JOURNAL
OF SURGICAL
RESEARCH
50, 163-169
(1991)
Crystalloid Resuscitation Restores But Does Not Maintain Cardiac Output Following Severe Hemorrhage PING WANG, Departments
of *Surgery
and
M.D.,* tPhysiology, Submitted
AND
IRSHAD H. CHAUDRY,
Michigan for publication
State
University,
January
PH.D.*+’ East
Lansing,
Michigan
48824
4, 1990
Furthermore, studies have shown that hemorrhage alone produces significant immunosuppression [ 1, 2, 331. A great deal of knowledge has been obtained concerning the decreased cardiac output (CO) following hemorrhage [4, 6, 22, 23, 29, 431. Although it is known that a marked depression in CO following hemorrhage results in decreased total organ blood flow and microcirculation, tissue ischemia, cellular dysfunction, and cell death [8], the progressive changes in CO following hemorrhage and crystalloid resuscitation have not been determined in a small animal such as the rat. In addition, although CO decreases following hemorrhagic shock, the relationship between the decrease in mean arterial pressure (MAP) and CO in the rat during hemorrhage is not known. Since a fiberoptic catheter and an in vivo hemoreflectometer (IVH) provides the means for repeated measurement of CO in small animals [14], we have used this technique to determine the progressive changes in CO in the same animal following hemorrhage and resuscitation. The results presented below indicate that resuscitation with 4 times (X) the volume of the shed blood with Ringer’s lactate (RL) restored but did not maintain CO following severe hemorrhage and may play a prominent role in the development of multiple organ failure observed following severe and prolonged hemorrhagic shock.
Although Ringer’s lactate (RL) is routinely used for resuscitation, it is not known whether this fluid alone restores and maintains the depressed cardiac output (CO) following severe hemorrhage. To study this, a fiberoptic catheter was inserted to the level of the aortic arch in rats. Following indocyanine green (0.05 mg) administration, CO was measured using an in viva hemoreflectometer (IVH). The rats were then bled to and maintained at a mean arterial pressure (MAP) of 40 mmHg until 40% of the shed blood volume was returned in the form of RL. They were resuscitated with 2,3, or 4 times (X) the volume of the shed blood with RL and CO recorded at various intervals thereafter. The results indicate that CO decreased significantly during hemorrhage and remained depressed following resuscitation with 2 or 3~ RL. CO was normal immediately after resuscitation with 4~ RL, but it was not sustained and decreased significantly 0.5 to 8 hr postresuscitation. This was not due to the decreased hematocrit since acute hemodilution did not decrease CO. These results indicate that: (1) the progressive changes in CO following hemorrhage and resuscitation can be measured in rats by using IVH; (2) resuscitation with 4~ RL restores total peripheral resistance to normal, but does not maintain CO, suggesting that pharmacological support may be needed under such conditions; (3) the lack of maintenance of CO following resuscitation may play an important role in the development of multiple organ failure after severe hemorrhage. o issi Academic PWS, IUC.
MATERIALS
AND
METHODS
Animal model of hemorrhage. A model of hemorrhagic shock in the rat was recently described by us [9] and was used in this study with some modification. Male Sprague-Dawley rats (281 * 20 g, mean f SD) were fasted overnight (16-20 hr) and lightly anesthetized with ether. In order to introduce trauma prior to hemorrhage, a 4-5 cm ventral midline laparotomy was performed following which the abdominal incision was closed in layers. A length of PE-50 tubing was inserted to the level of the right atrium via the right jugular vein. Heparin (1 unit/g BW) was given via the jugular catheter. Subsequent anesthesia was maintained by means of intravenous sodium pentobarbital (via the jugular
INTRODUCTION Severe hemorrhage due to trauma with the subsequent metabolic and immunological depression continues to remain one of the major problems in the surgical intensive care unit [ 7,121. In those patients who survive the initial hemorrhagic insult, there is an increased risk of sepsis, multiple organ failure, and late death [3,7,11]. 1 To whom correspondence and reprint requests should be addressed at Department of Surgery, B424 Clinical Center, Michigan State University, East Lansing, MI 48824-1315. 163
0022.4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
164
JOURNAL
OF
SURGICAL
RESEARCH:
