CHAPTER 133 HIGH-FREQUENCY VENTILATION: LESSONS LEARNED AND FUTURE DIRECTIONS

Transcription

1 CHAPTER 133 HIGH-FREQUENCY VENTILATION: LESSONS LEARNED AND FUTURE DIRECTIONS NIALL D. FERGUSON ARTHUR S. SLUTSKY Mechanical ventilation practices continue to evolve. In particular, the ventilatory care of patients with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) has been refocused on avoiding or minimizing further damage to the fragile, injured lung. Conventional ventilators have been used successfully to limit this ventilator-induced lung injury (VILI) (1 3), but questions remain regarding the optimal method for lung protection during mechanical ventilation. The recognition of the critical role that VILI can play in the outcome of adults with ALI/ARDS has, in part, spawned a renewed interest in alternative modes of ventilation that may be more lung protective. High-frequency ventilation (HFV) is just such a modality. In this chapter, we briefly review the different subtypes of HFV and then concentrate on high-frequency oscillation as the mode with the most supporting data and promise. As will be seen, a great deal of experience with HFV has been accrued in the neonatal and pediatric settings. We will draw on this knowledge where relevant, but focus this chapter from the viewpoint of HFV as an emerging therapy in the adult intensive care unit (ICU). DEFINITIONS, TERMINOLOGY, AND SUBTYPES OF HIGH- FREQUENCY VENTILATION High-frequency ventilation is a collection of ventilatory modes, grouped together by their common property of employing high respiratory rates, all greater than 60 breaths per minute (4,5). In addition to high-frequency oscillation (HFO), other modes in this group include high-frequency jet ventilation (HFJV) and high-frequency percussive ventilation (HFPV). High-frequency Jet Ventilation HFJV is a mode of ventilation in which gas is delivered through a small-bore catheter into the lungs at rates of 100 to 150 breaths per minute (5,6). Delivered tidal volume is still small, but higher than just the volume exiting the jet, as the jets entrain an additional flow of gas by the Venturi effect. Exhalation is passive during HFJV, and gas trapping or dynamic hyperinflation can be an issue. In practice, a conventional ventilator is set up as a slave to the jet to provide positive end-expiratory pressure, along with basic monitoring and alarms. HFJV is a very efficient mode for removing CO 2. Additionally, because of HFJV s very high flow rates and the differing pulmonary time constants in the clinical setting of bronchopleural fistulae, this mode of ventilation is purported to have beneficial effects in the presence of this disorder; however, these benefits have not been verified during objective testing (7,8). Concerns with HFJV relate to the delivery of high pressures (10 50 pounds per square inch), unpredictable tidal volumes, and the development of dynamic hyperinflation, all of which may worsen VILI rather than minimize it. In addition, problems with adequate humidification of inspired gas and the subsequent risks of tracheobronchitis have been noted and documented. High-frequency Percussive Ventilation HFPV is the newest and least well studied of the HFV modes. It combines a high-frequency rate of 200 to 900 breaths per minute superimposed on a conventional pressure mode of ventilation (9). HFPV is reported to enhance the clearance of respiratory secretions and has been successfully used in this regard in patients with burns and inhalational injury (9). High-frequency Oscillation HFO is a mode of mechanical ventilation that delivers very small tidal volumes around a set mean airway pressure at high respiratory rates of 3 to 15 Hz (equivalent to breaths per minute, as 1 Hz = 60 breaths per minute) (10). HFO has been widely and effectively used in neonates and children for close to 20 years (11 15), but only recently has become available in the adult ICU; previous versions of this ventilator were only capable of generating sufficient flow so as to provide adequate CO 2 elimination in patients under 35 kg. In contrast to other HFV modes, humidification is less of an issue during HFO, as a continuous bias flow of humidified gas is passed in front of an oscillating membrane (Fig ). This oscillating membrane pushes the humidified gas into the patient and also provides active expiration, a factor that likely accounts for the lack of important gas trapping that is observed when HFO is employed at adequate airway pressures (16). The elegance of HFO is that it allows for decoupling of oxygenation and ventilation. Alveolar ventilation, and thus carbon dioxide elimination, are dependent on the frequency and tidal volume but are relatively independent of lung volume (17). In contrast, oxygenation is proportional to mean airway pressure and lung volume (18,19). 2029

2 2030 Section XIII: Respiratory Disorders O 2 Blender Humidifier Bias Flow OSCILLATING MEMBRANE Patient RESISTANCE VALVE FIGURE Schematic overview of the high-frequency oscillation (HFO) circuit. (Reused with permission from Ferguson ND, Stewart TE. New therapies for adults with acute lung injury: high-frequency oscillatory ventilation. Crit Care Clin. 2002;18: ) RATIONALE FOR THE USE OF HIGH-FREQUENCY VENTILATION IN ACUTE LUNG INJURY Ventilator-induced Lung Injury Mechanisms of Ventilator-induced Lung Injury VILI is histologically indistinguishable from ARDS; three decades of experimental research has shown it to occur through a number of mechanisms, including (a) overdistention injury (volutrauma) (20 28), (b) collapse reopening injury (atelectrauma) (1,2,20,29 34), and (c) oxygen toxicity (35 37). Each of these can lead, in turn, to further injury termed biotrauma, the release of inflammatory mediators that may worsen pulmonary injury and propagate systemically to harm distant organs (1,2,31,32,38 45). Numerous studies, performed using both small and large animals, consistently show that ventilatory high end-inspiratory stretch can cause a clinical and histologic picture similar to ARDS even in the absence of any other noxious stimulus (20 28). Patients with ARDS are at increased risk of regional lung overdistention because of the patchy nature of ARDS (46,47); the small areas of relatively normal lung (the so-called baby lung ) receive the bulk of the tidal volume and are at particular risk of volutrauma (48,49). Repeated opening and closing of alveolar units can also cause VILI. In the injured lung, alveolar damage and absolute, or qualitative, deficiencies in alveolar surfactant lead to