Show Summary Details
Page of

High-frequency ventilation and oscillation 

High-frequency ventilation and oscillation
High-frequency ventilation and oscillation

Mireia Cuartero

and Niall D. Ferguson

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 29 November 2020

Key points

  • High-frequency oscillatory ventilation (HFOV) is a ventilatory mode that delivers pressure oscillations around a relatively constant mean airway pressure.

  • Resultant tidal volumes with HFOV are very small—often smaller than the anatomic dead space—but adequate ventilation is achieved through a number of mechanisms generally related to enhanced mixing of gas within the lung.

  • HFOV has been widely studied and adopted in the neonatal intensive care unit (ICU), where it may help reduce the incidence of chronic lung disease; it is also commonly used in the paediatric setting.

  • In adults with acute respiratory distress syndrome (ARDS), HFOV is often effective in improving oxygenation among patients failing conventional ventilation.

  • Despite being theoretically ideal for preventing ventilator-induced lung injury, recent trials indicate that HFOV does not reduce mortality in patients with moderate–severe ARDS.


High-frequency oscillatory ventilation (HFOV) is a ventilation mode that achieves adequate alveolar ventilation despite using very low tidal volumes (VT), at or below the dead space volume (VD) (approximately 1–2 mL/kg), at frequencies significantly above normal physiological values (more than 3 breaths/second) [1]‌. Theoretically, it presents some advantages in its ability to apply a lung protective ventilation strategy and thereby avoid ventilator-induced lung injury (VILI) [2,3].

The modern history of HFOV began in 1971 when Jonzon presented a study about the circulatory effects of high-frequency ventilation [4]‌. The next year, Lunkenheimer et al. described the use of a ventilator based on an electromagnetic vibrator at very high respiratory frequencies (40 Hz) to clear CO2 [5]. Previously, there had also been some studies reported in animal models. However, through his work in animals and neonates in the late 1970s and 1980s, A. C. Bryan is considered to be the father and developer of HFOV as it is currently known [6,7]. HFOV was initially introduced in the neonatal ICU, where it has been the subject of many randomized controlled trials (RCTs) [8], and it was also widely adopted in the paediatric intensive care unit (ICU). More recently, HFOV made the transition to adult critical care beginning in the late 1990s when technical improvements allowed for the oscillation of those weighing more than 35 kg. Since then, there have been many series and case reports of this ventilatory mode in adults failing conventional ventilation, and a few randomized trials in patients with acute respiratory distress syndrome (ARDS).

HFOV is one of a family of ventilatory modes termed high-frequency ventilation (HFV), and is the only one we will discuss in any detail here. Other members of this family include high frequency jet ventilation (HFJV) and high-frequency percussive ventilation (HFPV). Their differences are based on the manner of generating the high frequency—by an oscillating membrane with HFO, by jet catheter in HFJV, and with flow interruption in HFPV—and also by their ranges in frequency (2–15 Hz), on the type of wave (triangular or sinusoidal), the inspiratory–expiratory ratio (I:E) (constant or adjustable), and the type of expiration (active or passive) (Table 98.1).

Table 98.1 Main characteristics of the different types of HFV

Type of HFV


Hz (breaths/min)





Additional flow through a small-bore catheter placed within the endotracheal tube (Coanda effect)

1.7–2.5 (100–150)


Very efficient for removing CO2

  • Risk of dynamic hyperinflation

  • Unpredictable VT

  • Tracheobronchial injury


HFV + conventional pressure-control breathing pattern

3.4–15 (200–900)


Improve secretion clearance

Much less studied


Pressure oscillations around a mPaw, resulting in very small VT

3–15 (180–900)


Decoupling of ventilation (depending on VT and Hz) and oxygenation (depending on mPaw).

  • Air leak syndrome

  • Right heart failure

  • Pneumothorax

HFJV, high frequency jet ventilation; HFPV, high-frequency percussive ventilation; HFOV, high-frequency oscillatory ventilation.

Physiological effects of HFOV

The HFOV ventilator (see Fig. 98.1) is an oscillator with a diaphragm located in the inspiratory circuit. The ventilator pressurizes the patient circuit by means of a continuous flow of gas (bias flow) and a control valve allows modulation of the mean airway pressure (mPaw). The gas is actively pushed in and pulled out of the patient by diaphragm, which oscillates according to an electrically-driven magnet or by bi-directional blower, depending on the model. The amplitude of displacement of the diaphragm can be modified by adjusting the power, which generates a peak-to-peak pressure gradient, displayed as Δ‎P on the more commonly-used SensorMedics 3100B oscillator. The oscillation frequency on this ventilator can be adjusted to between 180 and 900 cycles/min (3–15 Hz), although in adults frequencies between 5 and 10 Hz are most commonly used. The amplitude displacement of the diaphragm, the oscillation frequency, and the characteristics of the airway and respiratory system compliance determines the amplitude of the oscillations (or pressure difference Δ‎P) around the mean airway pressure (mPaw).