catheter, total dose approximately 30 mg/kg BW). Both femoral arteries were then cannulated with PE-50 tubing. One of the catheters was used for monitoring mean arterial pressure (MAP) while the other catheter was used for bleeding the animal. The jugular catheter was also used for the measurement of central venous pressure (CVP, via a pressure monitor) and fluid resuscitation. Rats were then bled to a MAP of 40 mmHg over 10 min via a femoral arterial catheter. The MAP was monitored with a saline manometer via another femoral catheter and was maintained at 40 mmHg until 40% of the shed blood volume was returned in the form of RL. At that time (approximately 90 min following the onset of hemorrhage), the rats were resuscitated with 2,3, or 4~ (approximately 22,33, or 44 ml per rat, respectively) the volume of maximum bleedout (MB) with RL over 60 min via the jugular catheter. Maximal bleedout was the volume of blood withdrawn following which the animal could not maintain the MAP of 40 mmHg unless some fluid (RL) was returned. The average time required for MB was 48 min and the average MB volume was 61% of estimated effective blood volume as determined by indocyanine green (ICG) clearance technique [ 141. The shed blood was not given back in this hemorrhage model. Rats were sacrificed at the end of the experiment by an overdose of ether. There were 7 rats in each group (2,3, or 4~ RL resuscitation) and various measurements of CO, MAP, CVP, etc., were performed prehemorrhage, during hemorrhage (at MB), and at various time periods after resuscitation. An additional group of 6 control rats underwent the same surgical procedures, i.e., anesthesia and cannulation of blood vessels, but hemorrhage was not induced. CO and MAP were measured at the corresponding time points. None of the animals died during the course of the experiment. In another group (n = 8), rats were bled so as to decrease the MAP by lo-mmHg increments to a pressure of 40 mmHg within 10 min. CO was recorded at each lo-mmHg decrease in MAP. Indocyanine green (ICG) dilution curve measureof the right jugular ment. Following the cannulation vein and both femoral arteries as described above, the left carotid artery was isolated and topical lidocaine applied. A 2.4-French fiberoptic catheter (Hospex Fiberoptics, Chestnut Hill, MA) was passed down the carotid artery to the level of the aortic arch for the continuous measurement of ICG concentration ([ICG]) in blood using an in vivo hemoreflectometer (IVH, Hospex Fiberoptics). A total of 0.05 mg of ICG (in 0.05 ml solution) per rat (approximately 0.167 mg/kg BW) was administered. With the aid of a computer-assisted data acquisition (Asystant+, Asyst Software Inc., Rochester, NY), 20 ICG concentration points were recorded every second for approximately 30 set immediately after administration of ICG. A representative dilution curve of ICG from a control rat is shown in Fig. 1A. ICG dilution curve
VOL.
50,
NO.
2, FEBRUARY
1991
0.000 0 0.020
z
0
0.015
-5 E
0.010
6 x
0.005
0.000 -0.025
0.000
0.025
Time
0.050
0.075
0.100
in Minutes
FIG. 1. (A) Graph of a representative dilution curve of ICG concentration in blood ([ICG] in mg/ml, y axis) versus time (in minutes, x axis). ICG (0.05 mg/rat) was administered intravenously and 240 time-[ICG] points (from 0 to 0.2 min after the administration of ICG) were recorded and are presented. [ICG] measurement and data acquisition are described under Materials and Methods. (B) Graph of the area under the representative ICG dilution curve presented in Fig. 1A. The calculation for the cardiac output is described under Materials and Methods.
measurements were performed before hemorrhage (control), during hemorrhage (at MB), and 0, 0.5, 1, and 1.5 hr following 2,3, or 4~ RL resuscitation. In rats with 4~ RL resuscitation, ICG dilution curves were also recorded at 4-8 hr following resuscitation. Oxygen saturation (S,O,) was directly measured by the IVH [14]. Calculation of cardiac output and total peripheral resistance. As shown in Fig. lB, the area under the ICG dilution curve was calculated by using a computer-assisted data acquisition (AsystantS). Cardiac output can then be determined by the equation CO = (ICG injected)/(area
under the curve)
(1)
in which the cardiac output is in ml/min, ICG dose is in mg, and the area under the ICG dilution curve is in mg/ ml. min. The cardiac output (ml/min) was then converted to ml/min/lOO g BW by the equation CO = (CO in ml/min)/BW Total peripheral resistance cording to the equation TPR = (MAP
(TPR)
in 100 g. was calculated
- CVP)/CO,
(2) ac-
(3)
where TPR is in mmHg/ml/min/100 g, MAP and CVP are in mmHg, and CO is in ml/min/lOO g.