alveolar instability and localized lung unit collapse. Through each respiratory cycle, these unstable alveoli undergo collapse and reopening, a process that generates injurious mechanical forces and causes further lung injury. There is a substantive body of animal evidence showing that efforts to limit lung unit closing on expiration by maintaining an adequate positive end-expiratory pressure (PEEP) are relatively protective against atelectrauma (1,2,20,30 34). Here, the paradigm is one of opening the lung and keeping it open, thereby avoiding cyclic collapse and recruitment/derecruitment (2,50). The major cause of death among ARDS patients is not refractory hypoxemia, as one might expect, but rather multiorgan dysfunction syndrome (MODS) (51). One hypothesis to explain the link between VILI and multiorgan failure is biotrauma (38,39) the release of inflammatory mediators in response to injurious ventilation; these mediators may enter the systemic circulation, and hence lead to the development of MODS. Inflammatory mediators, such as interleukins, tumour necrosis factor-α, and platelet-activating factor, are released in response to VILI, oxygen-free radicals, and alveolar shearing. These mediators perpetuate the cascade of lung inflammation and worsen lung injury (31,32). Additionally, injurious ventilation has been shown to increase these inflammatory mediators in the peripheral blood of patients with ARDS (44,52). Through a variety of mechanisms, this inflammatory injury can lead to nonpulmonary organ dysfunction (41 43). With an increased appreciation of the mechanisms of VILI, the next logical step is to employ ventilatory strategies that attempt to decrease overdistention, minimize shear injury, and limit oxygen toxicity (53 55). However, these aims may be competing; increasing PEEP may increase the risks of regional alveolar overdistention, while lowering tidal volumes can result in progressive alveolar collapse, a reduction in total lung volume, and higher oxygen requirements. The goals of mechanical ventilation in a patient at risk of VILI therefore should be to ventilate and oxygenate the patient while staying within a safe window, avoiding both overdistention and derecruitment collapse as illustrated in the volume pressure curve of the lung shown in Figure (56). Impact of Limiting Ventilator-induced Lung Injury with Conventional Ventilation In the mid-1990s, given the expanding animal data on VILI outlined above and in light of initial promising, but uncontrolled,

3 Chapter 133: High-frequency Ventilation: Lessons Learned and Future Directions 2031 Expiratory Portion Zone of Overdistention Volume Zone of Derecruitment and Atelectasis Safe Window Inspiratory Portion Pressure FIGURE Volume pressure curve of the injured lung. (Reused with permission from Froese AB. Highfrequency oscillatory ventilation for adult respiratory distress syndrome: let s get it right this time. Crit Care Med. 1997;25[6]: ) human studies of lung-protective ventilation (57,58), a call was made for randomized trials (53). Initial trials focused on avoiding volutrauma, comparing strategies that limited tidal volumes and inspiratory pressures with traditional strategies, both using conventional ventilation. Three smaller studies did not find mortality differences between these approaches (59 61). However, a large, methodologically rigorous, and adequately powered trial conducted by the National Institutes of Health (NIH) ARDS Network did show important differences in mortality (3). In this study, 861 patients were randomly assigned to receive a low-stretch strategy with a targeted tidal volume of 6 ml/kg predicted body weight (PBW) and a plateau pressure limit of 30 cm H 2 O, or to a higher-stretch strategy using a targeted tidal volume of 12 ml/kg PBW and a plateau pressure limit of up to 50 cm H 2 O. The low-stretch strategy was associated with a mortality reduction from 40% in the control group to 31% in the experimental group (relative risk [RR] 0.78; 95% confidence interval [CI] ). This trial clearly indicates that avoiding volutrauma saves lives in patients with acute lung injury. The trial has subsequently generated significant discussion regarding its mechanisms of benefit (62) and its choice of control strategy (63 66), but nevertheless, 6 ml/kg PBW has emerged as a standard for tidal volume limitation against which other strategies are compared (67,68). Another early randomized controlled trial (RCT) published in the late 1990s demonstrated dramatic reductions in mortality using a lung-protective strategy whose goal was to limit both volutrauma and atelectrauma using low tidal volumes and higher PEEP compared with traditional ventilation (69). Amato et al. found a statistically significant reduction in 28-day mortality (11/29 [38%] vs. 17/24 [71%] deaths; RR 0.53; 95% CI ) favoring patients exposed to the lung-protective strategy. In light of the subsequent positive ARDS Network trial noted above, interpretation of this trial is confounded by the use of both lower PEEP levels and higher tidal volumes and inspiratory pressures in the control group; the relative contribution of efforts to avoid cyclic collapse and reopening is unclear. Drawing on the promise of the Amato trial, three RCTs have now been completed analyzing the effects of higher versus lower levels of PEEP, while limiting tidal volumes in all study patients. Two of these the ExPress trial by Mercat et al. and the Lung Open Ventilation Study conducted by Meade et al. are very recently completed and not yet published; both were presented at the 2006 European Society of Intensive Care Meeting. The one fully published trial, the ALVEOLI study, was conducted by the NIH ARDS Network investigators (70) and was stopped early, which likely contributed to the large baseline imbalance in age favoring the lower PEEP group. When we consider the relative risk after adjusting for baseline imbalances in prognostic factors (including age) from this trial, along with the preliminary findings from the two recent trials, a consistent trend favoring higher PEEP begins to emerge. Therefore, while no single trial has definitively shown an incremental mortality benefit with higher PEEP and attempts to limit atelectrauma while already avoiding overdistention injury, when viewed together, these trials suggest that this benefit may well exist. They also suggest that a higher level of PEEP with or without other maneuvers to open the lung, along with limited tidal volumes and inspiratory pressures, may be considered a very reasonable comparison