- Email: [email protected]
Urinary excretion and efflux from the leg of 3-methylhistidine before and after major surgical operation
Urinary Excretion and Efflux From the Leg of 3-Methylhistidine Before and After Major Surgical Operation M.
J. Rennie,
K. Bennegsrd,
E. Ed&,
P. W.
Emery,
and K. Lundholm
Changes in the &fluxes from the leg of 3-methylhistidine and tyrosine were studied in relation to alterations in the 24-hour excretion of 3-methylhistidine and total nitrogen in 11 patients before and after undergoing major surgical operation. On the first day after operation, efflux of 3-methylhistidine from the leg was significantly decreased by 40% compared to preoperative values. In contrast, tyrosine efflux was doubled at the same time as a transient 20% increase in oxygen uptake of the leg and a marked increase in catecholamine excretion were observed. These changes coincided with a 40% elevation in the excretion of both 3-methylhistidine and nitrogen. Leg metabolism returned to the preoperative pattern within a week. These results suggest that the loss of amino acids from the lean tissues of the leg is the result of a fall in protein synthesis accompanied by an adaptive fall in protein breakdown. Although the increase in nitrogen excretion in response to major surgical trauma reflects the negative amino acid balance of skeletal muscle, the changes in urinary 3-methylhistidine do not correlate with changes in efflux of 3-methylhistidine from the leg. These results suggest that the use of 3-methylhistidine excretion as a specific index of skeletal muscle protein breakdown in postoperative patients may be invalid. Tissues other than skeletal muscle appear to make a substantial contribution to the 3-methylhistidine excretion postoperatively.
INCE the breakdown of tissue protein is difficult to measure for technical reasons, the use of 3methylhistidine excreted in urine as a specific indicator of skeletal muscle protein breakdown’,2 rapidly became widespread; the method was easy and rapid. Initial expectations of the usefulness of the method have not been fulfilled partly because of doubts as to the specificity of 3-methylhistidine as an indicator of breakdown of skeletal muscle protein3 and also because in some cases the results indicated changes in skeletalmuscle protein breakdown at variance to those obtained by other reliable methods.4,5 A major problem in the interpretation of the measurements of the rate of excretion of 3-methylhistidine is the extent to which nonskeletal muscle tissues (which contain contractile protein in smooth muscle) contribute to the whole-body production.’ Evidence is accumulating that in animals gastrointestinal smooth muscle may contribute a large fraction to the wholebody production under normal circumstances and that during starvation this proportion may rise (6, PW Emery, M Holness, and MJ Rennie, unpublished).
S
From the Department of Medicine, University College London School of Medicine, The Rayne Institute, London, England; and the Department of Surgefly I and Anaesthesiology II. Sahlgrenska Hospital. Gothenburg, Sweden. This work was supported by grants from The Cancer Research Campaign, the United Kingdom Medical Research Council, The Welfcome Trust and the Swedish Cancer Society (project No 536), the Assar Gabrielsson’Foundation. the Serena Ehrenstriim Foundation, the Swedish Medical Socieiy, and the Gothenburg Medical Society. Address reprint requests to M. J. Rennie. Department of Physiology, University of Dundee, Dundee DDI IHN, Scotland, England and Kent Lundholm. Department of Surgery I, Sahlgrenska Hospital, S-413 45 Gothenburg. Sweden. 0 1984 by Grune & Stratton, Inc. 0026-0495/84/3303~12%01.00/0 250
Unfortunately, not all methods used in animal studies are applicable in humans and there is still considerable uncertainty about the possible importance of nonskeletal sources of 3-methydhistidine in humans. In order to minimize interference from nonskeletal muscle sources of 3_methylhistidine, we have developed a method to measure the efflux of 3-methylhistidine from tissues of the leg by comparison of arterial and femoral venous concentration of 3-methylhistidine.7 We expect that the general application of the method will provide some insight into possible changes in skeletal-muscle protein breakdown in circumstances of clinical interest. It is well known that major surgery results in increased loss of body nitrogen and an increase in the excretion of 3-methylhistidine.*~’ We have, therefore, applied our techniques to a study of the changes in leg efflux and excretion of 3-methylhistidine in patients undergoing major abdominal surgery. The aims of the study are dual: first, to examine the effects of surgical trauma on skeletal muscle amino acid balance to gain some insight into the mechanisms involved in muscle wasting; and second, to assess the role of the contribution of skeletal muscle to the increased whole-body production rate of 3-methylhistidine which occurs as a result of surgical trauma. MATERIALS
AND
METHODS
Patients Eleven patients admitted grenska Hospital
to the Department
the diagnoses and nutritional Six
patients
with
through abdominal formed
of Surgery at Sahl-
for major operations were studied. Table gastric
states of the patients tumors
underwent
total
gastrectomy
incisions. A classic hemihepatectomy
in the single case with liver tumors.
ulcerative colitis and rectal carcinoma
One patient with urinary bladder carcinoma with the Bricker-deviation
Two
underwent
1 shows
before operation. was per-
patients
with
proctocolectomy.
underwent cystectomy
procedure.
Metabolism, Vol 33, No 3 (March), 1984
84 62 k4
64
54
57 + 4
11. Gastric carcinosarcoma (MI
Mean k SE
10.
63
63
38
64
76
52
72
6. Urinary bladder carcinoma (M)
51 47
Gastric carcinoma (M)
29
5. Ulcerative colitis (F)
9. Gastric lymphosarcoma IF)
56
4. Gastric carcinoma (F)
75
75
41
3. Benign liver tumour (F)
54
63
65
51
1. Gastric Carclnoms (F)
2. Primary liver (F)
(kg)
7. Gastric carcinoma (M)
65
No. Diagnoses
Weight
8. Rectal carcinoma (F)
Age (Yf)
0.88
+ 0.05
0.94
0.83
0.93
0.56
0.89
1.1
0.73
0.76
1.1
0.85
0.95
(%)
Ideal Weight
Apparent:
Postoperatively
yrb
0
315 mo
vr 2.11 vr 5/l
513 mo
0
0
2
none
0
lo-1512
(kg)
LOSS
Weight
11.5 + 1.0
12.0
8.7
19.0
5.4
7.0
10.0
19.0
11.0
11.2
12.0
11.5
(mm)
Skinfold
Triceps
26 + 1
30.0
26.0
26.0
20.5
25.2
30.5
21.8
21.5
30.5
25.0
25.5
Icml
Ctrcumference
Mtid-Arm
36 + 2
23
42
38
44
45
37
23
33
39
41
28
(g/L)
Albumin
Serum
5.3 + 0.7
10.0
37.9 37 ” 0.1
5.5
3.0
2.5
6.5
8.5
3.0
6.5
3.5
5.5
37.3
36.6
36.8
37.0
37.9
37.1
36.8
37.2
36.6
4.0
(hr)
(“a 37.2
Time
Operation
Temperature
Body
Table 1. The Nutritional State of the Patients Before Operation and the Diagnosis, Operation Time, Blood Loss, and Ventilatory Support
1.7
1.6 f 0.4
1.7
0.4
0.7
0.2
3.1
3.0
2.5 0.2
0.4
4.0
12h
Jd
-
14h
Postoperatively
Support
Ventilatory
252
