High-frequency ventilation (HFV) is a relatively new form of mechanical ventilation that uses rapid rates and extremely small tidal volumes. Ventilator rates can he greater than 900 breaths per minute (hpm) (15 Hz). Tidal volumes are near, often less than anatomical dead space. Conventional wisdom dictates that alveolar ventilation is the result of the tidal volume delivered into the airways minus anatomical dead space (VA=VT- VD). If true, alveolar ventilation during HFV should be zero, no matter how fast a ventilator cycles. At present, there is no agreement as to the exact mechanism or mechanisms responsible for gas transport during HFV. However, most theories invoke some form of augmented diffusion.
Normally, gas moves through the large airways primarily by convection. Convective or bulk transport continues down the tracheobronchial tree until approximately the eighth bronchial generation. There, convective transport progressively decreases and transport by diffusion progressively increases. By the time gas reaches the alveolar duct, transport is entirely by diffusion. During spontaneous breathing and conventional mechanical ventilation, convective transport predominates in the upper airways; diffusion in the lower airways. During HFV, convection becomes less important and diffusion takes over as the major gas transport mechanism. As ventilator rates increase, the point of transition from convective transport to diffusive transport moves proximally up the airways. At frequencies above 900 hpm (15 Hz) diffusion appears to he the primary transport mechanism throughout the airways. A detailed analysis of gas transport during HFV is beyond the scope of this article. For those who are interested, there are a number of theoretical and practical reviews of these matters.1,2
TYPES OF HIGH-FREQUENCY VENTILATION
Currently there are four major categories of HFV:
1. High-frequency positive pressure ventilation (HFPPV);
2. High-frequency jet ventilation (HFJV);
3. High-frequency flow interruption (HFFI); and
4. High-frequency oscillatory ventilation (HFOV).
All deliver small tidal volumes and do so at supraphysiologic rates. However, these ventilators are all mechanically different from one another. Generalizations applied to one system may or may not apply to another.
High-Frequency Positive Pressure Ventilation
The term HFPPV describes conventional ventilators operating at unconventionally rapid rates, 60 bpm (1 Hz) to 150 bpm (2.5 Hz). The presumed advantage of HFPPV is its potential to produce gas exchange using lower proximal airway pressures. If airway pressures are lower, lung distention should be less and the risk of barotrauma should also be less. A few preliminary reports suggest that HFPPV, indeed, reduces incidence of lung injury.3'5 Unfortunately, definitive clinical studies have not yet confirmed these early reports.
Today's conventional ventilators have arbitrary frequency ceilings set at 150 bpm (2.5 Hz). Most, however, were not designed to operate at such rapid rates. Compared with ventilators specifically designed for HFV, these machines have relatively high internal compliances and less efficient valve systems. Recently, five different pressure preset infant ventilators were tested over a frequency range of 25 bpm (0.42 Hz) to 150 bpm (2.5 Hz). All had maximum effective rates, ranging from 75 bpm (1.25 Hz) to 100 bpm (1.67 Hz), beyond which minute volumes progressively decreased. At rates beyond a ventilator's maximum effective rate, maintaining given minute volumes actually required higher airway pressures than were necessary at slower, more conventional rates.6 At extreme rates, conventional ventilators' exhalation valves generate increased expiratory resistances. Such resistances produce inadvertent positive end-expiratory pressure (PEEP), which can steadily and silently overpressurize the patient circuit.7 Mechanical ventilators, like other machines, develop problems when pushed beyond their limits.
High-Frequency Jet Ventilation
The term HFJV describes a ventilator system that delivers rapid, high-velocity pulses of pressurized gas directly into the trachea or endotracheal tube through a small bore injector. The volume of the jet pulse is determined by the velocity of gas flow and the duration of the jet pulse (inspiratory time). Actual delivered volumes are greater than the volume of the jet pulse because of ambient gases entrained with each highvelocity burst. Tidal volumes are thought to be near anatomical dead space. HFJV frequencies range between 150 bpm (2.5 Hz) and 600 bpm (10 Hz) depending on the mechanics of the system. HFJV systems have no exhalation valves. Since exhalation is passive an open system or "controlled air leak" must be maintained to prevent lung overdistention. Airway pressure monitoring is difficult because of the extreme gas velocities in the upper airway and their subsequent venturi effects. To be meaningful, pressures must be measured downstream from the jet injector. Recently, a triple lumen endotracheal tube was developed specifically for HFJV. This tube incorporates a standard lumen for suctioning and entrainment, an injector port, and an accurately positioned pressure monitoring port. Both HFJV and HFFI systems employ the term "driving pressure." This is the pressure upstream from the injector valve, literally the pressure that "drives" the jet pulse. Driving pressure is strictly an internal ventilator pressure and should not be confused with or substituted for proximal airway pressure measurements.
In the past, HFJV humidification was a major problem.8,9 It is now clear that to humidify respiratory gases adequately both the jet pulse and the entrained gases must be humidified. I0 Since the development of new, improved HFJV humidity systems, this is no longer a major issue.
High'Frequency Flow Interruption
HFFI is a modified form of HFJV with injectors set back some distance from the trachea and endotracheal tube. Consequently, these ventilators have also been called "set-back jets." As with HFJV, tidal volume is the product of gas velocity, inspiratory time, endotracheal tube size, and entrained gases.11 Tidal volumes are thought to be near or somewhat greater than anatomical dead space. Operating frequencies range from 300 bpm (5 Hz) to 1,200 bpm (20 Hz). Some suggest that because the high velocity gas pulse enters the airway further upstream, HFFI is less likely to produce trachéal injuries than other forms of HFV.7 Despite the logic of this suggestion, recent animal studies have failed to demonstrate significant differences in airway pathology produced by the various forms of HFV. 12 In addition, a recent clinical trial of HFFl demonstrated a high incidence of airway damage.13
High-Frequency Oscillatory Ventilation
High-frequency oscillators are airway vibrators. Their energy sources, usually piston pumps or vibrating diaphragms, move volumes of gas first toward then away from the airways. This to and fro movement creates first a positive, then a negative airway pressure deflection. Fresh humidified gas enters the system via a continuous transverse, "bias flow" stream passing in front of the oscillatory source. Exhaled gases exit through a controlled leak or resistor called a low pass filter. Tidal volumes are said to he near or less than anatomical dead space. Delivered volumes are determined by the amplitude of the pressure wave oscillation, which in turn is determined hy the stroke of the piston or vibrating diaphragm; frequency and endotracheal tube size also determine volume to a large extent.11 HFOV is the fastest form of HFV. Operating frequencies range from 400 bpm (6.7 Hz) to 2,400 bpm 40 Hz).14 Today, the most commonly used HFOV frequency is 900 bpm (15 Hz).
Although HFV has been studied for more than ten years, it is still an experimental therapy. With the exception of HFPPV, HFV can still only be used under NlH approved research protocols and only following informed consent. At one time or another, either in clinical settings or in animal models, HFV has been employed in the treatment of most significant neonatal lung disorders. In some situations it has worked, in others it has not.
The most dramatic successes of HFV have been in the treatment of life-threatening pulmonary air leaks. Recurrent pneumothoraces, large bronchopleural fistulae, and pulmonary interstitial emphysema have all been successfully treated with various forms of HFV. At this writing, there are reports of 52 patients with severe pulmonary air leaks treated with HFV. Forty-four (85%) improved following HFV. Twentyfour (46%) survived.14 It is not clear why HFV is so superior to conventional mechanical ventilation in this condition. One theory suggests that during HFV, gas enters the lower airways at a more constant pressure. The pressure differential between lung and pleural space is less; therefore less gas escapes during peak inspiration,* There is less stretching of the injured lung and hence a better chance for self-repair.
Hyaline Membrane Disease
In 1980, Bland described the successful use of HFPPV in hyaline membrane disease (HMD). 3 Since then several clinical and laboratory studies have shown that the various forms of HFV can maintain adequate gas exchange in this condition. Studies of HFOV in an elegant baboon HMD model suggested HFOV may reduce the incidence of ventilator related lung injury.15 The initial enthusiasm generated by these primate studies cooled considerably following the recent completion of a nationwide collaborative study of HFOV in infants with HMD. This study involved ten major centers. Infants with HMD weighing 750 to 2,000 grams were randomly assigned to either HFOV or a conventional pressure preset infant ventilator. At the conclusion of the study, more than 600 infants had been enrolled. There were no significant differences in survival nor in the incidence of bronchopulmonary dysplasia. However, the incidence of intraventricular cerebral hemorrhage (grades III and IV) was significantly increased in the infants receiving HFOV. 16 The significance of these findings is not yet clear. The implications are obvious and disturbing. At this time, there is no evidence of any advantage to any form of HFV in uncomplicated, human HMD. Until such evidence appears, HFV should not be used.
Persistent Pulmonary Hypertension
Persistent pulmonary hypertension (PPH) or persistent fetal circulation is a reactive disorder of the pulmonary capillary bed characterized by hypoxemia, elevated pulmonary artery pressures, and right to left shunting through persisting fetal channels. The current treatment of choice is mechanical hyperventilation. Respiratory alkalosis is intentionally produced in hopes of dilating the pulmonary vascular bed. Here, HFOV, HFJV, and HFFI may have a theoretical advantage over conventional ventilation and HFPPV. Many infants with PPH have little, if any, underlying parenchymal lung disease. The absence of underlying lung disease increases vulnerability to lung overdistention and pressure injury. Since HFOV, HFJV, and HFFI remove CO? using lower airway pressures than either conventional mechanical ventilation or HFPPV, they may be safer, more effective therapies.
Meconium Aspiration Syndrome
Several recent case reports describe the successful use of HFJV and HFOV in infants with seemingly fatal meconium aspiration syndrome (MAS). In most, HFV was used as an intermediate measure between conventional mechanical ventilation and extracorporeal membrane oxygénation (ECMO). In many, it averted the need for ECMO. I7 Whether HFV has any place in the treatment of moderate MAS is not yet known. However, based on these clinical experiences, it seems that HFV can, at times, be useful in the treatment of severe MAS. Many now recommend a trial of some form of HFV prior to the use of ECMO in this condition.
The two major complications currently attributed to HFV are trachéal inflammations and intraventricular cerebral hemorrhage (IVH). Both are controversial- Neither has been adequately investigated or explained. A number of clinical and laboratory reports have linked trachéal damage and obstruction to various forms of HFV. Initially, these lesions, called necrotizing rracheobronchitis (NTB), were thought to be unique to HFJV. They have since been associated with all forms of HFV, including HFPPV. Although trachea! damage has occurred for as long as mechanical ventilators have been used, NTB appears to be an exaggerated form of that damage, magnified somehow by HFV." How and why these lesions occur is currently under study.
The increased incidence of significant IVH associated with HFOV during the national collaborative study of HFV was a surprising and disturbing finding. Whether an association or a cause-and-effect relationship exists is not yet known. Whatever the relationship ultimately turns out to be, serious questions must be answered. How do the various forms of HFV influence cerebral arterial blood flow? How do they affect cerebral venous pressures? Until the basic questions of safety are adequately answered, HFV must still be considered an experimental, potentially dangerous therapy. Accordingly, HFV should only be used in extremely high risk situations and only under supervised protocols.
1. Kamm RD, Slutsky AS. Draien JM: High-frequency ventilation. CRC Reviews in Biomedical Engineering 1984; 9:347-279.3
2. Drazen JM Kamm RD, Slutsky AS: High-frequency ventilation. Physiol Rev 1984; 64:505-543.
3. Bland RD, Kim MH, Light MJ, et al: High-frequency mechanical ventilation in severe hyaline membrane disease. Crit Care Med 1980; 8:275-280.
4. Bland RD, Sodin EG: High frequency mechanical ventilation in the treatment of neonatal respiratory distress. Int Anesthesiol Clin 1983; 21:125.
5. Heicher DA,. Kasting DS, Harrid JR: Prospettive clinical comparison of two methods for mechanical ventilation of neonates: Rapid rate and short inspiratory time versus slow rate and lone inspiratory time. J Pediatr 1981; 98:957-961.
6. Boros SJ, Bing DR, Mammel MC, et al: Using conventional infant ventilators at unconventional rates. Pediatrics 1984; 74:487-492.
7. Ackerman NB, DeLemos, RA: High-frequency ventilation. Adv Pediatr 1984; 31:259-293.
8. Carlon G, Kahn R, Howland W, et al: Clinical experience with high-frequency jet ventilation. Crit Care Med 1981; 9:1-6.
9. Pokora T, Bing D, Mammel M, et al. Neonatal high-frequency jet ventilation. Pediatrics 1983; 72:27-32.
10. Ophoven JP, Mammel MC, Gordon MJ, et al: Tracheobronchial histopathology associated with high frequency jet ventilation. Crit Care Med 1984; 12;829-832.
11. Fredberg JJ, Glass GM, Boynton BR, et al: Factors influencing mechanical performance of neonatal high-frequency ventilators. J Appl Physiol 1987; 62:2485-2490.
12. Mammel MC, Boms SJ: Airway damage and mechanical ventilation: A review and commentary. Pediatr Pulmonol 1987; 3:445-447.
13. Gaylord MS, Quissell BJ, Lair ME: High-frequency ventilation in the treatment of infants weighing less than 1500 grams with pulmonary intentitial emphysema: A pilot study. Pediatrics 1987; 79:915-921.
14. Mammel MC, Boros SJ: High frequency ventilation, in Goldsmith JP, Karotkin El(eds): Assisted Ventilation of the Neonate. Philadelphia, WB Saunders, 1988, pp 190-199.
15. DeLemos RA, Coalson JJ. Gerstman DR, et al: Ventilatory management of infant baboons with hyaline membrane disease: The use of high-frequency ventilation. Pediatr Res 1987; 21:594-602.
16. The HIFI Study Group and the Division of Lung Diseases, National Heart, Lung and Blood Institute. A collaborative trial of high frequency oscillatory ventilation in the treatment of respiratory failure in preterm infants. Pediatr Res 1988; 23:427A.
17. Carlo WA, Beoglos A, Walsh MC, et al: Can high-frequency jet ventilation avert the need for extracorporeal membrane oxygenation? Pediatr Res 1988: 23:403A.