Journal of Applied Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 102: 331-339, 2007. First published September 7, 2006; doi:10.1152/japplphysiol.00002.2006 Free Article
8750-7587/07 $8.00
This Article
Free upon publication Free Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
102/1/331    most recent
00002.2006v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Syring, R. S.
Right arrow Articles by Baumgardner, J. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Syring, R. S.
Right arrow Articles by Baumgardner, J. E.

Maintenance of end-expiratory recruitment with increased respiratory rate after saline-lavage lung injury

Rebecca S. Syring,1 Cynthia M. Otto,1 Rebecca E. Spivack,1 Klaus Markstaller,2,3 and James E. Baumgardner3,4

1Department of Clinical Studies, Section of Critical Care, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; 2Department of Anesthesiology, Johannes Gutenberg University, Mainz, Germany; 3Department of Anesthesiology and Critical Care, School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania; and 4Oscillogy, Folsom, Pennsylvania

Submitted 3 January 2006 ; accepted in final form 19 August 2006


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 
Cyclical recruitment of atelectasis with each breath is thought to contribute to ventilator-associated lung injury. Extrinsic positive end-expiratory pressure (PEEPe) can maintain alveolar recruitment at end exhalation, but PEEPe depresses cardiac output and increases overdistension. Short exhalation times can also maintain end-expiratory recruitment, but if the mechanism of this recruitment is generation of intrinsic PEEP (PEEPi), there would be little advantage compared with PEEPe. In seven New Zealand White rabbits, we compared recruitment from increased respiratory rate (RR) to recruitment from increased PEEPe after saline lavage. Rabbits were ventilated in pressure control mode with a fraction of inspired O2 (FIO2) of 1.0, inspiratory-to-expiratory ratio of 2:1, and plateau pressure of 28 cmH2O, and either 1) high RR (24) and low PEEPe (3.5) or 2) low RR (7) and high PEEPe (14). We assessed cyclical lung recruitment with a fast arterial PO2 probe, and we assessed average recruitment with blood gas data. We measured PEEPi, cardiac output, and mixed venous saturation at each ventilator setting. Recruitment achieved by increased RR and short exhalation time was nearly equivalent to recruitment achieved by increased PEEPe. The short exhalation time at increased RR, however, did not generate PEEPi. Cardiac output was increased on average 13% in the high RR group compared with the high PEEPe group (P < 0.001), and mixed venous saturation was consistently greater in the high RR group (P < 0.001). Prevention of end-expiratory derecruitment without increased end-expiratory pressure suggests that another mechanism, distinct from intrinsic PEEP, plays a role in the dynamic behavior of atelectasis.

acute lung injury; atelectasis; cyclical recruitment; intrinsic positive end-expiratory pressure; viscoelasticity; arterial oxygen oscillations

IN ACUTE LUNG INJURY (ALI), and experimental models of ALI, the application of positive end-expiratory pressure (PEEP) is a commonly used strategy to prevent end-expiratory collapse of alveoli. PEEP can maintain alveolar recruitment, which improves arterial oxygen concentrations in some patients (7, 22) and may limit ventilator-associated lung injury by decreasing cyclical recruitment (14, 18, 22, 45, 63, 65, 68, 69). The elevated intrathoracic pressure that results from PEEP, however, impairs venous return and may reduce cardiac output (7, 22, 53). Additionally, increased levels of PEEP are generally associated with higher airway pressures that may aggravate overdistension of alveoli (7, 22).

In surfactant depletion models of lung injury, several investigators have recently explored the use of short exhalation times to prevent end-expiratory collapse despite a low end-expiratory pressure (5, 39, 47). One mechanism proposed to explain this prevention of end-expiratory derecruitment is the generation of intrinsic PEEP. Intrinsic PEEP results from incomplete alveolar emptying during exhalation and can be exacerbated by shortened expiratory times. Both intrinsic PEEP and extrinsic PEEP increase alveolar and intrathoracic pressures (50), and equivalent levels of intrinsic and extrinsic PEEP should have similar hemodynamic consequences. Maintaining recruitment with short expiratory times may, therefore, have no hemodynamic benefit compared with maintaining recruitment with extrinsic PEEP (46, 47). None of the experimental studies in surfactant depletion models that directly evaluated the effects of a short exhalation time on end-expiratory atelectasis, however, have measured intrinsic PEEP (5, 39, 47). Thus the hypothesized role of intrinsic PEEP as the mechanism for the maintenance of recruitment with short exhalation times in saline-lavage injury (33, 47) has not been verified.

In a surfactant depletion model of acute lung injury in rabbits, we compared maintenance of lung recruitment with a slow respiratory rate and high extrinsic PEEP, to maintenance of lung recruitment with a fast respiratory rate and low extrinsic PEEP. Cyclical recruitment was continuously assessed with a fast responding arterial PO2 probe, which measured changes in shunt fraction throughout the respiratory cycle. After adjustment of either extrinsic PEEP or respiratory rate to levels that prevented end-expiratory collapse, we measured intrinsic PEEP, cardiac output, and mixed venous oxygen saturation. Our hypothesis was that the total PEEP (extrinsic plus intrinsic) required to maintain end-expiratory lung recruitment would be identical with both strategies, and depression of cardiac output and mixed venous saturation would be similar between these two ventilator settings.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 
Animal preparation.   The study protocol was approved by the Institutional Animal Care and Use Committee at the University of Pennsylvania. The animal preparation is similar to a previous report (5) with the following exceptions: general anesthesia was maintained with hydromorphone and midazolam infusions, a catheter was placed into the right ventricle for mixed venous blood gas sampling, cardiac output was monitored via continuous pulse contour analysis, and the fast response oxygen probe was positioned in the distal aorta.

Seven female New Zealand White rabbits, weighing 3.5–3.9 kg (mean weight 3.7 kg), were sedated with an intramuscular injection of 140 mg ketamine and 20 mg xylazine. Following sedation, the rabbits were positioned in dorsal recumbency for the remainder of the experiment. Intravenous access and arterial pressure monitoring were established via ear vessels, and a balanced electrolyte solution (Normosol-R, Abbott Laboratories, Chicago, IL) was administered at 10 ml/h for maintenance fluid requirements. General anesthesia was established with intravenous loading doses of hydromorphone (0.3 mg/kg) and midazolam (0.8 mg/kg) and maintained with infusions of 40 µg·kg–1·min–1 and 0.008–0.009 mg·kg–1·min–1, respectively. A tracheostomy was performed under a surgical plane of anesthesia, and a 4.0 mm uncuffed endotracheal tube was sealed in the trachea with umbilical tape. All mechanical ventilation (Servo 900 C, Siemens) was in pressure control mode throughout the experiment. A respiratory monitor (CO2SMO-Plus, Novametrix, Wallingford, CT) was attached to the end of the tracheostomy tube to monitor airway pressure at the mouth, end-tidal carbon dioxide (ETCO2), and dynamic intrinsic PEEP (PEEPidyn).

A 4-Fr. catheter was inserted via an introducer in the right jugular vein and advanced to the level of the right ventricle, as guided by pressure waveform analysis. This catheter was used for mixed venous blood gas sampling as well as iced saline injection for transpulmonary thermodilution cardiac output determination. Catheters (4-Fr., 8 cm; Pulsiocath, Pulsion Medical Systems) were inserted via surgical cutdown into both the left and right femoral arteries. The tips of those catheters extended into the distal aorta. One catheter was used for transpulmonary cardiac output, direct arterial blood pressure measurement, and continuous pulse contour cardiac output. The continuous pulse contour cardiac output was calibrated at the beginning of the experiment against transpulmonary thermodilution cardiac output. A fiber-optic fluorescence-quenching oxygen probe (FOXY AL-300, Ocean Optics, Dunedin, FL) was inserted through the other catheter, with the end of the probe extending 3–5 mm beyond the catheter tip in the distal aorta. The oxygen probe was calibrated in vivo against partial pressures of oxygen obtained via conventional arterial blood gas analysis (Stat Profile CCX, NOVA Biomedical) at two inspired oxygen concentrations (FIO2 = 0.21 and 1.0). Data acquisition software (OOISensors, Ocean Optics) displayed the arterial PO2 in real time at a digital sampling rate of 3.5 Hz.

Immediately prior to induction of lung injury, neuromuscular blockade was established via an intravenous loading dose (0.4 mg/kg) of pancuronium and maintained with a constant rate infusion at 0.2 mg·kg–1·h–1 for the duration of the experiment. Surfactant depletion was induced by instilling 26 ml/kg of warm, balanced electrolyte solution (Normosol-R, Abbott Laboratories, Chicago, IL) into the lungs via the endotracheal tube, followed by immediate drainage by gravity. This lavage was performed three times for all rabbits, with 3–5 min between each lavage.

All rabbits had a phenylephrine infusion instituted immediately following saline lavage to maintain hemodynamic stability following lung injury, and the infusion rate was fixed at 0.4 mg/h (the same dose for every rabbit) for the duration of the experiment. In addition, 10–20 ml hydroxyethyl starch boluses were administered, to a maximum cumulative dose of 50 ml, when the systolic blood pressure was <60 mmHg and respiratory variation in blood pressure suggested hypovolemia. Once the study protocol comparing the three ventilatory strategies commenced, additional fluid boluses were prohibited.

Study design.   For each of seven rabbits, three different ventilator strategies were investigated following surfactant depletion lung injury. Two settings investigated ventilatory strategies to avoid end-expiratory collapse: either high respiratory rate with low PEEP or low respiratory rate, high PEEP. A third setting, using a low respiratory rate and low PEEP, was used to demonstrate that in the absence of elevations in either respiratory rate or PEEP, end-expiratory collapse would occur as a result of lung injury. All rabbits were ventilated with an FIO2 of 1.0.

Ventilator settings for each rabbit were determined after lavage according to a defined protocol and then remained fixed for the rest of the experiment in that rabbit. Plateau pressure (Pplat), PEEP, respiratory rate (RR), and inspiratory-to-expiratory (I:E) ratio were selected individually at the beginning of each experiment, with a goal of achieving equivalent levels of lung recruitment in the two main groups despite individual variability in the pressure responsiveness and dynamics of atelectasis. First, Pplat was adjusted during a series of 5–8 s inspiratory pauses to find the region of Pplat, where arterial partial pressure of O2 (PaO2) indicated nearly maximal lung recruitment but further increases in Pplat produced only small changes in PaO2. The target Pplat was set to 2–3 cm above that point, up to a maximum limit allowed by protocol of 35 cmH2O. Next, PEEP for the high PEEP group was set by a series of end-expiratory pauses, searching for the level of PEEP where derecruitment began and then setting the target high PEEP at 2–3 cmH2O above this closing pressure, up to a limit of 15 cmH2O. Low PEEP was targeted at 2–4 cmH2O for all rabbits. Finally, RRs and I:E ratio were adjusted to find a combination of high rate, low PEEP that provided a PaO2 similar to the PaO2 in the low rate, high PEEP group, while allowing cyclical recruitment in the low rate, low PEEP group (Fig. 1). By protocol, high rate could be increased up to 24 breaths/min, low rate could be decreased to 6 breaths/min, and I:E could be varied over the range 1:2 to 4:1.


Figure 1
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 1. Data from 1 rabbit demonstrating the use of a rapid time-response intra-arterial PO2 probe to adjust ventilator settings to achieve equivalent alveolar recruitment. Upper graph [respiratory rate 7, inspiratory-to-expiratory ratio (I:E) 2:1, positive end-expiratory pressure (PEEP) 4 cmH2O, plateau pressure 30 cmH2O] shows substantial cyclical recruitment with large PaO2 oscillations from 170 to 410 mmHg within each breath. The bar marks the time period for 1 breath. Bottom graphs illustrate maintenance of end-expiratory alveolar recruitment (increased mean PaO2 and reduced amplitude of oscillations) by either increased respiratory rate (left) or increased extrinsic PEEP (right). Recruitment by respiratory rate settings: respiratory rate 24, I:E 2:1, PEEP 4 cmH2O, plateau pressure 30 cmH2O. Recruitment by extrinsic PEEP settings: respiratory rate 7, I:E 2:1, PEEP 14 cmH2O, plateau pressure 30 cmH2O.

 
After ventilator parameters for each rabbit were fixed, the high respiratory rate, low PEEP and low respiratory rate, high PEEP settings were performed in random order, followed by the low rate, low PEEP setting. Each group of three settings was performed in triplicate for each rabbit. Prior to each setting, the rabbits were briefly disconnected from the ventilator to achieve an equivalent lung volume status at zero end-expiratory pressure (ZEEP). During this disconnection, an additional dead space of 32 ml was inserted into the airway, between the Y piece of the ventilator circuit and the CO2SMO monitor, prior to ventilating with high respiratory rate, and removed prior to ventilating with low respiratory rate. Recruitment maneuvers were not performed prior to any ventilator setting.

Real-time assessment of lung recruitment was determined by average PaO2 and by the amplitude of PaO2 oscillations, as measured by the intra-arterial oxygen probe (Fig. 1) for all three ventilator settings. Peak and trough PaO2 concentrations were recorded from breath-to-breath oscillations, and the amplitude was calculated as the difference between these two values.

Simultaneous mixed venous and arterial blood gas samples were obtained from catheters in the right ventricle and distal aorta, respectively, for the high respiratory rate, low PEEP and low respiratory rate, high PEEP settings. Samples were collected over several respiratory cycles (52) and were analyzed within 5 min of collection. Venous admixture (Qs/Qt) was calculated according to the standard equation: Qs/Qt = (CcO2 – CaO2)/(CcO2 – CvO2), where CcO2 is the end-capillary, CaO2 is the arterial, and CvO2 is the mixed venous oxygen content.

