Original Contributions

A Study of the Physiologic Responses to a Lung Recruitment Maneuver in Acute Lung Injury and Acute Respiratory Distress Syndrome
Maureen O Meade MD, Deborah J Cook MD, Lauren E Griffith MSc, Lori E Hand RRT, Stephen E Lapinsky MD, Thomas E Stewart MD, Kieran J Killian MD, Arthur S Slutsky MASc MD, and Gordon H Guyatt MD MSc
OBJECTIVE: To determine the magnitude, duration, and consistency of the effects of lung recruitment maneuvers (RMs) on oxygenation, lung mechanics, and comfort in patients with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS). METHODS: We conducted a prospective physiologic study at 3 tertiary-care hospitals. We enrolled 28 consecutive eligible patients with ARDS or ALI and a ratio of PaO2 to fraction of inspired oxygen (PaO2/FIO2) < 250 mm Hg while receiving FIO2 > 0.50. We performed RMs twice daily for 3 days. The first RM was at 35 cm H2O for 20 s. If initial response was equivocal, the clinician immediately administered another RM at a higher pressure (40 cm H2O, then 45 cm H2O) or for longer period (30 s, then 40 s), in a randomized order. Each patient had up to 6 sets of up to 3 RMs. RESULTS: Twenty-seven patients met the criteria for ARDS at baseline; 1 had ALI. There was no net effect on oxygenation or pulmonary mechanics following the first or subsequent RMs. The largest rise in PaO2 was from 61 mm Hg to 71 mm Hg, and the largest decrease was 6 mm Hg following the first RM. Augmenting the inflation pressure or duration had no significant effect. These findings precluded analyses about predictors of response or consistency of response. Over the entire study of 122 RMs, 5 patients developed ventilator asynchrony, 3 appeared uncomfortable, 2 experienced transient hypotension, and 4 developed barotrauma that required intervention. CONCLUSIONS: These results do not support the addition of scheduled RMs to usual treatment for ALI or ARDS. Key words: recruitment maneuver, acute respiratory distress syndrome, acute lung injury, critical illness, respiration, mechanical ventilation. [Respir Care 2008;53(11):14411449. 2008 Daedalus Enterprises]

Introduction The therapeutic concept of lung-protective ventilation, which dates back to the 1960s,1-3 has emerged in the past decade as the most important advance in the management of acute lung injury (ALI) and acute respiratory distress syndrome (ARDS). The lung recruitment maneuver (RM) is one of many experimental lung-protective strategies.4

Often described as a sigh on the ventilator, an RM includes an inspiratory breath-hold. The goal is to open the lung by inflating it to total lung capacity.5

SEE

THE

RELATED EDITORIAL

ON

PAGE 1429

Maureen O Meade MD, Deborah J Cook MD, Kieran J Killian MD, and Gordon H Guyatt MD MSc are affiliated with the Department of Medicine, McMaster University, Hamilton, Ontario, Canada. Maureen O Meade MD, Deborah J Cook MD, Gordon H Guyatt MD MSc, Lauren E Griffith MSc, and Lori E Hand RRT are affiliated with the Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada. Stephen E Lapinsky MD, Thomas E Stewart MD, and Arthur S Slutsky MASc MD are affiliated with the Department of Medicine, University of Toronto, Toronto, Ontario, Canada.

Several mechanisms may underlie lung protection from RMs. First, recruiting collapsed alveoli may temporarily improve gas exchange and thus reduce the required fraction of inspired oxygen (FIO2). Second, with an open lung, a lower pressure can expand the lungs through the tidal range. Third, if applied with adequate positive end-expiratory pressure (PEEP), an RM may reduce repetitive opening and closing that causes shear stress that can damage terminal lung units.6 In summary, an RM that achieves an open lung may reduce the risk of oxygen toxicity, overdistention injury, and shear-stress injury.