strategy in future ventilation trials; this approach may be superior, and there is no suggestion of harm compared with lower PEEP levels. These studies clearly demonstrated that ventilatory strategy is important in patients with ARDS, and that lung-protective strategies can minimize VILI and decrease mortality in humans with ARDS. As such, and given the proposed mechanisms of lung protection (Fig ), a strategy that minimizes overdistention and allows the use of high PEEP should be the ideal mode in patients with ARDS. This is where HFO may have great clinical benefit. Animal Studies Comparing Conventional Ventilation to High-frequency Oscillation in Acute Lung Injury The very small delivered tidal volumes are the key to the lungprotective potential of HFO. Because cyclic alveolar stretch is minimal, clinicians are able to set the mean airway pressure (mp AW ) on HFO significantly higher than they are able to set PEEP on conventional ventilation, thereby avoiding cyclic collapse and atelectrauma and yet they are still able to avoid very high peak inspiratory pressures and subsequent

4 2032 Section XIII: Respiratory Disorders PRESSURE CMV HFOV terms of lung protection and mortality (101,102). Similarly, when making decisions regarding the best methods for applying HFO, we must draw inferences from multiple alternate sources (including animal studies, neonatal RCTs, and adult case series), since large RCTs comparing different HFO strategies are unavailable and would be, frankly, difficult to justify ethically in the absence of demonstrable efficacy over conventional ventilation. TIME FIGURE Pressure time curve contrasts tidal variations in airway pressure associated with conventional ventilation (dashed line) and high-frequency oscillation (solid line). HFOV, high-frequency oscillation ventilation; CMV, conventional mechanical ventilation. (Reused with permission from Ferguson ND, Stewart TE. New therapies for adults with acute lung injury: high-frequency oscillatory ventilation. Crit Care Clin. 2002;18: ) volutrauma (Fig ). This should allow a larger margin of error and make it easier to stay within the safe window of lung protection (Fig ) (56). Since the first descriptions of HFO (71,72), numerous studies have examined the lung-protective potential of HFO using animal models (Table 133.1) (73 89). Perhaps not surprisingly, earlier studies that compared HFO with potentially injurious conventional ventilation uniformly found improvements in physiology, decreased inflammatory markers, and/or improved histology in animals treated with HFO (73 83). More relevant today, however, are recent studies that have compared HFO with lung-protective conventional ventilation, the majority of which continue to favor HFO in terms of physiology, inflammation, and pathology (Table 133.1) (86 89). Mechanisms of Gas Transport during High-frequency Oscillation Tidal volumes are extremely small with HFO, often smaller than the anatomic dead space, in the range of 1 to 2 ml/kg (90). In addition to relying on bulk flow, adequate CO 2 removal during HFO is achieved through a number of alternative mechanisms including convective streaming due to asymmetric velocity profiles, Pendelluft, cardiogenic mixing, and diffusion (Fig ). A full explanation of these physiologic principles is outside the scope of this chapter; the reader is referred to a number of papers reviewing this physiology in detail (5,91 99). The major message from these theoretical and experimental studies is that CO 2 elimination which is inversely proportional to PaCO 2 is described by the relationship f a V b T, where a = 1 and b = 2. This is important clinically in that it indicates that increases in frequency may lead to increased PaCO 2 if V T falls (see below). IMPLEMENTING AND OPTIMIZING HIGH-FREQUENCY OSCILLATION IN ADULTS Because of the lack of an appropriate surrogate end point that correlates with mortality (3,100), a large multicenter trial will be needed to definitively determine the relative effects of HFO compared with conventional mechanical ventilation (CMV) in Animal Studies Too many animal studies have been conducted on highfrequency ventilation to list them in this review (4). Rather, in Table 133.2, we outline a number of these studies that are informative when considering how best to employ HFO in adults (72, ). Several key messages for HFO can be gleaned from these studies, including the following: 1. At adequate mp AW, gas trapping is not an issue despite very high respiratory rates. 2. Increasing mp AW leads to improved oxygenation through lung recruitment. 3. Tidal pressure swings measured in the trachea can be increased with either underrecruitment or overdistention of the lung. 4. Recruitment maneuvers (sustained inflations) may be required to adequately recruit severely injured lungs. Taken together, and in keeping with the current understanding of VILI mechanisms, these studies suggest that HFO in adults should be employed using an open-lung strategy, facilitated by higher mean airway pressures and lung recruitment maneuvers (56,119,120). Neonatal Studies The longest clinical experience and the most rigorous evaluation of HFO have both been in the neonatal population ( ). The initial large RCT evaluating HFO in this setting raised safety concerns, but it later became evident that here, too, the safe and effective application of HFO requires lung volume recruitment (121,131,132). Subsequent RCTs of HFO with an open-lung approach have demonstrated HFO to be safe and effective in improving oxygenation and, as indicated by the current Cochrane systematic review, may reduce the risk of death or chronic lung disease (15). Interpreting the neonatal HFO literature is challenging due to differences in study populations (preterm vs. term), interventions (degree of lung recruitment targeted), the timing of HFO (immediately after birth vs. later), and by the introduction of exogenous surfactant as a standard therapy in the 1990s. In addition, it is important to realize that the baseline mortality rate in infant respiratory distress syndrome is an order of magnitude less than that seen in adults with severe ARDS. All of these factors mean that results from neonatal studies cannot be directly extrapolated to adult populations. Important lessons can be learned from the neonatal HFO literature, however, including the importance of thoroughly understanding the underlying physiologic mechanisms of the therapy, and the recognition that a learning curve may exist in the initiation of an HFO program (131,132).

5 TABLE ANIMAL STUDIES COMPARING VILI WITH HFO VS. CMV HFO effects on HFO effects on HFO effects on Author Animal Lung injury HFO CMV LRM at onset Duration physiology inflammation pathology HFO vs. potentially injurious (high volume/pressure) CMV Hamilton 1983 Rabbits Saline lavage mpaw = 15, f = 15 Hz Tamura 1985 Rabbits IV starch particle emboli f = 15 Hz, VT 1 2 ml/kg McCulloch 1988 Rabbits Saline lavage High EELV PEEP 8, PIP for CO2 Bond 1993 Rabbits Saline lavage mpaw = 15 Intermittent LRM Imai 1994 Rabbits Saline lavage mpaw = 15, f = 15 Hz Matsuoka 1994 Rabbits Saline lavage mpaw = 15, f = 15 Hz Sugiura 1994 Rabbits Saline lavage mpaw = 15 Hz Intermittent LRM Takata 1997 Rabbits Saline lavage mpaw = 13, f = 15 Hz Gommers 1999 Rabbits Saline lavage f = 10 Hz mpaw for PaO2 Kerr 2001 Rabbits NNMU lavage f = 15 Hz mpaw = PIP-7 Noda 2003 Rabbits Saline lavage mpaw = 15, f = 15 Hz HFO vs. lung-protective CMV Vazquez de Anda Rats Saline lavage f = 10 Hz 1999 mpaw for PaO2 Rimensberger 2000 Rabbits Saline lavage f = 15 Hz mpaw = CCP + 4 Rotta 2001 Rabbits Saline lavage mpaw = 16, f = 10 Hz Imai 2001 Rabbits Saline lavage mpaw = 15, f = 15 Hz Sedeek 2003 Sheep Saline lavage f = 8 Hz mpaw = PMC Von der Hardt 2004 Piglets Saline lavage mpaw = 18, f = 15 Hz PEEP 6, PIP 25 Both 5 20 h O2 Mortality in 20-h model PEEP 2, PIP 20 No 3 h O2 EVLW, protein leak PEEP 8 PIP for CO2 mpaw = 15, PIP 25, PEEP 5 Both 7 h O2, compliance Both 6 h O2, compliance Both 4 h O2, compliance PMN, PAF, TXB2 in BAL PEEP 5, PIP 25 Both 2 4 h O2 BAL PMN # and activation PEEP 5, PIP 23 Both 4 h O2, compliance BAL PMN # and activation Hyaline membranes Hyaline membranes Intra-alveolar cellularity Lung injury PEEP 5, PIP 28 Both 1 4 h O2, TNF-α mrna, Hyaline membranes compliance PMN in BAL PEEP 6, PIP 26 HFO only 5 h O2 Lung injury PEEP 4, VT = 10 ml/kg No 1 2 h O2, resp. acidosis LA surfactant conversion/loss PEEP 5, PIP 25 Both 4 h O2 TNF-α mrna, BAL PMN # PEEP for PaO2, VT for CO2 VT = 5 ml/kg PEEP > CCP PEEP > Pflex, VT = 6 ml/kg PEEP 10, VT = 6 ml/kg PIP <35; PEEP = PMC Both 3 h Similar O2 Similar BAL protein levels Both 4 h Similar O2 and compliance Both 4 h resp. acidosis, inotropes Both 4 h O2, CO2 compliance Similar MPO levels Similar histology score Similar, elastase in BAL TNF-α, PMN in BAL Both; repeated 6 h Similar O2 Similar to cytokine mrna PEEP 4, PIP 20 No 8 h O2 mrna of IL-8, IL-1β IL-6, etc. MDA in lung homogenate Alveolar path. score Lung injury Macrophage IL-8 gene expr. VILI, ventilator-induced lung injury; HFO, high-frequency oscillation; CMV, conventional mechanical ventilation; LRM, lung recruitment manoeuvre; mpaw, mean airway pressure; f, frequency; PEEP, positive end-expiratory pressure; PIP, peak inspiratory pressure on CMV; EVLW, extravascular lung water; VT, tidal volume; EELV, end-expiratory lung volume; PMN, polymorphonuclear cells; PAF, platelet-activating factor; TXB2, thromboxane B2; BAL, bronchoalveolar lavage; TNF, tumor necrosis factor; NNMU, N-nitroso-n-methylurethane; resp., respiratory; MDA, malondialdehyde (marker of lipid peroxidation); LA, large aggregate surfactant (i.e., functionally active surfactant); CCP, critical closing pressure (i.e., pressure on deflation limb of volume pressure curve corresponding to 50% of total lung capacity); MPO, myeloperoxidase; PMC, point of maximal curvature (on the deflation limb of the volume pressure curve); IL, interleukin; gene expr., gene expression; exog surf., exogenous surfactant. Study selection criteria: Population: Animal models (excluding preterm animals) of acute lung injury; intervention: HFO vs. conventional ventilation, no additional cointerventions; outcome: any marker(s) of lung injury (not gas exchange alone); therefore, not crossover studies. 2033

6 2034 Section XIII: Respiratory Disorders Direct ventilation of close alveoli Turbulence Convection Pendelluft Turbulent flow and radial mixing Asymmetric inspiratory and expiratory velocity profiles Velocity profile on inspiration Velocity profile on expiration Convection and diffusion Laminar flow and radial mixing Diffusion Diffusion Collateral ventilation FIGURE Alternative mechanisms of gas exchange with high-frequency oscillation (HFO). (Reused with permission from Slutsky AS, Drazen JM. Ventilation with small tidal volumes. N Engl J Med. 2002;347[9]: ) Uncontrolled Adult Studies The studies reporting the clinical experience with HFO in adults, which are summarized in Table ( ), are also informative for optimizing an HFO protocol. Almost all of them studied HFO as rescue therapy for patients with extremely severe disease who were failing conventional ventilation. Despite being limited by their uncontrolled nature and by selection bias, consistent messages that arise from these rescue series include: 1. HFO is usually effective in improving oxygenation. 2. HFO appears safe, with no obvious increased rates of complications. 3. Both baseline oxygenation index and duration of CMV prior to HFO are associated with increased mortality. One prospective, multicentered study from our group is unique in that it tested an explicit protocol for applying HFO early in the course of ARDS, incorporating lung recruitment maneuvers (RMs) and a descending titration of mean airway pressure to optimize lung volumes for gas exchange and lung protection (140). RMs sustained inflation maneuvers with 30 to 40 cm H 2 O pressures for 30 to 40 seconds have been found to be safe in adults on CMV, but studies have shown mixed results in terms of efficacy and duration of their oxygenation effects ( ). Because of the small tidal volumes generated with HFO, there is very little tidal recruitment of the lung, creating a more compelling rationale for RMs during HFO compared with CMV (19,50,102,153,154). The main finding from a physiologic standpoint in our study was that the combination of HFO and RMs was well tolerated and resulted in rapid and sustained lung recruitment (140). Furthermore, the explicit HFO protocol appeared feasible; adherence was excellent, and the method for weaning HFO and transitioning to CMV appeared practical and safe. Tidal volume during HFO has always been assumed to be very low, but it is not routinely measured with currently available oscillators. Tidal volume is known to be inversely related to frequency due to a decreasing inspiratory time with increasing frequency (17), and in adults, typical frequencies have been significantly lower than those used in neonates (3 6 vs Hz) (132,139). Sedeek et al. explored this issue in a lunginjured sheep model, measuring tidal volumes of up to 4 ml/kg at high-pressure amplitudes and low frequencies (Table 133.2) (118). These tidal volume measurements may be overestimated because of technical reasons (155), but they do highlight the importance of efforts to minimize delivered tidal volumes with adult HFO, and they have spurred further investigation. Tidal volumes on HFO measured in adults using a very accurate device (156) were small 1.1 to 2.5 ml/kg and strongly (text continues on page 2038)

7 TABLE ANIMAL STUDIES INFORMING OPTIMAL HFO IMPLEMENTATION Author Animal Lung injury Intervention Design Outcomes Main findings Comment HFO proof of concept and basic settings Bohn 1980 Dogs None HFO at various frequencies Single cohort Gas exchange and airway pressures First description of HFO showing safety and efficacy Bryan 1986 Rabbits Saline lavage HFO at various mpaw Single cohort Occlusion pressures Gas trapping with HFO only seen at extremely low Bancalari 1987 Rabbits Meconium aspiration HFO (mpaw = 11 22; f = Hz) vs. HFJV Courtney 1992 Rabbits Saline lavage HFO at various I:E ratios (constant mpaw, P and f ) Crossover Gas exchange, lung volumes (plethys.) Single cohort Gas exchange and hemodynamics Pillow 1999 Rabbits None HFO at various I:E ratios Single cohort Airway alveolar pressure difference mpaw No significant gas trapping with HFO; yes with HFJV Varying I:E ratios had no significant effect on PaO2, PaCO2, BP, or CO PALV was lower than mpaw at I:E ratios below 1:1 HFO capable of supporting adequate gas exchange Gas trapping not an issue with HFO despite high frequencies Gas trapping not an issue with HFO despite high frequencies At constant mpaw, I:E ratio does not have a large effect on gas exchange With I:E of 1:2, alveolar pressures may be somewhat lower than set mpaw HFO volume-pressure and gas-exchange relationships Boynton 1991 Rabbits Saline lavage HFO at various mpaw Single cohort Gas exchange, lung volumes (plethys.) Goddon 2001 Sheep Saline lavage HFO at various mpaw Single cohort Gas exchange and mechanics van Genderingen 2002 (ICM) Dembinski 2002 Pigs Saline lavage HFO with stepwise increases in mpaw Pigs Oleic acid infusion HFO vs. CMV titrated to PaO2 Single cohort Oxygenation index and shunt fraction RCT Airway pressures and shunt fraction Luecke 2003 Pigs Saline lavage HFO at various mpaw Single cohort Gas exchange and mechanics mpaw leads to O2 through lung volume; hysteresis present Best O2 was at mpaw of PFLEX + 6 also equal to PMC on deflation mpaw with OI was close to shunt; large hysteresis mpaw needed with HFO to match PaO2 on CMV Increasing mpaw mirrored volume on deflation limb of pressure volume curve; no hysteresis on HFO Oxygenation response is a reasonable surrogate for recruitment Oxygenation response is a reasonable surrogate for recruitment Oxygenation response is a reasonable surrogate for recruitment High mpaw needed for lung opening on HFO may be reduced subsequently Lung opening may occur on HFO without RMs (Continued ) 2035

8 TABLE (CONTINUED) Author Animal Lung injury Intervention Design Outcomes Main findings Comment Intrathoracic tidal pressure swings and lung recruitment during HFO Sakai 1999 Rabbits Saline lavage HFO at various mpaw and VT van Genderingen 2002 Single cohort Pressure swings in airway and pleura Pigs Saline lavage HFO at various mpaw Single cohort Oscillatory pressure ratio (OPR) a At low mpaw pressure swings are increased; decreased after an RM OPR is increased when the lung is overdistended OR underrecruited Underrecruited lungs lead to higher pressure swings on HFO Underrecruited or overdistended lungs lead to higher pressure swings on HFO Recruitment maneuver effects with HFO Byford 1988 Rabbits Saline lavage + CMV HFO + RMs with oscillator on and off Sznajder 1988 Dogs Oleic acid HFO vs. high tidal volume CMV Crossover Gas exchange, lung volumes (plethys.) Crossover Gas exchange, lung volumes (plethys.) Walsh 1988 Rabbits Saline lavage HFO + various RMs Single cohort Gas exchange and mechanics Suzuki 1992 Rabbits Saline lavage + CMV HFO ± RMs Single cohort Lung volume and gas exchange Delivered tidal volumes with adult settings on HFO Sedeek 2003 Sheep Saline lavage HFO at various P and f Single cohort Delivered tidal volumes RMs with oscillator on yielded recruitment at pressures Full lung recruitment on HFO only achieved after a RM to TLC RMs improve gas exchange and compliance on HFO HFO without RMs was uniformly fatal; HFO + RMs improved gas exchange At very low f and high P, VT may not be negligible on HFO RMs with oscillator not paused may be more effective RMs on HFO needed for lung recruitment in severe lung injury model Lower pressures possible with HFO after RMs RMs on HFO needed for lung recruitment in severe lung injury model Need to target lowest VT possible with HFO (use highest tolerated frequency) a Oscillatory pressure ratio is the ratio of pressure swings at the proximal and distal end of the endotracheal tube during HFO. HFO, high-frequency oscillation; mpaw, mean airway pressure; f, frequency; HFJV, high-frequency jet ventilation; plethys., plethysmography; I:E, inspiratory:expiratory; BP, blood pressure; CO, cardiac output; PALV, alveolar pressure; PFLEX, pressure at the lower inflexion point of the static volume pressure curve; OI, oxygen index; RCT, randomized controlled trial; PMC, point of maximal curvature on the deflation limb of the static volume pressure curve; CMV, conventional mechanical ventilation; VT, tidal volume; RM, recruitment maneuver. 2036

9 TABLE UNCONTROLLED STUDIES HFO FOR ADULT ARDS Baseline severity Author N Population of illness Design Complications Main findings Comment HFO as rescue therapy Fort ARDS OI = 49 APACHE II = 23 Claridge ARDS, trauma P/F = 52 APACHE II = 28 Mehta ARDS OI = 33 APACHE II = 22 Prospective Barotrauma 6% ETT obstruction 6% Improved oxygenation; mortality 53% Not reported Not reported Improved oxygenation; mortality 20% Prospective Barotrauma 8% ETT obstruction 4% Improved oxygenation; mortality 67% Andersen 16 ARDS OI = 28 Retrospective Barotrauma 6% Improved oxygenation; 2002 APACHE II = 27 mortality 31% David ARDS OI = 23 Prospective Barotrauma 2% Improved oxygenation; APACHE II = 28 mortality 43% Cartotto 25 ARDS, burns OI = 27 Retrospective Barotrauma 0% Improved oxygenation; 2004 APACHE II = 16 mortality 28% Mehta ARDS OI = 31 Retrospective Barotrauma 21% Improved oxygenation; APACHE II = 24 mortality 62% Pachl ARDS Not reported Prospective Not reported Not reported Finkielman ARDS OI = 35 APACHE II = 35 Weiler ARDS OI = 28 APACHE II = 25 Retrospective Barotrauma 0% Improved oxygenation; mortality 57% Retrospective Barotrauma 0% Improved oxygenation; mortality not reported Baseline OI and duration of CMV associated with mortality HFO appears safe and improves O2 in rescue setting HFO appears safe and improves O2 in rescue setting HFO appears safe and improves O2 in rescue setting HFO appears safe and improves O2 in rescue setting HFO appears safe and improves O2 in rescue setting Baseline OI and duration of CMV associated with mortality HFO appears safe and improves O2 in rescue setting HFO appears safe and improves O2 in rescue setting HFO for lung protection Ferguson 25 Early ARDS OI = APACHE II = 24 Prospective, consecutive, protocolized Barotrauma 8% 20% Rapidly improved oxygenation; mortality 44%; excellent protocol adherence Pilot study of an explicit protocol including conversion criteria to CMV HFO, high-frequency oscillation; ARDS, acute respiratory distress syndrome; OI, oxygen index; ETT, endotracheal tube; CMV, conventional mechanical ventilation. 2037

10 2038 Section XIII: Respiratory Disorders influenced by frequency (90). Cumulatively, these data are reassuring in that HFO in adults can be set to deliver very small tidal volumes, but at the same time they highlight the potential importance of targeting the lowest tidal volume possible (155). This could be achieved by increasing frequency as high as tolerated while avoiding severe respiratory acidosis, a strategy that has been proposed as physiologically sensible (157) and recently shown to be feasible in most patients (158). Randomized Controlled Trials of High-frequency Oscillation in Adults To date, only two RCTs have compared HFO with CMV in adults (101,159). Both of these trials were planned and started prior to the completion of the first ARDS Network study that showed benefit from strict control of tidal volumes (3). Neither RCT demonstrated safety concerns (the primary outcome); rates of barotrauma and other complications were similar between groups in both studies. The larger of the two RCTs shows an impressive trend toward a mortality benefit with HFO (RR 0.72; 95% CI ) despite more than 10% of the control group crossing over to HFO (101). The second trial, which began accrual at the same time, was stopped because of slow enrollment. In fact, the enrollment rates per center per month were identical in both studies; the latter study simply highlights the need for extensive multicentered collaboration for a successful HFO RCT. Mortality results from the smaller study are less encouraging, but these are significantly confounded by large baseline differences favoring the control group, and by an almost 20% crossover rate. Nevertheless, pooled results from these trials still suggest a potentially important survival benefit with HFO (RR 0.899, 95% CI ; random effects model). Inferences from these RCTs are limited, however, by their methodology, small sample sizes, and, importantly, by use of now dated, potentially injurious conventional ventilation strategies. FUTURE DIRECTIONS A research focus on the ventilatory care of ARDS patients has, to date, paid significantly greater dividends than the disappointing results of pharmacotherapy testing (160). Following the completion of the three conventional ventilation RCTs comparing a lung open approach with a lower PEEP approach, we believe that the study of HFO represents the next logical and important question to be addressed in ARDS research. We ground this belief in: 1. A strong physiologic rationale and database of animal studies 2. Experience from the neonatal and pediatric arenas 3. An expanding clinical experience with adult HFO 4. Promising results from small nondefinitive RCTs Despite the very strong physiologic rationale and the encouraging clinical data to date, there are potential detrimental consequences of HFO. First, the physiologic benefit of HFO is likely derived from recruitment of the lung with higher mean airway pressures, while still not overdistending alveoli because of the very small tidal volumes. Hence, patients with mild disease and minimal collapsed or consolidated lung may not be good candidates for HFO (151). Second, the high mean airway pressures may negatively impact hemodynamics. Third, because the bias flow rate is insufficient to meet the inspiratory flow demands of adults in respiratory distress, all adults on HFO must have their respiratory efforts suppressed with intravenous sedation. This means that the majority of adults will be heavily sedated, and many may need transient neuromuscular blockade. Due to these concerns, and to target a population in need of recruitment, we believe that future studies should enroll patients with severe ARDS who are likely to have significant lung collapse and who frequently require significant amounts of sedation, with or without paralysis, on conventional ventilators. In summary, we believe that HFO and other forms of highfrequency ventilation should currently be reserved for rescue therapy in patients who are failing conventional ventilation. We do not believe that HFO can be recommended for routine use in adults with ARDS at the current time because: There are potential detrimental physiologic and clinical consequences as described above This would represent premature dissemination of a complex technology without rigorous evaluation of its risks, benefits, and indications A definitive RCT to establish the impact of HFO versus best current conventional ventilation on mortality is needed. References 1. Pinhu L, Whitehead T, Evans T, et al. Ventilator-associated lung injury. Lancet. 2003;361: Tremblay LN, Slutsky AS. Ventilator-induced lung injury: from the bench to the bedside. Intensive Care Med. 2006;32(1): 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(18): Froese AB, Bryan AC. High frequency ventilation. Am Rev Resp Dis. 1987;135(6): Drazen JM, Kamm RD, Slutsky AS. High-frequency ventilation. Physiol Rev. 1984;64(2): Klain M, Smith RB. High frequency percutaneous transtracheal jet ventilation. Crit Care Med. 1977;5(6): Roth MD, Wright JW, Bellamy PE. Gas flow through a bronchopleural fistula. Measuring the effects of high-frequency jet ventilation and chesttube suction. Chest. 1988;93(1): Bishop MJ, Benson MS, Sato P, et al. Comparison of high-frequency jet ventilation with conventional mechanical ventilation for bronchopleural fistula. Anesth Analg. 1987;66(9): Salim A, Martin M. High-frequency percussive ventilation. Crit Care Med. 2005;33(3 Suppl):S241 S Chan KP, Stewart TE. Clinical use of high-frequency oscillatory ventilation in adult patients with acute respiratory distress syndrome. Crit Care Med. 2005;33(3 Suppl):S170 S Froese AB, Butler PO, Fletcher WA, et al. High frequency oscillatory ventilation in premature infants with respiratory failure: a preliminary report. Anesth Analg. 1987;66: Bryan AC. The Oscillations of HFO. Am J Resp Crit Care Med. 2001;163: Bollen CW, Uiterwaal CS, van Vught AJ. Cumulative metaanalysis of highfrequency versus conventional ventilation in premature neonates. Am J Respir Crit Care Med. 2003;168(10): Arnold JH, Truog WE, Thompson JE, et al. High-frequency oscillatory ventilation in pediatric respiratory failure. Crit Care Med. 1993;21: Henderson-Smart DJ. Elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants. Cochrane Database Syst Rev. 2006; Bryan AC, Slutsky AS. Lung volume during high frequency oscillation. Am Rev Resp Dis. 1986;133(5):

11 Chapter 133: High-frequency Ventilation: Lessons Learned and Future Directions Slutsky AS, Kamm RD, Rossing TH, et al. Effects of frequency, tidal volume, and lung volume on CO2 elimination in dogs by high frequency (2 30 Hz), low tidal volume ventilation. J Clin Invest. 1981;68(6): Suzuki H, Papazoglou K, Bryan AC. Relationship between PaO 2 and lung volume during high frequency oscillatory ventilation. Acta Paediatr Jpn. 1992;34(5): Kolton M, Cattran CB, Kent G, et al. Oxygenation during high-frequency ventilation compared with conventional mechanical ventilation in two models of lung injury. Anesth Analg. 1982;61(4): 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 Resp Dis. 1974;110(5): Dreyfuss D, Soler P, Basset G, et al. High inflation pressure pulmonary edema. Respective effects of high airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Resp Dis. 1988;137(5): Parker JC, Hernandez LA, Longenecker GL, et al. Lung edema caused by high peak inspiratory pressures in dogs. Role of increased microvascular filtration pressure and permeability. Am Rev Resp Dis. 1990;142(2): Dreyfuss D, Basset G, Soler P, et al. Intermittent positive-pressure hyperventilation with high inflation pressures produces pulmonary microvascular injury in rats. Am Rev Resp Dis. 1985;132(4): Kolobow T, Moretti MP, Fumagalli R, et al. Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation. An experimental study. Am Rev Resp Dis. 1987;135(2): Tsuno K, Miura K, Takeya M, et al. Histopathologic pulmonary changes from mechanical ventilation at high peak airway pressures. Am Rev Resp Dis. 1991;143(5 Pt 1): Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am J Resp Crit Care Med. 1998;157: Hernandez LA, Peevy KJ, Moise AA, et al. Chest wall restriction limits high airway pressure-induced lung injury in young rabbits. J Appl Physiol. 1989;66(5): Steinberg J, Schiller HJ, Halter JM, et al. Tidal volume increases do not affect alveolar mechanics in normal lung but cause alveolar overdistension and exacerbate alveolar instability after surfactant deactivation. Crit Care Med. 2002;30(12): Slutsky AS. Lung injury caused by mechanical ventilation. Chest. 