Interpretation of capnography

Interpretation of capnography JONATHAN L. BENUMOF, MD San Diego, California Monitoring and interpretation of the end-tidal carThe anesthetist will ge...
Author: Kenneth Dean
5 downloads 0 Views 697KB Size
Interpretation of capnography JONATHAN L. BENUMOF, MD San Diego, California

Monitoring and interpretation of the end-tidal carThe anesthetist will get the most information out of a capnograph if it is examined systematically. First,the anesthetist must determine whether exhaled co 2 (i.e., a waveform) is present. The differential diagnosisof absent CO2 includes esophageal intubation, accidentaltrachealextubation, disconnection of the breathingcircuit, complete obstruction of the endotracheal tube or conducting system (kink, inspissated blood or secretions, extremely severe bronchospasm) or of the breathingcircuit, apnea, and cardiacarrest. Second, the shape of the waveform must be analyzed systematically by looking at, and in sequence, phase I (inspiratorybaseline, which should be zero); phase II (expiratory upstroke, which should be nearly perpendicularto the inspiratory baseline); phase III (expiratory or alveolarplateau, which should be a straight,nearly horizontal, line); andphase IV (inspiratorydownstroke, which should be nearly perpendicularto the inspiratory baseline). This discussion will follow this systematic approachbut will emphasize diagnosisthat can be obtained from the phase III alveolarplateau. Key words: Capnogram, end-tidal co2 concentration, ventilation.

April 1998/ Vol. 66/No. 2

bon dioxide (ETCO2) waveform are skills critical to

the conduct of a modern anesthetic. This review offers a systematic examination of the ETCO2 waveform and includes the relevant physiological and pathophysiological considerations. Basic principles The best measure of the pressure of end-tidal carbon dioxide (PETCO2) will be obtained when:

1. Tidal volumes are large enough to displace dead space. 2. Fresh gas flow rates are low enough to prevent dilution or washing out of co2. 3. Sample aspiration rates are low enough that they do not interfere with patient ventilation or entrain air that may dilute the co2. 4. The sampling site is close to the patient, minimizing the dead space. 5. The waveform is displayed for end-tidal alveolar plateau analysis.' 2 In any situation in which there is doubt about the validity of the PETCO2 value, Paco 2 should be

measured as a guide for tracking and interpretation. A change in the Paco 2-PETCO 2 gradient in itself may indicate an important pathophysiologic change (e.g., change in dead space, see below). Clinical use The anesthetist will get the most information out of a capnograph if it is examined systematically. First, the anesthetist must determine whether

169

exhaled co2 (i.e., a waveform) is present. Second, the shape of the waveform must be analyzed systematically by looking at, and in sequence, phase I (inspiratory baseline), phase II (expiratory upstroke), phase III (expiratory plateau), and phase IV (inspiratory downstroke) (Figure 1). This discussion will follow this systematic approach but will emphasize diagnoses that can be obtained from

0

x

w

IV.Inspiratory downstroke

/

c 0 0

C02

I. Exhaled co2 usually present, then absent 1. Accidental tracheal extubation

2. Disconnection of breathing circuit 3. Monitor failure 4. Complete obstruction of endotracheal tube 5. Cardiac arrest 6. Patient becomes apneic remains absent 1. Esophageal intubation

III. Alveolar plateau

C

Differential diagnosis of absent exhaled

II. Exhaled co2 usually absent/minimal initially,

Figure 1 The four phases of the capnogram

I. Inspiratory baseline

Table I

Time -

the phase III alveolar plateau. In the future, capnography may display co2 concentration as a function of the volume of exhaled gas from each breath (the single breath test or SBT) because the SBT allows online determination of Vco2 (co2 concentration x tidal volume) and is a more sensitive test of and reveals more information about gas elimination in the late phase of each breath.;

