CA2091622C - Method and apparatus for measuring central venous oxygen saturation during human cardiopulmonary resuscitation and clinical shock - Google Patents

Abstract

A fiber optic catheter for continuous measurement of central venous oxygen saturation (Scv02> during human cardiopulmonary arrest and shock. When applied in cardiopulmonary arrest (cardiac arrest), the catheter provides therapeutic and prognostic guidelines in the management of a patient in this condition. The catheter also serves as a conduit for fluid and drug infusion and as a sampling port for venous blood. The catheter comprises a body having a first port that exits through a connector body to the computer interface that provides the sending signal and receiving signal which generates central venous oxygen saturation readings. A fiberoptic bundle of afferent and efferent light-conducting fibers transverses the first port to provide signal generation and interpretation of oxygen measured in the blood. A second port is a lumen that traverses and exits at the distal port to allow for pressure measurement and sampling of the venous blood. A third proximal port provides an opening of a lumen that transverses and exits the catheter via a side port spaced from the skin insertion site.

Description

~0~~.~2~
This invention relates to a method and apparatus for measuring central venous oxygen saturation and treatment during human cardiopulmonary resuscitation (cardiac arrest) and clinical shock Baakgroaand of the In~entioa Cardiac arrest is one of the most dynamic pathophysiologic events in clinical medicine. An immediate cascade of pathologic processes is 'triggered secondary to 'the abrupt cessation of oxygen delivery. Since oxygen is not stored in sufficient quantities in the body, interruption of adequate oxygen transport to the cells for brief periods of time can result in death. Advanced cardiac life support (ACTS) attempts to provide oxygen delivery and thereby attenuate this cascade.
Rapid and substantial improvement in oxygen delivery is required to decrease the morbidity and mortality of ischemic organ injury.
Current monitoring techniques used in ACLS include continuous electrocardiographic monitoring and physical examination e.g.
palpation of the carotid or femoral pulse. Both of these techniques provide little information regarding hemodynamic status and oxygen delivery to the brain or body.
Clinical monitoring techniques used as prognostic and therapeutic indicators during ACTS include the coronary perfusion pressure (CPP) or aortic to right atrial relaxation phase pressure gradients, and end-tidal carbon dioxide concentration (ETC02). The importance of CPP as a prognostic indicator of return of spontaneous circulation (ROSC) during 20~~.~2~
animal and human CPR is well established. CPP is the '°gold"
standard for measuring hemodynamic response to therapy during CPR. Calculation of CPP requires placement of both an aortic artery and central venous catheter which may limit its applicability. ETC02 has been studied in animals and humans and has been proposed as a prognostic and therapeutic guide during CPR. Although ETC02 has the advantage of being non-invasive, it is influenced by multiple variables (i.e.
aspiration, pre-existing pulmonary disease) that may limit its reflection of blood flow and CPP in the setting of cardiac m arrest, Scv02 monitoring in accordance with the present invention requires only central venous cannulation while Ao-Ra requires venous and arterial cannulation. Scv02 monitoring is not affected by the same extrinsic variables that affect ETC02.
Mixed venous blood reflects tissue oxygen (02) delivery during cardiac arrest and circulatory failure.
Selective venous hypoxemia or low oxygen content when compared to arterial blood are characteristic during cardiac arrest.
Intermittent mixed venous oxygen saturation measurement during ACLS predicts outcome in cardiac arrest patients and hemodynamically unstable trauma patients. In studies made in accordance with the invention, patients successfully resuscitated had higher central venous oxygen saturations than non-resuscitated patients. As far as the present invention is aware, this is the first measurement of this parameter continuously in cardiac arrest patients with fiberoptic technology.

2~~:~.~~~
Ideally, mixed venous blood should be drawn from a pulmonary artery catheter. However, placement of such a catheter is unlikely during cardiac arrest. A number of studies have supported the substitution of central venous (right arterial or superior vena cave) blood for mixed venous blood (pulmonary arterial) during spontaneous circulation, circulatory failure and closed chest ACZS. There is reported no significant difference between pulmonary artery, central and femoral venous blood gases during closed chest AChS in animals.
l~ The catheter in accordance with the invention and methods using the catheter utilize the concept of Scv02. The information provided by such a catheter is a guide to the care of patients in cardiac arrest. It is believed that this is the first utilization of a catheter of this type in this clinical situation. The present invention includes a catheter and methods specific for utilization in cardiopulmonary resuscitation.
Swonnary of the Invention and Method It is an object of this invention to provide a catheter and method of utilizing continuous central venous oxygen measurement during cardiac arrest. This will enable a treating clinician to tailor therapy and formulate a prognosis in the treatment of a patient in shock and cardiac arrest, Using the concentration of oxygen in central venous blood, decisions can be made regarding hemodynamic status in response to therapy and predict outcome.

In accordance with the invention, a fiberoptic catheter is provided for continuous measurement of central venous oxygen saturation tScv02) during human cardiopulmonary arrest and shock. When applied in cardiopulmonary arrest, the catheter provides therapeutic and prognostic guidelines in the management of a patient in this condition. The catheter also serves as a conduit for fluid infusion, drug administration, and as a sampling port for venous blood. The catheter comprises a body having a first port that exits through a connector body to the computer interface and provides the sending signal and receiving signal that generates central venous oxygen saturation readings. A fiberoptic bundle of afferent and efferent light-conducting fiber transverses the first port to provide signal generation and interpretation of oxygen measured in the blood.
A second port is a lumen that traverses and exits at the distal port to allow for pressure measurement and sampling of the venous blood. A third proximal port provides an opening of a lumen that transverses and exits the catheter via a side port ZO
cm from the skin insertion site.

Descr~on ~f the Drawings FIG. 1 is a schematic of a fiber optic catheter embodying the invention and adapted to be used in the method.
FIG. 2 is a sectional view taken along the line 2-2 in FIG. 1.
FIG. 3 is a sectional view taken along the line 3-3 in FIG. 1.
FIG. 4 is an end view from the left as viewed in FIG. 1.

