JOURNAL HOME CME HOME THIS MONTH PAST ISSUES ETOC COLLECTIONS AUTHORS REVIEWERS EDITORIAL BOARD FEEDBACK RSS HELP
		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Pfeiffer, B. Articles by Baumgardner, J. E. Search for Related Content PubMed PubMed Citation Articles by Pfeiffer, B. Articles by Baumgardner, J. E. Related Collections Blood Equipment Technology

---

TECHNOLOGY, COMPUTING, AND SIMULATION

The Implications of Arterial Po₂ Oscillations for Conventional Arterial Blood Gas Analysis

Department of Anesthesia, Section of Critical Care, Department of Clinical Studies-Philadelphia, School of Veterinary Medicine, Center for Sleep and Respiratory Neurobiology, University of Pennsylvania, Philadelphia, Pennsylvania; Department of Anesthesiology, Johannes Gutenberg University, Mainz, Germany; SpectruMedix LLC, State College, Pennsylvania; Department of Anesthesiology and Intensive Care Medicine, Otto-von-Guericke-University, Magdeburg, Germany

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
In a surfactant-depletion model of lung injury, tidal recruitment of atelectasis and changes in shunt fraction lead to large Pao₂ oscillations. We investigated the effect of these oscillations on conventional arterial blood gas (ABG) results using different sampling techniques in ventilated rabbits. In each rabbit, 5 different ventilator settings were studied, 2 before saline lavage injury and 3 after lavage injury. Ventilator settings were altered according to 5 different goals for the amplitude and mean value of brachiocephalic Pao₂ oscillations, as guided by a fast responding intraarterial probe. ABG collection was timed to obtain the sample at the peak or trough of the Pao₂ oscillations, or over several respiratory cycles. Before lung injury, oscillations were small and sample timing did not influence Pao₂. After saline lavage, when Po₂ fluctuations measured by the indwelling arterial Po₂ probe confirmed tidal recruitment, Pao₂ by ABG was significantly higher at peak (295 ± 130 mm Hg) compared with trough (74 ± 15 mm Hg) or mean (125 ± 75 mm Hg). In early, mild lung injury after saline lavage, Pao₂ can vary markedly during the respiratory cycle. When atelectasis is recruited with each breath, interpretation of changes in shunt fraction, based on conventional ABG analysis, should account for potentially large respiratory variations in arterial Po₂.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Tidal recruitment of atelectasis refers to the re-expansion of atelectasis with each inspiration and collapse of the same lung regions with each expiration during positive pressure mechanical ventilation (1). Tidal recruitment has long been thought to play a role in ventilator-associated lung injury (VALI) (2,3), but the prevalence of tidal recruitment in mechanically ventilated patients is currently unknown. Only recently have techniques become available that have adequate temporal resolution to assess tidal recruitment in mechanically ventilated patients or in laboratory models of lung injury.

In experimental lung injury, particularly in surfactant depletion lung injury models, several techniques have recently demonstrated that alveolar recruitment and collapse occurs with each ventilatory cycle. In saline-lavaged pigs, Markstaller et al. (4) used dynamic computed tomography to demonstrate cyclical changes in atelectasis during uninterrupted mechanical ventilation. Neumann et al. (5) used dynamic computed tomography in 3 models of lung injury (oleic acid, saline lavage, and endotoxin infusion) and concluded that lung recruitment and collapse occur mainly within the first 4 s of sustained inspiration and sustained expiration. They also demonstrated that the dynamics of recruitment and collapse are not changed by different levels of airway pressure (6). Halter et al. (7) used subpleural vital microscopy to directly visualize tidal collapse and recruitment of individual alveoli during uninterrupted ventilation, after surfactant deactivation. In saline-lavaged pigs, rapid increases in regional impedance, indicating increases in regional gas volume, have been demonstrated by electrical impedance tomography after a step increase in airway pressure (8,9). Rapid recruitment of atelectatic lung regions after increases in airway pressure has also been confirmed in some mechanically ventilated patients in a comparison of electrical impedance tomography and dynamic computed tomography (10).

One of the pathophysiologic consequences of tidal recruitment is that shunt fraction changes throughout the respiratory cycle. These respiratory changes in shunt fraction should, in theory, result in changes in pulmonary venous Po₂ throughout the respiratory cycle. If these pulmonary venous Po₂ oscillations are preserved as blood transits through the left heart to the systemic circulation, tidal recruitment should result in very large oscillations in systemic arterial Po₂ (Pao₂). Williams et al. (11) reported modest respiratory changes in Pao₂ in saline-lavaged dogs, but the magnitude of Pao₂ oscillations they could measure was limited by the time response of their Po₂ probe. Baumgardner et al. (12) demonstrated, in surfactant depleted rabbits, huge changes in Pao₂ (maximal amplitude of 439 mm Hg) throughout the respiratory cycle during uninterrupted mechanical ventilation, with maximal Pao₂ coinciding with end-inspiration and minimal Pao₂ coinciding with end-expiration. In the online supplement to that report, the authors demonstrated with both experimental data and theoretical calculations that neither cyclical changes in cardiac output nor alveolar changes in Po₂ can give rise to such large Pao₂ oscillations. In contrast, very modest changes in shunt fraction within the breath cycle can give rise to very large Pao₂ oscillations.

These large Pao₂ oscillations could markedly affect results obtained from conventional arterial blood gases (ABG) in laboratory studies and potentially in patients with early acute lung injury, but the effect of these large Pao₂ oscillations on conventional ABG has not been reported. We undertook the current study to assess the impact of tidal recruitment on conventional, discontinuous ABG analysis of discrete samples. We used saline lavage to create a mild surfactant depletion lung injury in adult rabbits, and we used an experimental, continuous arterial Po₂ probe as a tool to adjust and monitor the amount of tidal recruitment. The aims of our study were to 1) assess the effect of tidal recruitment and its associated Pao₂ oscillations on the maximal variability of conventional, intermittent ABG analysis, when the conventional ABG samples are withdrawn at different times within a breath cycle; and 2) assess the ability of conventional, intermittent ABG sampling to measure tidal recruitment when the blood samples are intentionally timed to coincide with end-inspiration and end-expiration. We hypothesized that Pao₂ results from conventional blood gas analysis would vary according to the timing of sample collection and the magnitude of Pao₂ oscillations in this surfactant depletion model of mild acute lung injury.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
The study protocol was approved by the Institutional Animal Care and Use Committee at the University of Pennsylvania. Details of the animal preparation are identical to a prior report (12), with 2 exceptions: 1) general anesthesia in the current study was maintained by thiopental and fentanyl; 2) no pulmonary artery catheter was placed for the current study. Briefly, 6 adult female New Zealand White rabbits, 3-5 kg, were sedated with 140 mg ketamine and 20 mg xylazine by IM injection and positioned in dorsal recumbency for the remainder of the experiment. General anesthesia was maintained with an IV infusion of thiopental (30-70 mg · kg^–1 · h^–1) and intermittent IV boluses of fentanyl (25 µg every 30 to 60 min). After tracheostomy, the rabbits' lungs were mechanically ventilated (Servo 900C; Siemens, Munich, Germany) and neuromuscular blockade was provided by pancuronium (1.5 mg/h IV). A 22-gauge catheter was inserted in the femoral artery by surgical cut-down for continuous arterial blood pressure monitoring and ABG sampling. A 20-gauge catheter was inserted by cut-down into the right carotid artery for placement of the Pao₂ probe. Surfactant depletion lung injury was induced by instilling 100 mL of a warm balanced electrolyte solution (Normosol-R; Abbott Laboratories, North Chicago, IL) into the lungs via the endotracheal tube followed by immediate drainage by gravity. This lavage was repeated 3 times in all animals. A balanced electrolyte solution (Normosol-R) was administered IV at 20 mL/h throughout the study. Boluses of 10-20 mL colloid (6% Hetastarch, Abbott Laboratories), to a maximum cumulative dose of 100 mL, were administered when systolic blood pressure decreased below 60 mm Hg and respiratory variation in arterial blood pressure suggested hypovolemia. Hypotension unresponsive to colloid boluses was treated with titrated continuous infusion of epinephrine in doses ranging from zero (2 rabbits) to 1.04 µg · kg^–1 · min^–1 (one rabbit). At the end of the experiment, animals were euthanized with IV potassium chloride while under general anesthesia.

A fiberoptic, fluorescence-quenching oxygen probe (FOXY-AL300; Ocean Optics, Dunedin, FL) was inserted through the carotid artery catheter and advanced into the brachiocephalic artery according to previously described techniques (12). The Pao₂ probe was calibrated in vivo, before lavage, against the partial pressure of oxygen obtained from conventional blood gas analysis (Stat-9; NOVA Biomedical, Waltham, MA) at two different inspired oxygen concentrations (0.21 and 1.0). Data acquisition software (OOISensors; Ocean Optics Inc., Dunedin, FL) displayed the arterial Po₂ in real time at a digital sampling rate of 3.1 Hz.

For each rabbit, we adjusted the ventilator settings as needed to achieve the closest approximation possible to 5 targeted goals for the Pao₂. Pressure-controlled ventilation at a constant respiratory rate of 10 breaths/min was used throughout the study. For the targeted Pao₂ goals, we first adjusted positive end-expiratory pressure (PEEP) and plateau pressure and then adjusted I:E ratio until the Pao₂, as monitored by the brachiocephalic Po₂ probe, was as close as possible to the selected target. The 5 targeted Pao₂ goals were as follows:

1) Normal lung, high mean Pao₂—for this group, the goal was to make the largest Pao₂ oscillations possible at a high mean Pao₂;

2) Normal lung, low mean Pao₂—for this group, the goal was to make the largest Pao₂ oscillations possible at a low mean Pao₂;

3) Tidal recruitment—the goal for this group was to generate the largest Pao₂ oscillations possible, after lavage;

4) Recruited lung—the goal for this group was to generate the smallest Pao₂ oscillations possible at a high mean Pao₂ after lavage;

5) Fixed atelectasis—the goal for this group was to make the smallest Pao₂ oscillations possible at a low mean Pao₂ after lavage.

Blood gas samples were taken at each of these 5 individualized ventilator settings for each rabbit. The 2 goals before lavage and the 3 goals after lavage were chosen in random order.

Three different sampling techniques were used to obtain ABG samples for each of the 5 ventilator settings. The continuous arterial Po₂ probe was used to time the sampling. One ABG was drawn exclusively during the peak of Po₂ oscillation to obtain the highest possible Pao₂. Another ABG was drawn exclusively during trough of Po₂ oscillations to obtain the lowest possible Pao₂. To draw sufficient blood for analysis in the correct phase, sample collection was coordinated to aspirate only during the appropriate phase of 3 sequential respiratory cycles (Fig. 1). A third sample was drawn slowly and continuously over a minimum of 3 respiratory cycles to determine the mean Pao₂. The 3 different ABG were obtained in random order. Blood was drawn into heparinized syringes and analyzed immediately by a blood gas analyzer (Stat-9; NOVA Biomedical). For each ventilator setting, hemodynamic and respiratory data were collected. The total time for each ventilator setting (allowing for Pao₂ oscillations to reach a new steady state, recording the hemodynamic and respiratory data, and withdrawing and analyzing the ABG samples) was approximately 20 min.

View larger version (29K): [in this window] [in a new window]   Figure 1. Pao2 oscillations as recorded by the indwelling Po2 probe, after saline lavage, with illustration of the three strategies for obtaining arterial blood gas samples. In the peak sample, blood was withdrawn as close as possible to the time of maximal Pao2, for 3 successive peaks. Similarly, in the trough sample, blood was withdrawn as close as possible to the time of minimum Pao2. In the average sample, blood was withdrawn slowly and steadily over at least 3 respiratory cycles.

The Pao₂ values from conventional arterial blood gas samples were empirically transformed to stabilize the variance over the data set (13) with the power function transform y = (x – 55)^0.3333 + 0.00001(x + 55)². Transforms are commonly used for data sets, such as this one, in which the data cover a broad range of values, in this case from arterial Po₂ as small as 58 Torr to a Pao₂ of 517 Torr. Often when the data cover such a broad range, the variance in the larger ranges of data is larger than the variance in smaller ranges of data. Transformation of the data equalizes the variance across the data range, thereby satisfying a requirement for use of analysis of variance (13). The transformed data were tested for equal variance by the Levene Median test, and then the data were analyzed by two-way repeated-measures analysis of variance (Sigmastat 3.1, SPSS, Chicago IL), with targeted Pao₂ goal group as one factor and blood gas sampling method as the second factor. The residuals were inspected visually and tested for normality by the Kolmogorov-Smirnov test. Pairwise comparisons among the 3 levels of the sampling method factor (peak, average, and trough), and the 5 levels of the targeted Pao₂ goal factor (the column headings in Table 1) were performed post hoc by the Tukey test. The relationship between the amplitudes of Pao₂ oscillations (peak minus trough values), as determined by timed blood gas collection versus the indwelling Pao₂ probe, was tested by the Pearson product moment correlation (Sigmastat 3.1). The best-fit linear relationship between the amplitudes of Pao₂ oscillations was determined by linear regression.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Hemodynamic data, shown in Table 1, are reported as mean ± sd. Table 1 also presents the ventilator settings that resulted from adjustments to achieve the targeted Pao₂ goals, as well as the resulting Pao₂ data from the indwelling arterial Po₂ probe. The largest oscillations that could be created in the groups with normal lungs, before lavage, were, on average, only 54 and 15 Torr for Groups 1 and 2, respectively. In Group 2, a low mean Pao₂ could not be achieved without reducing Fio₂ to 0.21. In contrast, after lavage the maximum Pao₂ oscillations were on average 283 Torr. After lavage, even when the ventilator was adjusted aiming for minimal oscillations, the Pao₂ oscillation amplitudes were 75 and 97 Torr in groups 4 and 5, respectively. Table 1 also lists pH and Pco₂ data from conventional ABG analysis.

The Pao₂ data from the blood gas sampling are plotted in Figure 2 as mean ± 2 sem. After the variance stabilization transformation, the data set satisfied the test for equal variance at the P = 0.05 level. Two-way repeated-measures analysis of variance showed a highly significant (P < 0.001) effect for both factors, as well as factor interaction, with F values for sample method, targeted Pao₂ goal, and interaction of 101, 65, and 22, respectively. Residuals appeared to be normally distributed both visually and by the Kolmogorov-Smirnov test at the P = 0.05 level. Significant (P < 0.05) pair-wise differences between sample methods within each of the 5 targeted Pao₂ groups are marked in Figure 2.

View larger version (42K): [in this window] [in a new window]   Figure 2. Arterial Po2 results from conventional blood gas analysis for the two targeted goals for normal lungs (nml, hi and nml, lo) and the three targeted goals for the surfactant-depleted lungs (tdl rct, rcrtd, and fxd atel). Within each targeted goal group, results are displayed for the three sample methods: peak, average, and trough. Data are displayed as mean ± 2 sem. Markings refer to significant differences (P < 0.05) between sampling methods, within a targeted goal group: *for peak versus trough, +for peak versus average, and #for average versus trough.

The amplitude of Pao₂ oscillations (peak minus trough) as determined by timed blood gas collection versus the amplitude from the Pao₂ probe is shown in Figure 3. The correlation was statistically significant (r^–2 = 0.54; P < 0.001). The linear regression of amplitude from timed blood gas collection on amplitude from the probe (y = 0.624(x) + 4.19) is also plotted in Figure 3 (dashed line), along with the line of identity (solid line).

