v v
v 47 I u . 3'1‘13. .“f‘ 5“.’o o I
v 5 r 1' :01 <\ .( 1 .1 . . no ~ . 1!1l‘. . all
1 1 .r. V: 1. .1 x n... 151511111 1! . . .1

‘ 1 of on! u 1' .1.” o o ‘15.... .01....!¢111.J..1J.1lfi.1glls D I
. 1 .1 1 I l lo‘d l. (i!\.l.. _ l
1. 1 . 1741‘. .1. 1

v 4

.-
0 I1
. .1 - 1 . .1 1 .. . ..1 .21. . 1. 1.. 1..
:1I ; r v 1 .10 l. Itasr1o|.. v
a 11 11|1 1: .o 11 z .7... l .10 n. 51.1 .v
1 4o .11 11'; 1!. l 1 .. a. 1 .i1v'r..1.l 1.1.1 ....'~ol.~ 1‘1 .11
'- V I r a 1:1 1......1 . . o 1.1! (5.1.11 1
1 11. 1 .u; .11- 1 11:...) 1,111....11w14111111M 1911‘! 1 71.... 1.
1 o 1!». .u.-l1l|lr.o.l.ov,11U..n . o l. 1 1. 1
. .1?.‘I1!.lf!9.1«.h1k.ao...1.. at”
..|.\H§1)I.o1 Q‘v 1.1‘.-o.1r .

- n c ' '1‘! 1 I
. o .0” U1.oa> .fi!!! !.1U!1~;..\...l. .
1! 1". I‘o 1.. u l ! o

0"... VI . v1. 11 ..
1 1! _. 1 1 . I. . 11. 1 . 1, v
. 1! - 111l1r1!l . 1H! I 1! 1.0!! 1.. . 1
- .-11u11ul.|||!1‘hv.!4“1 I 101‘?! .1 .1711 11". ‘1':
a 1 . \ 11d. 0 ha I . .
. 111 .1o.1.!.fl.olo {H.Nflmb. 1.. H1111“ ..«11.1h11.l1 .l . .41 .1 Tobi...
4§f.1!’ !l l II 1 .1 .!. I! “(‘10: I,
1 If!!!” .ffiruqfduflgfi f1. 1 IT a -1 lo, I [.13 . .1
OHIIII' 2| D1.ll!'1.h.lll"”l ! r'1'. {:40 Elk! .Iw $1. .!lr-Jol}"o 'w".l
.1 o HR!JA.uIIVRo!l 1! J 1.! _ . 1 1 1.1 .l..u1q1.!..hl,.uuflflm.1.
. ovll.n1 .\r.. !.. 1 all , ‘1!!! 8.1). l 1! 1 1 1o». 1
. .11 .111! I. :11! 1!. !..I11H1!o!1 for”... N1!!!) .1111! I.!o.\~1!lol.1..b.r\.!!o._!!.1. 11......1L111L1 1....1» .1: . o1
’ 1.!.1I!.11! 1"» .JIHI!’ v! (Idirlf 1 <1 Lfl. .!I!C‘ [Alt v3.10! {1.14 13!..11J: .v u
I II 'II loll 1 I . . b o . V! 1|. 1
181:!1110I11;ol. Rololnioollofiu lo éiool. _ rind P14911911“! lIF 3111 1 .
r 93!- .11 If Rial} 19.1.IHIJW! {Uri-”11“}! II! 16") ‘1' ‘ cpl! ILIDOL. I‘ll: '1 1
! 1 11 11-111! 1 1! l .
mmflmflklo! .l!11.1.1!1..1..1 11 11 Nahum” 11.. . . , . 1
1
o, 3' 1 L
.117

.
'15.”,

111:] .1 Ito! . 1 yo 1
mm 1... . 1!!dR11:..1rF11fl111lo!11u.1u1u9u1
lbw!!!“ 111. .1111 !.Nuotrl.ol11!§.l!o
OJ oi O‘III b o l v '3”. «I 0“ 11
1. 1, o unflofldrflflhu l1u1mufls1
’ lo” I o ‘1! 1V.-
. H1! - ”g1 Romafiorwol .
1 lynRerumflwx l.-l!111fi1fld!1M1K 1
1 - - flg1lxull1illflbdm!
“Nico!!! (A Ell - v

,1!
!1 L1!
1 . o 1 1dnlo
%!Hol!
-Ol !.1‘ll1 #(T‘.

‘ 1 l
.0! ‘11} I I
. . r

I?!" 1H1!11H- !.I

I. I l
1 1 o p .
u 1 no. . N o a I .
n 1 1 1o 1 I! C
. I. 11 1 111.1 111 bu..- . c 14 1‘ 11!} .
v. .111 1 1o 1. 711. v1 . 1 I 1 1 1 loan”.
.1 1 . 1 . 1: . 1 . 1. 1. .
ontll. ~o I A: .10 | l O! 1 I . k?
l. . 1 . v 1 1 o o 1 g 5 1 '111
.1 .I .1 1. I. L. 1 1‘ l1 . fl... .1
1. . I 1.

o .\ 1.1
a I l
. a . .1 ..
I I: ‘ I 1! l 1
1‘11. TIMUUI‘ 2.111 1J1- "obfuwuv 1.
o 1. ll! "Ilh..f|.!o\olill'.l I‘RHN). IL? {In W l {no
1 n. rll o ... Iggth . “11 .r: u .
. .. IhIIHth3hfir-1llll1lhd 1 1 N1! 11 I14.
. I 4 .|o'n4‘!o‘}- t. .Iu.‘!|‘.l ”. .‘.ol..l’u” .‘ l
. u. k . . ouflmgrldfl W111R 1.11J11umr11a..uo!k1¢ll!1.1111,1l 1 .. 1
1 n I «11 V on "I! '1 1‘. (IIJHVJIY1 1. v
I n o l 1 111 Iowan. 11”"! .1! l !.
I l .

1
I I. . '
o.fo"1fl.1\ .0 o
l!- lvl..! 1
N I
‘ -

 

'¥;l-‘:w!2~

 

lIBRARY
Michigan State
¢ University

 

 

 

This is to certify that the

thesis entitled

Evaluation Of Mixed Venous Oxygen Tension As An
Estimate 0f Cardiac Output In Anesthetized Horses

presented by

Lois Ann Netmore

has been accepted towards fulfillment
of the requirements for

Master of Science degreein Large Animal
Clinical Sciences

/

Major professor

 

Date August 4, 1987

 

0-7 639

 

 

 

 

‘IVIESI—J RETURNING MATERIALS:
Place in book drop to
Lian/muss remove this checkout from
—-c—. your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

 

 

EVALUATION OF MIXED VENOUS OXYGEN TENSION
AS AN ESTIMATE OF CARDIAC OUTPUT
IN ANESTHETIZED HORSES

BY

Lois Ann Wetmore

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Large Animal Clinical Sciences

1987

ABSTRACT

EVALUATION OF MIXED VENOUS OXYGEN TENSION
AS AN ESTIMATE OF CARDIAC OUTPUT
IN ANESTHETIZED HORSES

BY

Lois Ann Wetmore

The relationship between mixed venous 02 tension and
cardiac output was studied in six anesthetized horses
breathing 100 percent 02. Cardiac output, 02 consumption,
mean arterial pressure, heart rate, and arterial and
venous blood gases were measured after administration of
xylazine or dobutamine to horses in lateral, eternal and
dorsal recumbencies. At the end of the approximately

three-hour experimental period, Escherichia coli endotoxin

 

was administered while horses were in dorsal recumbency,
and all measurements were repeated. Relationships between
cardiac index (CI) and P902, heart rate, mean arterial
pressure, jugular PVOZ, and PVO2 of blood from a super-
ficial limb vein were evaluated by linear regression
analysis. Mean arterial pressure was significantly

(P< 0.05) correlated with CI in horses in all positions
and after endotoxin administration. However, data points

were poorly grouped. Heart rate and CI were significantly

Lois Ann Wetmore

correlated in all horses in all positions but not after
endotoxin administration. The correlations between jugular
on2 and PVO2 of blood from a superficial limb vein were
not significant in horses in sternal recumbency, and PVO

of blood from a superficial limb vein was not significantly
correlated with CI in horses in lateral recumbency. There
was a significant and tight correlation between PVC and CI
in horses in all positions and after endotoxin administra-

tion.

DEDICATION

To my mother,

Hanna Ivory Wetmore

ii

ACKNOWLEDGMENTS

I wish to express my sincere gratitude to Dr. Frederik J.
Derksen, my major advisor, for his guidance, enthusiasm and
assistance in the preparation of this thesis. I am also
indebted to Drs. Cheryl A. Blaze, George E. Eyster, and
N. Edward Robinson, members of my graduate committee, for
their assistance, counsel and constructive criticism.

I am grateful to Roberta Milar and Cathy Berney,
Pulmonary Lab Technicians, for their assistance.

I wish to thank Margaret R. Hoffman, the graphic artist,
for her expertise and patience in preparing the graphs. I
also thank Kathryn Sayles Winsky, Antoinette Tenlen, Martha
Devlin and Karen Schiffer for their clerical expertise in
preparing the manuscript that is a portion of this thesis.

Funding for this project was provided in part by a

grant from the Jesse Smith Noyes Foundation.

iii

TABLE OF CONTENTS

LIST OF TABLES .'. . . . . . . . . . .
LIST OF FIGURES . . . . . . . . . . .
INTRODUCTION . . . . . . . . . . . . .
LITERATURE REVIEW . . . . . . . . . .

1. Cardiac Output Measurements
in Horses . . . . . . . . .

2. Application of the Pick Principle for

Estimation of Cardiac Output

a. The Use of A-VOz Difference
to Estimate CO . . . . .

b. The Use of 8602 to Estimate

c. The Use of PVOZ to Estimate
MATERIALS AND METHODS . . . . . . . .
Horses . . . . . . . . . . . . . .
Measurement techniques . . . . . .
Experimental design . . . . . . . .
Statistical analysis . . . . . . .

