Effect of Synchronized vs. Continuous High Flow Nasal Cannula Using Neurally Adjusted Ventilatory Assist on Work of Breathing in Infants With Bronchopulmonary Dysplasia

Prior to the study period each subject will have the following monitoring equipment placed (if not already present): Edi catheter, transcutaneous monitor (TCM4, Radiometer, Brea, CA, USA) to measure CO2 and O2 levels, pulse oximeter probe (MasimoSET, Irvine, CA, USA) to measure oxygen saturations and heart rate, and RIP bands (SleepSense, MFI Medical, San Diego, CA, USA) around the chest and abdomen to measure breathing movements and relative tidal volume.

Each subject will be randomized on the day the study is to occur to begin with either NAVA-synchronized or continuous HFNC before crossing over to the other mode to serve as his/her own control. The same RAM cannula will be used in both study arms and will provide a leak of 60-80% as recommended by the product manual. The delivery of high flow during both synchronized and continuous HFNC will be given at two commonly provided levels of high flow: 6 LPM and 8 LPM, given in the same order in each mode (6 LPM then 8 LPM). Each subject will receive 15-minute trials of each mode-level combination, for a total of four trials. During each trial, the first 10 minutes will be used for stabilization, and the last 5 minutes will be used for data collection, as has been done in previous trials. Thus, the mode-level combinations of the trials will be as follows: for infants randomized to begin with synchronized support: synchronized-6 LPM, synchronized-8 LPM, unsynchronized-6 LPM, unsynchronized-8 LPM. For infants randomized to begin with unsynchronized support: unsynchronized-6 LPM, unsynchronized-8 LPM, synchronized-6 LPM, unsynchronized-8 LPM.

The flows described in the NAVA-synchronized trials refer to the peak flow provided during inspiration. A baseline flow rate of 2 LPM will be provided expiration in these trials (using the PEEP setting corresponding to the appropriate flow rate). During the unsynchronized trials, the continuous high flow rate will be provided (as is common practice with the use of HFNC). During the NAVA-synchronized HFNC trials, the Edi trigger will be set to 0.5 microvolts, apnea time to 5 seconds, back up rate to 10 breaths per minute, and backup pressure settings will be set to provide an estimated peak flow of 6 or 8 LPM according to the designated trial (again, using the pressure setting corresponding to the appropriate flow rate).

During NAVA-synchronized HFNC, the NIV NAVA mode will be set in such a way that synchronized HFNC will be provided. A minimal end-expiratory flow of 2 LPM will be provided using the positive end expiratory pressure (PEEP) setting in the NIV NAVA mode on the ventilator. The PEEP setting corresponding to 2 LPM via the pneumotachograph will be used. In order to deliver the desired peak flow rate with each neurally-triggered breath, a NAVA level of 15 cmH2O/μV will be set, then the maximum pressure setting that corresponds to the desired flow rate using the pneumotachograph will be used for the study. The subject will thus be provided with "synchronized HFNC". This contrasts with the constant-flow trials when subjects will receive a constant and non-synchronized flow using the HFNC software on the ventilator.

Servo-u ventilators and Servo Tracker Software (Maquet Critical Care, Solna, Sweden) will be used in order to track Edi signal. The MP100 Biopac data acquisition (Biopac Systems Inc., Goleta, CA, USA) will be used to collect data from the monitoring devices. HFNC will be delivered using appropriately sized RAM cannula to allow for air leak around the subject's nares. Persistent bradycardia (less than 100 beats per minutes), desaturation (<85%), or hypercarbia (transcutaneous CO2 >70) will result in cessation of the study.