All patients except two had body weights less than the ideal.’ Six patients had lost weight (2 to 15 kg) over the year preceding operation; four of them (numbers I, 4, 5, and 11) had decreased concentrations (< 36 g/L) of serum albumin. Two patients were subfebrile (temperature 37.0 to 37.YC) before operation. All patients were operated on by senior and experienced surgeons. Patient #5 received steroids (500 mg SolucorteR, Upjohn) on each of 3 days preoperatively and on the day of operation. Patient #6 was given a selective B-blocker, metoprolol (Seloke#, H$ssle A.B., Gottenburg, 200 mg/d) postoperatively. The patients were admitted to the hospital at least five days before operation. The patients ate an ordinary hospital diet consisting of 40% of energy as fat, 50% as carbohydrates and 10% as proteins to supply at least 6.3 MJ and 75 g protein. During these days the patients underwent a program of routine preoperative investigations, which made it impossible to make complete collections of urine with confidence until the day preceding operation. From 24 hours before our first measurements of leg metabolism on the day before operation, the patients received only parenteral glucose solutions (10% weight per volume) except that for the last 8 to 13 hours (ie, overnight) before each set of measurements, only saline was infused. This was done to standardize, as much as was possible under the circumstances, the metabolic states of the patients before operation and to make them more directly comparable to the measurements taken in the postoperative state (see below). For urine collections made in the last 24 hours before operation and from then onwards it was possible to be sure of completeness of collection since these were made via urinary catheters. In addition to following the time-course of excretion of nitrogen and urinary metabolites from immediately before and during operation (ie, in urine collected from 8 pm on the day preceding to 8 pm on the operative day), and during five days postoperatively, we have included reference values for excretions collected from cancer and noncancer patients of similar age and body weight index studied in the same hospital under the same conditions. In all patients, leg exchange of amino acids and leg blood flow were measured between 7 AM and 8 AM (by techniques described elsewhere”) on at least four occasions: immediately before surgery, 24 hours after surgery (day 1). and three and six days after surgery. Total parenteral nutrition (TPN) was given to all patients during recovery from surgery. The TPN protocols were not planned by us. Some of the patients were under the medical care of independent colleagues who treated the patients solely according to their customary clinical practice. Two patients (63, 8) received only glucose solutions postoperatively. Inspection of their urinary excretion data showed no clear major differences from those of the other patients and we have, therefore, treated the group as if it were homogenous with respect to postoperative nutrition. Non-protein energy was given half as fat (20% Intralipid, Vitrum Co., Stockholm, Sweden) and half as D-glucose (I 5% to 30% in water). Mean energy intake was 148 * 20 KJ kg/body wt-l/24 h-l. Mean nitrogen intake, as crystalline amino acid solutions, (Vamine, Vitrum Co.) was 0.20 + 0.04 gN/kg-‘/24h-‘. Vitamins (Soluvit, Vitalipids) and traceelements (Addamel) were given in sufficient amounts according to the manufacturer’s recommendations (Vitrum Company). TPN was administered continuously over 24 hours, except when saline was infused overnight before our metabolic measurements were made. These measurements included measurement of whole-body gas exchange carried out over 30 minutes and they were used to gain insight into the energy-exchange status of the patients. The estimated marginally negative energy balance in some patients may be due to some extent to the infusion of saline alone for 8 to 12 hours during the night before measurements of leg metabolite exchange. Plasma from blood samples was analyzed for tyrosine and 3methylhistidine using an amino acid analyzer, with post-column detection of the orthophthalaldehyde derivatives of tyrosine and
RENNIE ET AL
3-methylhistidine.” This method enables 3-methylhistidine in plasma to be estimated in the presence of large amounts of histidine since the fluorescence of histidine is reduced by the addition to the post-column reaction system of 1% formaldehyde. Twenty-four-hour urine samples had 1% by volume of 6N hydrochloric acid added as preservative and aliquots of the urine samples were used for 3methylhistidine analysis as above and for analysis of creatinine by the alkaline picrate method” and for catecholamines and 3methoxy-4-hydroxymandeleic acid (vannillymandelic acid, VMA) by colourimetric methods.” Urinary nitrogen was estimated using a chemiluminescent method.14 Whole-body nitrogen loss was estimated to be the urinary nitrogen excretion plus 2 g of nitrogen to account for skin and gastrointestinal tract losses preoperatively and 3 g postoperatively to include the additional extrarenal losses observed in these circumstances.15 No patient had any extraordinary extrarenal losses of fluids postoperatively. The apparent nitrogen balance was then calculated as nitrogen intake (including blood and plasma transfusions on the day given) minus estimates of wholebody nitrogen loss. Plasma cortisol was estimated by a competitive protein-binding assay using a commercially available kit (Amersham, UK). Leg blood flow was estimated by plethysmography16; and oxygen consumption and CO2 production across the leg were calculated from the blood flow; arterial and venous blood oxygen and CO, content were estimated from hemoglobin; and PO,, PCO,, pH, and oxygen saturation were measured in an automated Radiometer blood gas analyzer. Whole-body resting energy expenditure before operation was calculated according to the Harris and Benedicts formulae” and postoperatively energy expenditure was measured using indirect calorimetry as described by Long and colleagues.’
Statistics Measurements of leg exchanges before and after operation were statistically compared with a student’s t-test for paired difference. Urine excretions were compared with a group comparison t-test,18 since urine specimens were unavoidably missed in some patients. RESULTS
Nitrogen Balance and Outcome
of Surgery
Urinary excretion values are shown in Table 2 for eight patients from whom a complete set of urine samples was collected. Comparison of the peroperative values (ie, those obtained on the day of surgery) with data for normal subjects show that the patients already had increased nitrogen excretion at this time. The mean intakes of protein and energy over the five-day postoperative period were 1.25 -c 0.25 g protein and 13 1 + 17 KJ per kg body weight per day, and these values did not differ significantly from day to day. All patients were in negative nitrogen balance throughout the postoperative period (mean over the five days of 9.6 * 0.8 g N per 24 hours) and in marginal energy balance by between - 8 k 12 and -25 + 21 MJ kg-‘/24-l. The severity of the operative trauma is to some extent indicated by the extent of nitrogen loss after surgery and also by the extent of the noradrenaline excretion which was similar to that observed in septic and traumatized patients studied by other workers.” It is likely that the extent of the negative energy balance is underestimated since indirect calorimetry was carried out after saline infusion.
URINARY AND LEG 3-METHYLHISTIDINE
Table 2.
EFFLUX
Mean Daily Excretion
253
for Eight Patients of 3-Methylhistidine,
Noradrenaline
Nitrogen,
Before and After Operation.
Creatinine,
Adrenaline.
and
Mean i SE. Postoperatively
Refwsnce 3-methylhistidine (~mol/24
Day 1
Praoperatively
152
+
19
168
+ 22
240
6.5
+ 1.5
16.2
? 3.5
21.2
+
Day 2
17’
220
+ 24*
233
+ 2.1t
20.8
? 2.lt
21.1
8.0 t 1.5
(mmol/24
48
(nmo1/24
8.6 t 1.1
8.3
* 0.6
8.7
+ 0.5
7.9
10
153
*
28
113
? 23’
101
-r 27
68
+ 50
520
+ 77
658
i
553
f
2
453
(nmol/24
-
(fimol/24
significantly
Reference
143’
17
i4
34
+ 6’
29
values
different
are for normal
from subjects
peroperative with
values,
similar
l
=
P -c 0.05,
age and weight
+ 30t
? 2.2t
20.6
t
2.4t
24.5
+ 1.6t
8.6 + 0.9
8.5
+ 11
70*
60
330
*
535
31
+ 5’
+
1.0
-r 1.1
13
+ 11
172
89
+ 158
418
? 44
+ 4*
35
27
+ 6’
? 2*
t = P -e 0.025
index
hydroxymandeleic acid (VMA) were significantly increased over the postoperative period when compared either to the preoperative values or, more markedly, to the reference values (Table 2). The creatinine excretion did not change appreciably in response to operation even in those patients with postoperative fever.
The error introduced in our calculations is, however, unlikely to have been more than 15% based on the reponse of resting energy expenditure to TPN.19 All patients survived the immediate postoperative period (> 30 days). Only three patients were in need of some ventilatory support postoperatively, and one of these (#7) was ventilated via a tracheostoma for more than seven days due to a complication (septicemia). Three patients (#2,4,7) had postoperative fever (> 38 “C for up to six days.
