Chapter 100: High-Frequency Oscillatory Ventilation

High-frequency oscillatory ventilation (HFOV) is a form of ventilatory support that delivers extremely small tidal volumes (V_Ts) at very high rates and maintains a relatively constant and higher mean airway pressure (mPaw) than mechanical ventilation. These properties make HFOV an ideal mode of ventilation for lung protection, because it can allow clinicians to operate in a “safe” zone of the volume-pressure curve, avoiding zones of overdistension and derecruitment/atelectasis, provided an optimal mPaw is set and very small V_Ts are delivered.

HFOV was first described in 1972 and was used to improve oxygenation in neonates with severe respiratory distress syndrome. This was gradually expanded into larger, more mature pediatric patients with severe respiratory failure. In 2001, the Food and Drug Administration approved HFOV devices for use in adult patients with acute respiratory distress syndrome (ARDS) who failed conventional mechanical ventilation (CMV).

Conventional ventilation mimics the respiratory cycle with inspiration and expiration via positive pressure inhalation compared with the negative pressure that drives normal respiration. HFOV is based on several features of nonlaminar flow of gas into and out of a circuit. High-pressure gas flows down the center of the airway, displacing lower pressure gas via coaxial flow and bulk flow. Between cycles and in distal airways, gases mix uniformly so that delivered, oxygen-rich gas saturates available alveoli to maximize the chances of diffusion of oxygen into the bloodstream. This mixing is called pendelluft and essentially accounts for dead space ventilation of any kind, but is especially useful in HFOV. “Exhaled” gases move by previously mentioned bulk and coaxial flows toward the negative pressure generated by the high-velocity “breaths” where they ultimately exit the endotracheal tube into the exhalation circuit. HFOV can be considered to have separated oxygenation and ventilation into 2 separate mechanisms.

Similar to CMV when the respiratory frequency exceeds the time for intrinsic exhalation phase, HFOV delivers breaths that can lead to developing auto–positive end-expiratory pressure (auto-PEEP). This can still occur with very small tidal volumes in HFOV. Whereas auto-PEEP in CMV can be deleterious, the typical small volumes and high rates used in HFOV allow the clinician to avoid auto-PEEP; however, patients with elevated airway resistance are relatively contraindicated for HFOV. The typical maneuver to reduce auto-PEEP is to increase the frequency, which decreases the inspiratory time, and allows for more exhalation and limits breath stacking. Other mechanisms of gas exchange during HFOV include cardiogenic mixing where the contracting heart causes mechanical agitation of gas, especially in lung units surrounding the heart, and molecular diffusion near the alveolar-capillary interface.

HFOV is indicated for patients with severe ARDS whose lungs cannot tolerate high tidal distending pressure. In these patients, HFOV may provide a modality to improve oxygenation and maintain peak airway pressures that are nearly the same as the mean airway pressures. Although the pressure wave at the level of the endotracheal tube has a large gradient, the distending pressures exerted on the alveoli are much closer to the mean airway pressure and the theory suggests that the lung is maintained in a constantly open state. Conversely, the “tidaling” of CMV still exerts a gradient on the distal airways and alveoli. This means that if underrecruited, there exists the possibility of “atelectrauma” as the lung is opened and closed with each cycle. In the setting of very poor lung compliance, barotrauma is common.

Patient selection begins with moderate to severe ARDS that by strict definition represents patients whose ratio of the partial pressure of oxygen from the arterial blood to the fraction of inspired oxygen (Pao₂:FIo₂) is less than 200, and who require PEEP greater than 10 to 15 cm H₂O to achieve this oxygenation. Patients with plateau pressures greater than 30 cm H₂O and/or mean airway pressures of greater than 24 cm H₂O can also be considered for HFOV. Similarly, patients ventilated by other alternative modes including airway pressure release ventilation (APRV) with P high greater than 35 cm H₂O can be considered. Though not specifically studied, case reports of patients with bronchopleural fistulas have been successfully treated with HFOV.

HFOV has not been studied in pregnant patients, those with severe chronic obstructive pulmonary disease (COPD), and in patients with hemoptysis or copious thick secretions. Patients with high airway resistance are at increased risk of developing auto-PEEP and should be carefully screened. Pneumothorax is not a contraindication to HFOV, and the clinician must be aware of the possibility that a pneumothorax can occur at any time.

Following patient selection for HFOV, the airway should be suctioned and adequate sedation and analgesia should be provided (Table 100–1). Many patients experience discomfort with the constant insufflation of the lung and vibrations associated with HFOV, and thus may require deeper sedation and occasionally neuromuscular blockade to ensure patient-ventilator synchrony and reduce peripheral oxygen consumption. However, complete cessation of patient respiratory efforts is not recommended and preferably, the patient should be allowed to generate small spontaneous breaths. Once the patient is stable on HFOV, discontinuation of the paralytic agent should be attempted daily.

Initial bias flow should be set to 35 L/min and may be adjusted according to patient’s needs. Inspiratory time is set as a percentage and 33% is recommended as an initial setting. Frequency should be set to 5 Hz, which correlates to a respiratory rate of 300 breaths/min. The main determinant of oxygenation during HFOV is the mPaw, and this is generally initiated at approximately 5 cm H₂O greater than the mPaw noted during conventional ventilation or equal to that of APRV. For severely hypoxemic patients, a recruitment maneuver of 40 cm H₂O for approximately 40 seconds can be considered. The Fio₂ should be set to 1 and then tapered using pulse oximetry to maintain SpO₂ greater than or equal to 88%.

The main determinants of Paco₂ removal are the pressure amplitude of oscillation (ΔP) and the frequency (Hz) setting. Increasing the ΔP and decreasing the frequency reduces the delivered tidal volume and allows Paco₂ to rise. The ΔP is usually started at either a value where the chest vibrates down to their midthigh or a value of 20 to 30 cm H₂O above what the patient’s Paco₂ was on conventional ventilation. For example, if the Paco₂ is 60 mm Hg, the ΔP is initiated at 80 cm H₂O. After approximately 30 minutes of HFOV, a repeat blood gas should be analyzed and the power setting should be titrated based on the desired Paco₂ level. The chest wall needs to be vibrating; if not, the power setting has to be increased.

On all patients placed on HFOV, it is crucial that oxygen is flowing to the patient. By convention, in the United States oxygen tubing and cylinders are color-coded green. Pressurized air is color-coded yellow, and contains only 20.9% oxygen, the same as room air. Most HFOV ventilators display the amplitude and frequency. While an unexplained rise in amplitude can suggest that the endotracheal tube is becoming obstructed, a sudden decrease in amplitude can suggest pneumothorax. Examining the ventilator tubing can sometimes reveal secretions or hemoptysis. When examining the patient, a decrease in the patient’s chest wiggle can suggest pneumothorax, and a prompt chest x-ray should be performed. Although suctioning and bronchoscopy can cause derecruitment, they can be very important in managing the patient on HFOV. Care should be taken to minimize the duration of suctioning and bronchoscopy, and recruitment maneuvers might be considered following any interruption in HFOV.

Routine measures such as obtaining chest x-rays and performing physical examinations are not contraindicated while the patient is on HFOV. Temporarily stopping the piston allows for auscultation of the patient’s heart, but stoppage should be minimized. Patients can be moved for imaging studies without stopping HFOV. Daily nursing care remains unaffected. Measures to ensure adequate sedation such as a Bispectral Index Sensor (BIS) monitor should be in place if an infusion of neuromuscular blocking agent is being used as well as routine train-of-four monitoring to minimize the side effects of neuromuscular blockade.

Complications that may occur during HFOV include hypotension, pneumothorax, and endotracheal tube obstruction. Hypotension occurs occasionally shortly following the switch to HFOV or as mPaw is increased and usually responds to intravenous fluid boluses. A pneumothorax during HFOV may be difficult to detect because of the background noise of the ventilator and the diffuse transmission of airway sounds. However, the loss of chest wiggle that usually occurs on the affected side is an important physical clue to the possibility of a pneumothorax. Obstruction of the endotracheal tube should be considered when there is an abrupt rise in Paco₂ during HFOV in an otherwise stable patient. In this circumstance, a suction catheter should be passed immediately to ensure patency of the endotracheal tube; occasionally, bronchoscopy may be required to visually inspect the airway.

Weaning attempts from HFOV usually commences when the patient responds with improved oxygenation and is able to maintain a SpO₂ greater than 90% on a Fio₂ of 0.4. When this occurs, the mPaw is gradually reduced by 2 to 3 cm H₂O every 4 to 6 hours as tolerated as alveolar derecruitment and desaturation may occur if the mPaw is decreased rapidly. As soon as a mPaw of 20 to 24 cm H₂O is achieved while maintaining an Fio₂ of 0.4, the patient can be switched back to a trail of conventional ventilation. The conventional ventilator is usually set to achieve a mPaw of 20 cm H₂O by using pressure control mode with peak pressure set to achieve a delivered tidal volume of 6 to 8 mL/kg of predicted body weight and inspiratory plateau pressure of less than 30 cm H₂O. An arterial blood gas is obtained 20 to 30 minutes after transfer to conventional ventilation to guide further ventilator adjustments.

Two large multicenter trials compared HFOV with conventional ventilation in patients with moderate-to-severe ARDS. In the OSCILLATE (Oscillation for ARDS Treated Early), the HFOV strategy with high mean airway pressures was associated with more deaths than the conventional ventilation strategy that used relatively aggressive high PEEP levels (47% vs. 35%) which led to premature termination of the trial. Hemodynamic compromise resulting from the elevated mean airway pressures was thought to be the mechanism accounting for the poor HFOV outcomes. In the OSCAR (High-Frequency Oscillation in ARDS) trial, there was no major difference in the outcome between the HFOV strategy and usual care with conventional mechanical ventilation. In both trials, there was a higher proportion of patients in the HFOV groups who received sedatives and muscle relaxants, which may have also contributed to the poorer outcomes.