Respiratory failure requiring mechanical ventilation

Transcription

1 Current Issues in Mechanical Ventilation for Respiratory Failure* Neil R. MacIntyre, MD The morbidity and mortality associated with respiratory failure is, to a certain extent, iatrogenic. Mechanical ventilation, although the mainstay of treatment for respiratory distress syndrome, can result in physical trauma to lung tissue (ventilator-induced lung injury [VILI]). Strategies to alleviate VILI are often termed lung-protective strategies and are aimed at reducing overstretching and shear stresses associated with repetitive alveolar collapse and reopening. Lower tidal volumes during ventilation, maintenance of positive-end expiratory pressure, and high-frequency ventilation are the best-studied lung-protective strategies that appear to reduce VILI. Faster withdrawal from mechanical ventilation could also improve outcomes and lower the costs associated with care. To enhance the success of weaning from mechanical ventilation, the cooperative efforts of physicians and respiratory therapists are needed. These efforts involve the initiation of spontaneous-breathing trials, implementation of systematic weaning protocols, and optimization of individual patient interventions. (CHEST 2005; 128:561S 567S) Key words: mechanical ventilation; respiratory failure; ventilator-induced lung injury; weaning mechanical ventilation Abbreviations: ARDSNet ARDS Network; Fio 2 fraction of inspired oxygen; HFOV high-frequency oscillatory ventilation; PEEP positive end-expiratory pressure; SBT spontaneous breathing trial; VILI ventilator-induced lung injury; VT tidal volume Learning Objectives: 1. To understand the mechanisms responsible for the development of ventilator-induced lung injury. 2. To review the key clinical approaches for reducing the incidence of ventilator-induced lung injury. 3. To discuss weaning from mechanical ventilation and interventions that may improve extubation success. *From the Department of Pulmonary and Critical Care Medicine, Duke University Medical Center, Durham, NC. This publication supported by an educational grant from Ortho Biotech Products, L.P. The following authors have disclosed financial relationships with a commercial party. Grant information and company names appear as provided by the author: Neil R. MacIntyre, MD, FCCP: Viasys Health Care - Consultant Fee; Ortho Biotech - Speaker bureau. The following authors have disclosed that he or she may be discussing information about a product/procedure/technique that is considered research and is not yet approved for any purpose: Neil R. MacIntyre, MD, FCCP: New modes of mechanical ventilation. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( org/misc/reprints.shtml). Correspondence to: Neil R. MacIntyre, MD, Clinical Chief, Pulmonary and Critical Care Medicine, Duke University Medical Center, Room 7453 Duke Hospital, Box 3911 Medical Center, Durham, NC 27710; neil.macintyre@duke.edumc Respiratory failure requiring mechanical ventilation has a large impact on hospital economics and resources and is associated with substantial morbidity and mortality. ARDS has an estimated incidence between 1.3 to 89/100,000 personyears, 1 3 and until recently has had a mortality rate of 40 to 50%. 4,5 One study 5 estimated the direct medical expenditures for mechanical ventilation of patients in the United States to be in the $4 to $5 billion range. Despite improvements in the understanding of the pathophysiology and treatment of ARDS, mortality has been slow to improve. Mechanical ventilation, which is the mainstay of management of acute lung injury and ARDS, may actually contribute to the morbidity and mortality associated with ARDS through a number of possible mechanisms. These putative mechanisms include direct mechanical damage, induction of surfactant failure, and stimulation of pulmonary and systemic inflammatory cytokines. 6 These mechanisms will be addressed in this review, as will other important issues regarding the management of patients receiving mechanical ventilation for respiratory failure. Ventilator-Induced Lung Injury Mechanical ventilation is associated with two primary types of pulmonary injury: volutrauma and atelectrauma. Volutrauma is easily understood and occurs when the lung is overinflated and alveoli are overstretched. Interestingly, it appears that not only is maximal stretch important, but also tidal stretch, rate of stretch, and frequency of stretch. Although animal experiments indicate that the inflation volumes necessary to cause overdistension injury in normal lungs are much larger than those commonly used clinically, it is important to note that the injured CHEST / 128 / 5/ NOVEMBER, 2005 SUPPLEMENT 561S

2 lung is not homogeneous. Thoracic CT images of ARDS patients reveal regional zones of collapse or consolidation interspersed with less injured regions. 6,7 Thus, delivering a normal tidal volume (VT) [eg, one based on the patient s weight] would end up primarily in the healthier regions, resulting in regional overinflation. 8 A similar regional overinflation may occur in mechanical ventilation in obstructive pulmonary disease. The second type of ventilator-induced lung injury (VILI), atelectrauma, is caused by the repetitive opening and closing of recruitable alveoli. In the injured lung, positive-pressure ventilation can force open some airless alveoli, but on expiration these same alveoli again collapse. This cycling between open and collapsed is often referred to as recruitment-derecruitment of alveoli. The shear stresses occurring during recruitment-derecruitment are high and can cause trauma, resulting in disruption of the surfactant monolayer, especially when the opening/closing cycle is repetitive. Loss or disruption of the surfactant monolayer will result in not only a requirement for higher pressures to achieve alveolar opening but may affect the permeability of the alveolar-capillary barrier to proteins and other solutes. Whether caused by overstretching or repetitive opening and closing of alveoli, VILI is associated with the production and release of inflammatory mediators and cytokines such as tumor necrosis factor-, interleukin-6, macrophage inflammatory protein-2, platelet-activating factor, and thromboxane B 2. 9,10 These inflammatory mediators not only contribute to the exacerbation of local pulmonary injury but may also lead to a systemic inflammatory response, ultimately resulting in multiple organ dysfunction syndrome. Lung-Protective Ventilation Strategies The contributions of VILI to the morbidity and mortality in ARDS or other acute lung injuries has resulted in the testing of several ventilation strategies aimed at minimizing lung injury. These strategies are designed to prevent recruitment-derecruitment and to minimize the potential for overinflation. Lower VTs The traditional approach to mechanical ventilation was to ensure adequate oxygenation as well as to control arterial carbon dioxide levels and ph. To reach these goals, VTs of 10 to 15 ml/kg of body weight were employed. 11 These volumes are greater than observed in normal subjects at rest (approximately 5 to 7 ml/kg), and because of the heterogeneity of the mechanical properties of the injured lung would be expected to contribute to regional overinflation injury during mechanical ventilation. Although smaller VTs would be expected to reduce lung injury in patients receiving mechanical ventilation, concern about causing respiratory acidosis, reducing arterial oxygenation, and creating discomfort discouraged its practice. Despite these misgivings, a large clinical study 12 was conducted to test the concept that lower VTs would decrease mortality in ARDS patients. In the ARDS Network (ARDSNet) study, patients with acute lung injury and ARDS were randomized to one of two ventilator protocols based on VT. In the high-vt group, patients received mechanical ventilation with a VT of 12 ml/kg (ideal or predicted body weight), with a maximum plateau pressure 50 cm H 2 O. In the low-vt group, a VT of 6 ml/kg with a maximum plateau pressure 30 cm H 2 O was used. In both groups, the ventilator rate was adjusted to achieve an arterial ph of 7.3 to 7.45, and fraction of inspired oxygen (Fio 2 ) and positive end-expiratory pressure (PEEP) were varied to control Pao 2 within 55 to 80 mm Hg. The study was stopped early because interim analysis revealed a highly significant reduction in 28-day mortality in the low-vt group. Mortality was 39.8% in the high-vt group and 31.0% in the low-vt group (p 0.007). Of particular interest was the effect of VT on the ratio of Pao 2 to Fio 2 in the two groups. As shown in Figure 1, the patients in the high-vt group had better oxygenation during the first 3 days of the study compared with the low-vt group. Thus, if the study had been designed to look at short-term improvements in oxygenation rather than long-term mortality, the conclusions would have been quite different. This emphasizes the concept that sometimes an intervention producing a physiologic benefit (more alveoli opened by the high-vt strategy) may actually cause overall harm (regional overdistention by the high-vt strategy). Figure 1. Efficiency of oxygenation in patients receiving mechanical ventilation with a VT of 6 ml or 12 ml. Data are from the ARDSNet study 12 ;p 0.05, low-vt group vs high-vt group. 562S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management

3 The positive results in the ARDSNet study 12 are in contrast to the failure of lung-protective ventilation strategy to influence mortality in three smaller trials Several reasons have been put forward to explain these differences. The difference between the VTs used in the two groups was largest in the ARDSNet study, 12 and the average lung-protective VT was smallest, suggesting that study design might partially explain the different outcomes. In addition, the ARDSNet study 12 required increases in ventilation rate and infusions of bicarbonate in response to mild-to-moderate acidosis, which resulted in smaller differences in Paco 2 and ph between study groups than in the other studies. Others 16 have argued that the VTs used in the high-vt group of the ARDSNet study 12 were in fact larger than those used in standard care and may have contributed to a higher mortality in this group. However, surveys of standard-care values for VT have shown substantial equivalence to the high-vt strategy. In addition, mortality was similar in the ARDSNet high-vt group compared to patients receiving traditional ventilation levels in other studies, further weakening this argument (Table 1). PEEP Early studies in animals demonstrated that VILI could be ameliorated if PEEP was maintained. For example, Webb and Tierney 18 showed that normal rat lungs receiving mechanical ventilation at high distending pressures suffered severe injury, characterized by edema and hemorrhage. However, when PEEP was maintained, the degree of injury was substantially reduced. Figure 2 shows several schematic representations of pressure-volume curves of a region of mechanically ventilated injured lung with different levels of PEEP. Curve A shows the pressure-volume relationship during inflation of an injured lung. At approximately 10 cm H 2 O, previously gasless alveoli open and the lung inflates. However, as noted above, this repetitive recruitment-derecruitment of alveoli during the ventilation cycle disturbs the surfactant layer. Figure 2. Schematic representations of pressure-volume relationships in an injured lung region of ARDS patients. Curve A shows the inflation of a lung that had been gasless at the end of expiration. A pressure of approximately 10 cm of H 2 O is required before any inflation occurs. Curve B shows the benefit of maintaining PEEP. In this case, the alveoli do not collapse and enhanced inflation can occur with smaller excursions of pressure. Curve C illustrates the potential for excessively high inflation pressure to contribute to VILI. By maintaining PEEP above the inflection point of the curve as shown in curve B, these recruitable alveoli are kept open at end-expiration. This strategy improves gas exchange, serves to maintain an intact surfactant layer within the alveoli, and moves the pressure-volume curve leftward. The benefit of PEEP does not occur without a cost, as shown in curve C. At higher PEEP levels, VT must be reduced or overstretch injury can occur. The optimum level of PEEP and how to determine that level in individual patients remains somewhat controversial, but it is important to remember that the goal is not necessarily to maximize Pao 2 but to get Pao 2 into an acceptable range (Pao 2 of 55 to 80 mm Hg and an oxygen saturation of 88 to 95%). Some have suggested that PEEP should be set at a level above the lower inflection point of the pressure-volume curve (Fig 2), but this approach has not been tested in clinical outcomes studies. 7,8,11 The ARDS Network used PEEP-Fio 2 tables guided by oxygenation criteria in both the original ventilator management trial as well as a later trial evaluating a Table 1 VTs and Mortality Rates in Randomized Trials of Low VTs* VT, ml/kg Mortality, % Study Patients, No. Low High Low High p Value Amato et al Stewart et al Brochard et al Brower et al ARDSNet et al *Data are presented as mean SD unless otherwise indicated. CHEST / 128 / 5/ NOVEMBER, 2005 SUPPLEMENT 563S

4 more aggressive PEEP strategy (the Assessment of Low Tidal Volume and Elevated End-Expiratory Volume to Obviate Lung Injury Study). 12,19 The two protocols for varying PEEP and Fio 2 are presented in Figure 3. The study was planned for enrollment of 750 patients but was stopped at the second interim analysis because mortality in the two groups was not different. Thus, while most investigators agree that maintenance of PEEP is beneficial, 6 the method of determining the appropriate level for maximum benefit has not yet been decided. High-Frequency Ventilation High-frequency ventilation using small VTs may be an appropriate strategy for minimizing VILI in ARDS patients. In the Multicenter Oscillatory Ventilation for ARDS Trial, ARDS patients were randomized to either conventional ventilation with a VT of 5 to 10 ml/kg and PEEP 10 cm H 2 Oorto high-frequency oscillatory ventilation (HFOV). The patients in the HFOV group received ventilation at 5 breaths/s at a mean airway pressure of 5 cm H 2 O higher than that observed during conventional ventilation. This elevated mean airway pressure serves the same purpose as PEEP in maintaining open alveoli. The pressure amplitude of ventilation was initially set to achieve vibration of the chest wall and was adjusted along with frequency to achieve Paco 2 from 40 to 70 mm Hg. After 30 days, mortality in the conventional group was 52%, whereas in the HFOV group it was 37% (p 0.102). No safety issues were observed during the study. Although the trend in mortality was not significant, these results indicate that HFOV is a safe and effective mode of ventilation in ARDS patients. Figure 3. Protocols for varying Fio 2 and PEEP according to oxygenation criteria in two ARDSNet trials. 12,18 The squares represent a more aggressive PEEP strategy, and the diamonds represent the original PEEP-Fio 2 strategy. Ventilator Discontinuation The timing and method of discontinuation from mechanical ventilation remains an important clinical problem. As noted above, mechanical ventilation can result in life-threatening complications and therefore should be discontinued as soon as possible. However, premature attempts at weaning from respiratory support can result in failure and reinstitution of mechanical ventilation, which carries an enhanced risk of morbidity and mortality Therefore, it is no surprise that many different strategies for successful weaning have been described in the medical literature. Two large (and related) evidence-based projects were conducted to review the results of published studies and evaluate the different strategies for successful weaning from mechanical ventilation. The first of these was commissioned by the Agency for Healthcare Policy and Research, who asked the McMaster University Evidence-Based Practice Center to evaluate the issues surrounding ventilation weaning and discontinuation. 24 This group was directed to address five specific questions: (1) when should weaning be initiated? (2) what criteria should be used to determine when to begin the weaning process? (3) what are the most effective methods of weaning? (4) what are the optimal roles for nonphysician health-care providers in the weaning process? and (5) what is the value of using clinical practice algorithms or protocols in the weaning process? The McMaster project reviewed 5,000 reports and identified 154 publications, which they used to create their comprehensive review. The second group was a task force put together by the American College of Chest Physicians, the Society for Critical Care Medicine, and the American Association for Respiratory Care. This group was charged to create clinical practice guidelines based on the McMaster report and their own literature review and expert consensus opinion. 21 Several of the important clinical practice guidelines are discussed below. When Should Weaning Be Considered? All patients should be evaluated daily to determine if they are candidates for discontinuation of mechanical ventilation. To be considered a candidate, a patient should meet four criteria: (1) evidence of reversal or stability of the cause of acute respiratory failure; (2) adequate oxygenation as indicated by Pao 2 /Fio to 200, PEEP in the range of 5to8cm H 2 O, Fio to 0.5, and ph 7.25; (3) hemodynamic stability (no active myocardial ischemia or clinically important hypotension requiring pressor drug therapy); and (4) ability to make an inspiratory effort. 564S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management

5 Other objective measurements that can be used to evaluate the candidacy of the patients for ventilator withdrawal may include stable heart rate, absence of fever, adequate hemoglobin ( 8 to 10 g/dl), and mental state. It is important to understand that the evaluation of each patient should be individualized. In general, however, patients who meet all four of the criteria above should be considered candidates for the next step in the evaluation, the spontaneous breathing trial (SBT). SBT: Although the criteria listed above indicate which patients are candidates for ventilator withdrawal, this clinical evaluation is not sufficient to predict success. Therefore, a patient who meets these criteria should then be formally evaluated with an SBT. During this critical test, close observation is important, especially during the first few minutes when the frequency/vt ratio should be utilized. If the frequency/vt remains from 60 to 105 breaths/l during the first few minutes, the patient should be allowed to continue the SBT. During the continuation of the SBT, which should last for 30 to 120 min, respiratory and hemodynamic parameters should be monitored, patient comfort should be assessed, and mechanical ventilation should be reinstated if abnormalities in any of these parameters are observed. This test may be conducted with a T-tube, with continuous positive airway pressure, or with low levels of pressure support (5 to 7 cm H 2 O); these strategies have little effect on the outcome. Successful completion of an SBT is highly predictive of successful weaning. 21 Some of the traditional measurements that can be made while a patient is on the ventilator eg, minute ventilation, negative inspiratory force, maximal inspiratory pressure are of little use in evaluating the potential for weaning of individual patients. 21 Failure of SBT: Patients receiving mechanical ventilation who fail an SBT should be evaluated and reversible causes of failure corrected. For example, respiratory mechanics may be improved by addressing fluid balance issues, the need for bronchodilators, or the adequacy of pain control. Provided the patient continues to meet the criteria for attempting weaning, the SBT should be repeated every 24 h. In the interim, patients should receive comfortable, nonfatiguing, and stable ventilatory support. It is important to note that no good evidence exists to support the notion that gradually reducing the support level of mechanical ventilation in patients who Figure 4. Kaplan-Meier analysis of the duration of mechanical ventilation after a successful evaluation for weaning. 24 A successful SBT was defined as the ability of a patient to breathe without mechanical ventilation for 2 h. Monitored by nurses, the SBT could be stopped for any of the following reasons: if the respiratory rate was 35 breaths/min for 5 min, if arterial oxygen saturation fell to 90%, if the heart rate was 140 beats/min, if the heart rate increased or decreased by 20%, if systolic BP was 180 mm Hg or 90 mm Hg, or if the patient experienced increased anxiety or diaphoresis. Patients in the intervention group were discontinued from mechanical ventilation sooner than those in the standard-care group (p 0.001, Cox-proportional hazard analysis). Data are used with permission from Ely et al CHEST / 128 / 5/ NOVEMBER, 2005 SUPPLEMENT 565S

6 have failed an SBT will decrease the time to discontinuation of mechanical ventilation. 21 Weaning Protocols and Nonphysician Health-Care Professionals: Clear evidence exists to support the use of weaning protocols to complement clinical judgment. Such protocols are used to systematically evaluate patients receiving mechanical ventilation to assess their potential ability to be removed from mechanical support. One such study was reported by Ely and colleagues. 25 In this study, 300 patients receiving mechanical ventilation were randomly assigned to a protocol similar to the daily SBT strategy described above or to standard care. The patients in both groups were evaluated daily by a respiratory therapist, but only patients in the intervention group who were candidates for weaning were administered SBTs. No specific weaning method was specified and the attending physician made all decisions regarding the approach to weaning and the discontinuation of ventilation in both treatment groups. Although the patients in the intervention group were more ill (based on acute physiology and chronic health evaluation II scores and mean acute lung injury scores), they were successfully liberated from mechanical ventilation sooner than those in the standard-care group (Fig 4). Other studies 26,27 have similarly demonstrated that protocol-driven strategies result in faster weaning from mechanical ventilation, lower costs, and reduced complications as compared to physician-directed approaches. These studies also clearly illustrated that nonphysician health-care professionals can successfully and safely play a major role in executing these protocols that improve clinical outcomes and reduce costs. Conclusions Despite continuing improvement in the understanding of the pathophysiology and treatment of ARDS, mechanical ventilation for its treatment remains associated with substantial morbidity and mortality and several clinical controversies remain. Because mechanical ventilation itself may contribute to lung injury in these patients, lung-protective ventilation strategies have been tested. These strategies, which employ smaller VTs along with PEEP, appear to result in reduced mortality in ARDS patients, as does the use of high-frequency ventilation. The method of successful withdrawal of mechanic ventilation from an individual patient remains driven by clinical judgment. However, protocols designed to identify the patients most able to be successfully liberated from mechanical support and executed primarily by nonphysician health-care professionals have consistently shown faster withdrawals in the protocol-intervention groups than in the traditionalcare patients. References 1 Goss CH, Brower RG, Hudson LD, et al. Incidence of acute lung injury in the United States. Crit Care Med 2003; 31: Lewandowski K, Metz J, Deutschmann C, et al. Incidence, severity, and mortality of acute respiratory failure in Berlin, Germany. Am J Respir Crit Care Med 1995; 151: Luhr OR, Antonsen K, Karlsson M, et al. Incidence and mortality after acute respiratory failure and acute respiratory distress syndrome in Sweden, Denmark, and Iceland. The ARF Study Group. Am J Respir Crit Care Med 1999; 159: Sloane PJ, Gee MH, Gottlieb JE, et al. A multicenter registry of patients with acute respiratory distress syndrome: physiology and outcome. Am Rev Respir Dis 1992; 146: Kurek CJ, Dewar D, Lambrinos J, et al. Clinical and economic outcome of mechanically ventilated patients in New York State during 1993: analysis of 10,473 cases under DRG 475. Chest 1998; 114: Pinhu L, Whitehead T, Evans T, et al. Ventilator-associated lung injury. Lancet 2003; 361: Rouby JJ, Puybasset L, Nieszkowska A, et al. Acute respiratory distress syndrome: lessons from computed tomography of the whole lung. Crit Care Med 2003; 31:S285 S295 8 Gattinoni L, Pesenti A, Avalli L, et al. Pressure-volume curve of total respiratory system in acute respiratory failure: computed tomographic scan study. Am Rev Respir Dis 1987; 136: Chiumello D, Pristine G, Slutsky AS. Mechanical ventilation affects local and systemic cytokines in an animal model of acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160: Tremblay L, Valenza F, Ribeiro SP, et al. Injurious ventilatory strategies increase cytokines and c-fos m-rna expression in an isolated rat lung model. J Clin Invest 1997; 99: Marini JJ. Evolving concepts in the ventilatory management of acute respiratory distress syndrome. Clin Chest Med 1996; 17: 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: Brochard L, Roudot-Thoraval F, Roupie E, et al. Tidal volume reduction for prevention of ventilator-induced lung injury in acute respiratory distress syndrome. The Multicenter Trial Group on Tidal Volume reduction in ARDS. Am J Respir Crit Care Med 1998; 158: Stewart TE, Meade MO, Cook DJ, et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. Pressure- and Volume-Limited Ventilation Strategy Group. N Engl J Med 1998; 338: Brower RG, Shanholtz CB, Fessler HE, et al. Prospective, randomized, controlled clinical trial comparing traditional versus reduced tidal volume ventilation in acute respiratory distress syndrome patients. Crit Care Med 1999; 27: Eichacker PQ, Gerstenberger EP, Banks SM, et al. Metaanalysis of acute lung injury and acute respiratory distress syndrome trials testing low tidal volumes. Am J Respir Crit Care Med 2002; 166: S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management

7 17 Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338: Webb HH, Tierney DF. Experimental pulmonary edema due to intermittent positive pressure ventilation with high inflation pressures: protection by positive end-expiratory pressure. Am Rev Respir Dis 1974; 110: Brower RG. NHLBI ARDS Network: report on ongoing clinical trials. Update on studies of ventilatory management in ARDS. Presentation at the American Thoracic Society 99th International Conference, Seattle, WA, May Derdak S, Mehta S, Stewart TE, et al. High-frequency oscillatory ventilation for acute respiratory distress syndrome in adults: a randomized, controlled trial. Am J Respir Crit Care Med 2002; 166: MacIntyre NR, Cook DJ, Ely EW Jr, et al. Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians, the American Association for Respiratory Care, and the American College of Critical Care Medicine 2001; 120:375S 395S 22 Epstein SK, Ciubotaru RL, Wong JB. Effect of failed extubation on the outcome of mechanical ventilation. Chest 1997; 112: Epstein SK, Ciubotaru RL. Independent effects of etiology of failure and time to reintubation on outcome for patients failing extubation. Am J Respir Crit Care Med 1998; 158: Agency for Healthcare Research and Quality. Criteria for weaning from mechanical ventilation. Rockville MD: Agency for Healthcare Research and Quality, 2000; ARHQ publication 00-E Ely EW, Baker AM, Dunagan DP, et al. Effect on the duration of mechanical ventilation of identifying patients capable of breathing spontaneously. N Engl J Med 1996; 335: Kollef MH, Shapiro SD, Silver P, et al. A randomized, controlled trial of protocol-directed versus physician-directed weaning from mechanical ventilation. Crit Care Med 1997; 25: Marelich GP, Murin S, Battistella F, et al. Protocol weaning of mechanical ventilation in medical and surgical patients by respiratory care practitioners and nurses: effect on weaning time and incidence of ventilator-associated pneumonia. Chest 2000; 118: CHEST / 128 / 5/ NOVEMBER, 2005 SUPPLEMENT 567S