WANG 40 I-
O
AND 2X RL:
CHAUDRY: ES
3XRL:
CARDIAC I
OUTPUT
AFTER
165
HEMORRHAGE
150
4X RL
0
2X RL; Ed 3X RL: I
4X RL.
1
cn 0"
120
30 (7,
c .G
0.05 for CO, MAP, g BW.
or TPR
1
- cn
30
g
25
5
20 i
Y=-1.969+0.267X 0 /
R=0.9778 P(O.001
In this study, we have used a fiberoptic catheter and an in viva hemoreflectometer (IVH) with computer-assisted data acquisition to measure the progressive changes in cardiac output following hemorrhage and fluid resuscitation in the rat. Although various techniques have been used to measure CO in the rat [5,6,17, 18, 20, 21, 25, 34, 411, the ICG dilution method offers several advantages over the other methods. Unlike radiolabeled microsphere, this method allows repeated CO determinations without blood sampling by using IVH. Thermodilution technique suffers from difficulties in precisely controlling the temperature of a small volume of injectate and possible errors that may occur when lung water volume increases [26], as it occurred following resuscitation with Ringer’s lactate in this study (un-
co Prehemodilution Posthemodilution
P 40
60
60
2
Effect of Acute Hemodilution on Cardiac Output (CO), Mean Arterial Pressure (MAP), and Systemic Hematocrit (H,,) (ml/min/lOO
20
100
following
DISCUSSION
/
OJ
point
significant difference in H,,, values in rats which received 2, 3, or 4~ RL and there was no further decrease in H,,, following resuscitation during the course of the study.
TABLE 35
at any time
31.462 + 2.339 34.089 + 3.152 0.5548
MAP g)
(mmHg)
I-4, (%)
114.3 ?z 2.3 114.5 f 3.7 0.9593
44.0 * 1.0 20.8 2 1.3
OF SURGICAL
RESEARCH
50, 163-169
(1991)
Crystalloid Resuscitation Restores But Does Not Maintain Cardiac Output Following Severe Hemorrhage PING WANG, Departments
of *Surgery
and
M.D.,* tPhysiology, Submitted
AND
IRSHAD H. CHAUDRY,
Michigan for publication
State
University,
January
PH.D.*+’ East
Lansing,
Michigan
48824
4, 1990
Furthermore, studies have shown that hemorrhage alone produces significant immunosuppression [ 1, 2, 331. A great deal of knowledge has been obtained concerning the decreased cardiac output (CO) following hemorrhage [4, 6, 22, 23, 29, 431. Although it is known that a marked depression in CO following hemorrhage results in decreased total organ blood flow and microcirculation, tissue ischemia, cellular dysfunction, and cell death [8], the progressive changes in CO following hemorrhage and crystalloid resuscitation have not been determined in a small animal such as the rat. In addition, although CO decreases following hemorrhagic shock, the relationship between the decrease in mean arterial pressure (MAP) and CO in the rat during hemorrhage is not known. Since a fiberoptic catheter and an in vivo hemoreflectometer (IVH) provides the means for repeated measurement of CO in small animals [14], we have used this technique to determine the progressive changes in CO in the same animal following hemorrhage and resuscitation. The results presented below indicate that resuscitation with 4 times (X) the volume of the shed blood with Ringer’s lactate (RL) restored but did not maintain CO following severe hemorrhage and may play a prominent role in the development of multiple organ failure observed following severe and prolonged hemorrhagic shock.
Although Ringer’s lactate (RL) is routinely used for resuscitation, it is not known whether this fluid alone restores and maintains the depressed cardiac output (CO) following severe hemorrhage. To study this, a fiberoptic catheter was inserted to the level of the aortic arch in rats. Following indocyanine green (0.05 mg) administration, CO was measured using an in viva hemoreflectometer (IVH). The rats were then bled to and maintained at a mean arterial pressure (MAP) of 40 mmHg until 40% of the shed blood volume was returned in the form of RL. They were resuscitated with 2,3, or 4 times (X) the volume of the shed blood with RL and CO recorded at various intervals thereafter. The results indicate that CO decreased significantly during hemorrhage and remained depressed following resuscitation with 2 or 3~ RL. CO was normal immediately after resuscitation with 4~ RL, but it was not sustained and decreased significantly 0.5 to 8 hr postresuscitation. This was not due to the decreased hematocrit since acute hemodilution did not decrease CO. These results indicate that: (1) the progressive changes in CO following hemorrhage and resuscitation can be measured in rats by using IVH; (2) resuscitation with 4~ RL restores total peripheral resistance to normal, but does not maintain CO, suggesting that pharmacological support may be needed under such conditions; (3) the lack of maintenance of CO following resuscitation may play an important role in the development of multiple organ failure after severe hemorrhage. o issi Academic PWS, IUC.
MATERIALS
AND
METHODS
Animal model of hemorrhage. A model of hemorrhagic shock in the rat was recently described by us [9] and was used in this study with some modification. Male Sprague-Dawley rats (281 * 20 g, mean f SD) were fasted overnight (16-20 hr) and lightly anesthetized with ether. In order to introduce trauma prior to hemorrhage, a 4-5 cm ventral midline laparotomy was performed following which the abdominal incision was closed in layers. A length of PE-50 tubing was inserted to the level of the right atrium via the right jugular vein. Heparin (1 unit/g BW) was given via the jugular catheter. Subsequent anesthesia was maintained by means of intravenous sodium pentobarbital (via the jugular
INTRODUCTION Severe hemorrhage due to trauma with the subsequent metabolic and immunological depression continues to remain one of the major problems in the surgical intensive care unit [ 7,121. In those patients who survive the initial hemorrhagic insult, there is an increased risk of sepsis, multiple organ failure, and late death [3,7,11]. 1 To whom correspondence and reprint requests should be addressed at Department of Surgery, B424 Clinical Center, Michigan State University, East Lansing, MI 48824-1315. 163
0022.4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
164
JOURNAL
OF
SURGICAL
RESEARCH:
catheter, total dose approximately 30 mg/kg BW). Both femoral arteries were then cannulated with PE-50 tubing. One of the catheters was used for monitoring mean arterial pressure (MAP) while the other catheter was used for bleeding the animal. The jugular catheter was also used for the measurement of central venous pressure (CVP, via a pressure monitor) and fluid resuscitation. Rats were then bled to a MAP of 40 mmHg over 10 min via a femoral arterial catheter. The MAP was monitored with a saline manometer via another femoral catheter and was maintained at 40 mmHg until 40% of the shed blood volume was returned in the form of RL. At that time (approximately 90 min following the onset of hemorrhage), the rats were resuscitated with 2,3, or 4~ (approximately 22,33, or 44 ml per rat, respectively) the volume of maximum bleedout (MB) with RL over 60 min via the jugular catheter. Maximal bleedout was the volume of blood withdrawn following which the animal could not maintain the MAP of 40 mmHg unless some fluid (RL) was returned. The average time required for MB was 48 min and the average MB volume was 61% of estimated effective blood volume as determined by indocyanine green (ICG) clearance technique [ 141. The shed blood was not given back in this hemorrhage model. Rats were sacrificed at the end of the experiment by an overdose of ether. There were 7 rats in each group (2,3, or 4~ RL resuscitation) and various measurements of CO, MAP, CVP, etc., were performed prehemorrhage, during hemorrhage (at MB), and at various time periods after resuscitation. An additional group of 6 control rats underwent the same surgical procedures, i.e., anesthesia and cannulation of blood vessels, but hemorrhage was not induced. CO and MAP were measured at the corresponding time points. None of the animals died during the course of the experiment. In another group (n = 8), rats were bled so as to decrease the MAP by lo-mmHg increments to a pressure of 40 mmHg within 10 min. CO was recorded at each lo-mmHg decrease in MAP. Indocyanine green (ICG) dilution curve measureof the right jugular ment. Following the cannulation vein and both femoral arteries as described above, the left carotid artery was isolated and topical lidocaine applied. A 2.4-French fiberoptic catheter (Hospex Fiberoptics, Chestnut Hill, MA) was passed down the carotid artery to the level of the aortic arch for the continuous measurement of ICG concentration ([ICG]) in blood using an in vivo hemoreflectometer (IVH, Hospex Fiberoptics). A total of 0.05 mg of ICG (in 0.05 ml solution) per rat (approximately 0.167 mg/kg BW) was administered. With the aid of a computer-assisted data acquisition (Asystant+, Asyst Software Inc., Rochester, NY), 20 ICG concentration points were recorded every second for approximately 30 set immediately after administration of ICG. A representative dilution curve of ICG from a control rat is shown in Fig. 1A. ICG dilution curve
VOL.
