Abstract
Ventilatory modes employing different inspiratory flow patterns and inspiratory to expiratory ratios may alter lung strain in acute lung injury patients. To determine whether variations in lung strain existed between pressure-controlled, volume-controlled, and pressure-controlled inverse ratio modes of ventilation, we randomly applied each for 30 minutes in 18 acute lung injury patients, keeping tidal volume, respiratory rate, fractional inspired oxygen, and total positive end-expiratory pressure constant. After each mode, a multiple linear regression analysis of dynamic airway pressure and airflow was performed with a volume-dependent single compartment model of the equation of motion, and an index of nonlinear elastic behavior was calculated. In five additional patients, concurrent dynamic computerized axial tomography scanning at juxtadiaphragmatic and subcarinal levels was added. Although static mechanics, oxygenation, and hemodynamics were no different between pressure-controlled, volume-controlled, and pressure-controlled inverse ratio ventilation, we found significant differences in nonlinear behavior. This was least with pressure-controlled followed by volume-controlled ventilation, and pressure-controlled inverse ratio ventilation had the greatest nonlinear elastic behavior. Dynamic computerized axial tomography analysis revealed more overinflated units in the left subcarinal slice with pressure-controlled inverse ratio ventilation. Ventilator flow pattern and inspiratory to expiratory ratio independently influence lung strain in acute lung injury; however, further studies are needed to determine the biologic significance.
Acute lung injury (ALI) results in an increase in the work of breathing and impairment of gas exchange that usually requires mechanical ventilation. However, the pathophysiology of ALI includes surfactant dysfunction, increased lung weight, and airspace collapse and consolidation, which predispose the lung to ventilation-induced lung injury (1). For example, mortality in patients with ALI is reduced by using small tidal volume (Vt) ventilation (2), and protective ventilatory strategies using positive end-expiratory pressure (PEEP), recruitment maneuvers, surfactant replacement therapy, and varying Vt may also reduce ventilation-induced lung injury (1). In general, these strategies aim to reduce ventilation-induced lung injury by reducing sheer stress or lung strain (3).
The quasistatic volume–pressure curve and the static distending pressure of the respiratory system (plateau pressure [Pplat]) are the most commonly used measures of lung strain. However, the dynamic elastic distending pressure, which is composed of the static pressure, and the viscoelastic pressure will better reflect the volume–pressure relationships and end-inspiratory pressure strain that are applied to the distal airspaces during mechanical ventilation (4, 5). The viscoelastic pressure is, in turn, composed of tissue viscance and the effects of time constant inequalities within the lung (6), which can lead to regional flow maldistribution.
When Vt, PEEP, and inspiratory to expiratory ratio are constant compared with volume-controlled ventilation, pressure-controlled ventilation does not improve oxygenation or lower Pplat (7–9). In addition, when Vt and total PEEP (PEEPtot) are constant, the inverse ratio ventilation at an inspiratory to expiratory ratio of 2:1 compared with 1:2 does not improve oxygenation or Pplat (7–10). However, the different inspiratory flow patterns and different inspiratory to expiratory ratios may alter the viscoelastic component of lung strain by altering the buildup of tissue resistance or the distribution of ventilation. Variation in the extent of nonlinear dynamic elastic volume–pressure behavior of the respiratory system will reflect the differences in viscoelastic strain between modes. To examine these issues, we used a volume-dependent single-compartment model to examine the extent of nonlinear dynamic volume–pressure relationships in 18 patients with ALI randomly receiving pressure-controlled ventilation at an inspiratory to expiratory ratio of 1:2 and 2:1 (pressure-controlled inverse ratio ventilation) and volume-controlled ventilation at an inspiratory to expiratory ratio of 1:2 with PEEPtot, Vt, and respiratory rate constant. To examine whether heterogeneity of regional ventilation contributed to differences in nonlinear volume–pressure behavior when these three modes of ventilation were applied, dynamic computerized axial tomography (CT) scans and dynamic volume–pressure relationships were examined in an additional five ALI patients.
The local ethics committee approval for the study was granted (#26/97 and #94/01), and written informed consent was obtained from the patient's next of kin. Inclusion criteria were ALI as defined by the American European Consensus Conference criteria (11) and mechanical ventilation. Exclusion criteria were late ALI (greater than 5 days from onset), hemodynamic instability, and anticipated intolerance of transient PEEP removal. Demographic data (sex, height, and weight), Acute Physiology and Chronic Health Evaluation II score, etiologic factors, baseline ventilator settings, as set by the attending clinician before this study, and lung injury score (12) were recorded at inclusion.
All patients were supine, sedated, paralyzed, and initially ventilated with PEEP either via a Puritan Bennett 7200ae ventilator or a Nellcor Puritan-Bennet 840 ventilator (Puritan-Bennet, Carlsbad, CA). Baseline Vt, respiratory rate, extrinsic PEEP, and fractional inspired oxygen were recorded. PEEPtot and Pplat were measured (13).
Thirty minutes of pressure-controlled ventilation with an inspiratory to expiratory ratio of 1:2, volume-controlled ventilation with an inspiratory to expiratory ratio of 1:2 using a constant inspiratory flow, and pressure-controlled inverse ratio ventilation were randomly applied in each patient. Vt, PEEPtot, fractional inspired oxygen concentration, and respiratory rate were unchanged. At the end of each 30-minute period, hemodynamic parameters, arterial and mixed venous blood gases, airflow (V̇), and airway pressure at the opening data were recorded as previously described (13). Passive expiration to relaxation volume (ΔEELV) was measured and taken to reflect lung recruitment (14). Thirty minutes of baseline ventilation were interposed before the next ventilatory mode.
Recorded breaths were averaged, and multilinear regression analysis was performed using the volume-dependent single compartment model (13, 15):
 |
These studies, using a similar protocol, were performed in the CT scanner after a frontal tomogram of the chest (Helical Aquilon CT Scanner; Toshiba, Nasu, Japan). At the end of each 30-minute period immediately before recording respiratory mechanics, dynamic slice CT scans were obtained every 100 ms for 10 seconds at subcarinal and juxtadiaphragmatic levels. Lung density between +100 and −1,000 Hounsfield units was classified (16) at both right and left end-inspiratory and end-expiratory slices. A more detailed description of the Dynamic CT Study is available in the online supplement.
Repeated-measures analysis of variance was performed to detect a difference between modes using SPSS Base 10.0 software (SPSS Inc., Chicago, IL). Post hoc comparisons were made using Fisher's protected least significant difference test (17). Probability levels of less than 0.05 were considered significant, and all data are presented as mean ± SD.
Eighteen ALI patients (67 ± 16 years), of whom seven patients (38%) had a direct etiology, were studied (see Table E1 in the online supplement). Their lung injury score at inclusion was 2.5 ± 0.5, and nine (50%) died. All patients were studied within 5 days of the onset of ALI.
Respiratory rate, Vt, fractional inspired oxygen concentration, and PEEPtot were constant between the study modes (Table 1)
PC 1:2 | PC 2:1 | VC 1:2 | |
|---|---|---|---|
| Respiratory rate, breaths/min | 13 ± 3 | 13 ± 3 | 13 ± 3 |
| VT, L | 0.59 ± 0.16 | 0.60 ± 0.16 | 0.60 ± 0.17 |
| PEEPe, cm H2O | 9.5 ± 2.5 | 8.0 ± 2.8* | 9.6 ± 2.7 |
| PEEPtot, cm H2O | 9.8 ± 2.5 | 10.0 ± 2.5 | 10.0 ± 2.7 |
| FIO2 | 0.56 ± 0.10 | 0.56 ± 0.10 | 0.56 ± 0.10 |
Figure 1. Airway pressures. The cross-hatched bars indicate pressure-controlled ventilation, and the black bars indicate volume-controlled ventilation. The white bars show pressure-controlled inverse ratio ventilation. Data are expressed as mean + SD. Peak airway pressure (Ppk) was greatest with volume-controlled ventilation, and mean airway pressure (Pmean) was greatest with pressure-controlled inverse ratio ventilation. Pplat was not different (*p < 0.001 compared with other modes).
[More] [Minimize]The coefficients of determination for the volume-dependent single compartment model were in excess of 0.993 in all modes, indicating an excellent model fit to the data (Table 2)
PC 1:2 (n = 18) | PC 2:1 (n = 18) | VC 1:2 (n = 18) | |
|---|---|---|---|
| RRS, cm H2O/L/s | 10.8 ± 2.3 | 10.7 ± 2.1 | 10.2 ± 2.2 |
| E1, cm H2O/L | 24.4 ± 10.5* | 17.8 ± 8.7 | 22.1 ± 8.7* |
| E2 | 11 ± 12*,‡ | 21 ± 22 | 16 ± 18 |
| E2VT, cm H2O/L | 5.0 ± 4.8*,§ | 10.2 ± 7.3 | 8.0 ± 6.6 |
| ERS, cm H2O/L | 29.3 ± 14.0 | 28.9 ± 13.0 | 30.1 ± 13.6 |
| %E2 | 15.6 ± 10.8*,‡ | 35.7 ± 13.2 | 25.1 ± 12.0† |
| P0, cm H2O | 10.0 ± 2.6 | 10.0 ± 2.6 | 10.1 ± 2.7 |
| CD | 0.995 ± 0.002 | 0.993 ± 0.003 | 0.995 ± 0.003 |
| ΔEELV, L | 0.58 ± 0.22† | 0.65 ± 0.25 | 0.58 ± 0.22† |
Although oxygenation was not altered by the mode of ventilation, PaCO2 was lower, and the corresponding arterial pH was greater, with pressure-controlled inverse ratio ventilation (Table 3)
PC 1:2 | PC 2:1 | VC 1:2 | |
|---|---|---|---|
| PaO2, mm Hg | 119 ± 22 | 114 ± 26 | 111 ± 24 |
| pH | 7.33 ± 0.10* | 7.34 ± 0.10 | 7.31 ± 0.10* |
| PaCO2, mm Hg | 48 ± 12* | 46 ± 13 | 49 ± 13* |
| SvO2, % | 77 ± 7 | 74 ± 7 | 76 ± 6 |
| PA, mm Hg | 80 ± 9 | 78 ± 10 | 80 ± 10 |
| PPA, mm Hg | 33 ± 8 | 34 ± 8 | 32 ± 7 |
| CO, L/min | 6.7 ± 3.5 | 6.4 ± 2.3 | 6.7 ± 2.5 |
| PRA, mm Hg | 13 ± 2 | 12 ± 2 | 13 ± 3 |
| PPAO, mm Hg | 16 ± 5 | 16 ± 4 | 16 ± 5 |
Five patients with early ALI were recruited for the dynamic CT study (see Table E2 in the online supplement). Their ventilatory parameters were similar to the initial cohort, apart from a smaller Vt, and greater respiratory rate (see Table E3 in the online supplement).
As in the initial cohort, there appeared to be the greatest nonlinear behavior with pressure-controlled inverse ratio ventilation followed by volume-controlled ventilation and then pressure-controlled ventilation. However, no statistical difference was found for %E2 or E2Vt (see Tables E3 and E4 in the online supplement). There was no difference in the ΔEELV or model fit between groups. Because viscoelastic buildup likely varies between the different modes of ventilation, we measured inspiratory and expiratory times and relevant volumes. As expected, the inspiratory time was longer for pressure-controlled inverse ratio ventilation, with a shorter expiratory time. The time taken to achieve 50% of inspired Vt was shortest with pressure-controlled ventilation and longest with volume-controlled ventilation.
Right and left juxtadiaphragmatic and subcarinal dynamic CT data were available for all five subjects, and the relative distribution of tissue and gas at end inspiration and end expiration is summarized in Figures 2 and 3
Figure 2. Distribution of CT attenuation at end inspiration and end expiration from dynamic CT scan at the subcarinal slice. The cross-hatched bars indicate pressure-controlled ventilation, and the black bars indicate volume-controlled ventilation. The white bars show pressure-controlled inverse ratio ventilation. Data are expressed as mean + SD. The distribution of Hounsfield units were classified as −100 to 100: nonaerated (non); −500 to −100: poorly aerated (poor); −900 to −500: normally aerated (normal); and −1,000 to −900: overinflated (over). There were significantly more overinflated pixels at end inspiration with pressure-controlled inverse ratio ventilation compared with both volume-controlled ventilation and pressure-controlled ventilation (*p = 0.044 and †p = 0.013, respectively).
[More] [Minimize]
Figure 3. Distribution of CT attenuation at end inspiration and end expiration from dynamic CT scan at the juxtadiaphragmatic slice. The cross-hatched bars indicate pressure-controlled ventilation, and the black bars indicate volume-controlled ventilation. The white bars show pressure-controlled inverse ratio ventilation. Data are expressed as mean + SD. The distribution of Hounsfield units were classified as −100 to 100: nonaerated (non); −500 to −100: poorly aerated (poor); −900 to −500: normally aerated (normal); and −1,000 to −900: overinflated (over). There was no difference in the distribution of lung density between the three modes of ventilation.
[More] [Minimize]We have shown that the inspiratory flow pattern and inspiratory to expiratory ratio influence nonlinear volume–pressure behavior of the lung in patients with ALI, which was greatest with pressure-controlled inverse ratio ventilation and least with pressure-controlled ventilation. Because PEEPtot and Vt were constant, these differences in nonlinear behavior are most likely due to differences in the viscoelastic behavior of the respiratory system. A greater buildup of tissue viscance with volume-controlled ventilation could explain why there was greater nonlinear behavior compared with pressure-controlled ventilation, whereas greater regional overinflation at end inspiration, as shown by dynamic CT, may explain why pressure-controlled inverse ratio ventilation had the greatest nonlinear behavior. Irrespective of the etiology of these differences, these data suggest the greatest lung strain was with pressure-controlled inverse ratio ventilation and the least was with pressure-controlled ventilation. However, further studies are needed to examine whether a lung protective inspiratory flow pattern and inspiratory to expiratory ratio result in less alveolar inflammation and improvement in outcome.
Consistent with previous studies, we found no advantage of pressure-controlled ventilation or pressure-controlled inverse ratio ventilation over volume-controlled ventilation with respect to oxygenation, hemodynamics, or Pplat and a lower PaCO2 with pressure-controlled inverse ratio ventilation (7–10, 18). Although peak airway pressure was greater with volume-controlled ventilation, this is due to earlier dissipation of flow resistance with pressure-controlled ventilation and pressure-controlled inverse ratio ventilation resulting in similar Pplat. Mean airway pressure was greatest with pressure-controlled inverse ratio ventilation; however, cardiac output and mixed venous oxygen saturation were no different, and both unchanged (7) and reduced (10, 18) cardiac output have previously been reported with pressure-controlled inverse ratio ventilation. Using inert gas analysis, Zavala and coworkers found that the reduction in PaCO2 with pressure-controlled inverse ratio ventilation was due to a reduction in dead space and a right shift of ventilation to perfusion distribution, but this was not reflected in an improvement in oxygenation (9). Because the effect on PaCO2 is small, it is unlikely to be of clinical significance.
