Published ahead of print on January 21, 2005, doi:10.1164/rccm.200407-940OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200407-940OC
Effects of High versus Low Positive End-Expiratory Pressures in Acute Respiratory Distress SyndromeDipartimento dell'Emergenza e Trapianti d'Organo, Sezione di Anestesiologia e Rianimazione, Università degli Studi di Bari; and Azienda Sanitaria Locale Bari-4, Ospedale Di Venere, Servizio di Anestesia e Rianimazione, Bari, Italy Correspondence and requests for reprints should be addressed to S. Grasso, M.D., Dipartimento dell'Emergenza e Trapianti d'Organo (DETO), Sezione di Anestesiologia e Rianimazione, Ospedale Policlinico, Piazza Giulio Cesare 11, Bari, Italy 70124. E-mail: grassos{at}libero.it
A recent study by the Acute Respiratory Distress Syndrome Network compared the traditional lower end-expiratory pressure strategy with a higher end-expiratory pressure strategy in patients with the acute respiratory distress syndrome ventilated with low tidal volumes. Clinical outcomes were similar whether lower or higher positive end-expiratory pressure (PEEP) levels were used. We applied both the lower (9 ± 2 cm H2O) and higher (16 ± 1 cm H2O) PEEP strategy in 19 patients. In nine recruiters, the higher end-expiratory pressure strategy resulted in significant alveolar recruitment (587 ± 158 ml), improvement in arterial oxygen partial pressure/inspired oxygen fraction ratio (from 150 ± 36 to 396 ± 138), and reduction in static lung elastance (from 23 ± 3 to 20 ± 2 cm H2O/L). In 10 nonrecruiters, alveolar recruitment was minimal, oxygenation did not improve, and static lung elastance significantly increased (from 26 ± 5 to 28 ± 6 cm H2O/L). The increase in oxygenation, the reduction in static lung elastance, and the shape of the volume–pressure curve during the lower PEEP strategy were independently associated with alveolar recruitment. In conclusion, the protocol proposed by the Acute Respiratory Distress Syndrome Network, lacking solid physiologic basis, frequently fails to induce alveolar recruitment and may increase the risk of alveolar overinflation.
Key Words: acute respiratory distress syndrome, human • positive-pressure respiration/methods • respiratory distress syndrome, adult/therapy The first lung-protective ventilatory strategy proposed by the Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network, based on low VT, resulted in a 22% mortality reduction in a prospective, multicenter, randomized study (2). Pivotal to that strategy were both prevention of tidal alveolar overstretch and limitation of tidal alveolar excursions. Positive end-expiratory pressure (PEEP) and inspired oxygen fraction (FIO2) were set according to traditional criteria, using a table of lower PEEP/higher FIO2 combinations, aiming at the lowest PEEP level compatible with an oxygenation target. Besides tidal alveolar overstretch, tidal alveolar recruiting/derecruiting has also been implicated in ventilator-induced lung injury (3, 4). Higher-than-traditional PEEP levels together with lung recruiting maneuvers have been proposed to reduce the portion of nonaerated lung, thus avoiding cyclic alveolar recruiting/derecruiting during mechanical ventilation (5). Combining low VTs and high PEEP levels blunted the inflammatory response (6) and reduced mortality in patients with ARDS (7). The ARDS Network Assessment of Low Tidal Volume and Elevated End-Expiratory Lung Volume to Obviate Lung Injury (ALVEOLI) study was designed to validate these results in a larger clinical study (8). Study-arm patients were ventilated with low VTs and with a higher PEEP strategy aiming to decrease the amount of nonaerated lung at end-expiration, whereas control-arm patients were ventilated according to the former ARDS Network protective protocol (lower PEEP strategy) (2). Technically, the higher PEEP strategy was implemented through a higher PEEP/lower FIO2 combination table with the addition of lung recruiting maneuvers. The ALVEOLI study was stopped after the enrollment of 549 patients, having crossed a protocol-specified futility-stopping boundary (8). For the effectiveness of lung protection, any higher PEEP approach should increase the end-expiratory lung volume through alveolar recruitment, avoiding lung overinflation (9). Furthermore, PEEP and lung recruiting maneuvers may increase the end-expiratory lung volume through two opposite mechanisms: (1) by increasing the proportion of aerated alveoli at end-expiration or (2) by further inflating already-ventilated lung regions (10, 11). The occurrence of one of these two mechanisms (or their combination) may depend on ARDS etiology (12), lung and chest wall mechanics (13), the stage of the disease (14), and lung morphology seen on the computed tomography scan (11). In addition, the optimal implementation protocol for higher PEEP and recruiting maneuver is still debated (15). Because, in the ALVEOLI study, the higher PEEP strategy was table-based, matched on an oxygenation target, and applied regardless of any patient-related variable, we set up the hypothesis that the effectiveness of this strategy in generating alveolar recruitment could be unpredictable.
Nineteen patients with ARDS were studied. Inclusion and exclusion criteria were those of the ALVEOLI study (8). The institutional review board approved the protocol, and written, informed consent was obtained from the subjects or their next of kin. Patients were ventilated according to the ARDS Network lower PEEP strategy before the study (2). After a brief (30–35 minutes) baseline ventilation period (PEEP, 0; VT, 10 ml/kg predicted body weight; FIO2, 1; respiratory rate, 15 breaths/minute; inspiratory-to-expiratory ratio, 1:2), each patient was ventilated with the lower PEEP strategy for 12 hours and subsequently with the higher PEEP strategy for 12 hours (Table 1). At the end of each study period, measurements of gas exchange, hemodynamics, and respiratory mechanics were obtained.
Flow, volume, airway opening pressure, and esophageal pressure were measured (pneumotachograph and differential pressure transducers). Values of airway opening pressure at end-expiration of a regular breath (PEEPexternal) and 3 to 5 seconds after the onset of an end-expiratory occlusion (PEEPtotal) were also measured. Static intrinsic PEEP (PEEPi,st) was obtained as the difference between PEEPtotal and PEEPexternal. Static elastance of the respiratory system, chest wall, and lung were calculated through standard formulae (16).
Quasi-static pressure–volume curves of the respiratory system during low-flow inflation of the VT (low-flow tidal inflation) were obtained during each experimental condition applying the same principles of the "low-flow technique" (17–19). Subsequently, the shape of each volume–pressure curve was quantitatively analyzed by fitting it to the following equation:
Volume–pressure curves obtained during each experimental condition were plotted on the same volume/pressure axis by referring them to the elastic equilibrium volume of the respiratory system. The recruited volume was measured as the difference in lung volume for the same static airway opening pressure read on the volume–pressure curves (20, 21). Patients were defined as recruiters if the higher PEEP strategy induced an alveolar recruitment of more than 150 ml and as nonrecruiters if alveolar recruitment was lower than 150 ml (22). Invasive arterial pressure, heart rate, right atrial pressure, continuous cardiac output (transesophageal Doppler, Doptek ODM1; Deltex Medical, Chichester, UK), and arterial blood gases (Rapid Lab 865; Bayer Diagnostics, Dublin, Ireland) were recorded.
Statistical Analysis
Moving from the lower to the higher PEEP strategy resulted in alveolar recruitment between 25 and 125 ml (mean, 70 ± 38 ml) in 10 nonrecruiters and between 425 and 850 ml (mean, 587 ± 158 ml; p < 0.01) in 9 recruiters. Compared with the baseline period, the lower PEEP strategy induced alveolar recruitment in both groups; however, the amount of alveolar recruitment induced by the lower PEEP strategy was significantly higher in recruiters than in nonrecruiters (Figure 1).
