Combined High Frequency Oscillation and Tracheal Gas Insufflation for Severe Acute Respiratory Distress Syndrome

OBJECTIVES AND UNDERLYING HYPOTHESIS High frequency oscillation (HFO) has beneficial physiological effects in acute respiratory distress syndrome (ARDS) \[1-4\]. A recent randomized controlled trial of HFO versus conventional mechanical ventilation (CMV) with traditional tidal volumes (10.1± 2.8 mL•kg-1 predicted body weight) reported a trend toward reduced 30-day mortality in the HFO arm (37% versus 52% in the CMV group; P = 0.10) \[3\]. To-date, HFO has not been compared to CMV with low tidal volumes (6.2 ± 1.1 mL•kg-1 predicted body weight) in ARDS \[5\], with respect to respiratory physiology and clinical course. The main mechanisms of gas exchange during CMV are bulk convection and diffusion. \[4, 6\]. Tracheal gas insufflation (TGI) promotes CO2 elimination during CMV \[7-9\]. During HFO, additional gas exchange mechanisms comprise asymmetric velocity profiles, Taylor dispersion/turbulence, cardiogenic mixing, pendelluft effect, and collateral ventilation \[4, 6\]. We have recently shown that in patients with moderate-to-severe, primary ARDS, HFO combined with TGI (HFO-TGI) substantially improves oxygenation relative to both standard HFO and CMV according to the ARDS network protocol \[5, 10\]. Mean airway pressure (mPaw) was set at 1 cm H2O above the point of maximal curvature (PMC) of the expiratory pressure volume curve. During HFO-TGI, it is probably feasible to reduce mPaw, while still achieving PaO2 and peripheral oxygen saturation (SaO2) targets similar to those set by the ARDS network protocol \[5\]. This may result in reduced and non-traumatic ventilation pressures during HFO-TGI. Furthermore, if HFO-TGI-related gas exchange benefits are maintained during post-HFO-TGI CMV \[10\], then a reduction of ventilation pressures relative to the pre-HFO-TGI CMV may be feasible as well. However, the use of reduced ventilation pressures may minimize ventilator associated lung injury. In this study, we will test the hypothesis that HFO-TGI may improve the respiratory physiology and clinical course of patients with severe ARDS. METHODS Patients The protocol has been approved by the Scientific Committee of Evaggelismos Hospital. Informed consent will be requested from the next-of-kin of participation-eligible patients. Eligible patients should fulfill the criteria: 1) early (diagnosis established within the preceding 72 h) ARDS according to the American-European Consensus Conference Criteria \[11\]; 2) severe oxygenation disturbances {defined as PaO2/inspired oxygen fraction (FiO2) \< 150 mm Hg}, while ventilated with a positive end-expiratory pressure (PEEP) of ≥ 8 cm H2O (criterion for severe ARDS); 3) age 18-75 years, body weight \> 40 kg, and absence of a) severe air leak (i.e. \> 1 chest tube per hemithorax with a persistent airleak for \> 72 h); b) systolic arterial pressure of \< 90 mm Hg while receiving fluids and norepinephrine at ≥ 0.5 μg/kg/min; c) significant heart disease (i.e. ejection fraction \< 40 %, history of pulmonary edema, and active coronary ischemia or myocardial infarction); d) significant chronic obstructive pulmonary disease (COPD) or asthma \[10\]; e) intracranial abnormalities causing uncontrollable intracranial hypertension; f) chronic interstitial lung disease associated with bilateral pulmonary infiltrates; g) lung biopsy or resection on current admission; h) previous lung or bone marrow transplant, or presence of immunosuppression; i) inability to wean from prone positioning or inhaled nitric oxide; k) pregnancy or morbid obesity (i.e. body mass index \> 40 kg/m2); and l) enrollment in another interventional study. Patient monitoring will include electrocardiographic lead II, hemodynamics, and SaO2. Deep sedation/neuromuscular blockade will be employed as previously described \[10, 12\]. CMV strategy In the 37-bed intensive care unit (ICU) of Evaggelismos hospital, a ventilatory strategy similar to the ARDS network protocol is routinely employed. Consequently, before randomization, patients will already be ventilated (Siemens 300C ventilator; or Galileo Gold, Hamilton Medical) with one of the network-protocol allowable combinations of FiO2 and PEEP \[5\]. Administered tidal volumes will be 5.5-7.5 mL•kg-1 predicted body weight, ventilatory rate (e.g. 20-35•min-1) will be adjusted so that pHa is kept within 7.20-7.45; the inspiration to expiration duration ratio will be 1:1 to 1:3; and the target plateau inspiratory pressure will be ≤ 30-35 cm H2O. Oxygenation goals will be PaO2 = 55-80 mm Hg or SaO2 = 88-95%. Randomization Patients will be randomly assigned to either the CMV-group or the HFO-TGI-group. CMV-group patients will continue to receive CMV as described above. In the HFO-TGI-group, 30 min prior to HFO-TGI initiation, an inspiratory and an expiratory pressure volume curve will be consecutively constructed with the linear pressure ramp technique (whenever clinically feasible \[10, 13\]. HFO-TGI strategy The Sensormedics 3100B ventilator \[10\] will be connected to the endotracheal tube. Also, a thin Vygon TGI catheter {internal / external diameter = 1.0 / 2.0 mm, respectively \[10\]} will be introduced into the endotracheal tube. TGI catheter tip will be placed 0.5-1 cm beyond the endotracheal tube tip. The proximal end of the TGI catheter will be connected to an O2 flowmeter. Initial HFO settings will be as follows: 1) FiO2 = 100%, later-on titrated toward the FiO2 of preceding CMV; 2) bias flow = 30 L•min-1; 3) oscillation frequency = 4.0-5.0 Hz; 4) oscillatory pressure amplitude = arithmetical PaCO2 value during preceding CMV + 20-30 cm H2O, maximal acceptable value = 95-100 cm H2O \[10\]; 5) inspiratory to expiratory time ratio = 1:2; and 6) mPaw adjusted so that mean tracheal pressure = 2-3 cm H2O above the mean tracheal pressure of the preceding CMV (corresponding to an HFO mPaw of 9-10 cm H2O above the preceding CMV mPaw). Sixty to 120 secs after HFO initiation, a recruitment maneuver will be performed by pressurizing the HFO circuit at 40-45 cm H2O for 20-30 secs with oscillator piston off. HFO will then be resumed, a 3-5 cm H2O endotracheal tube cuff leak will be placed. Immediately thereafter, continuous forward thrust TGI (flow = 50% of the minute ventilation of the preceding CMV \[10\]) will be superimposed on the HFO.The mPaw control knob will then be adjusted to return mPaw to its originally set value. Fifteen min thereafter, arterial blood gas analysis will be performed and oscillatory pressure amplitude and oscillation frequency will be adjusted, in order to achieve a PaCO2 of \< 10-15 mm Hg above the PaCO2 of the preceding CMV and to keep pHa \> 7.20. After another 30-60 min, blood gas analysis will be repeated, and then, mPaw will be gradually reduced (estimated reduction rate: 0-1 cm H2O•h-1) toward a target mean tracheal pressure of up to 2-3 cm H2O lower than the mean tracheal pressure of the preceding CMV (corresponding to an HFO mPaw of 3-4 cm H2O above preceding CMV mPaw). Subsequently, TGI will be discontinued and standard HFO will be continued for 30 min. The aforementioned adjustments should permit maintenance of SaO2 = 88-95% or PaO2 = 55-80 mm Hg and of the above-described target PaCO2/pHa. Return to CMV will be considered at the following HFO settings: mPaw adjusted so that HFO-mean tracheal pressure = up to 2-3 cm H2O lower than tracheal pressure of preceding CMV (corresponding to an HFO mPaw of 3-4 cm H2O above preceding CMV mPaw); FiO2 = FiO2 of preceding CMV; and TGI = 0 L•min-1. CMV will be resumed and continued, provided that PaO2/FiO2 is maintained at ≥ 150 mm Hg at a PEEP of ≥ 8 cm H2O. Return to HFO-TGI will be considered necessary if after 12-16 h of CMV, PaO2/FiO2 is \< 150 mm Hg at a PEEP of ≥ 8 cm H2O. In the HFO-TGI-group, patients will receive repeated daily sessions of HFO-TGI until they no longer satisfy the severe ARDS criterion during CMV for \> 24 h. Predicted minimum duration of HFO-TGI sessions is 6 h. For HFO-TGI-group members, total HFO-TGI duration will have to be \> 12 h. Recruitment Maneuvers: Recruitment maneuvers (with continuous positive airway pressure of 40-50 cm H2O and a duration of 20-30 secs) will be performed as follows: In the HFO-TGI group, recruitment maneuvers will be performed at the onset and at 3 h following the onset of each HFO-TGI session, and just prior to return to CMV. In the CMV-group, \> 3 recruitment maneuvers (one every 3 h; first maneuver at 9:00 am) will be performed daily. In both groups, recruitment maneuvers will be continued as part of the early intervention period of the study protocol until resolution of severe ARDS (if achieved) or death. Weaning from CMV: Weaning from CMV will be by pressure-supported ventilation when a PaO2 of ≥ 60 mm Hg can be maintained at an FiO2 of ≤ 50% and a PEEP of ≤ 8 cm H2O. Data Collection Data on demographic, physiologic, and radiographic characteristics, coexisting conditions, and medication will be recorded within 4 h prior to randomization. Physiologic, laboratory, and radiographic/imaging data will be collected daily up to 28 days post-randomization. Patients will be monitored daily for signs of failure of nonpulmonary organs and systems. Outcome Measures are reported in the dedicated section. In concordance with a suggestion of a recent Editorial (Intensive Care Med (2014) 40:743-745), the Original and (its revision to) the Final Form of the Study Protocol (also corresponding to NCT00637507) detailing the Pre-specified Study Planning (which explains the reason for any prior changes in the current registration data) can be found by scrolling down to the end of the following webpage: http://www.evaggelismos-hosp.gr/0010000688/%CE%B9%CF%83%CF%84%CE%BF%CF%81%CE%B9%CE%BA%CE%BF-%CE%B5%CE%B5%CF%80%CE%BD%CE%B5.html