4'

W

t

 

mar

 

THE"’

 

This is to certify that the

thesis entitled

MECHANISM OF SKELETAL MUSCLE CARBOHYDRATE
METABOLISM DURING ESCHERICHIA COLI

ENDOTOXIN SHOCK IN THE DOG
presented by

Richard Michael Raymond

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Physiology

 

 

/§§/1/MW4{/04.

Major professor

7/27/79

0-7 639

    
   

OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

Return to book drop to remove
this checkout from your record.

 

 

MECHANISM OE SKELETAL MUSCLE CARBOHYDRATE
METABOLISM DURING ESCHERICHIA COLI
ENDOTOXIN SHOCK IN THE DOG
BY

Richard Michael Raymond

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHYSIOLOGY
Department of Physiology

1979

ABSTRACT
MECHANISM OF SKELETAL MUSCLE CARBOHYDRATE

METABOLISM DURING ESCHERICHIA COLI ENDOTOXIN SHOCK
IN THE DOG '

BY

Richard M. Raymond

Carbohydrate metabolism by skeletal muscle was studied
in spontaneously respiring dogs following intravenous
administration of 2 mg/kg (LDlgg) Escherichia coli
endotoxin.

Close intra—arterial infusion of Endotoxin stimulated
skeletal muscle (gracilis muscle) to increase glucose
uptake, which was independent of shock alterations in
hemodynamics, blood gases, pH, insulin and oxygen uptake by
the gracilis muscle. These results indicate that endotoxin
has an insulin-like effect by increasing glucose transport
into skeletal muscle when infused locally. When Shock was
induced by the intravenous administration of Endotoxin
(2 mg/kg, Lqug) glucose uptake increased throughout the
six-hour experiment when gracilis muscle blood flow was
allowed to flow freely. The increase in skeletal muscle
glucose uptake occurred in the presence of muscle ischemia
and muscle hypoxia. However, during the constantly perfused
gracilis muscle, endotoxin shock failed to alter glucose
uptake by the skeletal muscle. Unlike the naturally
perfused acilis muscle preparation, constant flow to the

gracilis muscle prevented changes in blood flow or tissue

Richard M. Raymond
oxygenation. Gracilis muscle glucose uptake increased
significantly during the six-hour experiment when blood flow
to the muscle was reduced causing local gracilis muscle
ischemia. The increase in glucose uptake was independent of
systemic metabolic or hemodynamic alterations and occurred
in the absence of endotoxemia. The results gave evidence
that glucose uptake by skeletal muscle was mediated through
local tissue ischemia and/or hypoxia. To differentiate
between ischemia and hyposia in regulating skeletal muscle
glucose metabolism, two sets of experiments were conducted:
When gracilis muscle blood flow was held constant and made
hypoxemic via an extracorporeal heart-lung preparation,
glucose uptake by the gracilis muscle increased and remained
elevated above control for the duration of the six-hour
experiment; however, when gracilis muscle blood flow was
decreased to very low levels concommitant with hyperoxia,
no increase in glucose uptake was noted -- in fact, glucose
uptake decreased. These data, therefore, provide evidence
that endotoxin has an insulin-like action in promoting
glucose uptake when given locally, but failed to Change
glucose uptake by the constantly perfused gracilis muscle
when given intravenously. Under natural flow conditions,
endotoxin shock induction produced muscle ischemia and
hypoxia and caused an increase in skeletal muscle glucose
uptake, which was mediated by muscle hypoxia and not

ischemia.

DEDICATI ON
To my wife, Barbara, for her relentless patience
during the past five years and also for her love and caring
in giving me enough time by myself so I could finish this
degree. To my Children, Ricky and Amy for absorbing the

hard times with smiles and open arms and love.

ii

ACKNOWLEDGEMENTS

The author gratefully acknowledges the research
guidence from his main professor, Dr. Thomas E.Emerson, Jr.,
and also for his unending support when times were tough.
Dr. Emerson's trust, support and help were truly beyond what
was expected of him and for that, the author has gained
much, not only in prefessional development but also in
personal gratification and awareness. To Dr. Jerry B. Scott
for his help and support. To Dr. Lester Walterink who was
always there to listen and advise. To Amylou Davis who
helped guide the author through many difficult times and to
the skillful technical assistance of Ms. Marilynn A.

DeBussey for her help and patience.

iii

TABLE

LIST OF TABLES. . . . . . .

LIST OF FIGURES . . . . . .

INTRODUCTION. . . . . . . .

SURVEY OF THE LITERATURE. .

METHODS

Biochemical Composition of gram
microbial cell wall and the
relationship of chemical structure
to the pathophysiology of endotexemia

or shock . .

Hemodynamic alterations during endotoxin

shock. . . .

OF CONTENTS

negative

Carbohydrate Metabolism during shock.

Mechanism(s) of shock induced Hypoglyemia

Surgical preparation of the gracilis muscle
during either natural or constant flow

Blood gases and chemical analyses .

Experimental Protocol

Series I:

constant flow conditions

Series II:

Direct effects of E.
endotoxin on gracilis muscle hemo—
dynamics and metabolism during

coli

Hemodynamics and metabolic
alterations of the gracilis muscle
under natural flow conditions during
the intravenous administration of

2 mg/kg (LDIUO) E. Coli endotoxin. . .

iv

Page
vi

viii

10
19
28
36

36
39
40

40

41

Eége

Series III: Effects of constant gracilis
muscle blood flow on muscle hemo-
dynamics and metabolism during the
intravenous administration of

2 mg/kg (LDlBD

) E.Coli endotoxin . . .

Series IV: The effect of partial local
ischemia on muscle hemodynamics and

metabolism . . . . . . .

Series V: Effects of local muscle hypoxia,
idependent of ischemia, on muscle
metabolism and hemodynamics. . . . . .

Series VI: Effects of local gracilis
muscle ischemia independent of hypoxia

on muscle metabolism and
STATISTICAL ANALYSES. . . . . . . . . .
RESULTS . . . . . . . . . . . . . . . .
DISCUSSION. . . . . . . . . . . . . . .
SUMMARY AND CONCLUSIONS . . . . . . . .

BIBLIOGRAPHY. O O O O O O O O O O O O O

hemodynamics.

42

42

43

47
47

48
99
110
112

TABLE

LIST OF TABLES

Hemodynamic measurements of gracilis muscle
under constant flow conditions during
close intra-arterial infusion of
E. coli endotoxin. . . . . . . . . . .

Metabolic alterations of gracilis muscle
under constant flow conditions during
close intra-arterial infusion of
E. coli endotoxin. . . . . . . . . . .

Blood gases, Ph and O uptake measurements
of gracilis muscIe under constant flow
conditions during Close intra—arterial
infusion of E. coli endotoxin. . . . .

Systemic arterial and gracilis venous
blood gases, pH and hematocrit under

natural flow conditions during the I.V.

administration of 2 mg/kg (LD

)
E. coli endotoxin. . . . . . in?

Alterations in systemic arterial and
gracilis venous oxygen content and
gracilis venous glucose concentration
during the I.V. administration of
2 mg/kg (LDlafl) E. coli endotoxin. . .

Blood gases, pH and hematatocrit changes of
the gracilis muscle under constant
flow conditions during the I.V.
administration of 2 mg/kg (LDIOO)
E. coli endotoxin. . . . . . . . . . .

Oxygen content and venous glucose
measurements of the gracilis muscle
under constant flow conditions during
I.V. administration of 2 mg/kg (LDlgg)
E. coli endotoxin. . . . . . . . . . .

vi

Page

53

54

55

61

62

71

72

8.

10.

ll.

12.

13.

Page

Systemic arterial and gracilis venous blood
gas, pH and hematocrit changes during
local gracilis muscle ischemia,
independent of shock . . . . . . . . . - 79

Systemic arterial and gracilis venous
oxygen and gracilis venous glucose
concentration during local gracilis
muscle ischemia, independent of shock. 80

Gracilis arterial and venous blood gas, pH
and hematocrit during local gracilis
muscle hypoxia, independent of
ischemia . . . . . . . . . . . . . . . 87

Systemic arterial and gracilis venous O2
contentand gracilis venous glucose

concentration during local gracilis muscle
hypoxia, independent of ischemia . . . . . . 88

Gracilis arterial and venous blood gases,
pH and hematocrit during local
gracilis muscle ischemia, independent
of hypoxia . . . . . . . . . . . . . . 96

Gracilis arterial and venous oxygen
content, gracilis venous glucose
concentration and muscle vascular
resistance during local gracilis
muscle ischemia, independent of
hypoxia. . . . . . . . . . . . . . . . 97

vii

FIGURE

1.

2.

10.

LIST OF FIGURES

Schematic representation of murein or
peptidoglycan. . . . . . . . . . . . .

Schematic representation of the gracilis
muscle perfusion system utilizing an
extracorporeal heart—lung preparation.

Direct effects of E. coli endotoxin on
gracilis muscle glucose uptake and
arterio-venous glucose difference. . .

Metabolic alterations of the gracilis
muscle under natural flow conditions
during the intravenous administration
of 2 mg/kg (LDIGO) E. coli endotoxin .

Hemodynamic alterations of the gracilis
muscle under natural flow conditions
during the intravenous administration
of 2 mg/kg (LDIOO) E. coli endotoxin .

Metabolic changes of the gracilis muscle
under constant flow conditions during
the intravenous administration of
2 mg/kg (LDlgg) E. coli endotoxin. . .

Hemodynamic changes of the gracilis muscle
under constant flow conditions during
the intravenous administration of
2 mg/kg (LDlUO) E. coli endotoxin. . .

Arterio-Venous glucose difference of the
gracilis muscle under constant flow
conditions during the intravenous
administration of 2 mg/kg (LD )

109
E. coli endotoxin. . . . . . . . .

Metabolic alterations of the gracilis
muscle during local muscle ischemia. .

Hemodynamic alterations of the gracilis
muscle during local muscle ischemia. .

viii

45

50

57

59

65

67

70

76

78

FIGURE Page

11. Changes in gracilis muscle glucose uptake,
blood gases and oxygen uptake during
local muscle hypoxemia under constant
flow conditions. . . . . . . . . . . . 84

12. Hemodynamics of the gracilis muscle and
systemic arterial blood pressure
during local muscle hypoxemia under
constant flow conditions . . . . . . . 86

13. Representative tracing of gracilis muscle
hemodynamics and matabolism during the
infusion of 1 mg/min acetylcholine and
during hypoxemia coupled with
acetylcholine infusion under constant
flow conditons . . . . . . . . . . . . 90

l4. Gracilis muscle glucose uptake, oxygen
uptake and muscle arterial and venous
PO during local gracilis muscle
ischemia, independent of hypoxia . . . 93

15. Systemic arterial blood pressure and
gracilis muscle hemodynamics during
local gracilis muscle ischemia,
independent of hypoxia . . . . . . . . 95

INTRODUCTION

Endotoxin shock is a serious clinical problem with
recent reports estimating between 70,090 and 130,0U0
fatalities a year in the United States (McCabe, et. al.,
1970; McCabe, et al., 1972). Historically, early emphases
were placed on metabolic alterations of shock where it was
reported that endotoxins, given as crude vaccines, deplete
animals of their carbohydrate reserves, resulting in early
hyperglycemia and eventual hypoglycemia (Menten and Manning,
1924). The mechanism(s) responsible for endotoxin-induced
hypoglycemia has received much attention. While the
literature dealing with specific organ utilization of
carbohydrates in shock is sparse, several organ systems have
been ruled out. Hinshaw, et a1. (1975) reported that
glucose uptake by the heart—lung preparation is not
increased and Raymond and Emerson (1978) have shown that
glucose uptake by the central nervous system is decreased.
Filkins and Buchanann (1977) have shown the hemidiaphragms
taken form shocked rats Show marked increases in glucose
uptake; however, hemidiaphragms from normal, non-shocked
rats did not exhibit any glucose uptake when indotoxin was
added directly to the in vitro preparation. Recently, Furr,

et. a1. (1977) reported that the isolated, innervated,

2
constantly—perfused forelimb preparation of the dog did not
show altered glucose metabolism following endotoxin shock
induction. Their data are, however, vague and incomplete as
it seemed evident that venous samples of blood, taken to
calculate glucose uptake, were drawn from a common reservoir
that contained both cephalic and brachial venous blood which
came from skin and skeletal muscle respectively. This
common pooled venous efflux from the forelimb was used to
determine skeletal muscle glucose metabolism during
endotoxin chock. Their experimental design and results do
not seem to correctly report on glucose metabolism of
skeletal muscle following endotoxin shock.

The hypothesis being tested in this study is that
skeletal muscle, which represents approximately 50-60% body
mass, accounts for a marked increase in glucose uptake
during E. coli endotoxin shock in the dog, and could give a
better understanding of the mechanism of hypoglycemia during

endotoxin shock.

SURVEY OF THE LITERATURE

Biochemical composition of gram negative microbial cell

 

walls and the relationship of chemical structure to the

 

pathophysiology of endotoxemia or shock.

 

In order to provide a reliable structural defense to
withstand large changes in osmotic force in the surrounding
environment, many bacteria possess a cell wall component
which encapsulates and protects the inner cell membrane.

Universally present among bacterial cell walls is a
large single macromolecule, murein or peptidoglycan. This
structure consists of two repeating units: 1) a linear
polysaccharide component composed of N-acetylglucosamine and
N—acetylmuramic acid which are chemically bound to each
other; and 2) a small polypeptide which is attached to
N—acetylmuramic acid and consists of l-alanine,
d-isoglutamine, 1-lysine and d-alanine. A schematic
representation is seen below.

In many organisms, such as Escherichia coli,
diaminopimelic acid replaces lysine in the polypeptide
Chain.

Attached to the cell wall outer covering of many
bacteria, including E. coli, are complex structures composed
of lipids, proteins and polysaccharides. These peptido-
1ipopolysaccharides form elaborate capsules that cover the

murein macromolecule, or cell wall, and serve as the

Figure 1: Schematic representation of murein or
peptidoglycan.

N-Acetylglucosamine N-Acetylmunmic add

 

D-lsoglutamine NH
HN—CO—CHz—Cl-Iz—
( Hz); CONH-
L-Lysinc

—-NH-— -—H
o

N” n-Alanine
—OC—$—C}i3

characteristic somatic O antigens, of which some are highly
toxic to animals.

Peptido-lipopolysaccharides or endotoxins of many gram
negative bacteria have received much interest in regards to
their chemical structure and toxicity. Much debate has
occurred over the years concerning which fraction of the
endotoxin molecule possesses the toxic properties exhibited
during infection.

Early studies were done in which endotoxins from a
variety of gram negative bacteria (Raistric and Topley,
1934) were fractionated to characterize its structure and to
identify the acitve toxic component. Endotoxins from
enteric bacilli were extracted utilizing techniques from a
number of investigators (Boivin and Mesrobeanu, 1935, 1936;
Boivin, 1938, 1949, 1946; Haas, 1937, 1938; Wagner-Jauregg
and Helmert, 1942; Boroff, 1949; Morgan, 1937; Morgan and
Partridge, 1940, 1941). Results from these early studies
indicate that endotoxins have a lipid component composed of
palmitic acid, oleic acid, alpha-glycerophosphoric acid and
many nitrogenous compounds not readily identified. Attached
to the lipid moiety is a protein complex which, according to
Boivin, could be dissociated from the lipid moiety by
appropriate biochemical fractionating techniques. The
separation was based on qualitative tests in which the

dissociated lipid complex did not show any protein

characteristic; however, the lipid residue did behave as a
protein. In addition to the lipid and protein components
comprising 9-12% and 17-29% of the total crude endotoxin
molecule respectively, there was also a polysaccharide
complex. The chemical compostion of the saccharides is
variable from one type bacteria to another, but the main
products from hydrolysis extractions were:
N-acetylglucosamine, d—galactose, 1-rhamnose and a variety
of small monosaccharides. The polysaccharide fraction of
the endotoxin molecule was found to comprise 55—60% of the
total weight.

Biochemical studies of other enteric bacilli, such as
the Flexner dysentery bacilli have revealed important
information concerning the nature of endotoxin toxicity
(Goeble, et al., 1945; Binkley, et al., 1945). As
previously mentioned, fractionating techniques gave similar
lipid, protein and polysaccharide complexes. Tal and Goeble
in 1950, using different analytical procedures, found that
the polysaccharide complex contained a small (about 1%)
organically bound phosphorous ion. Later, it was shown that
the protein moiety also contained a small amount of
phosphorous. Toxicity studies in mice showed that the
active or toxic portion of endotoxin was associated with the
presence of organic phosphorous in combination with either
protein or polysaccharide portions of the endotoxin
molecule. In 1954, Westphal and Luderity, using acid

hydrolysis, identified a phosphorous-containing lipid from

8

purified endotoxin, which they named lipid A. Lipid A was
shown to have approximately one-tenth the pyrogenic activity
as the parent endotoxin molecule. It was thought that the
explanation for the reduced activity of the isolated lipid A
molecule was that it was no longer cOntianed in the
1yophilic polysaccharide carrier which makes it water
soluble (Westphal, et a1,. 1958; Westphal, 1969). Westphal
and associates therefore concluded that the lipid A portion
of endotoxin was the active portion of the molecule;
however, other investigators disagreed with the findings of
Westphal, in that endotoxins with varying concentrations of
lipid did not correlate with increases or decreases in
biolobical acitivity (Ribi, et al., 1961, 1962). Conclusive
proof could have been presented only by producing an
entirely lipid-free endotoxin molecule with full biologic
activity; however, such preparations still do not exist.
Ribi and co-workers felt that the biological activity of
endotoxin was determined by its size, degree of agregation
or degradation. They found this result to be consistent
with the hypothesis that a macromolecular complex of
critical size is one of the major requirements for endotoxin
to elicit its characteristic effects in the mammalian animal
host (Haskins, et al., 1961; Ribi, et al., 1962).

