. [21]. Glycogen was
measured on separate 1-4 mg samples of FFDM [37].
Statistical analysis
The malnourished and septic groups were very heterogeneous, and the number of patients in each group was
too small for the data to be normally distributed. Therefore, comparisons between groups were carried out by
non-parametric methods [38,39].
Wilcoxon’s rank sum test was used for unpaired data,
and the signed rank test for paired data. The latter test
was used to compare the results in vastus lateralis and
rectus abdominis samples taken at the same time from the
same patient. Where the significance level of a difference
was less than 0.01, or the number’of patients in the groups
was larger than in tables of critical points of rank sums,
the normal deviate was used to assess sigmficance.
In the subsequent tables the data for each group are
presented as the median and the approximate 95% confidence range of the median [38].
When examining the relationship between variables,
Kendall’s rank correlation test and rank correlation coefficient (z) were used, with significance tested by the
normal deviate.
RESULTS
Nutritional status
The nutritional indices in the three patient groups are
shown in Table 1. The variation in the number of patients
for the weight standards in each group is because some
patients were too ill to be weighed accurately, or could
not recall their previous usual weight. The indices of
nutrition in the malnourished and septic groups were all
significantly worse than in the control group. The overall
malnutrition score was similar for these two groups,
although the malnourished group lost proportionately
more fat than the septic whilst the septic group had significantly lower plasma albumin than the malnourished.
However, these two groups of patients were well matched
for the severity of their muscle wasting, as reflected by
their similar arm muscle circumference.
Biochemical variables
ATP and creatine phosphate. Vastus lateralis. The
results for this muscle are shown in Table 2. The ‘energy
charge’ [40] was calculated as [ATP+O.5 ADPI/
[AMP+ ADP+ ATP]. The ratio of creatine phosphate to
total creatine (CP/TC ratio) was also calculated. Using
FFDM as the reference base, the only significant changes
were in the septic group, where ATP and total adenine
nucleotide (TAN) concentrations were about 5- 10%
lower than control values. Using muscle DNA as the reference base, a similar pattern was seen, although the
changes were greater, namely ATP and TAN concentrations in the septic group were about 30% lower than in
controls. The change of reference base also depressed
J. C. Tresadern et al.
236
Table 1. Comparison of nutrition indices
Results are expressed as medians and 95% confidence ranges of the medians. Indices other than
malnutrition score are given as percentages of 'standard' values (see the text). Statistical sigmficance: *P<O.OOl compared with the control group; t P < 0 . 0 5 , ttP<0.01, tttP<0.001 compared with the septic group.
Standard weight (YO)
Usual weight (YO)
Triceps skinfold
thickness (O/.)
Arm muscle
circumference ( O h )
Plasma albumin ( O h )
Malnutrition score
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
Control
Malnourished
Septic
15
115.30
103.69-1 33.71
15
100
28
79.08*
70.26-87.02
21
79.52*?
67.97-83.09
28
45.56*tt
33.60-60.00
28
84.01*
80.95-89.21
28
87.06*ttt
80.00-89.41
28
9*
7-1 1
30
79.08*
72.86-85.36
23
85.00*
77.65-96.38
31
68.00*
56.00-80.80
31
84.70*
78.54-90.28
31
61.17*
54.1 1-68.23
30
1o*
8-1 1
-
15
102.43
84.85-1 33.33
15
101.42
96.85-1 12.21
15
101.17
94.13-103.53
15
0
0- 1
Table 2. Comparisons of high-energy phosphate systems in the vastus lateralis
Results are expressed as medians and 95% confidence ranges of the medians. Statistical significance: *P<0.05, **P<0.01 compared with the control group; tP< 0.05, ttP< 0.01 compared
with the septic group.