1999;116: 9S 15S. 30. Muscedere JG, Mullen JB, Gan K, et al. Tidal ventilation at low airway pressures can augment lung injury. Am J Resp Crit Care Med. 1994;149 (5): 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(5): Tremblay L, Govindarajan A, Veldhuizen R, et al. TNFa levels are both time and ventilation strategy dependent in ex vivo rat lungs. Am J Resp Crit Care Med. 1998;157(3):A Steinberg JM, Schiller HJ, Halter JM, et al. Alveolar instability causes early ventilator-induced lung injury independent of neutrophils. Am J Resp Crit Care Med. 2004;169(1): Dos Santos CC, Slutsky AS. Cellular responses to mechanical stress: invited review: mechanisms of ventilator-induced lung injury: a perspective. J Appl Physiol. 2000;89(4): Bryan CL, Jenkinson SG. Oxygen toxicity. Clin Chest Med. 1988;9(1): Nash G, Blennerhassett JB, Pontoppidan H. Pulmonary lesions associated with oxygen therapy and artificial ventilation. N Engl J Med. 1967; 276(7): Davis WB, Rennard SI, Bitterman PB, et al. Pulmonary oxygen toxicity. Early reversible changes in human alveolar structures induced by hyperoxia. N Engl J Med. 1983;309(15): Tremblay LN, Slutsky AS. Ventilation-induced lung injury: from barotrauma to biotrauma. Proc Assoc Am Phys. 1998;110: Slutsky AS, Tremblay LN. Multiple system organ failure. Is mechanical ventilation a contributing factor? Am J Resp Crit Care Med. 1998;157: 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(1): Guery BP, Welsh DA, Viget NB, et al. Ventilation-induced lung injury is associated with an increase in gut permeability. Shock. 2003;19(6): Choi WI, Quinn DA, Park KM, et al. Systemic microvascular leak in an in vivo rat model of ventilator-induced lung injury. Am J Respir Crit Care Med. 2003;167(12): Imai Y, Parodo J, Kajikawa O, et al. Injurious mechanical ventilation and end-organ epithelial cell apoptosis and organ dysfunction in an experimental model of acute respiratory distress syndrome. JAMA. 2003;289: Ranieri VM, Suter PM, Tortorella C, et al. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. JAMA. 1999;282(1): Dos Santos CC, Slutsky AS. The contribution of biophysical lung injury to the development of biotrauma. Annu Rev Physiol. 2006;68: Gattinoni L, Pelosi P, Vitale G, et al. Body position changes redistribute lung computed-tomographic density in patients with acute respiratory failure. Anesthesiology. 1991;74(1): Gattinoni L, Caironi P, Pelosi P, et al. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med. 2001;164(9): Roupie E, Dambrosio M, Servillo G, et al. Titration of tidal volume and induced hypercapnia in acute respiratory distress syndrome. Am J Resp Crit Care Med. 1995;152: Gattinoni L, Pesenti A. The concept of baby lung. Intensive Care Med. 2005;31(6): Lachmann B. Open up the lung and keep it open. Intensive Care Med. 1992;18: Montgomery AB, Stager MA, Carrico CJ, et al. Causes of mortality in patients with the adult respiratory distress syndrome. Am Rev Resp Dis. 1985;132(3): Parsons PE, Eisner MD, Thompson BT, et al. Lower tidal volume ventilation and plasma cytokine markers of inflammation in patients with acute lung injury. Crit Care Med. 2005;33(1): Slutsky AS. Mechanical ventilation. American College of Chest Physicians Consensus Conference. Chest. 1993;104(6): Stewart TE, Slutsky AS. Mechanical ventilation: a shifting philosophy. Curr Sci. 1995;1: Stewart TE. Lung protection during mechanical ventilation. Ontario Thorac Rev. 1997;9: Froese AB. High-frequency oscillatory ventilation for adult respiratory distress syndrome: let s get it right this time. Crit Care Med. 1997;25(6): Hickling KG, Henderson SJ, Jackson R. Low mortality associated with low volume pressure limited ventilation with permissive hypercapnia in severe adult respiratory distress syndrome. Intensive Care Med. 1990;16(6): Hickling KG, Walsh J, Henderson S, et al. Low mortality rate in adult respiratory distress syndrome using low-volume, pressure-limited ventilation with permissive hypercapnia: a prospective study. Crit Care Med. 1994;22(10): 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(6): 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 Resp Crit Care Med. 1998;158(6): 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(8): de Durante G, del Turco M, Rustichini L, et al. ARDSNet lower tidal volume ventilatory strategy may generate intrinsic positive end-expiratory pressure in patients with acute respiratory distress syndrome. Am J Resp Crit Care Med. 2002;165: Eichacker PQ, Gerstenberger EP, Banks SM, et al. Meta-analysis of acute lung injury and acute respiratory distress syndrome trials testing low tidal volumes. Am J Resp Crit Care Med. 2002;166: Brower RG, Matthay M, Schoenfeld D. Meta-analysis of acute lung injury and acute respiratory distress syndrome trials. Am J Resp Crit Care Med. 2002;166: Steinbrook R. How best to ventilate? Trial design and patient safety in studies of the acute respiratory distress syndrome. N Engl J Med. 2003;348: Drazen JM. Controlling research trials. N Engl J Med. 2003;348: Tobin MJ. Culmination of an era in research on the acute respiratory distress syndrome. N Engl J Med. 2000;342(18): Stewart TE. Controversies around lung protective ventilation. Am J Resp Crit Care Med. 2002;166: Amato MB, Barbas CS, Medeiros DM, et al. Effect of a protectiveventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med. 1998;338(6): Brower RG, Lanken PN, MacIntyre N, et al. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome [see comment]. N Engl J Med. 2004;351(4): Lunkenheimer PP, Rafflenbeul W, Keller H, et al. Application of transtracheal pressure oscillations as a modification of diffusion respiration. Br J Anaesth. 1972;44: Bohn DJ, Miyasaka K, Marchak BE, et al. Ventilation by high-frequency oscillation. J Appl Physiol. 1980;48(4): Hamilton PP, Onayemi A, Smyth JA, et al. Comparison of conventional and high-frequency oscillatory ventilation: oxygenation and lung pathology. J Appl Physiol. 1983;55: Tamura M, Kawano T, Fitz-James I, et al. High-frequency oscillatory