Comparison of Cardio-respiratory Variables Between Low-frequency High Frequency Oscillation With/Without Tracheal Gas Insufflation and High-frequency High-frequency Oscillation in Severe ARDS.

Rationale of the study Acute Respiratory Distress Syndrome (ARDS) is an acute inflammatory state of the pulmonary parenchyma that causes hypoxemia, atelectasis, pulmonary congestion, and reduction in pulmonary compliance. Mechanical ventilation is actually life-saving, but it may traumatize the lungs (e.g. volutrauma, barotrauma, atelectrauma, and biotrauma). The use of low tidal volumes and high positive end-expiratory pressure (PEEP), aims at attenuating ventilator-associated lung injury. However, ARDS mortality still remains high. High frequency oscillation (HFO) is an alternative ventilatory technique that employs very low tidal volumes (1-4 ml/kg) administered at high frequencies (3-15 Hz). Prior observational studies have reported improvements in oxygenation, whereas recent two-center data on severe ARDS suggest a survival benefit from the intermittent, combined use of low-frequency HFO with a cuff leak, Recruitment maneuvers (RMs), and tracheal gas insufflation (TGI). The addition of TGI improves oxygenation and CO2 elimination; however, it is still unclear whether it affects survival. Two recently published multicenter studies showed either neutral (10) or negative results (11) with respect to survival when high frequency HFO without a cuff leak was used in the treatment of early ARDS. However, these negative results may be partly due to right ventricular overload/dysfunction/failure caused by the combination of high intrathoracic pressures and hypercapnia, with consequent hemodynamic instability and increased need of inotropic/vasopressor support. Accordingly, the investigators hypothesize that a different HFO strategy [employing a combination of a low frequency and a cuff leak - which augments CO2 elimination and is associated with relatively low mean tracheal pressures - could lead to different results.

A high intrathoracic pressure may impede venous return and increase pulmonary vascular resistance. This concurrent right ventricular preload reduction and afterload increase might cause right ventricular dysfunction/failure. On the other hand, the placement of a cuff leak results in a lower (by approx. 5-6 cmH2O) mean tracheal pressure relative to the set HFO-ventilator mean airway pressure (mPaw), with consequent unloading of the right ventricle. In addition, further unloading of the right ventricle can be expected through the combined use of a cuff leak, TGI, and a high HFO bias flow; these measures improve PaCO2 control and may prevent excessive, hypercapnia-induced increases in pulmonary vascular resistance. Right ventricular function can be assessed by transesophageal echocardiography (TEE) as previously described. Right ventricular dysfunction and associated dilatation may cause cardiac output reduction and coronary hypoperfusion; the latter may further compromise right ventricular performance and contribute to right ventricular failure.

The main goal of the present study is to document and compare the effect of lung protective conventional mechanical ventilation (CMV) and of different HFO strategies (already clinically tested in trials with conflicting results) on right ventricular performance as determined by TEE. More specifically, we intend to compare high-PEEP, lung protective CMV with a "high" mPaw/no cuff leak, high-frequency HFO strategy, and a "high" mPaw with cuff leak, "low" frequency HFO strategy with and without TGI.

Methods METHODS Eligible patients (relevant criteria provided below) with early and severe ARDS will be enrolled after the obtainment of informed, written, next-of-kin consent, as well as attending physician's non-written consent..

Patient monitoring will include electrocardiographic lead II, intra-arterial pressure (+/- cardiac index with PICCO plus, Pulsion Medical Systems, Munich, Germany)], and peripheral oxygen saturation (SaO2). Anesthesia will be maintained with midazolam and/or propofol, and fentanyl or remifentanil. Nrutomuscular blockade will be accomplished with cisatracurium, which will be used in concordance with current recommendations and as part of the attending physician-prescribed medical treatment. During the study period, all patients will be receiving a continuous infusion of cisatracurium.