The novel aspect of this study is the quantitation of nonlinear volume–pressure behavior as inspiratory flow pattern and inspiratory to expiratory ratio was varied. We have used a volume-dependent single compartment model to quantitate nonlinear behavior as the volume-dependent component of elastance (%E2). The volume-dependent single compartment model has been widely used in patients with acute respiratory failure, improves model fit, and is robust, as it is not influenced by mode of ventilation, endotracheal tube size, or respiratory resistance (13, 15, 19–22). We found that %E2 was greatest with pressure-controlled inverse ratio ventilation and least with pressure-controlled ventilation. Irrespective of the etiology of these differences, this suggests that the effective distending pressure and lung strain would be greatest with pressure-controlled inverse ratio ventilation and least with pressure-controlled ventilation.
Nonlinear behavior may be due to resistive, elastic, or viscoelastic properties of the respiratory system. For example, increased PEEP or Vt may result in ventilation at a greater end-inspiratory volume on the curvilinear, rather than the linear, section of the static volume–pressure curve, resulting in an increase in %E2 (13, 15, 23). In patients with ALI, many factors, including surfactant dysfunction, pulmonary edema, dysfunction of parenchymal collagen and elastin, and pulmonary fibrosis, may alter the shape of the quasistatic volume–pressure curve. However, this is an unlikely explanation for the differences that we found because the shape of the quasistatic volume–pressure curve should have remained constant for each patient, and the experimental design, a constant Vt and PEEPtot, aimed to conduct this on the same part of the volume–pressure curve. Although PEEPtot was similar between the three ventilatory groups, ΔEELV was slightly greater with pressure-controlled inverse ratio ventilation, but pressure-controlled ventilation and volume-controlled ventilation were no different. Although this could have contributed to pressure-controlled inverse ratio ventilation having the greatest %E2, it cannot explain the difference in nonlinear behavior between pressure-controlled ventilation and volume-controlled ventilation, and the end-expiratory dynamic CT data are consistent with matched PEEPtot and a similar end-expiratory lung volume for each mode of ventilation. Consequently, the differences we found are likely due to differences in the viscoelastic behavior of the respiratory system.
Flow resistance and viscoelastic behavior are the major components of the characteristically widened dynamic volume–pressure curve. Both tissue factors such as pulmonary surfactant, elastin, contraction of alveolar duct muscles and changes in pulmonary blood volume, and altered distribution of ventilation due to time constant inequalities contribute to the viscoelastic behavior of the respiratory system. The relative contribution of these components can be difficult to separate, but in the healthy lung tissue factors appear to be the major factor, and after methacholine challenge, airway inhomogeneities dominate (24).
During constant flow inspiration, there is a buildup of viscoelastic pressure that is dissipated during expiration and any end-inspiratory pause (25, 26). Consequently, the inspiratory flow profile of pressure-controlled ventilation, which consists of a decelerating flow, will allow some dissipation of the viscoelastic pressure by end inspiration, and this would likely result in less nonlinear behavior and a lower %E2. This is consistent with and a likely explanation for the finding that %E2 was lower with pressure-controlled ventilation compared with volume-controlled ventilation. However, on the same basis, it would predict an even lower %E2, instead of the greatest %E2, with pressure-controlled inverse ratio ventilation, as there is a greater time for dissipation of the viscoelastic pressure during the longer inspiratory time. To examine whether the short expiratory time with pressure-controlled inverse ratio ventilation exacerbated the time constant inequalities present in patients with ALI, possibly leading to regional overinflation, we also performed dynamic CT in a further five patients subjected to a similar experimental protocol.
Numerous studies have used chest CT to measure lung density in patients with ALI and partitioned voxels into nonaerated, poorly aerated, normally aerated, and overinflated units based on their mean density (27, 28). Many of these studies have used static CT with the lung usually held at a given volume for a minimum of a few seconds up to 15–20 seconds for whole-lung CT. Because this pause at end inspiration or end expiration would allow some dissipation of viscoelastic forces and redistribution of time constant inequalities, we used dynamic CT to capture these lung volumes without a pause. However, in an effort to minimize radiation exposure, we performed dynamic CT at only subcarinal and juxtadiaphragmatic levels. Because the lung also moves in a cephalocaudal axis during ventilation, it is not possible to compare directly our end-inspiratory and end-expiratory slices. An alternative approach would have been to perform multiple contiguous slices, and this would have allowed the same region of lung to have been followed across the respiratory cycle. However, we did not have access to a multislice CT scanner, and this would have increased radiation exposure. Consequently, we have compared the three modes of ventilation at either end inspiration or end expiration at two separate slice levels.