Descri~t~.t~n In accordance with the invention, the method and apparatus for measuring central venous oxygen saturation comprises a fiber optic catheter 10 for continuous measurement of central venous oxygen saturation (~av02) during human cardiopulmonary arrest and shock. When applied in cardiopulmonary arrest, the catheter provides therapeutic and prognostic guidelines in the management of a patient in this condition. The catheter also serves as a conduit for fluid and drug infusion and as a sampling port for venous blood. The catheter comprises a body ~ having a first port 11 that exits through the connector body B to the computer interface and provides the sending signal and receiving signal that generates central venous oxygen saturation readings. A fiberoptic bundle of afferent and efferent light-conducting fibers transverses the first port 11 (diameter .6 mm) to provide signal generation and interpretation of oxygen measured in the blood. A second port 12 is a lumen that traverses and exits at the distal port P
to allow for pressure measurement and sampling of the venous blood. A third proximal 13 port provides an opening of the lumen from proximal infusion port ~ that transverses and exits the catheter via a side port 14 preferably less than 10 centimeters from the skin insertion site. These ports allow for pressure measurement, fluid or drug administration. The placement of port 14 is important to avoid signal artifact (distortion) when infusing large volumes of fluid or drugs.

Referring to FIGS. 1 and 2, the catheter 10 includes a body 20 including a first catheter portion 21 and a second frustoconical body portion 22. The first portion 21 is adapted to be inserted into the blood vessel and preferably has a length of at least 19 cm. The first portion 21 is capable of flexing during insertion but has sufficient rigidity that it will not be flexed during turbulent blood flow. The body 20 further includes a second portion 22 that is sufficiently rigid to maintain the lumens relatively fixed to one another and to withstand the external compression forces of an introduces of conventional construction. The lumens 11, 12, 13 extend externally from body 20 to the respective connector body 1B, distal port P and proximal infusion port I.
Referring to FIGS. 1 and 2, a preferred form of the catheter 10 comprises a body of urethane having three lumens extending therethrough. The polyuretk~~ane catheter may have a length of 19 centimeters long and a diameter of 7 French for optimal placement into the (patient) venous system. The 7 French diameter is needed for the appropriate compliance. It has a slight curvature to avoid contact with the vessel walls as it enters and resides in the venous system. It can be placed through an introduces, preferably 7.5 French, to avoid damage to the optic fibers that exit from the distal port. The proximal or first portion 21 aspect of the catheter as it enters the skin is reinforced and tapered for 3 centimeters to avoid damage of the fiberoptic bundles and compression of the infusion and sampling ports when a tight fitting introduces is used.

The proximal aspect of the catheter 10 has a first port 11 that exits through a connector body to the computer interface and provides the sending signal and receiving signal that generates central venous oxygen saturation readings. A
fiberoptic bundle of afferent and efferent light-conducting fiver transverses the first port 11 (diameter .6 mm) to provide signal generation and interpretation of oxygen measured in the blood: A second port 12 is a lumen that traverses all 19 centimeters (.6 mm) and exits at the distal part to allow for pressure measurement and sampling of the venous blood. A third proximal port 13 (.9 mm) provides an opening of the lumen that transverses and exits the catheter via a side port 10 centimeters from the skin insertion site.
In a study approved by the Henry Ford Hospital Institutional Review Baard for Human Research, one hundred consecutive patients presented to the emergency department (ED) in non-traumatic, normothermic cardiopulmonary arrest were studied. Cardiopulmonary arrest was defined as the absence of a palpable pulse and respirations and was later verified by the absence of spontaneous aortic pulsations on the pressure tracing.
Patients received only asic Cardiac Life Support (BCLS) by emergency medical service (EMS) personnel prior to admittance to the ED. Upon arrival, patients were clinically managed by emergency medicine physicians using Advanced Cardiac Life Support tACLS) guidelines. The epinephrine dose (.O1 to .2 mg/kg) was used at the clinician's discretion. CPR was performed by a pneumatic compression device (Thumper, Michigan _ g _ Instruments, Grand Rapids, MI ) at a compression rate of 80/minute and an excursion of 1$-2 inches. Ventilation of 80~ oxygen was provided between compressions at a 5:1 ratio. Catheter placement, data recording and blood sampling were performed by on-call research personnel, who had no active involvement in the clinical management of the patient.
All catheters were simultaneously placed. A 7.5 French introducer was percutaneously placed into the central venous system via the subclabian vein. °'French" is a customary unit of measure for catheter and needle diameter, one French being equal to a third of a millimeter. The side port 1~ was used for drug and fluid administration. The infrared, fiberoptic catheter 10 was advanced through the central port of the introduces into the central venous sy~~tem. The distal lumen 12 on the fiberoptic catheter was used for central venous gas sampling and pressure measurement.
For arterial pressure monitoring, a 60 cm 5.8 French catheter was placed in the aortic arch, either percutaneously or by cut down technique, via the femoral artery. The catheters were connected to precalibrated pressure transducers through a heparinized saline flush system. The resulting signals were amplified and pressure tracings were recorded throughout CPR
and during the initial post--resuscitation period. Correct catheter positions were confirmed by supine chest radiograph at the end of ACLS.
Hemoglobin, hematocrit, simultaneous arterial and venous blood gases were obtained every 10 minutes throughout _ g the resuscitation. The fiberoptic catheter 10 was precalibrated in vitro before insertion. During the cardiac arrest, the initially measured 02 saturation of the venous blood gas samples was used for an in vivo calibration of the Scv02 monitor.
Initial (the first values obtained in the arrest), mean (average of all values obtained during arrest) and maximal (higher value obtained during the cardiac arrest before return of spontaneous circulation (ROSC) were recorded.
ROSC was defined as the development of a spontaneous pulse wave form on the arterial tracing associated with a systolic blood pressure greater than 60 mm Hg for more than 5 minutes. Cardiac arrest was defined as the absence of a pulsatile axterial tracing or a systolic blood pressure of less than 60 mm Hg.
Sixty-eight episodes of ROSC: or blood pressure were observed. Patients with ROSC had higher initial, mean, and maximal Scv02 compared to those without ROSC (p-.23,.0001, and .0001, respectively). The positive predictive value for ROSC
with a maximum Scv02 of greater than 60$ and 72~ was .93 and 1e0, respectively. The negative predictive value for ROSC with a maximum Scv02 of less than 30~ and 40~ was 1.0 and .93 respectively. Only one of 6$ episodes of ROSC was attained with a maximum Scv02 of less than 40$. High dose epinephrine t.2 mg/kg) depresses Scv02 temporarily, and this effect decreases with the duration of arrest. Continuous Scv02 monitoring can serve as an adjunct in monitoring therapeutic response and as an prognosticator of ROSC during cardiopulmonary resuscitation (CPR).
Using the catheter of this invention and data obtained, one can determine the effectiveness of therapy in providing adequate circulation or blood flow during CPR. This is the amount of oxygen that is delivered throughout the body. When the Scv02 is above 60~, the chances of having a blood pressure or beating heart is 93~. When the Scv02 is above 72~, there is a 100 chance of having a blood pressure, If the patient never obtains an Scv02 of above 40~, there is no chance of survival and should not be resuscitated indefinitely beyond 30 minutes with a value below this number.
The specific findings of the above research are set forth in the attached Appendixy pages 14 to 41.
The catheter thus provides the following advantages:
1. A f fiber optic central venous catheter and methods specifically for use in cardiopulmonary resuscitation and shock.
2. A fiber optic catheter specific for use in the central venous position during cardiac arrest and shock states in which the patient is resuscitated from low or no blood pressures. In this condition the patient is in delivery dependent oxygen consumption state which allows for the use of Scv02 as a diagnostic and therapeutic modality. It has greater prognostic value than measurement of arterial blood pressure.
Thus, in accordance with the invention, the catheter comprises three lumens, a body, said body having a first portion that extends into the blood vessel far a length of at least 19 cm, said first portion being capable of flexing during insertion but will remain rigid and not flex during turbulent blood flow, a second portion being rigid to maintain the lumens relatively fixed to one another and to withstand the external compression forces of an introducer, said lumens extending externally to a connector body to a computer interface, a distal infusion port and a proximal infusion port, .respectively, said lumens for said proximal infusion port extending radially to a side surface of the first portion of the body.
The length of the catheter portion 21 is preferably less than 20 cm from the skin to the e:nd of the catheter. The lumen side port 14 must be at least 9 cm from the end of the catheter portion 21 and must be flexible to avoid contact with the blood vessel walls and must be sufficiently rigid that it will not flex due to variations in blood pressure.
The method of managing a patient in cardiac arrest who is undergoing CPR comprises placing first portion of a catheter in the central venous system, connecting one lumen to computer to measure central venous oxygen saturation, utilizing said second and third lumens for volume and fluid infusion, pressure measurement and sampling of venous blood in accordance with ACLS guidelines.
In accordance with the inventions (a) if Scv02 is greater than 50~, interrupting the CPR and obtaining the blood pressure such that advanced cardiac life support can be tailored to treating a spontaneously beating heart, (b) if the Scv02 is above 73~, interrupting CPR and treating the patient as one with a spontaneously beating heart, (c) if the patient does not attain an Scv02 of above 40~ during 30 minutes of recording and treatment :Eor cardiac arrest, continued attempts at resuscitation are not warranted.