View larger version (22K): [in this window] [in a new window]   Figure 3. Comparison of Pao2 oscillation amplitude from the indwelling probe versus timed blood gas sampling (peak minus trough). The solid line is the line of identity (y = x); the dashed line is the regression of y on x (y = 0.624(x) + 4.187).

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Changes in Pao₂ within a respiratory cycle have traditionally been thought to be small, based on measurements in normal lungs (14) and on mathematical models of alveolar Po₂ changes within a respiratory cycle (15,16). However, remarkably large oscillations in Pao₂ were demonstrated in a surfactant depletion model of acute lung injury (12). The maximum difference, in that study, between Pao₂ at its peak value at end-inspiration and the Pao₂ at its trough value at end-expiration was 439 mm Hg. These substantial oscillations in Pao₂ were interpreted to be the result of changes in shunt fraction resulting from tidal recruitment of atelectasis (12). It is unknown how commonly tidal recruitment and large Pao₂ oscillations occur in mechanically ventilated patients. Rapid changes in lung aeration, however, have been demonstrated in several animal models of surfactant depletion (1,4,5,7–9) and also in some patients in an intensive care unit (10).

Conventional ABG analysis could be influenced by large Pao₂ oscillations, but the magnitude of this effect has not been previously reported. There are several reasons why Pao₂ from intermittent, conventional ABG analysis might not match the results from an indwelling, fast-response, continuous arterial Po₂ probe, for example 1) mixing in the thoracic and abdominal aorta; 2) mixing in the arterial catheter, tubing, stopcock, and syringe; and 3) collection of the discrete ABG sample over a finite fraction of the breath cycle rather than nearly instantaneously. The effect of large Pao₂ oscillations on conventional ABG analysis is therefore difficult to predict without experimental data.

Our study identified two implications of Pao₂ oscillations for conventional arterial blood gas analysis. First, intentionally timed blood gas collection, and comparison of end-inspiratory Pao₂ versus end-expiratory Pao₂, can provide a rough estimate of the amplitude of Pao₂ oscillations. More importantly, the presence of oscillations can indicate that the ventilation strategy is not optimal i.e., the ventilation strategy does not prevent derecruitment during exhalation. Second, when tidal recruitment and large Pao₂ oscillations are present, collection of ABG samples at different time points in the respiratory cycle can result in large variability in Pao₂. If this phenomenon is not recognized, the resulting Pao₂ could potentially be misinterpreted.

Tidal recruitment has long been considered one of the mechanisms of VALI (2,3). Contemporary concepts in ventilator management in acute lung injury therefore aim to avoid tidal recruitment. Applying this concept in current clinical practice, however, is difficult because there is no bedside tool currently available to measure tidal recruitment. Pressure-volume (PV) curves have been used to identify a lower inflection point (17), but the PEEP required to keep the lung open is not necessarily predicted by this inspiratory lower inflection point. Experimental and theoretical evidence suggests that recruitment in acute lung injury might occur throughout the inspiratory limb of the PV curve (18). Moreover, time constants for recruitment and collapse after step changes in airway pressure are substantial compared to the time for one breath cycle (5,19), which implies that airway pressure alone, from a statically constructed PV curve, will not predict the amount of recruitment during dynamic changes in airway pressure. Airway pressure-time curve profiles have been correlated with recruitment as assessed by computed tomography (1), but the maneuver requires interruption of the ventilator settings of interest. Indwelling arterial Po₂ probes in clinical use (4), unlike the laboratory probe used in this study, do not have adequate temporal resolution to identify tidal recruitment. Preliminary results suggest some promise for rapidly responding pulse oximetry (20), but further validation studies will be required. Electrical impedance tomography has great potential as a bedside tool for assessing tidal recruitment and adjusting mechanical ventilation, but this technology is still undergoing development and is not yet generally available (8–10). Dynamic computed tomography (4,5) is readily applied in patients but also is not widely available.