Recent analyses of lipid A portion of endotoxins have
shown the major constituents to be a phosphomucolipid and
that the residual endotoxicity of the lipid A mixture is due

to the presence of a remnant from the parent endotoxin

9

(Nowotny, 1965). Nowotny also performed new, unique
detoxification studies of endotoxin and found that organic
chemical reactions which split O-acly linkages of
carbohydrates were able to diminish the lethality of
endotoxin preparations. These O-acly linkages were found to
exist throughout the entire lipopolysaccharide complex.
Recent evidence has confirmed that the lipopolysaccharide
portion of endotoxin in responsible for the
pathophysiological effects seen in mammals (Benjamin, et
al., 1974). It has also been shown that, during the
fractionating of endotoxin, both simple proteins and
lipopolysaccharides retained a portion of the lipid A
moiety. Although the lipopolysaccharide portion contained
only small amounts of lipid A, it exhibited high levels of
toxicity.

Complete fragmentation of endotoxin, using acetic
acid, produced fractions of conjugated protein, O-specific
side chains and a polysaccharide core (Wolber and Alaupovic,
1971). The conjugated protein consisted of intact lipid A
moieties which proved to be equal or greater in toxicity
than that of lipopolysaccharide.

The lipid portions of conjugated proteins or
lipopolysassharide resembled the lipid A configuration.
Mortality rates were only observed with whole endotoxin or
fragments that contained at least a portion of the lipid A
moiety. It has recently been demonstrated that lipid A,

isolated from E. coli bacteria and solubilzed by complexing

10
it with bovine serum albumin (BSA), is the component
responsible for pyrogenicity and toxicity (Galanos, et al.,
1972). In other experiments dealing with lipid A
preparatoins, it was found that the active sites of the
lipid A molecule were side branches Of ester-bound fatty
acids.

In summary, the early results which emphasized protein
as the active component were correct, but for the wrong
reason. The early experiments did not detect the lipid
component within the proteins, which has since been
identified as lipid A and is universally found throughout
the peptido-lipopolysaccharide molecule. The lipid A
moiety, which contains phosphorous, confirmed the results
presented earlier that phosphorous is the active component
responsible for the pyrogenicity of endotoxins, independent
of its presence within proteins or polysaccharides.
Detoxification studies of lipid A produced evidence that
implicated side branches of lipid A as the active component

which have been identified as ester-bound fatty acids.

Hemodynamic alterations during endotoxin shock

 

Shock induced by intravenous injection of Escherichia
coli endotoxin is a widely used model in experimental shock
work because it resembles clinical septic shock in a variety
of respects and is an easy and relatively reproducible way

of inducing hypotension (Gilbert, 1960).

11

When endotoxin is given intravenously to the dog,
there is an initial rapid decline in systemic arterial blood
pressure followed by a temporary recovery which is then
succeeded by a progressive hypotensive period. This early
fall in blood pressure must either be due to a decline in
total peripheral resistance or cardiac output. Weil,
et al., in 1956, found that endotoxin caused a decrease in
cardiac output which was accompanied by a slight elevation
in total peripheral resistance. Their results showed that a
decrease in cardiac output was the cause for the initial
decline in arterial blood pressure rather than peripheral
dilation being the cause.

These results were further corroborated with venous
return-constant flow experiments, that is, the total venous
return of the dog was allowed to fill an extracorporeal
reservoir from which it was returned at a constant rate to
the right atrium of the heart. (Weil et al., 1956, 1956a).
Under these conditions, left ventricular output was held
constant. The intravenous administration of endotoxin into
the dog during these venous return-constant flow
preparations failed to show any decrease in systemic
arterial blood pressure over the first few minutes. These
results demonstrated two important findings: 1) that
systemic arterial blood pressure did not decrease over the
initital endotoxin period, indicating no decrease in
peripheral resistance for if resistance vessels had dilated

under a constant cardiac output, arterial blood pressure

12

would also have decreased, and, 2) the results from these
observations also indicated that the heart was capable of
pumping the constant amount of blood returned to it, which
demonstrated that the decrease in cardiac output seen in the
intact dog was clearly not due to myOcardial failure or
weakness. These data, therefore, could only point to a
decrease in venous return, as witnessed by the rapidly
falling blood level in the reservoir, as the primary means
for the decrease in arterial blood pressure during endotoxin
shock in the intact dog. Throughout the years, many
investigators have also shown similar responses to endotoxin
administration in the dog. The overwhelming evidence shows
that the early decrease in arterial blood pressure is caused
by a fall in cardiac output subsequent to a decrease in
venous return (Chien, et al., 1966; Hinshaw and Nelson,
1962; Weil, et al., 1956). Similar results.were also
obtained following endotoxin administration in
unanesthetized dogs (Prinao, et al., 1971). These animals
demonstrated the classic decrease in mean arterial blood
pressure, accompnaied by diminished cardiac output,
increased total peripheral resistance, increased heart rate
and decreased stroke volume.

The mechanism by which venous return is decreased
inititally in the dog following endotoxin administration is
an interesting one. Gilbert, in 1960, preported that the
decrease in venous return was the result of marked hepatic

venoconstriction resulting in protal hypertension and venous

13

pooling of blood in the hepatosplanchnic region. Further
substantiation came from Meyer and Visscher in 1962 when
they reported on the hemodynamic responses of intestinal
vascular segments in the dog. Their results showed that
following E. coli endotoxin administration, pre-capillary
arterial resistance increased by 56% whereas it increased by
566% in the venous vascular segment and decreased by 46% in
the segment form small artery to venule (Pre/post capillary
resistance ratio). Chien, et al., in 1966, also presented
similar results following the administration of E. coli
endotoxin to dogs. Their results were broken down, however,
into two groups, one in which hemodynamic alterations were
studied over the first twenty minutes of shock (early
phase), and, second, where these alterations were studied
during the late phase of the shock period (greater than 26
minutes). Their data showed marked hepatic venoconstriction
which caused rises in portal and wedged hepatic venous
pressures. These increases in venous resistance were,
however, independent of the sympathetic nervous system for
they were also observed in totally sympathectomized dogs.

In the late phase of shock, there was a decrease in venous
resistance in the hepatosplanchnic areas as compared to the
early period of shock but was still high relative to
control. As shock progressed, intestinal venous resistance
increased significantly which was similar to the findings of
Meyer and Visscher in 1962. Other investigators have also

shown an increased hepatosplanchnic sequestration or pooling

14

of blood as the mechanism for decreased venous return in the
dog during endotoxin shock (Moreno, et al., 1962; Hinshaw,
et al., 1962; Lillehei, et al., 1964).

On the other hand, Blattberg and Levy in 1971 supplied
evidence that if portal venous pressUre was prevented from
rising utilizing the aid of a Servo-pump, endotoxin
administration to dogs still produced blood pooling. By
appropriately placed aortic ligatures, it was found that 15
minutes after endotoxin administration, approximately 8
ml/kg of blood was pooled above the celiac artery and 12
mg/kg was pooled below the superior mesynteric artery.

Total pooling was estimated to be 28 ml/kg. It was
concluded that more than half of the early pooling in
response to endotoxin took place outside of the
hepatosplanchnic area. Extrahepatic pooling of blood was
also demonstrated in evicerated dogs (Hinshaw, et. al.,
1958, Brockman, et al., 1967). i

In the early phase of endotoxin shock, the hemodynamic
responses in the dog are followed by a decrease in arterial
blood pressure subsequent to a fall in venous return caused
by hepatosplanchnic and extrahepatosplanchnic pooling of
blood. During the late phase of endotoxin shock, however,
hemodynamics in the dog are different from the early phase.

The initial hypotensive period following endotoxin
administration is uaually followed by a recovery of blood
pressure towards control values. This could be explained by

either a decrease in hepatovenous resistance which would

15
”free" some of the sequestered pooled blood or to increased
catecholamine release or both (Gilbert, 1966). Following
this recovery, blood pressure then declines with the animals
dying in shock. Wiggers, in 1956, reported that the final,
prolonged hypotensive period could nOt result from either
trapping of blood in the liver or result from the early
decline in arterial blood pressure for the initial fall in
blood pressure would have to be more prolonged to be a
factor of irreversibility. It has, however, been well
established that cardiac output is decreased during late
endotoxin shock (greater than 26 minutes) (Chien, et al.,
1966; Dedichen and Schenk, 1967; Martin, et al., 1965;
Freidberg, et al., 1944; Ebert, et al., 1955), but the
mechanisms for the decline in cardiac output is still
controversial. Late is shock, venous return is also
decreased as it was during the early period. This probably
resulted from pooling of blood in the gut and elsewhere, for
it has been shown that pooling of blood does occur in the
noneviscerated dog for at least thirty minutes and in the
eviscerated dog for one hour (Meyer and Visscher, 1962;
Chien, et al., 1966, Hinshaw, et al., 1958; Weil, 1956).
Venous return-constant flow experiments are very reliable in
determining the degree of blood pooling or sequestration as
evident by changes in reservoir blood volume. However,
these experiments are not usually maintained beyond one hour
because of deterioration in the preparation (Gilbert, 1966).

Other experimental evidence, such as low right and left

16
atrial pressures, increases in cardiac output, following an
infusion of fluid and small decremental changes in blood
volume producing large decreased in output have supplied
good evidence for a decrease in venous return in the late
phase of shock (Hagano, et al., 1974; Freedberg and
Altschule, 1945; Pennington, et al., 1973; Ebert, et al.,
1955; Lansing, 1963; Kuida, et al., 1958).

Decreases in venous return other than by blood pooling
or sequestration are possible. Blood volume changes in the
late phase of endotoxin shock could potentiate the effects
of vascular pooling and cause a further decline in venous
return. Early reports have stated that plasma volume
reduction must eventually occur during endotoxemia or shock
because of the occurrence of diarrhea and vomitting in
unanesthetized animals, including man. There are also
reports suggesting loss of fluid into the tissues, based on
histological studies where edema was evident in the

131-1abelled albumin

intestine along with the presence of I
(Pemer and Bernheim, 1942; Kraneberg and Sandritter, 1953;
Aust, et al., 1957; Aust 1959).

In the hepathosplanchnic bed of the dog, fluid
filtration has been shown to occur during endotoxin shock
resulting in a bloody diarrhea (Chien, et a1. 1964; Hinshaw,
et al., 1964). Chien, et al., in 1964 and 1966, have
reported evidence for fluid filtration in other vascular

beds, often based upon measurements of thoracic lymph flow,

increased hematocrit and decreases in plasma volume. When

17

hematocrit is inhibited form rising following splenectomy,
plasma volume actually increased during endotoxin shock in
the dog. In addition, the means by which plasma volume is
measured in endotoxin shock are questionable. These
techniques offer employ dye-dilution and it has been shown
by Horwitz, et al., in 1972, that dye-dilution techniques
are invalid because of slower, uneven mixing and
sequestration of blood during shock. In the dog, these
techniques for measuring blood volume may be of little
significance as opposed to primate studies, where the degree
of pooling or sequestration is limited.

If net fluid filtration in extra-hepatosplanchnic
areas does occur, the specific location of this efflux has
not yet been identified. On the contrary, net fluid
reabsorption has been reported throughout the entire shock
period (Hinshaw and Owens, 1971; Weidner, et al., 1971).
These reports were based on actual organ weights and
vascular resistances in skin and skeletal muscle. Weidner,
et al., in 1971, and Hinshaw and Owens, in 1971, reported
that decreases in canine forelimb weight were associated
with declines in vascular resistance. This suggests that
the weight loss was due to interstitial fluid reabsorption.
Similar changes in other species such as the cat and monkey
during endotoxin shock also supports this concept. Thus,
blood volume loss subsequent to transcapillary fluid loss
cannot explain the decrease in venous return in the dog

during the late phase of endotoxin shock.

18

A possibility which has to be considered as a probable
mechanism for the decrease in venous return in the late
phase of endotoxin shock is the involvement of the
myocardium. Most authors agree that the myocardium will
fail in shock, although reports vary as to the mechanism of
the dysfunction (Alican, et al., 1962; Boho, et al., 1973;
Brand and Lefer, 1966; Hinshaw, et al., 1974, 1971, 1972).
Hinshaw, et al., in 1971, 1972 and 1974 were unable to
demonstrate any cardiotoxic effects of endotoxin during the
first 4 hours of shock. Their preparation consisted of
isolated working hearts. Six hours after endotoxin shock
induction, consistent changes in myocardial failure were
reported. Their results, however, were not consistent with
the theory that enodtoxin produces direct myocardial failure
but rather its effects are indirect; failure arising from
prolonged hypotension and low flow. Contrary to their
findings, Brand and Lefer in 1966, reported on the discovery
of a myocardial depressant factor (MDF) which was obtained
from blood of shocked animals and pOssessed cardiodepressant
or negative inotropic actions (Lefer, 1976, 1973). They
later reported that the pancreas was the major site of MDF
production. The production was dependent on pancreatic
ischemia which caused tissue hypoxia. Pancreatic hypoxia is
reported to stimulate the release of lysosomal hydrolase,
particularly acid proteased (Glenn and Lefer, 1971; Herlihy
and Lefer, 1975). These enzymes were shown not only to

catalyse the proteolysis leading to MDF formation, but also

19

directly exert deleterious effects on the micorcirculation
and myocardium (Ferguson, et al., 1972, Glenn, et al.,
1972). In contrast, Hinshaw, et al., 1972a, reported that
although the heart failed after four to six hours of shock,
this did not result from the direct action of MDF on
myocardial cntractility. To demonstrate their hypothesis,
pancreotectomized animals were studied and found that the
presence of the pancreas bore no relationship to the degree
of myocardial failure in endotoxin shock (Hinshaw, et al.,
1974).

The role of the heart late in shock (greater than 4-6
hours) in regulating cardiac output and blood pressure is
clear, however, the mechanism(s) for dysfunction remians

controversial.

Carbohydrate Metabolism During Shock

 

Reports by Menten and Manning in 1924 and Zechmer and
Goodell in 1925 presented evidence concerning rabbits
injected with crude bacterial extracts from the
Enteritidis-Parathyphoid B group. Blood glucose
concentrations in these rabbits increased from approximately
136—166 mg/dl to 235-256 mg/dl over the first few hours
following injection of the crude bacterial extracts, then
progressibely decreased, with the rabbits dying with a blood
glucose condentration of 36—46 mg/dl.

One Characteristic hallmark of endotoxin of gram

negative septic shock is the appearance of hypoglycemia

20

(Beck, et al., 1976). Some reports have documented
hyperglycemia during endotoxin shock as well as during live
E. coli septicemia (Cryer, et al., 1971; Hinshaw, et al.,
1975). These differneces between hypoglycemia versus
hyperglycemia may reflect only different time courses of
shock.

Cryer, et al. in 1971, reported that during the

administration of live E. coli bacteria (1610

organisms/m1
given at 16 ml/kg body wt.) into unanesthetized baboons
(Papio Doguera) plasma glucose concentrations increased
significantly and remained elevated during the 6-hour
experimental protocol. Along with the measurements of
glucose, they also measured insulin, ll-hydroxycortis-
teroids, growth hormone and urinary catecholamines.

Glucose, ll-hydroxycortisteroids, growth hormone and
catecholamines were elevated, whereas insulin was decreased.
Based on their results, they concluded that the initial and
prolonged elevated glucose concentration was a result of
impaired insulin release from the pancreatic islet cells.
This is consistent with the fact that glucose appearance and
disappearance from plasma is modulated by insulin which not
only promotes peripheral glucose uptake but also depresses
hepatic glucose release (LeCocq, et al., 1964; Mack and
Egdahl, 1976; Madison, et al., 1959). Their findings
emphasized a major role for depressed insulin in causing
hyperglycemia. This was based on data showing a decrease in

plasma insulin concentration during E. coli septicemia in

21

the baboon and suggested that the observed hyperglycemia was
due, in part, to lack of insulin at the appropriate receptor
site. Their interpretation of the results may be
incomplete. They concluded that the increased catecholamine
levels may have played a role in the development of
hyperglycemia because it has been well-documented that
catecholamines (norepinpherine, epinepherine) released from
either sympathetic nerve terminals or the adrenal medulla or
both in shock have profound effects in stimulating
glycogenolysis from the liver (Filkins and Cornell, 1974;
McCallum and Berry, 1973; Zwadzk and Snyder, 1973; Groves,
et al., 1974; Blackwood, et al., 1973). Many investigators
have also shown that during endotoxin or septic shock,
plasma catecholamines are elevated (Sanford, et. al., 1966,
LeCocq, et al., 1964; Guetner, et al., 1969; Spink et al.,
1966; Nykiel and Glaviana, 1961; Rosenberg, et al., 1959,
Hinshaw, et al., 1964; Zweifach, et al., 1956; Thomas,
1956). Insulin depletion, together with the presence of
anti-insulin mediators, such as ll-hydroxycortisteroids,
catecholamines and growth hormone, need not be the only
answer to elevated plasma glucose concentrations associated
with endotoxin or septic shock in the baboon. Since Cryer,
et al., did not measure either the rate of appearance or
disappearance of glucose form the plasma, it connot be
assumed that glucose efflux from plasma is not increased
during low inculin cncentrations. It has been demonstrated

that glucose transport into skeletal muscle and cardiac

22

muscle can occur independent of insulin, provided the
tissues are hypoxic (Morgan, et al., 1958; Randle and Smith,
1958). Morgan, et al., in 1958, and Randle and Smith, in
the same year, demonstrated that cardiac or skeletal muscle
glucose uptake increased by 166% at any given insulin
concentration when the bathing fluid was made hypoxic. The
increase in glucose transport was dependent on the degree of
hypoxia. Cryer, et al., also showed that venous PO2 values
were greatly reduced, which would further support an
anaerobic membrane transport of glucose. It is possible
that the hyperglycemia could have resulted from glycogen
breakdown only, its rate of production of glucose being
greater than the rate of disappearance from plasma into
muscle via anaerobiosis--which is independent of plasma
insulin concentrations. In 1972, Cryer, et al., published
another report showing similar hyperglycemic responses to E.
coli septicemia in the baboon. This study was undertaken to
clarify the effects of hypoinsulinemic hyperglycemia
witnessed in the previous study. Their assumption, based on
the previous data, was that hypoinsulinemia was the cause
for hyperglycemia in the septic baboon. To eliminate
catecholamine inhibition of insulin release through
alpha—adrenergic stimulation of islet cells, baboons were
pre-treated with an alpha-adrenergic inhibitor,
phenotlanine. E. coli septicemia did not produce any
increase in glucose concentration or depression in the

plasma insulin. These results were interpreted to

23

substantiate previous data where decreases in insulin were
implicated as the major cause of hyperglycemia in septic
baboons. As in their previous data, these results only
demonstrated that insulin, allowed to increase in response
to the increased plasma glucose concentration, following E.
coli septic shock induction, responded predominantly to
facilitate transfer of glucose into insulin-dependent cells,
reducing blood glucose concentrations to control levels.
With the mass of information dealing with carbohydrate
metabolism in the shock state, it was generally accepted
that hyperglycemia is generally described in primates
(Cryer, et al., 1971, 1972) and hypoglycemia is emphasized
in canines after injection of live Escherichia coli
organisms or endotoxin (Berk, et al., 1976; Griffits, et
al., 1973; Groves, et al., 1974; Peyton, et al., 1974;
Raymond and Emerson, 1978). Hinshaw, et al., in 1975,
repeated Cryer's previous work in baboons and found that, if
the animals were monitored until death or recovery, blood
glucose concentrations and insulin concentrations were
similar over the first 4 hours of the study. As the time
course of shock progresses, 86% of the baboons died within
26 hours after E. coli septic shock induction (mean survival
time of 15 hours). There was also a progressive decline in
blood glucose concentration in the presence of a decreased
plasma insulin concentration. Hinshaw correlated death of
the animal to its blood glucose concentration, one annimal

dying with a blood glucose concentration of 6 mg/dl at 14

24

hours into the experiment. These data provided evidence
that hypoglycemia did occur in the baboon if the experiments
were followed long enough and the decrease in blood glucose
concentration occurred in the presence of a diminished
plasma insulin concentration. Hinshaw demonstrated clearly
the progressive debelopment of hypoglycemia in the baboon
administered lethal infusions of live E. coli bacteria. The
nature of hypoglycemia was that it was not merely a terminal
event reflecting obvious deterioration of major organ
systems, but rather it represented an early metabloic
deficiency in the animal, lasting many hours and becoming
progressibely worse with time. Further corroborative
indirect evidence for an increase efflux of glucose from
plasma was seen by the increased lactate concentrations in
the baboon studies, indicatng a shift from aerobic to
anaerobic metabolism in which glucose—derived pyruvate
produces lactate (O'Donnell, et al., 1974). Blackwood, et
al., in 1973, and Groves and associates, in 1974, reported
elevations in lactate 1 and 2 hours after endotoxin or live
E. coli organism admistration. O'Donnell, et al., in 1974,
and Rackwitz, et al., in the same year, have reported
significantly elevated lactate in septic shock, concimitant
with severe hypoglycemis.

Intravenous injections of live E. coli organisms into
dogs also resulted in hypoglycemia; glucose was reported to
decrease from 96 mg/dl to 76 mg/dl within 4 hours of

injection (Griffits, et al., 1973) and to decrease from 162

25

mg/dl to 62 mg/dl in 6 hours (Groves, et al., 1974). When

an LD injection of E. coli endotoxin was administered to

76
dogs, hyopglycemia occurred (Berk, et al., 1976). These
animals showed progressively developing systemic hypotension
and acidosis and, within three hours, dogs were dying in
convulsions with a plasma glucose concentration of 25 mg/dl.
Hinshaw, et al., in 1974, also demonstrated the occurrance
of hpyoglcemia in dogs given an LD7g concentraion of E. coli
endotoxin. In both studies by Berk, et al., and Hinshaw, et
al., there were animals occasionally developing initial,
transient hyperglycemia. Similar hypolycemia was also noted
in unanesthetized dogs given lethal doses of live E. coli
organisms (Archer, 1976). To summarize these events, it is
clear that following either endotoxin or live gram-negative
bacteria in a variety of animal species, the predominant
effect is the occurence of hypoglycemia occasionally
preseeded by a transient hyperglycemic period. The initial
hyperglycemia has been attributed to depletion of liver
glycogen subsequent to hypotensive—induced catecholamine
release. Hyperglycemia persists for longer periods of time
in the primate as compared to the dog, but beyond 6 hours,
progressive hypoglycemia ensues in the baboon, with animals
dying with very low plasma glucose concentrations.

In previous studies, Griffits, et al., in 1973,
demonstrated that hypoglycemia, hypertriglyceridemia and
hypoinsulinemia occur in dogs with E. coli bacteremia during

twelve hour esperiments. The hypoglycemia and

26

hypoinsulinemia suggest that the septic dogs had hepatic
dysfunction resulting in impaired gluconeogenesis. A key
amino acid precussor for gluconeogenesis is alanine. This
amino acid is deaminated to pyruvate and is subsequently
converted to glucose (Felig and Wahren, 1971). Impaired
hepatic utilization of alanine is thought to cause the
hyperalanemia which has been observed during lactic acidosis
and Shock (Marliss, et al., 1972; Marchidk, et al., 1976.
Decreased oxidation of lactate via pyruvate in the
tricarboxylic acid cycle is generally considered responsible
for the lactic acidosis of shock (Schumer, 1968). Recent
evidence from Wolfe, et al., 1977, has also shown an
increased lacaite in endotoxin shock in dogs together with
hypoglycemia. Their results provided indirect evidence that
endotoxin hypoglycemia developed because of an enhanced
peripheral uptake of glucose and a failure of the liver to
maintain an increased glucose production. Their results
also concluded that lactate became an important precussor
for gluconeogenesis and an important metabolic substrate.
Filkins, et al., in 1975, provided direct evidence for
impaired hepatic gluconeogenesis in endotoxin shock. The
estimates of gluconeogenesis in vivo were accomplished
utilizing alanine loading of insulin-suppressed rats. In
their model, alanine is used primarily by the liver to
produce glucose. However, under normal conditions, as the
hepatic glucose output increases, thereby transiently

increasing plasma glucose, insulin release will be augmented

27

while glycogen release is inhibited. These changes would
effectively blunt hepatic gluconeogenesis thereby augmenting
tissue use and storage. In order to estimate hepatic_
gluconeogenesis better, they used mannoheptulose to block
insulin secretion which would allow hyperglycemia to better
reflect the status of hepatic gluconeogenesis. Conversion

of 14 14

C-alanine to C-glucose was another technique used to
estimate hepatic gluconeogenesis. Their results provided
direct evidence that during endotoxin shock, gluconeogenesis
as measured either by the hyperglycemic response to alanine

14C-alanine conversion to 14

loading or by C—glucose, is
greatly depressed. Several other reports have also noted a
decrease in hepatic gluconeogenesis in a variety of animal
models in shock (Filkins and Cornell, 1974; Schumer, et
al., 1976; Mela, et al., 1971; Freinkel, et al., 1965;
LaNoue, et al., 1968).

A malfunctioning liver, however, need not be the sole
explanation of progressively developing hypoglycemia because
notable hypoglycemia has been reported in evicerated dogs
within two hours following endotoxin administration, whereas
control, evicerated dogs demonstrated relatively constant
levels of glucose (Peyton, et al., 1974). Experiments
carried out in evicerated rats by Russell, et al., in 1944,
revealed the occurrence of hypoglycemia within three hours

in contrast to control, non-hemorrhaged evicerated rats.

The liver and pancreas were removed in the studies and the

28

mechanism for the marked fall glucose was ascribed to
anaerotic metabolism occuring in the presence of hemorrhage
at rates exceeding those in control, evicerated,

non-hemorrhaged rats.

Mechanism(s) of Shock Induced Hypoglycemia

 

Lindsgaard-Hansen, et al. in 1972, considered
circulatory deterioration to be the primary determinant of
the degree of shock due to enodtoxin because of its adverse
effect on tissue metabolism. They ascribed a lesser
influence of early toxic effects of endotoxin on liver
metabolism. Glucose levels, however, were not significantly
altered in their studies. Impairment of hepatic oxidative
metabolism which could be accounted for on the basis of
circulatory-toxic interactions has been documented following
endotoxin injection in rats (Mela, et al., 1971).
Circulatory factors indirectly influencing the release of
insulin (via catecholamine release) from the pancreas were
suggested (Spath, et al., 1974; Lefer and Spath, 1974). The
low cardiac output stage of septic or endotoxin shock and
the extremely low plasma concentration of insulin may be
primarily due to the impoverished metabolic status of a
poorly perfused pancreas (Clowes, et al., 1974; O'Donnell,
et al., 1974).

Aside from circulatory deterioration which would cause
a decrease in pancreatic release of insulin, hypoglycemia in

shock may be ralated to a reduced rate of glucose production

29

because the tissue demand for blood glucose might then
outstrip the supply mechanism, eventually resulting in
lethally low concentrations of glucose in peripheral cells.
Berry, in 1971, concluded that endotoxin administration
depleted animals of their carbohydrate reserves by virtue of
an impaired ability to carry out glycogenesis and
gluconeogenesis. This statement is in agreement with other
reports, showing impaired hepatic metabolism (Filkins and
Cornell, 1974; Mela, et al., 1971). Williamson, et al., in
1976, provided direct evidence for a malfunctioning liver
(impaired gluconeogenesis) after endotoxin induction in
rats. Groves, et al., have recently documented impaired
hepatic gluconeogenesis in dogs administered live E. coli
organisms. Their work suggested the presence of a block in
gluconeogenesis between pyruvate and glucose, and they
ascribed death as primarily due to derranged liver
metabolsim. Blackwood and others, in 1973, found
significant depressions of both glucose and glycogen stores
in liver and muscle of dogs subjected to hemorrhage or
endotoxin shock.

Data from the literature on trauma studies in man
suggest an increased need for substrate. Evidence was
provided that during endotoxin shock, glucose efflux from
plasma occurs at a greater rate than would normally occur
under control conditions (Kinney, 1972). Sepsis has also
been shown to cause not only a greater turnover rate of

glucose but also to cause greater glucose oxidation. Energy

30

expenditure measurements were made in severely burned
patients with sepsis and it was found that the energy
expenditure exceeded the caloric equivalent of fever (Roe
and Kinney, 1965). Long, et al., in 1971, provided
corroborative evidence of Roe and Kinney by showing that
sepsis did not interfere with the body to utilize glucose,
but rather the rate of glucose oxidation was elevated.
Further studies dealing with increased energy expenditure in
shock (Hypermetabolic state) were described by Halmagyi, et
al., in 1974. Their results were broken down into two
categories: 1) high resting metabolic expenditure during
shock in normothermia, and, 2) a state with increased fever
with an elevated caloric requirement. Their results
provided evidence for increased glucose oxidation in shock
of normothermic patients and a further increase in glucose
utilization with increased fever. Their results were
consistent with the clinical Observations that large amounts
of glucose are needed intraveously to maintain blood glucose
concentrations constant during endotoxin shock (Hinshaw, et
al., 1974). Utilizing isolated mitochondrial preparations
it was also found that endotoxin or septic shock produced an
uncoupling of mitochondrial oxidative-phosphorylation. This
was accompanied by increased glucose oxidation, increased
heat production and decreased ATP production. The
prevailing results from the literature have documented an
increased disappearance of glucose from plasma in septic or

endotoxin shock in the presece of low plasma insulin

31

concentrations. The mechanism(s) for the disappearance of
glucose has been the subject of recent investigations.
Berry, in 1971, reported that endotoxin stimulated cellular
glycolysis in ”a fashion similar to that obtained with
insulin" and also pointed out that endotoxin caused an
increased glucose utilization by leucocytes and macrophages.
Experiments done recently to specifically document the
glucose utilization and role of blood in endotoxin shock
(Hinshaw, et al., 1977, 1977a) following the intravenous
injection of E. coli endotoxin or live E. coli organisms
into the dog, Show that white blood cell (WBC) concentra-
tions were significantly increased (from 18,666/mm3 to
29,666/mm3). The major fraction for the elevated WBC count
was due to elevated neurtophils and immature neutrophils.
Lymphocytes and monocytes were not Significantly altered
during endotoxin challenge. Hinshaw, et al., however, did
not demonstrate the mechanism for the leukocytosis, but it
is known that endotoxin administration promotes the entry of
new leukocytes from the bone marrow into the Circlation
(Herion, et al., 1965). Hinshaw's results also demonstrated
increased glucose utilization by WBC, particularly the
neutrophils. These calculations were based on the total
amount of glucose taken up by the WBC during leukocytosis.
When glucose uptake, occurring_during the first hour of
endotoxin administration, was divided by the average WBC
count during the same time sequence, it was found that the

quantity of glucose taken up per acitvated (endotoxin) WBC

32

was not different from non-activated WBC in vitro (11.7 x 16

’9 vs. 7.4 x 16 '9

mg of glucose/ WBC/66 min. respectively).
The increased glucose utilization by the WBC was related to
their numbers rather than enhanced metabolic activity of
individual WBC. The WBC types accounting for the total
increase in numbers seen in their study was shown to be the
mature and immature neutrophils, which have been reported to
be perticularly active in phagocytosing endotoxin (Balis, et
al., 1974) or live E. coli organisms (Postel, et al., 1975).
In relation to their findings, other reports have documented
beneficial effects of transfused white blood cells in
animals and patients in septic or endotoxin shock (Graw, et
al., 1972; Epstein, et al. 1974). Nemeth, et al., in 1972,
studied the metabolism of glucose in rats during
Nobel-Collip drum traumatic shock. Their interest in
studying glucose metabolism as a function of shodk was
stimulated by the discrepancy between hypoglycemia-starved
shocked rats as opposed to the hyperglycemia-fed shocked
rats. These early findings of the differnces in the
metabolic response to shock may also be the underlying
factor for the increased resistance of fed animals to shock
(Mraz, et al., 1959). During their Shock protocol, they
provided results in which hypoglycemia was induced during
Nobel-Collip drum traumatic shock in the rat. Because of
the technique used to traumatize the animals, only indirect
evidence was used to determine the means for the increased

rate of glucose efflux from plasma. Urinary loss was

33

excluded because of the low glucose concentration in plasma.
Plasma glucose values were below the renal threshold (Tm)
for glucose. Increased diffusion through damaged intestines
was also discarded because luminal glucose concentrations
were not significantly elevated. Measurements of free
glucose concentrations in the liver, skeletal muscle and
epididymal fat pads were also not significantly elevated
above control. Their final conclusion was that glucose was
being lost through anaerobic glycolysis. This was shown by
pretreating the rats with 2-deoxyglucose, which blocks
glycolysis. When this was done, rats traumatized by
Nobel-Collip drum tumbling were seen to prevent the
disappearance of the greater part of glucose. Nemeth, et
al., concluded that if the glucose disappearance had been by
some other route other than glycolysis, the fall in glucose
should have been similar in magnitude to that which occurred
without 2-deoxyglucose. Two mechanisms for the enhanced
glucose disappearance are possible: 1) insulin's effect,
and, 2) tissue hypoxia. Insulin's effects were discarded
because the same results occurred in alloxan diabetic rats.
Therefore, their conclusion was that tissue hypoxia caused
an elevated glucose transport into tissues. This was
substantiated by low arterial and vena caval P02, Engel, in
1946, and Russell, et al., in 1944, both observed a sharp
fall in blood glucose in evicerated rats subjected to
hemorrhagic shock in the absence of pancreatic insulin.

Their findings suggested that anaerobic metabolic pathways

34
were utilized in the rapid disappearance of carbohydrate in
the periphery during shock. This mechanism would explain
the extremely marked fall of blood glucose in evicerated
dogs receiving endotoxin in contrast to control,
non-evicerated dogs(Peyton et al., 1974). Drucker and
Dekiewiet, in 1964, measured glucose uptake of the isolated
diaphragm removed from rats subjected to hemorrhagic shock.
They observed an increased glucose uptake occurring in the
presence of a severe depletion of intracellural energy
primarily through anaerobic pathways. O'Donnell, et al., in
1974, reported that hindlimbs of animals maintained normal
glucose uptake during low flow septic shock state. Filkins
also has shown that hemidiaphragm muscles taken from shocked
rats exhibited elevations in glucose uptake. On the other
hand, a recent study by Furr, et al., in 1978, suggested
that glucose uptake is not affected in the isolated,
perfused dog forelimg preparation during endotoxin shock.
Their results were interpreted to mean that skeletal muscle
glucose uptake was not altered during endotoxin shock.
Methods for their experiment could produce significant
errors in that blood flow from the brachial vein, which
drains mainly muscle, and cepahlic vein flow, which drains
mainly skin, were pooled into a common reservoir before
glucose analysis was determined. Therefore, glucose
measurement came from blood of skin and skeletal muscle
combined. Because the effect of endotoxin shock on skin

glucose metabolism is unknown, combined with some error in

35

sampling techniques, it is possible that small, but
significant, changes in pure skeletal muscle glucose uptake
would be missed during endotoxin shock. It is also possible
that skin glucose uptake actually decreased during endotoxin
shock. If this happened, and increase in forelimb muscle
glucose could be offset, resulting in misleading data and
and erroneous interpretation.

A final possible mechanism for glucose uptake in shock
during the presence of extremely low insulin cncentrations
would be the actions of kinins in facilitating glucose
transport. Research by Nies, et al., 1968, documented the
release of kinins in monkeys administered endotoxin,
whereas, Haberland, in 1972, had demonstrated insulin-like
actions of kinins in glucose transport.

In summary, five mechanisms are possible for the
hypoglycemia of shock. These include the insulin-like
actions of endotoxin, the increased glucose utilization by
WBC, the effects of kinins on glucose transport, impaired
hepatic gluconeogenesis and the increased glucose

utilization by anaerobic pathways.

