| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
Anesth Analg 2002;95:192-197 © 2002 International Anesthesia Research Society
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
A recognized hazard of prolonged endotracheal intubation is progressive airway occlusion resulting from deposition of secretions on the inner surface of the endotracheal tube (ETT). When volume-controlled ventilation is used, progressive ETT occlusion may be detected by monitoring the difference between peak and plateau airway pressures. In pressure-controlled modes, however, inspiratory airway pressures are preset and thus cannot act as a warning indicator. Instead, changes in delivered tidal volumes may aid the diagnosis of ETT occlusion. To determine whether tidal volume monitoring effectively de-tects progressive ETT occlusion, we mathematically modeled the response of a ventilator operating in pressure-controlled mode to increasing airway resistance. To corroborate our model, we then bench-tested the Siemens 300 and Puritan-Bennett 7200 ventilators by using a test lung and a series of ETTs ranging in size from 9.0 to 3.5 mm inner diameter to simulate progressive occlusion. We found that when pressure-controlled mode was used, progressive ETT occlusion did not reduce delivered tidal volumes until occlusion was nearly complete. We conclude that prolonged use of pressure-controlled mode may allow significant ETT obstruction to build up undetected, risking complete ETT occlusion and complicating the perioperative care of patients ventilated with this mode.
IMPLICATIONS: Although increasing airway pressures during volume-controlled ventilation allow early recognition of endotracheal tube (ETT) obstruction, airway pressures with pressure-controlled ventilation are fixed. We found during tests of two intensive care unit ventilators that although ETT obstruction reduces delivered tidal volumes during pressure-controlled ventilation, reductions do not occur until occlusion is advanced.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
Critically ill patients with respiratory failure often require prolonged mechanical ventilation. Most often, the ventilator mode chosen for these patients is volume-controlled ventilation (1). In this mode, respiratory rate, tidal volume, and inspiratory gas flow patterns are preset. Because inspiratory flow rates are fixed, however, airway pressures may increase with increases in airway resistance or reductions in lung compliance. With evidence that limiting inspiratory pressures may improve outcomes (2), the use of less common ventilator modes, such as pressure control, to limit inspiratory airway pressures and tidal volumes has increased (3). In pressure-controlled modes, the ventilator functions as a pressure generator, continually modulating gas flow during inspiration to produce a constant airway pressure instead of delivering a set tidal volume. Because inspiratory airway pressures are fixed by design, dangerous increases are avoided with this mode. Delivered tidal volumes, however, are not preset and may fluctuate with increases in airway resistance, reductions in lung compliance, or patient-ventilator dyssynchrony.
When mechanical ventilation is prolonged, progressive endotracheal tube (ETT) occlusion may result from continuing deposition of respiratory secretions on the inner surface of the ETT (4). Clinically significant narrowing can occur in as little as 5 days (4) and may require urgent or emergent airway management. When patients are ventilated in volume-controlled mode, progressive ETT narrowing increases peak airway pressure by increasing the resistance to gas flow. This increase in airway pressure vanishes when inspiratory gas flow is stopped, leaving only the component of airway pressure produced by lung elastance. Briefly pausing ventilation at end inspiration then allows indirect measurement of airway resistance by calculating the difference between peak and plateau airway pressures. This difference is often used to diagnose the presence of ETT obstruction (5).
In pressure-controlled modes, however, inspiratory airway pressures are preset by the operator and do not vary with increased airway resistance. As a result, monitoring airway pressure is not helpful in diagnosing progressive ETT obstruction. We hypothesized that when pressure-controlled mode is used, delivered tidal volumes decrease with progressive ETT occlusion, thus allowing tidal volumes to be used as a warning indicator of worsening ETT obstruction. To test our hypothesis, we first modeled the flow and volume responses of a ventilator operating in pressure-controlled mode to progressive ETT occlusion. To confirm our model, we tested the ability of the Puritan Bennett 7200 (Nellcor Puritan Bennett, Pleasanton, CA) and the Siemens 300 (Siemens AB, Solna, Sweden) critical care ventilators to ventilate a test lung with an unobstructed 9.0-mm-inner-diameter (ID) ETT. We then measured delivered tidal volumes, inspiratory flow patterns, and inspiratory airway pressures during ventilation while progressively substituting ETTs with smaller IDs to simulate ETT occlusion.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
The following equations (6) were used to model the response of a critical care ventilator operating in pressure-controlled mode to progressively increasing ETT occlusion [see Appendix 1 for derivation.
At any time t in inspiration,
|
|
where P is set inspiratory pressure, R is airway resistance (an assumption of this model is that airway resistance is independent of flow rate), and C is lung compliance; and
|
|
Predictions of the model were then tested by using a critical care ventilator and a test lung. All experiments were performed in the Respiratory Therapy Laboratory at the University of Chicago. A Puritan Bennett 7200 ventilator was connected via a standard 72-in. disposable ventilator circuit (Allegiance Healthcare Co., McGaw Park, IL) to a 9.0-mm-ID ETT (Mallinckrodt, St. Louis, MO) cut to a length of 5 cm. The other end of the ETT was connected to a test lung (Model 1600I; MI Instruments, Grand Rapids, MI).
The ventilator was then set to volume-controlled mode at a rate of 14 breaths/min, a tidal volume of 700 mL, an inspired oxygen content (FIO2) of 40%, and an end-expiratory pressure (PEEP) of 5 cm H2O. Inspiratory flow rates were set to 50 L/min, a value within the range often used for mechanical ventilation of critically ill patients (1). The resulting inspiratory time was 0.84 s. Test lung compliance was then adjusted until the plateau inspiratory airway pressure reached 30 cm H2O. A 0.1-s end-inspiratory pause was set to allow measurement of the plateau airway pressure. Peak airway pressure with this size ETT exceeded plateau pressures by 0.7 cm H2O. A Bicore CP-100 respiratory monitor was placed in the inspiratory limb of the circuit to serve as an independent monitor of tidal volume.
After inspiratory flow patterns, delivered tidal volumes, peak and pause airway pressures, and inspiratory flow rates were recorded, the 9.0-mm-ID ETT segment was replaced by progressively smaller ETT segments (8.5, 8.0, 7.5, 7.0, 6.5, 6.0, 5.5, 5.0, 4.5, 4.0, 3.5, and 3.0 mm ID) cut to the same length. With each change of ETT, inspiratory flow patterns, delivered tidal volume peaks, and plateau inspiratory airway pressures were recorded after the ventilator and test lung were allowed to equilibrate for 60 s.
After all ETT segments had been tested, the 9.0-mm-ID ETT was replaced, and the ventilator was reset to pressure-controlled mode. PEEP was set to 5 cm H2O, inspiratory airway pressure was set to 25 cm H2O above PEEP (30 cm H2O total), FIO2 was set to 40%, and respiratory rate was set to 14 breaths/min to duplicate the volume-controlled settings. The inspiratory time was set to 1.2 s, with a resulting tidal volume of 680 mL. Progressively smaller ETTs were then substituted as previously, and ventilator variables were recorded for all ETT sizes.
After testing was complete in both volume- and pressure-controlled modes, the Puritan Bennett 7200 ventilator was replaced with the Siemens 300 critical care ventilator, and all experimental steps were repeated. All measurements were made with the on-board volume, pressure, and flow monitors of the Puritan Bennett 7200 or Siemens 300 ventilator, and confirmed with the Bicore CP-100 respiratory monitor. After each change of ETT size, the ventilator was allowed to equilibrate with the test lung for 60 s before measurement of respiratory variables. In volume-controlled mode, airway pressures were measured both at peak and during a 0.1-s end-inspiratory hold maneuver (pause). The resistance of each ETT fragment was calculated by subtracting the peak airway pressure from the plateau airway pressure and dividing by the fixed inspiratory flow rate of 50 L/min. Changes in tidal volumes and airway pressures in both volume- and pressure-controlled modes were compared. Correlations between predicted tidal volumes and those measured during actual bench-testing for all ETT diameters were calculated, as well as those between Siemens and Puritan Bennett ventilators.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
Our model simulating positive-pressure ventilation in pressure-controlled mode predicted three responses to progressive ETT occlusion. First, the model suggested that progressively smaller ETT segments would result in lower initial inspiratory flow rates (Fig. 1). Second, the model predicted that smaller ETT segments would cause inspiratory flow rates at the end of inspiration to be higher (Fig. 1). Finally, the model predicted that delivered tidal volumes would not decrease linearly with ETT obstruction, but would instead remain constant until ETT occlusion was severe (Fig. 2).
|
|
Bench-testing of two critical care ventilators confirmed these predictions. Inspiratory flow patterns in pressure-controlled mode changed qualitatively during progressive ETT occlusion (Fig. 3). With the ETT unoccluded, the flow-versus-time wave form had a down-sloping appearance during inspiration, with the ventilator delivering maximal flow at the beginning of the breath and gradually decreasing flow with continuing lung inflation. ETT occlusion changed the shape of the wave form, reducing gas flow rates at the beginning of the inspiratory cycle and increasing flow at the end of the breath to produce a more constant flow rate over the inspiratory cycle.
|
With both the Puritan Bennett and the Siemens 300 ventilators operating in pressure-controlled mode, nearly no change in inspiratory airway pressure or delivered tidal volumes was noted with smaller ETT segments until the ETT diameter was reduced to 5.0 mm (Table 1). This reduction corresponded to a 70% reduction in cross-sectional area and a 15-fold increase in airflow resistance. Tidal volumes remained >90% of baseline values until the ETT diameter had been reduced to 4.0 mm, producing a 40-fold increase in airflow resistance. When a 3.5-mm-ID ETT was used, delivered tidal volumes continued to be 69% of baseline values. With a 3.0-mm-ID ETT, a level of airway resistance at which persistent high-pressure alarms prevented normal ventilation in volume-controlled mode, tidal volumes in pressure-controlled mode remained at 50% of baseline values. No change in peak inspiratory pressure was noted at any ETT diameter. Measurements of inspiratory airway pressure and delivered tidal volumes differed by <5% between the Bicore CP-100 and the on-board monitors of the Puritan Bennett and the Siemens ventilators.
|
In contrast, when volume-controlled mode was used, peak airway pressures on both ventilators did not differ appreciably from plateau airway pressures when a 9.0-mm-ID ETT was used, but they increased significantly when the diameter of the ETT was reduced (Table 1). Reducing the ETT ID from 8.0 to 5.0 mm increased peak airway pressures from 28 to 38 cm H2O (36% increase). Further reducing the ID to 4.0 mm doubled airway pressures (from 28 to 56 cm H2O). Airway pressures tripled with a 3.5-mm-ID ETT, making ventilation nearly impossible because of repeated high-pressure alarms. When a 3.0-mm-ID ETT was tried, the ventilator was unable to reach the preset tidal volumes because the peak inspiratory pressure alarm sounded repeatedly and disabled set variables.
The correlation between the tidal volumes predicted with our theoretical model and those produced during bench-testing was high for the Siemens (0.998) and the Puritan Bennett (0.993) ventilators. The correlation between ventilators was also excellent (0.997 for delivered tidal volumes in pressure-controlled mode and 0.999 for peak airway pressures in volume-controlled mode).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
An important complication of prolonged mechanical ventilation is progressive airway occlusion due to continuing deposition of respiratory secretions on the inner surface of the ETT. Partial ETT occlusion from deposition of secretions may complicate weaning from mechanical ventilation (7) and predispose patients to atelectasis and pneumonia (8). Severe obstruction may also lead to complete airway obstruction, necessitating emergent airway management.
When patients are ventilated in a volume-controlled mode, the increased airway resistance caused by partial ETT obstruction provides a useful diagnostic tool. Because tidal volumes and inspiratory flow rates are preset, increased airway resistance from progressive ETT occlusion results in a growing gap between peak and plateau inspiratory pressures. This gradient is often used to detect changes in ETT resistance during prolonged mechanical ventilation (5). When patients are ventilated in a pressure-controlled mode, such as pressure control or pressure support, however, inspiratory airway pressures are preset and thus cannot be used to diagnose airway occlusion. Because tidal volumes are not preset in pressure-controlled mode but may vary with changes in lung compliance or airway resistance, we hypothesized that delivered tidal volumes could act as a warning indicator for progressive ETT occlusion.
We found that in pressure-controlled mode, progressive ETT occlusion is almost undetectable by inspection of airway pressures or tidal volumes until nearly complete. In both our mathematical and experimental models of pressure-controlled ventilation, inspiratory airway pressures did not change at any degree of occlusion, and delivered tidal volumes remained at 90% of their original level until the ETT ID had been reduced to 4.0 mm. In contrast, when a volume-controlled mode was used, peak inspiratory airway pressures increased to 115% of baseline levels with a 6.0-mm-ID ETT, 150% of baseline levels with a 5.0-mm-ID ETT, and 200% with a 4.0-mm-ID ETT. Ventilation in volume-controlled mode was impossible with a 3.0-mm-ID ETT because of persistent high airway pressure alarms.
Changes in the inspiratory flow-versus-time graph suggest an explanation for the stability of delivered tidal volumes with increasing ETT occlusion in pressure-controlled mode. Normally, in pressure-controlled mode inspiratory gas flow is greatest at the beginning of inspiration, when the ventilator seeks to increase airway pressure instantaneously from end-expiratory levels to the preset inspiratory value. Once the set inspiratory pressure has been reached, increasing lung recoil forces gradually diminish gas flow, producing the characteristic decelerating flow-versus-time wave form. We found that worsening ETT occlusion significantly changed the shape of the flow-versus-time wave form from the normal, decelerating flow pattern with an unobstructed airway to a flatter, more square-shaped flow pattern. Our mathematical model agreed, predicting that partial ETT obstruction would reduce gas flow at the beginning of the inspiratory cycle and increase flow at the end to flatten out the wave form.
These observations suggest that ETT occlusion causes the delivery of tidal volume to shift from occurring primarily in the beginning of inspiration to occurring evenly throughout inspiration. Because inspiratory airway pressures are fixed in pressure-controlled mode, increased airway resistance from ETT occlusion restricts maximum inspiratory gas flow. Thus, at the beginning of inspiration, the lung inflates at a reduced rate. The lower initial rate of lung inflation, however, also slows the increase in lung recoil forces. Reduced lung recoil forces then permit higher gas flow in the latter half of the inspiratory cycle. Our model demonstrates this change in gas flow distribution, noting that as airway resistance increases, delivery of tidal volume shifts into the latter half of the inspiratory cycle (Fig. 4). This redistribution then allows the ventilator to preserve tidal volumes in the face of increased airway resistance until resistance becomes severe.
|
Our model assumes that progressive ETT obstruction does not result in dynamic hyperinflation and that no spontaneous inspiratory efforts are made. The respiratory rate we used (14 breaths/min), however, is unlikely to induce progressive hyperinflation. Our model also used ETT resistance values obtained at a fixed flow rate of 50 L/min. Although resistance increases with increasing flow rates, such an effect would not significantly change our results, because actual flow rates rarely exceeded 50 L/min, even in pressure control (Fig. 3).
The ability of a ventilator operating in pressure-controlled mode to preserve tidal volume in the face of severe ETT occlusion has two important consequences for the anesthesiologist. First, by maintaining normal ventilatory variables in the face of near-total ETT occlusion, ventilators functioning for prolonged periods in pressure-controlled mode may allow dangerous levels of ETT occlusion to build up undetected. Because respiratory secretions can produce significant ETT obstruction in as little as five days (4), patients ventilated in pressure-controlled mode for such a duration should be considered at increased risk for undetected ETT obstruction. Anticipating this possibility will allow anesthesiologists involved in ventilator and airway management in the intensive care unit (ICU) to make more informed decisions regarding the need for urgent or emergent airway interventions.
The second important consequence involves mechanical ventilation for these patients in the operating room (OR) setting. Because of the disparity in the response of volume- and pressure-controlled modes to partial ETT obstruction, patients ventilated in pressure-controlled mode may have normal respiratory variables but may develop dramatically higher peak airway pressures when a switch to volume-controlled mode is made. With few exceptions, most anesthesia ventilators operate exclusively in a volume-controlled mode in which tidal volumes and inspiratory flow rates are preset (9). In the presence of ETT occlusion with increased airway resistance, transfer from pressure-controlled ventilation in the ICU to an OR ventilator operating in the volume-controlled mode will result in higher peak airway pressures. Because most anesthesia ventilators are unable to deliver adequate ventilation when airway pressures increase (10,11), this change may result in subsequent deterioration of blood gases.
Anesthesiologists should anticipate the possibility that patients ventilated in pressure-controlled mode for prolonged periods may have severe, undetected ETT obstruction, with important consequences for airway management and intraoperative ventilation. If possible, monitoring the flow-versus-time wave form can detect ETT obstruction even in the absence of changes in tidal volumes (Fig. 3). Another approach is to switch to volume-controlled mode before transport to the OR. If airway pressures required to sustain adequate ventilation exceed 50–65 cm H2O in volume-controlled mode, anesthesia ventilator failure is likely (10,11), and preemptive reintubation or an alternative ventilation strategy should be considered.
In summary, we conclude that as a result of redistribution of gas flow during the inspiratory cycle, ventilators operating in pressure-controlled mode may maintain nearly normal tidal volumes and inspiratory pressures despite near-total ETT occlusion. This potential for undetected ETT obstruction with the use of pressure-controlled mode has two important consequences. First, progressive ETT narrowing from deposition of respiratory secretions may occur unnoticed, particularly in patients for whom routine ETT exchange is risky. Second, undetected ETT obstruction should be considered when switching from pressure- to volume-controlled ventilation. With partial ETT occlusion, ventilators may maintain normal tidal volumes in pressure-controlled mode but produce high peak inspiratory pressures in volume-controlled mode. Transferring a patient with partial ETT obstruction from a pressure-controlled mode in the ICU to a volume-controlled mode in the OR may then lead to high inspiratory pressures and anesthesia ventilator failure. Anticipating these issues can allow anesthesiologists to make more informed choices about intraoperative ventilation in critically ill patients.
|
Appendix 1: Derivation of Mathematical Model Simulating Flow Versus Time and Volume Versus Time Curves for a Ventilator Operating in Pressure-Control Mode |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
Let Pvent = preset inspiratory pressure, Pelast = airway pressure resulting from lung elastance, and Presist = airway pressure resulting from resistance to gas flow in the endotracheal tube.
Then, at any time during inspiration in pressure-controlled mode, Pvent = Pelast + Presist.
If gas flow at time t = F(t) and ETT resistance = R, then at any time t, Presist = F(t) x R (this model assumes that R does not vary significantly with gas flow).
Now, if C = lung compliance, then Pelast = V(t)/C, where V(t) = lung volume at time t during inspiration.
Substituting for Pelast and Presist, Pvent = V(t)/C + F(t) x R, and thus F(t) = [Pvent - (V(t)/C]/R, where V(t) = 
0t F(t)dt.
Thus, F(t) = P/R - 1/RC 0t F(t)dt. Solving this equation for F(t) yields F(t) = instantaneous flow = (P/R) x e-t/RC V(t) = instantaneous delivered volume = 0t F(t)dt = PC(1 - e-t/RC).
|
Footnotes |
|---|
Presented in part at the 1999 annual meeting of the American Society of Anesthesiologists, October 11, 1999, Dallas, TX, and the 2000 International Anesthesia Research Society meeting, March 12, 2000, Honolulu, HI.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAppendix 1: Derivation of...References |
|---|
- Tobin MJ. Advances in mechanical ventilation. N Engl J Med 2001; 344: 1986–95.
[Free Full Text] - Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome: the Acute Respiratory Distress Syndrome Network. N Engl J Med 2000; 342: 1301–8.
[Abstract/Free Full Text] - O’Connor MF, Hall JB. ARDS: current treatment and ventilator strategies. Anesth Clin North Am 1998; 16: 155–80.
- Villafane MC, Cinnella G, Lofaso F, et al. Gradual reduction of endotracheal tube diameter during mechanical ventilation via different humidification devices. Anesthesiology 1996; 85: 1341–9.[Web of Science][Medline]
- Moon RE, Camporesi EM. Respiratory monitoring. In: Miller RD, ed. Anesthesia. 4th ed. New York: Churchill Livingstone, 1994: 1253–92.
- Ouellet P. Equation of motion for the respiratory system. In: Waveform and loop analysis in mechanical ventilation. Siemens-Elema AB, 1997:20–9.
- Manthous CA, Schmidt GA, Hall JB. Liberation from mechanical ventilation. In: Hall JB, Schmidt GA, Wood LDH, eds. Principles of critical care. New York: McGraw Hill, 1998: 653–67.
- Cohen IA, Weinberg PF, Fein A, Rowinski GS. Endotracheal tube occlusion associated with the use of heat and moisture exchangers in the intensive care unit. Crit Care Med 1988; 16: 277–9.[Web of Science][Medline]
- Dorsch JA, Dorsch SE. Anesthesia ventilators. In: Understanding anesthesia equipment. 4th ed. Baltimore: Williams & Wilkins, 1998: 309–54.
- Marks JD, Schapera A, Kraemer RW, et al. Pressure and flow limitations of anesthesia ventilators. Anesthesiology 1989; 71: 403–8.[Medline]
- Katz JA, Kallet RH, Alonso JA, Marks JD. Improved flow and pressure capabilities of the Datex-Ohmeda SmartVent anesthesia ventilator. J Clin Anesth 2000; 12: 40–7.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. K. Visaria and D. R. Westenskow Model-Based Detection of Endobronchial Intubation Anesth. Analg., October 1, 2006; 103(4): 888 - 893. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kawati, L. Vimlati, J. Guttmann, G. Hedenstierna, U. Sjostrand, S. Schumann, and M. Lichtwarck-Aschoff Change in expiratory flow detects partial endotracheal tube obstruction in pressure-controlled ventilation. Anesth. Analg., September 1, 2006; 103(3): 650 - 657. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