For assessing tidal recruitment in clinical practice, the data in Figure 3 suggests the possible utility of intentionally timed blood gas collection (end-inspiration versus end-expiration) as a crude and approximate tool that is nevertheless readily available with no special equipment. There was a statistically significant relationship between the timed blood gas results and the amplitude of oscillation as determined from the indwelling Pao₂ probe, although there is clearly a large amount of variability within this relationship. With one exception (which we believe is most likely attributable to a shift in probe position after calibration) the data points are near or below the identity line. It would be expected that timed collection of ABG will miss some of the amplitude information because the timing can never be so exact as to limit the blood sampling precisely to the peak or trough values. Nevertheless, the ratio of slopes between the regression line and identity line suggest that timed blood gas collection can reflect up to 62% of the real variations in Pao₂, when the lung is tidally recruited.

It should be noted that, in the clinical setting, the accuracy of timed blood gas collection in assessing tidal recruitment is likely to be less than our study suggests for three reasons. First, in our experiment the peak and trough samples were collected with the benefit of timing information from the indwelling Pao₂ probe. The use of timed blood gas collection without this information would be limited to collection at end-inspiration versus end-expiration, blinded as to the real Pao₂ oscillations. Furthermore, the phasic relationship between the ventilator cycle and Pao₂ changes at the arterial sampling location is unknown. Second, in our study respiratory rates were set to low values to minimize timing errors. Respiratory rates of 10, while not typical for management of acute respiratory distress syndrome (ARDS), are commonly used in intraoperative ventilator management. Third, the outlying data point well above the identity line, which is likely a result of error in the Pao₂ probe, tends to increase the slope of the regression line. Further work would be required to establish the utility of timed ABG collection as a tool for measuring arterial Po₂ oscillations outside of the tightly controlled conditions of our laboratory experiments.

Modern ABG analysis requires a fairly small volume of blood, and it is common clinical practice to aspirate this blood in a small fraction of the respiratory cycle, without regard for the timing within the respiratory cycle. One consequence of large Pao₂ oscillations is therefore a potentially large variability in Pao₂ as assessed by conventional blood gas analysis. In a surfactant depletion model of mild lung injury, we placed a rapidly responding laboratory Po₂ probe in a systemic artery, and used the probe to adjust mechanical ventilation to provide both large and small Pao₂ oscillations. Our results confirm the expectation that conventional blood gases drawn at different times during the respiratory cycle can reflect respiratory changes in Pao₂ measured by the experimental probe. When Pao₂ oscillations were negligible, for example in normal lungs in the first two targeted Pao₂ goals, there was no effect of the timing of sample collection on conventional Pao₂. When Pao₂ oscillations were large, however, as after surfactant depletion, the timing of blood gas sample withdrawal within the respiratory cycle had a large influence on conventionally measured Pao₂.

The experimental model for these studies was a model of mild lung injury from surfactant depletion in mature, adult lungs. Our group (21) and others (22) have documented that with mechanical ventilation, the injury progresses to a more complete model of acute lung injury, with inflammatory cytokine expression, neutrophil adhesion, and protein leakage. In our experiments, all animals' lungs were ventilated at a respiratory rate of 10 breaths per minute, and with Fio₂ of 1.0 after lavage. These ventilator settings are not commonly used in patients with ARDS but are commonly used in the intraoperative care of patients at risk for development of acute lung injury. Our results, therefore, are most relevant to early, mild lung injury in patients at risk for development of ARDS.

The impact of large variations in Pao₂ on conventional blood gas analysis would be especially significant in early lung injury and ARDS because blood gas results are frequently used in clinical practice to adjust mechanical ventilation. For example, modern protective ventilation strategies aimed at reducing VALI emphasize adequate PEEP to avoid tidal recruitment but not so much PEEP as to cause over-distention (17,23). One common approach to finding this PEEP setting is to use ABG as a measure of lung recruitment (24). At a fixed Fio₂, PEEP is increased while checking the Pao₂ for large increases that indicate lung recruitment. When further increases in PEEP cause only small increases in Pao₂, the lung is considered to be recruited, and further increases in PEEP would not be beneficial. If blood gas results are dependent on timing of collection in the respiratory cycle, and this factor is not accounted for, this approach could lead to erroneous conclusions. For example, collection of the sample near a peak value of Pao₂, followed by an increase in PEEP, followed by collection of the sample near a trough value of Pao₂, would lead to the erroneous conclusion that the lung is adequately recruited by the current PEEP.