METHODS

Fifty two mongrel dogs of either sex, weighing 26‘: 2
kg, were anesthetized with sodium pentobarbital (36 mg/kg),
intubated and ventilated spontaneously. Following the
surgical preparation, animals were anticoagulated with
16,666 units of sodium heparin. During the experimental
protocol, supplemental administration of sodium heparin (566
units) and sodium pentobarbital (averaging 56 mg) were given

hourly.

Surgical preparation of the gracilis muscle during either

natural or constant flow

 

A longitudinal skin incision, using a thermal cautery,
was made over the right gracilis muscle. The incision was
made from a point near the origin of the gracilis muscle to
approximately 2-3 cm below its insertion. Skin covering the
muscle and neighboring femoral artery and vein were removed
by cauterization. This was done to insure accessible
cannulations. The gracilis muscle was isolated from all
connective tissue and surrounding muscles. The obturator

nerve was carefully isolated from its conective tissue near

37

the gracilis muscle. Care was taken not to injure the nerve
during isolation. The entire length of the gracilis artery
and vein were completely isolated from conective tissue and
all side branches from these vessels were ligated. This was
done to insure that the gracilis muscle received no
collateral perfusion from non-gracilis arterial muscle
sources and also to insure that gracilis venous outflow was
free from venous blood not derived from the gracilis muscle.
The femoral artery and vein were isolated approximately 3-5
cm proximal and distal to the insertion of the gracilis
artery and vein. All side branches of the femoral vessels
were ligated to prevent collateral perfusion by non-gracilis
sources. To further decrease the possibility of collateral
perfusion of the gracilis muscle, heavy ligatures were
placed around the tendons at the insertion and origin of the
muscle. The left femoral artery was cannulated with
polyethylene tubing (PE 326) and the other end of the
cannula was inserted retrogradely into the right remoral I
artery at a point approximately 2 cm below the level of the
gracilis artery. The tip of the cannula in the right
femoral artery was positioned so that it was below the level
of the gracilis artery. A sigmamotor pump was interposed in
the perfusion circuit and was used for constant flow
perfusion of the gracilis muscle. The femoral artery, above
the gracilis artery, was cannulated with polyethylene tubing
(PE 246), and advanced to a level of the abdominal aorta.

This cannula was used to measure arterial blood pressure.

38
Interposed in the polyethylene tubing perfusion the right
femoral artery was a piece of rubber tubing. This rubber
tubing was placed close to the right frmoral artery. A
needle, attached to polyethylene tubing (PE 96) was inserted
into the rubber tubing to measure gracilis muscle perfusion
pressure. Perfusion pressure was set at a level
approximating arterial blood pressure. The right femoral
vein was connulated with polyethylene tubing (PE 326) at a
level 2-3 cm below the gracilis vein. Above the gracilis
vein, the femoral vein was ligated to prevent any backflow
from the systemic circulation. Gracilis venous outflow
drained by gravity into a reservoir that was primed with
approximately 266-366 ml physiological saline and/or
dextran, and was returned continually to the cannulated (PE
326) left femoral vein via a variable speed Holter blood
pump.

Completeness of vascular isolation was confirmed by
stopping the perfusion pump and observing the perfusion
pressure and gracilis venous outflow. A drop in muscle
perfusion pressure to 25 mm Hg or below during the stopping
of the perfusion pump was considered acceptable provided
venous outflow was zero. Muscle blood flow was determined
by timed collection of venous outflow with a graduated
cylinder and stopwatch and was expressed per 166 grams by
removing and weighing the muscle at the end of the
experiment. Muscle drying was prevented by bathing the

muscle with warm mineral oil and covering it with Saran

39

Wrap. Muscle temperature was maintained at 380 C with a
heat lamp. A Jugular vein cathether (PE 246) was used for
supplemental drug administration and endotoxin infusion.

All pressure recordings were made from appropriate
pressure transducers (Statham Laboratories, Model P23Gb)
connected to a Sanborn polygraph (Sanborn Co.,

Mode 166-1366).

During natural flow experiments, only the gracilis
vein was prepared and the arterial inflow remained intact.
The left femoral artery was used to measure systemic
arterial blood pressure. All other procedures listed above

were similar during the surgical preparation of the muscle.

BlOod gases and chemical analyses

 

During the appropriate experimental protocol, blood
samples were drawn anaerobically from the femoral artery and
gracilis vein. Blood gases and substances were analyzed as

follows: 1) PO PCO and pH with an Astrup Radiometer

2' 2

blood gas analyzer; 2) 02 content with a Lex-Oz-con oxygen
analyzer; 3) glucose concentration with a Yellow-Springs
glucose analyzer; 4) insulin by radioimmunoassay;

5) hematocrit by microcentrifugation; 6) muscle surface
temperatube with a celsius thermometer. Oxygen and glucose
uptake by the gracilis muscle were calculated as the product
of the appropriate arteriovenous difference muluiplied by

muscle blood flow. Resistance was calculated by dividing

perfusion pressure by muscle blood flow.

40

Experimental Protocol

 

Series I. Direct effects of E. coli endotoxin on gracilis

 

muscle hemodynamics and metabolism during constant flow

 

conditions.

 

To test the direct effects of E. coli endotoxin on
gracilis muscle glucose uptake and hemodynamics, the
inervated gracilis muscle, perfused under constant flow, was
used. E. coli endotoxin, prepared in physiological saline,
was infused at 6.382 mg/min at a volume flow rate of 6.382
ml/min via a needle inserted into the gracilis artery inflow
tubing behind the blood pump so as not to alter the constant
flow perfusion rate of the muscle. The saline vehicle
(control) was infused at an identical volume rate. The
amount of endotoxin infused approximated a gracilis artery
blood concentraion of endotoxin which occurs when 2 mg/kg E.
coli endotoxin is given intravenously as a-bolus injection.
The estimate was based upon an average blood volume of 8.6%
body weight, and average body weight of 26 kg, and an
average gracilis blood flow rate of 15 ml/min/166 grams.

The previously described hemodynamic and metabolic
parameters were determined before, during and after a five
to ten minute infusion of saline or purified E. coli
endotoxin.

The gracilis venous blood was diverted from the
reservoir and discarded during intra-arterial endotoxin
infusion in order to avoid any endotoxemia or shock in the

animal; however, uncontaminated blood from the venous

41
reservoir was returned continually to the dog to prevent any
systemic volume depletion. The initial blood volume in the
venous reservoir was approximately 266-366 ml and the amount
of blood returned to the animal during intra-arterial
endotoxin infusion never exceeded 166 ml. The reservoir was
primed with either normal physiological saline, dextran
(M.W. 46,666) or cross-matched whole blood from a donor
animal. No difference in any variables were noted when

saline, dextran or whole blood were used.

Series II. Hemodynamic and metabolic alterations of the

gracilis mescle under natural flow conditions, during the

 

 

intravenous administration of 2 mg/kg(LDlag) E. coli

endotoxin.

 

Experiments were conducted to determine the effects of
E. coli endotoxin shock on gracilis muscle hemodynamics and
metabolism. In this series of experiments, the innervated
naturally perfused gracilis muscle preparation was used.
Following a ten to fifteen minute stabilization period,
control blood samples and hemodynamic measurements were
taken and 2 mg/kg E. coli endotoxin, suspended in 26 c.c.
physiological saline, was infused intravenously. Endotoxin
was infused over a 5-minute period. All hemodynamic

measurements and chemical analyses, except insulin, as

42
described previously, were made during consecutive thirty
minute intervals for a total of six hours, or until death of

the animal.

Series III. Effects of constant gracilis muscle blood flow

on muscle hemodynamics and metabolism durng the intravenous

 

administration of 2/ mg/kg (LDlgg) E. coli endotoxin.

This series of experiments utilized the constant
flow-gracilis muscle preparation. Muscle blood flow was
maintained constant during endotoxin shock. Excluding
insulin, all chemical, hemodynamic and statistical
measurements were similar to those previously described in

Series II.

Series IV. The effect of partial local ischemia a muscle

hemodynamics and metobolism.

 

The innervated, constant flow gracilis muscle
preparation was used in this series of experiments. Similar
measurements were conducted as described in Series 11.,
except endotoxin was not administered, rather, muscle blood
flow was decreased to a similar level as that seen during
natural flow shock (Series II). Experiments lasted 6 hours
and similar hemodynamic and metabolic measurements were made

as in Series II.

43

Series V. Effects of local muscle hypoxia, independent of

 

ischemia, on muscle metabolism and hemodynamics.

 

The isolated, innervated, constantly perfused gracilis
muscle preparation was used as previously described except
that an extracorporeal heart-lung preparation was interposed
between the left femoral artery and right gracilis artery
(Figure 2).

A small donor dog (8—16 kg) was used for the
extracorporeal heart-lung preparation. The heart and lung
were removed from the donor dog by a left thoracotomy made
in the fourth intercostal space. Two lobes from the left
side of the lung were isolated and the remaining lobes on
the left side and all the lobes on the right side were
ligated and sectioned free from the main pulmonary tree.
The trachea was isolated and cut approximately 5-7 cm above
its insertion into the lung. The vena cava, aorta,
esophagus and all connective tissue and vessels were
sectioned and the heart and remaining lung were removed.
The heart was transversely sectioned at a position
approximately half-way between the base and apex. The left
atrium was cannulated with large bore tygon tubing having a
plastic fitting over the end of the tygon tubing. The
plastic fitting was flared so that its placement into the
left atrium could be secured by surrounding the atrium with

a tight ligature.

44

.cofiumammoam wcsfi-phmog Hwoaoqaoo
msuxo cm wcflufiafiu: Eoumxm coflmsmsoa OHUMsa
maawompm ecu mo :ofiumHCOmoaqos oaumEozom

”N Chamflm

4S

 

"v
I
f
I

 

 

 

1“”

l

“2%),

Min

 

 

 

 

R (SHE ex

46

The cannula from the left femoral artery (PE 326),
attached to a sigmamotor blood pump, was inserted into the
main pulmonary artery. Flow was adjusted so that
approximately 25—36 ml/min passed through the lung. As flow
from the lung filled the left atrial Cannula, it traversed
along polyethylene tubing (PE 326) to a plastic "Y”
connection. One section of the ”Y" connector was attached,
by polyethylene (PE 326) and rubber tubing to another
sigmamotor pump which was used to perfuse the right gracilis
artery via the right femoral artery as described earlier.
The other portion of the ”Y" connector drained excess blood
not perfusing the gracilis artery via the sigmamotor pump.
This blood drained into the reservoir which was also used
for the gracilis venous outflow. Arterial blood perfusing
the gracilis artery was sampled via the cannula draining
excess pulmonary blood. Pulmonary venous pressure was
measured via a cathether (PE 56) inserted into the large
bore tygon tubing. The cathether tip was advanced to a
level inside the left atrium, and pulmonary venous pressure
was set at approximately 6 - 5 mm Hg. This was done by
adding a variable resistance clamp which drained the excess
blood from the lung. During control measurements, the
extracorporeal heart-lung preparation was ventilated with
26% 02, 5% CO2 and 75% N2. Induction of hypoxia was
accomplished by vetilating the extracorporeal heart-lung

with 95% N2 abd 5% CO7. A typical schematic, representing

the surgical and perfusion set-up, is seen in Figure 2.

47

Following the induction of hypoxemia, hemodynamic and
chemical analyses, except insulin, were made as previously
described. All measurements were taken durng subsequent

thirty minute intervals for a total of six hours.

Series VI. Effects of local gracilis muscle ischemia

 

independent of hypoxia on muscle metabolism and

 

hemodynamics.

 

Previously described gracilis muscle and extracorporeal
heart-lung preparations (Series V) were used in this series
of experiments. Following a control period, the gracilis
muscle perfusing pump speed was decreased to induce muscle
ischemia. Concommitant with muscle ischemia, the extra-
corporeal heart—lung was ventilated with a gas mixture of
95% O2 and 5% CO2 (Hyperoxia). This allowed the arterial
blood perfusing the gracilis muscle, along with the gracilis
venous effluent, to remain at control, or higher P02,
thereby preventing ischemic—induced tissue hypoxia. Similar
chemical and hemodynamic measurements, as previously

described in Series V, were made during thirty minute

intervals for a total of six hours.

Statistical Analyses

 

Data were analyzed using one and two-way Analysis of
Variance. Means were compared with the Student-Newman-Keuls
test. A 'P' value less than or equal to 6.65 was considered

significant.

48

RESULTS

Series 1. Direct Effect of E. coli Endotoxin on Skeletal

 

Muscle Glusoce Uptake in the Dog.

 

The important findings of this study are illustrated
graphically in Figure 3. The upper portion of this figure
shows gracilis muscle glucose uptake, while the lower
portion indicates the arterio-venous glucose difference
under different conditions. Each value shown represents
steady state five-minute values, i.e., saline or endotoxin
was infused for a five-minute period and the values depicted
in Figure 4 represent the steady state value at the end of
the five-minute infusion period in each case. Glucose
uptake by the isolated gracilis muscle was 6.17 i 6.66
mg/min/166 grams tissue during the control period. Infusion
of saline was associated with no signicicant change in
glucose uptake. The post—control value represents post—
control value saline infusion and control for the endotoxin
infusion. This value was 6.26 i 6.64 mg/min 166 grams
tissue. During Close intra-arterial infusion of E. coli
endotoxin, glucose uptake by the gracilis muscle increased
significantly (P < 6.61) to 6.71 i 6.69 mg/min 166 grams
tissue, which represents an increase of 173% above control.
Glucose uptake returned to a level not different from
control immediately after stopping the endotoxin infusion (P

> 6.65). The bottom panel of this figure shows that the

Figure 3:

49

Direct effects of E. coli endotoxin on
gracilis muscle glucose uptake and
arterio-venous flucose difference.

Endo = endotoxin infhsion at .382 mg/min at
a volume flow rate of .382 ml/min.

Note: Saline post control is control for
endotoxin infusion.

11

D
II

3|-
II

p 0.01 compared to control and saline.

0.75
Gracilis
Muscle
Glucose
Uptake
(mg/min /100g)
0.0
6

Arterio-Venous
Glucose
Difference

(mg/(ll)

50

 

 

 

 

 

._l_ w I

 

 

 

 

 

Control Saline Post Endo Post
Control Control

51

arterio-venous glucose difference of the gracilis muscle did
not change significantly during saline infusion (P > 6.65).
There was a tendency for the arterio-venous difference to
increase; however, this was probably due to dilution of the
plasma glucose by the 6.382 ml/min infusion of saline. On
the other hand, endotoxin infusion was associated with an
increase in the arterio-venous glucose difference from 1.9
mg/dl to 5.2 mg/dl (P < 6.61). The arterio-venous
difference returned to a level not significantly different
from control immediately after stopping the endotoxin
infusion (P > 6.65). The increase in gracilis muscle
glucose uptake and arterio—venous difference during local
endotoxin infusion occurred with no significant change in
systemic arterial blood pressure, gracilis muscle perfusion
pressure, gracilis muscle vascular resistance, hematocrit,
muscle surface temperature, arterial glucose.concentration,
arterial or venous insulin concentration, gracilis muscle
arterial or venous blood gases, muscle oxygen uptake or pH
(P > 6.65). These values are summarized in Tables 1, 2 and
3.

Series II. Hemodynamic and Metabolic Alterations of the

Gracilis Muscle Under Natural Flow Conditions During

 

Intravenous Administration of 2 mg/kg (LDlgg) E. Coli

Endotoxin.

 

The intravenous administration of 2 mg/kg
(LDlgg) E. coli endotoxin produced both hemodynamic and

metabolic changes of the gracilis muscle under natural flow

52

conditions. The important findings are shown in Figures 5
and 6.‘ Endotoxin shock caused a marked increase in gracilis
muscle glucose uptake (P < 6.61), that was maintained above
control values, except at the fourth hour, where the mean
value was not different from control (P > 6.65.

Endotoxin administration caused the mean gluxose
uptake of the grecilis muscle to increase from a control
value of 6.687 mg/min/l66 grams to a value of 6.511
mg/min/l66 grams at sixty minutes post-endotoxin
administration (P < 6.61). This represented a 487% increase
as compared to control. Systemic arterial glucose
concentration decreased significantly 126 minutes following
shock induction and remained below control levels throughout
the remainder of the 6-hour experiment (P < 6.61). During
the initial ninety minute period of shock, however, arterial
glucose concentration did not differ from control (P 3 6.65)
although there was a tendency for the mean values to
increase. Gracilis venous glucose concentration, as seen in
Table 14 also decreased following endotoxin administration;
however, the decrease was seen during the first thirty
minute determination and remained below control values
throughout the remiander of the experiment ( < 6.61).

