Materials and Methods
A prospective observational study took place from March 2020 to December 2020 in the level III neonatal intensive care unit of Turku University Hospital, Finland. We included preterm infants born before 36 weeks of gestation who received at least 24 hours of ventilatory support with invasive or noninvasive NAVA using Servo-i or Servo-n ventilators (Getinge, Solna, Sweden).
As NAVA was the primary ventilation mode in this unit, for intubated preterm infants, we began NAVA whenever the infants had spontaneous breathing. As soon as possible we extubated them and applied noninvasive NAVA. Before the start of NAVA mode, a standard feeding tube was replaced with a dedicated electrode-equipped catheter to detect Edi (Edi catheter, Getinge, Solna, Sweden). The trigger Edi level was set as 0.5 μV above the minimum Edi. Attending clinicians decided and adjusted the NAVA level, positive end-expiratory pressure (PEEP), fraction of inspired oxygen (FiO2) and backup ventilation (pressure control mode) settings according to the infant’s condition. The apnea time was set as 2 to 10 seconds. After the set apnea time without the Edi signal, the backup ventilation started with a preset peak pressure and frequency until the Edi trigger resumed. Our usual backup rate was set as 20 to 40 breaths/min and the inspiratory time was set as 0.4 to 0.5 seconds. The safety limit for peak inspiratory pressure (PIP) was set as 25 to 35 cmH2O. If the infant lost his/her cardiorespiratory stability during NAVA, the ventilator mode was changed, and this episode was defined as ‘clinical deterioration.’
For the infant participating in the study, ventilator settings and respiratory variables were recorded every day and exported to a specific computer using Servo Record Viewer version 1.0 (Maquet Critical Care AB, Getinge, Solna, Sweden). The collected data provided the values for PIP, mean airway pressure, PEEP, expiratory tidal volume, peak Edi, minimum Edi, measured respiratory rate (RR), neural respiratory rate (nRR) and percentage time spent in backup ventilation for each minute. All the ventilator data were inspected and compared with the event logs recorded automatically from the ventilator, which included all alarm notifications, mode and setting changes, cable connections and disconnections. Data during disconnection of the Edi cable, malfunction or dislocation of the Edi catheter were excluded from the analysis. The mean values during each day were computed for each ventilatory variable. If there was a change in the ventilatory setting parameters, we chose the parameter that was applied for a longer duration in the 24-hour time period. For clinical deterioration, ventilator data 30 hours prior to the episode were reviewed at 6-hour intervals. The zero time point was defined as the moment when the ventilator mode was switched from NAVA to others because of clinical deterioration.
The following data were collected from medical records: gestational age at birth, birth weight, sex, postmenstrual age (PMA) and body weight at the study date. For clinical deterioration, information about its cause and interventions were collected. The study was approved by the Ethics Committee of the Hospital District of Southwest Finland as well as by the scientific research committee of the Pediatric Department of Turku University Hospital, Finland. Written informed consent was obtained from both parents according to the guidelines of the Research and Ethics Board before patient enrollment in this study.
Correlations between respiratory parameters and PMA were analyzed using the Spearman test. Changes in the proportion of backup ventilation before clinical deterioration were analyzed using the Friedman test. Ap value < 0.05 was considered to be statistically significant. Statistical analyses were performed using SPSS v27.0 (IBM, Armonk, NY, USA).