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

1441

LUNG RECRUITMENT MANEUVERS

There are several approaches to RMs. One of the most feasible and popular approaches is to increase the static airway pressure to a fixed level (eg, 40 cm H2O) for a fixed period (eg, 40 s). Amato et al used this approach in a randomized trial of a multifaceted lung-protective ventilation strategy with RMs following all ventilator disconnects.7 Their experimental open-lung strategy was associated with a large and statistically significant reduction in 28-day mortality (relative risk 0.53, 95% confidence interval 0.31 0.91), but the contribution of RMs to that survival benefit could not be independently evaluated. That work led to additional studies of RMs for patients with ALI. Historically, RMs were found to safely improve arterial oxygenation in patients who develop atelectasis due to general anesthesia,8 and in neonates undergoing high-frequency oscillation for respiratory distress syndrome.9 Case series in patients with ALI or ARDS, in which the RM inflation pressure range was 30 60 cm H2O and the duration 3 s to 3 min, have reported various improvements in gas exchange.10-16 Few adverse effects have been reported. Results from randomized trials of RMs have been less encouraging. A randomized trial of RMs versus a sham procedure was conducted with 80 patients in the context of a larger study. RMs with pressures of 35 40 cm H2O for 30 s had minimal effects on oxygenation and were associated with statistically significant decreases in blood pressure (mean 7 mm Hg, P .01).17 A related trial of RMs with 30 patients with extrapulmonary ARDS used inflation pressure of 50 cm H2O for 30 s and found no sustained oxygenation improvement.18 Our aim was to investigate the magnitude, duration, and consistency of effect of intermittent scheduled RMs on arterial oxygenation, FIO2, lung mechanics, heart rate, blood pressure, and patient comfort in patients with ARDS/ALI. We also measured the effects of increasing the RM inflation pressure and duration in patients whose oxygenation did not improve after the initial RM. Methods The study was conducted in the intensive care units at Hamilton Health Sciences and St Josephs Healthcare in Hamilton, Ontario, and at Mt Sinai Hospital in Toronto, Ontario, Canada. These units care for general medicalsurgical patients, trauma patients, neurosurgical patients,

IN

ALI

AND

ARDS

and patients with burns. The research ethics board at each hospital approved the study. Each patient (or legal surrogate) provided informed consent for participation. Patients We included patients who were receiving mechanical ventilation and had ALI or ARDS (PaO2/FIO2 250 mm Hg on FIO2 0.50, regardless of the level of applied PEEP, and bilateral infiltrates on a frontal chest radiograph). We excluded patients whose mean arterial pressure was 70 mm Hg, who had chronic interstitial lung disease, subcutaneous emphysema, pneumomediastinum, pneumothorax with a persistent air leak, previous pneumonectomy, exacerbation of asthma or chronic obstructive pulmonary disease, cardiogenic pulmonary edema (based on a physicians assessment), whose anticipated duration of mechanical ventilation was 72 h, or who had undergone lung biopsy or resection during the current hospital stay. Randomization We randomly allocated consecutive eligible patients to 2 groups, which corresponded to 2 methods of augmenting the RM pressure or duration if the first RM had an equivocal response. To conceal the blocked randomization scheme, we used a central telephone randomization system stratified by center. Intervention RMs were performed twice daily for 3 consecutive days, so each patient could receive up to 6 RMs. Research respiratory therapists (RTs) supervised many of the RMs. Clinical RTs conducted some of the study RMs without direct supervision from research personnel. For logistical reasons related to data collection, we generally did not initiate a study RM after 3 pm. We did not perform RMs after lung injury was resolved (in the judgment of the attending physician). The response to RM was rated positive, negative, or equivocal based on the following criteria. A positive response was an absolute increase in arterial oxygen saturation measured via pulse oximetry (SpO2) of 3% within 5 min of completing the RM. A negative response was an SpO2 decrease of 3% or systolic-blood-pressure decrease of 10 mm Hg. An equivocal response was neither positive nor negative. We used pressure-control ventilation, delivered with either a Dura 2 (Dra ger, Lubeck, Germany), a model 7200 (Puritan Bennett, Pleasanton, California), or (in 1 patient only) a model 840 (Puritan Bennett, Pleasanton, California) mechanical ventilator. Clinicians provided routine sedation at their discretion and did not administer additional

This research was supported by the Hamilton chapter of the Ontario Thoracic Society. The authors report no conflicts of interest related to the content of this paper. Correspondence: Maureen O Meade MD, Clinical Epidemiology and Biostatistics, McMaster University Medical Centre, 1200 Main Street W, Hamilton, Ontario L8N 3Z5 Canada. E-mail: meadema@hhsc.ca.