RESULTS . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . .
CONCLUSIONS . . . . . . . . . . . . .
FOOTNOTES . . . . . . . . . . . . . .

LIST OF REFERENCES . . . . . . . . . .

iv

CO .

CO .

page

vi

19
24
24
24
28
30

32
47
52
54

LIST OF TABLES

page

Table 1. Cardiac Output Values in
Standing Horses . . . . . . . . . . 4

Table 2. 3 x 3 latin square design
describing random sequencing
of recumbencies . . . . . . . . . . 29

Table 3. Cardiovascular measurements after
xylazine (0.65 to 1 mg/kg) and
after dobutamine administration
(1 to 7 ug/kg/min) and 15 minutes
after endotoxin administration
(100 ug/kg) in six anesthetized
horses . . . . . . . . . . . . . . 33

Table 4. The R values for linear regression
analysis of CO or CI and cardio-
vascular variables . . . . . . . . 45

Table 5. The R values for linear regression
analysis of cardiac index and
cardiovascular variables in each
recumbency . . . . . . . . . . . . 46

Figure

Figure

Figure

Figure

Figure

Figure

LIST OF FIGURES

Equipment used for measurement

of oxygen consumption .

The relationship between mixed

venous 02 tension (PV02) and‘

cardiac index (CI). . .

The relationship between jugular
venous 02 tension (PVO JUG) and

cardiac index (CI). . .

The relationship between peripheral
venous 02 tension (PV?2 PERI

and cardiac index (CI

The relationship between heart

PH)

rate (HR) and cardiac index (CI).

The relationship between mean

arterial blood pressure (MAP)

and cardiac index (CI).

vi

page

26

35

37

39

41

43

INTRODUCTION

A decrease in cardiac output is in part responsible for
complications related to equine anesthesia including post-
anesthetic recumbency myOpathy and increased alveolar-
arterial oxygen tension differences. Horses that require
emergency anesthesia for repair of acute gastrointestinal
or reproductive disorders frequently have concurrent hemo-
dynamic instability. In all these cases, intraoperative
monitoring of cardiac output is desirable.

Cardiac output in the horse has been measured directly
and indirectly. Direct measurement of cardiac output is not
clinically applicable. Indirect measurements of cardiac out-
put have been made using indicator dilution methods and the
Fick principle. Indicator dilution measurements of cardiac
output are not routinely performed on anesthetized horses
because these measurements require specialized equipment and
technical familiarity with catheter placement and use of
the equipment. Measurement of cardiac output (CO) using the
Pick principle requires measurement of oxygen consumption
(902) and the oxygen content of arterial and mixed venous

blood.

V02
CO = arterial 02 content - mixed venous 02 content

 

2
Where the content of oxygen (C02) in mixed venous or arterial
blood is calculated from the following formula requiring hemo-
globin content in gm/lOO ml blood (Hb), the 02 saturation of
the hemoglobin ($02) and the oxygen tension in the blood (P02).
C02 = (So2 x Hb x 1.34) + (PO2 x 0.003)

Measurement and calculation of these variables is too cumber-
some for routine clinical use.

In a steady state a decrease in CO is accompanied by a
decrease in SVO2 and PVO2 if oxygen consumption and arterial
oxygen content remain constant.

The purpose of the present study was to investigate the
use of P602 in the estimation of CO in anesthetized normal

horses and in horses given a bolus of Escherichia coli endo-

 

toxin. In addition, I examined the variability of oxygen
consumption in anesthetized horses because this variability
might interfere with the use of on2 in the estimation of

CO.

LITERATURE REVIEW

1. Cardiac Output Measurements in Horses

Cardiac output in the horse was first measured using
the Pick principle (Zuntz and Hagemann, 1892). Blood was
taken from an artery and from the right ventricle and blood
gas analyses were performed. Oxygen consumption was deter-
mined by examination of expired air. From these measurements
a value of approximately 75 ml/kg/min was calculated. Cardiac
output measurements in horses were not reported again until
1961 when Fisher and Dalton described results of a study using
an indicator dilution method previously employed in other
species. Dye (Evans blue 20 mg/ml) was injected into the
jugular vein over four seconds at a dose of l ml/SO kg body
weight. A series of arterial blood samples were then collected
from the brachial artery. Analysis of the dye content in
these samples was done and cardiac output was calculated.
Using fourteen horses the mean cardiac output value was again
75 ml/kg/min. Over the following fifteen years dye dilution
studies for determination of cardiac output in horses were
repeated with variable results (Table 1). It wasn't until
1976 that a thermodilution method for cardiac output deter-
mination was described in horses (Muir, Skarda, and Milne,
1976). Cardiac output was measured in each horse by both
dye dilution and thermodilution techniques. Since there are

3

Table 1.

Cardiac Output Values in Standing Horses

Investigators

 

Fick principle
Zuntz and Hagemann (1892)

Dye dilution technique
Fisher and Dalton (1961)
Eberly, Gillespie and Tyler (1964)
Hall, Gillespie and Tyler (1968)

Eberly, Gillespie, Tyler and
Fowler (1968)

Gillespie, Tyler and Hall (1969)
Kubo, Senta and Sugimoto (1973)
Bergsten (1974)

Hillidge and Lees (1975)

Muir, Skarda and Milne (1976)

Thermodilution Technique
Muir, Skarda and Milne (1976)

75.0

75.0
90.0
97.5

82.7
82.7
56.0
76.0
88.3
73.9

72.2

Cardiac output
ml/kg/min

(ponies)

5

many different types of thermodilution techniques and one had
not been previously described for use in horses, five differ-
ent volumes of injectate at two different temperatures were
studied. Results obtained by injecting 40 ml of 0°C 5%
dextrose in water into the right atrium were comparable to
results obtained from the dye dilution studies (Table 1).
Since the report of these results, other studies using the
thermodilution technique for cardiac output determination
have been done (Swanson, et al., 1985; Trim, et al., 1985).
Today, only thermodilution and dye dilution methods are
commonly used when measurement of cardiac output in horses
is required.
2. Application of the Fick Principle for

Estimation of Cardiac Output

In 1870, Adolf Fick (Wilson and Gibson, 1978) related

cardiac output (CO), tissue oxygen consumption (902), and
arteriovenous oxygen content difference (A-VOZ) in a formula,
today known as the Pick principle.

co=—Y(-)3—
A-VOZ

The A-V02 will change inversely with CO when 902 remains

constant. Unless a patient is hypermetabolic, febrile, or

in shock, 902 will be approximately 145 i 15 mllmin/m2 (Wilson

and Gibson, 1978). Therefore, A-VOZ and variables determining

this difference, such as mixed venous oxygen saturation (5x702)

and mixed venous oxygen tension (Pfioz) have been used to

6
estimate CO or changes in CO. In the next section of this
literature review, I will discuss the use of A-VOZ, $902

and P§02 in the estimation of CO.

a. The Use of A-VOziDifference to Estimate CO

In a normal resting animal A-VOZ is 4-5 ml 02/100 ml
blood. When CO does not meet the demand for 02, this dif-
ference will increase to greater than 6 ml 02/100 ml blood.

In 1978, Wilson et al reported results of a study that
investigated the use of A-VOZ difference to provide an
estimate of cardiac output in 200 critical patients. venous
blood from the superior vena cava rather than true mixed
venous blood from the pulmonary artery was used to measure
the A-V02 difference. Cardiac output was measured using a
dye dilution technique and was calculated using the measured
A-VOZ difference and an average V02 of 135 ml/min/mz.
Calculated and measured cardiac indices were compared and
calculated values were found to be approximately 23% higher
than measured values. Although thisdifference was signifi-
cant, there was a significant (r2 = 0.49) correlation
between the measured and calculated CI values. A-V02 dif-
ference and measured CI values were compared although no
statistical analysis was done. There was an inverse
relationship between A-VOZ difference and CI. Patients
with the lowest A-V02 differences had the highest CI values
and patients with the highest A-V02 differences had the

lowest CI values.

7

When measured CO and A-V02 difference were used to
valculate V02 and when V02 was measured in 42 patients, the
V02 values were found to be quite variable. Values ranged
from 63 i 32 ml OZ/min/m2 in one group to 158 1 51 m1
OZ/min/m2 in another group.

Results from this paper suggest that there is an
inverse relationship between A-V02 and CO, but whether
A-V02 can be used to estimate CO is questionable. For any
A-VOZ there is a wide range of possible CO values associ-
ated with it. Variation in V02 between patients contributes
to the range of values. In patients such as these with
cardiovascular instabilities and systemic disease, A-VOz
cannot accurately and consistently reflect CO without

simultaneous V02 measurement.

b. The Use of SVOZ to Estimate CO

 

Historically, as part of the method for Fick determina-
tion of cardiac output, 8602 has been recorded and noted to
change as a function of CO. In a study of 36 critical
patients, Cournand et al, in 1943, showed that SV02 de-
creased in proportion to the decrease in cardiac output.

It wasn't until the late 1950 ' s that monitoring SVO2 was pro-
posed asau1alternative method for CO measurement. Various
investigators studied the relationship between $602 and C1
or perfusion index in healthy subjects (Barratt-Boyes and
Wood, 1957) and patients undergoing heart-lung bypass for

cardiac defect repair (Moffitt et a1, 1959, and Boyd et a1,

8

1959). Each investigator noted the importance of V02 and
CaOz remaining constant or near constant in order for SVOZ to
accurately reflect the CO. An important point to note is
that both healthy subjects and surgical patients on heart
lung bypass have fairly stable respiratory and metabolic
functions. This accounts for the predictable relationship
the authors noted between SVOZ and CO. Boyd et al actually
separated his patients into two groups. Group 1 with cardiac
index values above 2 L/min/m2 and SVO2 values above 50% , and
group 2with CI and $902 below 2 L/min/m2 and 50%, respec-
tively. There was a 66% mortality (10/15) in group 2 as
compared to an 11% mortality (2/19) in group 1. This study
(Boyd et a1, 1959) demonstrated the importance of monitor-
ing either CO or $902 after Open heart surgery to prevent
the persistent low outflow state that precedes cardiac arrest.

