Short Term Effects of Synchronized vs. Non-synchronized NIPPV in Preterm Infants.

Respiratory problems are one of the major issues to deal with in preterm newborns.

Because of the immaturity of respiratory mechanisms and structures, the use of supporting devices is often necessary. These include both conventional mechanical ventilation (MV) techniques, which require the use of an endotracheal tube, as well as non-invasive ventilation (NIV) techniques that use softer ventilator-patient interfaces. Increasing attention is payed to the latter ones as less aggressive and associated with better outcomes both in terms of mortality and short and long-term complications, such as bronchopulmonary dysplasia (BPD).

Nasal intermittent positive pressure ventilation (NIPPV) is a NIV technique in which newborn airways are kept open between two pressure levels: peak inspiratory pressure (PIP) and positive end expiratory pressure (PEEP). The frequency and duration of each phase are defined by setting the inspiratory and expiratory times or the ventilation rate.

This technique has already shown its superiority in terms of reduced duration of MV, reduced necessity of intubation, decreased failure of extubating and reduced prevalence of BPD if compared with non-invasive techniques based on continuous pressure support, such as continuous positive airway pressure (CPAP). Recent meta-analyses of studies where NIPPV has been used as an alternative to CPAP following extubation show that it reduces need for re-ventilation and air leaks but without any reduction in BPD: there is insufficient evidence to recommend NIPPV as primary mode of respiratory support in the delivery room.

It should be specified that the ventilation rate on NIPPV does not reflect the real respiratory rate (RR) of the newborn, as the ventilator supplies the PIP regardless of newborn respiratory efforts. To reproduce a more physiological and gentle ventilation new devices able to detect newborn respiratory efforts and consequently supply a PIP have been developed to synchronize the ventilation rate with RR of the newborn.

The devices used for synchronization can identify newborn respiratory effort by detecting variation in flow or pressure. While in MV the exact beginning of inspiration can be detected through a continuous monitoring of pressure or through the precise interception of inspiratory and expiratory flow some difficulties occur in NIV where, as a consequence of the impossibility to detect expiratory flow, the moment of the exact beginning of spontaneous inspiration is hard to identify.

Recently, a new type of NIV ventilator equipped with a pressure sensor has been put on the market. The software of this ventilator is able to calculate the flow according to the pressure variations of the circuit and to capture the flow variations induced by spontaneous breathing allowing a synchronization of the flow with the patient's respiratory acts.

The use of a synchronized NIV technique would allow a more physiological respiratory support, reducing respiratory fatigue and improving newborn compliance. Despite these premises, the diffusion of synchronized NIPPV in neonatal intensive care units (NICUs) and works on its efficacy are limited.

Some authors have already demonstrated the benefits of using a synchronized NIV technique in terms of extubating success rate, BPD prevalence and mortality and neurocognitive development. Synchronized NIPPV (SNIPPV) seems more effective than NIPPV and NCPAP in reducing need for intubation in respiratory distress syndrome (RDS), in improving the success of extubation and in treating apnea of prematurity, with a reassuring absence of relevant side effects. Synchronised NIPPV delivered through a ventilator can reduce extubation failure but may not confer long-term advantages such as reduction in BPD. Other reported advantageous aspects of SNIPPV include improved thoraco-abdominal synchrony, reduced work of breathing (WOB) and reduced need of intubation.

It has already been shown that SNIPPV is more effective than NIPPV and CPAP in reducing the number of desaturations and apnoea in preterm infants undergoing CPAP treatment for prematurity apnoea. However, the effectiveness of SNIPPV compared to NIPPV in preterm infants with respiratory distress is still not completely clear.

Our study protocol was designed to evaluate the short-term effects of SNIPPV vs NIPPV on the major cardio-respiratory variables, trying to identify the best ventilation modality for preterm newborns at their first approach to NIV ventilation support, on the bases of cardio-respiratory events reduction and fraction of inspired oxygen (FiO2) request.

NUMBER OF PATIENTS The number of patients to be enrolled is calculated based on a predicted difference of 30% in cardio-respiratory events between the two ventilation modalities. Assuming a mean of 5 and a SD of 1.5 events/hour (based on available literature data), the number of patients to be enrolled is 30, to obtain an 80% power and a significance threshold of 0.05.

STUDY DESIGN The decision to use a NIV support is based on clinical evaluation. At starting of NIV, eligible patients will be allocated to one of the two arms (NIPPV or SNIPPV) by block randomization. A custom software will be used to obtain a casual sequence to randomize patients in both arms, creating a balance between patients needing NIV as first intention or after extubating.