Using this technique, we found an increase in the number of overinflated pixels at end inspiration in the left subcarinal slice with pressure-controlled inverse ratio ventilation but no difference between pressure-controlled ventilation and volume-controlled ventilation. Concurrently, the trend to greater nonlinear volume–pressure behavior was also present with constant Vt and PEEPtot. The demographics, ventilation, and mechanics of the patients in the nonlinear elastance study are similar with the dynamic CT study; however, the latter group was ventilated with a smaller Vt (approximately 500 versus 600 ml) and slightly greater PEEP (approximately 12 versus 10 cm H2O). Consequently, it is likely that the two groups were studied on similar parts of their quasistatic volume–pressure curves.
Consistent with Gattinoni and colleagues (28) and Vieira and colleagues (16), we defined overinflation as a density between −1,000 and −900 HU, although Dambrosio and colleagues used a density between −1,000 and −800 HU (29). Irrespective of the definition used, Gattinoni and colleagues have argued that excessive stretch (alveolar wall tension) may occur in acute respiratory distress syndrome without CT evidence of an excessive gas to tissue ratio (28). From this argument, CT evidence of overinflation may fail to reflect or underestimate nonlinear behavior due to excessive alveolar wall tension.
Regional lung aeration in patients with ALI is determined by many factors including the anteroposterior and cephalocaudal position and the heart. This results in reduced aeration of the lower lobes, particularly the left lower lobe, and the tendency to overinflate the upper lobes if excessive PEEP is applied (30, 31). Based on the whole-lung reconstruction data reported by Malbouisson and coworkers (31), our subcarinal slice would likely have sampled similar proportions of upper and lower lobes. Consequently, if the shortened expiratory time found with pressure-controlled inverse ratio ventilation were to lead to redistribution of regional ventilation, of the data we collected, this would most likely be found in the left subcarinal slice. Had our CT analysis been more comprehensive, we may have detected more widespread differences in regional ventilation. However, animal models where dynamic whole-lung CT can be performed would be needed to examine this issue.
The longer inspiratory time found with pressure-controlled inverse ratio ventilation has been thought to improve distribution of gas to slow time constant airspaces. However, this is at the cost of a short expiratory time often leading to intrinsic PEEP, which may be heterogeneously distributed. Ludwigs and colleagues and Neumann and coworkers used CT scans to examine inverse ratio ventilation in pigs after oleic acid and did not find any change in the distribution of aeration (32, 33); however, they did not specifically look for evidence of overinflation. In a lung model with different time constants, Kacmarek found that inverse ratio ventilation with a short expiratory time led to greater gas trapping and greater intrinsic PEEP in the slow lung units, with the measured tracheal PEEP the sum average of PEEPtot (34). Although our end-expiratory CT data were not statistically different, end-expiratory overinflation could precede the end-inspiratory overinflation that we found with pressure-controlled inverse ratio ventilation. Consistent with the notion that intrinsic PEEP may be maldistributed, Neumann and colleagues (33) have found better matching of the ventilation to perfusion distribution and more uniform aeration on CT scan with extrinsic PEEP than with intrinsic PEEP in an animal model of lung injury.