As the use of $902 to reflect CO became more common,
physicians began recognizing the limitation of this relation-
shipixlcritical patients with inadequate myocardial function
(Valentine et a1, 1966; Goldman et al, 1967; Goldman et a1,
1968). In a majority of patients SVOZ and CI were lowered
but concurrent respiratory disease frequently interfered with
the linear relationship between 8‘702 and CI that had previously
been described in healthy patients. The monitoring of SVO2
did not lose its importance at this time; rather, it began
to be considered an important early indicator of the deteri-
oration in oxygen delivery (Cao2 and CO) to the tissues. The

patient's clinical course correlated with 5602, and changes

9

in Sfioz were shown to be early indicators of changes in clinical
status or response to therapy (Goldman et al, 1967; Goldman
et a1, 1968; Kazarian and Del Guercio, 1980; de la Rocha et a1,
1978).

In contrast to these results, Muir et al in 1970 described
results of a study also designed to assess the use of $702 as
a measurement of CO in patients with varying degrees of acute
myocardial infarction. Indicator dilution measurements of CO
were made in 26 patients. Mixed venous blood gas samples were
taken just prior to determination of CO. Mixed venous oxygen
tension was measured directly and 5602 was derived from an
oxyhemoglobin dissociation curve. There was a significant
correlation (r = 0.74) between S§02 and CI. In individual
(patients, changes in SVO2 appeared to correlate well with
changes in CO. As was described in previous studies (Boyd
et a1, 1959; Goldman et a1, 1968), 8602 also appeared to
relate to the clinical status of the patient. In order to
explain the direct relationship between Sfioz and CO, Muir
et a1 assumed that V02 remained constant in patients with
myocardial infarction even during low output states and that
Saoz remained constant unless severe respiratory disease was
present. These assumptions are acceptable in cases of uncom-
plicated myocardial infarction or left ventricular failure
unless backward failure results in pulmonary edema. In
cases where respiratory disease is present or V02 is decreased
as may occur in cases with backward failure and shock, other

variables affecting 8902 must be evaluated before attributing

10
the decrease in $602 to inadequate CO alone (Hutter and Moss,
1970). The results of the study by Muir et al showed a large
variation in the CI values for a given SVOZ so a single measure-
ment of $602 was not shown to be useful in actual calculation
of CO. As has been suggested by other investigators (Butter
and Moss, 1970; Lee et a1, 1972), several measurements of
SVOz would give a more accurate indication of changes in CO
in response to therapeutic agents or due to depression of
myocardial function, a decreased preload, or an increased
afterload.

In 1978, de la Rocha et al published results of a study
done in children and infants to determine the relationship
between $602 and cardiac index. Eleven patients undergoing
surgery for congenital cardiac defects were monitored pre-
operatively, intraoperatively and post0peratively. Variables
monitored included 5902, cardiac index and various systemic,
pulmonary and cardiac pressures. Linear regression analysis
was done to evaluate the relationship between $902 and cardiac
index. A significant correlation was found (r =0.78, p <.001) .
All SVOZ values less than 65% corresponded to a cardiac index
of less than 2.5 L/min/m2 except in one case. Patients with
low cardiac output syndromes postoperatively (n=3) were treated
with positive inotropes. This resulted in an increase in both
8902 and cardiac index (r'==0.86,p)<.05). Patients with uncom-
plicated recoveries had 8602 values greater than 65% and normal
cardiac index values. The one exception to this was a patient

with a normal cardiac index with a Svoz lower than 65%. This

11
patient was also pyrexic. Oxygen consumption increases
10-13% for each degree Celsius elevation in body temperature
above 37°C and may be responsible in this case for lowering
SVOZ.

This study confirmed a significant correlation between
cardiac index and 8902. These results supported the results
of studies by Moffitt et a1 and Boyd et al in 1959 showing
that surgical patients appeared to be good cases to use SVOZ
for evaluation of CO. Generally, their metabolic rate is
stable which minimizes problems due to changing V02. This
is in contrast to septic shock patients where higher metabolic
rates and lower V02 occurs resulting in a decreased A-VOZ
(Duff et a1, 1969). Despite the significant relationship
between CI and $902, a specific $602 measurement had a range
of CI values associated with it and an increase of $602 less
than 10% did not associate well with a similar rise in CI
(de la Rocha, 1978). The results Of this study suggest that
an 8602 value less than 65% warrants further investigation
and in surgical patients with no preexisting metabolic or
respiratory disease emphasis can be placed on the patient's
cardiovascular function.

Investigations between 1959 and 1970 of the use of $602
as an indicator of cardiac output confronted the following
problems which precluded its wide acceptance into postoper-
ative and critical care. (1) Single 5602 measurements were
difficult to interpret because of the large variability in

cardiac index values for a specific 5602. (2) Changes in

12

cardiac output and SVO2 occurred rapidly, and single or inter-
mittent measurements were not adequate for monitoring of
unstable patients. (3) The number of samples taken could be
influenced by patient blood volume and hemoglobin concentra-
tion and (4) in vitro measurement of these samples was time
consuming.

The first report of continuous SVOZ measurement was made
in 1962 by McArthur et al. This study described the use of
a system whereby small amounts of mixed venous blood (18 ml
blood/hour) were pumped continuously from the pulmonary
artery to an anaerobic mixing cuvette equipped with an oxygen
electrode for measurement of 8602. It wasn't until 1972 that
a method for continuous in vivo measurement of $602 was
described. This technique involved the use of a fiberOptic
catheter oximeter which used alternating pulses of light at
either 670 and 950 or 660 and 805 nm wavelengths. Reflected
light was transmitted by a second set of fibers to a photo-
electric device outside the patient. The ratio of intensities
of the back scattered light of the two wavelengths was used to
determine oxygen saturation. Investigators (Martin et a1,
1973) showed a significant correlation (r = 0.978) between
simultaneously measured in vitro oximeter and fiberOptic
oximeter 8902 values.

In 1975 Krauss et al reported results of a study
describing the use of this catheter to measure $902 in
patients after cardiothoracic surgery. The intent of this

project was to determine the predictive value of 8602 during

13
the postoperative period after thoracotomy. The value of
8602 as a predictor of cardiac output was also investigated.
Fiberoptic catheters for continuous SVOZ measurement were
placed in 19 patients either prior to (12) or after (7)
surgery. Other variables monitored were cardiac index,
arterial blood gases and pH, systemic and various cardiac
pressures, and urine production. A correlation was found
between cardiac index and S§02 (r = 0.78, n = 28, 0.64 < p
<0.92) but only if 8902 were stable (change in $902 Of <3%)
for at least five minutes before and after the measurement
of CI. When CI was less than 2.0 L/min/mz, $902 was less
than 60% in six out of eight patients. When CI was less than
2.5 L/min/mz, all but one patient had an SVO2 less than 65%.
Urine production less than 20 ml/min, systemic blood pressure
under 90/60 mmHg or other clinical signs of shock were pre-
ceded by a fall in SVOZ of more than 5% or to values less
than 60%. In patients that experienced respiratory distress,
there was a simultaneous fall in 8602.

The results of this study support the contention that a
changing SVO2 reflects changes in variables determining this
value, namely cardiac output, SaOz, hemoglobin and V02. Try-
ing to evaluate Svo2 as an indicator of CO is impossible in
patients with potential for respiratory dysfunction or meta-
bolic instabilities. 8902 was shown in this study to change
prior to obvious changes in other monitoring variables such
as heart rate and blood pressure. Because of this, it may

be useful as an early indicator of cardiovascular or

14
respiratory dysfunction. This study supported the use of
continuous monitoring of SVOZ as a valuable monitoring tool
in the care of critical postsurgical patients. These con-
clusions were corroborated by results of similar studies by
Martin et a1, 1973, and McArthur et a1, 1962.

In 1978, a new system for reflection spectrOphotometry
was developed using light pulses of three different wave-
lengths between 600 and 1000 nm (Wilkinson et a1, 1979).
Reflected light is transmitted to a photodetector to measure
blood reflectance and to compute 8602 from the relative
intensities corresponding to the three different wave lengths.
A digital readout shows the average value for the preceding
five seconds and is updated every second. Although it is
more difficult to use (Schmidt and Staff, 1981), this catheter
is more accurate than the older one because three different
light wavelengths are used.

In 1982, Jamieson et al used this fiberOptic catheter
to compare hemodynamic parameters with 5902 in eighty-four
patients at six predetermined periods between preparation
for induction and recovery. Hemodynamic assessment included
measurement Of heart rate, arterial blood gases, systemic
blood pressure, right atrial (CVP), left atrial (LAP) and
pulmonary capillary wedge (PCWP) pressures, temperature and
cardiac output. Calculations of cardiac index, systemic (SVR)
and pulmonary (PVR) vascular resistances and left ventricular
stroke work index were then done. Observations were made in

the postoperative period to assess effects of endotracheal

15
tube suctioning, positioning, shivering, and chest physio-
therapy on the measured variables. During surgery a
decreased SVO2 was treated with inotropic agents and vaso-
pressors or vasodilators. Interpretation of therapeutic
needs based on 8902 correlated well with other hemodynamic
measurements. Changes in 8902 of more than 10% often pre-
ceded changes in the hemodynamic status of the patient.
When 8902 began to fall, the patient was evaluated for a
decrease in preload (CVP, PCWP, LAP), contractility (LV
stroke index) and an increase in afterload (SVR, PVR) and
treated appropriately until SVO2 improved. Shivering,
endotracheal tube suctioning, position changes and chest
physiotherapy markedly decreased 5702' $902 improved when
manipulation of the patient and/or shivering ceased.

This study confirmed the importance of $602 as an early
indicator of adequate tissue perfusion. In patients with
stable V02, hemoglobin concentration and respiratory func-
tion, SVOZ was also used to estimate cardiac output but
5702 and cardiac output had no "fixed" relationship.

In 1982, two studies evaluating technical problems
associated with placement of a new fiberOptic catheter were
published. This catheter had both fiberOptic SVOZ measure-
ment capabilities and the capacity for thermodilution
cardiac output measurement. The study by Baele et al used
critically ill adult patients requiring hemodynamic moni-
toring. Waller et a1 placed these catheters in patients

undergoing cardiac surgery. Measurements made included

16
in vivo fiberOptic measurement of $902, in vitro oximeter
measurement of $602, and cardiac output. Patients in both
studies were breathing an oxygen enriched gas. Baele et al
found a significant correlation between in vitro and in vivo
measurement of 5602 (r = 0.95). Waller et al found the
correlation to be significant as well (r = 0.92, n = 99).
This was true for a range of 8602 values from 60-95% measured
with hemotocrits of 20-42%.

Waller et al found a significant correlation between
$902 and hemodynamic changes. An increase or decrease in
Sfioz >5% correlated with a change in cardiac index (r = 0.69),
stroke index (r = 0.67), and left ventricular stroke work
index (r = 0.58). There were 39 instances when SVOZ values
changed by 5% or more. Eighteen of 21 decreases in 8902
were accompanied by corresponding decreases in cardiac index
and 14 of 18 8902 increases were accompanied by an increase
in cardiac index. In vivo 8902 values did not correlate
significantly with changes in mean arterial pressure, heart
rate, pulmonary capillary wedge pressure or systemic vascular
resistance. Baele et al showed by multivariant analysis that
hemoglobin concentration, body temperature and cardiac index
did not affect the accuracy of in vivo $602 measurement.

Both of these studies support the hypothesis that SGOZ
can be determined accurately in vivo by fiberOptic catheter
measurement. 8‘702 correlated with CO but the relationship
was not strong enough to define actual cardiac output values

from SVOZ measurement. However, if 8602 did decrease by 5%

17
or more, Waller et al showed there was an 86% chance that
cardiac index had also decreased. Because SVO2 is continu-
ously measured and cardiac output is not, 8602 can serve
as an early indicator of a diminished cardiac output. SVO2
also reflects changes in other important variables and may
be more useful than any one of the other variables that
determine $902 in critical patient care.

A more recent study using the fiberOptic continuous $602
measurement was by Schmidt et a1, 1984. This study compared
continuous SVO2 measurement with intermittently measured and
calculated oxygen transport variables in the preoperative
period of twenty adult cardiac surgery patients. All of
these patients had impaired left ventricular function.
Measurements made included continuous 8902, systemic blood
pressure, cardiac filling pressures and heart rate. Inter-
mittent arterial blood gases and cardiac output measurements
were made. Calculations included A-VOz, oxygen delivery
(D02 = C0 x Ca02), V02 and oxygen extraction ratio (OER =
VOZIDOZ). There was a significant negative correlation
(r = -0.84) between SVOz and percentage of 02 extracted from
the blood (OER). No other oxygen transport variables includ-
ing cardiac index showed significant correlation with SVOZ.
Changes in $902 on occasion preceded changes in cardiac out-
put but there was a large variation in the amount of change.
At the onset of anesthesia, Sfioz actually increased 15%

and cardiac output decreased.

18
As was described by Schmidt et al, the relationship of
SVOZ to OER is more obvious if the equation OER = VOZ/Doz is
simplified

CO (Ca02 - €602)

 

 

OER = C0 x CaO2
C902
OER = 1 - Caoz

since the majority of the oxygen in whole blood is bound to
hemoglobin Ca02 and C§02 can be simplified to Sa02 and $902.
SaOz is usually about 100%, so

OER = 1 - SVOZ
Using $902 to estimate OER is less variable than using SVO
to estimate CO and OER is more significant than CO in assess-
ing tissue oxygenation. In a patient with normal V02 a
decrease in CO can cause a decreased 8602. When V02 increases
and CO remains constant, 5902 will also decrease. Schmidt
et al claim that the cause of the decrease in SVOZ is not as
important as the level to which it decreases. Arterial hyp-
oxemia is just as significant a cause of tissue hypoxia as
is a decreased CO. It is difficult to use 8502 as an estimate
of CO since a multifactorial relationship exists. Other
variables (V02 and SaOz) involved in this relationship may
change at any time in critical care patients. Actual CO is
not as relevant in patient care as is adequate tissue oxygen-
ation. The reason emphasis is placed on maintaining or
increasing CO is to assure tissues have enough oxygen to

survive. The Pick Principle has been used widely in the past

19

to estimate cardiac output because CO has been so difficult
to measure. Rapid changes in C0 are difficult to detect
because indicator dilution measurement of CO has time and
indicator volume limits. If 8902 were difficult to measure
and CO was easy, CO measurement would likely be investigated
as a way to estimate SVOZ. 3902 reflects significant changes
in CO, V02 and arterial oxygenation and is an extremely use-
ful monitoring variable in critical patient care.

The purpose of S902 monitoring has evolved over the last
30 years. Initially it was monitored to reflect CO, a value
rarely measured clinically because indicator dilution tech-
niques were difficult to apply. Today 8902 is thought of as
a variable which if monitored continuously will reflect early
changes in a patient's tissue oxygenation which is a function
of CO, systemic vascular resistance, pulmonary gas exchange
and hemoglobin concentration. Svoz is believed to be more
sensitive and a better prognostic indicator than any other
single variable monitored in critical care patients (Schmidt
et a1, 1984). $902 is also still useful in estimation of
CO in patients with stable respiratory and metabolic functions
(Sottile et a1, 1982; Magilligan et a1, 1983; Sheldon et al,

1983).

c. The Use of P902 to Estimate CO

 

One other way to evaluate CO is by P602 measurement.

Pfioz is an indicator of tissue oxygenation and describes the

20
driving force moving oxygen from the blood to the tissues.
Diffusion of oxygen is entirely dependent on P02 and not on
the total amount of oxygen present. The speed and distance
over which diffusion occurs depends on a partial pressure
gradient. Oxygen dissociates from hemoglobin based on the
binding affinity of hemoglobin for oxygen. This affinity
is described by the oxyhemoglobin dissociation curve. Carbon
dioxide tension, pH, temperature and concentration of 2,3 DPG
inside the red blood cell determine the slope of this curve.
At a P02 of 40 mmHg approximately 75% of the hemoglobin is
bound. Between a P02 of 70 and 20 mmHg, there is a linear
relationship between $02 and P02. Over this portion of the
curve P02 determines $02. P902 is usually maintained between
36-40 mmHg. If this value decreases below 30 mmHg compen-
satory mechanisms come into play. These mechanisms include
a release of catecholamines which increase CO and P902.
When P602 decreases it is because the circulatory system is
failing to keep pace with the tissues' oxygen demands (Miller,
1982; Finch and Lenfant, 1972; Snyder and Carroll, 1982).

The first investigators to propose that PVOZ was a use—
ful indicator of the adequacy of tissue blood flow was
Stanley and Isern-Amaral in 1974. Their conclusion was
supported only by the fact that in patients where P992 was
kept between 38-42 mmHg during bypass there was a signifi-
cantly greater urine output and significantly lower

bicarbonate requirements than in patients perfused at a

21

fixed rate of 40-60 ml/kg/min. Statistical evidence to
support the use of P902 in estimation of CO was provided
later by Parr et al in 1975 and Kohanna et al in 1981. In
a study of 139 children less than two years of age and after
cardiac surgery, Parr et al showed that there was a statis-
tically significant correlation between CI and PVOZ. The
results also showed that acute cardiac death was more
reliably predicted using CI and PVOZ than either one alone.

Kohanna et al also discussed results supporting a
statistically significant correlation between PVC2 and CI.
This was shown in a study of 25 patients after cardiac oper-
ations. The thermodilution CO measurements were made and
compared to PVOZ, P302: PVCOz' Pacoz, venous and arterial
pH, packed red cell volume, temperature, left atrial preSe
sure, central venous pressure, urine output, heart rate, and
mean arterial pressure. Only P602, PVCOZ, PacoZ, venous and
arterial pH, and temperature had significant r values
(p < 0.01) and the highest r value was found with the cor-
relation between CI and P902 (r = 0.49). The authors
(Kohanna et al and Parr et a1) state that the correlation
between PVC2 and CI were too low to be relied on clinically
and that measurement of CO was the only accurate way to deter-
mine adequate cardiac function. As an alternate use for
P902 Kasnitz et al evaluated its use as a prognostic indi-
cator for critically ill patients with circulatory and/or
respiratory abnormalities. Results of the study showed

that all of the nine patients with P902 values less than,

22
28 mmHg died, whereas only two of the eleven patients with
P702 values greater than 28 mmHg died. They also showed
that measurement (If Pfioz alone was a better prognostic
indicator than measurement (If CO alone in critically ill
patients.

By 1978, continuous monitoring of 8602 had been reported
and shown to be superior to intermittent Sfio2 monitoring.
In an effort to apply this technique to P602 monitoring,
Armstrong et al described the use of continuous P602 measure-
ment in monitoring twenty-five patients with cardiorespi-
ratory disorders. A double lumen catheter was used with a
miniature polarographic electrode mounted on the end.
During periods of cardiorespiratory stability, the mean
PVC2 was 43,4 1 0.3 mmHg. Decreases in P602 below 40 mmHg
occurred in patients experiencing a reduction in F102!
acute respiratory failure, hypoventilation, hypovolemia
and cardiac arrhythmias. In four of five cases, the
decrease in PVOZ preceded clinical signs of deterioration.

It is clear from the results discussed that P602 cannot
be used for estimation of CO in patients with unstable
respiratory or metabolic function. When respiratory and
metabolic function are stable, then, as was demonstrated
with 8902 (Scottile et a1, Magilligan et a1, Sheldon et a1,
Boyd et a1), P602 is useful for estimation of CO.