A P P E N D I X
CONTINUOUS Scv02 DURING CPR

7 Emanuel P. Rivers, MD, MPH*,**

8 Gerard B. Martin, MD, FACEP*

9 Howard Smithline, MS, MD*

Mohamed Y. Rady, M.B, MD, FRCS*

11 Carol H oiby Schultz, MD*

12 Mark G. Goetting, MD** *

13 Timothy J. Appleton S.*
B.

14 Richard M. Nowak, MD, FACEP*

18 Departments of Emergency Medicine*, Surgery** and Pediatrics***
19 Henry Ford Health Systems 2799 West Grand Boulevard 21 Detroit, Michigan 48202 24 Presented at the Society for Academic Emergency Medicine, Annual Scientific Symposium, Washington, D.C., 26 May, 1991.

Address for correspondence:

32 Emanuel P. Rivers MD
33 Department of Emergency Medicine 34 Henry Ford Hospital 2799 West Grand Boulevard 36 Detroit, Michigan 48202 CONTINUOUS Scv02 DURING CPR

2 Study Objective: The purpose of this study was to observe, 3 measure and describe the changes in central venous oxygen satura-4 tion (Scv02) during CPR and immediately after ROSC. It was to also to examine the clinical utility of continuous Scv02 moni-6 toring as a indicator of return of spontaneous circulation (ROSC) 7 3uring human cardiopulmonary resuscitation (CPR).
8 Design: Eight month, prospective, non-outcome, observational, 9 nonrandomized case series.
Setting: Emergency department in a large urban hospital.
11 Types of patients: Adult normothermic, nontraumatic, out of 12 hospital cardiopulmonary arrests.
13 Interventions: All patients were managed according to advanced 14 cardiac life support (ACLS) guidelines. An proximal aortic and double lumen central venous catheter was placed. Scv02 was 16 continuously measured spectrophotometrically by a fiberoptic 17 catheter in the central venous location.
18 Measurements: Aortic blood pressure and Scv02 was simultaneously 19 measured throughout each resuscitation. ROSC was defined as a systolic blood pressure of greater than 60 mm Hg for more than 21 five minutes.
22 Results: One hundred patients who experienced 68 episodes of 23 cardiac arrest were studied. Patients with ROSC had a higher 24 initial, and statistically higher mean, and maximal Scv02 com-pared to those without ROSC (p=.23, .0001, and .0001, respec-26 tively;-p<.05 is significant). No patient attained ROSC without 1 reaching an Scv02 of at least 30%. Only one of 68 episodes of 2 ROSC was attained without reaching an Scv02 of at least 40%. An 3 Scv02 of greater than 72% was 100% predictive of ROSC.
4 Conclusion: Continuous Scv02 monitoring can serve as a reliable indicator of ROSC during CPR in humans.
6 Rey Words: CPR, continuous venous oxygen saturation, Scv02 8 In cardiac arrest there is an abrupt cessation of systemic 9 pulsatile blood pressure, flow and oxygen delivery (D02) that trigger a complex cascade of pathologic processes. Advanced 11 Cardiac Life Support (ACLS) attempts to immediately provide D02 12 to the cerebral and systemic circulation to attenuate this 13 cascade. Rapid and substantial improvement in D02 is required to 14 decrease the morbidity and mortality of ischemic organ injury.
Current monitoring techniques in ACLS (1) include continuous 16 electrocardiographic monitoring and physical examination (e. g.
17 palpation of the carotid or femoral pulse). Both of these tech-18 niques provide limited information regarding systemic blood flow 19 and D02.
Clinical monitoring techniques used as prognostic and 21 therapeutic indicators during ACLS include the coronary perfusion 22 pressure (CPP) or aortic to right atrial relaxation phase pres-23 sure gradients (2), and end-tidal carbon dioxide (ETC02) concen-24 tration (3). These variables do not necessary correlate with systemic D02 during CPR. In addition, continuous Scv02 measure-26 ment al-lows one to take advantage of a pre-existing (if per-CONTINUOUS Scv02 DURING CPR
1 formed) central venous cannulation; while CPP measurement re-2 quires both aortic and central venous cannulation. Because Scv02 3 is an intravascular device, it may not be affected by variables 4 that can alter pulmonary gas exchange and therefore the reli-ability of ETC02 in human CPR (4,5).
6 Mixed venous blood oxygen saturation is an index of the 7 balance between systemic D02 and oxygen consumption (V02) during 8 cardiac arrest (6,7) and circulatory failure (8). Selective 9 venous hypercarbia, hypoxia and acidosis when compared to arteri-al blood are characteristic and indicate the severity of derange-11 ment of tissue oxygenation during cardiac arrest (6,7,8,9).
12 Because of this, intermittent Scv02 measurement has been used to 13 predict outcome in cardiac arrest (10,11) and hemodynamically 14 unstable trauma patients (12,13). Patients successfully resus-citated have higher Scv02 than non-resuscitated patients.
16 The purpose of this study was to observe, measure and 17 describe the changes in continuous Scv02 during CPR and imme-18 diately after ROSC. This study will also explore its utility as 19 an indicator of ROSC.
MATERIALS AND METHODS
21 This study was approved by the Henry Ford Hospital Institu-22 tional Review Board for Human Research under waived consent. One 23 hundred consecutive patients presenting to the emergency depart-24 ment (ED) in nontraumatic, normothermic cardiopulmonary arrest were studied. Power calculations indicated that this was an 26 adequate sample size for statistical analysis. Patients received 1 only Basic Cardiac Life Support (BCLS) by emergency medical 2 service (EMS) personnel prior to admittance to the ED.
3 Upon arrival, patients were clinically managed by emergency 4 medicine physicians using Advanced Cardiac Life Support (ACLS) guidelines (1). CPR was performed by a pneumatic compression 6 device (Thumper, Michigan Instruments, Grand Rapids, MI) at a 7 compression rate of 80/minute and an excursion of 1 and 1/2-2 8 inches. Ventilation of 80°s oxygen was provided between compres-9 sions at a five to one ratio. Catheter placement, data recording and blood sampling were performed by on call research personnel, 11 who had no active involvement in the clinical management of the 12 patient.
