High-frequency ventilation in adults

INTRODUCTION — High-frequency ventilation (HFV) is a form of mechanical ventilation that combines very high respiratory rates (>60 breaths per minute) with tidal volumes that are smaller than the volume of anatomic dead space [1]. HFV is not a first-line mode of mechanical ventilation and is used rarely; its use should be coordinated by experienced clinicians only.

This topic review describes the different types of HFV, as well as patient selection, efficacy, and potential harms. Alternative modes of mechanical ventilation are described separately. (See "Modes of mechanical ventilation".)

TYPES OF HFV — There are four basic types of HFV: high-frequency jet ventilation, high-frequency oscillatory ventilation, high-frequency percussive ventilation, and high-frequency positive pressure ventilation (figure 1). Among these high-frequency oscillatory ventilation is the mode that is used most often. None of these modes are first line for the management of acute respiratory distress syndrome. (See 'Patient selection' below and 'Efficacy' below.)

High-frequency jet ventilation — High-frequency jet ventilation (HFJV) refers to HFV delivered using a jet of gas (figure 1). It is initiated by inserting into the lumen of the endotracheal tube a small (14 to 16 gauge) cannula, which is connected to a specialized ventilator. An initial pressure of approximately 35 pounds per square inch (psi) drives the jet of gas from the cannula with an initial respiratory rate of 100 to 150 breaths per minute and an inspiratory fraction less than 40 percent. The inspiratory fraction is the inspiratory time divided by the sum of the inspiratory and expiratory times. Applied positive end-expiratory pressure (PEEP) can be added if needed.

An arterial blood gas should be measured approximately 15 minutes after the initiation of HFJV:

●Appropriate adjustments when the arterial carbon dioxide tension (PaCO₂) is elevated include: increasing the driving pressure in 5 psi increments to a maximum of 50 psi, increasing the inspiratory fraction in 5 percent increments to a maximum of 40 percent, increasing the frequency in 10 breaths per minute increments to a maximum of 250 breaths per minute, or adding an another mode of mechanical ventilation (see "Modes of mechanical ventilation")

●Appropriate adjustments when the PaCO₂ is low include: decreasing the driving pressure in 5 psi decrements, decreasing the inspiratory fraction in 5 percent decrements to a minimum of 20 percent, or decreasing the frequency in 10 breaths per minute decrements to a minimum of 100 breaths per minute

●Appropriate adjustments when the arterial oxygen tension (PaO₂) is low include: adding applied PEEP in 3 to 5 cm H₂O increments, increasing the driving pressure by 5 psi increments to a maximum of 50 psi, or increasing the inspiratory fraction in 5 percent increments to a maximum of 40 percent

●Appropriate adjustments when the PaO₂ is high include: decreasing the fraction of inspired oxygen (FiO₂) or decreasing applied PEEP

A respiratory rate of approximately 150 breaths per minute is generally required during HFJV. Use of ultrahigh-frequency jet ventilation (180 to 400 breaths per minute) in patients with adult respiratory distress syndrome (ARDS) has been reported, although the study had numerous important limitations [2]. HFJV always requires sedation and usually requires pharmacologic paralysis also. (See "Sedative-analgesia in ventilated adults: Management strategies, agent selection, monitoring, and withdrawal" and "Neuromuscular blocking agents in critically ill patients: Use, agent selection, administration, and adverse effects".)

High-frequency oscillatory ventilation — High-frequency oscillatory ventilation (HFOV) uses an oscillatory pump to deliver a respiratory rate of 3 to 15 Hertz (up to 900 breaths per minute) through the endotracheal tube (figure 1). This rate is so fast that the airway pressure merely oscillates around a constant mean airway pressure. The respiratory rate (in Hertz) is set directly by the clinician. The mean airway pressure is set by adjusting the inspiratory flow rate and an expiratory back pressure valve (similar to applied PEEP) [3]. Some pumps allow the mean airway pressure to be set directly.

The constant mean airway pressure maintains alveolar recruitment, avoids low end-expiratory pressures, and avoids high peak airway pressures. It also impacts oxygenation. Specifically, a higher mean airway pressure is associated with better oxygenation. HFOV induces a higher mean airway pressure than most volume-or pressure-controlled modes of mechanical ventilation.

The tidal volume (also called amplitude) is small during HFOV, usually less than or equal to the anatomic dead space. The amplitude depends on the endotracheal tube size and respiratory frequency: a smaller amplitude results when the endotracheal tube is small or the respiratory frequency is high [4].

HFOV at a respiratory rate greater than 6 Hz may be required to reduce barotrauma because the usual respiratory rate of 3 to 6 Hz results in airway pressures that are potentially not lung protective. The feasibility of this approach was demonstrated by a single center, prospective cohort study of 30 patients with ARDS who were receiving HFOV after failing conventional lung protective ventilation [5]. Among patients whose respiratory rates exceeded 6 Hz (range 6 to 15 Hz), most were able to meet their oxygenation (PaO₂ 55 to 80 mmHg) and ventilatory goals (pH 7.25 to 7.35).

Tracheal gas insufflation (TGI) during HFOV may improve gas exchange. This was demonstrated by a single-center, randomized, cross-over trial of patients with early ARDS, in which TGI to a mean tracheal pressure 3 cm H₂O above mean airway pressure (Paw) increased the PaO₂/FiO₂ ratio and decreased the PaCO₂ compared to HFOV alone [6].

Step-wise increases in the Paw may be superior to other techniques of lung volume recruitment during HFOV, according to one animal study. The study found that an incremental increase in the Paw over six minutes (until it was 12 cm H₂O greater than the Paw achieved with conventional mechanical ventilation) improved lung volume, oxygenation, and ventilation to a greater degree than other techniques [7].

High-frequency percussive ventilation — High-frequency percussive ventilation (HFPV) combines HFV plus time cycled, pressure-limited controlled mechanical ventilation (ie, pressure control ventilation [PCV]). It can be conceptualized as HFOV oscillating around two different pressure levels, the inspiratory and expiratory airway pressures [8]. HFPV improves oxygenation, improves ventilation, and lowers airway pressures (peak, mean, and end-expiratory), compared to other modes of mechanical ventilation. (See "Modes of mechanical ventilation", section on 'Pressure-limited ventilation'.)

HFPV is possible because of a device called a phasitron. The phasitron is an inspiratory and expiratory valve located at the end of the endotracheal tube. High-pressure gas drives the phasitron to deliver small tidal volumes at a high frequency (200 to 900 beats per min), superimposed on the inspiratory and expiratory airway pressures of PCV. The PCV is typically delivered at a respiratory rate of 10 to 15 breaths per min.

HFPV does not require pharmacologic paralysis. In addition, it clears secretions more effectively than other types of HFV [8].

High-frequency positive pressure ventilation — High-frequency positive pressure ventilation (HFPPV) is rarely used anymore, having been displaced by the types of HFV discussed above. HFPPV is delivered through the endotracheal tube using a conventional ventilator whose frequency is set near its upper limits (figure 1).

PATIENT SELECTION — There are no universally accepted indications for HFV. Its use has been described in a variety of clinical situations, including acute respiratory distress syndrome (ARDS), bronchopleural fistula, inhalational injury, blunt trauma induced ARDS, and head injuries complicated by high intracranial pressure [8-11].

●ARDS – The theoretical benefit of using HFV in patients with ARDS relates to the small tidal volumes. A strategy of low tidal volume ventilation has been proven in randomized trials to improve mortality, possibly due to decreased alveolar distension and ventilator-associated lung injury. Although the trials did not use HFV, many clinicians suspect that HFV confers a similar benefit. Until this is proven, HFV should not be considered routine care for patients with ARDS. HFV is used by some clinicians when there is persistent hypoxemia during the first three days of mechanical ventilation despite maximal conventional therapy, although the data to support this are limited [12,13]. (See "Acute respiratory distress syndrome: Ventilator management strategies for adults", section on 'Low tidal volume ventilation: Initial settings' and "Ventilator-induced lung injury".)

●Severe airway compromise and stridor – The use of transtracheal HFJV in the management of patients with severe airway compromise has been described as a less morbid alternative to operative tracheostomy. In one study, the transtracheal catheter was placed with relative ease (84 percent on the first attempt) and adequate ventilation was achieved in 96 percent of patients [14]. Only minor complications were reported, including kinking of the catheter, bleeding, and surgical emphysema. There was no increased morbidity and no mortality.

●Bronchopleural fistula – HFJV is approved by the United States Food and Drug Administration for ventilating patients in whom a large and persistent bronchopleural fistula exists. However, the likelihood that HFJV will allow the bronchopleural fistula to close is unpredictable [9,10]. While HFJV may promote fistula closure by limiting alveolar distension, this may be outweighed in some patients by increased plateau airway pressure (alveolar pressure), decreased oxygenation, or worse hypercapnia [10]. (See "Management of persistent air leaks in patients on mechanical ventilation".)

HFV should be avoided in patients with obstructive lung disease. This is because the high respiratory rate used for HFV shortens the expiratory time, which can cause auto-positive end-expiratory pressure (PEEP) and related sequelae which can induce barotrauma in patients with obstructive lung disease. (See "Positive end-expiratory pressure (PEEP)", section on 'Auto (intrinsic) PEEP'.)

EFFICACY — This section describes the clinical evidence related to the different types of HFV. Generally speaking, there is evidence that high-frequency oscillatory ventilation (HFOV) and high-frequency percussive ventilation (HFPV) improve oxygenation, although neither has been shown conclusively to improve clinical outcomes (eg, mortality, duration of mechanical ventilation, or length of intensive care unit [ICU] stay).

HF jet ventilation — There is little moderate or high quality data evaluating the efficacy of high-frequency jet ventilation (HFJV) in adults. One trial randomly assigned a heterogeneous group of 309 patients with acute respiratory failure to receive HFJV or volume-limited mechanical ventilation [15]. There was no significant difference in mortality or the duration of ICU stay. (See "Modes of mechanical ventilation", section on 'Volume-limited ventilation'.)

In another trial, seven patients who were already receiving a traditional mode of mechanical ventilation for respiratory failure complicated by a bronchopleural fistula were randomly assigned to either switch to HFJV or continue their mode of ventilation [10]. There was no significant difference in the size of the chest tube leak (a measure of the bronchopleural fistula), but the HFJV group developed worse oxygenation and hypercapnia after switching to the HFJV.

HF oscillatory ventilation — Most studies of high-frequency oscillatory ventilation (HFOV) have been performed in adults with acute respiratory distress syndrome (ARDS) [16-22]. Despite evidence that HFOV improves oxygenation, it does not reduce, and may increase, in-hospital mortality compared with a ventilation strategy of low tidal volume and high positive end-expiratory pressure (ARDSNet strategy). Thus, we and others do not recommend HFOV as a routine ventilator strategy in patients with ARDS [23]. However, it is not known whether HFOV would have a role as a last resort strategy for those patients with severe ARDS who cannot be maintained on an ARDSNet strategy [22].

A number of studies and meta-analyses have examined whether HFOV reduces mortality [24,25]. In the largest, multicenter trial, the OSCILLATE investigators randomly assigned adult patients with new-onset moderate-to-severe ARDS to HFOV or to an ARDSNet ventilation strategy [26]. This well-designed study was terminated early after enrollment of 548 of a planned 1200 patients. The study was terminated for harm with an in-hospital mortality of 47 percent in the HFOV arm and 35 percent in the ARDSNet arm (relative risk of death with HFOV, 1.33; 95% CI 1.09-1.64; p = 0.005). Mortality in OSCILLATE did not appear related to risk of cor pulmonale, hemodynamics two hours after HFOV had been initiated, or obesity [27,28]. Thus, HFOV can NOT be recommended as the initial treatment strategy for adult patients with ARDS.