As pointed out earlier, a decrease in CO is a common

cause of life threatening complications in horses. Because

23
metabolic and respiratory function are likely to be stable
in anesthetized horses, I hypothesized that P602 is a
useful indicator of CO in anesthetized horses. Further—
more, in order to enhance the clinical applicability of
this research, the relationship between onz and CO was
studied with horses in dorsal, lateral, and sternal
recumbencies and in horses with a metabolic disturbance

induced by Escherichia coli endotoxin administration.

MATERIALS AND METHODS

Horses - Six healthy mixed breed horses (9.8 i 3.6 years
of age, weighing 466.2 I 18.5 kg [2 1 SEM]) were used.
Measurement techniques - For measurement of CO, a 110-cm
7-F balloon-tipped catheter with a thermistor near the tip3
was percutaneously placed in the pulmonary artery via the
jugular vein. A polyethylene catheter (inside diameter
1.67 mm, outside diameter 2.42 mm)b was percutaneously placed
in the right atrium also via the jugular vein. Catheters
were attached to pressure transducers,°and an oscillosc0ped
was used to verify placement by the presence of character-
istic pressures and wave form. The zero pressure reference
point was considered to be at the level of the sternum for
horses in lateral recumbency and at the point of the shoulder
for horses in sternal and dorsal recumbency. To measure CO,
40 ml of 0°C 5% dextrose in water was injected in <2 3 into
the right atrium using a pressure injectore at 27.2 atmo-
spheres. Falsely increased CO data were obtained when
indicator solution was allowed to warm in the atrial
catheter before making CO measurements. To prevent this,
the atrial catheter was filled with cold 5% dextrose in
water just before CO measurement. Injections of indicator

solution were made during exhalation at the onset of a QRS

24

25
complex. Cardiac output was calculated using a CO computer.a
Each data point was the mean of at least three CO deter-
minations.

Oxygen consumption was measured volumetrically (Brody,
1945). Briefly, at end exhalation, the endotracheal tubef
was connected through a three-way valveg to a 120-L spirom-
eterg containing a known volume of 100% 02 (Figure 1). The
horse was allowed to breathe from the spirometer through a
unidirectional Y piece for five minutes. Carbon dioxide was
removed from the circle system using a C02 absorbant.h At
the end of exhalation after approximately five minutes of
breathing through the spirometer, the three-way valve was
switched to allow the horse to breathe through the demand
valve.1 The change in gas volume in the spirometer was
measured, and a calculation of V02 was made. Values were
corrected to STPD using standard equations.

Blood samples were collected anaerobically from the
facial artery, pulmonary artery, jugular vein, and a super-
ficial limb vein (median or medial saphenous). Blood gas
tensions were determined using a blood gas analyzer.J Mean
arterial blood pressure was measured using a facial artery
catheterk and pressure transducer connected to a multichannel
physiograph. Heart rate was measured using a bipolar ECG
lead system.d From these values we calculated stroke volume
(SV), (SV = CO/HR) and peripheral vascular resistance (PVR)

PVR = MAP/CO

26

Figure 1.

Equipment used for measurement

of oxygen consumption.

27

 
 
 
    

 

 

 

 

 

SPIROMETER

11m (UHF?!

 

 

 

 

 

 

.. ..: coz ABSORBER

 

 

\/p .
DEMAND \UNIDIRECTIONAL Y

02 VALVE , PIECE
3WAY VALVE

  

 

 

28
where MAP is mean arterial blood pressure. Cardiac index
was calculated using the equations (Henness et al, 1977)
C1 = CO/body surface area
where
body surface area (m2) = (10.1 x W2/3)/104
and W is body weight in grams.

Experimental design - Horses were fasted for 12 hours
before anesthetic induction. After placement.of catheterS‘
in the pulmonary artery, right atrium and jugular vein,
anesthesia was induced with 1 mg of xylazineI/kg of body
weight IV and 2 mg of ketaminem/kg, IV. A surgical anes-
thetic plane was maintained with a mixture of 10%
guaifenesinn and 0.4% thiamylalo for 4.4 i 0.1 hours.

Horses ventilated spontaneously through an endotracheal

tube and a demand valve that delivered 100% 02. Horses

were placed in lateral, sternal, or dorsal recumbency as
predetermined by two 3 x 3 latin square designs (Table 2).

To achieve a low CO, xylazine (0.65 to 1 mg/kg, IV) was
administered. At least five minutes after xylazine adminis-
tration, CO and then V02 were measured. During the V02
measurement, blood samples were taken for 02 and C02 tension
measurements. Heart rate (HR) and MAP were also recorded
during the V02 measurement period. Subsequently, to increase

P

CO, dobutamine in 5% dextrose and water was administered by

an infusion pumpq at 1 to 7 ug/kg/min. In four instances during
dobutamine administration, a persistent second degree atrio-

ventricular (AV) block was observed. Because this arrhythmia

29

Table 2.

3 x 3 latin square design describing

random sequencing of recumbencies.

 

Recumbencies
First Second Third
Lateral Sternal Dorsal
Lateral Dorsal Sternal
Sternal Dorsal Lateral
Sternal Lateral Dorsal
Dorsal Lateral Sternal
Dorsal Sternal Lateral

30

interferes with indicator dilution CO measurement, 3 to 5 mg
of atropine sulfate was administered IV. After at least five
minutes of dobutamine administration, all measurements were
repeated in the sequence described. The horse was then moved
to the next position, and the series of measurements was
repeated as described. Steady state was determined through-
out each measurement period by constant monitoring of MAP
and HR. Repeated measurements of CO and.VOz were made
periodically at the end of the measurement periods and were
never significantly different from earlier CO and V02 measure-
ments made during that period.

To determine if the relationship between CO and P602
applied to horses with endotoxemia, horses were placed in
dorsal recumbency after the last measurement period and

n n o o r I o I
Escherichia c011 endotox1n was administered. An endotox1n

 

dose of 100 ng/kg was suspended in 60 ml of 0.9% NaCl and
was administered IV over a period of five minutes. Fifteen
minutes after endotoxin administration, measurements were
repeated.

Statistical analysis — The effect of position on all
variables was analyzed using a two-way analysis of variance.
If a significant F value were found, Tukey's w test was used
to compare means. The effect of dobutamine and endotoxin
administration was analyzed using paired t tests. Linear
regression analysis was used to relate CI and P602, HR, MAP,

jugular venous oxygen tension (on2 JUG) and superficial limb

31
vein 02 tension (PVO2 PERIPH). Significance was determined
by calculation of a t value for the regression coefficient (b).

Significance was set at P < 0.05.

RESULTS

Positioning had an effect on C0, CI, SV, PVR, MAP, and
Pa02 (Table 3). When compared with values in horses in lateral
recumbency, after xylazine administration, CO, CI and SV were
lower in horses in sternal recumbency whereas PVR was higher
in horses in sternal recumbency. In horses in dorsal recum-
bency, after xylazine administration, MAP, PVR and Pa02 were
significantly lower and CO and CI were significantly higher
than those in horses in sternal recumbency but not those in
horses in lateral recumbency.

In horses in all positions, dobutamine increased CO, CI,
HR, MAP, Pfioz, on2 JUG, and PV02 PERIPH (Table 3) but the
increase in HR was not significant in horses in lateral
recumbency. Dobutamine also increased V02 but the increase
was only significant in horses in sternal recumbency. In
horses in sternal recumbency, dobutamine increased SV and
decreased PVR. In horses in dorsal recumbency, dobutamine
significantly increased PaOZ. In horses in lateral recum-
bency, dobutamine significantly increased PaCOZ-

When compared with data obtained after xylazine adminis-
tration with horses in dorsal recumbency, endotoxin administra-
tion increased HR and decreased SV and PVO2 PERIPH. Other

variables were not changed significantly.

32

.cova-Luu.cvsco oc.~¢.»x sou»: mucoosauou pam.ov sou» Amo.o v av acotoeevu x—Acourcwcowm . A
.coAuALuA_c.sua ocvnapxx Lune. successunt potoum— sot» Amo.o v av acoLoAAAv xpucuuAAAcOAm A A

.33 2:22.. .8: Gas v 3 “carat... 2:81:83 . .
.cOAAcAA ~o a:0co> Lopsoan - gnu ~o>a icoAAcAA No

maoco> BEA. po.u.Atoa:A A :QHaua ~o>a acovmcou ~ou FAALAAL- u ~ou-a “co.AcAu ~o mango» vex_s . ~ooa acoAAcou ~c FAALAALA A ~o~g aco_uasamcou No . Nos
"Aug-AAAAAL sa—auuo> putosnvtoa

- ¢>n "usaunoua vac—a pAALoALA can: a a<t “AA-L About u a: noes—o> oxoguu - >m "nova. uupcgou u —u uuaauao OAADLAO A 09

 

14
DJ

 

 

A.v A m.om Ao.~ A ~.so n.n A a.nm ov.- A m.o- a.» A o.om co.o A ~.o~ c.n A o.~n “a: .0 say and ~o>a
c—.m A o.mn on.- A «.00 o.m A n.om co.m— A o.e- ~.~ A o.mm ou.o~ A a.onn a.m~ A n.~c— an: .0 sew xa_¢ua ~o>a
oé A m.~m m.~ A «.3 «J A ado vim A win ué A adv. c~.— A v.3 A; A n.3,. 3: Co 55 ~83
o.~ A o.nv o~.n A o.oo ~.~ A n.mv Ao.n~ A o.su n.~ A 9.»: c~.¢ A o.oo ~.~ A u.mv “a: A0 gas ~ooa
To... A m6: on;— A m.~n~ flow A 9.02 0.: A 9.3 +0.3 A .63 ~6~ A n63 can A mdow 3: Ac 55 3:.
o~.o A mo.~ -.o A c~.~ -.o A oo._ c-.o A un.~ o~.o A Aw.~ c~.c A ——.~ n~.o A as._ Ac_s\ug\—Ev ~oo
36 A 3..” 9A.... A 86 on... A 3.... .8... A 3.. t8.— A a; on... A 3." 36 A 3.... €2.33: .0 :5 a:
n.o_ A n.om cm.—~ A n.o~— o.s A o.- oo.n A n.om~ Aw.s A n.c- on.- A o.cn~ m.» A ~.~c Au: co gr; EA:
Ao.c A 5.5m on.“ A m.om n.n A c.»« co.» A u.nm —.~ A s.nn v.- A n.oo o.~ A ~.~n Acwsxmuooav a:
Ao.mm A ~.mwe «.sou A n.-o n.~a A m.~v~ o~.¢o A o.¢oo Ao.ms A ~.ovm A.~o A _.omm ~.c- A o.oMm Ape. >m
oo.o A am.v Aco.o A -.n mc.o A co.v coo.~ A oo.n ++cc.o A oo.n ovm.c A on.“ An.o A om.c A~E\cws\4. .u
n.~ A m.om A~.~. A v.co o.m A m.oo An.n~ A o.~o ++~.m A ~.on Am.- A o.m¢ n.e A o.vo AcAE\m4\_EV ou
cAAvoucu oc_eAuanoo Acm~APMM‘ ocssmuaooo ocANApaA AcAEaAaoom acu~a+mx acmeotamcmx
pomtoo postman putmuoa

