| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
POINT-COUNTERPOINT
Harvard Medical School
Massachusetts General Hospital
Boston, Massachusetts
e-mail: rkacmarek{at}partners.org
High-frequency ventilation has been around for a long time, in fact the first patent for high-frequency oscillation (HFO) was issued to Jack Emerson in 1959 (7). It is now almost 60 years since that patent was issued and still HFO has not become the accepted physiological approach for the management of adult ARDS! This is despite the fact that for decades during conventional ventilation (CV) we administered tidal volumes as high as 24 ml/kg actual body weight (9). In addition, epidemiologic data from Esteban et al. (8) indicate that during CV PEEP levels still appear to be inadequate to prevent injury from the repetitive opening and closing of unstable lung units.
Why has HFO not been able to establish itself as the optimal approach to ventilatory support in ARDS? It is true that theoretically HFO should be more lung protective then CV. Tidal stretch is dramatically limited, tidal volume even in adults is only 1–3 ml/kg PBW, and mean airway pressures are sufficient to avoid injury from recruitment/derecruitment and in most applications peak alveolar pressure can be maintained relatively low. In fact, all of the early animal data support this concept, study after study indicated that HFO was superior to CV (10, 15), a widely referenced study by McCulloch et al. (15), HFO was applied with an open lung approach at a mean airway pressure (MAP) to achieve a PaO2 in lung injured animals of >350 mmHg. But during CV, PEEP was set at 7.5 cmH2O and plateau pressure was not limited ranging from 27 to 37 cmH2O. With CV and a third arm using a low MAP approach to HFO, PEEP or MAP were titrated to achieve a PaO2 of 70–100. The results of this study were predictable: all outcome measurements, gas exchanges, lung mechanics, and histology favored open lung HFO. In shape contrast are animal studies by Vazquez et al. (26), Rimensberger et al. (20), and Sedeek et al. (23) where the approach to CV and HFO were similar. Both Vazquez and Rimensberger studied small animals (rats and rabbits) and were unable to identify any differences between CV and HFO. Sedeek similarly used an open lung approach in 40-kg sheep. After lung recruitment to a peak pressure of 50 cmH2O in both groups, PEEP or MAP were set with a decremental trial. The optimal PEEP level was that resulting in the best oxygenation. After 4 h of ventilation gas exchange, lung mechanics, leukocyte, or polymorphonuclear cell counts in pulmonary lavage fluid and pulmonary cytokine levels were similar. Histologic examination of lung tissue was also similar except that with HFO there was significantly less interstitial hemorrhage and alveolar septal expansion. However, the CV group was ventilated with a tidal volume of 9 ml/kg and plateau pressures throughout the study were above 30 cmH2O. From these data it appears that CV is as effective as HFO in preventing lung injury provided peak alveolar pressure is maintained below 30 cmH2O and PEEP is appropriately set to avoid recruitment/derecruitment injury.
An examination of the neonatal HFO vs. CV randomized control trials (RCT) yields the same equivocal results (3–5, 11–14, 16–19, 21, 22, 24, 25, 27). It is important to note that not a single RCT comparing high-frequency ventilation to conventional ventilation regardless of age group has yielded a mortality benefit to either approach! In neonates, HFO has been favored by many because of the perceived benefit of avoiding the development of chronic lung disease. This is only true for the RCTs in which CV was applied without the use of a lung protection approach (3, 11, 18). In those trials where HFO and CV were both applied without (19) lung protection or with (4, 13, 16, 24) the same lung protective strategy, no benefit from HFO has been identified! Similarly, in those RCTs in which HFO was applied without a high MAP, intraventricular hemorrhage and preventricular leukomalacia were more frequent than with CV (12, 19). However, this benefit of CV disappeared when both were applied without (3, 11, 18) a lung protective approach or with (4, 13, 16, 24) a similar lung protective approach.
The data in pediatrics are no better than that in neonates. A single RCT conducted by Arnold et al. (1) compared HFO to CV. Enrolled patients had an oxygenation index >13 on two consecutive arterial blood gas measurements over 6 h apart. A total of 58 well-matched patients were randomized. No differences existed between groups for duration of mechanical ventilation, frequency of air leak, or 30-day survival. However, fewer HFO patients (4 vs. 10; P < 0.039) required supplemental oxygen at 30 days. Unfortunately, it is impossible to determine why this difference occurred, because no data whatsoever regarding actual application of mechanical ventilation is provided; no PEEP, peak, or plateau pressure, no tidal volume or rate, and no FIO2! Considering the publication date of this study (1994), I can only assume that tidal volumes were 
10 ml/kg PBW and plateau pressures exceeded 30 cmH2O.
Two RCTs comparing HFO to CV in adults have been published, both negative (2, 6), Derdak et al. (6) randomized 148 patients to HFO or CV. Enrolled patients did have severe ARDS; PaO2/FIO2 < 200 on 10 cmH2O PEEP. There were no significant differences in any measured outcome variables; 30 day mortality, number of patients alive without mechanical ventilation at 30 days, barotrauma, or mucus plugging of the endotracheal tube. There was, however, a trend toward better survival (63% vs. 48%) with HFO. However, the approach to CV was not lung protective, tidal volumes were 10 ml/kg PBW and plateau pressure (pressure control ventilation with long inspiratory times) averaged 37 cmH2O. In addition, patients were enrolled late in the course of ARDS and some of the patients in the CV arm were ventilated for >2 wk before trial entry (2.7 ± 2.7 days HFO vs. 4.4 ± 7.8 days).
The second adult RCT was stopped early because of enrollment problems (2). A total of 61 well-matched adults was randomized. Enrolled patients simply had to meet ARDS criteria. There were no differences in mortality, survival with O2 or ventilatory support, therapy failure, air leak, or need for O2 therapy at 30 days. Contrary to the Derdak et al. (6) study, the survival trend favored CV (77% vs. 67%). This was despite the fact that this was an early intervention not a rescue trial (2.1 ± 2.6 days HFO vs. 1.5 ± 1.8 days CV on mechanical ventilation before randomization) and that in CV tidal volumes averaged 8–9 ml/kg PBW and plateau pressure were 30 cmH2O (pressure control ventilation).
In conclusion, there are no data to support the hypothesis that HFO is the optimal approach to ventilator ARDS patients! In none of the RCTs to date was conventional ventilation performed with an open lung protective strategy. No lung recruitment was used, PEEP was not set based on the patient's lung mechanics, and both tidal volumes and plateau pressures were higher than current recommendation, yet no benefits of HFO have been demonstrated. Considering the data available it is hard to imagine that HFO would perform better than in the past compared with our best approach to CV. Correctly and optimally applied HFO does work but not any better than even poorly applied CV!
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R. M. Muellenbach, C. Wunder, J. Brederlau, J. Villar, A. T. Rotta, S. Maruvada, G. Vento, M. Tana, C. Tirone, V. Vendettuoli, et al. High-frequency ventilation is/is not the optimal physiological approach to ventilate ARDS patients. J Appl Physiol, April 1, 2008; 104(4): 1236 - 1237. [Full Text] [PDF]
|
|
|
|
|
|
N. Ferguson and A. Slutsky Last Word on Point:Counterpoint: High-frequency ventilation is/is not the optimal physiological approach to ventilate ARDS patients J Appl Physiol, April 1, 2008; 104(4): 1240 - 1240. [Full Text] [PDF]
|
|
|
|
|
|
R. M. Kacmarek Last Word on Point:Counterpoint: High-frequency ventilation is/is not the optimal physiological approach to ventilate ARDS patients J Appl Physiol, April 1, 2008; 104(4): 1241 - 1241. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