50,
NO.
2, FEBRUARY
1991
0.000 0 0.020
z
0
0.015
-5 E
0.010
6 x
0.005
0.000 -0.025
0.000
0.025
Time
0.050
0.075
0.100
in Minutes
FIG. 1. (A) Graph of a representative dilution curve of ICG concentration in blood ([ICG] in mg/ml, y axis) versus time (in minutes, x axis). ICG (0.05 mg/rat) was administered intravenously and 240 time-[ICG] points (from 0 to 0.2 min after the administration of ICG) were recorded and are presented. [ICG] measurement and data acquisition are described under Materials and Methods. (B) Graph of the area under the representative ICG dilution curve presented in Fig. 1A. The calculation for the cardiac output is described under Materials and Methods.
measurements were performed before hemorrhage (control), during hemorrhage (at MB), and 0, 0.5, 1, and 1.5 hr following 2,3, or 4~ RL resuscitation. In rats with 4~ RL resuscitation, ICG dilution curves were also recorded at 4-8 hr following resuscitation. Oxygen saturation (S,O,) was directly measured by the IVH [14]. Calculation of cardiac output and total peripheral resistance. As shown in Fig. lB, the area under the ICG dilution curve was calculated by using a computer-assisted data acquisition (AsystantS). Cardiac output can then be determined by the equation CO = (ICG injected)/(area
under the curve)
(1)
in which the cardiac output is in ml/min, ICG dose is in mg, and the area under the ICG dilution curve is in mg/ ml. min. The cardiac output (ml/min) was then converted to ml/min/lOO g BW by the equation CO = (CO in ml/min)/BW Total peripheral resistance cording to the equation TPR = (MAP
(TPR)
in 100 g. was calculated
- CVP)/CO,
(2) ac-
(3)
where TPR is in mmHg/ml/min/100 g, MAP and CVP are in mmHg, and CO is in ml/min/lOO g.
WANG 40 I-
O
AND 2X RL:
CHAUDRY: ES
3XRL:
CARDIAC I
OUTPUT
AFTER
165
HEMORRHAGE
150
4X RL
0
2X RL; Ed 3X RL: I
4X RL.
1
cn 0"
120
30 (7,
c .G
0.05 for CO, MAP, g BW.
or TPR
1
- cn
30
g
25
5
20 i
Y=-1.969+0.267X 0 /
R=0.9778 P(O.001
In this study, we have used a fiberoptic catheter and an in viva hemoreflectometer (IVH) with computer-assisted data acquisition to measure the progressive changes in cardiac output following hemorrhage and fluid resuscitation in the rat. Although various techniques have been used to measure CO in the rat [5,6,17, 18, 20, 21, 25, 34, 411, the ICG dilution method offers several advantages over the other methods. Unlike radiolabeled microsphere, this method allows repeated CO determinations without blood sampling by using IVH. Thermodilution technique suffers from difficulties in precisely controlling the temperature of a small volume of injectate and possible errors that may occur when lung water volume increases [26], as it occurred following resuscitation with Ringer’s lactate in this study (un-
co Prehemodilution Posthemodilution
P 40
60
60
2
Effect of Acute Hemodilution on Cardiac Output (CO), Mean Arterial Pressure (MAP), and Systemic Hematocrit (H,,) (ml/min/lOO
20
100
following
DISCUSSION
/
OJ
point
significant difference in H,,, values in rats which received 2, 3, or 4~ RL and there was no further decrease in H,,, following resuscitation during the course of the study.
TABLE 35
at any time
31.462 + 2.339 34.089 + 3.152 0.5548
MAP g)
(mmHg)
I-4, (%)
114.3 ?z 2.3 114.5 f 3.7 0.9593
44.0 * 1.0 20.8 2 1.3