1442

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

LUNG RECRUITMENT MANEUVERS

analgesia, sedation, or neuromuscular blockade for the purpose of lung recruitment. We reasoned that any changes in the background level of sedation or paralysis following baseline assessments could confound the study results. If, however, the clinicians anticipated that a patient was insufficiently sedated for the procedure, additional background sedation or analgesia was administered and the baseline assessment was deferred until there was an apparent and consistent effect. Prior to an RM the patient was suctioned and the FIO2 was adjusted to achieve an SpO2 of 90 92%, to maximize the detection of any PaO2 improvement. Beginning at end-expiration we delivered continuous positive airway pressure (with no pressure support) at 35 cm H2O for 20 s, and then immediately restored the prior ventilation settings. If the response to an RM after 5 min was equivocal, we repeated the RM immediately, with either a greater inflation pressure (40 cm H2O) or a longer duration (30 s), as indicated by the randomization directives. If the response to the second RM was also equivocal, we conducted a third (and final) RM with either a greater inflation pressure (45 cm H2O) or a longer inflation period (40 s). The Puritan Bennett ventilators could deliver continuous positive airway pressure at 35 cm H2O, but the Dura 2 could not, so with the Dura 2 we adapted the airway-pressure-release-ventilation (APRV) mode to achieve continuous positive airway pressure in 20 patients.19 For the RM the APRV mode was set as follows. The high pressure parameter was set to the desired inflation pressure, and the low pressure parameter was set to 35 cm H2O, which was the maximum low pressure option. The high time parameter was set at 30 s, which was the longest inspiratory-time option. The low time parameter was set to 0.1 s, which was the shortest inspiratory-time option. We manually measured the inspiratory time. The Dura 2 reliably delivered the prescribed continuous positive airway pressure. In pre-study assessments we connected a Dura 2 to a calibration analyzer (Timeter RT200, Allied Healthcare Products, St Louis, Missouri). Pressure-control mode that simulated the patient experience was followed by APRV at inflation pressure of 35 cm H2O for 20-s, 30-s, and 40-s RMs. We also tested inflation pressures of 40 cm H2O and 45 cm H2O for 20 s each. The airway pressure always exceeded the desired pressure by 0.55 cm H2O, and never fell below the desired pressure. Safeguards We continuously monitored blood pressure and SpO2 during the RM. The RM was immediately terminated if mean arterial pressure fell below 65 mm Hg or 70% of the baseline pressure. Volume-depletion was considered and

IN

ALI

AND

ARDS

an intravenous fluid bolus was administered if the supervising clinician deemed it clinically appropriate. We resumed RMs after the patient regained stability, and only with the approval of the supervising clinician. RMs were also terminated if SpO2 decreased to 85% or if the patient appeared uncomfortable. Patients who experienced hypotension (mean arterial pressure 65 mm Hg), desaturation, or discomfort on 2 consecutive RMs received no further RMs. If a study patient became unstable at any time during the 3-day period, further study was either discontinued or postponed until the patients condition stabilized. Concurrent Respiratory Support FIO2 was increased if SpO2 decreased to 85% within 15 min of the RM, or to 88% thereafter. Use of inhaled nitric oxide was permitted during the study. Prone positioning was permitted only for refractory hypoxemia (SpO2 90%, FIO2 90%, and progressively increasing PEEP), during which further study was postponed. RM was not performed within 30 min of any change in mechanical ventilation. FIO2 and PEEP were not reduced for 6 h following RM. Measurements We collected demographic data, including patient age, sex, height, weight, Acute Physiology and Chronic Health Evaluation II score at admission to the intensive care unit, and all risk factors that contributed to lung injury. A designated study RT collected the following baseline data immediately after endotracheal suctioning and prior to each RM: PaO2, FIO2, arterial oxygen saturation (SaO2), E), mean airway pressure, intrinsic SpO2, minute volume (V PEEP, total PEEP, ratio of inspiratory time to expiratory time, plateau and mean airway pressures, ventilation mode, use of neuromuscular blockers, use of prone positioning, presence of subcutaneous emphysema/pneumothorax/other air leaks, and mean arterial blood pressure. The most recent chest radiograph from that day was considered the baseline chest radiograph. We also documented the Glasgow coma score and the amount of sedation, analgesia, and neuromuscular blocker administered. The study RT prospectively recorded data immediately after the RMs and at 5 min, 20 min, 40 min, and 1, 2, 3, 4, 5, and 6 h after the RMs. Analyses Starting with the first set of RMs with each patient, we evaluated changes from baseline (ie, immediately prior to the RM set) to 5 min after the final RM in a set. We evaluated PaO2 changes at constant FIO2 and PEEP. We

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

1443

LUNG RECRUITMENT MANEUVERS

IN

ALI

AND

ARDS

Met Inclusion Criteria (45) Excluded (8)

- Improved (4) - Declined consent (4)

Table 1.

Subjects* (n 28) Allocated to PressureAugmented RMs 5/9 28/0 59 19 25 7 Allocated to TimeAugmented RMs 5/9 27/1 55 19 22 10

Total

Allocation (37) Did not receive intervention (9)

- New hemodynamic instability - Clinical improvement

Received intervention (28) Lost to follow-up (0) Follow-Up (28) Analysis (28)
Fig. 1. Patient flow through the study.