Reference base: FFDM
ATP
(pmol/g of FFDM)
TAN
(pmol/g of FFDM)
Energy charge
Creatine phosphate
(pmol/g of FFDM)
Total creatine
(pmol/g of FFDM)
CP/TC ratio
Reference base: muscle DNA
ATP
(,umol/mg of DNA)
TAN
( pmol/mg of DNA)
Creatine phosphate
(pmol/mg of DNA)
Total creatine
(pmol/mg of DNA)
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Range
Control
Malnourished
Septic
12
22.98
21.83-25.75
11
25.88
24.79-30.95
27
23.15tt
21.15-25.59
25
27.19tt
24.54-29.72
26
0.930
0.904-0.934
27
70.40
60.94-83.31
26
128.19t
116.46-140.43
26
0.56
0.51-0.59
29
20.5 1**
19.05-21.72
28
24.58*
21.95-25.83
28
0.909
0.890-0.924
29
73.80
64.05-86.91
29
143.94
128.18-154.89
29
0.53
0.50-0.58
16
6.17
4.76-7.35
15
7.05
6.14-8.40
16
20.68
14.00-24.41
15
34.70
27.19-38.91
13
5.72*
3.04-6.99
13
6.98*
3.79-8.15
13
22.63
12.49-25.18
13
38.59
26.42-47.09
11
0.926
0.891-0.940
12
74.42
66.11-81.71
12
133.85
130.03-142.56
12
0.56
0.50-0.59
5
7.78
6.16-8.83
5
9.49
7.94-9.91
5
23.42
10.99-3 1.7 1
5
42.58
26.30-5 1.92
Muscle adenosine 5’-triphosphate and creatine phosphate
ATP and TAN concentrations in the malnourished group
by 20-25% of the control values, but these latter changes
were not statistically significant. There was no difference
in ‘energy charge’ or CP/TC ratio between the three
groups.
Rectus abdominis. The results for this muscle are
shown in Table 3. The pattern of change in the adenine
nucleotides was similar to that in the vastus lateralis
muscle, but the changes were greater. ATP and TAN concentrations in the septic group were about 30-35% lower
than controls using FFDM as the reference base and
about 55-60% lower using the DNA reference, base.
Additionally, in the rectus abdominis muscle ATP and
TAN concentrations in the malnourished group were
approximately 28% lower than controls using the FFDM
reference base and 55% lower using the DNA reference
base. These changes were statistically significant, as were
the lower creatine phosphate concentrations. As with
vastus lateralis, there was no difference in ‘energy charge’
between the three groups, but the CP/TC ratio was lower
in the two malnourished groups than in the controls.
There was no difference in the ATP and creatine phosphate concentrations of the vastus lateralis and rectus
abdominis muscles in the control group. Therefore the
237
lower values in the rectus abdominis in the other patient
groups represented a greater response to the factors
initiating the loss of ATP and creatine phosphate. However, their concentrations showed no correlation with our
indices of nutrition or sepsis.
Lactate and pyruvate. Although lactate and pyruvate
concentrations were both raised in the vastus lateralis
muscle of the malnourished and septic groups, the lactate/
pyruvate ratio was not significantly different from
controls (Table 4). In the rectus abdominis muscle, lactate,
pyruvate and lactate/pyruvate ratios were similar in all
three groups.
Glycogen. Glycogen was measured to assess the availability of an energy substrate (Table 5). The only statistically significant finding was that its concentration in the
rectus abdominis muscle of the septic group of patients
was lower than in controls. However, the glycogen concentrations in these septic patients showed no correlation
with ATP or creatine phosphate concentrations, nor with
the nutritional or sepsis indices.
Serial changes in individual patients. The results
reported above are for a cross-sectional study of different
patient groups. We were also able to study serial changes
in a small number of patients. Four of the malnourished
Table 3. Comparisons of high-energy phosphate systems in the rectus abdominis
Results are expressed as medians and as 95% confidence ranges of the medians. Statistical significance: *P<0.05, **P< 0.01 ,***P< 0.001 compared with the control group.
Reference base: FFDM
ATP
(pmol/g of FFDM)
TAN
( pmol/g of FFDM)
Energy charge
Creatine phosphate
(,umol/g of FFDM)
Total creatine
(pmol/g of FFDM),
CP/TC ratio
n
Median
Range
n
Median
Range
n
Median
Range
n
Median
Ranse
n
Median
n
Median
Range
Reference base: muscle DNA
n
ATP
Median
(pmollmg of DNA)
Range
n
TAN
(pmollmg of DNA)
Median
Range
n
Creatine phosphate
Median
(pmol/mg of DNA)
Range
n
Total creatine
(prnol/mgof DNA)