 

.Aomgog uo~_uozumoco x,» c_ on\a: ooflv co_bmtum,cAEum

cwonoucu LAAAA Auuacvs ma uco Acts\mx\o: A cu ~V cowuutumchEUA ocAEAASBou Levee vca onxus g o» mo.ov ocvumpxx touts mucocmtamome topaumququLuu

.n spank

34

Correlations between CI and HR, MAP, PVOZ , PVC2 JUG,
and PV02 PERIPH were statistically significant (Figures 2-6,
Tables 4 and 5). There was a significant relationship
between CI and either P902 or MAP in all positions and
after endotoxin administration. In horses in sternal
recumbency, on2 JUG and PVOZ PERIPH did not significantly
correlate with C1. In horses in lateral recumbency, the
correlation between PVO2 PERIPH and CI was also not sig-
nificant. There was a significant correlation between
HR and CI in horses in all positions, but not after endo-

toxin administration (Table 5).

35

Figure 2.

The relationship between mixed venous 02
tension (PV02) and cardiac index (CI).
Measurements made in six horses in lateral,
sternal, and dorsal recumbencies and in
dorsal recumbency after endotoxin admin-
istration are shown on upper panels;
cumulative data are shown in lower panel.
Regression coefficient (b) t SE of the
regression coefficient. All b and R
values are significant. ( A = xylazine,

O = dobutamine, [] = endotoxin.)

36

 

 

 

 

LATERAL [501: STERNléL
IOO~ IOO‘F
0 o o
75 ~ 0 75 ~
O
50- 50~ A
A AA A
25 - b=4.99+0.98 25 » b=946+238
*R = 0.85 *R=0.78
I I I l l l
2 6 I0 2 6 I0
RSA R 0
loop 00 L 'OOI' DO SALAN ENDOTDXIN

   

 

 

 

 

A
a; 25 - b: 5.832.096 25 b= 5.742090
:1: «R4139 *R=O.85
E l l l I l I
v 2 6 IO 2 6 IO
N ISO r-
.9 ° .
0.
I25 ~
IOO -

    
 

 

 

75 L
50 -
b=6.47.t HS
0
o *R= 0.66
25 ~
0 1 l I 1 I
2.0 40 6.0 8.0 I00

Cardiac Index (llmin/m2)

37

Figure 3.

The relationship between jugular venous 02
tension (PVO2 JUG) and cardiac index (CI).
Measurements made in Six horses in lateral,
sternal, and dorsal recumbencies and in
dorsal recumbency after endotoxin admin-
istration are shown on upper panels;
cumulative data are shown in lower panel.
Regression coefficient (b) t SE of the
regression coefficient. Significant b and
R values are indicated (*). ( A = xylazine,

O = dobutamine, E) = endotoxin.)

PV02 Jugular (mmHg)

38

 

 

 

 

 

 
 
  

    

 

 

 

 

  
    

 

 

LATERAL 200 _ O STERNAL> 0
HQ;
I00 — I00 ‘- o
75" 75 _.
A O
50 ~ 50 ~ A A o
A
b=6.26::l.34 b=9.89 14.6l
25 » *R=O.63 25 ~ R=O.57
4 I l J L I
2 6 IO 2 6 l0
0 XINO
'00 _ loo _ DORSAL mg ENDOTO
O
75 — 75 -
50 50 -
b=7.732.l96 ‘3 b=7.73il.66
25 _ *R-OJe 25 - 'R=0.76
I I I l 1 1
2 6 IO 2 6 I0
250 -
200 -
O
O
ISOr
O
O
I OO~
_ 0 ca, b=742 L l.83
50 ° *R=.O.54
O
O l I I l l
2.0 4.0 6.0 8.0 I0.0

Cardiac Index (I/min/mz)

39

Figure 4.

The relationship between peripheral venous Oz
tension (PV02 PERIPH) and cardiac index (CI).
Measurements in six horses in lateral, sternal,
and dorsal recumbencies and in dorsal recumbency
after endotoxin administration are shown on
upper panels; cumulative data are shown in lower
panel. Regression coefficient (b) t SE of the
regression coefficient. Significant b and R
values are indicated (*). ( A = xylazine,

O = dobutamine, C] = endotoxin.)

40

 

 

 

 

 

   

  
 
 

 

 

 

 

LATERAL STERNAL
O
A
200 ' 200 r
'50 ” I50 r
:9
O
'00 '- I00 I-
A O
A E A
50* b=I3.I3:6.I2 50 AB A b=9.2O:42I
R=O.46 R=O.57
l 1 l 1 J l
2 6 IO 2 6 l0
200 — OORSAL 200 - DORSAL AND ENDOTOXIN
I50 ~ I50 -
O O
A '00 " '00"
:‘n”
A
E 50 ~ 50 — A
E flII=95311253 A fl6:951: 2.I2
V J R=077 D R=0.75
" TL___.L __.5___.I% J
g 2 6 IO 6 l0
5 .-
2 250 o
I—
O
0' O
8‘
> 200 -
n.
O
0
I50 r o

I00 F

     
 

 

 

oO o b= l0.90*.3.0I
*R=O.5O
O
50 I. C!)
O O
O 1 1 _l 1 L
20 40 6.0 8.0 IOO

Cardiac Index (I/min/mz)

41

Figure 5.

The relationship between heart rate (HR)
and cardiac index (CI). Measurements in
six horses in lateral, sternal, and dorsal
recumbencies and in dorsal recumbency after
endotoxin administration are shown on upper
panels; cumulative data are shown in lower
panel. Regression coefficient (b) 1 SE

of the regression coefficient. Significant
b and R values are indicated (*).

( A = xylazine, O = dobutamine, D =

endotoxin, bpm = beats per minute.)

42

   
 

     

 

 

 

 

T
'00” LATERAL o '00 _ S ERNAL
O
60 I" 60'-
b=8 481l9| b=4.6610.97
20 - “2:031 20 - *R=O.84
J l l l l l
2 6 IO 2 6 IO
DORSAL DORSAL AND ENDOTOXIN
IOO " IOO _
O O

60

   

   

 

 

 

 

 

 

 

E b=4.9l'_£l80 A0 A b=3.5ItI.69
a 20 __ *R=0.66 20 " [2:045
‘9 4 J l J 1 I
33 2 6 IO 2 6 IO
2
.. IZO I
L.
(U
m
I
IOO )- O
80 -
60 ~
40»
0°
D: 5.20 1 0.90
20 * *R=O.66
0 I I I I I
2.0 4.0 60 8.0 I00

Cardiac Index (Ilmin/mz)

43

Figure 6.

The relationship between mean arterial blood
pressure (MAP) and cardiac index (CI).
Measurements in six horses in lateral, sternal,
and dorsal recumbencies and in dorsal recumbency
after endotoxin administration are shown on
upper panels; cumulative data are shown in lower
ipanel. Regression coefficient (b) 1 SE of the
regression coefficient. All b and R values are
significant. ( A = xylazine, O = dobutamine,

E] = endotoxin. )

Mean Arterial Pressure (mmHg)

200

l20

40

200

l20

40

200

I60

l20

80

40

 

  

44

LATERAL O 200 _ STE RNAL

I20

 

 

 

 

  
  

 

b=I332 :3I9 b=6.0612.40
b ”Fir-080 40 L *R= 0.62

l l l l I 41

2 6 IO 2 6 Io
_ DORSAL 200 .. DORSAL AND ENDOTOXIN

I20 —

   

b=l4.l8 :t 294

 

 

 

,_ 40 _ b=l38| $3.0l
*R=0.84 II R= 075
J I I L_.|_____|.___|__
2 6 IO 2 6 l0
0
o O
0 o O

   

b=9.391' 2.I2

   

 

o o *R=O.57
I— o O
Q)
o o
o
0
J I l I 4
20 4.0 6.0 8.0 I00

Cardiac Index (I/min/m2)

45

Table 4.

The R values for linear regression analysis
of CO or CI and cardiovascular variables.

 

 

Variable CO (ml/kg/min) CI (L/min/mz)
PVOZ 0.66 0.66
HR 0.69 0.68
MAP 0.57 0.57
va2 JUG 0.53 0.54
PVOZ PERIPH 0.49 0.50

 

CO = cardiac output, CI = cardiac index,

P002 = mixed venous 02 tension, HR = heart rate,
MAP = mean arterial blood pressure,

on2

vein and PVO2 PERIPH = 02 tension of blood from

JUG = 02 tension of blood from the jugular

a superficial limb vein.

All R values statistically significant.

46

Table 5.

The R values for linear regression analysis
of cardiac index and cardiovascular variables
in each recumbency.

 

Dorsal and

 

Variable Lateral Sternal Dorsal Endotoxin
on2 0.85* 0.78* 0.89* 0.85*
PV02 JUG 0.83* 0.57 0.78* 0.76*
PVOz PERIPH 0.46 0.57 0.77* 0.75*
HR 0.81* 0.84* 0.663 0.46
MAP 0.80* 0.62* 0.84* 0.75*

 

onz = mixed venous oxygen tension,

on2 JUG = jugular venous oxygen tension,

PVO2 PERIPH = Superficial limb vein oxygen tension,
HR = heart rate, and MAP = mean arterial blood
pressure in different recumbencies and in dorsal

recumbency after endotoxin administration.

