|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Institute of Critical Care Medicine (Drs. Povoas, Weil, Tang, and Kamohara and Mr. Bisera), Palm Springs, CA; and Optical Sensors Inc (Mr. Moran), Minneapolis, MN.
Correspondence to: Max Harry Weil, MD, PhD, Master FCCP, The Institute of Critical Care Medicine, 1695 North Sunrise Way, Building 3, Palm Springs, CA 92262-5309; e-mail: Weilm{at}aol.com
| Abstract |
|---|
|
|
|---|
Design: Prospective animal study.
Setting: Animal laboratory in a research institution.
Participants: Domestic pigs.
Interventions: Hemorrhagic shock was induced by a modification of the Wiggers method. BP was maintained at 50 mm Hg for 120 min followed by reinfusion of shed blood at a rate of 100 mL/min with the aid of an infusion pump.
Measurements and results: During bleeding, the mean arterial pressure decreased from an average of 127 to 42 mm Hg, and cardiac output decreased from 7.7 to 2.4 L/min. Arterial blood lactate concentration concurrently increased from 1.2 to 13.9 mmol/L. Sublingual PCO2 (PslCO2) increased from 59 to 105 mm Hg, and gastric PCO2 increased from 61 to 111 mm Hg. The correlation between time-coincident sublingual and gastric measurements of PCO2 was r = 0.91 (p < 0.0001). Bland-Altman analyses demonstrated a close correspondence between the two measurements. The reinfusion of shed blood promptly reversed the hemodynamic abnormalities and reestablished gastric and PslCO2 to near baseline values. This contrasted with a delayed reversal of lactic acidosis.
Conclusions: Under experimental conditions of hemorrhagic shock, sublingual capnometry yielded measurements that were interchangeable with those of gastric tonometry.
Key Words: carbon dioxide tension gastric tonometry hemorrhagic shock sublingual capnometry
| Introduction |
|---|
|
|
|---|
The use of gastric tonometers with saline solution-filled balloons required prolonged equilibration periods. The accuracy of PCO2 measurements on saline solution utilizing a conventional blood gas analyzer also was called into question.7 A new generation of tonometers that utilized an intermittent and automated gas-sampling system, simplified the procedure.8 Nevertheless, gastric tonometry remains a semi-invasive and intermittent measurement that is subject to errors because of the interfering effects of acid gastric fluid and delays in equilibration. It is in this context that we sought alternatives for the measurement of tissue PCO2.
Earlier studies had demonstrated that not only the stomach but also the esophageal wall served as appropriate sites for the measurement of tissue PCO2 to estimate the severity of circulatory shock.9 Indeed, increases in tissue CO2 appeared as a universal phenomenon when blood flow to tissues was critically reduced during circulatory shock states.10 This led us to examine the option of measuring PCO2 under the tongue. Sublingual measurements were initially performed on rats during hemorrhagic and septic shock. These measurements correlated highly with those of gastric wall PCO2 levels.11 Sublingual PCO2 (PslCO2) also emerged as a quantitative measurement of the severity of perfusion failure in that it predicted survival on a par with that of arterial blood lactate.12
In the present study, we compared the commercially available automated gastric gas tonometer with directly measured PslCO2. We anticipated that sublingual measurements of PCO2 would yield measurements that would correspond closely to those of gastric PCO2 and thereby support the measurement of sublingual tissue PCO2 as an alternative to more complicated and invasive gastric measurements.
| Materials and Methods |
|---|
|
|
|---|
Animal Preparation
Five domestic, male Yorkshire pigs, weighing 35 to 40 kg were
fasted overnight except for free access to water. The animals were
preanesthetized with IM injection of 20 mg/kg ketamine and were
positioned on a V-shaped surgical table in a supine posture. Anesthesia
was induced with IV injection of 30 mg/kg pentobarbital from a 16-gauge
cannula placed in an ear vein. A cuffed endotracheal tube then was
advanced into the trachea, and the animals were mechanically ventilated
with room air utilizing a tidal volume of 15 mL/kg and a peak flow of
40 L/min with the aid of a volume-controlled ventilator (model MA-1;
Nellcor Puritan Bennett; Carlsbad, CA). End-tidal carbon dioxide
(ETCO2) was monitored with an infrared
analyzer (model 601POET; Crit Care System; Milwaukee, WI). Respiratory
frequency was adjusted to maintain ETCO2
between 35 and 40 mm Hg and was unchanged thereafter. A bolus IV
injection of 8 mg/kg pentobarbital at approximately 60-min intervals
maintained anesthesia at levels that precluded spontaneous breathing.
Conventional ECG limb leads were continuously monitored. A 7F
pentalumen thermodilution catheter (Abbott Critical Care Systems; North
Chicago, IL) was advanced into a surgically exposed right femoral vein,
and the flow was directed into the pulmonary artery. For the monitoring
of aortic pressure, a 5F catheter (model DLR-7 h; Braintree Scientific,
Inc; Braintree, MA) was advanced into the descending thoracic aorta
through the right femoral artery. A 14F cannula (William Harvey model
1848 USCI; CR Bart Inc; Billerica, MA) was advanced into the abdominal
aorta through the left femoral artery for bleeding. Another 14F cannula
was inserted into the femoral vein for the reinfusion of shed blood.
