Effects of PERMISSive Lung-protective Ventilation on Outcome in Critically Ill Invasively Ventilated Patients

RATIONALE Lung-protective ventilation using a lower respiratory rate (RR) is an appealing strategy to reduce ventilation intensity, which may require permissive hypercapnia. However, the feasibility and safety of this so-called 'permissive lung-protective ventilation' must be investigated, before conducting a large randomized clinical trial to evaluate its effectiveness on patient-centered outcomes. OBJECTIVE To study the feasibility and safety of permissive lung-protective ventilation in adult critically ill patients receiving invasive ventilation for acute hypoxemic respiratory failure, and to inform the design of a future randomized clinical trial in this patient population. HYPOTHESIS Permissive lung-protective ventilation is a feasible and safe ventilation strategy. STUDY DESIGN Multicenter, randomized clinical pilot trial. STUDY POPULATION Critically ill patients, aged \> 18 years, intubated for acute hypoxemic respiratory failure, and expected to receive ventilation for \> 24 hours. METHODS Patients are randomized to permissive lung-protective ventilation wherein RR is stepwise reduced, or to conventional lung-protective ventilation. OUTCOME MEASURES The primary endpoint is feasibility, assessed by the difference in respiratory rate (RR) between the two groups, from the start of mechanical ventilation until first extubation. Secondary endpoints include protocol compliance and feasibility of collecting data, and safety, assessed by the occurrence of unacceptable hypercapnia and hypoxemia and the incidence of ventilator-associated complications SAMPLE SIZE To estimate the appropriate sample size for this pilot study, we considered the primary feasibility endpoint of detecting a difference in the respiratory rate (RR). Assuming an expected mean difference in RR of 7.5, based on previous studies \[1, 2\], with an SD of 10, a power of 90% and an alpha of 0.05, with a drop-out rate estimated at 10%, a two-tailed t-test was used. The required sample size is 84