* Significant R values.

DISCUSSION

The present study demonstrates that P602 was signifi—
cantly correlated with C1 in anesthetized horses. Moffitt
et al describes a significant correlation between mixed
venous oxygen saturation (8902) and perfusion index
(L/min/mz) in human patients on a cardiopulmonary bypass
machine. The correlation between SVO2 or P902 and CI have
been confirmed in a variety Of clinical settings. (Stanley
and Isern-Amaral, Armstrong et al, Parr et a1, Kohanna et a1,
Dyson et a1, de la Rocha et al, Muir AL et al, and Kasnitz
et a1). When SVO2 decreases to <60% or if Pfioz is decreased
to 30 mmHg, the occurrence of death because of acute cardiac
failure increases significantly (Armstrong et a1; Kasnitz

et al). In most studies, these values of SVOZ and PVC2

correlate with a CI of 2 L/min/m2 (Parr et al; Kasnitz et al).

In the present study, a P902 of 40 mmHg correlated with a
mean CI of 2 L/min/mz.

Xylazine and dobutamine were administered to give a
wide range of CI values over which to test the correlation
between CI and PVOZ. Xylazine was used to decrease CI where—
as dobutamine increased CI. Administration of xylazine IV
induces transient second degree AV blocks and depresses HR
(Kerr et a1). Systemic arterial blood pressure rises

initially, then gradually decreases to less than base line

47

fl .

48
values (Kerr et a1). Peripheral vascular resistance
increases twenty minutes after IM injection Of xylazine
at 2 mg/kg (McCashin and Gabel). In the present study, CI
decreased five to ten minutes after IV xylazine administra-
tion compared with the CI values calculated after dobutamine
administration. Second degree AV blocks did not occur
immediately after xylazine administration.

Dobutamine increases CI and may also increase systemic
arterial blood pressure and the rate of develOpment of left
ventricular pressure (dp/dt); there is no change in total
peripheral vascular resistance (Swanson et al). Effects
on HR are variable depending on the horse and the dose
administered (Swanson et a1). Sinus bradycardia occurs
frequently with infusion rates of 3 to 5 ug of dobutamine/
kg/min and a second degree AV block has been reported in
one horse (Swanson et al). In the present study, dobutamine
consistently increased CI as compared with CI values calcu-
lated after xylazine administration. Second degree AV
blocks occurred in four horses during dobutamine adminis-
tration. Previous xylazine administration may have been
partially responsible for the high occurrence of this
arrhythmia (Kerr et al).

In critical equine colic patients, endotoxemia is
frequently present. After administration of anesthetic
agents, CO may be further depressed (McDonell). In the

present study, endotoxin was administered and the

49
relationship between P902 and CI was evaluated. Endotoxin
administration (200 ug/kg) results in an increased HR and
MAP and a decreased CO fifteen minutes after administration
(Bottoms et al; Burrows). In the present study, endotoxin
increases HR, but did not change CO or MAP. This differ-
ence is likely related to the lower dose of endotoxin.

There is a significant correlation between HR and CO
in resting and exercising horses (r = 0.91) (Bergsten).

In the present study, the correlation between HR and CI
was significant (r = 0.68) and HR of twenty-five beats
per minute correlated with a CI of 2 L/min/mz; however,
endotoxin administration significantly increased HR with—
out changing CI, resulting in an insignificant correla-
tion between HR and CI. This finding may have clinical
importance because many horses with severe gastrointes—
tinal disease, which are anesthetized, have endotoxemia
(Moore).

The P602 was significantly correlated with CI in our
anesthetized horses regardless of positioning. In addi-
tion, the correlation between P902 and CI was unaltered
by endotoxin administration. Whereas, HR and P902 are
significantly correlated with CI in the healthy anesthe-
tized horse, only P602 correlates with CI in anesthetized
horses after endotoxin administration.

THE MAP was also significantly correlated with CI

but the data points were poorly grouped, making it

50
difficult to derive a specific CI value for a measured MAP.
Therefore, MAP was not as good an estimate of CI in anes-
thetized horses as was P602.

Because a catheter must be placed in the pulmonary
artery to collect mixed venous blood, we also evaluated
the correlation between PV02 in the jugular blood or
blood obtained from superficial limb veins and CI. In
horses in lateral and sternal recumbency, on2 PERIPH did
not correlate with C1, whereas in horses in sternal recum-
bency, PVO2 JUG and CI did not correlate. Therefore,

PVO2 PERIPH and PV02 JUG were poor indicators of C0 in
anesthetized horses.

Cardiovascular changes resulting from sternal recum-
bency were noticed. The decreased SV and CO may be
attributed to mechanical interference with venous return
or CO. Sympathetic response to the decreased CO would
account for the increased PVR, although an increased HR
would also be expected. This was not observed.

The V02 in our anesthetized horses ranged from 1.78
1 0.13 to 1.89 1 0.17 ml/kg/min and was unaffected by
positioning. These values are slightly lower than those
predicted for conscious horses (Brody). This slight
difference may be explained by the effects of general
anesthesia (Mikat et a1).

Present C0 values ranged from 39.1 1 5.1 to 64.0 1 4.3

ml/kg/min. These values are comparable with results of

51
studies (Hillidge and Lees) in which CO in halothane-
anesthetized ponies ranged from 51.7 1 8.7 to 68.9 1 6.1
ml/kg/min and CO in anesthetized horses was 46.3 m1/kg/min
(Gillespie et al).

In anesthetized horses, CO is frequently depressed
because of the myocardial depressant effects of anesthetic
agents or because of decreased venous return caused by
positive pressure ventilation (Lumb and Jones). A low
C0 contributes to the severity of alveolar-arterial 02
tension gradients in horses with intrapulmonary shunts
(Gillespie et a1; Kelman et a1) and may have a role in
the pathogenesis of postanesthetic recumbency myOpathy
(Lindsay et al). To prevent these complications, monitor-
ing of C0 and parameters that reflect CO is useful because
it allows for early detection and treatment of low output
states. In clinical practice, measurement of C0 is
difficult; therefore, measurement of P902, which reflects
C0, may be a useful adjunct to the management of anesthe-

tized horses.

CONCLUSIONS

From results presented in this thesis the following

conclusions can be made:

1)

2)

3)

A significant statistical correlation can be made
between 8902 (PV02 or A-VOZ) and C0 because of
the relationship described in the Pick Principle.
This relationship does not allow accurate calcu-
lation of C0 without measurement of all variables
included in this principle. An estimation of CO
from SVOZ, P902 and A-VOz can be made in patients
that are known to have stable metabolic and
respiratory function.

Continuous monitoring is more useful than inter-
mittent monitoring of SVO2 and P602 because it
allows these measurements to be used as early
indicators of changes in cardiorespiratory
function. A decrease in onz or SVO2 indicates
the need for further diagnostics.

There is a statistically significant relationship
between P902 and CI in anesthetized horses regard—
less of position of the horse or the presence of

endotoxin.

52

4)

5)

6)

7)

53
In anesthetized horses V02 does not vary enough
to interfere with the significant correlation
between onz and CI.
Oxygen tension of blood from the jugular vein or
from a superficial limb vein does not consistently
correlate with CI so is not useful in estimation
of CI.
Heart rate does not reflect the CO in anesthetized
horses given a bolus of endotoxin.
Mean arterial blood pressure cannot accurately
be used in estimation of CI because of the wide
scatter of data points around a particular CI

value.

FOOTNOTES

aEdwards Laboratories, Santa Ana, Calif.
bPE240, Intramedic, Clay Adams, Parsippany, NJ.
cP23 series, Gould Inc. Oxnard, Calif.

dSeries 702, Spacelabs Inc. Chatsworth, Calif.
eCordis Corp, Miami, Fla.

f . .

Aire-Cuf, Bivona Inc, Gary, Ind.

8Warren E. Collins Inc, Braintree, Mass.

hSodasorb, W.R. Grace and Co, Lexington, Mass.

1Hudson, Temecula, Calif.

jABLl, Radiometer, COpenhagen NV, Denmark.
kCathlon IV, Critkon, Tampa, Fla.

1Rompun, Haver Lockhart, Shawnee Mission, Kan.
mVetalar, Parke-Davis, Morris Plains, NJ.
nAceto Chemical Co, Inc, Flushing, NY.
°Bio-tal, Bio-Cutic, St. Joseph, Mo.
pDobutrex, Eli Lilly, Indianapolis, Ind.
q5000b, Valley Labs, Boulder, Co.

rSigma Chemical Co, St. Louis, Mo.

54

LI ST OF REFERENCES

LI ST OF REFERENCES

Armstrong RF, St Andrew D, Cohen SL, et al. Continuous
monitoring of mixed venous oxygen tension (onz) in
cardiorespiratory disorders. Lancet 1978;1:632-634.

Baele PL, McMichan JC, Marsh HM, et a1. Continuous monitor-
ing of mixed venous oxygen saturation in critically
ill patients. Anesth Analg 1982;61:513-517.

 

Barratt-Boyes BG, Wood EH. The oxygen saturation of blood
in the vanae cavae, right-heart chambers, and pulmonary
vessels of healthy subjects. J Lab Clin Med 1957;50:
93-106.

 

Bergsten G. Blood pressure, cardiac output, and blood-

gas tension in the horse at rest and during exercise.
Acta Vet Scand [Suppl] 1974;48:1-88.

Bottoms GD, Fessler JF, Roesel 0F, et a1. Endotoxin-
induced hemodynamic changes in ponies: Effects of
flunixin meglumine. Am J Vet Res 1981;42:1514-1518.

 

Boyd AD, Tremblay RE, Spencer FC, et al. Estimation of
cardiac output soon after intracardiac surgery with
cardiOpulmonary bypass. Ann Surg 1959;150:613—626.