Leg Exchange In the overnight-fasted state before operation, there was a measurable and significant efflux of both tyrosine and 3-methylhistidine from the leg (P < 0.01) (Table 3). The effect of surgery was to double the output of tyrosine from the leg but to decrease the
Urinary Excretion Twenty-four total nitrogen,
hour excretion noradrenaline,
of 3-methylhistidine, and 3-methoxy-4-
Table 3. Changes in arterial concentrations
and leg exchange
as a result of operation.
tyrosine
b.rmol
1
49
5
59
Day 3
+ 6
54
Day 6
* 5
58
t
6
‘)
Arterial-venous (firno
*
Values are means + SE for 11 patients
Day 1
Pre-operatively Arterial
tyrosine
-8
+ 2
-16
+ 4’
-16~4’
*
5
-46
t
-44
f
0.6
-11*3
1-l)
Efflux
tyrostne
(nmol
100
Arterial
-22
g-’
1
+ gb
-24
+ 9
3.5
3.2
2 0.4
3.5
* 0.5
3.6
t
0.6
‘)
Arterial-venous fnmol
10d
min-‘)
4-methylhistidine
(nmol
3-methylhistidine
-0.66
* 0.08
-0.38
+ 0.14’
-0.37
i_ 0.13”
-0.64
+ 0.10
-
+ 0.29
-
+ 0.33b
-1.17
+_ 0.30”
-
t
1 -‘I
of 3-methylhistidine
hwnol
100
gg’
min
1.88
100
uptake 100
Respiratory (ymol
g-’
< 0.10
b-p
< 0.05
‘--p
< 0.025
d--D
< 0.01
2.9
+ 0.3
3.3
* 0.6
2.8
& 0.3
2.7
+ 0.4
7.0
+ 1.4
8.4
YL 1.3”
6.5
t 0.8
6.4
+
+ 0.10
0.62
+ 0.10
0.57
+ 4
267
0.86
’ ~*mol 0,
198
‘I different
1.6
+ O.lob
0.82
+ 39
201
+ 0.10
1)
cortisol
Statistically
0.40
min-‘1
quotient CO;
1
1.74
’ min-‘)
g
(j.unolO,
1.04
‘)
flow
Oxygen
a-_P
245
hh’)
Statisttcally
(Irg
57’
h -‘I
VMA
Artenal
*
h-l)
Noradrenaline
(ml
327
‘)
h
Adrenaline
Blood
+ 46
Day 5
h~‘l
Creatinine
Efflux
Day 4
Day 3
h -‘)
Nitrogen (g/24
Values
pre-operatively
vs post-operatively
* 42
222
+ 27
RENNIE ET AL
254
output of 3-methylhistidine by 40%. The arteriofemoral venous differences of tyrosine before surgery were similar to those observed by other workers in normal overnight-fasted subjects, and the differences after surgery were similar to those observed in patients with clean trauma.*’ During recovery from surgery the extent of the negative tyrosine balance diminished and there was a normalization (ie, an increase) in the efflux of 3-methylhistidine within the first week. Leg blood flow was not significantly different under any of the circumstances examined but the oxygen uptake across the leg increased immediately postoperatively due to an increase in oxygen extraction. There was a marked decrease in the respiratory quotient (RQ) across the leg in the first three days postoperatively, indicating a switch to a more predominantly fat-based metabolism. The absolute values for muscle RQ may be underestimated because the electrode method used for measuring the total carbon dioxide content of the blood does not take into account the Bohr effect. The observed trend towards a more fat-based fuel metabolism probably reflects, however, adaptation to nutritional stress. Plasma cortisol was not significantly elevated as a result of surgery or at any time during recovery, compared to the preoperative values. DISCUSSION
The results of the present investigation confirm that major surgical trauma increases nitrogen and 3methylhistidine excretion. The elevated nitrogen excretion after surgical trauma has often been perceived as being due to a loss of skeletal proteins through the mechanism of increased protein breakdown and this is reflected in the emphasis on postoperative “catabolism.“2~8~20~2’ So far as whole-body protein turnover is concerned, isotopic labeling studies have suggested that elective surgery results in a fall in whole-body protein synthesis,22-24 especially if postoperative dietary intake is reduced, with little or no change in whole-body protein breakmay be different in patients down.25 The situation suffering burns or major accidental trauma in whom it has been suggested that both protein synthesis and protein breakdown in the whole body are elevated with, of course, breakdown increased more.25-28 No direct measurements of the effect of surgical trauma on skeletal-muscle protein breakdown or synthesis are available but it would be surprising if the changes in muscle protein synthesis were not in the same direction as those in the whole body since muscle is such an important contributor to whole-body protein turnover.29 A fall in protein synthesis would cause a release of amino acid from skeletal muscle, even if protein breakdown were unchanged. In contrast, many
workers have suggested, from their interpretation of the increase in urinary 3-methylhistidine excretion ,2~8.2’.27 that skeletal-muscle breakdown is actually elevated after major surgery. The present results show that such an interpretation is untenable here. The observation of a fall in the efflux from leg tissues of 3_methylhistidine, at a time when leg production of tyrosine, whole-body production of 3_methylhistidine, and total nitrogen loss are all increased, provide evidence that increased skeletal-muscle myofibrillar protein breakdown is not responsible for the increased nitrogen loss. Nevertheless, it must be emphasized that we do recognize that urinary excretion values represent production over 24 hours while our measurements of leg amino acid exchanges represent point observations made after temporary withdrawal of TPN. Our results cannot, however, easily be explained away as artifacts occurring as a result of short-term changes in muscle and whole-body metabolism of amino acids. For this to be the case it would be necessary for tyrosine efflux and 3-methylhistidine efflux from the leg to be temporally uncoupled, since tyrosine efflux is mirrored by nitrogen loss in the urine. The excretion of 3-methylhistidine, which is rapidly cleared from the body, does not show diurnal variation (eg, mean excretion in four successive six-hour periods in three normal healthy men on a meat-free diet 62 f 14,68 5 23,59 + 10, and 66 + 11 pmol). In addition, it is not responsive to acute withdrawal of food (up to four days)29‘3’ so that we may expect its production by muscle to be rather steady. Certainly, we have measured arterial and femoral venous blood concentrations of 3-methylhistidine in surgical patients, both before operation and after operation during TPN, which were steady (ie, within c 15%) for periods up to seven hours (Rennie, Emery, & Lundholm, unpublished results). We have, therefore, no factual basis for any suspicion that the isolated measurements of leg production of 3-methylhistidine are unrepresentative of mean changes over the whole day, although we cannot yet rule it out. Other studies32.33 have shown that the net output of tyrosine from muscle is rapidly responsive to nutrient supply, but these studies do not indicate whether this response represents an increase in muscle protein synthesis or a decrease in muscle protein degradation. Our own studies suggest that muscle protein synthesis increases on feeding, while muscle protein degradation does not change significantly.29 We are not aware of any direct evidence that muscle protein degradation changes in response to short-term fasting. With this proviso, however, our results strongly suggest that some other source of 3-methylhistidine besides skeletal muscle must be contributing to the whole-body pro-
URINARY AND LEG 3-METHYLHISTIDINE
EFFLUX
255
duction rate postoperatively, and furthermore, that, nonskeletal muscle contribution must be elevated as a result of surgical trauma. We have no direct indication of the precise site from which this 3-methylhistidine is being produced but work by ourselves and by others on animal models suggests that the smooth muscle of the gut, among other tissues, is a likely candidate.3.6.34 Although the present results probably rule out an increase in muscle-protein breakdown as the major mechanism resulting in an increased nitrogen loss consequent to surgical trauma, there is nevertheless an increase in the net loss from leg tissues, principally muscle, of amino acids as indicated by tyrosine loss. Results from our own” and other studies” have shown a highly significant correlation between net exchanges of tyrosine and of total amino acids in 45 fed and fasted subjects. The increases in amino acid efflux from muscle will of course contribute to an increased net negative nitrogen balance of the whole body as well as of skeletal muscle. Given the apparent decrease in myofibrillar protein degradation, indicated by the fall in 3-methylhistidine efflux, the only mechanism capable of producing the net loss of amino acids is a fall in muscle protein synthesis below the rate of muscle protein degradation. Similar circumstances of apparently decreased muscle protein synthesis in the presence of decreased muscle protein degradation have been deduced in our previous studies of tyrosine and 3-methylhistidine balance across the limbs of patients suffering from malnutrition and muscle wasting consequent upon malignancy.7 Thus, it seems reasonable to suggest from the present results that postoperative “catabolism” does not in general occur as a result of increasing catabolic processes per se, at least so far as muscle is concerned, but as a result of the decrease in protein synthesis. The discrepancy between the direction of change of 3-methylhistidine efflux from the limb and 3-methylhistidine excretion is strongly suggestive of the involve-
ment of nonskeletal muscle pools of 3-methylhistidine which contribute substantially to whole-body production of this amino acid. Such observations invalidate the definitive use of 3-methylhistidine in urine as a specific index of skeletal muscle protein breakdown. This evidence confirms the doubts raised on the basis of previous animal and human studies.3-6 In addition, however, it raises the question of the extent to which smooth-muscle-containing tissues, particularly the gut, contribute to the supply of protein-derived amino acids, eg, in acute-phase protein synthesis, wound healing, and gluconeogenesis. Theoretically, the contribution of the nonskeletal muscle tissues could be assessed by the difference between the total excretion rate and the efflux from skeletal muscle assuming that the fiber-type composition of leg muscle was representative of that in the whole body and calculating the whole-body skeletal muscle mass from creatinine excretion. Unfortunately, the plethysmographic method of measuring limb blood flow does not distinguish between skeletal muscle and nonmuscle blood flow. Moreover, we do not know whether the point measurements of leg efflux of 3-methylhistidine can be quantitatively extrapolated over 24 hours. In summary, the present results provide evidence that amino acid efflux from skeletal muscle in response to major surgery contributes to the negative nitrogen balance of the whole body mainly as a result of a fall in muscle-protein synthesis, with adaptive falls in muscleprotein breakdown, and that the observed increase in 3-methylhistidine excretion must be derived from tissues other than skeletal muscle.
