Does Automated Closed-Loop Ventilation Reduce the DRiving Pressure Levels in Patients With ARDS (AiRDRoP) - a Single-Center Crossover Study and Randomized Controlled Trial
Does Automated Closed-Loop Ventilation Reduce the DRiving Pressure Levels in Patients With ARDS (AiRDRoP) - a Single-Center Crossover Study and Randomized Controlled Trial
The objective of this crossover study and randomized controlled trial (RCT) is to compare ΔP levels during INTELLiVENT®-ASV with conventional lung protective ventilation in the acute and sub-acute phase of moderate or severe ARDS.
A total of 48 adult patients admitted to intensive care units with moderate or severe ARDS will be included. In the acute phase patients will receive 4 hours of INTELLiVENT-ASV ventilation and 4 hours of conventional lung protective ventilation in random order. After these two blocks the patients are allocated into either the INTELLiVENT-ASV arm or the conventional lung protective ventilation arm.
in the sub-acute phase patients will be assessed every day until day 7 or extubation, whichever comes first.
Primary endpoint is the transpulmonary transpulmonary (ΔP). Secondary endpoints of both studies include other ventilator settings and ventilation parameters, as well as time spent at a ΔP level of 15 cm H2O or higher.
- Objective:
The objective of this crossover study and randomized controlled trial (RCT) is to compare ΔP levels during INTELLiVENT®- adaptive support ventilation (ASV) with conventional lung protective ventilation in the acute and sub-acute phase of moderate or severe ARDS.
- Study design:
Single center crossover study (in the acute phase) and RCT (in the sub-acute phase).
- Study population:
The study population consists of 48 consecutive intubated and ventilation patients with moderate or severe ARDS according to the Berlin definition with an anticipated duration of mechanical ventilation of > 24 hours.
- Sample size calculation:
The sample size is computed based on the hypothesis that ventilation with INTELLiVENT®-ASV is associated with a reduction in the ΔP level of 5 cm H2O. We based this power calculation on unpublished data from a published cohort of ARDS patients29 and data from an abstract31.
For the RCT, a sample size of 20 patients in each group has 90% statistical power to detect a difference in the ΔP level of 5 cm H2O, with means of 15 and 10 cm H2O, respectively, assuming an effect size (f) of 0.4 using a repeated measures ANOVA with a 0.05 two-sided significance level.
The sample size is increased by 20% to correct for early dropouts (i.e., before day 7) and patients in whom it is impossible to measure the ΔP level, meaning that each group will contain 24 patients. Therefore, the total sample size of this study will be 48 patients.
We expect less variation in the crossover than in the RCT. Next, in this part of the study patients serve as their own control. Therefore, with this sample size we will have sufficient power for the crossover.
- Patient allocation:
Crossover study - Data will be collected during two blocks of four hours; one block of conventional lung protective ventilation, and one block of INTELLiVENT®-ASV. The two blocks will take place in randomized order. Both ventilation strategies are frequently used in our ICU. INTELLiVENT®-ASV is a relatively new ventilatory mode that was successfully implemented in our intensive care unit in 2016, especially in patients with ARDS.
Randomized controlled trial - After the crossover, patients are randomly allocated to conventional lung protective ventilation5, meaning that the ventilator is set to pressure controlled mode, or INTELLiVENT®-ASV.
- Ventilator settings:
Cross-over study
During conventional lung protective ventilation, the attending physician sets the ventilator according to the local ventilation protocol. The fraction of inspired oxygen (FiO2) is adjusted to maintain an oxygen saturation of 92 to 96% and/or a PaO2 of > 8 kPa. The respiratory rate is adjusted to maintain a blood pH of 7.25 to 7.45. The lowest level of positive end-expiratory pressure (PEEP) is 5 cm H2O; allowed PEEP-FiO2 are in concordance with the ARDS network recommendations;5 these combinations are based on two large RCTs in ARDS patients, and is standard practice in our unit.
During INTELLiVENT®-ASV, the attending physician sets the target for et-CO2 to maintain a blood pH of 7.25 to 7.45. The target for oxygen saturation is set at 92- 96%. The lowest level of PEEP is kept at 5 cm H2O.
Of note, INTELLiVENT®-ASV is available on all ventilators used in patients with ARDS. Since both ventilation strategies can be applied with these ventilators, there is no need to disconnect a patient from the ventilator. Also, switching between ventilatory modes is a standard procedure in our ICU. However, for the purpose of this study it is protocolized.
Randomized controlled trial
Settings are similar for the RCT as for the crossover study.
- Statistical analysis:
Primary study parameters
The primary outcome, the (transpulmonary) ΔP level, is analyzed using a repeated measures ANOVA. The effect mediation of ventilator parameters on the (transpulmonary) ΔP is analyzed by mediation analysis. P-values of 0.05 are used for statistical significance. When appropriate, statistical uncertainty will be expressed by the 95% confidence levels. All statistical analysis will be performed with the R language and environment for statistical computing.
Secondary study parameters
Continuous normally distributed variables will be expressed by their mean and standard deviation or when not normally distributed as medians and their interquartile ranges. Categorical variables will be expressed as n (%). To test groups Student's t test will be used, if continuous data is not normally distributed the Mann-Whitney U test will be used. Categorical variables will be compared with the Chi-square test or Fisher's exact tests. Time dependent data will be analyzed using a proportional hazard model adjusted for possible imbalances of patients' baseline characteristics. Analysis will be performed with R statistics version 3.0.2. Patient characteristics will be compared and described by appropriate statistics.
Inclusion Criteria:
Exclusion Criteria: