By Richard H. Kallet, MS, RRT, FAARC, FCCM, Director of Quality Assurance, Respiratory Care Services, San Francisco General Hospital
Mr. Kallet reports no financial relationships relevant to this field of study.
Synopsis: Two large, prospective, multicenter, randomized, controlled trails (OSILLATE and OSCAR) compared high-frequency oscillatory ventilation (HFOV) to either a high-PEEP approach to open-lung ventilation or usual care (lung-protective ventilation). In the first study, HFOV using relatively high mean airway pressures was stopped early due to significantly higher hospital mortality in the HFOV group. The second study found no difference in mortality.
Sources: Ferguson ND, et al. High-frequency oscillation in early acute respiratory distress syndrome. N Engl J Med 2013;368:795-805. Young D, et al. High-frequency oscillation for acute respiratory distress syndrome. N Engl J Med 2013;368:806-813.
These two reports of large, multicenter, randomized, controlled trials of high-frequency oscillatory ventilation (HFOV) as compared to conventional lung-protective ventilation (LPV) in patients with the acute respiratory distress syndrome (ARDS) were electronically published on the same day in The New England Journal of Medicine. The first, the Oscillation for Acute Respiratory Distress Syndrome Treated Early (OSCILLATE) trial, was conducted primarily in Canada, while the second, the Oscillation in ARDS (OSCAR) trial, was carried out at 29 hospitals in the United Kingdom.
The OSCILLATE trial used explicit ventilator management protocols for both the HFOV arm and the control arm, which consisted of a high positive end-expiratory pressure (PEEP), low tidal volume (Vt) strategy. There were no differences in baseline characteristics between treatment groups. The study was stopped early because of significantly higher mortality in the HFOV vs the control group (47% vs 35%, relative risk = 1.33 [confidence interval, 1.09-1.64]; P = 0.005). A larger proportion of patients in the HFOV arm required both vasopressor support and paralysis for longer duration, as well as more sedation. There also was a non-significant trend toward higher positive fluid balance among patients in the HFOV arm.
In contrast, the OSCAR trial did not find a difference in mortality between the HFOV and control group (41.7 vs 41.1% respectively, P = 0.85); neither was there a difference in requirements for sedatives, neuromuscular blocking agents, or vasopressors. There were no differences in baseline characteristics between groups. An explicit protocol was used for the HFOV arm, whereas patients in the control arm were managed by usual care, although the use of a Vt between 6-8 mL/kg ideal body weight and adherence to the NIH ARDSNet PEEP/FIO2 grid1 was encouraged. In each study, the median duration of HFOV was 3 days and the primary outcome was mortality.
In theory, HFOV seemed the ideal LPV mode for managing severe ARDS. Therefore, it is surprising that in these two large, well-designed, well-executed studies enrolling more than 1300 patients with moderate-to-severe ARDS, HFOV failed to improve meaningful outcomes. And in the case of the OSCILLATE trial, HFOV paradoxically was harmful. In retrospect, HFOV may be too simplistic an approach to mechanical ventilation, given the complexity of both pulmonary mechanics in ARDS and the mechanisms governing ventilator-induced lung injury (VILI).
The complexity of managing severe ARDS also includes other related factors. Survival is negatively influenced by the inability to clear pulmonary edema and by the development of acute cor pulmonale. Mortality associated with cor pulmonale in ARDS appears highly influenced by the magnitude of lung stress.2 Excessive airway pressures cause alveolar overdistension, diminished venous return, and increased right ventricular afterload that requires intravascular volume expansion to maintain perfusion, and thus may worsen pulmonary edema. Furthermore, lung injury in ARDS is heterogeneously distributed in a manner determined by the mechanism and severity of injury as well as abnormalities in chest wall compliance. Thus, the degree of lung recruitment vs overdistension, as well as hemodynamic compromise and fluid requirements, may vary considerably between patients.
Hemodynamic compromise is the archetypal signifier for pulmonary overdistension, and its prominence during HFOV in the OSCILLATE trial suggests a plausible, partial explanation for the higher mortality. The most cogent explanation of VILI (i.e., strain-stress theory) proposes that excessive end-inspiratory volume (or strain) relative to functional residual capacity causes abnormal stress to develop resulting in alveolar cell membrane failure and secondary inflammation. In addition, non- or poorly aerated areas adjoining normal lung tissue serve as “stress amplifiers” magnifying the inflammatory response because of sheering phenomena. However, the theory also posits that the duration of excessive stress is likely as important as the magnitude.
Therefore, in a heterogeneously injured lung with variable recruitability and a potentially wide spectrum of regional compliances and resistances (i.e., time constant distributions), less-injured and normal lung tissue alike are vulnerable to sustained overdistension during HFOV. Paradoxically, this could accelerate the development of VILI. For perspective, it is sobering that a 6 mL/kg Vt with moderate PEEP levels and a Pplat under 30 cm H2O nonetheless causes regional tidal hyperinflation and increased markers of inflammation.3
An inherent limitation of HFOV is the inability to monitor variables useful for assessing both lung recruitment and overdistension: plateau pressure, compliance, and dead space. Some newer ventilators already can measure functional residual capacity (and therefore can calculate a strain index). In the near future, other useful measurements for titrating LPV such as slope of Phase 3 (a volumetric capnography measure of ventilation-perfusion heterogeneity) and intra-tidal compliance (SLICE) should become available if clinical demand increases. Given the pathophysiologic complexity and variability of ARDS, it is more likely that titrating lung-protective ventilation according to both traditional and sophisticated measures of chest mechanics and gas exchange will further improve outcomes in ARDS rather than novel modes of ventilation.