13 All catheters were simultaneously placed. An 8 French 14 introducer (Cook Catheter, Bloomington, IN.) was percutaneously placed into the central venous system via the subclavian vein.
16 The side port was used for drug and fluid administration. A
17 double-lumen, infrared, fiberoptic catheter was advanced through 18 the central port of the introducer into the central venous 19 system. The open lumen on the fiberoptic catheter was used for central venous gas sampling and pressure measurement.
21 For arterial pressure monitoring, a 60 cm. 5.8F catheter 22 (Bunegin-Albin, Cook Catheter, Bloomington, IN.) was placed in 23 the aortic arch via the femoral artery. The aortic catheter was 24 connected to a manometrically precalibrated pressure transducer (Sorenson Transpac, Abbot Systems, Bloomington, IN.) through a 26 heparinized saline flush system. The resulting signals were CONTINUOUS Scv02 DURING CPR
1 amplified (Hewlett Packard, Sunnyvale, CA.), and pressure trac-2 ings were recorded (Hewlett Packard 7758 multichannel) throughout 3 CPR and during the initial post-resuscitation period. Correct 4 catheter positions were confirmed radiographically.
Cardiopulmonary arrest was clinically defined as the absence 6 of a palpable pulse and respirations and was later verified by 7 the absence of spontaneous pulsatile waveforms (systolic blood 8 pressure less than 60 mm Hg) on the aortic pressure tracing.
9 ROSC was defined as the development of spontaneous pulsatile waveforms greater than 60 mm Hg for 5 minutes. A repeat arrest 11 was defined as the lack of the same pressure criteria for 5 12 minutes while receiving ACLS.
13 The fiberoptic unit consists of a double-lumen poly-urethane 14 catheter (Baxter Edwards Laboratories, Irvine, CA.,) and com-13 puter. Scv02 was determined spectrophotometrically (22, 23).
16 oxyhemoglobin saturation is measured, computed, and averaged over 17 a five to ten second interval then subsequently updated each 18 second on a paper recorder and digital display.
19 Hemoglobin, simultaneous arterial and venous blood gases were obtained every 10 minutes throughout the resuscitation. The 21 fiberoptic catheter was precalibrated before insertion. During 22 cardiac arrest, the first central venous oxygen saturation 23 sampled (Instrumentation Laboratory Inc. Co-oximeter Model-282, 24 Lexington, MA.) was used for an in vivo calibration of the Scv02 computer.
26 Initial (the first Scv02 value obtained in the arrest), mean CONTINUOUS Scv02 DURING CPR
1 with an arterial partial pressure of oxygen of less than 50 mm Hg 2 were excluded.

4 Student's t-tests with correction for multiple comparisons were used to compare patient and study time characteristics in 6 addition to components of the blood gas analysis between the ROSC
7 and no-ROSC group's where applicable. A p value of less than .05 8 was defined prospectively as statistically significant.
9 The sensitivity of a variable using a given Scv02 is defined as the number of ROSC patients with values equal to or above the 11 cutoff point divided by the total number of ROSC patients. The 12 specificity of maximal Scv02 is defined as the number of no-ROSC
13 patients with values below the cutoff divided by the total number 14 of no-ROSC patients. The positive predictive value for a given Scv02 was calculated by dividing the number of ROSC patients with 16 Scv02's equal to or above the given Scv02 (true positives) by the 17 sum of this group and the patients without ROSC who had Scv02's 18 equal to or above the Scv02 (true positives and false positives).
19 The negative predictive value for a given Scv02 was calculated by dividing the number of patients without ROSC who had Scv02's 21 below the value (true negatives) by the sum of this group and the 22 ROSC patients with an Scv02 below this value (true negatives and 23 false negatives).
24 Regression analysis with Pearson's correlation was used to compare Scv02 obtained from the fiber-optic catheter and la-26 boratory co-oximetry measured Scv02. Repeated measure analysis 1 (average of each Scv02 per minute obtained during arrest) and 2 maximal (highest Scv02 obtained during the cardiac arrest before 3 ROSC) Scv02 measurements were obtained for each patient and each 4 episode of ROSC.
The Scv02 measurement over the course each resuscitation was 6 divided into four phases corresponding to the presence or absence 7 of a spontaneous pulsatile aortic pressure. These phases are the 8 ACLS, TRANSITION, ROSC AND IMMEDIATE POST-ROSC PHASES. In the 9 ACLS phase, patients had no spontaneous pulsatile aortic pressure and low Scv02. The TRANSITION phase is characterized by either 11 an acute or gradual upward trend in Scv02 without a spontaneous 12 pulsatile aortic pulse pressure. In the ROSC phase, the Scv02 13 obtained immediately after a spontaneous aortic pulse represents 14 this value following the transition phase. The POST-ROSC phase was characterized by a plateau of Scv02 in the presence of a 16 stable arterial blood pressure.
17 Down 'time interval (DT) was obtained when available from 18 family members and was the time from cardiac arrest until EMS
19 arrival. Bystander cardiopulmonary resuscitation (CPR) was sporadically performed during DT. Transport time interval (TT) 21 was the time from EMS arrival to ED arrival. Basic CPR was 22 performed by EMS personnel during TT. ACLS time interval repre-23 sents the time from ED arrival until ROSC, the time between a 24 repeat cardiac arrest or until ACLS was terminated. There was no prehospital ACLS performed in this study. The total time of 26 arrest interval (TTA) was the sum of DT, TT, and ACLS. Patients CONTINUOUS Scv02 DURING CPR
1 of variance (ANOVA) and student s t-test with correction for 2 multiple comparisons were used to compare Scv02 values between 3 the 4 phases of resuscitation.

Forty-nine percent of the 100 patients studied attained 6 ROSC. Nineteen patients experienced two or more episodes of 7 ROSC. There was a total of 68 episodes of ROSC. measured. While 8 this study was not a prospective outcome study, twenty-percent of 9 patients with ROSC were admitted to the intensive care unit.
There were 3 patients who were discharged neurologically intact.
11 There was no difference in mean age, and TT between the ROSC
12 and no ROSC groups (Table 1). The mean DT, ACLS, and TTA were 13 statistically greater in the no ROSC than ROSC groups. ROSC
14 patients had statistically higher initial venous p02, S02, and pH
compared to no-ROSC groups (table 2). ROSC patients also had a 16 lower initial venous pC02, arterial pH, serum bicarbonate and 02 17 extraction ratios (normal less than .25-.30) than the no-ROSC
18 group (table 2). ROSC patients had statistically higher mean and 19 maximal Scv02 than no-ROSC patients (table 3). These values were consistent with better systemic blood flow and tissue oxygenation 21 in the ROSC than no-ROSC group. The threshold Scv02 for ROSC was 22 not significantly different between patients treated with stan-23 dard dose (.O1 mg/kg) and high dose (.2 mg/kg) epinephrine.
24 The receiver operator characteristic curve compares the relationship of sensitivity and specificity for maximal Scv02 as 26 a predictor of ROSC using the values shown. The better the sen-CONTINUOUS Scv02 DURING CPR
1 sitivity and specificity of a variable the closer the curve lies 2 to the left and top of the graph (figure 1). The negative 3 predictive value for ROSC with a maximum Scv02 of less than 30%
4 was 1Ø The positive predictive value for ROSC with a maximum Scv02 of greater than 72% was 1.0 (figure 1). A scatter plot 6 (comparing maximal Scv02 of the ROSC and no-ROSC groups) and a 7 contingency figure (comparing ROSC with maximal Scv02) is shown 8 in figures 2 and 2a, respectively.
9 The mean Scv02 values during the ACLS, TRANSITION, ROSC and POST-ROSC phases of resuscitation were 24 +/- 13, 41 +/- 15, 66 11 +/-14 and 7o +/- 10% respectively. The differences in all 12 phases were statistically significant (p<.01). A representive 13 example of these different phases during resuscitation is shown 14 in figure 3.
In vivo calibration was performed for 87 patients. The mean 16 difference between the fiberoptic catheter and laboratory co-17 oximetry measured Scv02 was 3 +/- 3% (p=.05). The correlation 18 coefficient was .93 with a 95% confidence interval.

The prime goal of CPR is to provide adequate systemic D02 to 21 ischemic vital organs. Conventionally, closed chest CPR is moni-22 toyed by ETC02 and/or CPP (2). Although ETC02 was shown to 23 correlate with pulmonary blood flow in experimental models, such 24 correlation is diminished in human CPR (5). Ventilation-per-fusion abnormalities, alveolar gas exchange impairment due to 26 pulmonary aspiration, hemorrhage and atelectasis are pathologic CONTINUOUS Scv02 DURING CPR
1 factors (4,5). Neither ETC02 or CPP reflects the balance of 2 systemic D02 to demand during CPR. CPP is determined pre-3 dominantly by systemic vascular resistance during CPR (2). Vaso-4 pressor therapy (e. g. epinephrine) augments systemic vasocon-striction and CPP while temporarily decreasing systemic D02 6 during CPR. Therefore, CPP does not reflect the systemic ox-7 ygenation during CPR.
8 Ideally, mixed venous blood should be drawn from a pulmonary 9 artery catheter (14). However placement of such a catheter is difficult during CPR. A number of studies have supported the 11 substitution of central venous (right atrial or superior vena 12 cava) blood for mixed venous blood (pulmonary arterial) during 13 spontaneous circulation (15), circulatory failure (16,17,18) and 14 closed chest CPR (19). Emerman et a1 (19) reported no signifi-cant difference between pulmonary artery, central and femoral 16 venous blood gases during closed chest CPR in animals.
17 Arterial and mixed venous oxygen content (Ca02 or Ccv02) 18 takes into account the amount of oxygen bound to hemoglobin (Hb) 19 and dissolved in plasma (20), see appendix. The Fick equation represents the relationship between cardiac output (Qt), oxygen 21 delivery (D02), oxygen consumption (V02), and arterial-mixed 22 venous oxygen content difference, (Ca02-Ccv02). Thus, Qt(Ca02-23 Ccv02)=V02. If Hb is assumed to be constant, the majority of 24 oxygen available to tissues is bound to Hb. By transformation of the above equation, Scv02 = 1-V02/D02 (20). Therefore, Scv02 can 26 reflect-the systemic D02 and V02 relationship during CPR (10,21).