Fig. 98.1 Scheme of the HFOV circuit.

Fig. 98.1 Scheme of the HFOV circuit.

During HFOV there is relative decoupling of oxygenation and ventilation. Oxygen is generally directly dependent on FiO2 and mPaw. Ventilation meanwhile, is directly proportional to power (Δ‎P) and inversely proportional to the applied frequency. CO2 elimination is proportional to the frequency and the square of the tidal volume (Vco2 α‎ = f × VT2). Unlike conventional ventilation, where tidal volume (VT) is relatively independent of respiratory rate, during HFOV increasing frequency reduces inspiratory time and can have a strong influence on the resultant tidal volume, thus reductions in frequency usually lower PaCO2. The VT produced is often less than the volume of the anatomical dead space and is determined by the Δ‎P, the frequency, the airway resistance (mainly by endotracheal tube size), and respiratory system compliance. Even though the Δ‎P measured in the ventilator circuit can be as high as 90 cmH2O, the transmission of this pressure to the alveoli is significantly attenuated, and is dependent on the diameter of endotracheal tube, the oscillation frequency, the resistance of the airway, and respiratory system compliance.

Because tidal volumes with HFOV are often lower than dead space volumes, the mechanisms of gas-transport mechanisms are complex and multiple (see Fig. 98.2) [9,10]. Adequate ventilation is achieved through the following phenomena:

  • Direct alveolar ventilation: despite a very small VT proximal alveoli may still be ventilated directly by bulk convection (the usual method of gas transport within the lung).

  • Turbulence in the large airways: causing enhanced mixing.

  • Asymmetric velocity profiles: due to the rapidly alternating high velocities of gas in the airway, a summation vector is created with a parabolic shape of penetration greater into the centre of the air than in the periphery. This leads to fresh gas entering through the middle of the airway, while expired gas tracks up the outside of the airway.

  • Pendulum movement of air between alveolar units (pendelluft): this is the exchange of gas between adjacent lung units with differing time constants, due to the asymmetries in airflow impedances.

  • Increased dispersion: this refers to the molecular dispersion of gas in the airway increases the sinusoidal deformation and radial forces at gas diffusion movements caused by the turbulence produced in the bronchial branches (Taylor dispersion).

  • Molecular diffusion: displacement of the molecular gas from areas of higher to lower concentration. CO2 has a higher diffusion constant than O2.

Fig. 98.2 Gas transport mechanisms.

Fig. 98.2 Gas transport mechanisms.

From New England Journal of Medicine, Slutsky AS and Drazen JM, ‘Ventilation with small tidal volumes’, 347(9), pp. 629–33. Copyright © 2002 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society.

The key action of HFOV is the constant maintenance of the mPaw, in contrast to conventional modes, where even if the numerical value of mPaw is similar, airway pressure varies greatly over the course of the respiratory cycle. The relatively constant mPaw decreases the resistance to gas movement, facilitating and promoting the gas-transport phenomena described in the bulleted list previous to this paragraph. Moreover, the active recruitment of non-functional alveoli may be accomplished in a few hours, thanks to the constant high mean airway pressure.

Clinical implications

In 1980s Bryan and colleagues developed several physiological studies, which set the basis for our current knowledge of HFOV [7]‌. There have been several experimental animal studies to apply this mode to prevent VILI. Most of them, even when compared to ‘protective’ conventional ventilation, showed an improvement in pulmonary injury severity scores, in gas exchange and a decrease of some inflammatory markers (TNF, IL-1B, IL-6, IL-8, IL10, and TGF). Many case series were reported as a first clinical approach since that moment on.

From 1990s up to now, HFV has become standard practice in neonatal and paediatric ICU’s in the treatment of newborns with respiratory distress syndrome. A recent meta-analysis [11], however, and the latest Cochrane review [12] comparing HFOV with conventional ventilation in preterm infants do raise some doubts about the real effectiveness of HFOV in the neonatal population. The 15 RCTs included in the Cochrane review showed no clear benefit with HFOV when their results were pooled. In those studies there are data suggesting some reduction in the rate of chronic lung disease (CLD) with HFOV, but that was not consistent across all them. Moreover, the introduction of HFOV into neonatal ICU’s coincided with the simultaneous introduction of surfactant in the treatment of newborns with respiratory distress of prematurity. There is strong evidence that surfactant has led to a decreased mortality and morbidity, so a confounding effect cannot be ruled out. Therefore, further trials on elective HFOV in infants at higher risk of CLD could be conducted, although the feasibility of such trials may be questioned given the widespread adoption of HFOV in the neonatal and paediatric setting.