carbon dioxide. It should be noted that very severe bronchospasm (enough to cause complete airway obstruction) can prevent carbon dioxide from being registered by the capnograph even though the trachea has been properly intubated. 6 Only after all of the above life-threatening possibilities have been (quickly) ruled out and ventilation of the patient's lungs has been confirmed by clinical examination, should failure of the capnometer or capnograph be considered in the differential diagnosis. A rapid qualitative check of the capnograph consists of simply removing the co2 sensing or sampling port and exhaling into it. Of all the entities in the differential diagnosis listed above, monitoring cardiac output and cardiopulmonary resuscitation during cardiac arrest is a new application of capnography and therefore will be the only entity further discussed. * Monitoringcardiac output and cardiopulmonary resuscitation by end-tidal carbon dioxide concentration.

During steady state gas exchange equilibrium, the alveolar Pco2 (PACO2), tissue C02 production (Vco2),

Is exhaled co2 (waveform) present? If the presence or persistence of co02 is not detected by the capnometer or capnograph, failure to ventilate the patient's lungs must be assumed. The differential diagnosis of absent co2 includes esophageal intubation, accidental tracheal extubation, disconnection of the breathing circuit, complete obstruction of the endotracheal tube or conducting system (kink, inspissated blood or secretions, extremely severe bronchospasm) or of the breath4 ing circuit, apnea, and cardiac arrest (Table I). With respect to esophageal intubation, capnometry/capnography can make the diagnosis in one breath and is therefore far superior to pulse oximetry, which usually requires some time for desaturation to occur.5 If the stomach contains exhaled gas from previous mask ventilation attempts or carbonated beverages, a few tidal ventilations through an esophageal tube may contain minimal and progressively diminishing concentrations of

170

and alveolar ventilation (VA) are uniquely related as given by PACO2 = (K) VCO2/VA. During constant minute ventilation and Vco 2, an abrupt reduction in cardiac output (Qt) reduces PETCO2 by two mechanisms. 7' 8 First, a reduction in venous return causes a decrease in co02 delivered to the alveolar compartment, resulting in decreased PACO2. Second, the increase in alveolar dead space, which results from the reduced pulmonary vascular pressures, will dilute the C02 from normally perfused alveolar spaces to decrease PETCO2 below PACO2 (see below). During a sustained reduction in Qt, increasing co2 accumulation in the peripheral tissues and in venous blood will begin, after 10 to 20 minutes, to restore C02 delivery to the lung and PETC02 toward baseline levels. Reciprocal changes in PETCO2 will occur during acute increases in Qt. These Qt versus PETCO2 observations have been

made quantitatively and with very good correlation in both controlled experiments in animals 7

Journalof the American Association of Nurse Anesthetists

and in patients undergoing major vascular and cardiac surgery.8 With cardiac arrest, there is no pulmonary blood flow and therefore no delivery of carbon dioxide to the lungs. Consequently, PETCO2 exponentially decreases over a dozen breaths, and there is no steady state exhaled co2 (waveform).9 However, with external cardiac compression, pulmonary blood flow will begin again, and the amount

Figure 2 Infrared analyzer trace

Sc

8

study, the initial PETCO2 was 19 mmHg in those who

eventually regained spontaneous pulses, but only 5 mmHg in those who did not (P< .0001). 15 Furthermore, a sharp increase in ETCO2 often heralded and was the first indicator of the resumption of spontaneous circulation. 3, 5 Almost identical data were found in yet another study (Figure 2, bottom panel.)' 0 Figure 3 shows all of the above, in addition to the effects of sodium bicarbonate infusion, in a patient undergoing ventricular fibrillation, cardiopulmonary resuscitation, and resumption of a spontaneous circulation.' 2 Thus, capnography provides an instantaneous and continuous guide to the efficacy of external chest compression and the resumption of spontaneous pulmonary perfusion. Phase I:The inspiratory baseline The inspiratory baseline is traced as fresh gas moves over the co2 sensing or sampling site (Figure 1). The co2 level during this phase should be zero; if it is not, co2 is being rebreathed. This may be intentional and/or a characteristic (desirable or undesirable) of the equipment being used. The inspiratory baseline becomes elevated if co2 is added to the fresh inspired gas. A co2 rebreathing or bypass valve (which is present on older anesthesia machines) is open, if the co2 absorbent is partially exhausted or gas is channeling through the absorbent, if the expiratory valve is missing or incompetent (exhaled gas in the exhalation limb goes back into the patient's lungs during inhalation, thereby pulling co2-containing gas by the sampling site during inhalation), or a Bain circuit is being used. Phase II: The expiratory upstroke Soon after exhalation begins, co2-containing gas arrives at the co2 sampling site, and it quickly

April 19981 Vol. 66/No. 2

_

0

of CO2 excreted by the lungs (i.e., PETCO2) will be

proportional to the amount of pulmonary blood flow (see above). Indeed, the efficacy of external cardiac compression can be continuously and quantitatively followed by the amount of C02 excreted 11 (Figure 2, top panel) 10°, Exhaled co2 during cardiopulmonary resuscitation can also be used prognostically. 12 14 In one

2

Operator 2

Operator 1

Sc

Time (1 division = 2 seconds)

Recurrence of a SDefibrillation o

.5 --

-

STime

Sinus rhythm-

ventricular fibrillation

---------

(1 division = 2 seconds)

0 o

Top panel: Infrared analyzer trace showing end-tidal carbon dioxide concentration with two different physicians giving external chest compression. Bottom panel: Infrared analyzer trace showing changes inend-tidal carbon dioxide concentration with successful defibrillation and recurrence of ventricular fibrillation. 10 washes away the fresh gas from the previous inspiration. Thus, the expiratory upstroke is steep (Figure 1). When the expiratory upstroke phase of the capnogram becomes prolonged (i.e., the upstroke becomes less steep), delivery of co2 from the lungs to the CO2 sampling site is delayed. Possible causes include mechanical obstruction in the equipment, such as a kinked endotracheal tube, or slow emptying of the lungs, such as with chronic obstructive pulmonary disease or bronchospasm. The expiratory upstroke also becomes prolonged when a sidestream capnograph samples gas too slowly or when the capnograph has a slow response time and the respiratory rate is fast. Phase III: The expiratory (alveolar) plateau The expiratory plateau should detect mixed alveolar gas at the CO2 sampling site. During initial exhalation, elimination of gas from the anatomical dead space (infinite V/Q, zero co2 concentration) is followed by elimination of gas from the wellventilated, low resistance regions of the lung (relatively high V/Q, low co2 concentration). Later, gas from poorly ventilated, high resistance regions of the lung (relatively low V/Q, high co2 concentration) is eliminated. The continuum of V/Q ratios between high and low V/Q areas creates a positive slope (upward to the right) to the alveolar plateau part of the co2 elimination waveform, with the result that the ETCO2 concentration is the last and

171

Figure 3 Capnography and other parameters during cardiopulmonary resuscitation

COMPRESSION

-'_PRECORDIAL

VF

RESUSCITATION

EKG II 1•

200 -

04- _

fr.... .

.

...

:

..

I

.i

;-

DF-200 J

EPINEPHRINE 1mg

ARTERIAL, mmHg

.

NaHCO3, 50 mEq

END TIDAL C02, %

]

I 0958

pH, units PC02, mmHg P02, mmHg Hco3 mEq/L

1000

1002

1004

1006

A

PA

A

PA

A

PA

7.10 26 143 8

6.99 42 16 9

7.60 41 48 40

7.23 110 13 44

7.22 47 92 19

7.10 51 41 18

Serial changes inthe end-tidal carbon dioxide (ETC02) concentration and arterial (A)and mixed venous (PA) blood gases in a representative patient before and immediately after cardiac arrest, during precordial compression, and after defibrillation (DF) and resuscitation. The transient increase in the ETCO2 after the administration of sodium bicarbonate (NaHco3) is also demonstrated. The original tracing has been modified because of space limitation. 12 highest (the peak) concentration on the alveolar or the expiratory plateau (Figures 1 and 4). 16 17 In addition, continued production and evolution of co2 into the alveolar space during exhalation contributes to the rise in co2 concentration during exhalation. Theoretically, the slope is such that the endtidal value should be about 2% higher than the time-weighted mean in normal resting subjects and about 4% higher in exercising subjects, assuming tidal volumes are large enough to displace dead space.' Since the alveolar plateau expresses the V/Q continuum in the lung, analysis of the alveolar plateau may result in a wealth of diagnostic information. First, the steepness of the alveolar plateau is directly related to the degree of airway resistance. Second, biphasic waveforms may reveal the presence of a two-compartment lung. Third, leaks in the sampling system may alter the alveolar plateau in a characteristic, and at first glance, peculiar way. Fourth, the Paco2-PETC02 gradient is directly related to alveolar dead space.

172

* The steepness of the slope of the alveolarplateau is a function of expiratory resistance. When the lungs of a patient without lung disease are being mechanically ventilated, the V/Q units in the lung are relatively uniform and homogeneous (have the same C02 concentration), and the expiratory plateau is smooth and nearly horizontal. However, when there is significant lung disease and a wide spread in V/Q ratios within the lungs, very wellventilated, high V/Q, low C02 concentration areas will empty first, causing the alveolar plateau to be relatively low. Following this, very poorly ventilated, low V/Q, high co2 concentration areas empty causing the alveolar plateau to be relatively high. Thus, with a wide spread in the V/Q ratios, the upward positive slope of the alveolar plateau will be very steep to the right (it is possible, but unusual, for there to be simultaneous emptying of alveoli with very different V/Q ratios and therefore a minimal slope to the expiratory plateau). Thus, it is not surprising that the increase in airway resistance that is associated with bronchospasm,

Journalof the American Association of Nurse Anesthetists

which causes a wide spread in the V/Q ratios and emptying times, to be tightly correlated with an increase in the slope of the alveolar plateau (see Figure 4).18

Figure 5 Biphasic C02 excretion waveform during manual intermittent positive-pressure breathing

mm

Figure 4 V/Q ratio shifts as a result of bronchospasm

20 10

Bronchospasm spreads the

C02 Waveform

.m..

distribution of Rairw, V/Q ratios, and [002] -

slope phase III (Reprinted with permission from Nichols K,Benumof JL.19)

Figure 6 Biphasic and normal capnograms High Rairw

V/Q

o LOW Rairw

t[C02]

tV/Q \ [C02]

Time

The phase III alveolar plateau slopes upward to the right due to a spread in V/Q ratios from high (early phase III) to low (late phase III). Bronchospasm spreads the distribution of V/Q ratios and increases the slope of phase III. a Biphasic waveform in patients who have markedly different individual lungs. Theoretically, if the

continuum of the V/Q ratios is broken into two distinctly different lung regions (a low-resistance, high V/Q, low co2 concentration region and a highresistance, low V/Q, high co2 concentration region), a biphasic co2 excretion waveform might be expected. Situations in which such a biphasic co02 waveform have been described are the lateral decubitus position in which the nondependent lung has relatively low airway resistance, high V/Q ratio, and low co2 concentration, compared with the dependent lung, in a patient with severe rotary kyphoscoliosis causing severe compression of one lung (Figure 5),19 and a major mainstem bronchial intubation (Figure 6). 20

Some patients with chronic obstructive pulmonary disease may also display a slight biphasic expiratory plateau if they have, throughout both

April 1998/ Vol. 66/No. 2

Left panel (read right to left): Biphasic capnogram during right mainstem bronchial intubation. Each horizontal line indicates 10 mmHg o02 concentration. The first co2 concentration peak ranged from 21 to 23 mmHg, and the second co2 concentration peak ranged from 26 to 29 mmHg. No spontaneous ventilation was apparent during this period of controlled ventilation. Right panel (read right to left): Normal appearing capnogram after the tip of the endotracheal tube was pulled back above the tracheal carina. Each horizontal line indicates 10 mmHg o02 concentration; the endtidal co2 concentration at the end of this breath was 28 mmHg. (Reprinted with permission from Gilbert D,Benumof JL.20 )

lungs, two distinct populations of alveoli with very different time constants. In this situation, rapidly exchanging alveolar spaces are over-inflated during inspiration (their compliance is high) so that their co2 concentration is low, whereas slower exchanging alveoli empty only during the latter part of exhalation, releasing a higher co2 content. 21 In patients with active expiratory efforts, a similar pattern may also be precipitated by airway closure due to increased intrapleural pressure during expiration. 21 Finally, spontaneous breathing effort during a mechanical positive-pressure breath will create a cleft in the expiratory plateau and therefore a biphasic appearance.

173

Sampling line leak = peculiar waveform. When

there is a leak in the sampling line and the sampling line can entrain room air, the alveolar plateau will be artificially low. In addition, when the next inspiration forces gas through the sampling line at a faster rate so that room air is no longer entrained, a peak will follow the low plateau and the peak will be equal to the true ETCO2 (undiluted end-tidal gas being pushed through the sampling line) (Figure 7).22 The Paco2 -PETco2 gradient = alveolar dead

Figure 7 c02 excretion waveforms INTERNAsTIONAL SAeAc.e e.,,..,.

ALLEGENY

Dles

401-6

-

rimes

NEDIC

1284113t Ue.'r

12411912-01-86

TECHNOLOGY

ae

"

pIph

Pkrasel 5.0. Se

[a

hM

00.0 006-----25.5 SYS/DIA MAP/HR 02iN20 PiC02/PeC02 RRI -----99% 36.4 33.4 P 5 Case Time 02:05:441 002 Waveform

mm

.....

20.......

IYITENNATIOML

ALLEOHENv Dale+

ls e r84 6 ~m

1l-Ol-0

..................... Al ....... 25 seconds

~

30 20 ................

T

NMDlCAL

12:43:47 12-01-86

ft

TECNOLGOOGY

U3sers

1.0.

Pa

S.

1E1U

00.0 006------29.1 02'N20 PiC02/PeCO2 RR'P] SYS/DIA MAP/HR 72.1 35.7 99 1

Case Time02:08:13

mm

1

total dead space in the lung.3 "7 The alveolar concentration of C02 (PACO2) is ordinarily just slightly

less than the PVco2 but slightly higher than Paco2 (Figure 8). However, the alveolar concentration of C02 is ordinarily diluted by alveolar and anatomical dead space gas that has no C02 in it so that the ETCO2 ordinarily is less than the Paco2 (Figure 8). If the lungs are small (reduced functional residual capacity) and homogeneous (normal V/Q relationship and no alveolar dead space), and the anatomical dead space is small (these conditions occur in healthy, supine, term, pregnant women about to undergo cesarean section, patients having postpartum tubal ligations, and women in early pregnancy) 2 3 -2 5 and in exercise,2 6 then PACO2 is only diluted to a small extent by C02 free gas and PETCO2 may be greater than Paco2. The existence of negative arterial to end-tidal Pco2 gradients is best un-

derstood if it remembered that Paco2 represents the temporal and spatial mean alveolar Pco2 (i.e.,

.................................

..................................

............

3 0

space. The Paco2-PETC02 gradient is a function of the temporal sequence of alveolar emptying (i.e., the slope of the phase III of the single breath test and how high the C02 concentration rises) and the

--

BPS

C02 Waveform

physiologic integrator), whereas the ETCo2 is the highest (peak) alveolar Pco 2 coming from slow and low V/Q areas (especially when the tidal-volume is large and the respiratory rate is low).3'23-26 However, normally the amount of anatomical dead space

is large enough so that the Paco2-PETC02 is slightly 3 6 positive (by 2 to 4 mmHg). "1

Figure 8 Diagram of Paco2-PETco2 gradient Pa002 - PET0O2 gradient = dead space

25 seconds Top panel: Photograph of a C02 excretion waveform when the 002 sampling line was loosely connected to the 002 analyzer sample port. The C02 excretion waveform consists of a long low plateau followed by a brief peak. This patient was being ventilated with an inspired 02 concentration of 30% and an inspired N20 concentration of 70% (as indicated by flowmeter settings). Bottom panel: Photograph of the C02 excretion waveform from the same patient as inthe top panel, but when the connection between the C02 sample line and C02 analyzer sample port was made tight. The 002 excretion waveform is now almost square wave or rectangular. Note that the mean exhaled 02 and N20 concentrations are now near the inspiratory settings. (Reprinted with permission from Zupan J, Martin M,Benumof JL.22)

174

PET0O2

PET02

t

N

Dead space

j

C1 (a

PAC02

0 (O

co

Coa

o PA002

CD

The Pa002-PET0O2 gradient is directly proportional to the amount of alveolar dead space because 002-free gas from alveolar dead space dilutes 002-containing gas from gas exchanging alveolar ventilation (PA002) (The origin of PA02 is 002 in the mixed venous blood (PV00) and decreases the concentration, usually below the arterial level (Pa002).

Journalof the American Association of Nurse Anesthetists

As alveolar dead space progressively increases, the Paco 2-PETC0 2 gradient progressively increases." 27 In patients with pulmonary disease, the Paco 2-PETC02 gradient increases unpredictably by 10 to 20 mmHg or more with the result that PETC02 is no longer a reliable reflection of the

effectiveness of ventilation and Paco2. 28 Although trends in PETCO2 may still be

useful, 16 one author found that the trends were not useful in patients with severe lung disease. 29 The reasons trends in Paco2-PETC02 may not follow Paco 2 in patients with severe lung disease is because a fall in PETCO2 may be associated with an

equal fall in Paco 2 (no change in distribution of V/Q), with a greater fall in Paco 2 (due to better ventilation and improvement in V/Q matching), or with a constant or increased Paco2 (due to increased dead space and worsening of V/Q matching). All authors agree that in patients with severe respiratory failure, the Paco 2-PETC0 2 gradient is usually a good index of efficiency of ventilation and VD/VT. Causes of increased alveolar dead space include decreased pulmonary blood flow (decreased cardiac output, and/or pulmonary artery pressure), pulmonary embolization, kinking of pulmonary vessels, thrombosis in the pulmonary circulation, application of positive end-expiratory pressure, and significant dilation of the tracheal bronchial tree (Figure 8). A large shunt across the lungs will increase the Paco 2-PETC02 gradient to a small extent (due to an increase in Paco2 due to admixture of PVco 2 through the shunt).27 During thoracotomy, the Paco 2-PETC02 from the dependent and nondependent lungs alone (individually) is mainly a function of the pulmonary artery pressure. In one study in the lateral position, Paco2-PETC0 2 was zero for the lower lung and 11 mmHg for the upper lung. 30 It may be surmised that if PETCO2 had been measured in the combined expirate, the Paco2-combined PETCO2

difference would be large, expecially since the upper lung has more volume to empty (i.e., it makes the greater contribution to late expiration). Incision of the chest wall produced an increase in mean pulmonary artery pressure that was associated with an increase in c02 elimination by the upper lung and a decrease in dead space."' Consequently, a surgical stimulation-induced increase in mean pulmonary artery pressure causes a decrease in upper lung Paco2-PETCO 2 .

Phase IV: The inspiratory downstroke Soon after inspiration begins, fresh gas from the breathing circuit washes co2 from the previous exhalation away from the co2 sampling site. Because the volume of gas at the sampling site is April 1998/ Vol. 66/No. 2

small, the inspiratory downstroke (like the expiratory upstroke) is steep. A missing inspiratory valve or an inspiratory valve that is incompetent during exhalation allows co2-containing gas to go up the inspiratory limb. In this case, during inspiration the co2 previously exhaled into the inspiratory hose is pushed back into the patient's lung and the inspiratory downstroke becomes prolonged and slanted. Summary The capnogram is an extremely useful breathby-breath monitor of co2 exhalation. As such, it should be considered by anesthetists as a vital sign. Furthermore, the diagnosis of significant physiology and pathophysiology is contained within the shape of the capnogram. The diagnosis of this pathophysiology is best made by a systematic analysis of the four phases of the capnogram. REFERENCES (1) Clark JS, Votteri B, Ariagno RL, et al. Noninvasive assessment of blood gases. Am Rev Respir Dis. 1992;145:220-232. (2) Badgwell JM, McLeod ME, Lerman J, Creighton RE. End-tidal Pco2 measurements sampled at the distal and proximal ends of the endotracheal tube in infants and children. Anesth Analg. 1987;66:959964. (3) Fletcher R. The arterial-end-tidal co0 difference during cardiothoracic surgery. i CardiothoracVasc Anesth. 1990;4:105-117. (4) Murray IP, Modell JH. Early detection of tube accidents by monitoring carbon dioxide concentration in respiratory gas. Anesthesiology. 1983;59:344-346. (5) Guggenberger H, Lenz G, Federle R. Early detection of inadvertent oesophageal intubation: Pulse oximetry vs capnography. Acta AnaesthesiolScand. 1989;33:112-115.

(6) Dunn SM, Mushlin PS, Lind LJ, Raemer D. Tracheal intubation is not invariably confirmed by capnography. Anesthesiology. 1990;73: 1285-1287. (7) Isserles SA, Breen PH. Can changes in end-tidal Pco2 measure changes in cardiac output. Anesth Analg. 1991;73:808-814. (8) Shibutani K, Whelan G, Zung N, Ferlazzo P. End-tidal co,: A clinical noninvasive cardiac output monitor. Anesth Analg. 1991 ;72:52251. (9) Swedlow DB. Capnometry and capnography: The anesthesia disaster early warning system. Seminars in Anesthesia. 1986;5:194-205. (10) Nielsen MS, Fitchet A, Saunders DA. Monitoring cardiopulmonary resuscitation by end-tidal carbon dioxide concentration. BMJ. 1990;300:1012-1013. (11) Lambert Y, Cantineau JP, Merckx P, Bertrand C, Duvaldestin P. Influence of end-tidal cos monitoring on cardiopulmonary resuscitation. Anesthesiology. 1992;77:A1081. (12) Falk JL, Rackow EC, Weil MH. End-tidal carbon dioxide concentration during cardiopulmonary resuscitation. N Engl] Med. 1988;318: 607-611. (13) Garnett A, Ornato JP, Gonzalez ER, Johnson B. End-tidal carbon dioxide monitoring during cardiopulmonary resuscitation. JAMA. 1987;257:512-515. (14) Trevino RP, Bisera J, Weil MH, Rackow EC, (;rundler W(;. Endtidal co2 as a guide to successful cardiopulmonary resuscitation: A preliminary report. Crit CareMed. 1985;13:910-911. (15) Callaham M, Barton C. Prediction of outcome of cardiopulmonary resuscitation from end-tidal carbon dioxide concentration. Crit CareMed. 1990;18:358-362. (16) Tobin MJ. Respiratory monitoring in the intensive care unit. Am Rev Respir Dis. 1988;138:1625-1642. (17) Fletcher R, Jonson B. Deadspace and the single breath test for carbon dioxide during anesthesia and artificial ventilation. BrJAnaesth. 1984;56:109-119.

175

(18) Watson R, Benumof JL, Clausen J, Ozaki G. Expiratory cot plateau slope predicts airway resistance. Anesthesiology. 1989;71:A1072. (19) Nichols K, Benumof JL. Biphasic carbon dioxide excretion waveform in a patient with severe kyphoscoliosis. Anesthesiology. 1989;71: 986-987. (20) Gilbert D, Benumof JL. Biphasic carbon dioxide elimination waveform with right mainstem bronchial intubation. Anesth Analg. 1989;69:829-832. (21) Carlon GC, Ray C, Miodownik S, Kopec I, Groeger JS. Capnography in mechanically ventilated patients. Crit Care Med. 1988;16:550-556. (22) Zupan J, Martin M, Benumof JL. End-tidal co2 excretion waveform and error with gas sampling line leak. Anesth Analg. 1988;67:579581. (23) Shankar KB, Moseley H, Kumar Y, Vemula V. Arterial to endtidal carbon dioxide tension difference during cesarean section anesthesia. Anaesthesia. 1986;41:698-702. (24) Shankar KB, Moseley H, Kumar Y, Vemula V, Krishnan A. Arterial to end-tidal carbon dioxide tension difference during anaesthesia for tubal ligation. Anaesthesia. 1987;42:482-486. (25) Shankar KB, Moseley H, Vemula V, Ramasamy M, Kumar Y. Arterial to end-tidal carbon dioxide tension difference during anaesthesia in early pregnancy. Can JAnaesth. 1989;36:124-127.

(26) Jones NL, Robertson DG, Kane JW. Difference between endtidal and arterial co2 in exercise. JAppl Physiol. 1979;47:954-960. (27) Good ML. Capnography: Uses, interpretation and pitfalls. ASA Refresher Courses for Anesthesiology. 1990;18(12):175-193. (28) Yamanaka MK, Sue DY. Comparison of arterial-end-tidal Pco2 difference and dead space/tidal volume ratio in respiratory failure. Chest. 1987;92:832-835. (29) Hoffman RA, Krieger BP, Kramer MR, et al. End-tidal carbon dioxide in critically ill patients during changes in mechanical ventilation. Am Rev Respir Dis. 1989;140:1265-1268. (30) Werner O, Malmkvist G, Beckman A, et al. Carbon dioxide elimination from each lung during endobronchial anaesthesia. BrJAnaesth. 1984;56:995-1001. (31) Werner O, Malmkvist G, Beckman A, et al. Gas exchange and haemodynamics during thoracotomy. BrJAnaesth. 1984;56:1343-1349.

AUTHOR Jonathan L. Benumof, MD, is professor of Anesthesia at the University of California at San Diego Medical Center, Department of Anesthesia, San Diego, California.

Why Choose between a Great Career and a Great Location? At University of North Carolina Hospitals You'll Have Both. Immediate openings for CRNAs. Must have a minimum of a Master's degree and be a graduate of an accredited school of nurse anesthesia, a graduate nurse anesthetist with certification pending, or a current certified nurse anesthetist, eager to practice all types of anesthetic techniques including regional anesthesia, general anesthesia, and invasive monitoring. As a Level I Trauma Center with 24 ORs, UNC Hospitals offers the CRNA opportunities to provide care to all surgery and diagnostic services including cardiac, burns, neuro, oncology, pediatrics, ENT, and comprehensive solid organ transplantation. Enjoy an anesthesia care team environment that supports collegial relationships. Flexible self-scheduling provides time off to enjoy all the cultural, academic and leisure activities of a university setting, not to mention the exceptional quality of life in Chapel Hill. UNC Hospitals is located on the prestigious University of North Carolina Campus and is centrally located between serene mountains and uncrowded beaches. Excellent salary and comprehensive benefit package, including a sign-on stipend and five paid educational days. For more information, please call 1-800-852-6877. Send resume to CRNA Recruiters, UNC Hospitals, Nurse Employment Office, 101 Manning Drive, Chapel Hill, NC 27514 AA/EOE or fax to (919) 966-6475. Visit us on the Internet at http://www.med.unc.edu/hosphr/

The University of North Carolina Hospitals

1-800-852-NURSE 176

Journalof the American Association of Nurse Anesthetists