In patients with ALI, differences in inspiratory flow pattern and inspiratory to expiratory ratio alter nonlinear behavior, despite minimal differences in gas exchange and hemodynamics. Using a similar technique, Ranieri and coworkers found that nonlinear behavior was associated with alveolar inflammation in the rat (35). Consequently, regardless of the mechanism, our data suggest that there may be reduced lung strain with pressure-controlled ventilation and greatest strain with pressure-controlled inverse ratio ventilation compared with volume-controlled ventilation. However, Esteban and coworkers (36) found no difference in outcome when 37 patients with ARDS were randomized to pressure-controlled ventilation and 42 to volume-controlled ventilation. Consequently, further studies are needed to examine the possible clinical significance of our finding.
| 1. | Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am J Respir Crit Care Med 1998;157:294–323. |
| 2. | 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;342:1301–1308. |
| 3. | Wirtz HR, Dobbs LG. The effects of mechanical forces on lung functions. Respir Physiol 2000;119:1–17. |
| 4. | Servillo G, Svantesson C, Beydon L, Roupie E, Brochard L, Lemaire F, Jonson B. Pressure-volume curves in acute respiratory failure: automated low flow inflation versus occlusion. Am J Respir Crit Care Med 1997;155:1629–1636. |
| 5. | Jonson B, Richard J-C, Straus C, Mancebo J, Lemaire F, Brochard F. Pressure-volume curves and compliance in acute lung injury: evidence for recruitment above the lower inflection point. Am J Respir Crit Care Med 1999;159:1172–1179. |
| 6. | Otis AB, McKerrow CB, Bartlett RA, Mead J, McIlroy MB, Selverstone NJ, Radford EP. Mechanical factors in distribution of ventilation. J Appl Physiol 1955;8:427–443. |
| 7. | 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. Anesthesiology 1994;80:983–991. |
| 8. | Mang H, Kacmarek RM, Ritz R, Wilson RS, Kimball WP. Cardiorespiratory effects of volume- and pressure-controlled ventilation at various I/E ratios in an acute lung injury model. Am J Respir Crit Care Med 1995;151:731–736. |
| 9. | Zavala E, Ferrer M, Polese G, Masclans JR, Planas M, Milic-Emili J, Rodriguez-Roisin R, Roca J, Rossi A. Effect of inverse I:E ratio ventilation on pulmonary gas exchange in acute respiratory distress syndrome. Anesthesiology 1998;88:35–42. |
| 10. | Mercat A, Titriga M, Anguel N, Richard C, Teboul J-L. Inverse ratio ventilation (I/E = 2/1) in acute respiratory distress syndrome: a six-hour controlled study. Am J Respir Crit Care Med 1997;155:1637–1642. |
| 11. | Bernard G, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg R. The American-European Consensus Conference on ARDS: Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;149:818–824 |
| 12. | Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis 1988;138:720–723. |
| 13. | Bersten AD. Measurement of overinflation by multiple linear regression analysis in patients with acute lung injury. Eur Respir J 1998;12:526–532. |
| 14. | Ranieri VM, Eissa NT, Corbeil C, Chasse M, Braidy J, Matar N, Milic-Emili J. Effects of positive end-expiratory pressure on alveolar recruitment and gas exchange in patients with the adult respiratory distress syndrome. Am Rev Respir Dis 1991;144:544–551. |
| 15. | Kano S, Lanteri C, Duncan AW, Sly PD. Influence of nonlinearities on estimates of respiratory mechanics using multilinear regression analysis. J Appl Physiol 1994;77:1185–1197. |
| 16. | Vieira SRR, Puybasset L, Richecoeur J, Lu Q, Cluzel P, Gusman PB, Coriat P, Rouby JJ. A lung computed tomographic assessment of positive end-expiratory pressure-induced overdistension. Am J Respir Crit Care Med 1998;158:1571–1577. |