ACKNOWLEDGMENTS We thank our clinical colleagues for their help in these studies. We are grateful for the help and encouragement of Professors T. ScherstCn and R. H. T. Edwards, and for the first-class technical assistance of Margaret Nathan in running the amino acid analyzer.
REFERENCES I. Bilmazes C, Uuauy R, Haverberg LN, et al: Muscle protein breakdown rates in humans based on N’-methylhistidine (3-methylhistidine) content of mixed proteins in skeletal muscle and urinary output of N’-methylhistidine. Metabolism 27:525-530, 1978 2. Young VR, Munro HN: N’-methylhistidine (3-methylhistidine) and muscle protein turnover: An overview. Federation Proc 37:2291-2300, 1978 3. Millward DJ. Bates PC, Broadbent P, et al: Sources of 3-methylhistidine in the body (Chap 2l), in Wesdorp RIC (ed): Clinical Nutrition ‘8 I. Edinburgh, Churchill-Livingstone, 1982 4. Rennie MJ, Edwards RHT, Halliday E, et al: The effect of Duchenne muscular dystrophy on muscle protein synthesis. Nature 296165-167. 1982 5. Griggs RC, Rennie MJ: Muscle wasting in muscular dystrophy: Decreased protein synthesis or increased degradation? Ann Neurol 13:125-132, 1983
6. Cotellessa L, Emery PW, Rennie MJ: Paradoxical changes in 3-methylhistidine production and muscle protein degradation in fasted rats. Proc Nutr Sot 42:26A, 1983 7. Lundholm K, Benneglrd K, Eden E, et al: Efflux of 3methylhistidine from the leg in cancer patients who experience weight loss. Cancer Res 42:4807-48 I 1, 1982 8. Long CL, SchatTel N, Geiger JW, et al: Metabolic response to injury and illness: Estimation of energy and protein needs from indirect calorimetry and nitrogen balance. J Par Ent Nutr 3:452456,1979 9. Bengtsson C, Hulttn B, Larsson B, et al: Nya Ilngdvikttabeller fiir modelalders och aldre man och kvinnor. Llkartidningen 78:3152-3154, 1981 IO. BennegHrd K, Eden E, Ekman L, et al: Metabolic balance across the leg in weight-losing cancer patients compared to depleted patients without cancer. Cancer Res 42:4293-4299. 1982
256
11. Emery PW, Rennie MJ: Elimination of interference of histidine by use of formaldehyde in the measurement of 3-methylhistidine: Application to study of muscle protein degradation. Anal Biochem 126:67-73, 1982 12. Varley H: Practical Clinical Biochemistry (4th Ed). London, Heinemann, 1967 13. Euler von US, Lishajro F: The estimation of catecholamines in urine. Acta Physiol Stand 45: 122-l 32, 1959 14. Ward MWN, Owens CWI, Rennie MJ: Nitrogen estimation in biological samples by use of chemiluminescence. Clin Chem 26:1336-1339, 1980 15. Caldwell MD, Kennedy-Caldwell C: Normal nutritional requirements. Surg Clin North Am 61:489-507, 1981 16. Aist, Needham TM: The measurement of blood flow: Strain gauge plethysmography. Bio-Medical Engineering, July 1972, pp 266-269 17. Harris JA, Benedict FG: Biometric studies of basal metabolism in man. Carnegie Institute of Washington, pub1 279, 1919 18. Woolf CM: Statistics for Biologists. Principles of Biometry. New York, D. van Nostrand Co, Inc, 1968 19. Askanazi J, Carpentier YA, Elwyn DH, et al: Influence of total parenteral nutrition on fuel utilization in injury and sepsis. Ann Surg 191:40-46, 1980 20. Clowes GHA, Randall HT, Cha CJ: Amino acid and energy metabolism in septic and traumatized patients. J Parent Ent Nut 4: 195-205, 1980 2 1. Niehluser M, Bergstrom J, Chao L, et al: Urinary excretion of 3-methylhistidine as an index of muscle protein catabolism in post-operative trauma: The effect of parenteral nutrition. Metabolism 29:1206-l 213, 1980 22. O’Keefe SJD, Sender PM, James WPT: Catabolic loss of body nitrogen in response to surgery. Lancer 2:1035, 1974 23. Crane CW, Picou D, Smith R, et al: Protein turnover in patients before and after elective orthopaedic operations. Br J Surg 64:129-133,1977 24. Kien CL, Young VR, Rohrbaugh DK, et al: Whole body
RENNIE ET AL
protein synthesis and breakdown rates in children before and after reconstructive surgery of the skin. Metabolism 27:27-34, 1978 25. Clague MB: Turnover in pathological states (Chap. 41) in Waterlow JC and Stephen JML (eds). Nitrogen Metabolism in Man. London, Applied Science Publishers, 1981 26. Kien CL, Young VR, Rohrbaugh DK, et al: Increased rates of whole body protein synthesis and breakdown in children recovering from burns. Ann Surg 187:383-391, 1978 27. Long CL, Jeevanandam M, Kim BM, et al: Whole body protein synthesis and catabolism in septic man. Am J Clin Nutr 30:1340-1352, 1977 28. Birkhahn RH, Long CL, Fitkin D, et al: Effects of major skeletal trauma on whole body protein turnover in man measured by L-(1“‘C) leucine. Surgery 88:294-299, 1980 29. Rennie MJ, Edwards RHT, Halliday D, et al: Muscle protein synthesis measured by stable isotope technique in man: The effects of feeding and fasting. Clin Sci 63:519-523, 1982 30. Marliss EB, Murray FT. Nakbooda AF: The metabolic response to hypocaloric diets in obese man. J Clin Invest 62:468479,197s 31. Elia M, Carter A, Bacon C, et al: Clinical usefulness of urinary 3-methylhistidine excretion in indicating muscle protein breakdown. Br Med J 282:351-354,1981 32. Clowes GHA, Heideman M, Lindberg B, et al: Effect of parenteral alimentation on amino acid metabolism in septic patients, Surgery 88:531-543, 1980 33. Abumrad NN, Rabin D, Wise KL, et al: The disposal of an intravenously administered amino acid load across the human forearm. Metabolism 31:463-470, 1982 34. Wassner SJ, and Li JB: N’-methylhistidine release: Contributions of the rat skeletal muscle, G-I tract and skin. Am J Physiol 243:E293-E297, 1982 35. Bennegard K, Lindmark L, Ed&n E, et al: Flux of amino acids across the leg in weight-losing cancer patients. Cancer Res (in press)
J. Rennie,
K. Bennegsrd,
E. Ed&,
P. W.
Emery,
and K. Lundholm
Changes in the &fluxes from the leg of 3-methylhistidine and tyrosine were studied in relation to alterations in the 24-hour excretion of 3-methylhistidine and total nitrogen in 11 patients before and after undergoing major surgical operation. On the first day after operation, efflux of 3-methylhistidine from the leg was significantly decreased by 40% compared to preoperative values. In contrast, tyrosine efflux was doubled at the same time as a transient 20% increase in oxygen uptake of the leg and a marked increase in catecholamine excretion were observed. These changes coincided with a 40% elevation in the excretion of both 3-methylhistidine and nitrogen. Leg metabolism returned to the preoperative pattern within a week. These results suggest that the loss of amino acids from the lean tissues of the leg is the result of a fall in protein synthesis accompanied by an adaptive fall in protein breakdown. Although the increase in nitrogen excretion in response to major surgical trauma reflects the negative amino acid balance of skeletal muscle, the changes in urinary 3-methylhistidine do not correlate with changes in efflux of 3-methylhistidine from the leg. These results suggest that the use of 3-methylhistidine excretion as a specific index of skeletal muscle protein breakdown in postoperative patients may be invalid. Tissues other than skeletal muscle appear to make a substantial contribution to the 3-methylhistidine excretion postoperatively.
INCE the breakdown of tissue protein is difficult to measure for technical reasons, the use of 3methylhistidine excreted in urine as a specific indicator of skeletal muscle protein breakdown’,2 rapidly became widespread; the method was easy and rapid. Initial expectations of the usefulness of the method have not been fulfilled partly because of doubts as to the specificity of 3-methylhistidine as an indicator of breakdown of skeletal muscle protein3 and also because in some cases the results indicated changes in skeletalmuscle protein breakdown at variance to those obtained by other reliable methods.4,5 A major problem in the interpretation of the measurements of the rate of excretion of 3-methylhistidine is the extent to which nonskeletal muscle tissues (which contain contractile protein in smooth muscle) contribute to the whole-body production.’ Evidence is accumulating that in animals gastrointestinal smooth muscle may contribute a large fraction to the wholebody production under normal circumstances and that during starvation this proportion may rise (6, PW Emery, M Holness, and MJ Rennie, unpublished).
S
From the Department of Medicine, University College London School of Medicine, The Rayne Institute, London, England; and the Department of Surgefly I and Anaesthesiology II. Sahlgrenska Hospital. Gothenburg, Sweden. This work was supported by grants from The Cancer Research Campaign, the United Kingdom Medical Research Council, The Welfcome Trust and the Swedish Cancer Society (project No 536), the Assar Gabrielsson’Foundation. the Serena Ehrenstriim Foundation, the Swedish Medical Socieiy, and the Gothenburg Medical Society. Address reprint requests to M. J. Rennie. Department of Physiology, University of Dundee, Dundee DDI IHN, Scotland, England and Kent Lundholm. Department of Surgery I, Sahlgrenska Hospital, S-413 45 Gothenburg. Sweden. 0 1984 by Grune & Stratton, Inc. 0026-0495/84/3303~12%01.00/0 250
Unfortunately, not all methods used in animal studies are applicable in humans and there is still considerable uncertainty about the possible importance of nonskeletal sources of 3-methydhistidine in humans. In order to minimize interference from nonskeletal muscle sources of 3_methylhistidine, we have developed a method to measure the efflux of 3-methylhistidine from tissues of the leg by comparison of arterial and femoral venous concentration of 3-methylhistidine.7 We expect that the general application of the method will provide some insight into possible changes in skeletal-muscle protein breakdown in circumstances of clinical interest. It is well known that major surgery results in increased loss of body nitrogen and an increase in the excretion of 3-methylhistidine.*~’ We have, therefore, applied our techniques to a study of the changes in leg efflux and excretion of 3-methylhistidine in patients undergoing major abdominal surgery. The aims of the study are dual: first, to examine the effects of surgical trauma on skeletal muscle amino acid balance to gain some insight into the mechanisms involved in muscle wasting; and second, to assess the role of the contribution of skeletal muscle to the increased whole-body production rate of 3-methylhistidine which occurs as a result of surgical trauma. MATERIALS
AND
METHODS
Patients Eleven patients admitted grenska Hospital
to the Department
the diagnoses and nutritional Six
patients
with
through abdominal formed
of Surgery at Sahl-
for major operations were studied. Table gastric
states of the patients tumors
underwent
total
gastrectomy
incisions. A classic hemihepatectomy
in the single case with liver tumors.
ulcerative colitis and rectal carcinoma
One patient with urinary bladder carcinoma with the Bricker-deviation
Two
underwent
1 shows
before operation. was per-
patients
with
proctocolectomy.
underwent cystectomy
procedure.
Metabolism, Vol 33, No 3 (March), 1984
84 62 k4
64
54
57 + 4
11. Gastric carcinosarcoma (MI
Mean k SE
10.
63
63
38
64
76
52
72
6. Urinary bladder carcinoma (M)
51 47
Gastric carcinoma (M)
29
5. Ulcerative colitis (F)
9. Gastric lymphosarcoma IF)
56
4. Gastric carcinoma (F)
75
75
41
3. Benign liver tumour (F)
54
63
65
51
1. Gastric Carclnoms (F)
2. Primary liver (F)
(kg)
7. Gastric carcinoma (M)
65
No. Diagnoses
Weight
8. Rectal carcinoma (F)
Age (Yf)
0.88
+ 0.05
0.94
0.83
0.93
0.56
0.89
1.1
0.73
0.76
1.1
0.85
0.95
(%)
Ideal Weight
Apparent:
Postoperatively
yrb
0
315 mo
vr 2.11 vr 5/l
513 mo
0
0
2
none
0
lo-1512
(kg)
LOSS
Weight
11.5 + 1.0
12.0
8.7
19.0
5.4
7.0
10.0
19.0
11.0
11.2
12.0
11.5
(mm)
Skinfold
Triceps
26 + 1
30.0
26.0
26.0
20.5
25.2
30.5
21.8
21.5
30.5
25.0
25.5
Icml
Ctrcumference
Mtid-Arm
36 + 2
23
42
38
44
45
37
23
33
39
41
28
(g/L)
Albumin
Serum
5.3 + 0.7
10.0
37.9 37 ” 0.1
5.5
3.0
2.5
6.5
8.5
3.0
6.5
3.5
5.5
37.3
36.6
36.8
37.0
37.9
37.1
36.8
37.2
36.6
4.0
(hr)
(“a 37.2
Time
Operation
Temperature
Body
Table 1. The Nutritional State of the Patients Before Operation and the Diagnosis, Operation Time, Blood Loss, and Ventilatory Support
1.7
1.6 f 0.4
1.7
0.4
0.7
0.2
3.1
3.0
2.5 0.2
0.4
4.0
12h
Jd
-
14h
Postoperatively
Support
Ventilatory
252
All patients except two had body weights less than the ideal.’ Six patients had lost weight (2 to 15 kg) over the year preceding operation; four of them (numbers I, 4, 5, and 11) had decreased concentrations (< 36 g/L) of serum albumin. Two patients were subfebrile (temperature 37.0 to 37.YC) before operation. All patients were operated on by senior and experienced surgeons. Patient #5 received steroids (500 mg SolucorteR, Upjohn) on each of 3 days preoperatively and on the day of operation. Patient #6 was given a selective B-blocker, metoprolol (Seloke#, H$ssle A.B., Gottenburg, 200 mg/d) postoperatively. The patients were admitted to the hospital at least five days before operation. The patients ate an ordinary hospital diet consisting of 40% of energy as fat, 50% as carbohydrates and 10% as proteins to supply at least 6.3 MJ and 75 g protein. During these days the patients underwent a program of routine preoperative investigations, which made it impossible to make complete collections of urine with confidence until the day preceding operation. From 24 hours before our first measurements of leg metabolism on the day before operation, the patients received only parenteral glucose solutions (10% weight per volume) except that for the last 8 to 13 hours (ie, overnight) before