Female/male (n) ARDS/ALI (n) Age (mean SD y) APACHE-II Score at ICU admission (mean SD) ARDS risk factors per patient (mean SD) ARDS risk factors Severe sepsis (n) Pneumonia (n) Shock (n) Gastric aspiration (n) Burn injury (n) Pulmonary contusion (n) Acute pancreatitis (n) Multiple transfusions (n) Drug overdose (n) Smoking history (n)

10/18 27/1 57 19 23 8

1.5 0.5

1.5 0.5

1.5 0.5

15 8 7 3 3 2 2 1 1 9

8 1 7 1 0 2 1 0 1 2

7 7 0 2 3 0 1 1 0 7

evaluated changes in lung mechanics, as reflected by change E over that brief period, at constant driving pressure in V and sedation level. Compliance was not specifically measured. We repeated these assessments for the subsequent sets of RMs, from set 2 to 6, as applicable. Next we evaluated changes in oxygenation and ventilation over the 6-hour period following each RM set. Lack of response rendered irrelevant our planned regression modeling to evaluate potential predictors of response. We did use regression modeling to test for an interaction between RM responses and baseline PEEP. Results Over the 1-year period June 1999 through May 2000, 45 patients met the inclusion criteria and 28 (62%) completed the study (Fig. 1 and Table 1). Table 2 shows the numbers of RM sets conducted. We conducted a total of 122 RM sets. Reasons for not administering 46 scheduled RMs included clinical instability or death in 7 of 17 patients (41%), barotrauma in 5 (29%), and resolution of lung injury in 5 (29%). Table 3 summarizes the baseline data (from before the first set of RMs). The severity of lung injury ranged from mild to very severe. The mean P a O 2 /F I O 2 was 118 39 mm Hg, and mean PEEP was 10 4 cm H2O. The patients were heavily sedated; their mean abbreviated Glasgow coma score (ie, only the motor and eye components of the Glasgow coma scoring system) was 4.7 3.3.

* Data collected prior to the first set of recruitment maneuvers Unknown for 7 patients RM recruitment maneuver ARDS acute respiratory distress syndrome ALI acute lung injury APACHE Acute Physiology and Chronic Health Evaluation

Table 2.

Recruitment Maneuver Implementation* Patients Receiving RM Sets n (%) 28 (100) 26 (93) 21 (75) 20 (71) 16 (57) 11 (39) 122 (100)

RM Sets 1 2 3 4 5 6 Total

* 14 patients were allocated to pressure-augmented recruitment maneuvers, and 14 were allocated to time-augmented recruitment maneuvers. NA not applicable

Twelve patients (43%) received neuromuscular paralysis: 10 prior to baseline, one 4 h after RM, and one 5 h after RM. Eight patients (29%) received inhaled nitric oxide: 7 prior to RM, and one 4-h after RM. One patient was ventilated in the prone position throughout the study. The respiratory data (not shown) taken just before the subsequent RMs for each patient were similar to the data

1444

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

LUNG RECRUITMENT MANEUVERS

Table 3. Baseline Data From Before the First Recruitment Maneuver* Total PaO2/FIO2 (mean SD mm Hg) FIO2 (mean SD) PaO2 (mean SD mm Hg) SaO2 (mean SD %) SpO2 (mean SD %) Set PEEP (mean SD cm H2O) Total PEEP (mean SD cm H2O) Plateau pressure (mean SD cm H2O) Mean airway pressure (mean SD cm H2O) Minute ventilation (mean SD L/min) Total respiratory rate (mean SD breaths/min) Tidal volume (mean SD mL) Indexed tidal volume (mean SD mL/kg) Abbreviated Glasgow coma score Ventilator dyssynchrony (n, %) Recent increase in analgesia (n, %) Recent increase in sedation (n, %) Paralyzed (n, %) Prone positioning (n, %) Inhaled nitric oxide (n, %) Chest-wall restriction (n, %) 118 39 0.60 0.19 65 9 94 3 92 2 10 4 10 4 33 6 20 5 12.9 3.6 23 7 603 186 8.4 3.0 4.7 3.3 1 (4) 2 (7) 2 (7) 10 (36) 1 (4) 7 (25) 1 (4)

IN

ALI

AND

ARDS

Allocated to Pressure-Augmented RM 117 35 0.59 0.17 64 9 93 2 93 2 94 10 4 33 8 19 5 12.9 3.8 21 8 649 204 8.7 3.4 3.9 3.2 0 (0) 1 (7) 1 (7) 6 (43) 0 (0) 5 (36) 0 (0)