Median
Range
Control
Malnourished
Septic
14
22.62
18.45-23.78
14
25.91
22.28-27.88
14
0.9 15
0.895-0.940
14
80.45
72.79-95.83
14
132.67
115.52-15 1.61
14
0.63
0.58-0.65
11
16.30**
13.91-22.19
11
18.71**
16.66-25.66
11
0.918
0.891-0.928
11
66.27*
56.95-83.58
11
124.99
96.57-1 5 1.01
11
0.53*
0.50-0.60
11
14.50***
10.56-17.24
11
17.56***
12.13-2 1.43
11
0.910
0.874-0.935
11
56.87*
40.06-86.08
11
111.69
90.19-140.49
11
0.54*
0.40-0.61
10
8.54
6.32-10.94
10
10.26
7.10-13.19
10
30.36
26.86-42.95
10
53.67
42.18-64.91
11
3.80**
2.50-7.21
11
4.65**
2.88-8.43
11
17.56**
8.94-25.42
11
32.96**
16.20-50.03
10
3.50**
2.70-5.30
10
4.29***
3.33-6.47
10
16.07***
10.59-20.27
10
29.33***
10.35-38.2 1
J. C. Tresadern et al.
238
Table 4. Comparisons of lactate and pyruvate
Results are expressed as medians and as 95% confidence ranges of the medians. Statistical significance: *P< 0.05, **P< 0.01, ***P< 0.001 compared with the control group; tP<O.O1 compared
with the septic group.
Lactate
(,umol/g of FFDM)
Pyruvate
(pmol/g of FFDM)
Lactate/pyruvate ratio
Rectus abdominis
Lactate
(pmol/g of FFDM)
Pyruvate
(pmol/g of FFDM)
Lactate/pyruvate ratio
Control
Malnourished
Septic
n
Median
Range
n
Median
Range
n
Median
Range
9
2.51
2.09-4.82
9
0.151
0.131-0.236
9
17.25
13.23-22.27
25
4.81* t
4.0 1-5.83
25
0.3 12*
0.242-0.434
25
17.21
13.70-18.72
23
7.28***
6.03- 10.43
23
0.331**
0.289-0.409
23
20.83
16.98-30.57
n
Median
Range
n
Median
Range
n
Median
Range
13
4.92
1.66-7.32
13
0.229
0.152-0.342
13
17.15
11.60-23.49
11
4.34
2.34-6.99
11
0.240
0.186-0.554
11
14.07
9.75-23.33
11
5.4
1.51-9.08
11
0.305
0.174-0.435
11
15.70
7.93-38.78
Table 5. Muscle glycogen
Results are expressed as medians and as 95% confidence ranges of medians. Statistical significance: *P< 0.05 compared with the control group.
Glycogen (mg/g of FFDM)
Control
Malnourished
Septic
Vastus lateralis
n
Median
Range
5
50.56
45.81-86.4
19
56.38
34.25-66.29
15
53.62
47.23-56.56
Rectus abdominis
n
Median
Range
14
50.94
4 I .07-59.68
11
41.71
30.86-58.33
3 1.89*
16.56-67.90
non-septic patients had a second biopsy carried out when
their nutritional status changed. The changes in their ATP
and creatine phosphate concentrations are shown qualitatively in Table 6. It can be seen that with the exception of
creatine phosphate in patient 4, ATP and creatine phosphate concentrations moved in the same direction as
nutritional status.
Eight of the patients in the septic group also had a
second biopsy taken. in contrast to the non-septic
patients, there was no consistent pattern in the direction
of change of ATP and creatine phosphate concentrations.
However, in most of these patients there were simultaneous changes in both nutritional status and degree of
sepsis, in addition to surgical intervention.
DISCUSSION
Our results in the malnourished patients can be compared
with two previous reports of changes in ATP and creatine
11
Table 6. Serial changes in individual patients
Subjects were all malnourished and non-septic. t,
Improved; 1, deteriorated.
Patient no.
Nutritional
status
,4TP
Creatine
phosphate
1
2
3
4
1
1
1
t
1
1
1
t
1
1
1
1
phosphate concentrations in the vastus lateralis. Furst et
al. [ 181 studied three patients with malnutrition associated
with the short-bowel syndrome and noted lower concentrations of both ATP and TAN compared with controls.
Symreng et af. [ 191 studied a larger group of 26 patients
with gastric carcinoma and demonstrated that the concen-
Muscle adenosine 5’-triphosphate and creatine phosphate
trations of ATP, TAN, creatine phosphate and total
creatine were lower in these patients than in controls.
We found no significant changes in the vastus lateralis
muscle in our non-septic malnourished group. However,
the rectus abdominis muscle did show changes similar to
those reported in the vastus lateralis by the other authors.