After 2 hours of stabilization (stabilization phase) in NIV, enrolled patients will be alternatively ventilated with 2 different techniques for 2 time frame of 4 hours each.

In case of needing surfactant, the stabilization phase will be 4 hours after its administration.

Infants will be kept supine throughout the study. During the whole study duration (including stabilization phase) all patients will be continuously monitored with a multiparametric monitor, recording also data from the ventilator. The first hour of each NIPPV/SNIPPV phase will be considered as wash-out phase and named "adaptation phase": data recorded during this phase are excluded from the analysis. Milk meals will be administered during the adaptation phase.

Pain and compliance scales will be filled in by nurses every 60 minutes. EGA values will be recorded at the end of the stabilization phase, at the end of Phase A (first NIV modality) and Phase B (second NIV modality).

Patients will considered drop out of the study in case of:

1. NIV failure criteria: FiO2 > 40%, pH \< 7.2, pCO2 > 65mmHg, ≥ 3 episodes of desaturations (transcutaneous O2 saturation (SatO2 TC) \< 80%) per hour, ≥ 3 episodes of apnea (> 20 s) and/or bradycardia (heart rate (HR) \< 80 beats per minute (bpm)) per hour, Silverman score >6. Necrotizing enterocolitis, bowel perforation, and hemodynamic instability as indications of NIV failure
2. Air leak syndrome (i.e. pneumothorax)
3. Needing of invasive procedures during the study
4. Needing of surfactant during the study
5. Development of hemodynamic instability or surgical problems during the study
6. Death Data obtained from dropouts (patients who drop out of the study) will be analyzed separately.

After 8 hours of study, each patient will be ventilated with the best NIV modality according to clinical data and cardio-respiratory parameters observed during the study.

MONITORING Nurse staff will continuously monitor patients to avoid biases due to device wrong positioning

DATA

For each patient enrolled the following variables will be collected:

• ANAMNESTIC VARIABLES:
• Gestational age at birth
• Birth weight
• Delivery type
• APGAR at 1/5 minutes (and 10 minutes if available)
• Presence of intrauterine growth restriction
• Maternal administration of magnesium sulphate
• Steroid prenatal prophylaxis (number of doses)
• Intrapartum antibiotic prophylaxis (if indicated)
• Presence of intraamniotic infection and administration of intrapartum antibiotic therapy
• CLINICAL VARIABLES
• Surfactant administration (time and number of doses)
• Type and duration of MV previously administered (if any)
• Type and duration of NIV previously administered (if any)
• Corrected GA at enrolling
• Caffeine doses administered (if any)
• Neonatal Pain Scale score
• CARDIO-RESPIRATORY VARIABLES
• FiO2 to maintain SatO2 TC 90-94% (as weighted mean)
• NIV failure and endotracheal intubation
• Number of cardiorespiratory events, defined as episodes of apnea lasting more than 20 seconds or over 5 seconds if followed by desaturation or bradycardia and/or episodes of desaturation with blood oxygen saturation below 80% for 4 sec. or more and/or episodes of bradycardia with heart rate below 80 bpm
• OTHER POLYGRAPH VARIABLES (continuous monitoring)
• Heart rate
• Respiratory rate
• SatO2 TC
• Thoracic impedance
• From ventilator: trigger, pressure, flow
• Pressure in ventilator circuit
• LABORATORISTIC VARIABLES
• EGA values at the end of stabilization phase, Phase A and Phase B

RESULTS INTERPRETATION The main result will be the difference in cardio-respiratory events during SNIPPV versus NIPPV.

Tolerance to each of the two NIV modalities will be evaluated by evaluating the number of failure episodes and of cardio-respiratory events and analysing the scores for individual compliance and pain. The individual need for oxygen under the two modalities of NIV will be evaluated as a known risk factor for premature retinopathy and various other complications.

STATISTICAL ANLYSES Descriptive variables will be analyzed in function of their distribution. T student test or Mann Whitney U test in case of continuous variables (if normally or not normally distributed respectively) and chi-squared or fisher test for qualitative ones. All test will be two-sided with a significance threshold of 0.05.

EXPECTED RESULTS Identifying the best NIV modality for preterm newborns at their first approach to NIV ventilation support, on the bases of cardio-respiratory events reduction and fraction of inspired oxygen (FiO2) request.