Allocated to Time-Augmented RM 120 44 0.60 0.20 66 9 94 3 91 2 11 4 11 5 33 5 21 4 12.9 3.4 24 6 556 160 8.2 2.6 5.4 3.3 1 (7) 1 (7) 1 (7) 4 (29) 1 (7) 2 (14) 1 (7)

* All patients were ventilated with a pressure-control mode. Total PEEP was not measurable in 4 patients. The abbreviated Glasgow Coma Score includes only the motor and eye components. Data is missing for one patient. RM recruitment maneuver FIO2 fraction of inspired oxygen SaO2 arterial oxygen saturation SpO2 arterial oxygen saturation measured via pulse oximetry PEEP positive end-expiratory pressure

in Table 3. The remaining data presented herein refer to the first set of RMs for each patient, unless otherwise specified. Figure 2 shows the changes in oxygenation following the first RM, in each patient. Following the first RM, 4 (14%) of 28 patients had a positive response (increase oxygen saturation). The largest corresponding PaO2 improvement was 10 mm Hg (from 61 mm Hg to 71 mm Hg). Sixteen of 28 patients (57%) had a negative response (decreased oxygenation). The largest PaO2 decrease was 6 mm Hg (from 62 mm Hg to 56 mm Hg in one patient, and from 70 mm Hg to 64 mm Hg in another). Eight patients (29%) had an equivocal response. Augmenting the inflation pressure or duration gave no benefit. Overall, the mean change in PaO2 5 min after the first set of RMs was 2.4 5.0 mm Hg. Among the 7 patients who were receiving inhaled nitric oxide at baseline, the mean change in PaO2 after the first RM set was 1.7 4.8 mm Hg, which did not differ from those who were not receiving inhaled nitric oxide (P .68). The results of subsequent sets of RMs were very similar; there was no apparent benefit (data not shown). Figure 3 summarizes the oxy-

genation responses to the RMs. We found no effect of baseline PEEP on any of the results reported above. E were similarly small. The mean change Changes in V E in VE at 5 min was 0.08 0.79 L. In 20 (71%) patients V did not change by 0.5 L/min in either direction. In 5 E increased, which indicates improved patients (18%) V E rise was from 11.6 L/min lung mechanics; the largest V E decreased, which to 14.2 L/min. In 3 patients (11%) V E decrease indicates worse lung mechanics; the largest V was from 17.3 L/min to 15.9 L/min. The results of subsequent RMs were very similar; there was no appreciable improvement after RM. Figure 4 shows the FIO2 changes over the 6 h following the first RM. In one patient FIO2 was reduced from 1.0 to 0.9 at 1 h and further to 0.80 after 5 h. Five patients (18%) had FIO2 increased. No patients had a PEEP decrease. One patient required a PEEP increase after 4 h. Subsequent sets of RMs had a similar lack of effect. The ventilator respiratory rates and driving pressures were not appreciably influenced by RM during the 6 h that followed (data not shown). The levels of sedation, analgesia, and paralysis were essentially constant during that time frame.

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

1445

LUNG RECRUITMENT MANEUVERS

100

IN

ALI

AND

ARDS

90

80

70

60

50 Baseline 5 min

mothorax reported on computed tomogram that should have excluded him from the study. Within 6 h of his first RM there was clinical evidence of barotrauma, and bilateral chest tubes were inserted. A 79-year-old male with aspiration pneumonitis developed barotrauma after his second RM. The RM appeared uncomplicated at the time, but within hours he had bilateral chest tubes inserted for pneumothorax, pneumomediastinum, and pneumopericardium. Finally, a 45-year-old woman developed pneumothorax and subcutaneous emphysema following her fifth study RM, with no change in heart rate or blood pressure up to the time of chest tube insertion, 2 h after the RM. Because of progressing subcutaneous emphysema, a second chest tube was inserted. Three patients (11%) experienced discomfort during RM, manifested primarily as coughing, but this was not associated with subsequent barotrauma. Moreover, none of the 5 instances of patient-ventilator asynchrony resulted in barotrauma. Two patients developed substantial hypotension (defined as a mean arterial pressure 65 mm Hg), and no patients developed substantial desaturation during an RM (defined as an oxygen saturation of 85%). Discussion This randomized trial was unique in the conduct and evaluation of scheduled RMs in addition to usual care for patients with ALI and ARDS. On average, RM had no significant benefits for gas exchange or pulmonary mechanics. Even the greatest improvements in oxygenation and pulmonary mechanics in any individual patient were small, and there was no evidence that more aggressive RMs had any benefit. In patients with an initial equivocal response to the RM, increasing the inflation pressure to 45 cm H2O or the duration from 20 s to 40 s did not achieve noteworthy responses. The lack of response to RM performed as a routine intervention in usual care precluded subsequent attempts to assess potential predictors of response or consistency. Though benefit was absent, 14% of patients developed barotrauma, and 7% developed substantial hypotension during RMs. Approximately 50% of RMs were associated with a decrease in arterial saturation, by at least 3%. The failure of RMs to improve oxygenation in this study contrasts with several reported series.14,20-24 However, we performed regularly scheduled RMs in stable patients across the spectrum of ALI, whereas other investigators used RM for patients with deteriorating gas exchange, where greater effects may be expected.12,25 Other contrasting case series may have been subject to selective reporting of patients with positive findings or selective publication of positive reports. At any rate, many of the reports of improved oxygenation with RM found that this response was very transient.11,12,23,25,26

Fig. 2. Initial changes in PaO2 during the first lung recruitment maneuvers. There were no significant changes in fraction of inspired oxygen or positive end-expiratory pressure in any patient 5-min after the lung recruitment maneuver.

Table 4 summarizes the adverse events associated with the 122 RMs. The average change in heart rate over 5 min was 1 5.2 beats/min, and the average change in mean arterial blood pressure was 0.1 8.0 mm Hg. Though those mean values did not change appreciably, individual fluctuations occurred. In 2 patients (7%) mean blood pressure fell 16 mm Hg (from 102 mm Hg to 86 mm Hg, and from 80 mm Hg to 64 mm Hg). The maximum increase in mean blood pressure was 23 mm Hg (from 70 mm Hg to 93 mm Hg). In the 5 min following RM the maximum decrease in mean arterial pressure was 18 mm Hg (from 82 mm Hg to 64 mm Hg), and the maximum increase was 35 mm Hg (from 77 mm Hg to 112 mm Hg). During RM, heart rate fell as much as 36 beats/min in 1 patient (from 126 beats/min to 90 beats/min) and rose by as much as 13 beats/min in another (from 106 beats/min to 119 beats/ min). Four patients (14%) developed barotrauma, all of which were clinically detected and confirmed on repeat chest radiography on the day of the RM. One 46-year-old male with septic shock who had bilateral chest tubes previously placed for pleural fluid drainage developed, during his first RM, a new air leak through one tube, and a new basal pneumothorax was noted on his next scheduled chest radiograph. A 20-year-old male with pulmonary contusion had a pre-existing, undocumented, and undrained pneu-

PaO2 (mm Hg)

1446

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

LUNG RECRUITMENT MANEUVERS

Pressure-Augmented RM

IN

ALI

AND

ARDS

Time-Augmented RM

Positive response to 1st RM

Positive response to 2nd or 3rd RM Equivocal response to 2nd or 3rd RM Negative response to 2nd or 3rd RM

Positive response to 1st RM

Positive response to 2nd or 3rd RM Equivocal response to 2nd or 3rd RM Negative response to 2nd or 3rd RM 1 RM not done

Negative response to 1st RM

Negative response to 1st RM

Fig. 3. Oxygenation response to 122 sets of lung recruitment maneuvers (RMs). A plus symbol denotes a positive response, defined as an absolute increase in arterial oxygen saturation measured via pulse oximetry (SpO2) of 3%. A minus symbol denotes a negative response, defined as SpO2 decrease of 3% or a systolic-blood-pressure decrease of 10 mm Hg. E denotes an equivocal response (neither positive nor negative).

80

PaO (mm Hg)

2

FIO (%)
2

PaO or FIO

70

60

50

Baseline

End of RM

20 min

40 min

1h

2h

3h

4h

5h

6h

Fig. 4. Mean changes in oxygenation and fraction of inspired oxygen (FIO2) at baseline (ie, just prior to) and over the 6 h after the lung recruitment maneuver (RM).

Table 4.

Adverse Events Associated With 106 Recruitment Maneuvers Type of Adverse Event Number of Adverse Events 5 4 4 2 Number of Patients 5 4 3 2

Patient-ventilator dysynchrony Barotrauma on same-day chest radiograph Patient discomfort Mean arterial pressure decreased to 65 mm Hg or 70% of baseline

Our findings in the present study are consistent with 2 other randomized trials. Oczenski and colleagues found a significant short-term oxygenation improvement in patients with early ARDS who underwent an RM with inflating pressure of 50 cm H2O, but that resolved within 30 min.18 In the ARDS Network trial the failure of RM may have been attributable to the high baseline PEEP (13.8

3 cm H2O), which was even higher than our mean PEEP (10.1 4 cm H2O).17 Unique features of the present study include the randomized design, our focus on scheduled rather than ad hoc RMs, and our use of increased inflation pressure or duration if initial RM response was equivocal. The strengths of our study include minimal confounding effects due to different ventilator modes or setting changes. Patients were ventilated with a pressure-control mode, and the ventilator settings were not substantially changed during the 6 h following RM. There were no changes in FIO2 or PEEP that might mask improvements in gas exchange due to RM. Differences in sedation and paralysis during the 6-h period after RM were largely avoided. Trained study RTs ensured standardization among the patients and sites in this multicenter randomized trial. We frequently collected data on ventilation parameters and gas exchange, and assessed for both improvement and adverse effects.

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

1447

LUNG RECRUITMENT MANEUVERS

Limitations We did not evaluate RMs in the setting of acute or subacute deterioration of gas exchange, so we cannot comment on the effect of RMs under those circumstances. Similarly, our findings do not apply to RMs after ventilator-disconnect or derecruitment caused by suctioning.25 The aim of this trial was to test whether routine application of RMs in standard practice affects oxygenation, lung mechanics, or comfort in patients with ALI or ARDS. In addition, this trial preceded publication of the ARDS Network trial that found that mechanical ventilation with a target tidal volume of 6 mL/kg and airway pressure 30 cm H2O saves lives. Our mean tidal volume at baseline was 8.45 2.95 mL/kg, and mean plateau airway pressure was 33.2 6.1 cm H2O, and it is plausible that RMs may be more effective in the setting of lower tidal volume. It is conceivable that the inflation pressure or duration were on average too low to open the lungs and achieve important benefit. More recent studies used much higher RM pressure and PEEP,8 and some case series have reported improved oxygenation after RM if PEEP was adjusted to avoid derecruitment.10,27,28 Indeed, in the present study the average baseline plateau airway pressure was relatively high, such that high recruiting pressure might be anticipated to have greater effect on oxygenation. However, the rate of negative responses and the lack of response to augmenting the inflating pressure do not support that notion. Logically, adverse effects must increase with pressure and duration; over 14% of our patients suffered barotrauma. Finally, our sample size of 28 patients is relatively small, and arguably underpowered to detect an important effect. However, the total lack of RM benefit in our patients makes it unlikely that we have missed major benefits. Conclusions The findings from this randomized trial provide no support for the addition of routinely scheduled RMs (with pressure up to 45 cm H2O or duration of 40 s) to the current management of critically ill patients with ALI or ARDS.
REFERENCES 1. Greenfield LJ, Ebert PA, Benson DW. Effect of positive pressure ventilation on surface tension properties of lung extracts. Anesthesiology 1964;25:312-316. 2. Dreyfuss D, Basset F, Soler P, Saumon G. Intermittent positivepressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am Rev Respir Dis 1985;132(4): 880-884. 3. Kolobow T, Moretti MP, Fumagalli R, Mascheroni D, Prato P, Chen V, Joris M. Severe impairment in lung function induced by high peak 16. 4.

IN

ALI

AND

ARDS

5. 6.

7.

8.

9. 10.

11.

12.

13.

14.

15.

17.

18.

19.