| 17. | Snedecor GW, Cochran WC. Statistical methods, 7th ed. Ames, IA: Iowa State University Press; 1980. |
| 18. | Mercat A, Graini L, Teboul J-L, Lenique F, Richard C. Cardiorespiratory effects of pressure-controlled ventilation with and without inverse ratio in the adult respiratory distress syndrome. Chest 1993;104:871–875. |
| 19. | Lanteri CJ, Kano S, Duncan AW, Sly PD. Changes in respiratory mechanics in children undergoing cardiopulmonary bypass. Am J Respir Crit Care Med 1995;152:1893–1900. |
| 20. | Peslin R, Felicio da Silva J, Chabot F, Duvivier C. Respiratory mechanics studied by multiple linear regression in unsedated ventilated patients. Eur Respir J 1992;5:871–878. |
| 21. | Peslin R, Gallina C, Saunier C, Duvivier C. Fourier analysis versus multiple linear regression analysis to analyse pressure-flow data during artificial ventilation. Eur Respir J 1994;7:2241–2245. |
| 22. | Rousselot JM, Peslin R, Duvivier C. Evaluation of the multiple linear regression method to monitor respiratory mechanics in ventilated neonates and young children. Pediatr Pulmonol 1992;13:161–168. |
| 23. | Wagers S, Lundblad L, Moriya HT, Bates JHT, Irvin CG. Nonlinearity of respiratory mechanics during bronchoconstriction in mice with airway inflammation. J Appl Physiol 2002;92:1802–1807. |
| 24. | Lutchen KR, Hantos Z, Petak F, Adamicza A, Suki B. Airway inhomogeneities contribute to apparent lung tissue mechanics during constriction. J Appl Physiol 1996;80:1841–1849. |
| 25. | 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 1993;75:132–140. |
| 26. | Beydon L, Svantesson C, Brauer K, Lemaire F, Jonson B. Respiratory mechanics in patients ventilated for critical lung disease. Eur Respir J 1996;9:262–273. |
| 27. | Rouby JJ, Lu Q, Goldstein I. Selecting the right level of positive end-expiratory pressure in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;165:1182–1186. |
| 28. | Gattinoni L, Caironi P, Pelosi P, Goodman LR. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med 2001;164:1701–1711. |
| 29. | Dambrosio M, Roupei E, Mollet JJ, Anaglde MC, Vasile N, Lemaire F, Brochard L. Effects of positive end-expiratory pressure and different tidal volumes on alveolar recruitment and hyperinflation. Anesthesiology 1997;87:495–503. |
| 30. | Vieira SRR, Puybasset L, Lu Q. Richecoeur, Cluzel P, Coriat P, Rouby J-J. A scanographic assessment of pulmonary morphology in acute lung injury: significance of the lower inflection point detected on the lung pressure-volume curve. Am J Respir Crit Care Med 1999;159:1612–1623. |
| 31. | Malbouisson LM, Busch CJ, Puybasset L, Lu Q, Cluzel P, Rouby JJ. A role of the heart in the loss of aeration characterizing lower lobes in acute respiratory distress syndrome: the CT scan ARDS group. Am J Repir Crit Care Med 2000;161:2005–2012. |
| 32. | Ludwigs U, Klingstedt C, Baehrentz S, Wegenius G, Hedenstiena G. Volume-controlled inverse ratio ventilation in oleic acid induced lung injury: effects on gas exchange, hemodynamics, and computed tomographic density. Chest 1995;108:804–809. |
| 33. | 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 2000;161:1537–1545. |
| 34. | Kacmarek R, 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 1995;108:1073–1079. |
| 35. | Ranieri VM, Zhang H, Mascia L, Aubin M, Lin CY, Mullen J, Grasso S, Binnie M, Volgyesi G, Eng P, et al. Pressure-time curve predicts minimally injurious ventilatory strategy in an isolated rat lung model. Anesthesiology 2000;93:1320–1328. |
| 36. | Esteban A, Inmaculada A, Federico G, Raul DP, Gumersindo G, Blanco J. Prospective randomized trial comparing pressure-controlled ventilation and volume-controlled ventilation in ARDS: for the Spanish Lung Failure Collaborative Group. Chest 2000;117:1690–1696. |