each set of measurements, only saline was infused. This was done to standardize, as much as was possible under the circumstances, the metabolic states of the patients before operation and to make them more directly comparable to the measurements taken in the postoperative state (see below). For urine collections made in the last 24 hours before operation and from then onwards it was possible to be sure of completeness of collection since these were made via urinary catheters. In addition to following the time-course of excretion of nitrogen and urinary metabolites from immediately before and during operation (ie, in urine collected from 8 pm on the day preceding to 8 pm on the operative day), and during five days postoperatively, we have included reference values for excretions collected from cancer and noncancer patients of similar age and body weight index studied in the same hospital under the same conditions. In all patients, leg exchange of amino acids and leg blood flow were measured between 7 AM and 8 AM (by techniques described elsewhere”) on at least four occasions: immediately before surgery, 24 hours after surgery (day 1). and three and six days after surgery. Total parenteral nutrition (TPN) was given to all patients during recovery from surgery. The TPN protocols were not planned by us. Some of the patients were under the medical care of independent colleagues who treated the patients solely according to their customary clinical practice. Two patients (63, 8) received only glucose solutions postoperatively. Inspection of their urinary excretion data showed no clear major differences from those of the other patients and we have, therefore, treated the group as if it were homogenous with respect to postoperative nutrition. Non-protein energy was given half as fat (20% Intralipid, Vitrum Co., Stockholm, Sweden) and half as D-glucose (I 5% to 30% in water). Mean energy intake was 148 * 20 KJ kg/body wt-l/24 h-l. Mean nitrogen intake, as crystalline amino acid solutions, (Vamine, Vitrum Co.) was 0.20 + 0.04 gN/kg-‘/24h-‘. Vitamins (Soluvit, Vitalipids) and traceelements (Addamel) were given in sufficient amounts according to the manufacturer’s recommendations (Vitrum Company). TPN was administered continuously over 24 hours, except when saline was infused overnight before our metabolic measurements were made. These measurements included measurement of whole-body gas exchange carried out over 30 minutes and they were used to gain insight into the energy-exchange status of the patients. The estimated marginally negative energy balance in some patients may be due to some extent to the infusion of saline alone for 8 to 12 hours during the night before measurements of leg metabolite exchange. Plasma from blood samples was analyzed for tyrosine and 3methylhistidine using an amino acid analyzer, with post-column detection of the orthophthalaldehyde derivatives of tyrosine and
RENNIE ET AL
3-methylhistidine.” This method enables 3-methylhistidine in plasma to be estimated in the presence of large amounts of histidine since the fluorescence of histidine is reduced by the addition to the post-column reaction system of 1% formaldehyde. Twenty-four-hour urine samples had 1% by volume of 6N hydrochloric acid added as preservative and aliquots of the urine samples were used for 3methylhistidine analysis as above and for analysis of creatinine by the alkaline picrate method” and for catecholamines and 3methoxy-4-hydroxymandeleic acid (vannillymandelic acid, VMA) by colourimetric methods.” Urinary nitrogen was estimated using a chemiluminescent method.14 Whole-body nitrogen loss was estimated to be the urinary nitrogen excretion plus 2 g of nitrogen to account for skin and gastrointestinal tract losses preoperatively and 3 g postoperatively to include the additional extrarenal losses observed in these circumstances.15 No patient had any extraordinary extrarenal losses of fluids postoperatively. The apparent nitrogen balance was then calculated as nitrogen intake (including blood and plasma transfusions on the day given) minus estimates of wholebody nitrogen loss. Plasma cortisol was estimated by a competitive protein-binding assay using a commercially available kit (Amersham, UK). Leg blood flow was estimated by plethysmography16; and oxygen consumption and CO2 production across the leg were calculated from the blood flow; arterial and venous blood oxygen and CO, content were estimated from hemoglobin; and PO,, PCO,, pH, and oxygen saturation were measured in an automated Radiometer blood gas analyzer. Whole-body resting energy expenditure before operation was calculated according to the Harris and Benedicts formulae” and postoperatively energy expenditure was measured using indirect calorimetry as described by Long and colleagues.’
Statistics Measurements of leg exchanges before and after operation were statistically compared with a student’s t-test for paired difference. Urine excretions were compared with a group comparison t-test,18 since urine specimens were unavoidably missed in some patients. RESULTS
Nitrogen Balance and Outcome
of Surgery
Urinary excretion values are shown in Table 2 for eight patients from whom a complete set of urine samples was collected. Comparison of the peroperative values (ie, those obtained on the day of surgery) with data for normal subjects show that the patients already had increased nitrogen excretion at this time. The mean intakes of protein and energy over the five-day postoperative period were 1.25 -c 0.25 g protein and 13 1 + 17 KJ per kg body weight per day, and these values did not differ significantly from day to day. All patients were in negative nitrogen balance throughout the postoperative period (mean over the five days of 9.6 * 0.8 g N per 24 hours) and in marginal energy balance by between - 8 k 12 and -25 + 21 MJ kg-‘/24-l. The severity of the operative trauma is to some extent indicated by the extent of nitrogen loss after surgery and also by the extent of the noradrenaline excretion which was similar to that observed in septic and traumatized patients studied by other workers.” It is likely that the extent of the negative energy balance is underestimated since indirect calorimetry was carried out after saline infusion.
URINARY AND LEG 3-METHYLHISTIDINE
Table 2.
EFFLUX
Mean Daily Excretion
253
for Eight Patients of 3-Methylhistidine,
Noradrenaline
Nitrogen,
Before and After Operation.
Creatinine,
Adrenaline.
and
Mean i SE. Postoperatively
Refwsnce 3-methylhistidine (~mol/24
Day 1
Praoperatively
152
+
19
168
+ 22
240
6.5
+ 1.5
16.2
? 3.5
21.2
+
Day 2
17’
220
+ 24*
233
+ 2.1t
20.8
? 2.lt
21.1
8.0 t 1.5
(mmol/24
48
(nmo1/24
8.6 t 1.1
8.3
* 0.6
8.7
+ 0.5
7.9
10
153
*
28
113
? 23’
101
-r 27
68
+ 50
520
+ 77
658
i
553
f
2
453
(nmol/24
-
(fimol/24
significantly
Reference
143’
17
i4
34
+ 6’
29
values
different
are for normal
from subjects
peroperative with
values,
similar
l
=
P -c 0.05,
age and weight
+ 30t
? 2.2t
20.6
t
2.4t
24.5
+ 1.6t
8.6 + 0.9
8.5
+ 11
70*
60
330
*
535
31
+ 5’
+
1.0
-r 1.1
13
+ 11
172
89
+ 158
418
? 44
+ 4*
35
27
+ 6’
? 2*
t = P -e 0.025
index
hydroxymandeleic acid (VMA) were significantly increased over the postoperative period when compared either to the preoperative values or, more markedly, to the reference values (Table 2). The creatinine excretion did not change appreciably in response to operation even in those patients with postoperative fever.
The error introduced in our calculations is, however, unlikely to have been more than 15% based on the reponse of resting energy expenditure to TPN.19 All patients survived the immediate postoperative period (> 30 days). Only three patients were in need of some ventilatory support postoperatively, and one of these (#7) was ventilated via a tracheostoma for more than seven days due to a complication (septicemia). Three patients (#2,4,7) had postoperative fever (> 38 “C for up to six days.