airway pressure during mechanical ventilation. Am Rev Respir Dis 1987;135(2):312-315. McCulloch PR, Forkert PG, Froese AB. Lung volume maintenance prevents lung injury during high frequency oscillatory ventilation in surfactant-deficient rabbits. Am Rev Respir Dis 1988;137(5):11851192. Lachmann B. Open the lung and keep the lung open. Intens Care Med 1992;18(6):319-321. Muscedere JG, Mullen JB, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 1994;149(5):1327-1334. Amato MB, Barbas CS, Medeiros DM, Magaldi RB, Schettino GP, Lorenzi-Filho G, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998;33(6):347-354. Rothen HU, Sporre B, Engberg G, Wegenius G, Hedenstierna G. Re-expansion of atelectasis during general anaesthesia: a computed tomography study. Br J Anaesth 1993;71(6):788-795. Froese AB. Role of lung volume in lung injury: HFO in the atelectasis-prone lung. Acta Anaesthesiol Scand Suppl 1989;33:126-130. Borges JB, Okamoto VN, Matos GF, Caramez MP, Arantes PR, Barros F, et al. Reversibility of lung collapse and hypoxemia in early acute respiratory distress syndrome. Am J Respir Crit Care Med 2006;174(3):268-278. Po voa P, Almeida E, Fernandez A, Mealha R, Moreira P, Sabino H. Evaluation of a recruitment maneuver with positive inspiratory pressure and high PEEP in patients with severe ARDS. Acta Anaesthesiol Scand 2004;48(3):287-293. Lapinsky SE, Aubin M, Mehta S, Boiteau PE, Slutsky AS. Safety and efficacy of a sustained inflation for alveolar recruitment in adults with respiratory failure. Intens Care Med 1999;25(11):12971301. Pelosi P, Bottino N, Chiumello D, Caironi P, Panigada M, Gamberoni C, et al. Sigh in supine and prone position during acute respiratory distress syndrome. Am J Respir Crit Care Med 2003; 167(4):521-527. Grasso S, Mascia L, Del Turco M, Malacarne P, Giunta F, Brochard L, et al. Effects of recruiting maneuvers in patents with acute respiratory distress syndrome ventilated with protective ventilatory strategy. Anesthesiology 2002;96(4):795-802. Foti G, Cereda M, Sparacino ME, De marchi L, Villa F, Pesenti A. Effects of periodic lung recruitment maneuvers on gas exchange and respiratory mechanics in mechanically ventilated acute respiratory distress syndrome (ARDS) patients. Intens Care Med 2000;26(5): 501-507. Barbas CSV, Silva E, Garrido A, Assuncao M, Hoelz EC, Knobel E. Recruitment maneuvers with different pressure control levels in ARDS patients (abstract). Am J Respir Crit Care Med 2001;163:A767. The ARDS Clinical Trials Network; National Heart, Lung, and Blood Institute, National Institutes of Health; Brower RG, Morris A, MacIntyre N, Matthay MA, Hayden D Thompson T, et al. Effects of recruitment maneuvers in acute lung injury and acute respiratory distress syndrome ventilated with high positive end-expiratory pressure. Crit Care Med 2003;31(11):2592-2597. Erratum in Crit Care Med 2004;32(3):907. Oczenski W, Ho rmann C, Keller C, Lorenzi N, Kepka A, Schwarz S, Fitzgerald RD. Recruitment maneuvers after a positive endexpiratory pressure trial do not induce sustained effects in early adult respiratory distress syndrome. Anesthesiology 2004;101(3): 620-625. Hand L, Meade MO. Lung recruitment maneuvers utilizing airway pressure release ventilation (OPEN FORUM abstract). Respir Care 2000;

1448

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

LUNG RECRUITMENT MANEUVERS

45(8):1012. http://www.rcjournal.com/abstracts/2000/?ida00000320. Accessed September 10, 2008. Zavala E, Roca J, Adalia R, Herring C, Valera JL, Alcon A, et al. Long lasting effects of recruitment manoeuvres on pulmonary gas exchange in ALI/ARDS patients (abstract). Am J Respir Crit Care Med 2001;163:A295. Bugedo G, Bruhn A, Herna ndez G, Rojas G, Varela C, Tapia JC, Castillo L. Lung computed tomography during a lung recruitment maneuver in patients with acute lung injury. Intensive Care Med 2003;29(2):218-225. Wauer H, Groll G, Krausch D, Lehmann C, Kox WJ. [Clinical experiences with the open lung concept.] Anaesthesiol Reanim 2003; 28(2):38-44. Article in German. Li MQ, Zhang Z, Li SM, Shi ZX, Xu JY, Lu F, et al. [Comparative study on recruitment maneuvers in acute respiratory distress syndrome with pulmonary and extrapulmonary origin.] Chin Crit Care Med 2006;18(6):355-358. Article in Chinese.

IN

ALI

AND

ARDS

20.

21.

22.

23.

24. Talmor D, Sarge T, Legedza A, ODonnell CR, Ritz R, Loring SH, Malhortra A. Cytokine release following recruitment maneuvers. Chest 2007;132(5):1434-1439. 25. Dyhr T, Bonde J, Larsson A. Lung recruitment manoeuvres are effective in regaining lung volume and oxygenation after open endotracheal suctioning in acute respiratory distress syndrome. Crit Care 2003;7(1):55-62. 26. Schreiter D, Reske A, Scheibner L, Glien C, Katscher S, Josten C. [The open lung concept: a clinical trial in severe chest trauma.] Chuirg 2002;73(4):353-359. Article in German. 27. Tugrul S, Akinci O, Ozcan P, Ince S, Esen F, Telci L, et al. Effects of sustained inflation and postinflation positive end-expiratory pressure in acute respiratory distress syndrome: focusing on pulmonary and extrapulmonary forms. Crit Care Med 2003;31(3):738-744. 28. Girgis K, Hamed H, Khater Y, Kacmarek RM. A decremental PEEP trial identifies the PEEP level that maintains oxygenation after lung recruitment. Respir Care 2006;51(10):1132-1139.

RESPIRATORY CARE NOVEMBER 2008 VOL 53 NO 11

1449