In general, therefore, the pattern of change was similar in
the three studies. The difference between the findings for
vastus lateralis in our study and the two quoted above
may reflect differences in the severity of malnutrition
exhibited by the patients. It is evident that the group
studied by Symreng et af. [ 191 had a higher proportion of
more severely malnourished patients.
Our results for septic patients can also be compared
with those in two previous studies. Bergstrom et af. [16]
studied a group of intensive care patients, of whom four
were septic. These septic patients had lower concentrations of ATP, TAN, and creatke phosphate than did
control subjects, with the adenine nucleotides being more
severely affected. Liaw et af. [ 171 also studies four septic
patients and noted lowered concentrations of ATP, TAN
and creatine phosphate. Both studies reported a low
‘energy charge’ in their patients. The results in our 31
patients are in general agreement with those quoted
above, except that we found no change in ‘energy-charge’.
However, the ratio of creatine phosphate to total creatine
was lower than controls for the rectus abdominis (but not
the vastus lateralis)of both our malnourished groups.
In conditions associated with muscle hypoxia [ 10- 161,
there is a marked fall in intramuscular creatine phosphate
concentrations, with relative preservation of ATP. The
creatine phosphate concentrations are directly related to
the adequacy of muscle oxygenation [ 11,131.
In contrast, our study and the others cited above all
show that in patients with malnutrition, with or without
sepsis, the concentrations of ATP, TAN and creatine
phosphate are lower than in controls, but the fall is generally more marked for ATP and TAN than for creatine
phosphate. We did note that there was a parallel rise in
lactate and pyruvate concentrations in the vastus lateralis
of both our malnourished groups (but not in the rectus
abdominis). Whilst this could be compatible with hypoxia,
the observations that the lactate/pyruvate ratio did not
change would make this less likely. The finding that the
CP/TC ratio was unchanged in the vpstus lateralis would
support this view. These features, together with the comments above on the different pattern of change seen
in conditions of known hypoxia, suggest that the changes
in adenine nucleotides and creatine phosphate seen in the
malnourished and septic patients were probably not a
consequence of muscle hypoxia.
Symreng et af. [19] noted that ATP and TAN in their
malnourished patients showed significant correlations
with serum albumin and the degree of weight loss. However, we have been unable to find any significant correlation between the concentrations of ATP and creatine
phosphate and our index of malnutrition, the sepsis score
or the duration of sepsis. Nevertheless, our results suggest
that sepsis exerts an independent effect on intramuscular
concentrations of ATP and creatine phosphate, in addi-
239
tion to the effects of malnutrition, as the changes were
more marked in the septic than in the non-septic patients,
even though both groups were similar in their degree of
malnutrition.
We have also shown, in four patients, that ATP and
creatine phosphate concentrations changed in the same
direction as changes in nutritional status (as defined by
the variables we measured) in serial biopsies on individual
patients. We were unable to show such consistent effects
of changing sepsis status. A possible explanation is that
the septic patients who had serial biopsies also demonstrated changes in their nutritional status, as well as
undergoing surgical intervention.
Symreng et af. [ 191 reported reduced concentrations of
intramuscular glycogen in their malnourished subjects,
who had a reduced calorie intake as a result of the
anorexia associated with their carcinomas. This might
suggest that a reduced supply of energy substrate could be
a limiting factor in the formation of ATP and creatine
phosphate in these patients. However, with the exception
of the rectus abdominis muscle of the septic patients we
found no reduction in intramuscular glycogen concentrations in our patients.
A feature of particular interest in the septic and malnourished patients was that the loss of ATP was not solely
a consequence of a shift in the equilibrium between ATP
and ADP as there was an absolute loss of TAN. Sahlin et
af. [41] showed that the loss of TAN seen in exercising
subjects was the result of deamination of AMP to inosine
monophosphate and suggested this was initiated by the
fall in ‘energy-charge’ that occurs in exercising muscle.
Presumably the authors were using the ‘energy charge’
calculation as in index of the movement away from equilibrium of the adenine nucleotide/creatine phosphate
system in the hypoxic muscle. However, this mechanism
may not be applicable in malnutrition and sepsis, where
there is probably no significant hypoxia (as detailed
above). Also, we have found no change in ‘energy charge’
in our groups of 28 non-septic and 31 septic malnourished
subjects. This contrasts with the studies of Bergstrom
et af. [16] and Liaw et af. [17], who each reported
decreased ‘energy charge’ in groups of four septic patients.