At all three ventilator settings, numerous directly monitored and calculated hemodynamic parameters were recorded from the PiCCO monitor (Pulsion Medical Systems, Munich, Germany). Cardiac output, averaged over several respiratory cycles, was recorded from continuous pulse contour analysis. Blood pressure (systolic and diastolic), heart rate, core body temperature, and systemic vascular resistance were also recorded from the PiCCO monitor. At each ventilator setting, ETCO2, RR, expiratory tidal volume (Vte), Pplat, mean airway pressure (Pmean), PEEP, and I:E ratio were measured (CO2SMO-Plus). Intrinsic PEEP was measured at each setting by the dynamic method (58).

Data were analyzed by two-way, repeated-measures ANOVA (SigmaStat 3.1, SPSS, Chicago, IL). The first factor was treatment group, with two levels (high rate, low PEEP vs. low rate, high PEEP for arterial and venous blood gas data) or three levels (high rate, low PEEP vs. low rate, high PEEP vs. low rate, low PEEP for hemodynamic and respiratory data). The second factor was replicate number with three levels (first, second, or third). ANOVA was preceded by normality testing by Kolmogorov-Smirnov tests and equal variance testing by the Levene median test. Data were transformed with power transforms as necessary to satisfy the normality and equal variance assumptions (13). Post hoc testing was carried out by Tukey's test. Residuals were inspected visually and with Kolmogorov-Smirnov testing.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 
The postlavage protocol for setting the ventilator resulted in the parameters shown in Table 1. I:E ratio was 2:1 in all rabbits. On average, low respiratory rate was near the minimum of 6 breaths/min and high respiratory rate was near the maximum of 24 breaths/min allowed by protocol. Pplat, and PEEP in the high PEEP group, were both on average substantially less than the maximum pressures allowed by protocol. The similar Pplat and high PEEP among rabbits suggests that the lavage injury on our model was reproducible. The average volume of supplemental hydroxyethyl starch was 41 ± 9 ml (mean ± SD).


View this table:
[in this window]
[in a new window]

  		Table 1. Ventilator settings

 
All variables satisfied the Levene median equal variance testing without data transformation. Several variables required power transformations to satisfy normality conditions: cardiac output was transformed by x2, systemic vascular resistance by x1.9, base excess by x0.5, lactate by x0.3, and PaO2 oscillation amplitude by (x – 8)0.33. After transformations, all variables satisfied both normality and equal variance testing before the analysis with two-way ANOVA. Differences in heart rate were not tested for significance because three rabbits had heart rates that exceeded the maximum measurable value of 240 beats/min. Differences between replicate number for the three sets of measurements in each rabbit were not significant for any of the variables.

Cyclical recruitment was assessed by the amplitude of PaO2 oscillations measured by the fast intra-arterial probe (Fig. 2) The PaO2 oscillation amplitude in the low rate, low PEEP group was substantially larger than the oscillations in the other two groups (Fig. 2, P < 0.001). The smaller difference in amplitude between the high rate, low PEEP group and the low rate, high PEEP group was also significant (P < 0.001), indicating slightly more cyclical recruitment in the high PEEP group.


Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 2. Amplitude of PaO2 oscillations within each breath, as an index of cyclical recruitment. Rate, respiratory rate. Error bars represent mean + SD. Data for 7 rabbits, with each setting replicated 3 times per rabbit. *Difference high rate, low PEEP vs low rate, high PEEP; {dagger}difference high rate, low PEEP vs. low rate, low PEEP; {ddagger}difference low rate, high PEEP vs. low rate, low PEEP; P ≤ 0.05, ANOVA.

 
Equivalence of average lung recruitment between the high rate, low PEEP group and the low rate, high PEEP group was assessed by three measures: 1) PaO2 from arterial blood gas analysis (Fig. 3 and Table 2); 2) calculation of venous admixture from the arterial and mixed venous blood gas analysis (Table 2); and 3) alveolar dead space fraction (Table 2). The higher average PaO2 in the high PEEP group (Fig. 3 and Table 2) was not significantly different between the two groups (P = 0.07), although the power to detect a difference in PaO2 was low (0.37). Assessment of average recruitment by PaO2 alone, however, does not take into account potential average changes in mixed venous saturation (SvO2) between the groups as cardiac output changes with different ventilator settings. Equivalence of average recruitment was also assessed by calculated venous admixture, which does account for changes in venous saturation (as well as alveolar PO2 changes), and by this criteria the average recruitment was better in the high PEEP group (P = 0.006). Average lung recruitment is expected to alter efficiency of CO2 exchange as well as O2 exchange. Recently, a simple formula to approximate alveolar dead space fraction (25), Vd,alv/Vt,alv {approx} (PaCO2 – PETCO2)/PaCO2, has been shown to be associated with changes in average recruitment as assessed by computed tomography (CT) analysis (17). By this criterion (Table 2), there was no difference in average lung recruitment (P = 0.23) between the two groups, although the power to detect a difference in alveolar dead space was small (0.11). Table 2 presents hemodynamic, respiratory, and blood gas results. Cardiac output was significantly greater in the high rate, low PEEP group than in the low rate, high PEEP group (P < 0.001) and venous saturation was similarly improved (P < 0.001). Also, systolic blood pressure (P = 0.002) and diastolic blood pressure (P = 0.008) were increased in the high rate, low PEEP group. These results demonstrate substantially improved hemodynamics when end-expiratory recruitment was maintained by increased respiratory rate vs. increased PEEP.


Figure 3
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 3. Comparison of arterial oxygen tensions (PaO2) from arterial blood gas samples for paired ventilator settings of high rate, low PEEP and low rate, high PEEP. The solid line is the line of identity. bullet, Paired PaO2 tensions. Data are from 7 rabbits, with settings replicated 3 times per rabbit.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Cardiovascular, respiratory, and blood gas results.

 
Figure 4 shows the paired cardiac outputs compared with the line of identity. Cardiac output was greater in the high rate, low PEEP group for all 21 of these comparisons. The percent increase in cardiac output in the high rate group averaged 13% (range 1–34%). Similar pairing for the venous saturation data is shown in Fig. 5.


Figure 4
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 4. Comparison of cardiac output for paired ventilator settings. The solid line is the line of identity. bullet, Paired cardiac outputs. Data are from 7 rabbits, with settings replicated 3 times per rabbit.

 

Figure 5
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 5. Comparison of mixed venous oxygen saturation (SvO2) from mixed venous blood gas samples for paired ventilator settings. The solid line is the line of identity. bullet, paired SvO2 concentrations. Data are from 7 rabbits, with settings replicated 3 times per rabbit.

 
Intrinsic PEEP was below the detection threshold at all ventilator settings (Table 2). The absence of significant gas trapping is supported by finding of equivalent tidal volumes between the high rate, low PEEP group and the low rate, low PEEP group. The results confirm that dynamic compliance was equivalent between these groups (because tidal volume was the same at the same Pplat and PEEP), which is contrary to the lower compliance expected with gas trapping and increased end-expiratory residual volume.

Figure 6 examines the potential influence of the differences in average recruitment (as assessed by differences in shunt fraction) on the observed differences in cardiac output. The slightly negative slope in this relationship is not significantly different from zero (P = 0.11, power = 0.36).


Figure 6
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 6. The relationship between the difference in shunt fraction compared with the difference in cardiac output for the high respiratory rate, low PEEP setting vs. the low respiratory rate, high PEEP setting. Circles represent paired data points. The solid line is the regression line, y = 0.0842 – 0.0039x. The negative slope is not significantly different from zero (P = 0.11).

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 
This study in a surfactant depletion lung injury model in rabbits demonstrated that approximately equivalent reductions in cyclical recruitment could be achieved using either moderate elevations in respiratory rate or application of extrinsic PEEP. The mechanism for maintenance of end-expiratory recruitment using moderately elevated respiratory rates, however, was not related to generation of intrinsic PEEP. In addition, maintenance of recruitment via increased respiratory rates afforded improved cardiac output, mixed venous saturation, and systolic blood pressure, compared with results obtained when PEEP was used to maintain lung recruitment.

Limitations of the study.   Differences in cardiac output between the paired ventilator settings (high rate, low PEEP vs. low rate, high PEEP) were determined by use of arterial pulse contour analysis. This technique uses a proprietary, and confidential, algorithm to calculate cardiac output from the waveform of the arterial pressure tracing and an assessment of arterial input impedance and aortic compliance (21, 54). The influence of ventilator settings, such as PEEP, on the accuracy of this method has not been reported. Cardiac output via the pulse contour analysis method, however, does compare favorably with thermodilution over a range of doses of a variety of vasoactive drugs (21) that would be expected to change impedance and compliance much more than variations in ventilator settings. Also, measurements of mixed venous saturation provided a supplemental assessment of cardiac output. For cardiac outputs above a critical range, systemic oxygen metabolism (VO2) becomes independent of systemic oxygen supply (DO2), and increases of cardiac output above that critical range result in decreased extraction to maintain VO2 constant (61). In this supply-independent range of DO2, therefore, decreases in cardiac output are reflected in decreases in mixed venous saturation. The critical oxygen delivery, below which VO2 becomes supply dependent, has not been reported in rabbits. In multiple species, however, the onset of supply dependence is associated with reductions of mixed venous saturations to the range of 30–50% (1, 8, 36, 60), well below the lowest venous saturation of 64% observed in the current study. The paired differences in SvO2 between the two groups (high rate, low PEEP vs. low rate, high PEEP), therefore, support the cardiac output data measured by pulse contour analysis. Finally, the finding of increased cardiac output in the high rate, low PEEP group is consistent with the finding of lower end-expiratory pressure in this group, although other mechanisms such as reflex changes in cardiac output cannot be ruled out in this intact animal preparation.

Precisely matching arterial CO2 tensions over a wide range of ventilator settings is technically quite difficult. We attempted to adjust for the markedly larger minute ventilation in the high rate, low PEEP group by insertion of an additional dead space volume in the airway. Nevertheless, the arterial CO2 tension in the low rate, high PEEP group was systematically greater than the arterial CO2 tension in the high rate, low PEEP group. Increased arterial CO2 levels, however, are associated with increased cardiac output, a result of catecholamine release and systemic vasodilation (51, 55, 67). If any effect would be expected from the difference in PaCO2 between the two strategies, it would be to increase the cardiac output in the low rate, high PEEP group. Therefore this difference in PaCO2 between the groups probably did not contribute to the higher cardiac output in the high rate, low PEEP group; it actually would be expected to reduce the paired differences in cardiac output.

We measured intrinsic PEEP by the dynamic method (PEEPidyn), as described by Rossi et al. (58) and implemented in the CO2SMO-Plus respiratory monitor. The dynamic method has been shown to underestimate intrinsic PEEP as measured by the static, end-expiratory occlusion method (26). For lungs with homogeneous airways resistances, however, the values measured by PEEPidyn are consistently ~76% of the values obtained by static measurements (26). Even in the worst case scenario of very heterogeneous lungs and a detection threshold of 0.2 cmH2O, the maximum intrinsic PEEP measured by static determinations would be estimated to be <2 cmH2O. Additionally, in a series of pilot studies in our model, we confirmed that a measured PEEPidyn of zero was accompanied by an end-expiratory flow of zero, at respiratory rates of up to 30 breaths/min and I:E ratios of 2:1.

Our study used arterial oxygenation in rabbits breathing 100% oxygen as an index of lung recruitment. At the beginning of each experiment, we adjusted the PEEP in the low rate, high PEEP group and the respiratory rate in the high rate, low PEEP group, with the goal of achieving the same PaO2 in the two groups. During ventilation with 100% oxygen, PaO2 correlates with the percentage of atelectatic lung measured by CT (38, 40, 49) and with true shunt fraction measured by the multiple inert gas elimination technique (46). Matching PaO2 with such different ventilator settings and maintaining this matching with exact precision over the course of the experiment, however, is nearly impossible, even with the real-time information provided by the PaO2 probe. In our study, there was a small and nonsignificant trend for the PaO2 to be greater in the low rate, high PEEP group than in the high rate, low PEEP group (Fig. 3). Additionally, because the different ventilator settings affected cardiac output and venous saturation as well as PaO2, an identical PaO2 did not necessarily reflect identical venous admixture. In our study, the venous admixture was systematically lower in the low rate, high PEEP group (Table 2), indicating better average recruitment in this group. A precise equivalence of average recruitment between the two groups, as assessed by venous admixture, would have resulted in less PEEP in the low rate, high PEEP group, which might have increased cardiac output in this group and made the difference in cardiac output between the groups less marked. However, the data of Fig. 6 argues against a major role of this difference in recruitment in determining the differences in cardiac output between the groups. There was no significant relationship between the magnitude of the differences in shunt fraction between the groups and the magnitude of the differences in cardiac output.

Comparison to previous studies.   Several prior studies have suggested that shortened expiratory times can prevent end-expiratory derecruitment. Neumann et al. (47) were the first investigators to explore the potential impact of short exhalation times on end-expiratory collapse. They used dynamic CT to measure time-dependent atelectasis during prolonged expiratory pauses in pigs with saline lavage, oleic acid, or LPS lung injury. Their results suggested that short expiration times (<0.6 s for oleic acid injury) could allow exhalation but maintain end-expiratory recruitment without extrinsic PEEP. In a subsequent study, Neumann et al. (46) investigated the effects of short exhalation times on average lung recruitment during tidal breathing in pigs after oleic acid injury. Exhalation times of 0.5 and 1.0 s with ZEEP reduced shunt fraction compared with control, but did not reduce shunt fraction as much as 20 cmH2O of extrinsic PEEP. Despite providing less average lung recruitment than extrinsic PEEP, the short exhalation times were associated with generation of significant intrinsic PEEP. Reductions in cardiac output, compared with control, were similar in all three study groups. There are several possible reasons for the contrasting results of Neumann's study and our current study. Most notably, the animal species, lung size, and lung injury models are different between these studies. Mishima et al. (44) presented evidence, for example, that during some phases of oleic acid injury, airway resistance is increased, a phenomenon that would certainly encourage generation of intrinsic PEEP.

Markstaller et al. (39) used dynamic CT to study time-dependent collapse in saline-lavaged pigs. Their results suggested that exhalation times >1 s predispose to end-expiratory collapse. In a following study, Markstaller et al. (40) extended their work with dynamic CT in saline lavaged pigs to examine tidal breathing. Their data demonstrate substantial end-expiratory collapse with an exhalation time of ~1.9 s, confirming that this commonly used exhalation time is associated with cyclical recruitment of atelectasis.

In one of our prior studies (5), we used a rapidly responding PaO2 probe to assess cyclical recruitment in saline-lavaged rabbits. Maintenance of end-expiratory recruitment was strongly influenced by both PEEP and respiratory rate and substantial recruitment could be maintained, during tidal breathing, at an exhalation time of 1.0 s. We did not measure intrinsic PEEP or cardiac output at individual ventilator settings in that study.

In summary, these prior studies of the effects of exhalation time on maintenance of end-expiratory recruitment have suggested that exhalation times in the range of 0.5 to 1.0 s can prevent end-expiratory collapse. Our finding that a mean exhalation time of 0.83 s (range 0.83–0.91 s) prevented end-expiratory collapse is consistent with these prior reports.

Other previous studies have suggested that even shorter expiration times are required to generate intrinsic PEEP in normal rabbit lungs. Cartwright et al. (9) demonstrated a negligible increase in functional residual capacity in normal rabbits ventilated through a 3.0 mm endotracheal tube when the exhalation time was decreased from 1.8 to 0.8 s, indicating that very little gas trapping was induced by the reduced exhalation time. Fujino et al. (16) reported that ventilation of normal rabbits through a 4.0 mm endotracheal tube at a rate of 50 breaths/min and I:E ratio of 2:1 (an exhalation time of 0.4 s) did not generate static intrinsic PEEP. Ludwigs et al. (37) reported intrinsic PEEP ≤5 cmH2O in normal rabbits ventilated through a 3.0 mm endotracheal tube at an exhalation time of 0.4 s (37).

Our results are also consistent with these previous reports of the short exhalation times required to generate intrinsic PEEP in normal rabbit lungs. In our model, nondependent regions of the lung are expected to be nearly normal and dependent regions are expected to have reduced compliance. The relevant time constant for the generation of intrinsic PEEP, however, is the local RC time constant, i.e., local airways resistance times local compliance (6). Reduction of compliance in the dependent regions would be expected to reduce the RC emptying time constant and make these regions less prone to develop intrinsic PEEP (3, 31), compared with the more normal, nondependent parts of the lung.

Implications—clinical ventilator management.   Our results suggest that in surfactant depletion lung injury, cyclical recruitment of atelectasis can be avoided by judicious choice of exhalation time without generation of intrinsic PEEP. This strategy for maintaining end-expiratory recruitment, compared with maintaining recruitment with extrinsic PEEP, results in significantly improved cardiac output and would have obvious advantages in clinical ventilator management of acute respiratory distress syndrome (ARDS). Several limitations, however, preclude direct translation of our results to clinical ventilator management.

First, there currently is no widely adopted method to directly measure cyclical recruitment at the bedside in routine care, and it is therefore difficult to determine or predict this optimum exhalation time. Static pressure volume curves have been advocated to predict end-expiratory recruitment by examining the measured end-expiratory pressure in relation to the critical closing pressure on the expiratory limb of the static pressure-volume curve (56, 57) or, alternatively, in relation to the maximal tidal compliance in a decremental PEEP trial (12, 27, 29). The prediction of dynamic behavior based on static pressure-volume relationships, however, does not take into account the dynamics of end-expiratory collapse. Whether collapse occurs at a particular end-expiratory pressure depends not just on that pressure and the volume history, but also on how long the airway has remained at that pressure (5). Several methods that have the potential to directly assess cyclical recruitment have been applied in recent research studies, including electrical impedance tomography (15, 32, 71), dynamic CT (39, 40, 46, 47, 71), subpleural vital microscopy (24, 65), rapid continuous PaO2 monitoring (5), timed blood gas collection (52), and fast pulse oximetry (66). None of these methods have been used, however, to study the dynamics of end-expiratory collapse in ARDS. It therefore remains unknown if end-expiratory collapse in ALI and ARDS demonstrates dynamic behavior similar to the dynamics observed in our saline lavage model. It is also currently unknown how prevalent cyclical recruitment is with ventilator strategies commonly used in ARDS.

Second, our model of saline lavage lung injury in rabbits is not a complete representation of ARDS. The impacts of species, lung size, lung maturity, and type of lung injury on the dynamics of end-expiratory collapse have not been investigated and we cannot extrapolate the data of our model to predict the dynamics of atelectasis in ARDS. The saline lavage model replicates the surfactant dysfunction of ARDS (34, 70), but other features such as epithelial and endothelial damage and alveolar inflammation take several hours to develop after lavage (41, 59). Our model of mild injury with surfactant depletion in mature lungs is most likely relevant to early stages of injury in patients at risk for development of ARDS and ALI (52, 57, 70).

Third, the ventilator settings in our study were chosen for optimal study of the dynamics of end-expiratory collapse, not to replicate clinically relevant ventilator settings for ARDS. The large tidal volume and low respiratory rate in our low PEEP, low rate group (the group that demonstrated cyclical recruitment) have no counterpart in clinical ventilator management of ARDS. Slow respiratory rates and large tidal volumes are common, however, in the management of patients at risk for ALI, such as operative management of trauma patients.

Implications—dynamics of atelectasis.   Most of the recent experimental (37, 46) and clinical (2, 11, 28, 35, 42, 43, 72) studies investigating the effects of a short exhalation time on average lung recruitment, and more specifically on maintenance of end-expiratory recruitment (5, 39, 47), have proposed intrinsic PEEP as the mechanism for prevention of end-expiratory derecruitment. There are, however, other time-dependent phenomena relevant to both recruitment and collapse. For example, several investigators have studied the kinetics of surfactant transport and adsorption to a gas-liquid interface (20, 23), which would have a rapidly increasing area during lung recruitment (20). Also, considerable attention has been given to the intrinsic mechanical properties of lung tissue (4, 10, 30, 44, 62, 64). Viscoelastic and plastoelastic models suggest characteristic time constants for expansion and contraction of lung tissue, independent of alveolar pressure-flow relationships. In this regard, it is of interest that time constants estimated on the basis of intrinsic tissue mechanics are generally larger than the RC time constant (airways resistance times regional compliance) relevant to intrinsic PEEP. D'Angelo et al. (10), for example, reported a viscoelastic time constant for normal rabbits of 0.81 s, substantially slower than the time constants of <0.4 s required to generate intrinsic PEEP (16).

Airway closure during exhalation may also be delayed by the time it takes for airway lining fluid to flow, coalesce, and form liquid bridges (48). Because these flows are driven by surface tension instability, increases in surface tension after saline lavage are expected to speed the formation of liquid bridges and airways occlusion (48). Similarly, recruitment of a closed airway could be delayed by the time it takes for an air-liquid meniscus to transit the airway, given finite surface tension and viscosity (19), and a similar phenomenon might introduce time dependence in airway closure. Although airway closure and acinar air trapping per se would not give rise to rapid increases in shunt fraction, similar time-dependent phenomena in the more complex geometry of the alveolar duct might also be rate limited by finite times for fluid flow.

In summary, we demonstrated, in saline-lavaged rabbits, that a respiratory rate of 24 breaths/min and a low PEEP of 3 cmH2O could limit cyclical recruitment as well as a slower rate of 7 breaths/min and a larger PEEP of 14 cmH2O. At a low PEEP of 3 cmH2O, an exhalation time of 0.83 s prevented end-expiratory derecruitment but a longer exhalation time of 2.9 s did not. The mechanism of maintained end-expiratory recruitment with short exhalation times, however, was not intrinsic PEEP. Consistent with this finding, maintaining end-expiratory recruitment with rapid respiratory rates produced better cardiac outputs and mixed venous saturations than maintenance of end-expiratory recruitment with high PEEP. Our results suggest that an optimally chosen expiratory time can prevent end-expiratory collapse but still allow sufficient time for complete exhalation, as implied by the absence of intrinsic PEEP. The finding of negligible intrinsic PEEP in turn suggests that other mechanisms, for example, lung tissue mechanics, kinetics of surfactant transport, or finite time for the flow of airway lining fluid, were responsible for delaying end-expiratory collapse in this surfactant depletion model of lung injury.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 
This study was supported in part by National Institutes of Health GM-64486 [C. M. Otto, principal investigator (PI)], DFG-MA2398/3 (K. Markstaller, PI), and the Department of Anesthesiology and Critical Care, University of Pennsylvania.


   DISCLOSURES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 DISCLOSURES
ACKNOWLEDGMENTS
 REFERENCES
 
Dr. Baumgardner is president and sole owner of Oscillogy LLC, a small business that manufactures and sells scientific research equipment. Oscillogy has a long-term strategic commitment to development and commercialization of new technologies that emphasize high temporal resolution for the study of time-dependent phenomena in physiology research. In the short term, however, Oscillogy has only one product available, a system for performing MIGET under the usual steady-state assumptions. Currently the company does not own or license any new technology directed at time-dependent measurements, and there are no patents pending in this area. In particular, Oscillogy has no commercial interest in or connection to the fast PaO2 probe used in this study (sold by Ocean Optics) or the fast pulse oximeter recently used by our research group (sold by Masimo) in a study reported in abstract form (Ref. 66).


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
REFERENCES
 
The authors thank the Emergency Service at the Ryan Veterinary Hospital, University of Pennsylvania, for the use of the blood gas machine (Stat Profile CCX, NOVA Biomedical) in this study. The authors also thank Dr. Vicki Campbell for assistance with developing the anesthetic protocol.

Current affiliation for Dr. Markstaller: Department of Anesthesia, University of Bern, Switzerland.


   FOOTNOTES
 

Address for reprint requests and other correspondence: J. E. Baumgardner, Dept. of Anesthesiology and Critical Care, Univ. of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104 (e-mail: baumgarj{at}uphs.upenn.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 DISCLOSURES
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Adams RP, Dieleman LA, Cain SM. A critical value for O2 transport in the rat. J Appl Physiol Respir Environ Exercise Physiol 53: 660–664, 1982.[Abstract/Free Full Text]
  2. Armstrong BW Jr, MacIntyre NR. Pressure-controlled, inverse ratio ventilation that avoids air trapping in the adult respiratory distress syndrome[see comment]. Crit Care Med 23: 279–285, 1995.[CrossRef][ISI][Medline]
  3. Bancalari E. Inadvertent positive end-expiratory pressure during mechanical ventilation. J Pediatr 108: 567–569, 1986.[CrossRef][ISI][Medline]
  4. Bates JH, Brown KA, Kochi T. Respiratory mechanics in the normal dog determined by expiratory flow interruption. J Appl Physiol 67: 2276–2285, 1989.[Abstract/Free Full Text]
  5. Baumgardner JE, Markstaller K, Pfeiffer B, Doebrich M, Otto CM. Effects of respiratory rate, plateau pressure, and positive end-expiratory pressure on PaO2 oscillations after saline lavage. Am J Respir Crit Care Med 166: 1556–1562, 2002.[Abstract/Free Full Text]
  6. Brochard L. Intrinsic (or auto-) PEEP during controlled mechanical ventilation. Intensive Care Med 28: 1376–1378, 2002.[CrossRef][ISI][Medline]
  7. Brower RG, Lanken PN, MacIntyre N, Matthay MA, Morris A, Ancukiewicz M, Schoenfeld D, Thompson BT, and National Heart, Lung and Blood Institute ARDS Clinical Trials Network. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med 351: 327–336, 2004.[Abstract/Free Full Text]
  8. Cain S. Oxygen delivery and uptake in dogs during anemic and hypoxic hypoxia. J Appl Physiol 42: 228–234, 1977.[Abstract/Free Full Text]
  9. Cartwright DW, Willis MM, Gregory GA. Functional residual capacity and lung mechanics at different levels of mechanical ventilation. Crit Care Med 12: 422–427, 1984.[ISI][Medline]
  10. D'Angelo E, Prandi E, Tavola M, Robatto FM. Assessment of respiratory system viscoelasticity in spontaneously breathing rabbits. Respir Physiol 114: 257–267, 1998.[CrossRef][ISI][Medline]
  11. De Durante G, del Turco M, Rustichini L, Cosimini P, Giunta F, Hudson LD, Slutsky AS, Ranieri VM. ARDSNet lower tidal volume ventilatory strategy may generate intrinsic positive end-expiratory pressure in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 165: 1271–1274, 2002.[Abstract/Free Full Text]
  12. Downie JM, Nam AJ, Simon BA. Pressure-volume curve does not predict steady-state lung volume in canine lavage lung injury. Am J Respir Crit Care Med 169: 957–962, 2004.[Abstract/Free Full Text]
  13. Draper NR, Smith H.Applied Regression Analysis. New York: Wiley, 1981.
  14. Dreyfuss D, Saumon G. Ventilator-induced lung injury—lessons from experimental studies. Am J Respir Crit Care Med 157: 294–323, 1998.
  15. Frerichs I, Dargaville PA, Dudykevych T, Rimensberger PC. Electrical impedance tomography: a method for monitoring regional lung aeration and tidal volume distribution? Intensive Care Med 29: 2312–2316, 2003.[CrossRef][ISI][Medline]
  16. Fujino Y, Nishimura M, Uchiyama A, Taenaka N, Yoshiya I. Dynamic measurement of intrinsic PEEP does not represent the lowest intrinsic PEEP [see comment]. Intensive Care Med 25: 274–278, 1999.[CrossRef][ISI][Medline]
  17. Gattinoni L, Caironi P, Cressoni M, Chiumello D, Ranieri VM, Quintel M, Russo S, Patroniti N, Cornejo R, Bugedo G. Lung recruitment in patients with the acute respiratory distress syndrome [see comment]. N Engl J Med 354: 1775–1786, 2006.[Abstract/Free Full Text]
  18. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med 151: 1807–1814, 1995.[Abstract]
  19. Gaver DP III, Samsel RW, Solway J. Effects of surface tension and viscosity on airway reopening. J Appl Physiol 69: 74–85, 1990.[Abstract/Free Full Text]
  20. Ghadiali SN, Gaver DP III. An investigation of pulmonary surfactant physicochemical behavior under airway reopening conditions. J Appl Physiol 88: 493–506, 2000.[Abstract/Free Full Text]
  21. Goedje O, Hoeke K, Goetz AE, Felbinger TW, Reuter DA, Reichart B, Reinhard F, Hannekum A, Pfeiffer UJ. Reliability of a new algorithm for continuous cardiac output determination by pulse-contour analysis during hemodynamic instability. Crit Care Med 30: 52–58, 2002.[CrossRef][ISI][Medline]
  22. Grasso S, Fanelli V, Cafarelli A, Anaclerio R, Amabile M, Ancona G, Fiore T. Effects of high versus low positive end-expiratory pressures in acute respiratory distress syndrome. Am J Respir Crit Care Med 171: 1002–1008, 2005.[Abstract/Free Full Text]
  23. Halpern D, Bull JL, Grotberg JB. The effect of airway wall motion on surfactant delivery. J Biomech Eng 126: 410–419, 2004.[CrossRef][ISI][Medline]
  24. Halter JM, Steinberg JM, Schiller HJ, DaSilva M, Gatto LA, Landas S, Nieman GF. Positive end-expiratory pressure after a recruitment maneuver prevents both alveolar collapse and recruitment/derecruitment. Am J Respir Crit Care Med 167: 1620–1626, 2003.[Abstract/Free Full Text]
  25. Hardman JG, Aitkenhead AR. Estimating alveolar dead space from the arterial to end-tidal CO(2) gradient: a modeling analysis. Anesth Analg 97: 1846–1851, 2003.[Abstract/Free Full Text]
  26. Hernandez P, Navalesi P, Maltais F, Gursahaney A, Gottfried SB. Comparison of static and dynamic measurements of intrinsic PEEP in anesthetized cats. J Appl Physiol 76: 2437–2442, 1994.[Abstract/Free Full Text]
  27. Hickling KG. Best compliance during a decremental, but not incremental, positive end-expiratory pressure trial is related to open-lung positive end-expiratory pressure: a mathematical model of acute respiratory distress syndrome lungs. Am J Respir Crit Care Med 163: 69–78, 2001.[Abstract/Free Full Text]
  28. Hough CL, Kallet RH, Ranieri VM, Rubenfeld GD, Luce JM, Hudson LD. Intrinsic positive end-expiratory pressure in Acute Respiratory Distress Syndrome (ARDS) Network subjects. Crit Care Med 33: 527–532, 2005.[CrossRef][ISI][Medline]
  29. Jonson B. Elastic pressure-volume curves in acute lung injury and acute respiratory distress syndrome. Intensive Care Med 31: 205–212, 2005.[CrossRef][ISI][Medline]
  30. Jonson B, Beydon L, Brauer K, Mansson C, Valind S, Grytzell H. Mechanics of respiratory system in healthy anesthetized humans with emphasis on viscoelastic properties. J Appl Physiol 75: 132–140, 1993.[Abstract/Free Full Text]
  31. Kacmarek RM, Kirmse M, Nishimura M, Mang H, Kimball WR. The effects of applied vs auto-PEEP on local lung unit pressure and volume in a four-unit lung model. Chest 108: 1073–1079, 1995.
  32. Kunst PW, Vazquez de Anda G, Bohm SH, Faes TJ, Lachmann B, Postmus PE, de Vries PM. Monitoring of recruitment and derecruitment by electrical impedance tomography in a model of acute lung injury. Crit Care Med 28: 3891–3895, 2000.[CrossRef][ISI][Medline]
  33. Lachmann B. Open the lung and keep it open. Intensive Care Med 18: 319–321, 1992.[CrossRef][ISI][Medline]
  34. Lachmann B, Robertson B, Vogel J. In vivo lung lavage as an experimental model of the respiratory distress syndrome. Acta Anaesthesiol Scand 24: 231–236, 1980.[ISI][Medline]