CONTINUOUS Scv02 DURING CPR
1 If V02 remains unchanged then an increase in D02 can result in a 2 corresponding increase in Scv02 during CPR (22).
3 Scv02 was measured in addition to central venous partial 4 pressure of oxygen (Pcv02). In cardiac arrest Pcv02 is low and the amount of oxygen dissolved in plasma is negligible. The 6 majority of venous oxygen content is hemoglobin bound and is 7 represented by Scv02. Although there is a linear relationship 8 between Scv02 and Pcv02 in the range of 75% to 35%, below this 9 range the relationship is nonlinear (22). Additionally, the relationship of Scv02 to Pcv02 is sensitive to alterations in 11 venous pH (22).
12 The ACLS phase (figure 3) is characterized by low cardiac 13 outputs, estimated to be 20-30% of normal (23). The Scv02 values 14 in this phase are consistent with previous studies measuring Scv02 in cardiac arrest patients (10). This phase continues 16 until there is an increase in D02 due to improved effectiveness 17 of ACLS or ROSC. Persistence in this phase is associated with 18 lactic acidosis and high mortality (24). No patient attained 19 ROSC with a maximum Scv02 of less than 30%. Only 1 of 68 epi-sodes of ROSC was attained with a maximum Scv02 of less than 40%.
21 The TRANSITION phase (figure 3) characterizes an acute or 22 gradual upward trend in Scv02, and possibly systemic D02 which 23 heralds the onset of ROSC. The rate of increase in Scv02 is 24 dependent on the rate of improvement of systemic D02 as the patient develops ROSC.
26 The ROSC phase (figure 3) characterizes the maximum Scv02 1 attained following the transition phase and immediately before 2 ROSC. All patients attaining an Scv02 of greater than 72% had 3 RoSC. Only one of 68 episodes of ROSC was attained with an Scv02 4 of less than 40%. The patient who developed ROSC with a maximal Scv02 of 30 had a high central venous pressure and a pulsus 6 paradoxus of 30 mm Hg. She was found to have a tension pneumo-7 thorax.
8 The POST ROSC phase (figure 3) characterizes a plateau of 9 Scv02 in the presence of a stable systemic D02. "Venous hyper-oxia" was defined as supra-normal Scv02 (greater than 75%) and 11 was observed in 40% of patients with ROSC. This indicated either 12 a low systemic V02 and/or high systemic D02. Such an observation 13 in spite of the preceding oxygen debt accrued during cardiac 14 arrest requires further study.
Ten of 51 patients without ROSC attained a maximum Scv02 of 16 greater than 50%. Five of these 10 patients had values greater 17 than 60% (figure 2). In spontaneous circulation elevated Scv02 18 can be attributed to various clinical conditions such as cirrho-19 sis, sepsis, and hypothermia (25). However, in cardiac arrest, the impact of these conditions on this observation are probably 21 marginal except for acute structural defects in the cardiac 22 chambers (acute left to right shunt).
23 Lewinter et al (26), described a case of return of con-24 sciousness during CPR. There was a loss of consciousness after discontinuation of CPR suggesting that there was D02 to support 26 cerebral function. This patient had an Scv02 of 47% during CONTINUOUS Scv02 DURING CPR
1 consciousness which is identical to the mean Scv02 of the ROSC
2 group (table 2). In this case, elevated Scv02 during CPR may be 3 the result of higher systemic D02.
4 A further cause of elevated Scv02 in the absence of sponta neous circulation may be technical or "catheter-wall artifact".
6 If the fiberoptic catheter is adjacent to a vessel wall, light is 7 abnormally reflected and is read by the receiving fibers as 8 increased light reflection. If unrecognized, this can lead to an 9 erroneous elevation of Scv02. The computer is programmed to indicate this artifact in spontaneous circulation but has not 11 been adequately tested during cardiac arrest. Fibrin or platelet 12 thrombi at the optic fiber tip as well as infusion of large 13 volumes of fluid will for the same reason falsely elevate Scv02.
14 The potential for erroneous values using this fiberoptic tech-nology are represented in the references (27,28,29).
16 The accuracy of using Scv02 compares favorably to previous 17 studies. Similar comparisons using the pulmonary artery catheter 18 in spontaneous circulation with a correlation coefficients of .95 19 have been shown (29). This compares to the correlation coeffi-cient of .93 in this study.
21 The use of Scv02 may have a cost-benefit advantage. Scv02 22 would be a greater diagnostic value during CPR since arterial 23 blood gases often fail to reflect the abnormalities seen in 24 central venous gases which axe more hypoxic, hypercarbic and acidotic indicating severe systemic perfusion deficits (18).
26 Improvement in systemic D02 would be reflected in the venous CONTINUOUS Scv02 DURING CPR
1 gases more than arterial. Thus continuous measurement of Scv02 2 may obviate the need for multiple arterial blood gases during the 3 course of a resuscitation.
4 In utilizing its prognostic value, the visual impact of a pathologically low Scv02 over the course of a sometimes chaotic 6 resuscitation will provide objective evidence for deciding when 7 to terminate a fruitless resuscitation attempt. Outcome studies 8 are needed to determine more definitive time intervals of irre-9 versibility. Using information gained from these studies may help avoid resuscitating an irreversibly damaged individual thus 11 preserving ED manpower and health care resources for more sal-12 vageable clinical situations.

14 Continuous Scv02 monitoring can provide on-line oxygen transport information during human CPR. It may provide immediate 16 feedback regarding the effectiveness of ACLS in providing system-17 is D02, serve as an indicator of ROSC and reflects D02/V02 in the 18 immediate post-resuscitation state. Even with its current 19 technical limitations, continuous Scv02 monitoring is accurate and reliable as a indicator of deterioration or improvement of 21 cardiopulmonary function in the course of resuscitating a patient 22 during cardiac arrest.

24 ACRNOWhEDGEMENTS: The authors would like to thank the nursing and all support personnel in the Department of Emergency Medicine 26 for their support in completion of this project.

CONTINUOUS Scv02 DURING CPR
1 REFERENCES:

3 1. Standards and guidelines for cardiopulmonary resuscita-4 tion (CPR) and emergency cardiac care (ECC). JAMA
1986;255:2905-2984.

7 2. Paradis NA, Martin GB, Rivers EP,.Goetting MG, Appleton 8 TJ, Feingold M, Nowak RM: Coronary perfusion pressure 9 and the return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990;263:1106-1113.

12 3. Sanders AB, Ewy GA, Bragg S, Mathew A, Kern KB: Expired 13 ~ pCO2 as a prognostic indicator of successful resusi-14 tation from cardiac arrest. Ann Emerg Med 1985;14:948-952.
16 4. Martin GB, Gentile NT, Paradis NA, Moeggenber J, Apple-17 ton TJ, Nowak RM: Effect of Epinephrine on End-Tidal 18 Carbon Dioxide Monitoring During CPR Ann Emerg Med.
19 1990;19:396-398.
21 5. Callahan M, Barton C: Effect of epinephrine administra-22 tion on ability of end-tidal carbon dioxide readin3s to 23 predict outcome of cardiac arrest. Ann Emerg Med 1990;
24 19:4. (Abstract) 26 6. Weil MH, Rackow EC, Trevino R, Grundler W, Falk JL, CONTINUOUS Scv02 DURING CPR
1 Griffel MI: Difference in acid-base state between 2 venous and arterial blood during cardiopulmonary resus-3 citation. N Engl J Med 1986;315:153-156.

7. Capparelli EV, Chow MSS, Kluger J, Feildman: Differenc-6 es in systemic and myocardial blood acid-base status 7 during cardiopulmonary resuscitation. Crit Care rted 8 1989;17:442-446.

8. Adrogue' HJ, Rashad N, Gorin AB, Yacoub J, Madias NE:
11 Assessing acid-base status in circulatory failure. N
12 Engl J Med 1989;320:1312-1316.

14 9. Nowak RM, Martin GB, Carden DL, Tomlanovich MC. Selec-tive venous hypercarbia during human cpr: Implications 16 regarding blood flow. Ann Emerg Med 1987;16:527-530.

18 10. Rivers EP, Paradis NA, Martin GB, Goetting MG, Rosenb-19 urg J, Appleton, Nowak RM: Systemic oxygen extraction during prolonged CPR in humans. Crit Care Med 1989;17-21 :S72. (Abstract) 23 11. Snyder AB, Salloum LJ, Barone JE, Conley M, Todd M, 24 Digiacomo: Predicting short-term outcome of cardiopul-monary resuscitation using central venous oxygen ten-26 sion measurement. Crit Care Med 1991;19:111-113.

CONTINUOUS
Scv02 DURING
CPR

1 12. Scalea TM, Hartnett RW, Duncan AO, Atweh NA, Phillips TF, 2 Sclafani SJ, Fuortes M, Shaftan GW: Central venous oxygen 3 saturation: a useful tool in trauma patients. J Trauma 4 1990;30:1539-1543.

6 13. Kazarian KK, Del Guercio LRM: The use of mixed venous blood 7 gas eterminations in traumatic shock. Ann Emerg Med 1980;

8 9:179-182.

14. Kandel G, Aberman A: Mixed venous oxygen saturation-its 11 role in the assessment of the crtically ill patient.

12 Arch Intern Med 1983;143:1400-1402.

14 15. Tahvananinen J, Meretoja O, Nikki P: Can central venous blood replace mixed venous blood samples? Crit Care Med 16 1982;10:758-761.

18 16. Lee JF, Wright R, Barber R, Stanley L: Central venous 19 oxygen saturation in shock-a study in man. Anesthesiol-ogy 1971;36(5):472-470.

22 17. Scheinman MM, Brown MA, Rapaport E: Critical assessment 23 of use of central venous oxygen saturation as a mirror 24 of mixed venous oxygen in severely ill cardiac pa-tients. Circulation XL 1969;165-172.

CONTINUOUS Scv02 DURING CPR
1 18. Goldman RH, Klughhaupt M, Metcalf T, Spivack AP, Harri-2 son DC. Measurement of central venous oxygen saturation 3 in patients with myocardial infarction. Circulation.
4 1968;XXXVIII:941-947.
6 19. Emerman CL, Pinchak AC, Hagen JF, Hancock D: A compari-7 son of venous blood gages during cardiac arrest. Am J
8 Emerg Med 1988;6:580-583.

20. Divertie MB, McMichan JC: Continuous monitoring of 11 mixed venous oxygen saturation. 1984 Chest 85;3:423-12 428.
13 21. Martin GB, Rivers EP, Paradis NA, Goetting MG, Appleton 14 TJ, Nowak RM: Systemic oxygen utilization during car-diopulmonary bypass in cardiac arrest patients. Crit 16 Care Med 1990;18:S276. (Abstract) 18 22. Miller MJ: Tissue oxygenation in clinical medicine-an 19 historical review. Anesth & Anal 1982 6;61:527-535.
21 23. Del Guercio LRM, Coomaraswamy RP, State D: Cardiac 22 output and other hemodynamic variables during external 23 cardiac massage in man. N Engl J Med 1963;269:1398-404.
24 24. Kasnitz P, Druger GL, Yorra, Simmons DH: Mixed venous oxygen tension and hyperlactemia. JAMA 1976;236:570-26 574.

CONTINUOUS Scv02 DURING CPR
1 25. Miller M, Cook W, Mithoefer J: Limitations of the use 2 of mixed venous p02 as an indicator of tissue hypoxia.
3 Clin Res 1979;27:401A.

26. Lewinter JR, Carden DL, Nowak RM, Enriquez E, Martin 6 GB. CPR-dependent consciousness: evidence for cardiac 7 compression causing forward flow. Ann Emerg Med 1989;1-8 8:1111-1115.

27. Krouskop RW, Cabatu EE, Bhaktharaj CP, McDonnell FE, 11 Brown EG: Accuracy and clinical utility of an oxygen 12 saturation catheter. Crit Care Med 1984;11:744-749.

14 28. Birman H, Haq A, Hew E, Aberman: Continuous monitoring of mixed venous oygen saturation in hemodynamically un-16 stable patients. Chest 1984;86:753-756.

18 29. Baele PL, McMichan JC, Marsh HM, Sill JC, Southorn PA
19 Continuous monitoring of mixed venous oxygen saturation in critically ill patients. Anesth Analg 1982;61:513-21 17.

CONTINUOUS Scv02 DURING CPR
Figure 1. RECEIVER OPERATOR CURVES FOR MAXIMAL Scv02: The sen-sitivity and 1-specificity for ROSC are compared at various values (in parentheses) of the maximal Scv02.
1.o 0,9' O.B
0.7 r 0.6 f-N
N
0.5 z w 0.4 0,3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 (441 (30) CONTINUOUS Scv02 DURING CPR
Figure 2. SCATTERPLOT COMPARING MAXIMUM Scv02 IN THE NO-ROSC VERSUS
THE ROSC GROUPS: The mean of the maximal Scv02 is 65 +/- 13 and 38 +/- 14 % in the ROSC versus NO-ROSC groups, respectively.
Sv02(~) g0 80 ~ ~ ~ ~

p O
70 ~

...... 70 6 ~ " Q ~
0 ~
''.