We did, however, note a reduced CP/TC ratio (suggesting
a move away from equilibrium) in some of our malnourished and septic patients, but only in the rectus
abdominis muscle, and not in the vastus lateralis (which
was the muscle studied by Bergstrom et af. [16] and by
Liaw et af. [ 171).
Bergstrom et af. [16] suggested that the loss of TAN
could be a result of decreased synthesis of purine precursors, a process that requires large amounts of energy. If
the mechanism suggested by Bergstrom et af. [16] is
correct, a vicious circle will ensue, as the synthesis of purine precursors would require amounts of energy which
these patients may not have. The failure to synthesize adequate purine precursors would then lead to a further loss
of TAN and hence ATP, and so on.
An alternative explanation for the decrease in A V ,
creatine phosphate and TAN concentrations in malnourished and septic patients may lie in a change in the
240
J. C. Tresadern et al.
fibre composition of muscle as it wastes. It has been
reported that wasting of muscle is associated with selective atrophy and loss of type I1 fibres (‘white’ or ‘fasttwitch fibres) in a number of conditions such as neuromuscular disease, disuse atrophy [42] and old age
[43, 441. Edstrom et al. [45] have reported that A P ,
creatine phosphate and TAN concentrations are higher in
those muscles which have a higher proportion of type I1
fibres. Rehunen & Harkonen [46] reported no difference
in the ATP and creatine phosphate concentrations of fastand slow-twitch fibres in males, but in females the
creatine phosphate concentration in fast-twitch (type 11)
fibres was lower than in slow-twitch fibres. There is
clearly disagreement in the literature on the ATP and
creatine phosphate contents of the two fibre types. However, if the results of Edstrom et al. [45] are correct, one
could propound a hypothesis that in malnourished and
septic patients disuse atrophy of their skeletal muscle
results in a selective atrophy and death of fast-twitch
fibres. This in turn results in a lowering of the concentrations of ATP, TAN and creatine phosphate, perhaps in
relation to a disproportionate loss of mitochondria1
activity.
Such a disproportionate loss of fast-twitch fibres has
been reported in the vastus lateralis muscle in the elderly
[43,44]and there is a small loss of ATP and creatine phosphate in this same age group [22]. In the elderly, a course
of physical training has been shown to increase the proportion of fast-twitch fibres in muscle [47] and also raise
the concentrations of ATP and creatine phosphate [48].
With regard to our own patients, we did not unfortunately
keep individual records of previous levels of activity.
However, most of the septic patients were confined to bed
for long periods before biopsy and many of the malnourished non-septic patients had reduced levels of activity because of general weakness and lethargy. Also, both
groups of malnourished patients had significantly
decreased arm muscle circumferences, suggesting there
had been muscle atrophy.
It was noted earlier that we found a greater loss of ATP
and creatine phosphate in the rectus abdominis muscle of
the malnourished and septic patients that in the vastus
lateralis muscle, although there was no difference
between these two muscles in the control group. This
finding could be explained by the different ATP and
creatine phosphate contents of fast- and slow-twitch
fibres if there is a greater proportional loss of the former
in the wasted rectus abdominis than in the vastus lateralis.
Certainly there is such a differential response to wasting
in various muscle groups in the rat [49], but such data are
lacking for man at present.
The physiological significance of the low intramuscular
concentrations of ATP and creatine phosphate in malnourished or septic patients is uncertain. Very low concentrations are associated with the inability of muscle to
continue contracting [lo, 14, 50-521, but ATP and creatine phosphate concentrations in septic and malnourished
patients rarely fall so low. The maximal relaxation rate
and endurance of muscle, which are dependent on adequate intramuscular concentrations of ATP and creatine
phosphate are impaired in malnourished subjects [53,54],
and can be restored by nutritional supplementation [45,
53, 551. It has also been suggested that the general fatigue
and malaise associated with major illness may be related
to low concentrations of ATF’ and creatine phosphate [ 17,
191.
ACKNOWLEDGMENTS
We thank Professor H. B. Stoner and Dr R. A. Little for
general guidance, Dr D. Heath for statistical advice, and
Mrs P. Maycock for additional technical help. This work
was carried out as part of the requirements (by J.C.T.) for
the degree of Ch.M. of the University of Manchester.
REFERENCES
1 . McShan, W.H., Potter, V.R., Goldman, A., Shipley, E.G. &
Meyer, R.H. ( 1945) Biological energy transformation during
shock as shown by blood chemistry. American Journal of
Physiology, 145,93-106.