Leg Exchange In the overnight-fasted state before operation, there was a measurable and significant efflux of both tyrosine and 3-methylhistidine from the leg (P < 0.01) (Table 3). The effect of surgery was to double the output of tyrosine from the leg but to decrease the
Urinary Excretion Twenty-four total nitrogen,
hour excretion noradrenaline,
of 3-methylhistidine, and 3-methoxy-4-
Table 3. Changes in arterial concentrations
and leg exchange
as a result of operation.
tyrosine
b.rmol
1
49
5
59
Day 3
+ 6
54
Day 6
* 5
58
t
6
‘)
Arterial-venous (firno
*
Values are means + SE for 11 patients
Day 1
Pre-operatively Arterial
tyrosine
-8
+ 2
-16
+ 4’
-16~4’
*
5
-46
t
-44
f
0.6
-11*3
1-l)
Efflux
tyrostne
(nmol
100
Arterial
-22
g-’
1
+ gb
-24
+ 9
3.5
3.2
2 0.4
3.5
* 0.5
3.6
t
0.6
‘)
Arterial-venous fnmol
10d
min-‘)
4-methylhistidine
(nmol
3-methylhistidine
-0.66
* 0.08
-0.38
+ 0.14’
-0.37
i_ 0.13”
-0.64
+ 0.10
-
+ 0.29
-
+ 0.33b
-1.17
+_ 0.30”
-
t
1 -‘I
of 3-methylhistidine
hwnol
100
gg’
min
1.88
100
uptake 100
Respiratory (ymol
g-’
< 0.10
b-p
< 0.05
‘--p
< 0.025
d--D
< 0.01
2.9
+ 0.3
3.3
* 0.6
2.8
& 0.3
2.7
+ 0.4
7.0
+ 1.4
8.4
YL 1.3”
6.5
t 0.8
6.4
+
+ 0.10
0.62
+ 0.10
0.57
+ 4
267
0.86
’ ~*mol 0,
198
‘I different
1.6
+ O.lob
0.82
+ 39
201
+ 0.10
1)
cortisol
Statistically
0.40
min-‘1
quotient CO;
1
1.74
’ min-‘)
g
(j.unolO,
1.04
‘)
flow
Oxygen
a-_P
245
hh’)
Statisttcally
(Irg
57’
h -‘I
VMA
Artenal
*
h-l)
Noradrenaline
(ml
327
‘)
h
Adrenaline
Blood
+ 46
Day 5
h~‘l
Creatinine
Efflux
Day 4
Day 3
h -‘)
Nitrogen (g/24
Values
pre-operatively
vs post-operatively
* 42
222
+ 27
RENNIE ET AL
254
output of 3-methylhistidine by 40%. The arteriofemoral venous differences of tyrosine before surgery were similar to those observed by other workers in normal overnight-fasted subjects, and the differences after surgery were similar to those observed in patients with clean trauma.*’ During recovery from surgery the extent of the negative tyrosine balance diminished and there was a normalization (ie, an increase) in the efflux of 3-methylhistidine within the first week. Leg blood flow was not significantly different under any of the circumstances examined but the oxygen uptake across the leg increased immediately postoperatively due to an increase in oxygen extraction. There was a marked decrease in the respiratory quotient (RQ) across the leg in the first three days postoperatively, indicating a switch to a more predominantly fat-based metabolism. The absolute values for muscle RQ may be underestimated because the electrode method used for measuring the total carbon dioxide content of the blood does not take into account the Bohr effect. The observed trend towards a more fat-based fuel metabolism probably reflects, however, adaptation to nutritional stress. Plasma cortisol was not significantly elevated as a result of surgery or at any time during recovery, compared to the preoperative values. DISCUSSION
The results of the present investigation confirm that major surgical trauma increases nitrogen and 3methylhistidine excretion. The elevated nitrogen excretion after surgical trauma has often been perceived as being due to a loss of skeletal proteins through the mechanism of increased protein breakdown and this is reflected in the emphasis on postoperative “catabolism.“2~8~20~2’ So far as whole-body protein turnover is concerned, isotopic labeling studies have suggested that elective surgery results in a fall in whole-body protein synthesis,22-24 especially if postoperative dietary intake is reduced, with little or no change in whole-body protein breakmay be different in patients down.25 The situation suffering burns or major accidental trauma in whom it has been suggested that both protein synthesis and protein breakdown in the whole body are elevated with, of course, breakdown increased more.25-28 No direct measurements of the effect of surgical trauma on skeletal-muscle protein breakdown or synthesis are available but it would be surprising if the changes in muscle protein synthesis were not in the same direction as those in the whole body since muscle is such an important contributor to whole-body protein turnover.29 A fall in protein synthesis would cause a release of amino acid from skeletal muscle, even if protein breakdown were unchanged. In contrast, many
workers have suggested, from their interpretation of the increase in urinary 3-methylhistidine excretion ,2~8.2’.27 that skeletal-muscle breakdown is actually elevated after major surgery. The present results show that such an interpretation is untenable here. The observation of a fall in the efflux from leg tissues of 3_methylhistidine, at a time when leg production of tyrosine, whole-body production of 3_methylhistidine, and total nitrogen loss are all increased, provide evidence that increased skeletal-muscle myofibrillar protein breakdown is not responsible for the increased nitrogen loss. Nevertheless, it must be emphasized that we do recognize that urinary excretion values represent production over 24 hours while our measurements of leg amino acid exchanges represent point observations made after temporary withdrawal of TPN. Our results cannot, however, easily be explained away as artifacts occurring as a result of short-term changes in muscle and whole-body metabolism of amino acids. For this to be the case it would be necessary for tyrosine efflux and 3-methylhistidine efflux from the leg to be temporally uncoupled, since tyrosine efflux is mirrored by nitrogen loss in the urine. The excretion of 3-methylhistidine, which is rapidly cleared from the body, does not show diurnal variation (eg, mean excretion in four successive six-hour periods in three normal healthy men on a meat-free diet 62 f 14,68 5 23,59 + 10, and 66 + 11 pmol). In addition, it is not responsive to acute withdrawal of food (up to four days)29‘3’ so that we may expect its production by muscle to be rather steady. Certainly, we have measured arterial and femoral venous blood concentrations of 3-methylhistidine in surgical patients, both before operation and after operation during TPN, which were steady (ie, within c 15%) for periods up to seven hours (Rennie, Emery, & Lundholm, unpublished results). We have, therefore, no factual basis for any suspicion that the isolated measurements of leg production of 3-methylhistidine are unrepresentative of mean changes over the whole day, although we cannot yet rule it out. Other studies32.33 have shown that the net output of tyrosine from muscle is rapidly responsive to nutrient supply, but these studies do not indicate whether this response represents an increase in muscle protein synthesis or a decrease in muscle protein degradation. Our own studies suggest that muscle protein synthesis increases on feeding, while muscle protein degradation does not change significantly.29 We are not aware of any direct evidence that muscle protein degradation changes in response to short-term fasting. With this proviso, however, our results strongly suggest that some other source of 3-methylhistidine besides skeletal muscle must be contributing to the whole-body pro-
URINARY AND LEG 3-METHYLHISTIDINE
EFFLUX
255
duction rate postoperatively, and furthermore, that, nonskeletal muscle contribution must be elevated as a result of surgical trauma. We have no direct indication of the precise site from which this 3-methylhistidine is being produced but work by ourselves and by others on animal models suggests that the smooth muscle of the gut, among other tissues, is a likely candidate.3.6.34 Although the present results probably rule out an increase in muscle-protein breakdown as the major mechanism resulting in an increased nitrogen loss consequent to surgical trauma, there is nevertheless an increase in the net loss from leg tissues, principally muscle, of amino acids as indicated by tyrosine loss. Results from our own” and other studies” have shown a highly significant correlation between net exchanges of tyrosine and of total amino acids in 45 fed and fasted subjects. The increases in amino acid efflux from muscle will of course contribute to an increased net negative nitrogen balance of the whole body as well as of skeletal muscle. Given the apparent decrease in myofibrillar protein degradation, indicated by the fall in 3-methylhistidine efflux, the only mechanism capable of producing the net loss of amino acids is a fall in muscle protein synthesis below the rate of muscle protein degradation. Similar circumstances of apparently decreased muscle protein synthesis in the presence of decreased muscle protein degradation have been deduced in our previous studies of tyrosine and 3-methylhistidine balance across the limbs of patients suffering from malnutrition and muscle wasting consequent upon malignancy.7 Thus, it seems reasonable to suggest from the present results that postoperative “catabolism” does not in general occur as a result of increasing catabolic processes per se, at least so far as muscle is concerned, but as a result of the decrease in protein synthesis. The discrepancy between the direction of change of 3-methylhistidine efflux from the limb and 3-methylhistidine excretion is strongly suggestive of the involve-
ment of nonskeletal muscle pools of 3-methylhistidine which contribute substantially to whole-body production of this amino acid. Such observations invalidate the definitive use of 3-methylhistidine in urine as a specific index of skeletal muscle protein breakdown. This evidence confirms the doubts raised on the basis of previous animal and human studies.3-6 In addition, however, it raises the question of the extent to which smooth-muscle-containing tissues, particularly the gut, contribute to the supply of protein-derived amino acids, eg, in acute-phase protein synthesis, wound healing, and gluconeogenesis. Theoretically, the contribution of the nonskeletal muscle tissues could be assessed by the difference between the total excretion rate and the efflux from skeletal muscle assuming that the fiber-type composition of leg muscle was representative of that in the whole body and calculating the whole-body skeletal muscle mass from creatinine excretion. Unfortunately, the plethysmographic method of measuring limb blood flow does not distinguish between skeletal muscle and nonmuscle blood flow. Moreover, we do not know whether the point measurements of leg efflux of 3-methylhistidine can be quantitatively extrapolated over 24 hours. In summary, the present results provide evidence that amino acid efflux from skeletal muscle in response to major surgery contributes to the negative nitrogen balance of the whole body mainly as a result of a fall in muscle-protein synthesis, with adaptive falls in muscleprotein breakdown, and that the observed increase in 3-methylhistidine excretion must be derived from tissues other than skeletal muscle.
ACKNOWLEDGMENTS We thank our clinical colleagues for their help in these studies. We are grateful for the help and encouragement of Professors T. ScherstCn and R. H. T. Edwards, and for the first-class technical assistance of Margaret Nathan in running the amino acid analyzer.
REFERENCES I. Bilmazes C, Uuauy R, Haverberg LN, et al: Muscle protein breakdown rates in humans based on N’-methylhistidine (3-methylhistidine) content of mixed proteins in skeletal muscle and urinary output of N’-methylhistidine. Metabolism 27:525-530, 1978 2. Young VR, Munro HN: N’-methylhistidine (3-methylhistidine) and muscle protein turnover: An overview. Federation Proc 37:2291-2300, 1978 3. Millward DJ. Bates PC, Broadbent P, et al: Sources of 3-methylhistidine in the body (Chap 2l), in Wesdorp RIC (ed): Clinical Nutrition ‘8 I. Edinburgh, Churchill-Livingstone, 1982 4. Rennie MJ, Edwards RHT, Halliday E, et al: The effect of Duchenne muscular dystrophy on muscle protein synthesis. Nature 296165-167. 1982 5. Griggs RC, Rennie MJ: Muscle wasting in muscular dystrophy: Decreased protein synthesis or increased degradation? Ann Neurol 13:125-132, 1983
6. Cotellessa L, Emery PW, Rennie MJ: Paradoxical changes in 3-methylhistidine production and muscle protein degradation in fasted rats. Proc Nutr Sot 42:26A, 1983 7. Lundholm K, Benneglrd K, Eden E, et al: Efflux of 3methylhistidine from the leg in cancer patients who experience weight loss. Cancer Res 42:4807-48 I 1, 1982 8. Long CL, SchatTel N, Geiger JW, et al: Metabolic response to injury and illness: Estimation of energy and protein needs from indirect calorimetry and nitrogen balance. J Par Ent Nutr 3:452456,1979 9. Bengtsson C, Hulttn B, Larsson B, et al: Nya Ilngdvikttabeller fiir modelalders och aldre man och kvinnor. Llkartidningen 78:3152-3154, 1981 IO. BennegHrd K, Eden E, Ekman L, et al: Metabolic balance across the leg in weight-losing cancer patients compared to depleted patients without cancer. Cancer Res 42:4293-4299. 1982
256
11. Emery PW, Rennie MJ: Elimination of interference of histidine by use of formaldehyde in the measurement of 3-methylhistidine: Application to study of muscle protein degradation. Anal Biochem 126:67-73, 1982 12. Varley H: Practical Clinical Biochemistry (4th Ed). London, Heinemann, 1967 13. Euler von US, Lishajro F: The estimation of catecholamines in urine. Acta Physiol Stand 45: 122-l 32, 1959 14. Ward MWN, Owens CWI, Rennie MJ: Nitrogen estimation in biological samples by use of chemiluminescence. Clin Chem 26:1336-1339, 1980 15. Caldwell MD, Kennedy-Caldwell C: Normal nutritional requirements. Surg Clin North Am 61:489-507, 1981 16. Aist, Needham TM: The measurement of blood flow: Strain gauge plethysmography. Bio-Medical Engineering, July 1972, pp 266-269 17. Harris JA, Benedict FG: Biometric studies of basal metabolism in man. Carnegie Institute of Washington, pub1 279, 1919 18. Woolf CM: Statistics for Biologists. Principles of Biometry. New York, D. van Nostrand Co, Inc, 1968 19. Askanazi J, Carpentier YA, Elwyn DH, et al: Influence of total parenteral nutrition on fuel utilization in injury and sepsis. Ann Surg 191:40-46, 1980 20. Clowes GHA, Randall HT, Cha CJ: Amino acid and energy metabolism in septic and traumatized patients. J Parent Ent Nut 4: 195-205, 1980 2 1. Niehluser M, Bergstrom J, Chao L, et al: Urinary excretion of 3-methylhistidine as an index of muscle protein catabolism in post-operative trauma: The effect of parenteral nutrition. Metabolism 29:1206-l 213, 1980 22. O’Keefe SJD, Sender PM, James WPT: Catabolic loss of body nitrogen in response to surgery. Lancer 2:1035, 1974 23. Crane CW, Picou D, Smith R, et al: Protein turnover in patients before and after elective orthopaedic operations. Br J Surg 64:129-133,1977 24. Kien CL, Young VR, Rohrbaugh DK, et al: Whole body
RENNIE ET AL
protein synthesis and breakdown rates in children before and after reconstructive surgery of the skin. Metabolism 27:27-34, 1978 25. Clague MB: Turnover in pathological states (Chap. 41) in Waterlow JC and Stephen JML (eds). Nitrogen Metabolism in Man. London, Applied Science Publishers, 1981 26. Kien CL, Young VR, Rohrbaugh DK, et al: Increased rates of whole body protein synthesis and breakdown in children recovering from burns. Ann Surg 187:383-391, 1978 27. Long CL, Jeevanandam M, Kim BM, et al: Whole body protein synthesis and catabolism in septic man. Am J Clin Nutr 30:1340-1352, 1977 28. Birkhahn RH, Long CL, Fitkin D, et al: Effects of major skeletal trauma on whole body protein turnover in man measured by L-(1“‘C) leucine. Surgery 88:294-299, 1980 29. Rennie MJ, Edwards RHT, Halliday D, et al: Muscle protein synthesis measured by stable isotope technique in man: The effects of feeding and fasting. Clin Sci 63:519-523, 1982 30. Marliss EB, Murray FT. Nakbooda AF: The metabolic response to hypocaloric diets in obese man. J Clin Invest 62:468479,197s 31. Elia M, Carter A, Bacon C, et al: Clinical usefulness of urinary 3-methylhistidine excretion in indicating muscle protein breakdown. Br Med J 282:351-354,1981 32. Clowes GHA, Heideman M, Lindberg B, et al: Effect of parenteral alimentation on amino acid metabolism in septic patients, Surgery 88:531-543, 1980 33. Abumrad NN, Rabin D, Wise KL, et al: The disposal of an intravenously administered amino acid load across the human forearm. Metabolism 31:463-470, 1982 34. Wassner SJ, and Li JB: N’-methylhistidine release: Contributions of the rat skeletal muscle, G-I tract and skin. Am J Physiol 243:E293-E297, 1982 35. Bennegard K, Lindmark L, Ed&n E, et al: Flux of amino acids across the leg in weight-losing cancer patients. Cancer Res (in press)