0 ..
.. .. ~ -.~ ~- ..
O ~ ~

VS
D
0 ~ 0 ~

3 '..' _ 30 ~

2 ~
0 ~

~ _. .- ' ( 2 0 ~ ~ ~

NO-ROSC
ROSC

CONTINUOUS Scv02 DURING CPR
Figure 2a. CONTINGENCY FIGURE COMPARING MAXIMAL Scv02 AND PERCENT
OF PATIENTS WITH ROSC:

100°~6 (n=1 s) O 91 °,6 84°~6 (n=
f- (n=25) 74°.6 (n=19) Z
W
Q
O 33°.6 f' (n=18) w 20 V 5 °.6 w 0°~ 0°~ (n=19) a 0 (n=4) (n=6) MAXIMAL Scv02(%) CONTINUOUS Scv02 DURING CPR
Figure 3. THE PHASES OF Scv02 DURING RESUSCITATION: This patient received 10 minutes of ACLS and was successfully resuscitated to ROSC. While in the ROSC phase, the patient developed ventricular fibrillation (v-fib). Immediate cardioversion and administration of vasopressor therapy restored ROSC. Note the venous hyperoxia during RoSC.
Sv02 RoSC RosC POST-ROSC

20 V Fts A(~LS

TIME (MINUTES) CONTINUOUS Scv02 DURING CPR

1 Table 1. PATIENT AND ST UDY TIME CHARACTERISTICS: an Scv02 Me 2 between ROSC and no-ROS C groups are compared using Student t-3 test with significance at p<.05. andard deviations are St in 4 parentheses. Age is in years and me intervals minutes.
ti in (DT-down time interval, TT-transport time interval, ACLS-6 emergency department AC LS interval, TTA-total time of arrest 7 interval) 9 MEAN ROSC No ROSC p Age 63 (14) 63 (15) 62 (14) .6 11 DT 12 (13) 8 (8) 18 (15) .0001 12 TT 12 (7) 11 (7) 13 (7) .16 13 ACLS 23 (16) 18 (17) 31 (9) .0001 14 TTA 46 (26) 35 (24) 61 (20) .0001 CONTINUOUS Scv02 CPR
DURING

1 Table 2: COMPARISON INITIAL ARTERIAL AND VENOUS BLOOD
OF

2 GASES: The means between all, ROSC and No-ROSC groups are 3 compared by Student ests with significance t-t at p<.05.

4 Standard deviations in parentheses.
are All ROSC No-ROSC p 6 Arterial:

7 p02 (mm Hg) 247(148) 239 (145) 254 (152) .63 8 S02 (%) 95(11) 95(.12) 95(.10) .97 9 pC02(mm Hg) 31(19) 34(19) 27(18 .09 pH 7.16(.25) 7.09(.23) 7.21(.26) .02 11 HC03-(meq/L) 10(5) 9(5) 11(5) .30 13 Venous:

14 p02 (mm Hg) 35(22) 43(24) 26(14) .003 S02 (%) 34(27) 47(29) 21(18) .0001 16 PC02 (mm Hg) 89(46) 71(30) 107(53) .0005 18 pH 6.90(.23) 6.98(.2) 6.85(.25) .O1 19 HC03-(meq/L) 17(12) 14(5) 20(15) .03 21 02 ext. (%) 67 (25) 55 (26) 77 (18) . 0001 22 Hemoglobin 11.3(2.5) 11.4(2.3) 11.1(3) .56 CONTINUOUS Scv02 DURING CPR

1 Table 3: COMPARISON MEAN AND
OF INITIAL, MAXIMAL
Scv02: Mean 2 Scv02 (%) are compared between the ROSC No ROSC groups and 3 using Students t-tests with sig nificance p<.05. Standard at 4 deviations are in parentheses.

6 ROSC No ROSC p 7 Initial 25(10) 21(9) .23 8 Mean 41(15) 27(10) .0001 9 Maximal 65(13) 38(14) .0001 CONTINUOUS Scv02 DURING CPR
1 APPENDIX: SUMMARY OF OXYGEN TRANSPORT VARIABLES AND EQUATIONS:

3 Ca02 = (Sa02 x Hb x 1.34)+(0.0031 x Pa02) 4 Ccv02 = (Scv02 x Hb x 1.34)+(0.0031 x Pcv02) D02 = Qt x Ca02 6 V02 = Qt x (Ca02-Ccv02) 7 Qt = V02/(Ca02-Ccv02) 8 Scv02 = Ca02 - V02/Qt 9 Scv02 = 1 - V02/D02 11 D02: 02 delivery (ml/min) 12 V02: 02 consumption (ml/min) 13 Qt: cardiac output (L/min) 14 Hb: hemoglobin (gm/dl) Sa02: arterial oxygen saturation (%) 16 Pa02: partial pressure of arterial 02 (mm Hg) 17 Scv02: mixed venous 02 saturation (%) 18 Pcv02: partial pressure of venous 02 (mm Hg) 19 Ca02: arterial 02 content (ml/dl) Ccv02: central venous 02 content (ml/dl)

Claims (5)

1. A central venous oxygen saturation catheter for use in treating a patient for human cardiopulmonary resuscitation and shock during and after cardiac arrest comprising:
a body, a first lumen traversed by a fiber optic bundle of afferent and efferent light-conducting fiber means for sending signals and receiving signals for generating central venous oxygen saturation readings, said body having a first portion which has a distal end and a diameter enabling insertion into a central venous blood vessel, said body having a substantially frustoconical second portion adjacent said first portion, said first portion being capable of flexing during insertion but remaining sufficiently rigid and not flexing during turbulent blood flow, said first portion of said catheter body having a length that is less than 19 cm.

2. The central venous oxygen saturation catheter set forth in Claim 1 including a connector means connected to said first lumen for attachment to a fiber optic computer interface.

3. The central venous oxygen saturation catheter set forth in any one of Claims 1 and 2 including a second lumen extending to a distal port to allow for pressure measurement and sampling of venous blood.

4. The central venous oxygen saturation catheter set forth in Claim 3 including a third lumen having a proximal infusion port for fluid or drug administration.

5. The central venous oxygen saturation catheter set forth in Claim 4 wherein said proximal infusion port is at least cm from said distal end of said first portion of the catheter.