In adults, the first studies were set as a rescue therapy for patients with severe ARDS who were ‘failing’ conventional ventilation. Fort et al. published the first of those case series, which suggested an improvement in oxygenation with HFOV [13]. Later on, Mehta and others showed similar results in subsequent case series [14]—HFOV appeared to be safe and effective in improving oxygenation in these extremely severe ARDS patients.

During the late 1990s several small RCT set out to explore those promising results further. The largest of these [15] randomized 148 patients with ARDS to either conventional mechanical ventilation (MV) with a VT target of 6–10 mL/kg actual body weight, or to HFOV. There was no significant difference between groups in the primary outcome measure of survival without need for mechanical ventilation at 30 days. There was, however, an improvement in oxygenation and a non-significant trend towards lower mortality at 30 days with HFOV compared with the conventional MV (37 versus 52%, P = 0.102). Ten years later, however, it was clear that this and other trials were confounded by both small sample sizes and outdated control ventilation strategies. Thus, despite a meta-analysis showing a significant reduction in mortality [11], calls went out for larger trials comparing HFOV with lung-protective conventional ventilation in adults with ARDS. These trials were conceived on the premise that HFOV could be theoretically ideal for preventing VILI, based on the very small tidal volumes delivered, and the minimal swings in alveolar airway pressure, which might allow one to use higher pressures to recruit the atelectatic lung, while simultaneously avoiding tidal overdistention [16,17].

Recently, two large multicentre RCTs have been published, which shed light on the effects of HFOV on all-cause mortality. Both the Oscillation for ARDS (OSCAR; n = 795) trial [18] and the Oscillation for ARDS Treated Early (OSCILLATE; n = 548) trial [19] recruited patients with moderate–severe ARDS and assigned them to relatively early support with HFOV versus protective conventional MV (with VT target 6 mL/kg and in the case of OSCILLATE, higher positive end-expiratory pressure). In contrast to prior studies, both of these trials found no evidence of improved survival in patients assigned to HFOV. Indeed, the OSCILLATE trial was stopped early because of concerns of a significantly higher mortality rate in the HFOV group (47 versus 35%).


HFOV, like any ventilatory mode, has related potential complications. None of the reported RCTs or case series, however, suggested any specific and clear complications at a higher rate than might be expected with conventional ventilation. The specific reasons for the higher mortality seen in the OSCILLATE trial remain uncertain, but among the postulated explanations are:

  • Haemodynamic consequences related to higher airway pressures and sedation levels.

  • Increased barotrauma and paradoxical increase in VILI.

  • Increased sedation use.

  • Chance alone.

A known complication of HFOV (and of conventional ventilation) is barotrauma, with the occurrence of air leak being reported in up to 25% of patients in some case series of rescue therapy. This ARDS population, however, is at higher risk of developing pneumothorax. The two recent large trials did not show statistically significant differences in barotrauma, although in OSCILLATE barotrauma rates did trend higher in the HFOV group.

Secondly, because the bias flow rate is insufficient to meet ventilatory demands, adults on HFOV usually requires larger doses of sedation and sometimes paralysis in order to blunt or eliminate their respiratory efforts, something that is often not the case in paediatric and neonatal populations. Transient muscle relaxation is often recommended during the initiation of HFOV, during recruitment manoeuvres and, in some specific conditions, such as air leak syndromes and severe hypoxaemia. In fact, a moderate amount of spontaneous breathing does not usually interfere with gas exchange during HFOV; changes in the mPaw of plus or minus 5 cmH2O can often be allowed. Moreover, although the use of neuromuscular paralysis and sedation are related with an overall increased mortality in the overall ICU population, they may improve survival when used in short-term and early stages of ARDS.

Another potential complication is the obstruction of the airway due to the lack of clearance of secretions. This effect is uncommon, and can be reduced by means of adequate humidification [13,19]. Nevertheless a sudden increase in PaCO2 or unexplained increase in Δ‎P on the ventilator should be investigated with bronchoscopy to rule out endotracheal tube obstruction.

Finally, as with any kind of positive pressure ventilation, HFOV increases intrathoracic pressure, reducing venous return and even may potentially promote right heart failure, perhaps because of the constant mean airway pressure to which the right ventricle is exposed [15,20]. However, the presence of acute cor pulmonale is usually reversible in patients who recover from ARDS and does not increase long-term mortality.


HFOV produces effective exchange of CO2 and O2 with lower peak pressures at the alveolar level and a higher mean airway pressure. Indeed, applying a stable mPaw produces minimal variations in pressures and ventilation volumes, keeping the lung volume above functional residual capacity largely constant. Therefore, HFOV is an interesting ventilatory modality, which has the potential to minimize ventilator-induced lung injury, at least in theory.

HFOV has become an established lung-protective modality in neonatal and paediatric intensive care, although further studies to support an improvement of mortality and morbidity could be conducted. In adults recent publications do not support the routine use of HFOV in patients with moderate-severe ARDS as there was no signal for benefit and even a suggestion of harm from one trial. While these findings do not necessarily apply to patients with severe hypoxaemia failing conventional ventilation, they do increase uncertainty about the role of HFOV even in these patients. In carefully selected patients who respond to lung recruitment, HFOV may still have a role in severe ARDS, but only after conventional ventilation settings have been optimized and after prone positioning has been considered.


1. Froese AB and Ferguson ND. (2012). Unconventional methods of ventilator support. In: Tobin MJ (ed.) Principles and Practice of Mechanical Ventilation, pp. 1–25. New York, NY: McGraw-Hill.Find this resource:

2. Amato MB, Valente CS, Machado D, et al. (1998). Effect of a protective-ventilation strategy on mortality in the acute respiratory distress. New England Journal of Medicine, 338, 347–54.Find this resource:

3. Lee WL and Slutsky AS. (2001). Ventilator-induced lung injury and recommendations for mechanical ventilation of patients with ARDS. Seminars in Respiratory and Critical Care Medicine, 22(3), 269–80.Find this resource:

4. Jonzon A, Öberg P, Sedin G, and Sjöstrand U. (1970). High frequency low tidal volume positive pressure ventilation. Acta Physiologica Scandinavica, 80(Vi), 21–2.Find this resource:

5. Lunkenheimer PP, Rafflenbeul W, Keller H, Frank I, Dickhut HH, and Fuhrmann C. (1972). Application of transtracheal pressure oscillations as a modification of ‘diffusion respiration’. British Journal of Anaesthesia, 44, 627.Find this resource:

6. Bryan A and Slutsky A. (1986). Long volume during high frequency oscillation. American Review of Respiratory Disease, 5, 928–30.Find this resource:

7. Bryan AC. (2001). How it really happened the oscillations of HFO. American Journal of Respiratory and Critical Care Medicine, 163(4), 816–17.Find this resource:

8. Parad RB. (2010). HFOV in preterms: an individual patients’ data meta-analysis. Lancet, 375(9731), 2054–5.Find this resource:

9. Jonzon A, Rondio Z, and Sedin G. (1983). Alveolar deadspace during high frequency positive pressure ventilation. British Journal of Anaesthesia, 55, 1133.Find this resource:

10. Slutsky AS and Drazen JM. (2002). Ventilation with small tidal volumes. New England Journal of Medicine, 347(9), 629–33.Find this resource:

11. Sud S, Sud M, Friedrich JO, et al. (2010). High frequency oscillation in patients with acute lung injury and acute respiratory distress syndrome (ARDS): systematic review and meta-analysis. British Medical Journal, 340, c2327.Find this resource:

12. Henderson-Smart D, Cools F, Bhuta T, and Offringa M. (2008). Cochrane review: elective high frequency oscillatory ventilation versus conventional ventilation for acute pulmonary dysfunction in preterm infants. Evidence-Based Child Health: A Cochrane Review Journal, 3(3), 731–808.Find this resource:

13. Fort P, Farmer C, Westerman J, et al. (1997). High-frequency oscillatory ventilation for adult respiratory distress syndrome—a pilot study. Critical Care Medicine, 25(6), 937–47.Find this resource:

14. Mehta S, Granton J, MacDonald RJ, et al. (2004). High-frequency oscillatory ventilation in adults: the Toronto experience. Chest, 126(2), 518–27.Find this resource:

15. Derdak S, Mehta S, Stewart TE, et al. (2002). High-frequency oscillatory ventilation for acute respiratory distress syndrome in adults: a randomized, controlled trial. American Journal of Respiratory and Critical Care Medicine, 166(6), 801–8.Find this resource:

16. Hager DN, Fessler HE, Kaczka DW, et al. (2007). Tidal volume delivery during high-frequency oscillatory ventilation in adults with acute respiratory distress syndrome. Critical Care Medicine, 35(6), 1522–9.Find this resource:

17. Fessler HE, Derdak S, Ferguson ND, et al. (2007). A protocol for high-frequency oscillatory ventilation in adults: results from a roundtable discussion. Critical Care Medicine, 35(7), 1649–54.Find this resource:

18. Young D, Lamb SE, Shah S, et al. (2013). High-frequency oscillation for acute respiratory distress syndrome. New England Journal of Medicine, 368(9), 806–13.Find this resource:

19. Ferguson ND, Cook DJ, Guyatt GH, et al. (2013). High-frequency oscillation in early acute respiratory distress syndrome. New England Journal of Medicine, 368(9), 795–805.Find this resource:

20. Guervilly C, Forel J-M, Hraiech S, et al. (2012). Right ventricular function during high-frequency oscillatory ventilation in adults with acute respiratory distress syndrome. Critical Care Medicine, 40(5), 1539–45.Find this resource: