| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2003;97:1680-1685 © 2003 International Anesthesia Research Society
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
We designed this study to ascertain whether, for the purpose of clinical interpretation, the direct measurement of O2 consumption with the PhysioFlex closed-circuit anesthesia machine and with the Deltatrac II indirect calorimeter are interchangeable. Oxygen consumption was measured using the two instruments successively in critically-ill, mechanically-ventilated patients. Measurements were recorded as the mean of 10 consecutive, minute-by-minute, stable readings. The degree of agreement between the measurements obtained with the two systems was estimated using Bland-Altman analysis and the intraclass correlation coefficient. Forty-four pairs of measurements made in 21 patients were analyzed, yielding a mean bias of 6.32 mL/min and limits of agreement of 40.28 and -27.63 mL/min. The intraclass correlation coefficient was 0.95, and the 95% confidence interval ranged from 0.91 to 0.97. The measurement of O2 consumption obtained with the PhysioFlex anesthesia machine is interchangeable with that obtained by indirect calorimetry.
IMPLICATIONS: The PhysioFlex anesthesia machine (Dräger Inc., Lübeck, Germany) is a closed circuit anesthesia delivery device. The oxygen delivered by this device to maintain a steady-state inspired oxygen concentration is therefore a measure of the patient’s oxygen consumption. This study was designed to evaluate the accuracy of the PhysioFlex for measuring oxygen consumption by comparing it with an established technology (Deltatrac II Calorimeter) for making this measurement.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Oxygen consumption (
O2) is a physiological variable that has been used as a target for goal-directed therapy in the perioperative period (1,2) and is useful for assessing the risk of postoperative complications (2). Two methods are generally used to determine O2 during the perioperative period: its estimation using the Fick equation (O2-Fick) and indirect calorimetry. The former is familiar to most clinicians, although it has some drawbacks such as the morbidity associated with insertion of a Swan-Ganz thermodilution catheter. Moreover, the method does not take into account O2 by the lungs (3), and because the calculations of O2 and of O2 delivery (DO2) share some variables (arterial O2 content and cardiac output), mathematical coupling of the data could make it difficult to interpret the O2/DO2 relationship (4). It has been recommended that interventions based on DO2 and O2 during the perioperative period should be based on direct O2 measurement (5). Indirect calorimetry is the method of choice for measuring O2 in clinical practice (6), although its use in patients on mechanical ventilation may be complicated by numerous inconveniences (7) that must be considered, because not all commercially available calorimeters deal with them successfully. The Deltatrac II is an indirect calorimeter that has been validated in the laboratory and clinical settings in intubated patients undergoing mechanical ventilation (8–10), although its application requires technical conditions and expertise that are difficult to satisfy in standard anesthetic practice.
Different techniques have been described to measure O2 on the basis of gas exchange during anesthesia, although none of them have come to be widely used in routine clinical practice. With the PhysioFlex (Dräger Inc., Lübeck, Germany) anesthesia machine, the fresh gas flow rate adjusts automatically to that taken up by the patient, thereby affording a continuous, noninvasive measurement of O2 (11,12).
This study was designed to ascertain whether the measurement of O2 with the Deltatrac II calorimeter (Datex Instrumentation, Helsinki, Finland; O2-Deltatrac) and the PhysioFlex anesthesia machine (O2-PhysioFlex) are interchangeable for the purpose of clinical interpretation in critically-ill patients on mechanical ventilation.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The Hospital Research Ethics Committee approved the study protocol, and written informed consent was obtained from the nearest patient relatives.
The study was performed in the Surgical and Medical Critical Care Units of the Fundación Hospital Alcorcón. Patients admitted from June 2000 to July 2001 were considered. Those older than 18 yr of age, subjected to tracheal intubation, and on volume-controlled mechanical ventilation were eligible. Exclusion criteria included pregnancy, fraction of inspired oxygen (FIO2) more than 0.6, positive end-expiratory pressure higher than 10 cm H2O, the presence of air leaks (i.e., around the endotracheal tube via thoracostomy tubes or bronchopleural fistulae), and concomitant use of renal replacement therapies. When more than one measurement was performed in the same patient, they were separated by at least 24 h. Aerosols were not used during the study period, and at least 6 h were allowed to elapse before testing in patients who had undergone anesthesia involving vapors.
To achieve stable O2 during measurements, the following precautions were taken: (a) during measurements and for the 30 min before testing, no nursing activities such as tracheal aspiration or physiotherapy were performed; (b) ventilator settings were not changed 90 min before and during measurements; (c) the required hemodynamic stability was assessed according to the criteria of the attending physician, the rate of vasoactive drug infusion was not modified, and no new drugs were administered during the period of the determinations; (d) the absence of changes in body temperature of more than 0.5°C was confirmed during the measurements; (e) in no case was the schedule for analgesic and sedative administration changed 30 min before or during the determinations, and none of the patients received neuromuscular blocking drugs during measurements or in the 12 h preceding them; and (f) in patients receiving enteral or parenteral nutrition, the rate and schedule for feeding were not modified 4 h before or during the determinations.
All patients were ventilated using a volume-controlled mode (intermittent positive-pressure ventilation) provided by an Evita-2-dura ventilator (Dräger). During measurement with the Deltatrac II, all ventilator settings were maintained. When the PhysioFlex was used, the ventilatory settings were the same as those programmed on the patient’s ventilator.
The sample size was predetermined on the basis of an expected degree of agreement of 0.85, estimated with the intraclass correlation coefficient, a minimum acceptable agreement of 0.70, a confidence level of 5%, and a power of 80%. It was considered that 46 observations would be required to satisfy these preconditions.
The PhysioFlex anesthesia machine is designed to work as a closed-anesthetic circuit, the functioning of which has been described in detail previously (11,12) and is based on that of the closed-circuit spirometer. In the latter system, an individual breathes connected to an O2-filled spirometer, and the emitted CO2 is trapped by an absorbing unit. Based on volume reduction as a function of time, O2 consumption can be determined. In the PhysioFlex system, the spirometer is replaced by a chamber, the expired CO2 is completely eliminated by means of an absorber consisting of soda lime installed in the expiratory limb of the circuit, and the O2 equals the O2 flow rate required to keep the O2 concentration in the system constant.
Inspiratory O2 concentration is continuously measured by means of a paramagnetic O2 sensor, whereas a triple-channel infrared spectrometer monitors expiratory CO2, inspiratory N2O, and the inspiratory and expiratory concentrations of volatile anesthetic by sampling from the patient connection piece. The PhysioFlex system compares the O2 concentration measured in the circuit with the prefixed FIO2 value. If the measured value is comparatively less, exact-volume O2 is injected in the circuit to reach the prefixed value. Because this is a closed circuit, the injected O2 will be equal to the patient O2 uptake.
Before each measurement, the system was calibrated according to the instructions and using the calibration gas provided by the manufacturer. The PhysioFlex displayed minute-by-minute data on O2, expressed in milliliters per minute, and stored them in memory so that they could be loaded on a personal computer using an Excel 97 spreadsheet (Microsoft Corp, Redmond, WA).
The Deltatrac II Metabolic Monitor is an open-circuit indirect calorimeter, the operation of which has been described in detail elsewhere (13). Deltatrac II measures O2 in mechanically ventilated patients as follows: the expired gas from the ventilator passes into a 4-L mixing chamber from which samples are obtained to analyze the mixed expired O2 concentration (FeO2) and mixed expired CO2 concentration (FeCO2). Expired gas leaves the chamber and is mixed with a flow of room air large enough to ensure that the total flow (Q) is constant and equal to the flow produced by the constant flow generator of the apparatus (40 L/min). The concentration of CO2 expired in this gas flow is measured (Fe*CO2), and CO2 production is calculated using the formula: CO2 = Q x Fe*CO2. The respiratory quotient (RQ) is calculated using the Haldane transformation: RQ = [1 - FIO2 ]/[((FIO2 - FeO2)/FeCO2) - FIO2 ]. FIO2 is measured in the inspiratory limb of the ventilator circuit. O2 is calculated according to the formula: O2 = CO2/RQ.
Alcohol-burning flow-testing, pressure calibration, and, before each measurement, warmup and gas calibrations were performed according to the instructions of the manufacturer. Deltatrac II prints out O2 minute-by-minute in milliliters per minute. Only those determinations in which the RQ was between 0.67 and 1.3 were accepted in this study. Both Deltatrac II and PhysioFlex report O2 for Standard Temperature and Pressure, Dry conditions.
The features of the two measurement systems prevent them from being used simultaneously. Thus, determinations of O2 were performed successively. The order in which they were performed followed a random number table. Each measurement lasted long enough to achieve 10 min of stable recording (a variation of <10% in the minute-by-minute recordings). The mean of the findings in these 10 consecutive, minute-by-minute readings was considered to be the result of the measurement. The RQ was calculated as the ratio between mean CO2 and mean O2, measured with the Deltatrac II calorimeter, over the 10-min period.
To estimate the degree of agreement, the difference between each set of two consecutive measurements was determined, and the mean difference (bias) was calculated. The mean bias represents the degree of systematic difference between methods of measurement and is determined by summing the differences between paired measurements and dividing by the number of paired measurements (14). The limits of agreement (bias ± 2 SD) defined the concordance interval, which encompassed 95% of the differences between each set of two consecutive measurements in each patient. Confidence intervals were estimated for the limits of agreement.
In the graphical representation, the difference between each set of two consecutive measurements was determined and entered on the ordinate axis; the mean value of the measurements with both methods was entered on the abscissa axis. The regression line in the plot of the difference against the average was presented to know if there was a trend in the bias (a tendency for the mean difference to increase or decrease with increasing magnitude of measurements) (15). The relative difference ((O2 Deltatrac–O2 PhysioFlex) x 100/[(O2 PhysioFlex + O2 Deltatrac)/2]) was estimated and expressed graphically.
The level of agreement was also estimated with a summary variable, the intraclass correlation coefficient and its 95% confidence interval. An absolute value between 0.80 and 1 was considered a very good agreement.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The following analysis was applied to 44 measurements, performed in 21 patients, which fulfilled all the requested criteria. One pair of measurements was performed in 5 patients, 2 pairs in 11, 3 pairs in 3, and 4 pairs in 2. Patient characteristics appear in Table 1, whereas Table 2 shows the determinations of
O2 with the two instruments.
|
|
Figure 1 represents the differences related to the means of measurements with the two systems. It also shows a linear regression analysis that allows us to consider an association between the differences and the magnitude of the measurements (P = 0.4148) to be unlikely. Figure 2 corresponds to the same graph with the differences expressed in relative terms (%). In 36 of the 44 pairs of measurements (81.8%), the relative difference was <10%, and in no case was the relative difference more than 20%; the mean relative difference was 2.9%, and the limits of agreement were 17.7% and -11.8%.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In this study
O2 was measured in critically ill patients using a closed anesthetic circuit, (O2-PhysioFlex) and compared with the calorimetric measurement (O2-Deltatrac), which has been validated in intubated patients undergoing mechanical ventilation (8–10). We found a bias of 6.32 mL/min, limits of agreement of 40.28 and -27.63 mL/min, and an intraclass correlation coefficient of 0.95, which corresponds to a very good degree of agreement. O2-Deltatrac was higher than O2-PhysioFlex, with a mean relative difference of 2.9%.
Two methods of measurement can be considered interchangeable if the difference observed is not more than the difference considered to be clinically acceptable. In this respect, Weyland et al. (16) accepted limits of agreement of less than ±20%, whereas Epstein et al. (17), on comparing O2 calorimetric measurement and O2-Fick, regarded as clinically acceptable a bias between the 2 methods of no more than 20 mL · min-1 · m-2, a difference that can represent 14%–15% of the total O2 of the body. Our study has estimated limits of agreement less than those established as acceptable in the literature. This positive assessment of the agreement between the two systems, by means of limits of agreement, is reinforced by the intraclass correlation coefficient, which was found to be excellent.
There is no "gold standard" for measuring O2 under the conditions of our study. In practice, indirect calorimetry is considered the reference method for O2 measurement comparison. Indirect calorimetry requires specific equipment and skilled personnel; this and the technical difficulties posed by its application to patients on mechanical ventilation have caused estimation based on the Fick method to be the most widely used alternative. Different authors (3,8,18–20) that have compared O2-Deltatrac and O2-Fick in patients undergoing controlled mechanical ventilation have found poor agreement between the two methods. This lack of agreement is caused partly by errors in the measurement techniques and partly by the fact that the O2-Fick does not take into account lung O2 because lung O2 represents only approximately 5% of whole body O2 in anesthetized healthy patients (21) but can reach 30% (22) in patients with lung infection. The O2-Fick measurement has many potential sources of error (hemoglobin concentration, O2 saturation, O2 partial pressures in arterial and venous blood measurements used, and the technique used to determine cardiac output), and consequently, variability in O2-Fick measurement can contribute to the differences observed when compared with the calorimetry recordings (6,20). However, the measurement of O2 via indirect calorimetry in patients subjected to mechanical ventilation must deal with a number of sources of error (7), such as high or fluctuating FIO2, humidified gases, high pressure in the ventilator circuit, positive end-expiratory pressure, leaks in the system preventing complete gas collection, and increased respiratory rate. As has been mentioned, correct functioning of the Deltatrac II system has been confirmed in patients in situations similar to those found in our own study (8–10).
One limitation to our comparison is the impossibility of simultaneously performing the measurements with a closed anesthetic circuit and an open-circuit indirect calorimeter. Thus, the possibility that patient-related factors may have caused the differences in measurements cannot be eliminated. It was decided to make this comparison in critical care patients rather than in anesthetized patients because of the difficulty in achieving periods of stable O2 during surgery. Moreover, by performing the study in patients who were critically ill because of a variety of causes, we attempted to record measurements in a wide range of O2. We consider that the patient selection criteria, the precautions taken to achieve periods of stable O2, and the fact that the order of the measurements followed a sequence obtained from a random number table sufficed to prevent systematic error in the results.
Use of the Deltatrac II to measure O2 during general anesthesia poses inconveniences such as the need to use an open anesthetic circuit, the impossibility of using anesthetic gases and vapors (23) because they interfere with the function of the O2 sensor, and the fact that the Haldane equation is not applicable if the respiratory gases contain components other than O2, CO2, and N2. Moreover, because O2 consumption is measured based on the formula O2 = CO2/RQ, the O2 value may be influenced by elimination of the CO2 insufflated for performing laparoscopic surgical techniques. If the above considerations are taken into account, Deltatrac II can be used to measure O2 during anesthesia.
This is the first report on the comparison of O2 measurement obtained with a closed anesthesia circuit and with a calorimeter in the clinical setting. Recently, Schindler et al. (24) in an in vitro study with a calibrated lung model, evaluated O2-PhysioFlex and found it recommendable to average the O2-PhysioFlex values for 10 minutes to avoid the minute-by-minute changes. When that measure was adopted and FIO2 was <0.85, O2-PhysioFlex agreed with the lung model O2 and rapidly responded to changes in the latter. However, when FIO2 exceeded 0.85, PhysioFlex overestimated O2. Furthermore, in anesthetized dogs, the authors compared O2-PhysioFlex with O2-Fick, reporting similar results when FIO2 was <0.85. In contrast, in anesthetized patients, they found O2-PhysioFlex to systematically overestimate O2-Fick (bias, 52 mL/min; SD, 40). The authors attributed this to the methodological imprecision of O2-Fick and to error caused by small arteriovenous differences in O2 content and high cardiac output. The loss of accuracy of the O2-PhysioFlex system with FIO2 values more than 0.85 was attributed to systematic error resulting from imprecise oxygen replacement and, in the in vivo tests, from the accumulation of foreign gases. Brandi et al. (25) compared O2-PhysioFlex with O2-Fick in 5 patients during anesthesia, finding a mean bias of 5.6 mL · min-1 · m-2 (SD, 16 mL · min-1 · m-2).
The measurement of O2 in patients during anesthesia has not been implanted as habitual monitoring in routine anesthetic practice. The reasons that might explain this situation include the fact that the measurement reference method (indirect calorimetry) presents technical difficulties that make it difficult to apply in anesthetized patients. However, O2-Fick estimation requires the placing of a pulmonary artery catheter that would not be indicated in most cases, implying a risk, and measurement is not as accurate as in the case of calorimetry. Under the conditions of our study, we found continuous and noninvasive measurement of O2 by the PhysioFlex anesthesia machine to be no different from that obtained with indirect calorimetry using the Deltatrac II system, and consequently, it may be useful in clinical practice.
|
Acknowledgments |
|---|
The authors would like to thank Dräger Medical Hispania, S.A. (Madrid, Spain), for the generous loan of the PhysioFlex anesthesia machine used in this study.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Shoemaker WC, Kram HB, Appel PL, Fleming AW. The efficacy of central venous and pulmonary artery catheters and therapy based upon them in reducing mortality and morbidity. Arch Surg 1990; 125: 1332–7.
[Abstract/Free Full Text] - Shoemaker WC, Appel P, Bland R. Use of physiologic monitoring to predict outcome and to assist in clinical decisions in critically ill postoperative patients. Am J Surg 1983; 146: 43–50.[Web of Science][Medline]
- Keinänen O, Takala J. Calculated versus measured oxygen consumption during and after cardiac surgery: is it possible to estimate lung oxygen consumption? Acta Anaesthesiol Scand 1997; 41: 803–9.[Web of Science][Medline]
- Phang PT, Cunningham KF, Ronco JJ, et al. Mathematical coupling explains dependence of oxygen consumption on oxygen delivery in ARDS. Am J Respir Crit Care Med 1994; 150: 318–23.[Abstract]
- Pulmonary Artery Catheter Consensus Conference. Consensus statement. Crit Care Med 1997; 25: 910–25.[Web of Science][Medline]
- Hanique G, Dugernier T, Laterre PF, et al. Evaluation of oxygen uptake and delivery in critically ill patients: a statistical reappraisal. Intensive Care Med 1994; 20: 19–26.[Web of Science][Medline]
- Weissman C, Kemper M. Metabolic measurements in the critically ill. Crit Care Clin 1995; 11: 169–97.[Web of Science][Medline]
- Takala J, Keinänen O, Väisänen P, Kari A. Measurement of gas exchange in intensive care: laboratory and clinical validation of a new device. Crit Care Med 1989; 17: 1041–7.[Web of Science][Medline]
- Ronco JJ, Phang PT. Validation of an indirect calorimeter to measure oxygen consumption in critically ill patients. J Crit Care 1991; 6: 36–41.[Web of Science]
- Tissot S, Delafosse B, Bertrand O, et al. Clinical validation of the Deltatrac monitoring system in mechanically ventilated patients. Intensive Care Med 1995; 21: 149–53.[Web of Science][Medline]
- Verkaaik APK, Van Dijk G. High flow closed circuit anaesthesia. Anaesth Intensive Care 1994; 22: 426–34.[Web of Science][Medline]
- Nathan N, Sperandio M, Erdmann W, et al. PhysioFlex: a target-controlled self-regulating closed-circuit inhalation anesthesia regulator. Ann Fr Anesth Reanim 1997; 16: 534–40.[Web of Science][Medline]
- Branson RD. The measurement of energy expenditure: instrumentation, practical considerations and clinical applications. Respir Care 1990; 35: 640–59.
- Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–10.[Web of Science][Medline]
- Bland JM, Altman DG. Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 1995; 346: 1085–7.[Web of Science][Medline]
- Weyland W, Weyland A, Gefeller O, et al. A ventilator with an integrated gas-exchange monitoring function. Crit Care Med 1994; 22: 864–71.[Web of Science][Medline]
- Epstein CD, Peerless JR, Martin JE, Malangoni MA. Comparison of methods of measurements of oxygen consumption in mechanically ventilated patients with multiple trauma: the Fick method versus indirect calorimetry. Crit Care Med 2000; 28: 1363–9.[Web of Science][Medline]
- Bizouarn P, Soulard D, Blanloeil Y, et al. Oxygen consumption after cardiac surgery: a comparison between calculation by Fick’s principle and measurement by indirect calorimetry. Intensive Care Med 1992; 18: 206–9.[Web of Science][Medline]
- Bizouarn P, Blanloeil Y, Pinaud M. Comparison between oxygen consumption calculated by Fick’s principle using a continuous thermodilution technique and measured by indirect calorimetry. Br J Anaesth 1995; 75: 719–23.
[Abstract/Free Full Text] - Walsh TS, Hopton P, Lee A. A comparison between the Fick method and indirect calorimetry for determining oxygen consumption in patients with fulminant hepatic failure. Crit Care Med 1998; 26: 1200–7.[Web of Science][Medline]
- Loer SA, Scheeren TW, Tarnow J. How much oxygen does the human lung consume? Anesthesiology 1997; 86: 532–7.[Web of Science][Medline]
- Hensel M, Kox WJ. Increased intrapulmonary oxygen consumption in mechanically ventilated patients with pneumonia. Am J Respir Crit Care Med 1999; 160: 137–43.
[Abstract/Free Full Text] - Scheeren TWL, Krossa M, Meriläinen P, Arndt JO. Error in measurement of oxygen and carbon dioxide concentrations by the DeltatracII metabolic monitor in the presence of desflurane. Br J Anaesth 1998; 80: 521–4.
[Abstract/Free Full Text] - Schindler AW, Scheeren TWL, Picker O, et al. Accuracy of feedback-controlled oxygen delivery into a closed anaesthesia circuit for measurement of oxygen consumption. Br J Anaesth 2003; 90: 281–90.
[Abstract/Free Full Text] - Brandi LS, Giunta F, Oleggini M, et al. Measured and predicted values of oxygen consumption during isoflurane anesthesia in man. Adv Exp Med Biol 1994; 345: 763–73.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. D. Branson and J. A. Johannigman The Measurement of Energy Expenditure Nutr Clin Pract, December 1, 2004; 19(6): 622 - 636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hofland, R. Tenbrinck, A. Gonzalez-Arevalo, J. I. Gomez-Arnau, and S. Garcia-del-Valle Oxygen Consumption: Validation of the Closed-Circuit PhysioFlex(R) Anaesthesia Machine * Response Anesth. Analg., July 1, 2004; 99(1): 306 - 306. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