Brody S. Bioenergetics and growth. New Yorszafner
Publishing Co Inc, 1945;313-319.

Burrows GE. Hemodynamic alterations in the anesthetized
pony produced by slow intravenous administration of
Escherichia coli endotoxin. AmJVet Res 1970;31:1975-1932.

 

Cournand A, Riley RL, Bradley SE, et a1. Studies of the

circulation in clinical shock. Surgery 1943;13:964-
995.

de la Rocha AG, Edmonds JF, William WG, et a1. Importance
of mixed venous oxygen saturation in the care of
critically ill patients. Can J Surg 1978;21:227-229.

 

Duff JH, Groves AC, McLean APH, et a1. Defective oxygen
consumption in septic shock. Surg Gynecol Obstet
1969;128:1051-1060.

 

55

56

Dyson DH, Allen DG, McDonell WN. Comparison of three
methods for cardiac output determination in cats.
Am J Vet Res 1985;46:2546-2552.

 

Eberly VE, Gillespie, JR, Tyler WS. Cardiovascular param-
eters in the thoroughbred horse. Am J Vet Res 1964;
25:1712-1715.

 

Eberly VE, Gillespie JR, Tyler WS, et a1. Cardiovascular
values in the horse during halothane anesthesia.
Am J Vet Res 1968;29:305-314.

 

Fegler G. Measurement of cardiac output in anesthetized
animals by a thermodilution method. QfJ Exp Physiol
1954;39:153-164.

 

Finch CA, Lenfant C. Oxygen transport in man. N Engl
J Med 1972;286:407-415.

Fisher EW, Dalton RG. Determination of the cardiac output
of cattle and horses by the injection method. Br Vet J
1961;118:143-151.

Gillespie JR, Tyler WS, Hall LW. Cardi0pulmonary dysfunction
in anesthetized, laterally recumbent horses. Am J Vet
Res 1969;30:61-72.

Goldman RH, Braniff BA, Harrison DC, et a1. Early detec-
tion of heart failure by central venous oxygen
saturation monitoring. J Am Cardiol (Abst) 1967;21:100.

Goldman RH, Klughaupt M, Metcalf T, et al. Measurement
of central venous oxygen saturation in patients with
myocardial infarction. Circulation 1968;38:941-946.

 

Hall LW, Gillespie JR, Tyler WS. Alveolar-arterial oxygen
tension differences in anaesthetized horses. Br J
Anaesth 1968;40:560-567.

Henness AM, Theilen GH, Madewell BR, et al. Use of drugs
based on square meters of body surface area. J Am
Vet Med Assoc 1977;171:1076-1077.

 

Hillidge CJ, Lees P. Cardiac output in the conscious
and anaesthetized horse. Equine Vet J 1975;7:16-21.

 

Hodges M, Downs JB, Mitchell LA. Thermodilution and Pick
cardiac index determination following cardiac surgery.
Crit Care Med 1975;3:182-184.

 

57

Hutter AM, Moss AJ. Central venous oxygen saturations:
Value of serial determinations in patients with acute
myocardial infarction. JAMA 1970;212:299-303.

Jamieson WRE, Turnbull KW, Larrieu AJ, et a1. Continuous
monitoring of mixed venous oxygen saturation in cardiac
surgery. Can J Surg 1982;25:538-543.

Kasnitz P, Druger GL, Yorra F, et al. Mixed venous oxygen
tension and hyperlactatemia: Survival in severe
cardiopulmonary disease. JAMA 1976;236:570—574.

Kazarian KK, Del Guercio LRM. The use of mixed venous
blood gas determinations in traumatic shock. Ann
Emerg Med 1980;9z179-182.

Kelman GR, Nunn JF, Prys-Roberts C, et al. The influence
of cardiac output on arterial oxygenation: A theoretical
study. Br J Anaesth 1967;39:450-458.

 

Kerr DD, Jones EW, Huggins K, et al. Sedative and other
effects of xylazine given intravenously to horses.
Am J Vet Res 1972;33:525-532.

Kohanna FH, Cunningham JN, Catinella FP, et a1. Cardiac
output determination after cardiac operation: Lack
of correlation between direct measurements and indirect
estimates. J Thorac Cardiovasc Surg 1981;82:904-908.

Krauss XH, Verdouw PD, Hugenholtz PG, et al. On-line
monitoring of mixed venous oxygen saturation after
cardiothoracic surgery. Thorax 1975;30:636-643.

Kubo K, Senta T, Sugimoto 0. Cardiac output in the
Thoroughbred horse. Exp Rep Equine Health Lab 1973;
10:84-89.

 

Lee J, Wright F, Barber R, et a1. Central venous oxygen
saturation in shock: A study in man. Anesthesiology
1972;36:472-478.

 

Lindsay WA, McDonell W, Bignell W. Equine postanesthetic
forelimb lameness: Intracompartmental muscle pressure
changes and biochemical patterns. Am J Vet Res 1980;
41:1919-1924.

 

Lumb WV, Jones EW. Veterinary anesthesia. 2nd ed.
PhiladelphiazLea and Febiger, 1984;95-99, 147-153,
510-512.

 

58

Magilligan DJ, Eisiminger R, Fried M. Use of inline
venous oxygen saturation to predict post bypass cardiac
output. Circulation (Abst) 1983;68:111-152 (608).

Manley SV. Monitoring the anesthetized horse. Vet Clin
North Am [Large Anim Pract] 1981;3:111-133.

Martin WE, Cheung PW, Johnson CC, et a1. Continuous

monitoring of mixed venous oxygen saturation in man.
Anesth Analg 1973;52:784-793.

 

McArthur KT, Clark LC, Lyons C, et a1. Continuous record-
ing of blood oxygen saturation in open-heart operations.
Surgery 1962;51:121-126.

McCashin FB, Gabel AA. Evaluation of xylazine as a seda-
tive and preanesthetic agent in horses. Am J Vet Res
1974;36:1421-1429.

 

McDonell WN. General anesthesia for equine gastrointestinal
and obstetric procedures. Vet Clin North Am [Large
Anim Pract] 1981;3:163-194.

 

 

Mikat M, Peters J, Zinder M, et al. Whole body oxygen
consumption in awake, sleeping, and anesthetized dogs.
Anesthesioloqy 1984;60:220-227.

 

Miller MJ. Tissue oxygenation in clinical medicine: An
historical review. Anesth Analg 1982;61:527-535.

Moffitt EA, Patrick RT, Swan HJC, et al. A study of blood
flow, venous blood oxygen saturation, blood pressure
and peripheral resistance during total body perfusion.
Anesthesiology 1959;20:18-26.

Moore JN. Management of pain and shock in equine colic.
Compend Contin Educ Prac Vet 1985;7:5169-Sl77.

 

Muir AL, Kirby BJ, King AJ, et al. Mixed venous oxygen
saturation in relation to cardiac output in myocardial
infarction. Br Med J l970;4:276-278.

Muir WW, Skarda RT, Milne DW. Estimation of cardiac out-
put in the horse by thermodilution techniques. Am J
Vet Res 1976;37:697-700.

Nunn JF. Applied respiratory physiology. 2nd ed. London:
Butterworth and Co Ltd, 1981;388-395.

 

Parr GVS, Blackstone EH, Kirklin JW. Cardiac performance
and mortality early after intracardiac surgery in infants
and young children. Circulation 1975;51:867-874.

 

 

59

Schmidt CR, Frank LP, Forsythe SB, et al. Continuous
3V0 measurement and oxygen transport patterns in
caréiac surgery patients. Crit Care Med 1984;12:
523-527.

 

Schmidt CR, Starr NJ. Evaluation of a continuous SVO2
monitoring system. Anesthesiology 1981;55:A125.

 

Sheldon CA, Cerra FB, Bohnhoff N, et al. Peripheral post-
capillary venous pressure: A new, more sensitive
monitor of effective blood volume during hemorrhagic
shock and resuscitation. Surgery 1983;94:399-406.

Snyder JV, Carroll GC. Tissue oxygenation: A physiologic
approach to a clinical problem. Curr Prob Surg 1982;
19:650-719.

 

Sottile FD, Durbin CG, Hoyt JW, et al. Evaluation of
pulmonary artery oximetry as a predictor of cardiac
output. Anesthesiology 1982;57:A127.

 

Stanley TH, Isern-Amaral J. Periodic analysis of mixed
venous oxygen tension to monitor the adequacy of
perfusion during and after cardiOpulmonary bypass.
Can Anaesth Soc J 1974;21:454-460.

Stowe CM, Good AL. Estimation of cardiac output by the
direct Fick technique in domestic animals, with obser-
vations on a case of traumatic pericarditis. Am J
Vet Res 1961;22:1093-1096.

Swanson CR, Muir WW, Bednarski RM, et a1. Hemodynamic
responses in halothane-anesthetized horses given
infusions of dopamine or dobutamine. Am J Vet Res
1985;46:365-370.

Trim CM, Moore JN, White NA. CardioPUlmonary effects of
dompamine hydrochloride in an anesthetized horse.
Equine Vet J 1985;17:41-44.

 

Valentine PA, Fluck DC, Mounsey JPD, et a1. Blood-gas
changes after acute myocardial infarction. Lancet
1966;2z837-841.

Waller JL, Kaplan JA, Bauman DI, et al. Clinical evalua-
tion of a new fiberOptic catheter oximeter during
cardiac surgery. Anesth Analg 1982;61:676-679.

 

Wilkinson AR, Phibbs RH, Gregory GA. Continuous in vivo
oxygen saturation in newborn infants with pulmonary
disease. Crit Care Med l979;7:232-236.

 

60

Wilson RF, Gibson D. The use of arterial-central venous
oxygen differences to calculate cardiac output and
oxygen consumption in critically ill surgical patients.
Surgery 1978;84:362-369.

Zuntz N. Die ernahrung des herzens und ihre beziehung
zu seiner arbeitsleistung. Dtsch Med Wochenschr
1892;18:109-111.