This study was designed to demonstrate the maximum differences in Pao₂ that could be obtained by purposely timed blood gas collection. The results do not estimate the amount of variability in conventional blood gas analysis that might be expected specifically the result of Pao₂ oscillations, when samples are drawn at random times in the respiratory cycle. The results do suggest, however, that when tidal recruitment is present, the variability in Pao₂ by conventional blood gas analysis can be substantial. The ideal timing of sample collection depends on the expected use of the information. If the goal of Pao₂ measurement is to assess the degree of end-expiratory collapse and atelectasis, the most relevant sample would be collected at end-expiration. If the goal of Pao₂ measurement is to assess average oxygen delivery, the most relevant sample would be collected slowly over several breath cycles.

	Footnotes

 
Accepted for publication January 19, 2006.

Supported, in part, by NIH GM59274; NIH HL59052; AHA 0151528U; DFG PF424/1; DFG Ma2398/3; SpectruMedix LLC, State College PA; the Departments of Anesthesia at the University of Pennsylvania, the Johannes Gutenberg-University of Mainz, and the University of Greifswald; and the Emergency Service at the School of Veterinary Medicine, University of Pennsylvania.

Address correspondence and requests for reprints to James E. Baumgardner, MD, PhD, Department of Anesthesia, University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104. Address e-mail to baumgarj{at}uphs.upenn.edu.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Grasso S, Terragni P, Mascia L, et al. Airway pressure-time curve profile (stress index) detects tidal recruitment/hyperinflation in experimental acute lung injury. Crit Care Med 2004;32:1018–27.[ISI][Medline]
2. Muscedere JG, Mullen JB, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care 1994;149:1327–34.[Abstract]
3. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model of pulmonary elasticity. J Appl Physiol 1970;28:596–608.[Free Full Text]
4. Markstaller K, Kauczor HU, Weiler N, et al. Lung density distribution in dynamic CT correlates with oxygenation in ventilated pigs with lavage ARDS. Br J Anaesth 2003;91:699–708.[Abstract/Free Full Text]
5. Neumann P, Berglund JE, Mondejar EF, et al. Dynamics of lung collapse and recruitment during prolonged breathing in porcine lung injury. J Appl Physiol 1998;85:1533–43.[Abstract/Free Full Text]
6. Neumann P, Berglund JE, Mondejar EF, et al. Effect of different pressure levels on the dynamics of lung collapse and recruitment in oleic-acid-induced lung injury. Am J Respir Crit Care 1998;158:1636–43.[Abstract/Free Full Text]
7. Halter JM, Steinberg JM, Schiller HJ, et al. Positive end-expiratory pressure after a recruitment maneuver prevents both alveolar collapse and recruitment/derecruitment. Am J Respir Crit Care 2003;167:1620–6.[Abstract/Free Full Text]
8. Kunst PW, Vazquez de Anda G, Bohm SH, et al. Monitoring of recruitment and derecruitment by electrical impedance tomography in a model of acute lung injury. Crit Care Med 2000;28:3891–5.[ISI][Medline]
9. Frerichs I, Dargaville PA, Dudykevych T, Rimensberger PC. Electrical impedance tomography: a method for monitoring regional lung aeration and tidal volume distribution? Intens Care Med 2003;29:2312–6.[ISI][Medline]
10. Victorino JA, Borges JB, Okamoto VN, et al. Imbalances in regional lung ventilation: a validation study on electrical impedance tomography. Am J Resp Crit Care 2004;169:791–800.[Abstract/Free Full Text]
11. Williams EM, Viale JP, Hamilton RM, et al. Within-breath arterial PO₂ oscillations in an experimental model of acute respiratory distress syndrome. Br J Anaesth 2000;85:456–9.[Abstract/Free Full Text]
12. Baumgardner JE, Markstaller K, Pfeiffer B, et al. Effects of respiratory rate, plateau pressure, and positive end-expiratory pressure on PaO2 oscillations after saline lavage. Am J Respir Crit Care 2002;166:1556–62.[Abstract/Free Full Text]
13. Draper NR, Smith H. Applied regression analysis, 2nd ed. New York: John Wiley and Sons, 1981.
14. Folgering H, Smolders FDJ, Kreuzer F. Respiratory oscillations of the arterial PO₂ and their effects on the ventilatory controlling system in the cat. Pflugers Arch 1978;375:1–7.[ISI][Medline]
15. Hlastala MP. A model of fluctuating alveolar gas exchange during the respiratory cycle. Respir Physiol 1972;15:214–32.[ISI][Medline]
16. Lin KH, Cumming G. A model of time-varying gas exchange in the human lung during a respiratory cycle at rest. Respir Physiol 1973;17:93–112.[ISI][Medline]
17. Amato MBP, Barbas CSV, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998;338:347–54.[Abstract/Free Full Text]
18. Hickling KG. Re-interpreting the pressure-volume curve in patients with acute respiratory distress syndrome. Curr Opin Crit Care 2002;8:32–8.[Medline]
19. Markstaller K, Eberle B, Kauczor HU, et al. Temporal dynamics of lung aeration determined by dynamic CT in a porcine model of ARDS. Br J Anaesth 2001;87:459–68.[Abstract/Free Full Text]
20. Syring R, Otto CM, Campbell V, Baumgardner JE.. Respiratory variations in SpO2 with a fast pulse oximeter [abstract]. Anesthesiology (Suppl) 2003:A-375.
21. Markstaller K, Otto CM, Karmrodt J, et al. Regional and temporal distribution of iNOS activity in a model of acute lung injury. Shock 2005;23(Suppl 3):84.
22. Rotta AT, Gunnarsson B, Fuhrman BP, et al. Comparison of lung protective ventilation strategies in a rabbit model of acute lung injury. Crit Care Med 2001;29:2176–84.[ISI][Medline]
23. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301–8.[Abstract/Free Full Text]
24. Brochard L, Roudot-Thoraval F, Roupie E, et al. Tidal volume reduction for prevention of ventilator-induced lung injury in acute respiratory distress syndrome. Am J Respir Crit Care 1998;158:1831–8.[Abstract/Free Full Text]