2. Le Page, G.A. ( 1946) Biological energy transformation during shock as shown by tissue analysis. American Journal of
Physiology, 146,267-281.
3. Goranson, E.S., Hamilton, J.E. & Haist, R.E. (1948)
Changes in phosphate and carbohydrate metabolism in
shock. Journal of Biological Chemistry, 174, 1-9.
4. Stoner, H.B. & Threlfall, C.J. (1954) The effect of nucleotide and ischaemic shock in the level of energy-rich phosphates in the tissue. Biochemical Journal, 58, 1 15- 122.
5. Threlfall, C.J. & Stoner, H.B. (1957)Studies on the mechanism of shock. The effect of limb ischaemia on the phosphates of muscle. British Journal of Experimental Pathology,
38,339-356.
6. Chaudry, LH., Sayeed, M.M. & Baue, A.E. (1976) Alterations in high-energy phosphates in haemorrhagic shock as
related to tissue and organ. Surgery, 79,666-668.
7. Stoner, H.B., Threlfall, C.J. & Green, H.N. (1952) Studies
on the mechanism of shock. Carbohydrate metabolism in
nucleotide and ischaemic shock. British Journal of Experimental Pathology, 33,131-156.
8. Stoner, H.B. & Threlfall, C.J. (1959) In: The Biochemical
Response to Injury, pp. 105-208. Ed. Stoner, H.B. &
Threlfall, C.J. Blackwell Scientific Publications, Oxford.
9. Chaudry, I.H., Wichterman, K.A. & Baue, A.E. (1979)
Effect of sepsis on tissue adenine nucleotide levels. Sutgery,
85,205-21 1.
10. Hultman, E., Bergstrom, J. & McLennan-Anderson, N.
( 1967) Breakdown and re-synthesis of phosphorylcreatine
and adenosine triphosphate. Scandinavian Journal of
Clinical and Laboratory Investigation, 19,56-66.
11. Karlsson, J. ( 1 971) Lactate and phosphagen concentrations
in working muscle of man. Acta Physiologica Scandinavica,
Suppl. 358.
12. Hams, R.C., Hultman, E., Kaijser, L. & Nordesjo, L.-0.
(1975) The effect of circulatory occlusion on isometric
exercise capacity and energy metabolism of the quadriceps
muscle in man. Scandinavian Journal of Clinical and Laboratory Investigation, 35,87-95.
13. Bylund-Fellenius, A.-C., Walker, P.M., Elander, A., Holm,
S., Holm, J. & Schersten, T. (1981) Energy metabolism in
relation to oxygen partial pressure in human skeletal muscle
during exercise. Biochemical Journal, 200,247-255.
14. Boobis, L., Williams, C. & Wooton, S.A. (1983) Human
muscle metabolism during brief maximal exercise. Journal
of Physiology (London), 338,21 P-22P.
Muscle adenosine 5'-triphosphateand creatine phosphate
15. Karlsson, J., Willerson, J.T., Leshin, S.J., Mullins, C.B. &
Mitchell, J.H. (1975) Skeletal muscle metabolites in patients
with cardiogenic shock or severe congestive heart failure.
Scandinavian Journal of Clinical and Laboratory Investigation, 35,73-79.
16. Bergstrom, J., Bostrom, H., Furst, P., Hultman, E. &
Vinnars, E. ( 1 976) Preliminary studies of energy-rich phosphagens in muscle from severely ill patients. Critical Care
Medicine, 4, 197-204.
17. Liaw, K.Y., Askanazi, J., Michelsen, C.B., Kantrowitz, L.R.,
Furst, P. & Kinney, J.M. ( 1 980) Effect of injury and sepsis
on high-energy phosphates in muscle and red cells. Journal
of Trauma, 20,755-759.
18. Furst, P., Bergstrom, J., Hultman, E. & Vinnars, E. (1976)
In: Metabolism and the Response to Injury, pp. 94-122. Ed.
Wilkinson, A.W. & Cuthbertson, D. Pitman Medical, Tunbridge Wells, Kent.
19.. Symreng, T., Larsson, J., Schildt, B. & Wetterfors, J. (1983)
Nutritional assessment reflects muscle energy metabolism in
gastric carcinoma. Annals of Surgery, 198,146-50.