  35. Lessard MR, Guerot E, Lorino H, Lemaire F, Brochard L. Effects of pressure-controlled with different I:E ratios versus volume-controlled ventilation on respiratory mechanics, gas exchange, and hemodynamics in patients with adult respiratory distress syndrome [see comment]. Anesthesiology 80: 983–991, 1994.[ISI][Medline]
  36. Lieberman JA, Weiskopf RB, Kelley SD, Feiner J, Noorani M, Leung J, Toy P, Viele M. Critical oxygen delivery in conscious humans is less than 7.3 ml O2-kg–1-min–1. Anesthesiology 92: 407–413, 2000.[CrossRef][ISI][Medline]
  37. Ludwigs U, Philip A, Robertson B, Hedenstierna G. Pulmonary epithelial permeability. An animal study of inverse ratio ventilation and conventional mechanical ventilation [see comment]. Chest 110: 486–493, 1996.
  38. Malbouisson LM, Muller JC, Constantin JM, Lu Q, Puybasset L, Rouby JJ, Group CTSAS. Computed tomography assessment of positive end-expiratory pressure-induced alveolar recruitment in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 163: 1444–1450, 2001.[Abstract/Free Full Text]
  39. Markstaller K, Eberle B, Kauczor HU, Scholz A, Bink A, Thelen M, Heinrichs W, Weiler N. Temporal dynamics of lung aeration determined by dynamic CT in a porcine model of ARDS. Br J Anaesth 87: 459–468, 2001.[Abstract/Free Full Text]
  40. Markstaller K, Kauczor HU, Weiler N, Karmrodt J, Doebrich M, Ferrante M, Thelen M, Eberle B. Lung density distribution in dynamic CT correlates with oxygenation in ventilated pigs with lavage ARDS. Br J Anaesth 91: 699–708, 2003.[Abstract/Free Full Text]
  41. Markstaller K, Otto CM, Karmrodt J, Pfeiffer B, Syring R, Baumgardner JE. Regional and temporal distribution of iNOS activity in a model of acute lung injury (Abstract). Shock 23 (Suppl3): 84, 2005.
  42. Mercat A, Diehl JL, Michard F, Anguel N, Teboul JL, Labrousse J, Richard C. Extending inspiratory time in acute respiratory distress syndrome. Crit Care Med 29: 40–44, 2001.[CrossRef][ISI][Medline]
  43. Mercat A, Titiriga M, Anguel N, Richard C, Teboul JL. Inverse ratio ventilation (I/E = 2/1) in acute respiratory distress syndrome: a six-hour controlled study. Am J Respir Crit Care Med 155: 1637–1642, 1997.[Abstract]
  44. Mishima M, Balassy Z, Bates JH. Temporal response of lung impedance after iv oleic acid in dogs. Respir Physiol 103: 177–185, 1996.[CrossRef][ISI][Medline]
  45. Muscedere JG, Mullen JB, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 149: 1327–1334, 1994.[Abstract]
  46. Neumann P, Berglund JE, Andersson LG, Maripu E, Magnusson A, Hedenstierna G. Effects of inverse ratio ventilation and positive end-expiratory pressure in oleic acid-induced lung injury. Am J Respir Crit Care Med 161: 1537–1545, 2000.[Abstract/Free Full Text]
  47. Neumann P, Berglund JE, Mondejar EF, Magnusson A, Hedenstierna G. Dynamics of lung collapse and recruitment during prolonged breathing in porcine lung injury. J Appl Physiol 85: 1533–1543, 1998.[Abstract/Free Full Text]
  48. Otis DR Jr, Johnson M, Pedley TJ, and Kamm RD. Role of pulmonary surfactant in airway closure: a computational study. J Appl Physiol 75: 1323–1333, 1993.[Abstract/Free Full Text]
  49. Pelosi P, Goldner M, McKibben A, Adams A, Eccher G, Caironi P, Losappio S, Gattinoni L, Marini JJ. Recruitment and derecruitment during acute respiratory failure: an experimental study. Am J Respir Crit Care Med 164: 122–130, 2001.[Abstract/Free Full Text]
  50. Pepe PE, Marini JJ. Occult positive end-expiratory pressure in mechanically ventilated patients with airflow obstruction: the auto-PEEP effect [see comment]. Am Rev Respir Dis 126: 166–170, 1982.[ISI][Medline]
  51. Pfeiffer B, Hachenberg T, Wendt M, Marshall B. Mechanical ventilation with permissive hypercapnia increases intrapulmonary shunt in septic and nonseptic patients with acute respiratory distress syndrome. Crit Care Med 30: 285–289, 2002.[CrossRef][ISI][Medline]
  52. Pfeiffer B, Syring RS, Markstaller K, Otto CM, Baumgardner JE. Implications of arterial PO2 oscillations for conventional arterial blood gas analysis. Anesth Analg 102: 1758–1764, 2006.[Abstract/Free Full Text]
  53. Pinsky MR. The hemodynamic consequences of mechanical ventilation: an evolving story. Intensive Care Med 23: 493–503, 1997.[CrossRef][ISI][Medline]
  54. Pinsky MR. Probing the limits of arterial pulse contour analysis to predict preload responsiveness (editorial). Anesth Analg 96: 1245–1247, 2003.[Free Full Text]
  55. Prys-Roberts C, Kelman GR, Greenbaum R, Kain ML, Bay J. Hemodynamics and alveolar-arterial PO2 differences at varying PaCO2 in anesthetized man. J Appl Physiol 25: 80–87, 1968.[Free Full Text]
  56. Rimensberger PC, Cox PN, Frndova H, Bryan C. The open lung during small tidal volume ventilation: concepts of recruitment and "optimal" positive end-expiratory pressure. Crit Care Med 27: 1946–1952, 1999.[CrossRef][ISI][Medline]
  57. Rimensberger PC, Pache JC, McKerlie C, Frndova H, Cox PN. Lung recruitment and lung volume maintenance: a strategy for improving oxygenation and preventing lung injury during both conventional mechanical ventilation and high-frequency oscillation. Intensive Care Med 26: 745–755, 2000.[CrossRef][ISI][Medline]
  58. Rossi A, Gottfried SB, Higgs BD, Zocchi L, Grassino A, and Milic-Emili J. Respiratory mechanics in mechanically ventilated patients with respiratory failure. J Appl Physiol 58: 1849–1858, 1985.[Abstract/Free Full Text]
  59. Rotta AT, Gunnarsson B, Fuhrman BP, Hernan LJ, Steinhorn DM. Comparison of lung protective ventilation strategies in a rabbit model of acute lung injury. Crit Care Med 29: 2176–2184, 2001.[CrossRef][ISI][Medline]
  60. Schou H, Perez de Sa V, Sigurdardottir M, Roscher R, Jonmarker C, Werner O. Circulatory effects of hypoxia, acute normovolemic hemodilution, and their combination in anesthetized pigs. Anesthesiology 84: 1443–1454, 1996.[CrossRef][ISI][Medline]
  61. Schumacker P, Cain S. The concept of a critical oxygen delivery. Intensive Care Med 13: 223–229, 1987.[CrossRef][ISI][Medline]
  62. Similowski T, Levy P, Corbeil C, Albala M, Pariente R, Derenne JP, Bates JH, Jonson B, and Milic-Emili J. Viscoelastic behavior of lung and chest wall in dogs determined by flow interruption. J Appl Physiol 67: 2219–2229, 1989.[Abstract/Free Full Text]
  63. Slutsky AS. Lung injury caused by mechanical ventilation. Chest 116: 9S–15S, 1999.
  64. Stamenovic D, Glass GM, Barnas GM, Fredberg JJ. Viscoplasticity of respiratory tissues. J Appl Physiol 69: 973–988, 1990.[Abstract/Free Full Text]
  65. Steinberg JM, Schiller HJ, Halter JM, Gatto LA, Lee HM, Pavone LA, Nieman GF. Alveolar instability causes early ventilator-induced lung injury independent of neutrophils. Am J Respir Crit Care Med 169: 57–63, 2004.[Abstract/Free Full Text]
  66. Syring R, Otto CM, Campbell V, Baumgardner JE. Respiratory variations in SpO2 with a fast pulse oximeter (Abstract). Anesthesiology: A–375, 2003.
  67. Thorens JB, Jolliet P, Ritz M, Chevrolet JC. Effects of rapid permissive hypercapnia on hemodynamics, gas exchange, and oxygen transport and consumption during mechanical ventilation for the acute respiratory distress syndrome [see comment]. Intensive Care Med 22: 182–191, 1996.[CrossRef][ISI]