The OSCAR trial, involving nearly 800 patients in the United Kingdom, also failed to demonstrate a mortality benefit at 30 days although the detrimental effect of HFOV on mortality observed in the OSCILLATE trial above was not in evidence [29].

The results of the OSCILLATE and OSCAR trials differ from both a Cochrane analysis of eight randomized trials (419 patients), and a meta-analysis of six randomized trials (365 patients) [19,30]. Neither analysis included OSCILLATE. Both meta-analyses of adults with ARDS who received HFOV had significantly lower hospital mortality or 30-day mortality than those who received conventional mechanical ventilation alone (39 versus 49 percent, RR 0.77, 95% CI 0.61-0.98) [19,30]. A limitation of these meta-analyses was that some of the trials that were included did not use low tidal volume ventilation in their control groups, which could bias the results in favor of HFOV. When trials that allowed tidal volumes ≥8 mL/kg were excluded and the meta-analysis repeated, there was a trend toward lower mortality among patients who received HFOV (RR 0.67, 95% CI 0.44-1.03). These results indicate that the repeat meta-analysis was too small to exclude or confirm a clinically important effect and additional trials are necessary to compare the effects of HFOV and low tidal volume ventilation on mortality. In contrast, two meta-analyses of six and seven randomized trials, respectively that did include OSCILLATE and OSCAR confirmed no mortality benefit with HFOV [21,25], while another meta-analysis of four trials (that included OSCILLATE and OSCAR) reported increased mortality in those with mild to moderate ARDS and possibly decreased mortality in those with severe ARDS [22]. (See "Acute respiratory distress syndrome: Ventilator management strategies for adults", section on 'Low tidal volume ventilation: Initial settings'.)

It was hypothesized that an explanation for a lack of mortality benefit to HFOV might come from the duration of conventional mechanical ventilation prior to HFOV [31]. However, a meta-analysis of nine studies (two randomized trials and seven observational studies) found no such relationship, even after confounding variables were considered [32]. Only the oxygenation index was independently associated with mortality (see "Measures of oxygenation and mechanisms of hypoxemia", section on 'Oxygenation index'). If HFOV is of any benefit in any patients with ARDS it is potentially only among the most hypoxemic patients: a post hoc analysis of over 1500 patients from four trials found that while mortality was increased in patients treated with HFOV whose partial arterial pressure of oxygen/fraction of inspired oxygen (PaO₂/FiO₂) ratio on conventional ventilation was greater than 100, mortality was actually decreased when the PaO₂/FiO₂ ratio was below 100 and significantly so when below 50 [22].

The effects of HFOV combined with another intervention have also been evaluated. Generally speaking, HFOV improves oxygenation when combined with inhaled nitric oxide or recruitment maneuvers [31,33]. It may also prevent worsening of hypoxemia when a patient returns to the supine position following prone ventilation [34]. None of these combinations have been shown to improve important clinical outcomes and all of the studies had significant methodologic limitations.

HF percussive ventilation — High-frequency percussive ventilation (HFPV) improves both oxygenation and ventilation without hemodynamic instability or clinically evident pulmonary barotrauma [8,35,36]. It may also decrease intracranial pressure in patients with head injuries [8].

●One systematic review of seven studies reported that compared with conventional mechanical ventilation, HFPV may decrease in-hospital mortality and incidence of pneumonia in patients with smoke inhalation injury [37].

●A single center, randomized trial of 62 patients with burns and respiratory failure compared HFPV to conventional low tidal volume ventilation [38]. There were no differences in ventilator-free days or mortality, and there was less need for rescue ventilation in the HFPV group (6 versus 29 percent). In addition, there was a trend toward a lower incidence of ventilator-associated pneumonia in the HFPV group. Only 39 to 45 percent of patients in this study had acute lung injury (ALI)/ARDS.

●A trial randomly assigned 35 patients with inhalational injury to undergo HFPV or volume-limited mechanical ventilation [35]. The HFPV group had significant improvement in the arterial oxygen tension/fraction of inspired oxygen (PaO₂/FiO₂) ratio of the initial 72 hours, compared to the volume-limited ventilation group.

●An uncontrolled trial of 54 patients with ALI/ARDS demonstrated improved oxygenation, decreased physiologic shunting, and decreased peak airway pressures after changing the mode of ventilation to HFPV [36].

●In a retrospective analysis of 12 patients with class 2 or 3 obesity, HFPV was used as a rescue strategy following failure of conventional mechanical ventilation and prior to extracorporeal membrane oxygenation [39]. HFPV was associated with improvement in oxygenation parameters and a 67 percent survival rate.

HARMS — HFV is not risk free. The high respiratory rate shortens the expiratory time, potentially causing auto-positive end-expiratory pressure (PEEP) and dynamic hyperinflation. The plateau airway pressure (alveolar pressure) and mean airway pressure are likely to increase if auto-PEEP and dynamic hyperinflation develop, elevating the risk of pulmonary barotrauma and hemodynamic instability. This occurs despite a lower peak airway pressure conferred by the smaller tidal volumes. In many trials, the risk of pulmonary barotrauma or hemodynamic instability was the same for patients receiving HFV compared to those receiving an alternative mode of mechanical ventilation [2,25]. (See "Diagnosis, management, and prevention of pulmonary barotrauma during invasive mechanical ventilation in adults" and "Clinical and physiologic complications of mechanical ventilation: Overview", section on 'Hypotension'.)

There are also complications unique to type of HFV. As an example, high-frequency jet ventilation (HFJV) is associated with necrotizing tracheobronchitis, endotracheal tube mucus inspissation, and variability of cardiac output [40]. Proper gas humidification reduces the likelihood of necrotizing tracheobronchitis or endotracheal tube mucus inspissation.

SUMMARY AND RECOMMENDATIONS

●High-frequency ventilation (HFV) combines a very high respiratory rate with tidal volumes that are smaller than the volume of anatomic dead space. (See 'Introduction' above.)

●There are four types of HFV: high-frequency jet ventilation (HFJV), high-frequency oscillatory ventilation (HFOV), high-frequency percussive ventilation (HFPV), and high-frequency positive pressure ventilation (HFPPV). HFOV has been most commonly used. However, none of these modes can be recommended as first line modes of mechanical ventilation. (See 'Types of HFV' above.)

●There are no universally accepted indications for HFV. Its use has also been described in a variety of clinical situations. HFV should be avoided in patients with obstructive lung disease. (See 'Patient selection' above.)

●Despite evidence that HFOV improves oxygenation, it does not reduce, and may increase, in-hospital mortality in adult patients with adult respiratory distress syndrome (ARDS) compared with a ventilation strategy of low tidal volume and high positive end-expiratory pressure. Thus, HFOV is not recommended as a routine ventilatory strategy in patients with ARDS. (See 'HF oscillatory ventilation' above.)

●HPFV also appears to improve oxygenation, although evidence that it improves clinical outcomes (eg, mortality, duration of mechanical ventilation, or length of intensive care unit [ICU] stay) is lacking. (See 'Efficacy' above.)

●HFV is not risk free. Potential harms include intrinsic positive end-expiratory pressure (auto-PEEP), dynamic hyperinflation, and related sequelae (eg, pulmonary barotrauma, hemodynamic instability). In addition, there are specific risks associated with each type of HFV. (See 'Harms' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Anthony Courey, MD, who contributed to an earlier version of this topic review.