Table 2 shows the ventilatory settings and gas exchange parameters at baseline and at the end of each 12-hour study period. During the lower PEEP strategy, there were no differences in PEEPexternal and FIO2 between the two groups, whereas during the higher PEEP strategy, PEEPexternal and FIO2 were significantly higher in nonrecruiters than in recruiters. The PaO2/FIO2 ratio significantly increased in both groups moving from baseline to the lower PEEP strategy. Moving from the lower to the higher PEEP strategy, this ratio remained unchanged in nonrecruiters, whereas it increased in recruiters. PaCO2 increased significantly in nonrecruiters moving from the lower to the higher PEEP strategy, although minute volume remained unchanged between the two strategies (9.2 ± 1.9 vs. 8.3 ± 1.1 L/minute, p = 0.129).
Demographic and clinical characteristics of the studied patients are shown in Table 3. ARDS had a pulmonary origin in 4 of 9 recruiters, and in 6 of 10 nonrecruiters. The male/female ratio was significantly different in the two groups (8/1 in recruiters and 5/5 in nonrecruiters, p < 0.05). The Acute Physiology, Age, and Chronic Health Evaluation III score (23) taken on the day before the study was not significantly different between the two groups. Age was not significantly different between the groups, although there was a strong trend toward older ages in nonrecruiters.
At baseline, static lung elastance was significantly higher in nonrecruiters than in recruiters (29.6 ± 5.8 vs. 25.3 ± 3.9 cm H2O/L), whereas static chest wall elastance was not significantly different between the two groups (6.8 ± 2.5 in nonrecruiters and 7.8 ± 2.3 cm H2O/L in recruiters) and remained unchanged during the different experimental conditions (Figure 2). Compared with baseline measurements, the lower PEEP strategy induced a significant reduction of static lung elastance in both groups. The higher PEEP strategy further reduced static lung elastance in recruiters (from 23.2 ± 3.1 to 20.2 ± 2.4 cm H2O/L, p < 0.05), whereas it increased it in nonrecruiters (from 26.8 ± 5.5 to 28.3 ± 6.4 cm H2O/L).
In a representative recruiter, the quasi-static volume–pressure curve obtained during the lower PEEP strategy showed a progressive decrease in elastance (i.e., an upward concavity; Figure 3), with coefficient b less than 1 (see METHODS, Equation 1). In the same patient, the curve obtained during the higher PEEP strategy was shifted upward along the volume axis, suggesting alveolar recruitment. In a representative nonrecruiter, the quasi-static volume–pressure curve obtained during the lower PEEP strategy showed a progressive increase in elastance during low-flow tidal inflation (i.e., an upward convexity; Figure 3), with coefficient b higher than 1; the quasi-static volume–pressure curve obtained during the higher PEEP strategy was almost superimposed to the volume–pressure curve obtained during the lower PEEP strategy, suggesting an increase in end-expiratory lung volume without lung recruitment.
At baseline, the coefficient b was lower than 1 in both groups (mean, 0.967 ± 0.062 in nonrecruiters and 0.889 ± 0.098 in recruiters; p < 0.05). In recruiters, it remained lower than 1 during lower PEEP strategy (mean, 0.908 ± 0.081; p = not significant as compared with baseline), whereas in nonrecruiters, it increased significantly (1.050 ± 0.06). During the higher PEEP strategy, the mean coefficient b value significantly increased in both groups (1.133 ± 0.089 in nonrecruiters and 1.026 ± 0.050 in recruiters, p < 0.05 between the two groups). The hemodynamic pattern remained unchanged in both groups during the different experimental conditions, except for a significant increase in stroke volume in recruiters during the higher PEEP strategy (Table 4). Fluid balance during the 12 hours of lower PEEP strategy was not significantly different in recruiters and nonrecruiters, whereas during the higher PEEP strategy, significantly more fluids were administered to nonrecruiters as compared with the recruiters (Table 4).
The relationship between alveolar recruitment induced by the higher PEEP strategy (dependent outcome variable) and four potentially relevant independent variables was examined by multivariate regression analysis followed by a stepwise regression with backward elimination. The independent variables entered in the model were ARDS underlying disease, coefficient b value during the lower PEEP strategy, and variations in static respiratory system elastance and in PaO2/FIO2 ratio induced by the higher PEEP strategy as compared with the lower PEEP strategy. We found that coefficient b value during the lower PEEP strategy and variations in static respiratory system elastance and in the PaO2/FIO2 ratio induced by the higher PEEP strategy were independently associated with alveolar recruitment. On the contrary, the ARDS underlying disease was not independently associated with alveolar recruitment (Table 5).
The implementation protocol for the higher PEEP strategy used in the ALVEOLI study resulted in alveolar recruitment only in 9 of 19 patients, confirming our hypothesis that such a protocol may have unpredictable effects on alveolar recruitment. Some limitations of this study must be addressed:
According to the ALVEOLI protocol (8), in our study the applied PEEP level was significantly different between the two strategies (9 ± 2 vs. 16 ± 1 cm H2O, p < 0.05). However, in nonrecruiters, PEEP had to be increased significantly more than in recruiters to match the oxygenation target (Table 2). These findings can be explained if it is considered that the higher PEEP/lower FIO2 combination table adopted in the ALVEOLI study favors the increase of PEEP against the increase of FIO2 (Table 1). In nonrecruiters lacking alveolar recruitment, the oxygenation was most likely dependent on the applied FIO2, and to reach a higher FIO2 level in the table, PEEP had to be increased to higher levels than in recruiters. In a recent study, De Durante and coworkers (28) found significant intrinsic PEEP in patients ventilated according to the ARDS Network lower PEEP strategy, although we were not able to replicate these findings (Table 2). The difference in respiratory rate between ours (23 ± 5 breaths/minute) and De Durante's study (34 ± 1 breaths/minute) may explain these discrepancies. This hypothesis is confirmed by another study demonstrating that the use of high respiratory rates to compensate for VT reduction, whereas maintaining a constant inspiratory-to-expiratory ratio may generate intrinsic PEEP (29). Furthermore, if the increase in respiratory frequency is associated with an increase in inspiratory-to-expiratory ratio (the inspiratory time remaining constant despite the shortening of duty cycle), the use of high respiratory rate further reduces the time for lung emptying, generating significant air trapping (30). The physiologic effects of the lung recruiting maneuvers were not specifically assessed in our study. However, our data suggest that lung recruiting maneuvers were not effective in nonrecruiters. A possible explanation for this finding is that lung recruiting maneuvers were applied by following the protocol after the initial PEEP setting (Table 1). In fact, experimental data suggest that the effects of a lung recruiting maneuver on alveolar recruitment are transient if the preceding PEEP levels are maintained after the maneuver (22) and that, once the alveoli have been recruited, higher PEEP levels are required to keep them aerated (31). Confirming our explanation, an ancillary study performed during the ALVEOLI trial showed that the improvement in oxygenation induced by the lung recruiting maneuvers was only transient (32). Gattinoni and coworkers (12) showed that PEEP improves lung mechanics and induces significant alveolar recruitment in patients with extrapulmonary ARDS, whereas it results in lung overinflation and worsening of lung mechanics in patients with pulmonary ARDS. Although in our study the number of patients with pulmonary ARDS was higher in nonrecruiters than in recruiters, the underlying disease was not independently associated with alveolar recruitment (Table 5). On the other hand, Rouby and coworkers (11) were not able to confirm the correlation between ARDS etiology and PEEP-induced alveolar recruitment, whereas they clearly demonstrated that the distribution pattern of the loss of lung aeration seen on the computed tomography scan may predict the potential for lung recruitment. The same group has recently shown that applying a PEEP level of 15 cm H2O results in maximal recruitment in patients with "diffuse" loss of lung aeration and in mild recruitment plus lung overinflation in patients with "lobar" loss of lung aeration (33). Although we have no data to show which was the distribution pattern of the loss of lung aeration in our patients, in nonrecruiters we found both lung recruitment at lower PEEP levels and an increase in end-expiratory lung volume without further lung recruitment at higher PEEP levels (Figure 4).
In recruiters, the shape of the static volume–pressure curve during the lower PEEP strategy showed a pattern of ongoing elastance reduction during low-flow tidal inflation, with a coefficient b lower than 1 in the power equation fitting. Because this finding suggests the occurrence of intratidal recruitment (20, 34), we can speculate that, in recruiters, the lower PEEP strategy was not able to completely meet the potential for lung recruitment. Confirming this hypothesis, we found that lower coefficient b values during the lower PEEP strategy were independently associated with alveolar recruitment induced by the higher PEEP strategy (Table 5). On the contrary, the pattern of an ongoing increase in elastance with tidal inflation in nonrecruiters suggests that tidal overinflation was already present during the lower PEEP strategy, thus predicting a very low potential for further alveolar recruitment with the higher PEEP strategy (20, 34). In nonrecruiters, PaCO2 increased significantly during the higher PEEP strategy as compared with the lower PEEP strategy (Table 2). Because minute ventilation was not significantly different between the two conditions, the increase in PaCO2 suggests a PEEP-induced increase in alveolar dead space, which is an indirect sign of alveolar overinflation (35). However, we have compared PaCO2 levels recorded with a time interval of 12 hours and the effects of differences in metabolic CO2 production cannot be ruled out. Although PEEP may critically affect venous return, particularly if it does not induce alveolar recruitment (36), we were not able to show systemic hemodynamic impairment in nonrecruiters. Because nonrecruiters received significantly more intravenous fluids during the higher PEEP strategy period as compared with recruiters (Table 4), we speculate that blood volume expansion could have compensated for the PEEP-induced systemic hemodynamic impairment (37). These results are strengthened by the standardization of the hemodynamic management throughout the study period (see online supplement). Higher PEEP levels may increase right ventricular afterload, which induces right ventricular impairment (37). However, right ventricular function was not directly assessed in this study, so we have no data to show if our patients developed right ventricular dysfunction. In conclusion, our data suggest that the standardized higher PEEP and lower FIO2 implementation protocol proposed by the ARDS Network in the ALVEOLI study lacks a solid physiologic basis and therefore frequently fails to induce alveolar recruitment. In addition, this protocol may increase the risk of alveolar overinflation, favoring the use of higher PEEP levels whenever the lungs are not prone to be recruited. We found that some physiologic variables (the increase in oxygenation, the reduction in static lung elastance, and the shape of the volume–pressure curve) could be useful in predicting alveolar recruitment when applying higher-than-traditional PEEP levels. Further studies are needed to fully elucidate the role of higher-than-traditional PEEP levels in patients with ARDS.
The authors thank Lidia Dalfino, M.D., for her help in statistical analysis, and Caterina Brindicci, M.D., and Sarah Esselfie-Quaye, M.D., for their help in English-language editing.
Some of the results of this study have been previously reported in the form of an abstract (1). This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Conflict of Interest Statement: S.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; V.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; A.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; G.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; T.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 20, 2004; accepted in final form January 18, 2005
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
p < 0.05 higher versus lower PEEP strategy in the same group; ¶p < 0.05 between the two groups in the same experimental condition. As compared with the baseline, the lower PEEP strategy induced significant alveolar recruitment in both groups, although the amount of recruited volume was significantly higher in recruiters than in nonrecruiters. The higher PEEP strategy induced further alveolar recruitment in recruiters only.
p < 0.05 higher versus lower PEEP strategy in the same group; ¶p < 0.05 between the two groups, at baseline. Estrs was significantly higher at baseline in nonrecruiters as compared with recruiters. This was because of a higher EstL, because Estcw was similar in the two groups of patients. Estrs and EstL were significantly reduced by the lower PEEP strategy in both groups of patients. Moving from the lower to the higher PEEP strategy further reduced Estrs and EstL in recruiters, whereas Estrs and EstL in nonrecruiters increased. Estcw remained unchanged during the two ventilatory strategies in both groups.