20.. Milewski, P.J. (1980)The significance of muscle free amino
acid measurements in critically ill surgical patients. M.Ch
Thesis, Cambridge University.
21. Milewski, P.J., Threlfall, C.J., Heath, D.F., Holbrook, I.B.,
Wilford, K. & Irving, M.H. (1982) Intracellular free amino
acids in undernourished patients with or without sepsis.
CIinicalScience,62,83-91.
22. Moller, P., Bergstrom, J, Furst, P. & Hellstrom, K. (1980)
Effect of aging on energy-rich phosphagens in human skeletal muscle. ClinicalScience, 58,553-555.
23. Blackburn, G.L., Bistrian, B.R., Maid, B., Schlamm, H.T. &
Smith, M.F. (1977) Nutritional and metabolic assessment of
the hospitalized patient. Journal of Parenteral and Enteral
Nutrition, 1, 11-22.
24. Jelliffe, D.B. ( 1 966) The Assessment of the Nutritional Status
of the Communi@. World Health Organisation Monograph
no.53, WHO, Geneva.
25. Gray, G.E. & Gray, L.K. (1979) Validity
norms used in the assessment of ho
Journal of Parenteral and Enteral Nutrition, 3,366-368.
26. Symreng, T. (1983)Arm anthropometry in a large reference
population and in surgical patients. Clinical Nutrition, 1,
211-219.
27. Elebute, A. & Stoner, H.B. (1983) The grading of sepsis.
British Journal of Surgery, 70,29-3 1.
28. Stoner, H.B., Little, R.A., Frayn, K.N., Elebute, A.E.,
Tresadern, J. & Gross, E. (1983)The effect of sepsis on the
oxidation of carbohydrate and fat. British Journal of Surgery,
70,32-35.
29. White, R.H., Frayn, EN., Little, R.A., Threlfall, C.J., Stoner,
H.B. & Irving, M.H. (1987) Hormonal and metabolic
responses to glucose infusion in sepsis studied using the
hyperglycemic glucose clamp technique. Journal of
Parenteraland Enteral Nutrition, 11,345-353.
30. Bergstrom, J. (1 962) Muscle electrolytes in man. Scandinavian Journal of Clinical and Lpboratory Investigation,
Suppl. 68.
31. Harris, R.C., Hultman, E. & Nordesjo, L.-0. (1974) Glycogen, glycolytic intermediates and high-energy phosphates
determined in biopsy samples of musculus quadriceps
femoris of man at rest. Methods and variance of values.
Scandinavian Journal of Clinical and Laboratory Investigation, 33,109-120.
32. Tresadern, J.C. ( 1 984) Changes in intramuscular concentrations of high-energy phosphates in malnourished and septic
patients. Ch. M. Thesis, University of Manchester.
33. Passonneau, J.V. ( 1 984) In: Methods of Enzymatic Analysis,
2nd edn, vol. 3, pp. 1468-1472. Ed. Bergmeyer, H.U.
Academic Press, New York, San Francisco, London.
34. Passonneau. J.V. & Lowry, O.H. (1974) In: Methods of
Enzymatic Analysis, 2nd edn, vol. 3, pp. 1452-1456. Ed.
Bergmeyer, H.U. Academic Press, New York, San
Francisco, London.
24 1
35. Gutmann, 1. & Wahlefeld, A.W. (1974)In: Methods ofEnzymatic Analysis, 2nd edn, vol. 3, pp. 1464-1468. Ed.
Bergmeyer, H.U. Academic Press, New York, San Francisco, London.
36. Jaworek, D., Gruber, W. & Bergmeyer, H.U. (1974) In:
Methods of Enzymatic Analysis, 2nd edn, vol. 4, pp.
2127-2131. Ed. Bergmeyer, H.U. Academic Press, New
York, San Francisco, London.
37. Good, C.A., Kramer, H. & Somogyi, M. (1933) The determination of glycogen. Journal of Biological Chemistry, 100,
485-491.
38. Snedecor, G.W. & Cochran, W.G. ( 1 980) Statistical Methods,
7th edn, Iowa State University Press, Ames, Iowa.
39. Siegal, S. (1956)Non-Parametric Statistics for the Behavioura1 Sciences. McGraw-Hill Kogakusha, London.
40. Atkinson, D.E. & Walton, G.M. (1967) Adenosine triphosphate conservation in metabolic regulation. Journal of
Biological Chemistry,2 4 2,32 3 9- 324 1.