The position of the catheters was confirmed by both characteristic
pressure-pulse morphology and with the aid of fluoroscopy. The aortic
cannula was flushed with 5 mL physiologic salt solution containing 10
IU/mL bovine heparin at intervals of 30 min. This cannula was connected
to a 2-L reservoir for bleeding and reinfusion. Blood temperature was
continuously measured in the pulmonary artery and was maintained at
38 ± 0.5°C utilizing an infrared surface heating lamp.
Two gastric tonometer catheters (TRIP NGS Catheter; Tonometrics Division, Instrumentarium Corp; Helsinki, Finland) were introduced into the stomach of each animal. The first was introduced by the orogastric route and was advanced into the antrum of the stomach. A midline epigastric miniceliotomy was performed for the direct insertion of a second tonometer catheter (TRIP Sigmoid Catheter; Tonometrics Division, Instrumentarium Corp) into the stomach through an antral gastrotomy and immobilized with a purse-string suture. These tonometric catheters were designated as gastric PCO2 (PgCO2) sensor 1 (Pg1CO2) and sensor 2 (Pg2CO2). The abdomen was closed in one layer. Correct positioning of the catheters was confirmed by fluoroscopy.
For the measurements of PslCO2, two optical sensor prototypes (CapnoProbes; Optical Sensors Inc; Minneapolis, MN), which were designated as sublingual sensor 1 for PCO2 (Psl1CO2) and sublingual sensor 2 for PCO2 (Psl2CO2), were advanced under the tongue into the right inferior sublingual space.
Measurements
The tonometer monitors (Tonocap TC-200; Tonometrics Division,
Instrumentarium Corp) were calibrated according to manufacturers
instructions. Room air and a 5% CO2 gas source
were utilized for calibration. The tonometer monitors automatically
delivered 5 mL room air to the balloon of the tonometer tube. This air
was allowed to equilibrate for an interval of 10 min. At the end of
this 10-min interval, an aliquot of the gas in the tonometer balloon
was withdrawn and was delivered to an infrared capnometer.
Gastric fluid was aspirated from a second lumen of the tonometer catheter for the measurement of gastric luminal pH with a pH-sensitive paper (Short Range Alkacid; Fisher Scientific; Pittsburgh, PA). The gastric luminal pH was > 4.0 U in only one animal and ranged from 2.0 to 3.0 U in the remaining four animals. In those four animals, 100 mg of the H2 blocker ranitidine (Glaxo Pharmaceuticals; Research Triangle Park, NC) was injected IV. Within 60 min after the injection of the H2 blocker agent, the gastric fluid pH exceeded 4.0 in each instance.
The sublingual CO2 probes were calibrated in a closed system in which PCO2 in saline solution was maintained at a pressure of 78.1 mm Hg (SensiCath Level 2 Calibrants; Optical Sensors Inc).
Aortic and pulmonary arterial pressures were measured with the aid of strain gauge pressure transducers (TRANSPAC; Abbott Critical Care Systems). Cardiac output was measured by the thermodilution technique with the aid of a cardiac output computer (model 3300; Abbott Critical Care Systems) after a bolus injection into the right atrium of 5 mL normal saline solution, which had been maintained at a temperature between 0°C and 2°C. All measurements, including sublingual and gastric PCO2, aortic pressure, ETCO2, and ECG lead II were recorded on a 16-channel personal computer-based data acquisition system supported by appropriate software (CODAS; DATAQ Instruments; Dayton, OH).
Aortic and mixed venous (pulmonary artery) blood gases were measured with the aid of an automated pH/blood gas analyzer (Nova Stat Profile Ultra C; Nova Biomedical; Waltham, MA) and a co-oximeter system (model 482; Instrumentation Laboratories; Lexington, MA). The lactate concentration in blood was measured with an electrode-based lactate analyzer (model 23 L; Yellow Springs Instruments; Yellow Springs, OH).
Experimental Procedures
Hemorrhagic shock was induced by a modification of the Wiggers
method followed by the reinfusion of shed blood. Blood was allowed to
flow aseptically through preheparinized catheters into a sterile 2-L
reservoir containing 4,000 IU heparin. A hydraulic device of our design
provided for the fine adjustment of pressure within the reservoir,
which was maintained at 50 mm Hg for 120 min.13
The blood
contained in the reservoir was reinfused after 120 min at a rate of 100
mL/min with the aid of an infusion pump. Both
PslCO2 and
PgCO2 values were measured at
baseline and at 10-min intervals. Cardiac output, arterial and mixed
venous blood gas levels, and arterial blood lactate levels were
measured at baseline and at 30-min intervals for a duration of 3.5
h. ETCO2 levels, aortic pressure, and ECG
were continuously recorded. The animals were monitored for an
additional hour after the reinfusion of shed blood and then were killed
by IV injection of 150 mg/kg pentobarbital.
Statistical Analysis
Time-based hemodynamic measurements were compared by analysis of
variance for repeated measurements. A p value of < 0.05 was
considered to be significant.
Relationships between hemodynamic measurements and simultaneous sublingual and gastric tonometric measurements were analyzed with linear regression analyses.
For estimating the variances inherent in concurrent measurements of two gastric and two sublingual sensors, the difference between the average of Psl1CO2 and Psl2CO2 for 22 individual time-coincident measurements in each of five pigs was computed. The differences between Pg1CO2 and Pg2CO2 measurements also represent the averages of 22 individual time-coincident measurements in five pigs. Both the differences between the two sublingual and the two gastric measurements and the difference between Psl1CO2 and Pg1CO2 were analyzed by paired t tests and by the Bland and Altman14 procedure. The Bland-Altman analysis is the currently accepted statistical method for testing the assumption that time-coincident values obtained with one method of measurement are interchangeable with those of an alternative method of measurement. In the present study, we applied the Bland-Altman analysis to a comparison between sublingual and gastric wall tonometry. The mean differences (bias) and the standard deviation of the differences (precision) together with 95% confidence intervals are presented. This analysis overcomes the limitations of traditional linear regression analyses, which only demonstrate correlations but not interchangeability of two time-coincident methods of measurements.
| Results |
|---|
|
|
|---|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
Increases in tissue PCO2 are associated with decreases in oxygen availability or utilization by cells.17 Hydrogen ions are anaerobically generated as a byproduct of lactic acid and from the hydrolysis of adenosine triphosphate and adenosine diphosphate.17 When the hydrogen ions are buffered by intracellular HCO3-, CO2 is generated.18 The high diffusability of CO2 allows for the rapid equilibration of PCO2 throughout tissues, thereby facilitating surface measurements.19
The results of the present study support those of Kivilaakso et al,20 who found that the reduction of blood flow to the stomach was proportional to the reduction in total cardiac output during hemorrhagic shock. The reductions in intestinal blood flow during hemorrhagic shock were coincident with and proportional to reductions in cardiac output.21 Decreases in intestinal oxygen delivery to and oxygen consumption by the intestines were also proportional to decreases in total systemic oxygen delivery and systemic oxygen consumption during hemorrhagic shock, according to Nelson et al.22 When systemic and organ blood flows were measured by the microspheres technique during hemorrhagic shock in rats by our own group, reductions in blood flow to the tongue, stomach, jejunum, colon, and kidneys were each proportional to the measured decreases in cardiac output.10 It is to this extent that the present studies do not support the traditional assumption that the viscera, and specifically the stomach, represent unique sites for the measurement of increases in tissue PCO2 or decreases in tissue pH.
The model employed provided for rapid and severe blood loss, and the findings are therefore applicable to hypovolemic shock. As yet, there are no comparable data for the distributive types of shock, specifically septic shock, excepting significant correlations between PslCO2, cardiac index, ETCO2, and arterial blood lactate levels.11 12
For the purposes of assessing the severity of perfusion failure in settings of hemorrhagic shock, PslCO2 measurements demonstrated increases that were comparable to those of PgCO2 measurements in porcine models. Since the sublingual measurement also has the advantages of noninvasiveness, rapid availability, and technical simplicity without the need for medication to block gastric acid production, it provides a promising alternative to gastric tonometry.
| Footnotes |
|---|
Supported, in part, by grants from Optical Sensors, Inc (Minneapolis, MN), The Rosse Family Charitable Foundation, and the National Institutes of Health (National Heart, Lung, and Blood Institute grant No. RO1 HL-54322).
Received for publication November 24, 1999. Accepted for publication March 22, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. A. A. M. Cammarata, M. H. Weil, M. Fries, W. Tang, S. Sun, and C. J. Castillo Buccal capnometry to guide management of massive blood loss J Appl Physiol, January 1, 2006; 100(1): 304 - 306. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Bridges and S. Dukes Cardiovascular Aspects of Septic Shock: Pathophysiology, Monitoring, and Treatment Crit. Care Nurse, April 1, 2005; 25(2): 14 - 40. [Full Text] [PDF] |
||||
![]() |
K. R. Denninghoff, M. H. Smith, A. Lompado, and L. W. Hillman Retinal venous oxygen saturation and cardiac output during controlled hemorrhage and resuscitation J Appl Physiol, March 1, 2003; 94(3): 891 - 896. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. C. Rackow, P. O'Neil, M. E. Astiz, and C. M. Carpati Sublingual Capnometry and Indexes of Tissue Perfusion in Patients With Circulatory Failure Chest, November 1, 2001; 120(5): 1633 - 1638. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Marik Sublingual Capnography : A Clinical Validation Study Chest, September 1, 2001; 120(3): 923 - 927. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |