- Email: [email protected]
Improved flow and pressure capabilities of the Datex-Ohmeda SmartVent anesthesia ventilator
Original Contributions Improved Flow and Pressure Capabilities of the DatexOhmeda SmartVent Anesthesia Ventilator Jeffrey A. Katz, MD,* Richard H. Kallet, MS, RRT,† James A. Alonso, RRT,‡ James D. Marks, MD, PhD§ Department of Anesthesia, UCSF/Mount Zion Medical Center; Department of Anesthesia, San Francisco General Hospital; Department of Anesthesia, University of California School of Medicine, San Francisco, CA
*Professor of Clinical Anesthesia, University of California, San Francisco; Chief, Department of Anesthesia, UCSF/Mount Zion Medical Center †Clinical Research Coordinator, University of California, San Francisco ‡Clinical Research Associate, University of California, San Francisco §Professor of Anesthesia, University of California, San Francisco Address correspondence to Dr. Katz at the Department of Anesthesia, UCSF/Mount Zion Medical Center, Hellman Bldg., Room C355, 1600 Divisadero St., San Francisco, CA 94143-1605, USA. E-mail: katzj@anesthesia. ucsf.edu Supported in part by a grant from DatexOhmeda, Madison, WI. Received for publication June 25, 1999; revised manuscript accepted for publication November 22, 1999.
Study Objective: To compare the flow and pressure capabilities of the Datex-Ohmeda SmartVent (Ohmeda 7900, Datex-Ohmeda, Madison, WI) to previous Ohmeda (7810 and 7000, Datex-Ohmeda, Madison, WI) anesthesia ventilators. To determine airway pressure and minute ventilation thresholds for intraoperative use of a critical care ventilator. Design: Three anesthesia ventilators and one critical care ventilator (Siemens Servo 900C, Siemens, Solna, Sweden) were studied in a lung model. Retrospective medical record review. Setting: Research Laboratory and Critical Care Unit of a Level I Trauma Center. Patients: 145 mechanically ventilated patients treated for acute respiratory failure who underwent 200 surgical procedures. Interventions: The effect of increasing pressure on mean inspiratory flow was determined by cycling each ventilator through increasing restrictors. Maximum minute ventilation was measured at low compliance (10 –30 mL/cm H2O), positive end-expiratory pressure (PEEP) (0 –20 cm H2O), and increased airway resistance (⬃19 and ⬃36 cm H2O/L/sec) in a mechanical lung model. Measurements and Main Results: Flow, volume, and pressure were measured with a pulmonary mechanics monitor (BICORE CP-100, Thermo Respiratory Group, Yorba Linda, CA). Preoperative peak airway pressure and minute ventilation (V˙E) were extracted from the medical record. Mean inspiratory flow declined with increasing pressure in all anesthesia ventilators. The SmartVent and the 7810 produced greater mean inspiratory flow than did the 7000 ventilator. As compliance progressively decreased, the Siemens, the SmartVent, and the 7810 ventilators maintained V˙E compared to the 7000 ventilator. The Siemens and the SmartVent maintained V˙E with PEEP, compared to the 7810 and 7000 ventilators. During increased airway resistance, maximal V˙E was lower for all ventilators. The SmartVent met the ventilation requirements in 90% of the patients compared to 67% of patients with the 7000 ventilator. Conclusion: The improved pressure and flow capabilities of the SmartVent increase the threshold for using a critical care ventilator intraoperatively to a peak airway pressure ⬎65 cm H2O and/or V˙E ⬎18 L/min. © 2000 by Elsevier Science Inc. Keywords: Equipment, ventilators: flow characteristics; lung: acute respiratory failure, airway resistance, compliance; ventilation: positive end-expiratory pressure.
Journal of Clinical Anesthesia 12:40 – 47, 2000 © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
0952-8180/00/$–see front matter PII S0952-8180(99)00140-3
Flow capability of anesthesia ventilators: Katz et al.
Figure 1. Model for determining effects of increasing airway pressure on mean inspiratory flow and maximum minute ventilation. Each ventilator/anesthesia machine was connected to a mechanical lung with adjustable compliance. Resistance was altered by changing the mechanical resistor (R). A separate transducer measured pressure inside the mechanical lung.
Introduction
Materials and Methods
Patients with acute lung injury requiring mechanical ventilation often undergo general anesthesia for necessary surgical procedures. Ventilatory support of these patients may require a high minute ventilation (V˙E), increased airway pressure, and high levels of positive end-expiratory pressure (PEEP).1 We previously reported on the pressure and flow capabilities of anesthesia ventilators showing that maximal mean inspiratory flow, and therefore V˙E capacity, decreased with increasing airway pressure.2 Additionally, maximum V˙E can be calculated for a specific anesthesia ventilator using the mean inspiratory flow (tidal volume [VT] divided by inspiratory time [TI]) versus airway pressure curves, and the maximum allowable inspiratory duty cycle (the ratio of TI to total breathing cycle duration [TTOT]). Thus, V˙E could be expressed as a function of a driving (VT/TI) and timing (TI/TTOT) mechanisms: V˙E ⫽ VT/TI ⫻ [TI/TTOT].3 This approach permits prediction of ventilator performance for patients with high V˙E requirements. For example, with the Ohmeda 7000 ventilator (Datex-Ohmeda, Madison, WI), a maximal mean inspiratory flow of 42 L/min could be delivered at a pressure of 50 cm H2O.2 Thus, with a peak airway pressure of 50 cm H2O and a TI/TTOT of 0.33, the Ohmeda 7000 ventilator could deliver a maximum of 14 L/min. Because many patients with acute respiratory failure have V˙E requirements greater than 15 L/min and peak airway pressures that exceed 50 cm H2O,1 we previously recommended use of a critical care ventilator when preoperative V˙E and airway pressures exceeded these values.2 Since our recommendation in 1989, Datex-Ohmeda has introduced the SmartVent anesthesia ventilator, which is purported to have improved flow and pressure capabilities. This ventilator should be able to deliver higher V˙E at increased airway pressures, thus minimizing the need for critical care ventilators for patients with acute respiratory failure requiring surgery. This study compares the pressure and flow capacity of the SmartVent to previous Ohmeda 7000 series anesthesia ventilators.
Three anesthesia ventilators (the Datex-Ohmeda SmartVent™, the Ohmeda 7810™, and the Ohmeda 7000™) and one critical care ventilator (the Siemens Servo 900C™, Siemens, Solna, Sweden, adapted for delivery of inhalational anesthetics) were studied in a lung model. All ventilators were inspected and calibrated to manufacturer specifications prior to the study. Each ventilator was tested as part of the anesthesia machine provided by the manufacturer. An oxygen gas flow of 5 L/min was provided via the fresh gas outlet for all ventilators except the Siemens Servo 900C, which was powered by its own blender. Low-compliance tubing of 0.6 mL/cm H2O (Connect Company™, Lea & Kay, San Francisco, CA) was used to complete the breathing circuit. Gas flow and pressure were measured by a flow sensor (Varflex, Bicore CP100, Thermo Respiratory Group, Yorba Linda, CA) attached to a pulmonary mechanics monitor (BICORE CP-100, Allied Healthcare Products, Riverside, CA). Tidal volume (VT) was obtained by time integration of the flow signal. The flow sensor was positioned immediately distal to the Ypiece of the circuit (Figure 1), so that delivered VT excludes volume loss due to compressibility. Prior to each ventilator study, the flow sensor was electronically calibrated, and verified with a Fisher Porter flow tube for flow and a 1-L super-syringe for volume. Pressure measurements were calibrated against a water column. A difference of less than 5% was regarded as sufficiently accurate. The accuracy of the measurements provided by this monitoring system has been found to be satisfactory.4 – 6 The BICORE CP-100 provided a real-time display of gas flow, VT, respiratory frequency, and pressure. The V˙E and breathing pattern, i.e., VT, respiratory frequency, the duration of inspiration and expiration, and the TI/TTOT were analyzed based on the flow signal. The data of ten consecutive respiratory cycles were averaged. The study was conducted in three parts. In part one, we measured the effect of increasing pressure on mean inspiratory flow by cycling each ventilator/anesthesia machine through progressively increasing resistors into a J. Clin. Anesth., vol. 12, February 2000
41
Original Contributions
Figure 2. Effect of increasing airway pressure on mean inspiratory flow for each ventilator/anesthesia machine. The Siemens 900C delivered pressure-independent flow at airway pressures less than 80 cm H2O. While mean inspiratory flow declined with increasing airway pressure in all anesthesia ventilators, both the Datex-Ohmeda SmartVent and the Ohmeda 7810 produced larger flows at all pressures as compared to the Ohmeda 7000.
mechanical lung (Model 2600i, Michigan Instruments Training Test Lung, Grand Rapids, MI) at a compliance of 500 mL/cm H2O. For this determination, inspiratory flow was set at maximum for each ventilator and VT was maintained at 800 mL by adjusting TI. Mean inspiratory flow was calculated as VT/TI. In part two of the study, we measured maximal V˙E under three different conditions by connecting the breathing circuit to the mechanical lung having adjustable compliance and resistance (Figure 1). During all three study conditions, maximal V˙E was obtained by increasing respiratory frequency at a VT of 800 mL (when permitted by the pressure limit of the ventilator). Tidal volume of 800 mL was maintained by setting inspiratory flow at its maximum and increasing inspiratory time. Maximal V˙E was defined by the occurrence of gas trapping as reflected by an end-expiratory lung pressure exceeding 5 cm H2O. This pressure was measured inside the test lung using a separate tube connected to the BICORE CP-100 monitor. For condition 1, maximal V˙E was measured over a series of decreasing compliance (30, 20, 15, and 10 mL/cm H2O) during low airway resistance. Low airway resistance was achieved using a Rp5 resistor placed between the flow sensor and the test lung (Figure 1). This resistor and connection tubing produced a 5 cm H2O pressure decrease at 60 L/min. For condition 2, maximal V˙E was measured at 0, 10, and 20 cm H2O of PEEP at a compliance of 20 mL/cm H2O and low airway resistance. A Vital Signs™ spring-loaded PEEP valve (Totawa, NJ) was used with the Ohmeda 7810 and Ohmeda 7000 ventilators; for the other ventilators, the PEEP valve incorporated into their machines was used. For condition 3, maximal V˙E was measured at increased airway resistance, at a compliance of 30 mL/cm H2O. Increased airway resistance was 42
J. Clin. Anesth., vol. 12, February 2000
achieved using a Rp20 and a Rp40 resistor. These resistors and connection tubing produced a 19 cm H2O and 36 cm H2O pressure drop at 60 L/min. In part three of the study, we attempt to predict the percentage of critically ill patients whose level of preoperative V˙E might not be achieved with the various anesthesia ventilators studied. With approval from the University of California, San Francisco Committee on Human Research, we reviewed the medical records of 145 mechanically ventilated patients for acute respiratory failure who required 200 surgical procedures over a 3-year period. Preoperative peak airway pressure and V˙E were extracted from the medical record. The mean inspiratory flow versus airway pressure curves were converted to V˙E versus airway pressure curves using a TI/TTOT of 0.5, 0.33, and 0.25. The V˙E versus airway pressure curves were compared to the preoperative V˙E and peak airway pressure obtained from the mechanically ventilated patients to determine the percent of patients that fell above the curve. Comparisons of V˙E and TI/TTOT between ventilators, levels of compliance, resistance, and PEEP were analyzed using analysis of variance with differences between groups detected by Newman-Keuls testing; a p-value ⬍0.05 was considered significant. Data are reported only as means without respective standard deviation (SD) because the average coefficient of variation (SD/mean) was less than 0.1%.
Results Flow/Pressure Performance of the Anesthesia Ventilators Mean inspiratory flow decreased steadily with increasing airway pressure in the anesthesia ventilators (Figure 2).
Flow capability of anesthesia ventilators: Katz et al.
Table 1. Effect of Decreasing Compliance on Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) During Low Airway Resistance
Compliance (mL/cm H2O)
V˙E (L/min)
TI/TTOT
Peak Airway Pressure (cm H2O)
41.0*† 43.5† 42.9† 45.0*†
0.35*† 0.36† 0.36† 0.47*
46 56 62 84
28.0† 30.2*† 28.8*† 28.2
0.38* 0.40* 0.43* 0.51*†
34 47 60 84
27.0*† 27.5† 27.5† 28.0*
0.38* 0.40* 0.43* 0.47*
34 47 60 83
22.4† 22.4† 20.0*† 16.0*†
0.44*† 0.48† 0.46*† 0.48†
33 45 56 67§
Siemens 900C 30 20 15 10 Datex-Ohmeda SmartVent 30 20 15 10 Ohmeda 7810 30 20 15 10 Ohmeda 7000 30 20 15 10
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of ventilator at different compliance levels. †p ⬍ 0.05; comparison of different ventilators at same compliance level. §Pressure limit of the ventilator.
ventilators maintained V˙E while the Ohmeda 7000 ventilator showed a decrease (Table 1). Maintaining V˙E with decreasing compliance was accomplished by an increase in TI/TTOT. The Siemens 900C ventilator delivered higher V˙E than the other ventilators during decreased compliance with low airway resistance (Table 1). For the DatexOhmeda SmartVent, there was minimal discrepancy between ventilator set VT and the delivered VT (measured by the BICORE) as compliance decreased; for the Ohmeda 7810 and the Ohmeda 7000 ventilators, set VT progressively exceeded the delivered VT (Table 2). With the application of PEEP, both the Siemens 900C and the Datex-Ohmeda SmartVent maintained V˙E, while
The Datex-Ohmeda SmartVent and Ohmeda 7810 produced nearly identical mean inspiratory flow versus pressure curves. These two ventilators produced greater mean flow at increasing pressure compared to the Ohmeda 7000 ventilator. The Siemens 900C ventilator delivered the highest mean inspiratory flow at all airway pressures, and its inspiratory flow was pressure-independent at pressures less than 80 cm H2O.
Minute Ventilation with Decreasing Compliance, PEEP; and Increasing Resistance As compliance progressively decreased, the Siemens 900C, the Datex-Ohmeda SmartVent, and the Ohmeda 7810
Table 2. Effect of Decreasing Compliance on Delivered Tidal Volume (VT)* Datex-Ohmeda SmartVent Compliance (mL/cm H2O) 30 20 15 10
Ohmeda 7810
Ohmeda 7000
Set VT (L)
Delivered VT (L)
Set VT (L)
Delivered VT (L)
Set VT (L)
Delivered VT (L)
0.80 0.78 0.78 0.80
0.80 0.82 0.82 0.81
0.97 1.10 1.18 1.32
0.79 0.81 0.81 0.79
1.0 1.1 1.2 1.2
0.79 0.81 0.80 0.65†
Note: Mean data, n ⫽ 10. *Set VT is ventilator control setting; delivered VT is that measured by the BICORE distal to the Y-piece. †VT limited by pressure limit of ventilator. J. Clin. Anesth., vol. 12, February 2000
43
Original Contributions
V˙E (L/min)
TI/TTOT
43.5* 43.5* 43.6*
0.36*† 0.50* 0.50*
30.2* 30.4* 30.3*
0.40† 0.45* 0.45*
to the preoperative V˙E and peak airway pressure obtained from the mechanically ventilated patients (Figure 3). The percent of patients exceeding the predicted V˙E versus airway pressure curves varied as a function of TI/TTOT, and the anesthesia ventilator (Figure 4). At a TI/TTOT of 0.33, approximately 10% of critically ill patients exceeded the curves for the Datex-Ohmeda SmartVent and Ohmeda 7810 ventilators, while 34% of patients exceeded that constructed for the Ohmeda 7000 ventilator (Figure 3 and Figure 4). The V˙E versus airway pressure curves for the Siemens 900C exceeded the preoperative peak airway pressure and V˙E data for nearly all 200 patients (Figure 3 and Figure 4).
27.5*† 25.0*† 22.7*†
0.40† 0.39*† 0.37*†
Discussion
22.4*† 21.0*† 18.5*†
0.48* 0.48* 0.47*
Table 3. Effect of Positive End-Expiratory Pressure (PEEP) on Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) at a Compliance of 20 mL/cm H2O and Low Airway Resistance PEEP (cm H2O) Siemens 900C 0 10 20 Datex-Ohmeda SmartVent 0 10 20 Ohmeda 7810 0 10 20 Ohmeda 7000 0 10 20
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of different ventilators at same PEEP level. †p ⬍ 0.05; comparison of ventilator at different PEEP levels.
V˙E progressively decreased with the Ohmeda 7810 and 7000 ventilators (Table 3). During increased airway resistance, maximal V˙E was lower and TI/TTOT decreased in all ventilators (Table 4).
Predicting Intraoperative Requirement for Critical Care Ventilator The V˙E versus airway pressure curves (constructed from the mean inspiratory flow vs. pressure curves) were compared
Each manufacturer specifies maximal obtainable pressures and inspiratory flows for its ventilators (Table 5). These maximal peak inspiratory flow values are for the isolated ventilator, but when integrated into the anesthesia system, the actual delivered peak and mean inspiratory flow decline with increasing pressure. Inspiratory flow decreases primarily because of compressibility (compression of gases and distension of the breathing circuit).2,7 Both the Datex-Ohmeda SmartVent and the Ohmeda 7810 have improved maximal inspiratory flow and airway pressure capabilities, as compared to the Ohmeda 7000. Our results indicate that the Datex-Ohmeda SmartVent and the Ohmeda 7810, as compared to the older Ohmeda 7000 ventilator, delivered substantially larger mean inspiratory flows at increased airway pressures (Figure 2), and delivered larger V˙E at lower compliance and increased resistance (Table 1 and Table 4). In contrast to the anesthesia ventilators studied, the Siemens 900C ventilator
Table 4. Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) During Low and Increased Airway Resistance
Siemens 900C Rp5 Rp20 Rp40 Datex-Ohmeda SmartVent Rp5 Rp20 Rp40 Ohmeda 7810 Rp5 Rp20 Rp40 Ohmeda 7000 Rp5 Rp20 Rp40
V˙E (L/min)
TI/TTOT
Peak Airway Pressure (cm H2O)
41.0*† 26.7*† 21.1*†
0.35*† 0.34† 0.34†
46 61 70
28.0*† 22.8*† 18.2*†
0.38* 0.37*† 0.33*†
34 58 71
27.0*† 21.6*† 17.5*†
0.38* 0.35*† 0.31*†
34 56 73
22.4*† 18.4*† 16.0*†
0.44*† 0.40*† 0.38*†
33 43 53
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of ventilator at different resistor levels. †p ⬍ 0.05; comparison of different ventilators at same resistor level. Compliance ⫽ 30 mL/cm H2O. 44
J. Clin. Anesth., vol. 12, February 2000
Flow capability of anesthesia ventilators: Katz et al.
Figure 3. Calculated minute ventilation (V˙E) as a function of airway pressure at an inspiratory duty cycle (TI/TTOT) ⫽ 0.33. Superimposed are the preoperative V˙E and peak airway pressure values of 200 patients with acute respiratory failure requiring preoperative mechanical ventilation. Minute ventilation ⫽ mean inspiratory flow ⫻ inspiratory duty cycle.
maintained mean inspiratory flow up to airway pressures of nearly 80 cm H2O. Flow is maintained because the Siemens 900C has minimal compressible volume and a flow generator that is pressure-independent until the working pressure limit is approached. While the Siemens 900C has a maximum working pressure limit of 120 cm H2O, in the ventilator adapted for delivery of inhalational anesthetics, a working pressure of only 100 cm H2O was achieved because of limitations in the air/oxygen blender. Compressibility influences not only the apparent power of the ventilator but also the delivery of VT during mechanical ventilation. The actual VT delivered (to the
patient) may differ from the VT electronically set on the ventilator or from VT apparent by the excursion of the bellows.7 Under certain conditions, delivered VT is augmented and in others it is decreased.8 In anesthesia systems, delivered VT is augmented by the product of the fresh gas flow and the inspiratory time. In contrast, delivered VT is decreased due to compression of gases and distension of the breathing circuit (ventilator bellows, absorber, and connecting tubing and ventilator flow generator).2,7,9 A typical adult circle circuit with ventilator has a compression volume of 6 to 7 L and a compressibility of 6 to 12 mL/cm H2O.7
Figure 4. Percent of patients with acute respiratory failure with preoperative minute ventilation and/or airway pressure values that exceed a ventilator’s predicted minute ventilation and airway pressure capability (see also Figure 3). J. Clin. Anesth., vol. 12, February 2000
45
Original Contributions
Table 5. Maximum Pressure and Inspiratory Flow Using Manufacturers’ Specifications
Ventilator Siemens 900C Datex-Ohmeda SmartVent Ohmeda 7810 Ohmeda 7000
Pressure (cm H2O)
Inspiratory Flow (L/min)
Predicted V˙E* (L/min)
120 100 100 70
200 120 100 60
100 60 50 30
*Predicted maximum minute ventilation (V˙E) ⫽ inspiratory flow ⫻ (TI/TTOT); TI/TTOT ⫽ 0.5.
This discrepancy between ventilator set VT and the delivered VT measured at the endotracheal/circuit interface was significant for the Ohmeda 7810 and Ohmeda 7000 but not the Datex-Ohmeda SmartVent despite having similar compressibilities of approximately 8 mL/cm H2O (Table 2). With the SmartVent ventilator, there was close approximation between the electronically set VT and the VT measured at the endotracheal/circuit interface. This VT compensation with the SmartVent is a result of the microprocessor VT measurements on the inspiratory limb of the circuit comparing it to the electronically set VT with feedback loop increasing flow output to compensate for compressibility. Tidal volume compensation also occurs for changes in fresh gas flow, although we did not test for this action in our study.10 In typical anesthesia ventilators, the decrease in mean inspiratory flow at increasing airway pressure may limit maximal V˙E. Minute ventilation could be maintained by increasing TI/TTOT, but this action may lead to gas trapping, depending on the lung-thorax compliance and airway resistance.11 Our data indicate that when compliance is reduced without an increase in airway resistance (Table 1), V˙E can be maintained by increasing TI/TTOT provided that the time constant (compliance times resistance) for lung emptying is short. In contrast, when airway resistance is increased (Table 4), V˙E declines; the time constant for lung emptying is prolonged, requiring an increased expiratory time to avoid air trapping. While the Siemens 900C ventilator delivered the largest V˙E under conditions of increased resistance, the difference between it and the Datex-Ohmeda SmartVent and the Ohmeda 7810 was considerably less than in the condition of similar compliance and low resistance. Thus, increases in airway resistance have a dramatic effect on the ability to adequately provide mechanical ventilation without causing gas trapping. The strategy of low volume ventilation in severe asthmatics is deliberately designed to limit V˙E so as to avoid the clinical consequences of pulmonary hyperinflation and gas trapping.12 Positive end-expiratory pressure might affect the ability to maintain V˙E by increasing flow resistance and prolonging the time constant for exhalation.13 For the anesthesia ventilators, two types of PEEP valves were tested, including a mechanical spring-loaded device (Vital Signs), and electronic PEEP in the (Datex-Ohmeda SmartVent). The SmartVent applies continuous pressure to the bellows drive circuit. Minute ventilation was maintained with the SmartVent with increasing PEEP up to 20 cm H2O, but it declined in the anesthesia ventilators incorporating the 46
J. Clin. Anesth., vol. 12, February 2000
Vital Signs Valve. Electronic PEEP in the SmartVent was efficient and appears to perform as an ideal threshold resistor.14 The pressure and flow limitations of anesthesia ventilators have important clinical implications when patients with increased V˙E requirements and high airway pressures require anesthesia and surgery. Commonly used anesthesia ventilators may be unable to deliver the required V˙E, or they can do so only by using a longer TI/TTOT.2 Failure to deliver the required V˙E results in intraoperative hypercapnia and possible hypoxemia.1,15 A prolonged TI/TTOT may result in gas trapping and potential barotrauma or hemodynamic compromise.16,17 A critical care ventilator will avoid these complications.1 Thus, in planning an anesthetic for patients with acute respiratory failure, ventilator selection is important. Selection of the intraoperative ventilator should be based on each patient’s ventilation requirements. Airway pressure, V˙E, and TI/TTOT can be obtained from preoperative ventilator settings. These values can be compared to the corresponding V˙E versus airway pressure curves (Figure 3). Expiratory time requirements vary greatly depending on the nature and severity of pulmonary disease. In general, a TI/TTOT greater than 0.33 at high levels of V˙E frequently results in gas trapping.18,19 The improved pressure and flow capabilities observed with the Datex-Ohmeda SmartVent and the Ohmeda 7810 would be predicted to have met the ventilation requirements of approximately 90% of our critically ill patients. In contrast, the Ohmeda 7000 ventilator would be predicted to have met the ventilation requirements of approximately 66% of these critically ill patients undergoing surgery. Considering the improved flow and pressure capabilities of the Datex-Ohmeda SmartVent and its ability to deliver PEEP without compressible volume loss, the criteria for intraoperative use of a critical care ventilator could be increased to a peak airway pressure exceeding 65 cm H2O and/or V˙E exceeding 18 L/min.
References 1. Biery DR, Marks JD, Schapera A, et al: Factors affecting perioperative pulmonary function in acute respiratory failure. Chest 1990;98:1455– 62. 2. Marks JD, Schapera A, Kraemer RW, Katz JA: Pressure and flow limitations of anesthesia ventilators. Anesthesiology 1989;71:403– 8. 3. Milic-Emili J, Aubier M: Some recent advances in the study of the control of breathing in patients with chronic obstructive lung disease. Anesth Analg 1980;59:865–73.
Flow capability of anesthesia ventilators: Katz et al. 4. Petros AJ, Lamond CT, Bennett D: The Bicore pulmonary monitor. A device to assess the work of breathing while weaning from mechanical ventilation. Anaesthesia 1993;48:985– 8. 5. Blanch PB, Banner MJ: A new respiratory monitor that enables accurate measurements of work of breathing: a validation study. Respir Care 1994;39:897–905. 6. Tschernko EM, Gruber EM, Jaksch P, et al: Ventilatory mechanics and gas exchange during exercise before and after lung volume reduction surgery. Am J Respir Crit Care Med 1998;158:1424 –31. 7. Cote´ CJ, Petkau AJ, Ryan JF, Welch JP: Wasted ventilation measured in vitro with eight anesthetic circuits with and without inline humidification. Anesthesiology 1983;59:442– 6. 8. Gravenstein N, Banner MJ, McLaughlin G: Tidal volume changes due to the interaction of anesthesia machine and anesthesia ventilator. J Clin Monit 1987;3:187–90. 9. Patterson JR, Russell GK, Pierson DJ, et al: Evaluation of a fluidic ventilator: a new approach to mechanical ventilation. Chest 1974;66:706 –11. 10. Rothschiller JL, Uejima T, Dsida RM, Cote CJ: Evaluation of a new operating room ventilator with volume-controlled ventilation: the Ohmeda 7900. Anesth Analg 1999;88:39 – 42. 11. Scott LR, Benson MS, Pierson DJ: Effect of inspiratory flow rate and circuit compressible volume on auto-PEEP during mechanical ventilation. Respir Care 1986;31:1075–9.
12. Darioli R, Perret C: Mechanical controlled hypoventilation in status asthmaticus. Am Rev Respir Dis 1984;129:385–7. 13. Pan PH, van der Aa JJ: Positive end-expiratory pressure: effect on delivered tidal volume [Letter]. J Clin Anesth 1995;7:443– 4. 14. Banner MJ, Lampotang S, Boysen PG, et al: Flow resistance of expiratory positive pressure valve systems. Chest 1986;90:212–7. 15. Schapera A, Marks JD, Minagi H, Goodman P, Katz JA: Perioperative pulmonary function in acute respiratory failure: effect of ventilator type and anesthetic gas mixture. Anesthesiology 1989;71: 396 – 402. 16. Bergman NA: Intrapulmonary gas trapping during mechanical ventilation at rapid frequencies. Anesthesiology 1972;37:626 –33. 17. Pepe PE, Marini JJ: Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction: the auto-PEEP effect. Am Rev Respir Dis 1982;126:166 –70. 18. Brown DG, Pierson DJ: Auto PEEP is common in mechanically ventilated patients: a study of the incidence, severity, and detection. Respir Care 1986;31:1069 –74. 19. Rossi A, Gottfried SB, Zocchi L, et al: Measurement of static compliance of the total respiratory system in patients with acute respiratory failure during mechanical ventilation. The effect of intrinsic positive-end-expiratory pressure. Am Rev Respir Dis 1985; 131:672–7.
J. Clin. Anesth., vol. 12, February 2000
47
*Professor of Clinical Anesthesia, University of California, San Francisco; Chief, Department of Anesthesia, UCSF/Mount Zion Medical Center †Clinical Research Coordinator, University of California, San Francisco ‡Clinical Research Associate, University of California, San Francisco §Professor of Anesthesia, University of California, San Francisco Address correspondence to Dr. Katz at the Department of Anesthesia, UCSF/Mount Zion Medical Center, Hellman Bldg., Room C355, 1600 Divisadero St., San Francisco, CA 94143-1605, USA. E-mail: katzj@anesthesia. ucsf.edu Supported in part by a grant from DatexOhmeda, Madison, WI. Received for publication June 25, 1999; revised manuscript accepted for publication November 22, 1999.
Study Objective: To compare the flow and pressure capabilities of the Datex-Ohmeda SmartVent (Ohmeda 7900, Datex-Ohmeda, Madison, WI) to previous Ohmeda (7810 and 7000, Datex-Ohmeda, Madison, WI) anesthesia ventilators. To determine airway pressure and minute ventilation thresholds for intraoperative use of a critical care ventilator. Design: Three anesthesia ventilators and one critical care ventilator (Siemens Servo 900C, Siemens, Solna, Sweden) were studied in a lung model. Retrospective medical record review. Setting: Research Laboratory and Critical Care Unit of a Level I Trauma Center. Patients: 145 mechanically ventilated patients treated for acute respiratory failure who underwent 200 surgical procedures. Interventions: The effect of increasing pressure on mean inspiratory flow was determined by cycling each ventilator through increasing restrictors. Maximum minute ventilation was measured at low compliance (10 –30 mL/cm H2O), positive end-expiratory pressure (PEEP) (0 –20 cm H2O), and increased airway resistance (⬃19 and ⬃36 cm H2O/L/sec) in a mechanical lung model. Measurements and Main Results: Flow, volume, and pressure were measured with a pulmonary mechanics monitor (BICORE CP-100, Thermo Respiratory Group, Yorba Linda, CA). Preoperative peak airway pressure and minute ventilation (V˙E) were extracted from the medical record. Mean inspiratory flow declined with increasing pressure in all anesthesia ventilators. The SmartVent and the 7810 produced greater mean inspiratory flow than did the 7000 ventilator. As compliance progressively decreased, the Siemens, the SmartVent, and the 7810 ventilators maintained V˙E compared to the 7000 ventilator. The Siemens and the SmartVent maintained V˙E with PEEP, compared to the 7810 and 7000 ventilators. During increased airway resistance, maximal V˙E was lower for all ventilators. The SmartVent met the ventilation requirements in 90% of the patients compared to 67% of patients with the 7000 ventilator. Conclusion: The improved pressure and flow capabilities of the SmartVent increase the threshold for using a critical care ventilator intraoperatively to a peak airway pressure ⬎65 cm H2O and/or V˙E ⬎18 L/min. © 2000 by Elsevier Science Inc. Keywords: Equipment, ventilators: flow characteristics; lung: acute respiratory failure, airway resistance, compliance; ventilation: positive end-expiratory pressure.
Journal of Clinical Anesthesia 12:40 – 47, 2000 © 2000 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010
0952-8180/00/$–see front matter PII S0952-8180(99)00140-3
Flow capability of anesthesia ventilators: Katz et al.
Figure 1. Model for determining effects of increasing airway pressure on mean inspiratory flow and maximum minute ventilation. Each ventilator/anesthesia machine was connected to a mechanical lung with adjustable compliance. Resistance was altered by changing the mechanical resistor (R). A separate transducer measured pressure inside the mechanical lung.
Introduction
Materials and Methods
Patients with acute lung injury requiring mechanical ventilation often undergo general anesthesia for necessary surgical procedures. Ventilatory support of these patients may require a high minute ventilation (V˙E), increased airway pressure, and high levels of positive end-expiratory pressure (PEEP).1 We previously reported on the pressure and flow capabilities of anesthesia ventilators showing that maximal mean inspiratory flow, and therefore V˙E capacity, decreased with increasing airway pressure.2 Additionally, maximum V˙E can be calculated for a specific anesthesia ventilator using the mean inspiratory flow (tidal volume [VT] divided by inspiratory time [TI]) versus airway pressure curves, and the maximum allowable inspiratory duty cycle (the ratio of TI to total breathing cycle duration [TTOT]). Thus, V˙E could be expressed as a function of a driving (VT/TI) and timing (TI/TTOT) mechanisms: V˙E ⫽ VT/TI ⫻ [TI/TTOT].3 This approach permits prediction of ventilator performance for patients with high V˙E requirements. For example, with the Ohmeda 7000 ventilator (Datex-Ohmeda, Madison, WI), a maximal mean inspiratory flow of 42 L/min could be delivered at a pressure of 50 cm H2O.2 Thus, with a peak airway pressure of 50 cm H2O and a TI/TTOT of 0.33, the Ohmeda 7000 ventilator could deliver a maximum of 14 L/min. Because many patients with acute respiratory failure have V˙E requirements greater than 15 L/min and peak airway pressures that exceed 50 cm H2O,1 we previously recommended use of a critical care ventilator when preoperative V˙E and airway pressures exceeded these values.2 Since our recommendation in 1989, Datex-Ohmeda has introduced the SmartVent anesthesia ventilator, which is purported to have improved flow and pressure capabilities. This ventilator should be able to deliver higher V˙E at increased airway pressures, thus minimizing the need for critical care ventilators for patients with acute respiratory failure requiring surgery. This study compares the pressure and flow capacity of the SmartVent to previous Ohmeda 7000 series anesthesia ventilators.
Three anesthesia ventilators (the Datex-Ohmeda SmartVent™, the Ohmeda 7810™, and the Ohmeda 7000™) and one critical care ventilator (the Siemens Servo 900C™, Siemens, Solna, Sweden, adapted for delivery of inhalational anesthetics) were studied in a lung model. All ventilators were inspected and calibrated to manufacturer specifications prior to the study. Each ventilator was tested as part of the anesthesia machine provided by the manufacturer. An oxygen gas flow of 5 L/min was provided via the fresh gas outlet for all ventilators except the Siemens Servo 900C, which was powered by its own blender. Low-compliance tubing of 0.6 mL/cm H2O (Connect Company™, Lea & Kay, San Francisco, CA) was used to complete the breathing circuit. Gas flow and pressure were measured by a flow sensor (Varflex, Bicore CP100, Thermo Respiratory Group, Yorba Linda, CA) attached to a pulmonary mechanics monitor (BICORE CP-100, Allied Healthcare Products, Riverside, CA). Tidal volume (VT) was obtained by time integration of the flow signal. The flow sensor was positioned immediately distal to the Ypiece of the circuit (Figure 1), so that delivered VT excludes volume loss due to compressibility. Prior to each ventilator study, the flow sensor was electronically calibrated, and verified with a Fisher Porter flow tube for flow and a 1-L super-syringe for volume. Pressure measurements were calibrated against a water column. A difference of less than 5% was regarded as sufficiently accurate. The accuracy of the measurements provided by this monitoring system has been found to be satisfactory.4 – 6 The BICORE CP-100 provided a real-time display of gas flow, VT, respiratory frequency, and pressure. The V˙E and breathing pattern, i.e., VT, respiratory frequency, the duration of inspiration and expiration, and the TI/TTOT were analyzed based on the flow signal. The data of ten consecutive respiratory cycles were averaged. The study was conducted in three parts. In part one, we measured the effect of increasing pressure on mean inspiratory flow by cycling each ventilator/anesthesia machine through progressively increasing resistors into a J. Clin. Anesth., vol. 12, February 2000
41
Original Contributions
Figure 2. Effect of increasing airway pressure on mean inspiratory flow for each ventilator/anesthesia machine. The Siemens 900C delivered pressure-independent flow at airway pressures less than 80 cm H2O. While mean inspiratory flow declined with increasing airway pressure in all anesthesia ventilators, both the Datex-Ohmeda SmartVent and the Ohmeda 7810 produced larger flows at all pressures as compared to the Ohmeda 7000.
mechanical lung (Model 2600i, Michigan Instruments Training Test Lung, Grand Rapids, MI) at a compliance of 500 mL/cm H2O. For this determination, inspiratory flow was set at maximum for each ventilator and VT was maintained at 800 mL by adjusting TI. Mean inspiratory flow was calculated as VT/TI. In part two of the study, we measured maximal V˙E under three different conditions by connecting the breathing circuit to the mechanical lung having adjustable compliance and resistance (Figure 1). During all three study conditions, maximal V˙E was obtained by increasing respiratory frequency at a VT of 800 mL (when permitted by the pressure limit of the ventilator). Tidal volume of 800 mL was maintained by setting inspiratory flow at its maximum and increasing inspiratory time. Maximal V˙E was defined by the occurrence of gas trapping as reflected by an end-expiratory lung pressure exceeding 5 cm H2O. This pressure was measured inside the test lung using a separate tube connected to the BICORE CP-100 monitor. For condition 1, maximal V˙E was measured over a series of decreasing compliance (30, 20, 15, and 10 mL/cm H2O) during low airway resistance. Low airway resistance was achieved using a Rp5 resistor placed between the flow sensor and the test lung (Figure 1). This resistor and connection tubing produced a 5 cm H2O pressure decrease at 60 L/min. For condition 2, maximal V˙E was measured at 0, 10, and 20 cm H2O of PEEP at a compliance of 20 mL/cm H2O and low airway resistance. A Vital Signs™ spring-loaded PEEP valve (Totawa, NJ) was used with the Ohmeda 7810 and Ohmeda 7000 ventilators; for the other ventilators, the PEEP valve incorporated into their machines was used. For condition 3, maximal V˙E was measured at increased airway resistance, at a compliance of 30 mL/cm H2O. Increased airway resistance was 42
J. Clin. Anesth., vol. 12, February 2000
achieved using a Rp20 and a Rp40 resistor. These resistors and connection tubing produced a 19 cm H2O and 36 cm H2O pressure drop at 60 L/min. In part three of the study, we attempt to predict the percentage of critically ill patients whose level of preoperative V˙E might not be achieved with the various anesthesia ventilators studied. With approval from the University of California, San Francisco Committee on Human Research, we reviewed the medical records of 145 mechanically ventilated patients for acute respiratory failure who required 200 surgical procedures over a 3-year period. Preoperative peak airway pressure and V˙E were extracted from the medical record. The mean inspiratory flow versus airway pressure curves were converted to V˙E versus airway pressure curves using a TI/TTOT of 0.5, 0.33, and 0.25. The V˙E versus airway pressure curves were compared to the preoperative V˙E and peak airway pressure obtained from the mechanically ventilated patients to determine the percent of patients that fell above the curve. Comparisons of V˙E and TI/TTOT between ventilators, levels of compliance, resistance, and PEEP were analyzed using analysis of variance with differences between groups detected by Newman-Keuls testing; a p-value ⬍0.05 was considered significant. Data are reported only as means without respective standard deviation (SD) because the average coefficient of variation (SD/mean) was less than 0.1%.
Results Flow/Pressure Performance of the Anesthesia Ventilators Mean inspiratory flow decreased steadily with increasing airway pressure in the anesthesia ventilators (Figure 2).
Flow capability of anesthesia ventilators: Katz et al.
Table 1. Effect of Decreasing Compliance on Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) During Low Airway Resistance
Compliance (mL/cm H2O)
V˙E (L/min)
TI/TTOT
Peak Airway Pressure (cm H2O)
41.0*† 43.5† 42.9† 45.0*†
0.35*† 0.36† 0.36† 0.47*
46 56 62 84
28.0† 30.2*† 28.8*† 28.2
0.38* 0.40* 0.43* 0.51*†
34 47 60 84
27.0*† 27.5† 27.5† 28.0*
0.38* 0.40* 0.43* 0.47*
34 47 60 83
22.4† 22.4† 20.0*† 16.0*†
0.44*† 0.48† 0.46*† 0.48†
33 45 56 67§
Siemens 900C 30 20 15 10 Datex-Ohmeda SmartVent 30 20 15 10 Ohmeda 7810 30 20 15 10 Ohmeda 7000 30 20 15 10
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of ventilator at different compliance levels. †p ⬍ 0.05; comparison of different ventilators at same compliance level. §Pressure limit of the ventilator.
ventilators maintained V˙E while the Ohmeda 7000 ventilator showed a decrease (Table 1). Maintaining V˙E with decreasing compliance was accomplished by an increase in TI/TTOT. The Siemens 900C ventilator delivered higher V˙E than the other ventilators during decreased compliance with low airway resistance (Table 1). For the DatexOhmeda SmartVent, there was minimal discrepancy between ventilator set VT and the delivered VT (measured by the BICORE) as compliance decreased; for the Ohmeda 7810 and the Ohmeda 7000 ventilators, set VT progressively exceeded the delivered VT (Table 2). With the application of PEEP, both the Siemens 900C and the Datex-Ohmeda SmartVent maintained V˙E, while
The Datex-Ohmeda SmartVent and Ohmeda 7810 produced nearly identical mean inspiratory flow versus pressure curves. These two ventilators produced greater mean flow at increasing pressure compared to the Ohmeda 7000 ventilator. The Siemens 900C ventilator delivered the highest mean inspiratory flow at all airway pressures, and its inspiratory flow was pressure-independent at pressures less than 80 cm H2O.
Minute Ventilation with Decreasing Compliance, PEEP; and Increasing Resistance As compliance progressively decreased, the Siemens 900C, the Datex-Ohmeda SmartVent, and the Ohmeda 7810
Table 2. Effect of Decreasing Compliance on Delivered Tidal Volume (VT)* Datex-Ohmeda SmartVent Compliance (mL/cm H2O) 30 20 15 10
Ohmeda 7810
Ohmeda 7000
Set VT (L)
Delivered VT (L)
Set VT (L)
Delivered VT (L)
Set VT (L)
Delivered VT (L)
0.80 0.78 0.78 0.80
0.80 0.82 0.82 0.81
0.97 1.10 1.18 1.32
0.79 0.81 0.81 0.79
1.0 1.1 1.2 1.2
0.79 0.81 0.80 0.65†
Note: Mean data, n ⫽ 10. *Set VT is ventilator control setting; delivered VT is that measured by the BICORE distal to the Y-piece. †VT limited by pressure limit of ventilator. J. Clin. Anesth., vol. 12, February 2000
43
Original Contributions
V˙E (L/min)
TI/TTOT
43.5* 43.5* 43.6*
0.36*† 0.50* 0.50*
30.2* 30.4* 30.3*
0.40† 0.45* 0.45*
to the preoperative V˙E and peak airway pressure obtained from the mechanically ventilated patients (Figure 3). The percent of patients exceeding the predicted V˙E versus airway pressure curves varied as a function of TI/TTOT, and the anesthesia ventilator (Figure 4). At a TI/TTOT of 0.33, approximately 10% of critically ill patients exceeded the curves for the Datex-Ohmeda SmartVent and Ohmeda 7810 ventilators, while 34% of patients exceeded that constructed for the Ohmeda 7000 ventilator (Figure 3 and Figure 4). The V˙E versus airway pressure curves for the Siemens 900C exceeded the preoperative peak airway pressure and V˙E data for nearly all 200 patients (Figure 3 and Figure 4).
27.5*† 25.0*† 22.7*†
0.40† 0.39*† 0.37*†
Discussion
22.4*† 21.0*† 18.5*†
0.48* 0.48* 0.47*
Table 3. Effect of Positive End-Expiratory Pressure (PEEP) on Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) at a Compliance of 20 mL/cm H2O and Low Airway Resistance PEEP (cm H2O) Siemens 900C 0 10 20 Datex-Ohmeda SmartVent 0 10 20 Ohmeda 7810 0 10 20 Ohmeda 7000 0 10 20
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of different ventilators at same PEEP level. †p ⬍ 0.05; comparison of ventilator at different PEEP levels.
V˙E progressively decreased with the Ohmeda 7810 and 7000 ventilators (Table 3). During increased airway resistance, maximal V˙E was lower and TI/TTOT decreased in all ventilators (Table 4).
Predicting Intraoperative Requirement for Critical Care Ventilator The V˙E versus airway pressure curves (constructed from the mean inspiratory flow vs. pressure curves) were compared
Each manufacturer specifies maximal obtainable pressures and inspiratory flows for its ventilators (Table 5). These maximal peak inspiratory flow values are for the isolated ventilator, but when integrated into the anesthesia system, the actual delivered peak and mean inspiratory flow decline with increasing pressure. Inspiratory flow decreases primarily because of compressibility (compression of gases and distension of the breathing circuit).2,7 Both the Datex-Ohmeda SmartVent and the Ohmeda 7810 have improved maximal inspiratory flow and airway pressure capabilities, as compared to the Ohmeda 7000. Our results indicate that the Datex-Ohmeda SmartVent and the Ohmeda 7810, as compared to the older Ohmeda 7000 ventilator, delivered substantially larger mean inspiratory flows at increased airway pressures (Figure 2), and delivered larger V˙E at lower compliance and increased resistance (Table 1 and Table 4). In contrast to the anesthesia ventilators studied, the Siemens 900C ventilator
Table 4. Minute Ventilation (V˙E) and Inspiratory Duty Cycle (TI/TTOT) During Low and Increased Airway Resistance
Siemens 900C Rp5 Rp20 Rp40 Datex-Ohmeda SmartVent Rp5 Rp20 Rp40 Ohmeda 7810 Rp5 Rp20 Rp40 Ohmeda 7000 Rp5 Rp20 Rp40
V˙E (L/min)
TI/TTOT
Peak Airway Pressure (cm H2O)
41.0*† 26.7*† 21.1*†
0.35*† 0.34† 0.34†
46 61 70
28.0*† 22.8*† 18.2*†
0.38* 0.37*† 0.33*†
34 58 71
27.0*† 21.6*† 17.5*†
0.38* 0.35*† 0.31*†
34 56 73
22.4*† 18.4*† 16.0*†
0.44*† 0.40*† 0.38*†
33 43 53
Note: Mean data, n ⫽ 10. *p ⬍ 0.05; comparison of ventilator at different resistor levels. †p ⬍ 0.05; comparison of different ventilators at same resistor level. Compliance ⫽ 30 mL/cm H2O. 44
J. Clin. Anesth., vol. 12, February 2000
Flow capability of anesthesia ventilators: Katz et al.
Figure 3. Calculated minute ventilation (V˙E) as a function of airway pressure at an inspiratory duty cycle (TI/TTOT) ⫽ 0.33. Superimposed are the preoperative V˙E and peak airway pressure values of 200 patients with acute respiratory failure requiring preoperative mechanical ventilation. Minute ventilation ⫽ mean inspiratory flow ⫻ inspiratory duty cycle.
maintained mean inspiratory flow up to airway pressures of nearly 80 cm H2O. Flow is maintained because the Siemens 900C has minimal compressible volume and a flow generator that is pressure-independent until the working pressure limit is approached. While the Siemens 900C has a maximum working pressure limit of 120 cm H2O, in the ventilator adapted for delivery of inhalational anesthetics, a working pressure of only 100 cm H2O was achieved because of limitations in the air/oxygen blender. Compressibility influences not only the apparent power of the ventilator but also the delivery of VT during mechanical ventilation. The actual VT delivered (to the
patient) may differ from the VT electronically set on the ventilator or from VT apparent by the excursion of the bellows.7 Under certain conditions, delivered VT is augmented and in others it is decreased.8 In anesthesia systems, delivered VT is augmented by the product of the fresh gas flow and the inspiratory time. In contrast, delivered VT is decreased due to compression of gases and distension of the breathing circuit (ventilator bellows, absorber, and connecting tubing and ventilator flow generator).2,7,9 A typical adult circle circuit with ventilator has a compression volume of 6 to 7 L and a compressibility of 6 to 12 mL/cm H2O.7
Figure 4. Percent of patients with acute respiratory failure with preoperative minute ventilation and/or airway pressure values that exceed a ventilator’s predicted minute ventilation and airway pressure capability (see also Figure 3). J. Clin. Anesth., vol. 12, February 2000
45
Original Contributions
Table 5. Maximum Pressure and Inspiratory Flow Using Manufacturers’ Specifications
Ventilator Siemens 900C Datex-Ohmeda SmartVent Ohmeda 7810 Ohmeda 7000
Pressure (cm H2O)
Inspiratory Flow (L/min)
Predicted V˙E* (L/min)
120 100 100 70
200 120 100 60
100 60 50 30
*Predicted maximum minute ventilation (V˙E) ⫽ inspiratory flow ⫻ (TI/TTOT); TI/TTOT ⫽ 0.5.
This discrepancy between ventilator set VT and the delivered VT measured at the endotracheal/circuit interface was significant for the Ohmeda 7810 and Ohmeda 7000 but not the Datex-Ohmeda SmartVent despite having similar compressibilities of approximately 8 mL/cm H2O (Table 2). With the SmartVent ventilator, there was close approximation between the electronically set VT and the VT measured at the endotracheal/circuit interface. This VT compensation with the SmartVent is a result of the microprocessor VT measurements on the inspiratory limb of the circuit comparing it to the electronically set VT with feedback loop increasing flow output to compensate for compressibility. Tidal volume compensation also occurs for changes in fresh gas flow, although we did not test for this action in our study.10 In typical anesthesia ventilators, the decrease in mean inspiratory flow at increasing airway pressure may limit maximal V˙E. Minute ventilation could be maintained by increasing TI/TTOT, but this action may lead to gas trapping, depending on the lung-thorax compliance and airway resistance.11 Our data indicate that when compliance is reduced without an increase in airway resistance (Table 1), V˙E can be maintained by increasing TI/TTOT provided that the time constant (compliance times resistance) for lung emptying is short. In contrast, when airway resistance is increased (Table 4), V˙E declines; the time constant for lung emptying is prolonged, requiring an increased expiratory time to avoid air trapping. While the Siemens 900C ventilator delivered the largest V˙E under conditions of increased resistance, the difference between it and the Datex-Ohmeda SmartVent and the Ohmeda 7810 was considerably less than in the condition of similar compliance and low resistance. Thus, increases in airway resistance have a dramatic effect on the ability to adequately provide mechanical ventilation without causing gas trapping. The strategy of low volume ventilation in severe asthmatics is deliberately designed to limit V˙E so as to avoid the clinical consequences of pulmonary hyperinflation and gas trapping.12 Positive end-expiratory pressure might affect the ability to maintain V˙E by increasing flow resistance and prolonging the time constant for exhalation.13 For the anesthesia ventilators, two types of PEEP valves were tested, including a mechanical spring-loaded device (Vital Signs), and electronic PEEP in the (Datex-Ohmeda SmartVent). The SmartVent applies continuous pressure to the bellows drive circuit. Minute ventilation was maintained with the SmartVent with increasing PEEP up to 20 cm H2O, but it declined in the anesthesia ventilators incorporating the 46
J. Clin. Anesth., vol. 12, February 2000
Vital Signs Valve. Electronic PEEP in the SmartVent was efficient and appears to perform as an ideal threshold resistor.14 The pressure and flow limitations of anesthesia ventilators have important clinical implications when patients with increased V˙E requirements and high airway pressures require anesthesia and surgery. Commonly used anesthesia ventilators may be unable to deliver the required V˙E, or they can do so only by using a longer TI/TTOT.2 Failure to deliver the required V˙E results in intraoperative hypercapnia and possible hypoxemia.1,15 A prolonged TI/TTOT may result in gas trapping and potential barotrauma or hemodynamic compromise.16,17 A critical care ventilator will avoid these complications.1 Thus, in planning an anesthetic for patients with acute respiratory failure, ventilator selection is important. Selection of the intraoperative ventilator should be based on each patient’s ventilation requirements. Airway pressure, V˙E, and TI/TTOT can be obtained from preoperative ventilator settings. These values can be compared to the corresponding V˙E versus airway pressure curves (Figure 3). Expiratory time requirements vary greatly depending on the nature and severity of pulmonary disease. In general, a TI/TTOT greater than 0.33 at high levels of V˙E frequently results in gas trapping.18,19 The improved pressure and flow capabilities observed with the Datex-Ohmeda SmartVent and the Ohmeda 7810 would be predicted to have met the ventilation requirements of approximately 90% of our critically ill patients. In contrast, the Ohmeda 7000 ventilator would be predicted to have met the ventilation requirements of approximately 66% of these critically ill patients undergoing surgery. Considering the improved flow and pressure capabilities of the Datex-Ohmeda SmartVent and its ability to deliver PEEP without compressible volume loss, the criteria for intraoperative use of a critical care ventilator could be increased to a peak airway pressure exceeding 65 cm H2O and/or V˙E exceeding 18 L/min.
References 1. Biery DR, Marks JD, Schapera A, et al: Factors affecting perioperative pulmonary function in acute respiratory failure. Chest 1990;98:1455– 62. 2. Marks JD, Schapera A, Kraemer RW, Katz JA: Pressure and flow limitations of anesthesia ventilators. Anesthesiology 1989;71:403– 8. 3. Milic-Emili J, Aubier M: Some recent advances in the study of the control of breathing in patients with chronic obstructive lung disease. Anesth Analg 1980;59:865–73.
Flow capability of anesthesia ventilators: Katz et al. 4. Petros AJ, Lamond CT, Bennett D: The Bicore pulmonary monitor. A device to assess the work of breathing while weaning from mechanical ventilation. Anaesthesia 1993;48:985– 8. 5. Blanch PB, Banner MJ: A new respiratory monitor that enables accurate measurements of work of breathing: a validation study. Respir Care 1994;39:897–905. 6. Tschernko EM, Gruber EM, Jaksch P, et al: Ventilatory mechanics and gas exchange during exercise before and after lung volume reduction surgery. Am J Respir Crit Care Med 1998;158:1424 –31. 7. Cote´ CJ, Petkau AJ, Ryan JF, Welch JP: Wasted ventilation measured in vitro with eight anesthetic circuits with and without inline humidification. Anesthesiology 1983;59:442– 6. 8. Gravenstein N, Banner MJ, McLaughlin G: Tidal volume changes due to the interaction of anesthesia machine and anesthesia ventilator. J Clin Monit 1987;3:187–90. 9. Patterson JR, Russell GK, Pierson DJ, et al: Evaluation of a fluidic ventilator: a new approach to mechanical ventilation. Chest 1974;66:706 –11. 10. Rothschiller JL, Uejima T, Dsida RM, Cote CJ: Evaluation of a new operating room ventilator with volume-controlled ventilation: the Ohmeda 7900. Anesth Analg 1999;88:39 – 42. 11. Scott LR, Benson MS, Pierson DJ: Effect of inspiratory flow rate and circuit compressible volume on auto-PEEP during mechanical ventilation. Respir Care 1986;31:1075–9.
12. Darioli R, Perret C: Mechanical controlled hypoventilation in status asthmaticus. Am Rev Respir Dis 1984;129:385–7. 13. Pan PH, van der Aa JJ: Positive end-expiratory pressure: effect on delivered tidal volume [Letter]. J Clin Anesth 1995;7:443– 4. 14. Banner MJ, Lampotang S, Boysen PG, et al: Flow resistance of expiratory positive pressure valve systems. Chest 1986;90:212–7. 15. Schapera A, Marks JD, Minagi H, Goodman P, Katz JA: Perioperative pulmonary function in acute respiratory failure: effect of ventilator type and anesthetic gas mixture. Anesthesiology 1989;71: 396 – 402. 16. Bergman NA: Intrapulmonary gas trapping during mechanical ventilation at rapid frequencies. Anesthesiology 1972;37:626 –33. 17. Pepe PE, Marini JJ: Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction: the auto-PEEP effect. Am Rev Respir Dis 1982;126:166 –70. 18. Brown DG, Pierson DJ: Auto PEEP is common in mechanically ventilated patients: a study of the incidence, severity, and detection. Respir Care 1986;31:1069 –74. 19. Rossi A, Gottfried SB, Zocchi L, et al: Measurement of static compliance of the total respiratory system in patients with acute respiratory failure during mechanical ventilation. The effect of intrinsic positive-end-expiratory pressure. Am Rev Respir Dis 1985; 131:672–7.
J. Clin. Anesth., vol. 12, February 2000
47