Gracilis muscle oxygen uptake decreased from a
control value of 6.448 ml/min/166 grams to a level of 6.226
m1/min/l66 grams 6 hours following endotoxin administration
(P < 6.61). At the sixth hour of shock, oxygen uptake

represented a 45% reduction as compared to control. The

53

 

 

 

 

.m3HMHmo moocmwc ca consumes ocsumcoasou oommcsm oHomsz u o

.:HE\HE Nmm. mo mums 30am m on :HE\ms Nmm. pm mum; :onsmcH chOpoccw u c

.CHE\HE Nwm. um mum; conzmcH ocflamm u o

.mucosacoaxo mo Lopez: mucomocnog m." n

.ucoswcoaxo some ocouon >Hamu cocmqocq mm: saxouocco “Hoe .m n m

S. H megm .t. H megm t. H Shem t. .+. Shem t. H Chem 0 “1838853 0823 38::

m; H $.91 m; H $.91 m; .+. $.91 m; H 8.9. m; H $.91 : 520826:
I I I I I RSEEEESEV

wo._ + =o.o ea. + oo.a m._ + 00.01 AA. + mm.o om. + oe.m FF ooemumamme Imasomm> maaaomtu

E .H ST E u mmé P; H Ami E H omé E .+. AT? 2 $022225 33... 8on 2:85

9m H S: SS M 9m: is H 8.8, 0.: H To: Em H 0.0m? : Bess 238.5 22828 33::

Em H cum: 0.: .+. is: Em M 9m: 9m H mg: m.m H mama 2 3:55 953?; 8on 323.2

Hocpcou Hocucou
umom ccflxououcm anon oocfiamm Hocpcou n:

 

 

m

:onuoncm HHoo .m mo :onsucH HmHLopL<ImchH

omoHo mcwcsa meoH»Hucoo 30am ucmumcou Lone: mHomsz mHHHomco mo mucosocsmmoz OHEmcmcosom .P moprmh

54

.consucfi CHxOuouco Lou Hocucoo mm com: um: HOchoo anon ocuamm "ouoz

cHE\Hs mmm. um casst mwm. am mpmc scamscca cHxOuoocm

Hocucoo new ocflamm op cocmasoo Fo.o v m

:HE\HE Nmm. um mum; :ofimsmcfl ocHHmm

I!
.0

II
CU

 

 

 

 

 

mm; H NS: R2 + 8;: S; + m3: E; + ems: em; + 8.? o 29:; 238.6

E...“ £5 E; H 5.9 me; H of: 11m M £4: SS H 3.3 a $23.2
Aas\=:v eaasmeH

85 H To .85 H :..o .66 H 8.0 85 H SMS 85 H 2.0 2 813a: 88:8

aim H smde .mm.m H CNS aim H 8.3. mad H Si. mod H 91:. Z 98...; 228.6

afim H 3.3. om.m.+. ems; 8.m H was; firm H 8.? mad H 3.2. : 623.2
:32: a

Hocucoo Hocucoo
pmom ncflxouoccm umom mocHHmm Hocucoo :

 

 

.cfixOuoccm waoo .w mo :onSCCH HmHLOBL<ImchH mmoHu
mcficsn mcofiuficcoo 30am ucmpmcoo Lone: Hoomsz mHHwomcu mo mcofiumcoaa< OHHonmuoz .m canoe

55

:onuouco com Hogucoo an own: mm: Hocacoo anon ocfiamm ”ouoz

cas\ae mam. an eas\ms mam. an user eoansaea saxopoecm

n

 

 

 

 

 

 

casxae mmm. um can; conauca meadow u m
so.o.u o~.o SS.H =m.o 20.0.“ =~.o mo.o.u m~.o oe.o.u om.o PF Amoop\easxasv No >
ma.o.n mz.m_ ee.o.u mm.mp mo.o.u sa.mF mo.o.n op.m1 _e.o.u me.m1 PF Aae\aev 5:65:00 No

.o..H o=.e 10..“ mm.» Po..n as.» Po..u P=.e Po. u. m:.e _F lasagna ma

em.mu mo.mm m=.~.u mm.mm ap.m.u mo.mm am._.n ow.sm mo.m.u ._.8m _1 szssv Nous

ww.a.u NF.:m mo.~ H se.mm oo._ H.3w.mm 24.1 H.om em mm._ H.wm.mm FF Ammeev mos
msoco> mHHHomLo

em.o.u Pm.o. mm. u.oe.mp mo.o.u mm.mF oe.o “.mm.m_ Fm.o.u mm.81 PP Aae\aesaemucoo mo

Po..u we.» 10..“ Pe.e Po. u. _=.e ..o H. mz.e P.o.u m=.e FF Ampacsv ma

mm.a.u mm.mm as. n.0m.mm oo.F.H 11.4m oe.F.u 81.3m mo.__u mm.em _P Ammssv Neon

me.m H.mm.oe ma.m.u me.me me.=.n mm.ow wo.eu.me.ae e..=.n ow.=w FF Amzasc mom
Hmficouc< ofiEoume

Hocucoo Hocucou
boom ncfixouoccm pmom mosaamm Hocpcou :

 

 

.eaxouoocm “Hon .m co eoamsceH Haste»

30am ucmpmcou Loos: oHomsz mHHHomLu mo mucoEossmmoz oxmua:

N

L<ImcucH omoHu mcficsa meoHuHocoo
0 new ma .mommo nooam .m manna

56

Figure 4: Metabolic alterations of the gracilis muscle
under natural flow conditions during the
intravenous administration of 2 mg/kg (Lolgg) E.
coli endotoxin.

P 02 = Gracilis venous P02 in mmHg

Number of animals surviving at the
indicated time intervals

:5
II

*p < 6.61

57

n=7 (6) (5) (4)
Endotoxin

0.6
Glucos ./ \
Uptakeeo'3 i * *

(mg/min/loogz,

105 *

i’}\§"2\ *
Arterial 70
Glucose
Gone. 35

(mg %) 0

05

02 Uptake

(ml 02 Min/1005

40

PV 02
(mm/u!)

 

58

Figure 5: Hemodynamic alteration of the gracilis muscle
under natural flow conditions during the
intravenous administration of 2 mg/kg (LD ) E.

166
coli endotoxin.

n = Number of animals surviving at the indicated

time intervals.

*p < 6.61

59

II =7 (6) (5) (4)

Endotoxin

120 0

Systemic \* , . * ,, a
Arterial * ,, ,, * ’ __... \ __ \* 4
Inc" ‘° kH—I’H i i i i H

0
6
Gracilis \3
Blood
“0“ * * * t t fi *
(ml/min/lllllg) 3r \§\ \§__§-—— ;‘§-§~§~;——.~§~§
o _
40

W I/i\i/§“"Li—eH/i’f A H
.m ' )

 

60

decrease in oxygen uptake was noted during the first thirty
minute calculation and remained below control levels
throughout the experriment (P < 6.61). Concommitant with
the decrease in muscle oxygen uptake was a decrease in
gracilis venous P02 (P < 6.61) throughout the 6—hour
protocol.

Systemic arterial blood pressure and gracilis muscle
blood flow were decreased initially thirty minutes following
endotoxin administration (P < 6.61) and remained at these
low levels throughout the 6-hour shock experiment (P <
6.61). Gracilis muscle vascular resistance increased during
the first thirty minute calculation and remained elevated
for the duration of the experiment (P < 6.61).

Systemic arterial and gracilis venous blood gases, 0?

‘

content, pH, hematocrit, muscle surface temperature and
gracilis venous glucose concentration are summarized in
Tables 4 and 5. Systemic arterial oxygen content increased
during the sixty minute time period and remained elevated
throughout the remainder of the experimetn (P < 6.61).

Systemic arterial PO increased only during the later

2
portion of the experiment, where it became significant at
the 366 minute time interval and remained elevated during
the last two time periods (336 and 366 minutes) (P < 6.61).
Gracilis venous oxygen content decreased initially thirty
minutes post-endotoxin infusion and remained depressed

throughout the remainder of the experiment (P < 6.61).

Together with the reduction in gracilis venous oxygen

61

namsficm mo amass: u c

 

 

 

 

.0 0mm aaoumsfixocaam um ucmpncoo nae: ocaumgoasou momucsm oaomsz .zmm can mononucmcma :H memnssz ”ouoz
Po.o v a u .
Hocncoo on Hmaao.m n F
a m o s u :
sz sz Aav Amy Amy Amy Amv Amy Amy Amy Amv Amv Amy
cam tam 3mm cam amm nmm 3mm tam mmm amm epm om , m: Auv go:
Amp.v APF.V Aoo.v Amo.v Awo.v Apo.v Aeo.v Aso.v Aoo.v Aoo.v Amo.v Amo.v Amo.v
smo.o azm.o .mo.> c>0.> amc.> amo.> smo.> :mo.> mmo.> two.» 6P.> aw.> mm.> Ampficzv In
Ampv AoFV Ampv AMPS Azpv Amps Aspv Amy Amy Any Rev Amy Ans m
mo 8 so am mm mm mm E mm mm mm me me 3:25 8a

 

msoco> mHHHomLu

Amo.v Amo.v ANP.V AP.V Awo.v A>0.V Aho.v Aoo.v Aoo.v Amo.v Amo.v Azo.v Amo.v

amp.> amp.> amp.» cup.» amp.» cop.w tap.» am~.> :z—.~ uhp.b s>—.> mop.» zm.> Amuficav mo

Amv on sz Amy APFV AFFV AFC Asv va Asv any on Any

:m .5 .mm :m S 3 mm mm mm om mm me me Ammeavmoua

ANFV Aav Ampv Amy Amv hwy Aev any Apv Amy sz on Amy

.2. 1:. S 8 ma 8 E S S me me E S Amzsecmoa
Hmucoucm OHEmume

 

 

com omm oom Ohm cam oFN omp om? ONP om 00 cm 0 A.:HEV OEHH

 

 

.eaxoaoucm aaou .m Aooaoav mx\me m co coaumtumaeaeea .>.H on» mcaesa meofiuaueoo
30am Hmczumz Love: nwcooumsom can an .mommu voon msocm> mHHHomcu new Hmwcouc< oHEoum>m

.2 magma

62

 

 

 

 

 

Hospcoo o» Hando.m u H
save".
a m o s u :
Abmv hmmv Apmv APNV “spy Awpv Ampv Aowv Ampv Azpv Amy Amrv APFV AaC\mEv
am: com com gem as: amm rpm coo :50 two emo awn mm cowumcucoocoo omoosau
A>.FV Am.mv AP.NV AP.NV A>.PV A>.Pv Ac.Fv Am.PV Am.Pv Am.Fv Am.Fv AF.—v Ao.v “HU\HEM
aw.m tm.o sm.m 60.: cm.= aw.m cm.o ao.m um.m uo.o up.o aw.o 0.0— acoucoo o
msoco> mwafiomso
A>.V Aw.v AP.FV A:.V Ao.v Ao.v Ao.v Ao.v Ae.v Aw.v A~.v A:.V Am.v Aao\asm
mN.ON cm.om to.mp ao.om mm.om mm.ON mN.ON no.0N mh.op n=.om m>.mp m.wP m.mp ucwucoo O

 

Hmwcoacm OHEoume

 

 

com 0mm com cum ozm oFN owp omp GNP co co cm 0 A.:wsv oEHH

 

.caxouoocm HHOQ .m Aoopaqv mx\ms N no cofiumcumficasc< .>.H on» mcaczo cofiumgpcoocou
omoosao msoco> naaaomcu new pcoucoo comaxo mzoco> mHHHomLo new Hmwcouc< oasoumzm cw macaumcoua< .m magma

63

content was an elevated venous PCO (P < 6.61). Systemic

2
arterial PC02 and pH, together with gracilis venous pH were
all significantly decreased following endotoxin
administration (P < 6.61). Gracilis venous hematocrit was

increased throughout the 6-hour experimental period. Muscle

surface temperature was maintained approximately at 380 C.

Series III. Hemodynamic and Metabolic Alterations of the

 

Gracilis Muscle Under Constant Flow Conditions

 

During Intravenous Administration of 2 mg/kg

Q1qu

The important results concerning changes in the

) E. Coli Endotoxin.

 

hemodynamics and metabolism of the gracilis muscle,
following shock induction, are shown graphically in Firuges
7 and 8. Glucose uptake by the graiclis muscle was not
significantly different from control following endotoxin
administration (P > 6.65). Glucose uptake during control
was 6.61 i 6.63 mg/min/166 grams and at 36 minutes post
endotoxin administration was 6.28 i 6.11 mg/min/166 grams.
This represented the largest mean difference in glucose
uptake from control. The arterio-venous (A-V) difference in
glucose across the gracilis muscle is seen in Figure 9.

This figure graphically illustrates the non-significant
Changes in A-V difference of glucose (P > 6.65) across the
muscle. Control A-V glucose difference was 1.58 i .48 mg/dl
and at 36 minutes post endotoxin was 3.58 i .94 mg/dl.

Because muscle blood flow was held constant, any change in

64

Figure 6: ‘Metabolic Changes of the gracilis muscle
under constant flow conditions during the

intravenous administration of 2 mg/kg (LDIOO)
E. coli endtooxin.

n = NUmber of animals surviving at the
indicated time intervals.

*p ).01

PVO2 = gracilis venous P02 in mmHg

65

 

0.4 "48) (7) (6) (4) (3)

Glucose

Up take \

(mg/minnow / f i
o I \i VI I\,/ \i\{/i/1
8° ./1\.

3335?: i\1\* . * fi * A * *
ggcigtration §\f_f_{_}—{——T\}_.}
O I.

40
02 Uptake __ __.. x \
("z'loz/mitl/l f‘§/§\§—f\i__§/i { A f {—f
O.

Pyflz
(M No)

 

Figure 7:

66

Hemodynamic changes of the gracilis muscle
under constant flow conditions during the
intravenous administration of 2 mg/kg (LD )

. . 100
E. C011 endotox1n.

n = Number of animals surviving at the indacted
time intervals .

*p 0.01

Note: Numbers in parenthesis represent the
number of animals surviving at the
given time interval.

67

n= (8) . (7) (6) (4) (3)

160
/§/;\i-—’f\* ‘ ,— ____ t\
Perfusion i/ i i f f\ \_'} I/§ i
Pressure
(mm Hg)
40 t * t * *
Resistance f/{/}‘}’1\{\{’f\
("mile/"1W \
i—e—i—i
O
120
Systemic
new. \ * * «~
. (mmHg) —§\§/§—§\§/;_ E\{\{\§~;
O
16
Gracilis
Blood /§\§—§_—§

I—I—i—I—I—I—I—I/i

Flow
(ml/min/ioog)

 

68

A-V difference of glucose across the muscle would reflect
changes in glucose uptake. Although this A-V glucose
difference was large at 36 minutes, it was not significantly
different from control (P > 6.65). Arterial glucose
concentration decreased and remained below control values
throughout the experimental period (P < 6.61). Oxygen
uptake and gracilis venous PO2 were not different from
control during the entire six hour experiment (P > 6.65).
Systemic arterial blood pressure decreased during the
first thirty minute measurement (P < 6.61) and remained
below control during the experiment (P < 6.61). Gracilis
blood flow, being held constant, was not significantly
different from control (P > 6.65). Because this experiment
caused the dogs to die at different time periods during the
6-hour observation, any mean values that were significantly
altered as a consequence of small numbers of experimental
animals, due to the lethality of endotoxin shock, were
compared back to their original control values using paired
statistics. The time periods in question in this study are
form 246 minutes through 366 minutes inclusive. Gracilis
muscle perfusion pressure increased iitially following
endotoxin administration (P < 6.61) then returned to a level
not different from control at 276 minutes (P > 6.65) and
remained at control values through the 366 minute protocol
(P > 6.65). Muscle vascular resistance increased from 36
minutes to 246 minutes (PC 6.61) then decreased at 276

minutes to control levels where it was not significantly

69

Figure 8: .Arterio-venous glucose difference of the

gracilis muscle under constant flow conditions
during the intravenous administration of
2 mg/kg (LDlOO) E. coli endotoxin.

70

 

 

 

 

I- 6
awn—yam. ovum

m 3.98

cocoa—G
Paw

.Amo.o A av Hocucoo sLou ucoeommau >HucmoHuwcmwm no: mew: ma new moon mo monocouuao naoco>I0aLopL< ”one:
.0 0mm uaoumsfixocaam um ucmumcoo cam: ocsumcoasou oommesm maomsz

SS vm ..

Hocucoo on szoo o P

 

 

 

 

 

71

 

 

 

m a o e m n :
Ans on on Aev Rev Amy Ame Amy Ame Amy Ame Amy Amy
smm :Nm umm mam mam cam tam nmm «mm u—m com :0: mm Anv go:
“mo.v Awo.v Aop.v Amp.v AmF.v Amo.v Aeo.v Amo.v Amo.v Aeo.v Amo.v Amo.v Amo.v
PF.» .31.» as.» m~.e Fm.e op.» me.» :_.e m_.e m_.e mm.e om.e Pm.e Ampa:=V ma
Ace Ans Ame Rev Amy ASS Ase Ame Ame Ass Amy Ame Ame m
.om mam .em .mm .mm .1: .8: .oz .0: am: so: .om mm szesv one
mzocm> mHH HomLO
Aoo.v AP.V 11.8 Amp.v ANF.V Aeo.v Amo.v 140.3 Amo.v Amo.c Amo.v Amo.v Aeo.v
s>r.b sop.> n>—.> ¢=N.> smN.> 609.5 uhw.> tow.» :Pm.> amN.» um.> SN.> mm.> Amuwcav ma
ASS Ame Awe Ass Rev Ame has any Ass Ass Ass Ame ASS m
Sam .cm .em .Pm .om as am: me as as am: 3: mm Ramses ooa
Aomv Aopv Apps Aspv Aopv Any Any Any any sz Amv Ame Rev m
.5 PS .8 .mm as 2. S S S we :. S E. 3:55 8
Hmacoucm OHEopmmw
com omm com oem cam cam oma om, owe co co om u 1.:aev mesa

 

 

meoauaccou 20am

.eaxouooem aaoo .m A

00—

aav mS\ms m co coaumtpmaeaeoa .>.H an» meatsa

acmumcou coca: edema: mwawomcu on» no momcmno uHLooumEos cam ma .mommu vooam .0 dance

.o owm aaoumsfixocaqm um ucmumcoo cam: ocsumgoosou oomuczm oHomnz

Fo.o v_m .
Hocucoo 0» Hence 0 P

 

 

 

 

72

Abpv Aw—v Ammv ANNV Aomv AFPV on va va Amv va Amy ANV AaC\mEv
sup amp com smm m:m mam cam 3mm ram ea: cm 2» he :oHumLucoocoO omoosao
Amv Amy ANV Amy ANV Amv APV APV AFV Amv APV A—V APV N

amp amp cm? amp amp amP amp amp mm— amp app app or haC\HEV pcoucoo o
mzoco> mHHHomLc

 

Amv Amy Amy Amy Amv Amv AFV AFC APV AFV Aw.v APV Ao.v m

6:, :2, map am? am— :mF amF SF SF SF asp MP m? AHC\HEV acoucoo o
Hmficougm OHEOpmaw

 

 

com 0mm oom Ohm cam oFN owp omp omp cm 00 0m 0 A.:HEV mafia

 

 

.eaxoaoncm aaoo .m Aooaaav mx\ms m co nonparamaeaoee< A>.H wastes ”coauaecoo
30am ucmumcoo cone: oaomzz mHHHomco ecu mo mucmsmcsmmoz omoosao msoco> new pcoucoo comzxo .e magma

73

different and remained at control values during the
remaining experimental period (P > 6.65). Arterial and
gracilis venous blood gases, oxygen content, pH, hematocrit,
venous glucose and muscle surface temperature are listed in
Tables 6 and 7. Gracilis venous gluCose concentration
decreased at ninetly minutes and remained below control
values throughout the remainder of the experiment (P <
6.61). Systemic arterial and gracilis venous pH and PCO2
both decreased following the intravenous administration of
endotoxin (P < 6.61); however, the A-V difference of muscle
pH and PCO2 were not different from control throughout the
experiment (P > 6.65). Systemic arterial P02, oxygen
content, gracilis venous oxygen content and hematocrit all
increased significantly from control values (P < 6.61).
Muscle surface temperature was held constant.