This article has been cited by other articles:

				C. M. Otto, K. Markstaller, O. Kajikawa, J. Karmrodt, R. S. Syring, B. Pfeiffer, V. P. Good, C. W. Frevert, and J. E. Baumgardner Spatial and temporal heterogeneity of ventilator-associated lung injury after surfactant depletion J Appl Physiol, May 1, 2008; 104(5): 1485 - 1494. [Abstract] [Full Text] [PDF]

				G. Musch, G. Bellani, M. F. Vidal Melo, R. S. Harris, T. Winkler, T. Schroeder, and J. G. Venegas Relation between Shunt, Aeration, and Perfusion in Experimental Acute Lung Injury Am. J. Respir. Crit. Care Med., February 1, 2008; 177(3): 292 - 300. [Abstract] [Full Text] [PDF]

				R. S. Syring, C. M. Otto, R. E. Spivack, K. Markstaller, and J. E. Baumgardner Maintenance of end-expiratory recruitment with increased respiratory rate after saline-lavage lung injury J Appl Physiol, January 1, 2007; 102(1): 331 - 339. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Pfeiffer, B. Articles by Baumgardner, J. E. Search for Related Content PubMed PubMed Citation Articles by Pfeiffer, B. Articles by Baumgardner, J. E. Related Collections Blood Equipment Technology

Anesthesia & Analgesia® is published for the International Anesthesia Research Society® by Lippincott Williams & Wilkins with the assistance of Stanford University Libraries' HighWire Press®. Copyright 2006 by the International Anesthesia Research Society. Online ISSN: 1526-7598   Print ISSN: 0003-2999