41. Sahlin, K., Palmskog, G. & Hultman, E. (1978) Adenine
nucleotide and IMP contents of the quadriceps muscle in
man after exercise. Pfligers Archiv. European Journal of
Physiology, 374,193-198.
42. Engel, W.K. (1970) Selective and non-selective susceptibility of muscle fibre type. Archives of Neurology, 22,97-117.
43. Tomonaga, M. (1977) Histochemical and ultrastructural
changes in senile human skeletal muscle. Journal of the
American Geriatrics Sociey, 25, 125-131.
44. Larsson, L., Sjodin, B. & Karlsson, B. (1978)Histochemical
and biochemical changes in human skeletal muscle with age
in sedentary males, age 22-65 years. Acta Physiologica
Scandinavica, 103,31-39.
45. Edstrom, L., Hultman, E., Sahlin, K. & Sjoholm, H. (1982)
The contents of high-energy phosphates in different fibre
types in skeletal muscles from rat, guinea-pig and man.
Journal of Physiology (London), 332,47-58.
46. Rehiinen, S. & Harkonen, M. (1980) High-energy phosphate compounds in human slow-twitch and fast-twitch
muscle fibres. Scandinavian Journal of Clinical and Laboratory Investigation, 40,45-54.
47. Anainsson, A. & Gustafsson, E. (1981) Physical training in
elderly man with special reference to quadriceps muscle
strength and morphology. Clinical Physiology, 1,87-98.
48. Moller, P. & Brandt, R. (1982)The effect if physical training
in elderly subjects with special reference to energy-rich
phosphagens and myoglobin in leg skeletal muscle. Clinical
Physiology, 2,307-314.
49. Bass, A., Gutmann, E. & Hanzlikova, V. ( 1 975) Biochemical
and histochemical changes in energy-supply enzyme pattern
of muscles of the rat during old age. Gerontologia, 21,
31-45.
50. Eggleton, P. & Eggleton, M.G. (1927) The significance of
phosphorus in muscular contraction. Nature (London), 1 19,
194-195.
5 1 . Lundsgaard, E. (1930) Weitere Untersuchungen uber
Muskel-contraktionen ohne Milchsaurebildung. Biochemishe Zeitschrij?, 227,51-83.
52. Hohorst, U.J., Reim, M. & Bartels, H. (1962)Studies on the
creatine kinase equilibrium in muscle and the significance of
ATP and ADP levels. Biochemical and Biophysical Research
Communications, 7,142- 146.
53. Lopes, J., Russell, D.M.cR., Whitwell, J. & Jeejeebhoy, K.N.
(1982) Skeletal muscle function in malnutrition. American
Journal of Clinical Nutrition, 36,602-6 10.
54. Brough, W., Horne, G., Blount, A., Irving, M.H. &
Jeejeebhoy, K.N. (1986) Effects of nutrient intake, surgery,
sepsis and long-term adminstration of steriods on muscle
function. British Medical Journal, 293,983-988.
5 5 . Chan, S.T.F., McLaughlin, S.J., Ponting, G.A., Biglin, J. &
Dudley, H.A.F. (1986) Muscle power after glucosepotassium loading in undernourished patients. British
Medical Journal, 293,1055- 1056.
J. C. Tresadern et al.
242
APPENDIX: DERIVATION OF THE MALNUTRITION SCORE
Each individual nutrition index, as a percentage of ‘standa r d (see the text), is scored on a points system as below,
and the indices are then summed.
70of ‘standard’
Points
90 +
80-89.99
70-79.99
60-69.99
50-59.99
< 50
0
1
2
3
4
5
Example
Patient M10. Female, aged 30 years.
Present weight = 4 1.8 kg
Usual weight = 57.0 kg
‘Standard’ weight for height and sex = 54.9 kg
Nutrition
index*
Actual
value
sw
uw
0.6 cm
19.11 cm
37 8/1
TSF
AMC
Alb
Oh
of standard
76.13
73.33
36.36
82.37
87.05
Point
score
2
2
5
1
1
Malnutrition score = total = 1 1
*Abbreviations: SW, patient’s weight as percentage of ‘standard’ weight-for-height; UW, patient’s weight as percentage of his
usual weight; TSF, triceps skin-fold thickness as percentage of
‘standard‘; AMC, arm muscle circumference as percentage of
‘standard‘; Alb, plasma albumin as percentage of ‘standard’. See
the text for ‘standards’.