Series IV. Hemodynamic and Metabolic Differences of the

Gracilis Muscle During Local Muscle Ischemia,

Independent of Shock.

 

Unlike Series II and III, which dealt with systemic
shock induction, this series of experiments was undertaken
to look at the response of local ischemia on gracilis muscle
glucose uptake.

Local ischemia produced alterations in both gfracilis
muscle metabolism and hemodynamics. These important Changes
are shown in Figures 16 and 11. Following the local
induction of gracilis muscle ischemia, muscle glucose uptake

increased during the sixty minute time period and remained

74

elevated throughout the six hour observation period (P <
6.61). The maximal increase in glucose uptake during local
ischemia was seen during the 246 minute time period.
Glucose uptake had increased from a control value of 6.695
mg/min/166 grams to 6.564 mg/min/l66 grams which represents
a 431% increase. Oxygen uptake and gracilis venous PO2
decreased at the first thirty minute time period (P < 6.61)
and remained below control throughout the remainder of the
experimental protocol (P < 6.61). Arterial glucose
concentration did not differ from conrol during the six hour
observation period (P > 6.65).

Systemic arterial blood pressure did not
significantly Change from control values during the
experiment (P > 6.65). Gracilis muscle blood flow decreased
and remained below control throughout the experimental
protocol (P < 6.61). Gracilis muscle perfusion pressure
decreased initially at the thirty minute time period and
remained below control levels throughout the six hour
observation period (P < 6.61). Muscle vascular resistance,
on the other hand, increased at the first thirty minute
calculation (P < 6.61) and remained elevated throughout the
experiment (P < 6.61). Remaining systemic arterial and
gracilis venous blood gases, pH, oxygen content, hematocrit,
gracilis venous glucose concentraiton and muscle surface
temperature are summarized in Tables 8 and 9. Systemic

arterial p02, PCOz, pH and oxygen content did not

significantly differ from control during the six hour

75

Figure 9: Metabolic alterations of the gracilis
muscle during local muscle ischemia.

vaZ = Gracilis venous P02 in mmHg
*p 0.01

0.5

Glucose
Uptake

(mg/min/ 100g)
0

Arterial
Glucose
Conc.

(mg 7o) 0

4.5

02 Uptake

II Uz/min/ 1002)
0

60

onz
(mm/Hg)

IIsG

l

76

*\*/§"
:4) i

t
\ t t *

 

t

*
l~|

*

t

«It
.

fl

*
.—

*
._.

i

77

Figure 10: Hemodynamic alterations of the gracilis
muscle during local muscle ischemia.

*p 0.01

78

*=p<0.01
n=G

120

Arterial
Pressure
(mm Hg)

0

120r

Perfusion
Pressure
(lilfllflg) * * *

U

 

 

0..
15
Gracilis
l Elm1’00 i
(m MW g) \* t * g» t t t t t a t t
() iI-- --!I ————————— ll
25 * * t * * * * * * t t
i'iri/i\3/§—§~i’§’§\§4
Resistance i/
(mm Hg/mymin/iofl
o J 1 l 1 n l 1 1 l n a e
O 1 2 3 .4 5 I6
ililE

79

.o no ucmumcoo mm: mamsficm Hmucosficoqxo mo Lonssz

SS v m
Hocucoo on Hmsoo o

a
P

 

 

oz m: a: m: m: o: m: m: m: 0: o: m: o: ARV no:
Ame; .ANoJ 23$ So; AmoJ Ame; So; :5; Ame; Amoé No; No; :0;
35—.5 ION.» mom.> mww.h mar.» amp.> mop.> mmp.h 0N.» mN.> om.~ mN.> aN.> Amuwcsv an
A3 A8 Amv 3v 3V A8 A3 at ANV ANV Amy AmV Amy N
tom mom :3 gem :mm cam *mm com om a: a: om S 3:85 Cum

 

m302w> mHHHOwLO

25$ Ame; No; No; No; ANoJ No; No; No; No; 240; Ame; No;

mm.» mm; mm.» mm.» mm; om.» om.» wmg em.» mm.» omC. 2mg Pm; 33:3 ma
2: AC AC AC AC cmv AC AC A8 A8 A8 Amy 2: m
3 3 mm 3 S em em cm em mm mm .3 me SEE 8a
2: 3V 2: AS Amv A9 A3 Amy 2; Amy :3 Amy S m
S. F S S 8 Z. i. me 8 me E 8 S 3:55 on.

 

Hmficoucm OHEOpmam

 

 

oom 0mm com CNN ozN oFN owp om— om— ON 00 cm 0 A.:Hsv OEHH

 

 

.xoosm mo ucmocoqoocH mfismnomH edema: mHHflomco
Hmooq mcacsa momcmno paeooumso: new :a .mmo coon msocm> maafiomco cam HmHLouL< ofisoumam .w manna

.0 pm pcmpmcoo mm: mameficw Hmucosfigoaxo mo Lonssz

56 v m
Hocpcoo 0» Hence 0

s
P

 

 

 

 

 

 

Amv Any Aav Amy asv Amy Amy A:v Amy Amy any Amv ANV AHC\mEV
so: am: o== mm: 3mm :3: as: as: am: tom mom :90 N» coHumcucoocoo omoozau
APV APV APV A—V Am.v APV Am.v APV Am.v Apv AFV APV APV N
em am am am mm am am am am an as am Mr Aac\HEv 9:09:00 o
mso=o> mHHHomLu
APV APV AFV Am.v A—v AFV Apv APV APV APV Am.v AFV AFV N
m— or mp wP w— m— mp or mp m— mF mF m— AHU\HEV ucwucoo O
HmHLouLm OHEOpmmw
com 0mm com CNN cam OFN oww Omp ONP O@ 00 cm 0 A.:HEV OEHH

 

 

.xoocm no pcoocoaoucH .mfisonomH manna: mfiaaomco Hmooq
mcacsa coHumLucoocoo omooSHo msoco> waaaomcu new :omsxo msocm> mHHfiomLo new Hmficouc< oasOpm>m .a canoe

81
experimental period (P > 6.65). Gracilis venous PCO2
increased during the first thirty minute time period (P <
6.61) and remained elevated throughout the six hour
experiment (P < 6.61). Gracilis venous exygen content,
together with pH, were both decreased following induction of
local muscle ischemia (P < 6.61) and remained below control
values during the six hour experimental protocol (P < 6.61).
Gracilis venous glucose concentration was decreased
throughout the six hour experiment (P < 6.61), whereas
venous hematocrit and muscle surface temperature were not

significantly different form control (P > 6.65).

Series V. Effects of Local Gracilis Muscle Hypoxia,

 

Independent of Ischemia, on Muscle Metabolism and

Hemodynamics.

 

Metabolic and hemodynamic changes seen during this
experimental series is graphically illustrated in Figures 12
and 13. Under constant flow, hypoxic coonditions, gracilis
muscle glucose uptake increased at the sixty minute time
interval and remained elevated throuthout the experimental
protocol (P < 6.61). At the termination of the experiment
(6 hours) glucose uptake by the gracilis muscle had
increased from the control value of 6.14 i 6.67 mg/min/166
grams to a level of 1.63 i 6.44 mg/min/l66 grams. This
represents an increase of 1664% above control. On the
contrary, muscle oxygen uptake decreased during the first
thirty minute time calculation and stayed below control

levels during the remainder of the experiment (P < 6.61).

82
Gracilis arterial and venous PO2 both decreased
significantly throughout the six hour observation period (P
< 6.61). Gracilis arterial glucose concentration, systemic
arterial blood pressure and gracilis muscle blood flow were
not significantly different from control (P > 6.65).
Gracilis muscle perfusion pressure decreased initially at
the first thirty minute time period (P < 6.61) and remained
below control until at the sixth hour, where it increased to
a level not significantly different from control (P > 6.65).
Gracilis muscle vascular resistance initially decreased (P <
6.61) until the 216 through the 366 minute calculations
where it was not significantly different form control (P >
6.65).

In two dogs, acetylcholine was infused locally at a
concentraiton of 1 ug/min. During the first hour only
acetylcholine was infused and hemodynamics and metabolic
(glucose) variables were measured. At the first hour, along
with acetylcholine infusion, the gracilis muscle was made
hypoxic and monitored for an additional five hours. A
typical tracing and results are shown in Figure 14. It can
be seen that acetylcholine infusion did not cause an
appreciable increase in muscle glucose uptake; however,
muscle hpyoxia plus acetylcholine during the first through
the sixth hour observation period. caused a marked increase
in muscle glucose uptake. It should also be noted that when

the muscle was made hypoxic concommitant with acetylcholine

infusion, (66 minutes) muscle perfusion pressure decreased

83

Ho.o V.Q«

.mcofluflncou

30am ucwumcoo some: mfisoxomxc oHumDE
HwooH wcflhsw oxmua: comxxo wcm mommw wopo
.oxmua: omousfiw oHumse mHHHumpm :H momcwcu

”as opswam

Aces:
a

32:33 .5.
oxu~==

a
23...:

 

84
2
3
=
S

Wyn—8. m___e..5

a, O
m\ . 38:33

.... .... \m\. .5...
fill W El, . ”was.“
. . £25

ON
.. o ..... o 2.2.: 3:85 up-..
olo .252 ”:33

85

86 V a?

.mcofiuflvcoo

zofim ucmumcoo sows: masoxomxn HmooH
wcflpsv thmmopm wooHn Hmfisoppm ufisoumzm
pew oHumSE mwfiwomsw may mo mofiemczwoeo:

”NH manuam

86

A252:
a

m _ a

m m. e

 

mlmlmlmlnlelul.lalmlmlmlm

l..fil.......,ll..fil/
a. i m

a: 5.5

2:32..

:83
.235
2523

02
o
a: .55
2.5:...
5353..
3.35

00.

o

32:55};
a:

as...
3.33

NP

.o omm zaoumefixotaam um oocfimpcwme mm: oLSmeoasou oommnzm

odomzs ocm NF um acmpmcoo mew: ucoswtooxo on» one uzonmzotnp mamsficm Hmpcosfitoaxm no Lopez: "ouoz

.o.o v_m
Hogpcoo on Hmsao o

o
F

 

 

87

“so . AFV AFV APV APV AFV Arc AFV AFV APV APV Amy Amy

 

P: F: o: .3 o: o: o: o: o: m: m: N: ,2 Age so:
Amo.v Aao.v Amo.v Amo.v Amo.v Amo.v Amo.v Amo.v Amc.v Aso.v Amo.v Amo.v Amo.v
:wo.> umo.b c—F.> :mP.> :NP.> mop.b tar.» umF.> amp.» sop.» amp.» smm.> om.> Amuficsv ma
Amv Amy Amy Amv Amy Amy Amy Amy Amy Amv ANV Amv Amv m
too umm uwm mom tom aom tam cam cam mmm tam azm cm Awmsev cum
msocw> mHHfiOMLu
mm.s Pm.» om.s om.» om.» om.» mm.s Fm.s Fm.» Pm.s mm.» =m.> mm.~ Anuficsv as
Amy Amv “NV Amv Amy Amv Amv Amy Amy Amy Amy Amy Amv m
mm mm Pm Fm om _m mm mm mm .m _m m: s: Ammssv ooa

 

HmHLome ofismumzm

 

 

com omm oom Ohm ozm oPN ow? omp ONF om oo om u A.:HEV mafia

 

.masocomH mo pcoocoooocH .meoaam
odomsz mHHHomLu Hmooq weapon uwtooumso: new :a .mmo oooam msoco> ocm Hmwtopt< mfiawomgo .oP manna

88

.o omm saoumsfixotnam

um owcfimpcams monetsm wHomss ocm .ucmefigooxw on» osocmsotnu NF pm ucmumcoo cams mamsficm mo Looszz
5.0 v m

Hogpcoo op Amoco o

a
—

 

 

Amy Asv Any sz Amy Amv Aav Aav Amv Amy Aav Any Amy Aao\msv
hm Fe Fe me no me P» woo woo m» ow cm w» cofiumgucoocoo omoozao
A=o.v Aoo.v Auo.v Aso.v Apo.v Amo.v Aoo.v Azo.v Aso.v Aoo.v Aoo.v Amo.v AFV m
.mm. .mz. .mm. .m:. .om. .mm. .m:. .m:. .mm. .mm. .sm. 0:. =_ AHu\Hsv pcmucoo o
msoco> mHHHomLo

 

Am.v Am.v Am.v Am.v AP.V Am.v Am.v Am.v AP.V Am.v Am.v Am.v Am.v N
cw.P a>.F :m.F um.P u=.F a=.— :m.F sm.F am.P aw.P aw.— mm.F w? Aao\asv ucmucoo o
Hmwgougm ofiEopmhw

 

 

 

oom 0mm oom Ohm o:N OPN ow— om— omp om co om u A.cHBv mafia

 

 

.mfisonomH mo acmocoomoCH .meoa>: oHomsz
mHHHomLo Hmooq mcfitsa coHumLucoocoo omoooao msoco> mfiaaomto ocm msocm> ocm HmHLopL< mfiawomto .F— magma

89

.mcoflpflocou 30am pcmumcou
sown: :owmswcfi mafiaozuquoom aw“: toadsou
mweomomx: mcfipsb wcm ocflfiogofixuoom :«E\w: H

mo :ofimsmcw ecu wcfissw Emfiaonmuoe bum mowEmcchEo:
oHomSE mwaflompw mo wcfiomsu o>wpmucomosmom

”ma magmas

90

:3.

8n

.ga.»=°u
a.»

Sign?
:: a... £35

. .u: .55
c— —— «m . no em «as 2.23» «:32»

nan—\sii
. . . . 2.8 = 38.5
=3. 8.." «3 v: a: 23.... 3:85

3. .55
2.52..
57.33..

2%....

«=33

 

2.3.5. 3355. .35

SN 2. 8 8 3.5 2.:

91
only slightly (approximately 16 mmHg), indicating that the

muscle bed was nearly maximally dilated during acetylcholine
infusion.

Remaining gracilis arterial and venous blood gases,
oxygen content, pH, hematocrit, gracilis venous glucose
concentration and muscle surface temperature are listed in
Tables 16 and 11. Gracilis areterial P02, pH and oxygen
content, together with venous hematocrit, were not different
from control (P > 6.65). Gracilis venous PCO2 increased
while pH decreased from control (P < 6.61). Gracilis venous
oxygen content and glucose concentration were decreased and
remained depressed throughout the six hour experimental
period (P < 6.61). Muscle surface temperature and pulmonary
venous pressure in the extracorporeal heart-lung preparation

were not significantly altered (P > 6.65).

Series VI. Effects of Local Gracilis Muscle Ischemia,

 

Independent of Hypoxia, on Muscle Metabolism and

 

Hemodynamics.

 

Metabolic and hemodynamic changes associated with
local gracilis muscel ischemia, independent of hypoxia are
graphically presented in Figures 15 and 16. Following local
ischemia together with hyperoxia, gracilis muscel glucose
uptake decreased (P < 6.61) and remained depressed for the
duration of the experiment. Gracilis muscle blood flow was
decreased by the slowing of the blood infusion pump (P <
6.61) and was held at a constant level of approximately 2.5

m1/min/166 grams during the experimental protocol. Muscle

92

 

quva...

.mflxomxn wo ucowcoaowcfl
.mflsocomfi oHumSE mwawomhw Hmoofi wcfls3b
mom msoco> paw Hmfipouhm oHomJE mew oxmums
:omxxo .oxmpm: omoous oHumSE mHHHomso

“ea manage

93

 

mwmmmmm w W. . 8. a.

u 23...:

000 ”=32“—

m 0 EE
WIW%WW-WIW/W\./El/ .22
w . ”=33

ole .25: 2:85 m...

94

Ho.ouv a u «

.mfixoqxc mo unobcomobcfl .meocumfi odomss
mfiflfiomsw Hmuoa wcflpsb muflsmcxpoEo: oHomse
mfififiumsm paw opsmmosm vocab Hmfihoupm ufieoumxm

”ms ossmam

95

 

m\m+I/m|m \mlllw\w/m

a: .55
2:32..

:83
.25:
2533

our
0

9.5.5

238...
5.3::
£25

09.

0

38:55.5
2.:

2......
3:3

NF

96

.mst m com o coozuon

cocampcfims mmzogsmmogq msocw> >meosasm .0 0mm aaoumswxotqom um ucmumcoo cam: mm: ogsumtoosmu
oommgsm oaomsz .Hooopoto Hmucmsfitooxm manage com on» poonmsoggu .w um ucmpmcoo mm: mameficm mo embasz
Po.o v.m

Hotucoo op Hmzco o

o
—

 

on Amv Amy APV A—v APV Amy Amy Amy Amv Amv Amv any

 

am. am 3 0.. S m: .3 .2. m: m.. z. m: .3 .5 so:
Amo.v Amo.v Amo.v Amo.v Awo.v Amo.v Amo.v Amo.v Amo.v Amo.v Amo.v .mo.v Amo.v
asp.» sop.» asp.» ass.» ape.» .mF.s .FP.~ asp.» mo..~ asp.» mpw.> .om.» mm.» Amuacsv me
am. Amv Amy Amv Amy Amy Amy Amv Amv APV “NV Amv A_v m
mom mmm cam yam tam :mm u3m umm mam :mm mam amm m: Awmssv com
msocm> mHHfiomLD
mmé mm... mm.» 8... mm... mm... mm... mm... mm... .m... as Fm; mmg .323 :a
Amy ANV Amv Amv Am. .mv Amy Amy Amy .m. .m. Amv Amv m
om 9. om om Fm mm om mm om .m om m: m... 3:58 8..

 

HmHLouLm oHsmume

 

 

com 0mm com Ohm cam opm om, omp GNP co co om o A.Cwsv mafia

 

 

.meoazz mo ucoccoqoocH masonomH
odomsz mHHHomLo Hmooq wcfigso ufitoopmsmm ocm :o .mom<o coon msoco> ucm Hmagoug< mHHHomLu .NP magma

97

.mzsss m new 0 cwozuwn cocamucwms mm: mgsmmoto maoco> >meosasm .u owm
zaoumsfixogoom pm ucmumcoo cam: mm: otsumtoqsou oommgzm oHomsz .w um acmpmcoo mew: mamsficm mo Lmnssz

.o.o v_m
Hogucoo on Hmsco u

c
P

 

 

.moo.\cfis\ae\mmssv
mocmumfimom
Luasomm> mHomsz

on Amy Amy Amy on on Aev any hey Amv Am. Amy Amy Aao\msv
ms ms es me me ms ms ms ms ms zs as as cofimtpcmocoo mmoosao

Ao.v A>.V As.v .>.v Aw.v Am.v .m.v A>.v Am.v Am.v Aw.v as.v Am.v m

.o.o. .=.o. .o.o. .=.o. .o.o. .m.o. .m.o. .o.o. ...o. .m.o. .:.o. .o.o. ..m. .Hc\asv acmpcoo o
mzoco> mHHHomLo

 

 

 

A:.V Am.v Am.v Am.v Am.v Am.v Ao.v Am.v Am.v A=.v Ao.v Am.v An.v m
:w.om :P.om ao.Fm aw.mF mP.om am.om am.om ww.om ec.mp ap.om so.om nm.mF :.>P AHo\HEV acoucoo o
Hmfigouxm owsounww

 

 

 

com omm oom CNN cam o—N om? om— omp om 00 cm 0 A.:HEV mafia

 

 

.mfixoazm mo ucoccoaoocH .mflsonomH oaomsz mHHHomLo Hmooq wcfigzo oocmumwmom cmasomm>
odomsz new :oHmeucoocoo omoozao msoco> mfiawomtu .ucoucoo com>xo msoco> cam Hmfitout< mHHHomLo .mF manna

98

perfusion pressure initially decreased during the first 90
minutes post ischemia (P > 6.61) then increased back to
control levels (P > 6.65) until 366 minutes when it became
significantly elevated above control values (P < 6.61) and
remained high for the duration of the six hour experimet.
Systemic arterial blood pressure, muscle oxygen uptake and
arterial glucose concentraiton were not different from
control (P > 6.65). Gracilis arterial and venous PO2 both
increased following ischemia plus hyperoxia (P < 6.61).
Remaining gracilis arterial and venous blood gases, oxygen
content, pH, venous glucose concentration, hematocrit,
muscle surface temperature and pulmonary venous pressure are
summarized in Tables 12 and 13. Gracilis arterial PC02 and
pH were not significantly changed during the six hour
experiment (P > 6.65). Gracilis arterial oxygen content was
increased (P < 6.61) wheareas gracilis venous oxygen content
and pH were decreased throughout the entire experiment (P <
6.61). Gracilis venous PCO2 increased initially and
remained elevated throughout the experimental protocol (P <
6.61). Gracilis venous glucose concentration, muscle
surface temperature and pulmonary venous pressure in the

extracorporeal heart-lung preparation were not statistically

different from control (P > 6.65).

DISCUSSION

Four important findings resulted from this study: 1)
E. coli endotoxin infused locally into the constant flow
perfused gracilis muscle promoted glucose uptake by the
muscle; this was not associated with alterations in muscle
metabolism or insulin concentration across the muscle; 2)

when 2 mg/kg (LD coli endotoxin was infused

100) 3'
itravenously to induce shock, gracilis muscle glucose uptake
was not significantly different from control provided the
gracilis muscle blood flow was held constant; 3) under
natural flow conditions to the gracilis muscle, glucose
uptake increased markedly during endotoxin shock; 4) the
mechanism by which glucose uptake is increased under natural
flow conditions during endotoxin shock was shown to be
related to local tissue hypoxia and not to any metabotite
that might be released during shock.

Hypoglycemia is a characteristic feature of gram-
negative endotoxin or septic shock in a number of animal
species (Hinshaw, 1976). While it is well known that
glucose production by the liver is depressed during
endotoxin or septic shock, the study of Peyton, et al.,
1974, suggested that this alone cannot account for the
severity of hypoglycemia.. Therefore, and increased uptake
of glucose by peripheral organ system or systems is likely
involved. Hinshaw and colleagues, in 1975, showed that the

99

100

heart-lung system is not involved in this response. Raymond
and Emerson, in 1978, demonstrated that glucose uptake by
the centrol nervous system decreased during endotoxin shock
in the dog. 0n the other hand, Hinshaw, et al., in 1977,
observed an increase in glucose uptake by white blood cells
during E. coli endotoxin shock, which may account for part
of the hypoglycemia. More recently, spitzer, et al., in
1978, demonstrated that enodtoxin directly stimulated
glucose uptake in isolated epididymal fat pad cells cultered
in vitro. Although it has been suggested that endotoxin may
exert an insulinomimetic acion which could increase skeletal
muscle glucose uptake, this possibility has been relatively
unexplored to date. Furr, et al., in 1977, recently
reported that glucose uptake in the dog forelimb does not
increase during endotoxin shock, which was interpreted to
mean no increased glucose uptake by skeletal muscle.
However, the degree of shock produced in their study was
moderate and many of the hemodynamic and metabolic
alterations normally associated with endotoxin shock were
(very mild. Also, results obtained from the perfused
forelimb preparation are not necessarily representative of
pure skeletal muscle since contamination of muscle flow by
skin and bone cannot be excluded.

Results from Series I demonstrated that E. coli
endotoxin, given locally to the gracilis muscle under
constant flow conditions, increased glucose uptake by the

muscle independent of muscle metabolism. This study,

10]

therefore, supports the hypothesis that endotoxin has
insulinomimetic properties, as speculated by other
investigators (Wolfe, et al., 1977; Hinshae, 1976).
Endotoxin infusion caused an icrease in gracilis muscle
glucose uptake of 173% on the average and occurred with no
change in muscle arterio-venous difference of oxygen, carbon
dioxide, insulin, pH, oxygen uptake, vneous hematocrit or
muscle surface temperature. Hence, the observed phenomenon
does not appear to be dependent on insulin or alterations in
gracilis muscle metabolism. These data, therefore, support
the hypothesis that endotoxin acts like insulin to directly
increase uptake by skeletal muscle; however, in vitro
experiments by Filkin, et al., in 1977, using hemidiaphragms
of rats do not support a direct insulin-like action. No
explanation for this discrepancy is apparent, unless
differences in methodology between this present in vivo
study and their in vitro preparation is involved.

While these data clearly demonstrate that endotoxin
directly increased glucose uptake by skeletal muscle, the
involvement of this mechanism in the progressive, long-term
hypoglycemia associated with endotoxin shock in the dog is
uncertain. Studies by Zlydaszyk and Moon (1976), Buchanan
and Filkins (1976), and Starzecki, et al., (1967) have shown
that endotoxin is taken up rapidly by the
reticuloendothelial system (RES) -- the plasma half-life
averaging between 8 to 66 minutes. It has also been

reported by these investigators that approximately 86% of

102

the injected endotoxin is removedwithin the first 45
minutes.

In the typical experimental model in which endotoxin
is injected as a bolus, endotoxin whold likely be taken up
rapidly by the RES and hence the plasma concentration of
endotoxin would progressively decrease, thereby contributing
less to the direct action of endotoxin on skeletal muscle
glucose uptake causing a dminished direct effect in the
production of hypoglycemia of endotoxin shock. 0n the other
hand, during gram-negative septic shock in the experimental
animal or man, the bacteria are being continually destroyed
and thereby are continually releasing endotoxin into the
blood; the plasma concentration of endotoxin would be
continually increasing and might contribute significantly to
the progression of hypoglycemia. This would become
particularly significant as the RES is overwhelmed and
becomes less effective as septic shock progresses (Berry,
1972).

Experiments in which gracilis muscle blood flow was
held constant while the animals were in endotoxin shock
(2mg/kg, Series III) showed that muscle glucose uptake was
maintained at control levels during 6-hour protocols.
Although muscle glucose uptake was not significantly altered
during endotoxin shock, a tendency for glucose uptake to
increase was noticed at the first thirty minute time
interval post-endotoxin administration. These data supply

enidence that, although endotoxin exerted direct

103

"insulinomimetic” effect in causing muscle glucose uptake
when given locally, its insulin-like effect when given
intravenously was without effect. It should be mentioned
that the concentration of endotoxin given intravenously
(2mg/kg) was equal to approximately 6.38 mg/min/ 166 grams
reaching the gracilis muscle. This is based on the
assumption that all the endotoxin given itravenously would
remain in the plasma; however, as stated earlier, most of
the endotoxin given intravenously is removed early via the
RES, thereby decreasing the plasma levels of endotoxin.
Since the first determination of glucose uptake was made 36
minutes after endotoxin was given intravenously, it is
likely that any direct effect of endotoxin on glucose uptake
would be missed. It is conceivable that early after
endotoxin administration (approximately 1-5 minutes), the
concentration of endotoxin reaching the gracilis muscle
would be sufficient to increase gracilis muscle glucose
uptake. During constant flow conditions, endotoxin plays a
very minor role (< 36 minutes) in contributing to the
hypoglycemia of shock when given intravenously. Also, prior
to 36 minutes, endotoxin administration usually produces an
increase or no change in plasma glucose concentration.
Glucose uptake by the gracilis muscle increased
during endotoxin shock under natural flow conditions (Figure
5) and occurred in the presence of a decreased arterial
plasma glucose concentraiton. Many reports have cited

evidence that, during endotoxin shock in a variety of animal

104

species, plasma insulin concentration is depressed (via
decreased arterial glucose concenteration and increased
catecholamine levels). Based on previous data from the
literature concerning the levels of plasma insulin during
shock, it does not seem likely that enhanced skeletal muscle
glucose uptake is mediated through an increased plasma
insulin secretion. Direct effects of endotoxin, as shown in
Series I, could produce the increased glucose uptake seen in
Series II.

To test the hypothesis that endotoxin directly
stimulated glucose uptake of skeletal muscle in shock, a
third series of experiments was conducted (Series III).
These experiments utilized the gracilis muscle preparation
under constant flow conditions and endotoxin was given I.V.
(2 mg/kg/(LDlgg) to induce shock. Metabolic alterations of
this series are summarized graphically in Figures 7 and 9.
These results provided evidence that, although endotoxin
shock stimulated glucose uptake by the gracilis muscle
during natural flow conditions, it had no effect during
constant flow. These data show that endotoxin does not
exert an insulin like effect during endotoxin shock, and
that the numerous metabolites reported to be released during
endotoxin shock (i.e., histamine and bradykinin) also do not
stimulate glucose uptake during this condition when muscle
blood flow is held constant. This point is interesting
because the study by Nies, et al., (1968) demonstrated the

release of kinins during endotoxin shock. Haberland (1972)

105
demonstrated insulin-like acitons of kinins in increasing
skeletal muscle glucose uptake. The inference from these
two reports was that kinins, released during endotoxin
shock, caused increased skeletal muscle glucose uptake.
This speculation, along with the reSults of Haberland, has
been the subject of much debate over the years. As
previously mentioned, endotoxin removal by the RES --
decreasing the concentration of circulating plasma
endotoxoin to such a low level that it may not have any
insulin-like actions to increase skeletal muscle glucose
uptake. The results from this present study, however, do
not support the concept that any metabolite released during
shock has a direct effect in stimulating glucose uptake by
skeletal muscle.

Differences in glucose uptake by the gracilis muscle
during constant and natural flow conditions seem to imply
local muscle metabolism. Obvious local differences include
decreased oxygen uptake and skeletal muscle hypoxia in the
natural flow series, whereas these changes were prevented by
maintaining constant flow. Morgan, et a1. (1958) and Rendle
and Smith (1958) provided evidence that skeletal muscle
glucose uptake increases substantially during hypoxia and
that this increase is independent of insulin. Since the
gracilis muscle was hypoxic during natural flow shock
experiments, a possible mechanism for increased muscle
glucose uptake during shock is muscle hypoxia.

Changes in systemic arterial blood gases and pH

106
include an elevated P02 and decreased PC02 and pH. The
increases in arterial P02 and decreased in PCO2 can be
attributed to an increased respiratory rate, as was reported
by Raymond and Emerson (1978), Parker and Emerson (1977) and
others (Gilbert, 1966). Although respiratory rate was not
recorded in this study, many of the animals, if not all,
showed early and maintained increased respiratory rates
following the intravenous infusion of endotoxin. Decreased
systemic arterial pH has been reported by Emerson and Gill
in 1968 and Emerson and Kelly in the same year.

The decrease in gracilis venous oxygen content and
increase in PC02 most probably is ascribed to the low muscle
flow which would cause an elevated "lag-time” for oxygen
extraction and CO2 production. Similar reasoning could also
account for a decreased gracilis venous pH but this may also
be produced by an increased anaerobic metabolism by the
muscle during reduced oxygen uptake. During endotoxin
shock, many reports have indicated an increase in venous
lactate concentration and concommitant increase in the
lactate/pyruvate ratio. These findings were interpreted as
reflecting a generalized increase in anaerobic metabolism.
Therefore, the decrease in gracilis venous pH probably
resulted both from an increase in PC02 and an accumulation
of lactic acid via anaerobic glycolysis. Substantiation of
anaerobic glycolysis during natural flow whock experiments

(Series II) is seen by an increase in muscle glucose uptake.

To test the hypothesis that local muscle hypoxia

107

and/or ischemia may be responsible for the increased muscle
glucose uptake, gracilis muscle blood flow was reduced by
decreasing the rate of the perfusion pump. This muscle
ischemia (Series IV) is independent of metabolic and
hemodynamic changes normally associated with endotoxin shock
and gives a beter understanding of local control mechanisms
regulating skeletal muscle metabolic changes associated with
low flow states. Glucose uptake and gracilis venous PO2
were similar to that produced during shock induction in the
naturally perfused gracilis muscle (Figures 5 and 16).

These data confirm the hypothesis that local muscle hypoxia
and/or ischemia cause increases in glucose uptake. To
differentiate between gracilis muscle ischemia or hypoxia in
causing the increased muscle glucose uptake, two series of
experiments were conducted (Series V and VI). During
constant flow gracilis muscle perfusion, local hypoxemia in
lthe inflowing blood stimulated the muscle to increase
glucose uptake. Increased glucose uptake was maintained
throughout the 6-hour hypoxemic, constant flow experiments.
Also, muscle vascular resistance decreased during the period
of local muscle hypoxia. Therefore, the increased glucose
uptake by the gracilis muscle could have resultaed simply
from an increased muscle surface area. To answer this
question, acetylcholine was infused (lug/min) behind the
perfusion pump during constant flow conditions. Initially,
acetylcholine was infused for one hour and glucose uptake

was determined (Figure 14). During this period, neither

108

glucose uptake nor gracilis venous P02 varied from control.
During hypoxemic conditions along with acetylcholine
infusion and under constant flow, gracilis muscle glucose
uptake increased substantially and remained elevated. When
blood perfusing the gracilis muscle was made hypoxemic,
gracilis vascular perfusion pressure did not decrease
substantially (16 mmHg), indicating that acetylcholine
infusion nearly maximally dilated the gracilis muscle
vasculature. These results have shown that vasodilation per
se has no effect in increasing muscle glucose uptake above
control levels, but hypoxia, together with maxial
vasodilation, greatly increased glucose uptake of skeletal
muscle. An explanation for these data is that under normal
aerobic conditions, skeletal muscle, like cardiac muscle,
utilized primarily fatty acids as an energy source and use
only trivial amounts of glucose. Under hypoxic conditions,
however, glucose transport and uptake by skeletal muscle
increases. Increased glycolysis in the absence of oxygen is
termed the Pasteru effect, and explains the increase in
glucose uptake by the gracilis muscle during natural flow
shock, local ischemia, and in Series V, where flow was held
constant but arterial PO2 was decreased. Since it is well
known that tissue hypoxia stimulated glycolysis (Randle and
Smith, 1958), the effect of ischemia, independent of hypoxia
in regulating glucose uptake is vague. Because endotoxin
shock causes muscle ischemia together with hypoxia,

experiments were completed to determine what role a

109

decreased substrate delivery via ischemia would have on
glucose metabolism under normoxic and hyperoxic conditions
(Series VI). Induction of severe ischemia of the gracilis
muscle, together with local hyperoxia, did not result in
increased uptake of glucose by the gracilis muscle -- in
fact, glucose uptake decreased during the hyperoxic state
(increased gracilis venous P02). As mentioned previously,
skeletal muscle metabolism is similar to cardiac metabolism.
Under aerobic conditions, skeletal and cardiac muscle
preferentially utilize fattly for energy production. While
skeletal muscle utilized only small amounts of glucose
during control situations, increases in oxygen delivery or
increases in tissue PO2 would result in a greater efficient
utilization of fatty acids rather than glucose (White, et
al., 1973). The arterio-venous difference of glucose across
the gracilis muscle remained unchanged during
ischemic-hyperoxia which resulted in a calculated decreased

in glucose uptake (Pasteur effect).

SUMMARY AND CONCLUSIONS

The purpose of this study was to define the role
played by skeletal muscle in regulating glucose metabolism
during endotoxin shock. It was found that E. coli endotoxin
had insulinomimetic activity by increasing glucose uptake of
the gracilis muscle when infused locally. Direct
insulin-like actions of endotoxin were not noted when
endotoxin was given intravenously to induce shock.
Intravenously administered lethal doses of E. coli endotoxin
did cause a sharp fall in gracilis muscle blood flow
resulting in tissue hypoxia. During constant flow
conditions, no oxygen differences were seen. In experiments
where blood flow was held constant, endotoxin shock did not
alter glucose metabloism of the gracilis muscle, which
provided evidence that metabolites, released as a
consequence of shock induction also failed to alter glucose
metabolism in the gracilis muscle, which provided evidence
that metabolites, released as a consequence of shock
induction also failed to alter glucose metabolism in the

gracilis muscle. A possible explanation for endotoxin not

110

111

causing an increase in muscle glucose uptake during constant
flow is related to the fact that the half-life of endotoxin
in plasma is short -- approximately 8 to 45 minutes. The
first measurement of glucose uptake during constant flow
shock was made at 36 minutes post-endotoxin infusion. Any
glucose uptake that may have occurred earlier than 36
minutes would have been missed.

Local muscle hypoxia was found to be the major, if
not the only, mechanism for increased glucose uptake in
shock, and ischemia, alone, did not cause an elevated
glucose uptake, provided muscle tissue was not hypoxic. In
fact, glucose uptake decreased during ischemic hyperoxic
condtions.

In conclusion, it is felt that occurrence of
hypoglycemia associated whit endotoxin shock arises from two
major sources. Hepatic alteration in gluconeogenesis have
been well defined in shock. Although it is depressed, the
amount of glucose produced is still above control levels;
however, increased peripheral utilizatjion of glucose by
muscle may be the key factor which causes plasma glucose to

decrease early following endotoxin shock in the dog.

BIBLIOGRAPHY

112

Alican, F.M.; Dalton, M.L., Jr.; and Hardy, J.O.:
Experimental endotoxin shock. Am. J. Surg.
163:762, 1962.

Aust, J.B.: Distribution kinetics of various substances in
endotoxin shock. Fed. Proc. 18:5, 1959.

Archer, L.T.: Hypoglycemia in conscious dogs in live
Escherichia coli septicemia: achronic study.
Circ. Shock 3:93, 1957.

Balis, J.V.; Rapport. E.S.; Gerber, L.; and Buddengh, F.:
Continuous endotoxemia in Rhesus monkeys as a
clinically relevant model for shock-lung. Am. J.
Pathol. 74:96, 1974.

Benjamin, B.; Alaupovic, P.; Wang, C.S.; and Hinshaw, L.B.:
Relationship of chemical structure to
pathophysiologic properties of endotoxin from
serratia marcescens. Circ. Shock 1:61-69, 1974.

Berk, J.L.; Hagen, F.F.; Beyer, W.H.; and Gerber, M.J.:
Hypoglycemia of shock. Ann. Surg. 171:466, 1976.

Berry, L.J.: Metabolic effects of bacterial endotoxins.
Page 165 in S. Kadis, G. Weinbaum, and S.J. Ajl,
eds. Microbial Toxins. Vol. V. Academic Press,
New York. 1971.

Berry, J.L.: Factors influencing the function of the
reticuloendothelial. J. Reticuloendothelial Soc.
11:456, 1972. ’

Binkley, F; Perlman, E.; andf Goebel, W.F.: jThe chemical
degradation of the specific antigen of type Z
Shigella paradysenteriae (Flexner) J. Exptl. Med.
81:331-347, 1945.

Blackwood, J.M.; Hsieh, J.; Fewel, J.; and Rush, B1F., Fr.:
Tissue metabolites in endotoxin and hemorrhagic
shock. Arch. Surg. 167:181, 1973.

Blattberg, B. and Levy, M.N.: Sites of extra hepatic
pooling in early response to endotoxin. Am. J.
Physiol. 226:1267, 1971.

Boho. C.T.; Turbow, M.E.; and Kolmen, S.N.: Improvement of
coronary blood flow in endotoxin shock without
improvement of myocardial conrtractility.
Physiologist 16:268, 1973.

113

Boivin, A.; Mesrobeanu, I.; and Mesrobeanu, L.: Extraction
d'un complexe toxique et antigenique a partir du
bacile d'aertrycke. Compt. Rend. Soc. Biol.
114:367, 1933.

Boivin, A.; Mesrobeanu, I.; and Mesrobeanu, L.: Technique
pour la preparation des polysaccharides microbiens
specifiques. Soc. Biol.‘ (Paris) 113:496, 1933a.

Boroff, D.A.: Study on toxins and antigens of shigella
dysenteriae I. toxicity and antigenicity of whole
organisms and various fractions of shigella
dysenteriae. J. Bact. 57:617-632, 1949.

Brochman, S.K.; Thomas, C.S., Jr.; and Vasko, J.S.: The
effect of Escherichia coli endotoxin on the
circulation. Surg. Gynec. Obstet. 125:763, 1967.

Brand, E.D. and Lefer, A.M.: Myocardial depressant factor
in plasma from cats in irreversible post-oligenic
shock. Proc. Soc. Exptl. Biol. Med. 122:226,
1966.

Chien, 8.; Chang, C.; Dellenback, R.J.; Vsami, 8.; and
Gregersen, M.I.: Hemodynamic changes in endotoxin
shock. Am. J. Physiol. 216:1461, 1966.

Chien, 8.; Dellenback, R.J.; Vsami, S.; Treitel, K.; Chang,
C.; and Gregersen, M. I.: Blood volume and its
distributjion in endotoxin shock. Am. J. Physiol.
216: 1411, 1966.

Chien, 8.; Sinclair, D.G.; Dellenback, R.J.; Chang, jC.;
Peric, B.; Vsami, 8.; and Gregersen, M.I.: Effect
of endotoxin upon capillary permeability to
macromolecules. Am. J. Physiol. 267:518, 1964.

Clowes, G.H.A., Jr.; O'Donnell, T.F., Jr.; Ryan, N.T.; and
Blackburn, G.L.: Energy metabolism in sepsis:
Treatment based on different patterns in shock and
high output state. Ann. Surg. 179:684, 1974.

Cryer, P.E.; Coran, A.G.; Sode, J.; Herman, C.M.; andf
Horwitz, D.L.: Lethal Escherichia coli septicemia
in the baboon: Alpha adremergic inhibition of
insulin secretion and its relationship to the
duration of survival. J. Lab. Clin. Med.

79:622, 1972.

Cryer, P.E.; Herman, C.M.; and Sode, J.: Carbohydrate
metabolism in the baboon subjected to
gram-negative (E. coli) septicemia. I.
Hyperglyceria with depressed (plasma) insulin

114
concentrations. Ann. Surg. 174:91, 1971.

Dedichen, H. and Schenk, W.G.: Hemodynamics of endotoxin
shock in the dog. Arch. Surg. 95:1613, 1967.

Drucker, W.R., and Dekiewiet, J.C.: Glucose uptake by.
diaphragms from rats subjected to hemorrhagic
shock. Am. J. Physiol. .266:317, 1964.

Ebert, R.V., Borden, C.W., Hall, W.H., and Gold, 0.: A
study of hypotension (shock) produced by
meningecoccus toxin. Circ. Res. 3:378, 1955.

Emerson, T.E., Jr., and Gill, C.C.: Effect of slow
intravenous infusjion of endotoxin in dogs. J.
Appl. Physiol. 22:874,1967.

Emerson, T.E., Jr., and Kelly, K.C.: Effect of living E.
coli cells on hemodynamics and mortality in the
dog. J. Appl. Physiol. 23:669, 1967.

Engel, F.L.: Metabolic aspects of the shock problem. J.
Mt. Sinai Hosp. 12:152, 1946.

Epstein, R.B., Waxman, F.J., Bennett, B.T., and Andersen, B.
R.: Pseudomonas septicemia in neutropenic dogs.
I. Treatment with granulocyte transfusions.
Transfusion 14:51, 1974.

Felig, P., and Wahren, J.: Amino acid metabolism in
exercinsing man. J. Clin Invest. 56:2763, 1971.

Ferguson, W.W., Glenn, T.M., and Lefer, A.M.: Mechanisms of
production of circulatory shock factors in
isolated perfused pancreas. Am. J. Physiol.
222:456, 1972.

Filkins, J.P., and Buchanan, B.J.: In vivo vs. in vitro
effects of endotoxin on glycogenolysis,
gluconeogenesis and glucose utilization. Proc.
Soc. Exptl. Biol. Med. 155:216, 1977.

Filkins, J.P., Buchanan, B.J. and Cornell, R.P.: Hepatic
carbohydrate metabolism alterations during
endotoxin and traumatic shock. Circ. Shock
2:129, 1979.

Filkins, J.P. and Cornell, R.P.: Depression of hepatic
gluconeogenesis and the hypoglycemia of endotoxin
shock. Am. J. Physiol. 227:778, 1974.

Freedberg, A.S., Haimovici, H., and Glumgart, H.L.: The
cardiovascular dynamics in shock dur to infection.
J. Clin. Invest. 23:936, 1944.

115

Guenter, C.A., Fiorica, V., and Hinshaw, L.B.:
Cardiorespiratory and metabolic responses to live
E. coli and endotoxin in the monkey. J. Appl.
Physiol. 26:786, 1969. .

Haberland, G.L.: Selected references in G.L. Haberland and
J.W. Rohen, eds. Kininogenases, first symposium
on physiological properties and pharmacolocical
retionale. F.K. Schattauer Verlag, New York,
1972.

Haddy, F.J., Overbeck, H.W., and Daugherty, R.M.:
Peripheral Vascular Resistance. Annu. Rev. Med.
19:167, 1968.

Hall, R.C., and Hodge, R.L: Vasoactive hormones in
endotoxin shock: A comparative study in cats and
dogs. J. Physiol. (London) 213:69, 1971.

Halnagyi, D.F.J., Broell, J., Gump, F.E., and Kinney, J.M.:
Hypermetabolism is sepsis: correlation with fever
and tachycardia. Surgery 75:763, 1974.

Haskins, W.T., Landy, M., Milner, R.C., and Ribi, E.:
Biological properties of parent endotoxins and
lipid fractions, with kinetic study of
acid-hydrolyzed endotoxins. J. Exper. Med.
114:665-684, 1961.

Herion, J.C., Walder, R.I., Herring, W.B., and Palmer, J.C.:
Effects of endotoxin and nitrogen mustarf on
leukoctye kinetics. Blood 25:522, 1965.

Herlihy, B.L., and Lefer, jA.M.: Alteratjions in pancreatic
acinar cell organelles during circulatory shock.
Circ. Shock. 2:143, 1975.

Hinshaw, L.B.: The role of glucose in endotoxin shock.
Circ. Shock 3:1, 1976.

Hinshaw, L.B., Archer, L.T., Beller, B.K., White, G.L.,
Schroeder, T.M., and Holmes, D.D.: Glucose
utilization and tole of blood in endotoxin shock.
Am. J. Physiol. 233:E 71, 1977.

Hinshaw, L.B., Archer, L.T., Black, M.R., Elkins, R.C.,
Brown, P.P., and Greenfield, L.J.: Myocardial
function in shock. Am. J. Pysiol. 226:357, 1974.

Hinshaw, L.B., Archer, L.T., Greenfield, L.J., and Guenter,
C.A.: Effects of endotoxin on myocardial
hemodynamics, performance and metabolism. Am. J.
Physiol. 221:564, 1971.

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Hinshaw,

Horwitz,

116

L.B., Archer, L.T., Greenfield, L.J., Miller, J.A.,
and Guenter, C.A.: Effects of endotoxin on
myocardial performance. J. Trauma. 12:1656,
1972.

L.B., Beller, B.K., Archer, L.T., and White, G.L:
Associated leukocyte responses in the lethal
aspects of E. coli shock. -Proc. Soc. Exptl. Biol.
Med. 155:179, 1977.

L.B., Benjamin, B., Coalson, J.J., Elkins, R.C.,
Taylor, F.B., Jr., Price, J.T., Smith, C.W., and
Greenfield, L.J.: Hypoglycemia in lethal subhuman
primate septic shock. Circ. Shock 2:197, 1975.

L.B., Brake, C.M., Emerson, T.E., Jr., Jordan,
M.M., and Masucci, F.D.: Participation of the
sympathoadrenal system in endotoxin shock. Am. J.
Physiol. 267:925, 1964.

L.B., Gilbert, R.P., Kuida, H., and Visscher, M.B.:
Peripheral resistance changes and blood pooling
after endotoxin in evicerated dogs. Am. J.
Physiol. 195:631, 1958.

L.B., Greenfield, L.J., Owens, W.E., Archer, L.T.,
and Guenter, C.A.: Cardiac response to
circulating factors in endotoxin shock. Am. J.
Physiol. 222:1647, 1972.

L.B., and Nelson, D.L.: Venous response of the
intestine to endotoxin. Am. J. Physiol. 263:876,
1962.

L.B, and Owens, S.E.: Correlation of pooling and
resistance changes in the canine forelimb in
septic shock. J. Appl. Physiol. 36:331, 1971.

L.B., Peyton, M.D., Archer, L.T., Black, M.R.,
Coalson, J.J., and Greenfield, L.: Prevention of
death in endotoxin shock by glucose
administration. Surg. Gynec. Obstet. 139:851,
1974.

L.B., Vick, J.A., Jordan, M.M., and Wittmers, L.B.:
Vascular changes associated with development of
irreversible endotoxin shock. Am. J. Physiol.
262:163, 1972.

D.L., Ballantine, T.V.N., and Herman, C.M.: Acute
effects of septic shock on plasma and red cell
volumes in baboons. J. Appl. Physiol. 33:326,
1972.

117

Kinney, J.M.: Energy expenditure and tissue fuel in the
septic patient. Page 145 in B.K. Forsoher, R.C.
Lillehei, and 3.8. Stubbs, eds. Shock in Low and
High-Flow States. Excerpta Medica, Amsterdam.

Kraneberg, 6., and Sandritter, W.: Zur toxikoligie der
Bakterienendotoxine. Arch. exper. Path. u.
Pharmakol. 218:111, 1953.

Kuida, H., Hinshaw, L.B., Gilber, R.P., and Visscher, M.B.:
Effect of gram-negative endotoxin on pulmonary
circulation. Am. J. Physiol. 192:335, 1958.

LaNoue, K., Mason, D.D., Jr., and Daniels, J.: The
impairment of gluconeogenesis by gram-negative
infection. Metabolism 17:666, 1968.

Lansing, A.M.: Septic shock. J. Canad. Med. Assoc.
89:583, 1963.

Lecocq, F.R., Mebane, D., and Madison, L.L.: The acute
effect of hydrocortisone on hepatic glucose output
and peripheral glucose utilization. J. Clin,
Invest. 43:237, 1964.

Lefer, A.M.: Role of myocardial depressant factor in the
pathogenesis of circulatory shock. Fed. Proc.
29:1836, 1976.

Lefer, A.M.: Role of a myocardial depressant factor in
shock states. Modern Con. Cardiovasc, Dis.
42:59, 1973.

Lefer, A.M., and Spath, J.A., Jr.: Pancreatic hypoperfusion
and production of a myocardial depressant factor
in hemorrhagic shock. Ann. Surg. 179:868, 2974.

Lillehei, R.C., Longerbeam, J.K., Block, F.H., and Manax,
W.G.: The nature of irreversible shock:
Experimental and clinical observations, Ann. Surg.
166:682, 1964.

Long, C.L., Spencer, F.L., Gump, F.E., and Kinney, J.M.:
Hypermetabolism in sepsis: correlation with fever
and tachycardi. Surgery. 75:763, 1974.

Lundgren, O.J., Lundwall, J., and Mellander, 8.: Range of
sympathetic discharge and reflex vascular
adjustment in skeletal muscle during hemorrhagic
hypotension. Acta. Physiol. Scand. 62:386, 1964.

118

Lundsgaard—Hansen, P., Pappova, B., Urbaschek, B., Heitmann,

Mack,

B.,

L., Leaderach, A., Manes, N., Oroz, M., and Wirth,
V.: Circulatory deterioration as the determinant
of energy metabolism in endotoxin shock. J. Surg.
Res. 13:282, 1972.

and Egdahl, R. H.: Adrenal blood flow and
corticosteroid secretion in hemorrhagic shock.
Surg. Gynec. Obset. 131: 65, 1976.

Madison, L.L., Combes, B., Strickland, W., Unga, R., and

Adams, R.: Evidence for a direct effect of
insulin on heptic glucose output. Metabolism.
8:469, 1959. .

Marchick, J.B., Finley, R.J., Groves, A.C., Wolfe, L.I.,

Holliday, R.L., and Duff, J.H.: Catabolic
hormones and substrate patterns in septic
patients. J. Surg. Res. 23:177, 1977.

Marlis, E.B., Aoki, T.T., Toews, C.J., Felig, P., Cannon,

J.J., Kyner J., Huckabee, W.E., and Cahill, G.F.:
Amino acid metabolism in lactic acidosis. Am. J.
Med. 52:474, 1972.

Martin, D.S., Cassisi, N.J., and Pickens, J.L.: Endotoxin

shock: a collective review. Rev. Surg. 311,
1965.

McCallum, R.B., and Berry, L.J.: Effects Of endotoxin on

gluconeogenesis, glycogen synthesis and liver
glycogen synthase in mice. Inf. Immunity. 7:642,
1973.

Mela, L., Bacalzo, L.U., and Miller, L.D., Defective

oxidative metabolism in rat liver mitochondria in
hemorrhagic and endotoxin shock, Am. J. Physiol.
226:571, 1971.

Mellander, 8., and Lewis, D.H.: Effect on hemorrhagic shock

on the reactivity of resistance and capacitance
vessels and on capillary filtration transfer in
cat skelltal muscle. Circ. Res. 13:165, 1963.

Menten, M.L., and Manning, H.M.: Blood sugar studies on

rabbits infected with organisms of the
enteritidfis-parathyphoid B group. J. Med. Res.
44:675, 1924.

"11111111111111.111111111“