LABORATORY I N V E S T I G A T I O N hemorrhage; hypertonic saline/dextran; lactated Ringer's

Use of Hypertonic Saline/Dextran Versus Lactated Ringer's Solution as a Resuscitation Fluid After Uncontrolled Aortic Hemorrhage in Anesthetized Swine

From the Division of Military Trauma Research, Letterman Army Institute o f Research, Presidio o f San Francisco, California;t and DEpartment of Emergency Services, Saint Francis Hospital, Tulsa, Oklahoma. *

William H Bickell, MD* Stephen P Bruttig, PhDt Gregory A Millnamow, MA t John O'Benar, PhDt Charles E Wade, PhDt

Study objective: We tested the hypothesis that following aortotomy, administration of hypertonic saline/dextran increases hemorrhage and mortality. We also compared hypertonic saline/dextran with the standard therapy of attempting to replace three times the amount of lost blood with lactated Ringer's solution.

Received for publication November 15, 1990. Revisions received July 1 an d December 12, !991, and February 28, 1992. Accepted for publication March 26, 1992.

Design: In this model of uncontrolled arterial hemorrhage resulting from aortotomy, 24 anesthetized Yorkshire swine underwent splenectomy, stainless steel wire placement in the infrarenal aorta, and instrumentation with Swan-Banz and carotid artery catheters. The wire was pulled, producing a 5-mm aortotomy and spontaneous intraperitoneal hemorrhage.

The opinions and assertions contained herein are the private views of authors and are not to be construed as official nor do they reflect the views of the Department of the Army or the Department of Defense (AR 360-5).

Interventions: The animals were randomly assigned to one of three study groups: control; hypertonic saline/dextran group in which six minutes after aortotomy a 4-mL/kg mixture of IV 7.5% NaCI and 6% Dextran-70 was given over one minute; or lactated Ringer's group in which six minutes after aortotomy 80 mL/kg IV lactated Ringer's was given over nine minutes.

The studies described in this report were reviewed and approved by the Institutional Review Committee/Animal Care and Use Committee at Letterman Army Institute of Research. The manuscript was peer-reviewed for compliance before submission for publication. In conducting the research herein, the investigators adhered to the "Guide for the Care and Use of Laboratory Animals," DHEW Publication (NIH) 85-23.

The volume of hemorrhage and the mortality rate in hypertonic saline/dextran-treated animals were significantly greater than in the nonresuscitated controls (1,340 +230 mL versus 783 +85 mL and five of eight versus zero of eight, respectively; P< .05). Although the mortality rate in the lactated Ringer's group was not significantly different from the hypertonic saline/dextran group, survival time was significantly shorter than in the hypertonic saline/dextran group. Measurements and main results:

Conclusion: In this model of uncontrolled hemorrhage, immediate IV administration of hypertonic saline/dextran significantly increased hemorrhage volume and mortality. However, the accentuation of hemorrhage and reduction in survival were not as great as that produced by the standard practice of attempting to replace the lost blood with three times that volume of lactated Ringer's. [Bickell WH, Bruttig SP, Millnamow GA, O'Benar J, Wade CE: Use of hypertonic saline/dextran versus lactated Ringer's solution as a resuscitation fluid after uncontrolled aortic hemorrhage in anesthetized swine. Ann EmergMed September 1992;21:1077-1085.]

SEPTEMBER1992

21:9 ANNALSOF EMERGENCYMEDICINE

1077/47

HYPERTONIC SALINE/DEXTRAN BickeU et al

INTRODUCTION

The combination of hypertonic saline and dextran has been shown to improve the survival of animals subjected to blood loss that would otherwise be fatal.]-3 After a fixed-volume hemorrhage, in which blood is traumatically withdrawn through a surgically implanted catheter, hypertonic saline/dextran (HSD) has been shown to r e t u r n cardiac output, arterial pressure, and tissue blood flow to normal or near-normal levels in pigs, dogs, and sheep. 1-5 This hemodynamic response is believed to occur through a fluid mobilization from the extravascular compartment, which in t u r n increases venous r e t u r n and cardiac output. 2,4,5 The encouraging results from the aforementioned laboratory studies have led to the use of a combined solution of HSD for the out-of-hospital resuscitation of patients with hemorrhagic hypotension. 6-a However, in this clinical setting, blood loss most often results from an interruption in the vascular circuit that is inaccessible to external control.9,10 Moreover, it is doubtful that fluid resuscitation in an experimental model of controlled blood withdrawal through a catheter will produce the same outcome as that chnically observed when the body is attempting to control hemorrhage through a variety of hemostatic mechanisms.

Figure 1. Survival of animals during the experiment: A. Survival curves, data points for the experimental groups indicate the actual times at which each loss occurred B. Treatment groups' respective mean and median survival times after aortotomy % Survival 100

~_. . . . . . . . ~_-

"L'_

~

....

1

[] . . . .

[] . . . . .

[]

5O

MATERIALS AND METHODS S u r g i c a I P r e p a r a t i o n Twenty-four immature Yorkshire

gilt swine were obtained from a commercial b r e e d e r and maintained in a common indoor holding area until used for the study two to four weeks after arrival. They were fed a commercial ration and allowed water a d l i b i t u m . The pigs were 4 to 5 months old and weighed 23 to 40 kg when the studies were conducted. After an overnight fast, each pig received a preanesthetic intramuscular injection of 2.2 mg/kg ketamine HCI, 2.2 mg/kg xylazine HC1, and 1 mg/kg midazolam HCI. Halothane was given by face mask, and endotracheal intubation was performed. The animals then were administered oxygen (FIo 2 = 0.6), nitrous oxide, and 1% halothane. A celiotomy was performed, and the spleen was removed according to s t a n d a r d techniques with double ligation of all vascular pedicles. The retroperitoneal fascia was incised, and the ventral surface of

Figure 2. •

0

In an effort to create a clinically relevant study design, investigators have used uncontrolled hemorrhage models in which blood loss results from an i n j u r y to the vascular circuit. 1146 In these previous studies, the IV administration of isotonic (0.9%) saline or hypertonic (7.5%) saline was shown to increase hemorrhage and mortality. To our knowledge, no studies have evaluated the effect of administering a combined HSD solution after vascular injury and subsequent hemorrhage. We recently developed a model of uncontrolled arterial hemorrhage in which blood loss results from an abdominal aortotomy. 17 We used this model in our study to test the hypothesis that, after aortotomy, the administration of HSD will increase hemorrhage and mortality. We also c o m p a r e d the effects of HSD t h e r a p y to the s t a n d a r d therapy of replacing three times the amount of lost blood with lactated Ringer's solution. ]8

........

Q

I

I

~

0

5

15

i

30 Time

60

90

Cardiac output (mLIminlkg) during the experiment. All points are means + SEM. Fine dotted lines indicate the times at which the number of animals dropped below the valid sample size for the statistical analysis.

120

(rain)

3ardiac ouput/weight (mL/min/kg) [] .....

300

~

Time (rain)

[]

Control

~ LHaSctDatedR!nger's

200

I00

0

Median

Mean [ ] Control

48/1078

[ ] HSD

~ Lactated Ringer's

i

i 5

i 15

i 30

i 60

r 90

i __ 120

Time (rain) ~* Significantly different from control. # Significantly different between HSD and lactated Ringer's. 1" Significantly different between times. P< ,05

ANNALS OF EMERGENCY MEDICINE

21:9

SEPTEMBER1992

HYPERTONIC SALINE/DEXTRAN BickeU et al

the aorta was exposed. Ten centimeters proximal to the aortic bifurcation, on the'ventral median surface of the aorta, a 4-0 monofilament stainless steel wire was inserted through the aortic wall into the aortic lumen and then exteriorized through the same surface 5 mm cephalad. The details of the stainless steel wire placement have been described previously.17 The free ends of the aortic stainless steel suture were exteriorized on the ventral abdominal wall. The abdominal surgical incision was then closed in two layers with size 0 l)exon. Through a midline neck incision, the right common carotid a r t e r y was exposed and a polyvinyl catheter was inserted to the level of the aorta and secured by ligatures around the vessel. A 7.5-flow-directed thermodilution SwanGanz catheter (Gould Inc, Cleveland, Ohio) was inserted through the right internal j u g u l a r vein and positioned with the distal p o r t in the pulmonary artery. An 8F polyvinyl catheter was inserted into the right external j u g u l a r vein. The surgical p r e p a r a t i o n required 35 to 45 minutes. The lumina of all catheters were Filled with normal saline. The arterial and Swan-Ganz catheters were connected to Figure 3. Relationships between various cardiovascular pressures (mm Hg) and time daring the experiment: A. Mean systemic arterial pressure B. Mean pulmonary arterial pressure

Statham 23 Db pressure transducers (Viggo-Spectramed, Inc, Critical Care Division, Oxnard, California), a Gould ESI000 multichannel polygraph, and a Gould cardiac output computer. The animals then were given 50 mL of 0.5% chloralose in 3% urethane solution. The halothane, nitrous oxide, and oxygen were discontinued, and the animals were allowed to breathe room air spontaneously through the endotracheal tube. Twenty-five milliliters of the chloralose/ urethane mixture was administered as needed whenever the animal demonstrated spontaneous extremity or truncal motion. 19 Experimental P r o t o c o l One hour after the halothane had been discontinued, baseline hematocrit, arterial blood gases, and cardiodynamics (ie, phasic aortic, p u l m o n a r y arterial, central venous, p u l m o n a r y capillary wedge pressures, and cardiac output) were recorded. 2o The stainless steel wire was removed from the abdominal cavity by simultaneously pulling the two ends of the wire. The withdrawal of the wire suture resulted in a 5-mm laceration in the long axis of the aorta and spontaneous i n t r a - a b d o m i n a l C. Pulmonary capillary wedge pressure D. Central venous pressure AU points represent means + SEM. Fine dotted lines indicate the times at which the number of animals dropped below the valid sample size for the statistical analysis.

M e a n arterial pressure (mm Hg) 150

• ..... : ±

Pulmonary capillary wedge pressure (mm Hg) ~ ~ m . . . . . • control : : HSD I ± Lactated Ringer's

• Control : HSD -" Lactated Ringer's

100

~ . . . . ~->.,:.:.~

* 50

0

A

J

Basal

i

I

5

t5

i

I

r

i

30 60 90 120 Time (rain) * Significantly different from control. # Significantly different between HSD and lactated Ringer's. t Significantly different between times. P< .05

o

_l r i i 30 60 90 120 Time (rain) * Significantly different from control. # Significantly different between HSD and lactated Ringer's. 1" Significantly different between times. P< .05

M e a n pulmonary artedal pressure (ram Hg)

4°1 30

• ..... : }

~ ,

i Basal

i 5

t 15

Central v e n o u s pressure (mL/min/kg)

• Control : HSD J- Lactated Ringer's

15

7~ *

• ..... • : :

Control HSD

~

Lactated Ringer's

/_L\ 20

.

"

:'":"" ~

~°

-

,~,

5

t0

0

B

~ .........