Acute Hypoxemic Respiratory Failure (AHRF) is a leading cause of admission in Pediatric Intensive Care Unit (PICU). Traditional treatment includes endotracheal intubation and mechanical ventilation, that are invasive and not free from risks. Recent experiences from pediatric studies showed that Non-Invasive Pressure Support Ventilation (NIV-PS) has been associated with lower intubation rate, adverse events and mortality compared to mechanical ventilation delivered by an endotracheal tube. Nonetheless, in pediatric ARF, the application of a well-synchronized NIV-PS is technically challenging due to the presence of leaks and the age-specific characteristics of pediatric respiratory pattern (high respiratory rate, short inspiratory/expiratory time and weak inspiratory effort). Consequently, NIV-PS often results in difficult patient-ventilator interaction, with a failure rate up to 43% . Neurally Adjusted Ventilatory Assist (NAVA) is a new form of ventilatory assistance wherein the ventilator applies positive pressure throughout inspiration synchronously and proportionally to the Electrical Diaphragm activity (Edi). Thus, NAVA is not influenced by large leaks around uncuffed endotracheal tubes or noninvasive interfaces. Studies in intubated children found that NAVA improved interaction by reducing asynchronies and optimizing ventilator cycling.Two recent studies showed that the application of Non-Invasive NAVA (NIV-NAVA) in children with ARF is feasible and may reduce asynchronies as compared to NIV-PS.

More recently, in a recent RCT of our group, we demonstrated that NIV NAVA in children was associated with lower asynchronies, longer synchronization time between patient and ventilator at lower peak and mean airway pressure.

However no data are published to address the question if the better synchronization between patients and ventilator obtained with NIV NAVA could lead to a reduction in intubation rate and PICU outcomes.

To address this question, we designed an observational retrospective study aiming to define if early delivered NIV NAVA could reduce intubation rate if compared with traditionally flow-cycled NIV PS in hypoxemic children admitted to PICU.

End point

Primary end point is the intubation rate between groups. Secondary end points include: days on mechanical ventilation, number o invasive devices, nosocomial infections, PICU and Hospital length of stay and survival at 2-6 months.

Study design

Single center retrospective study

Sample size calculation and statistical analysis

According to previous studies, the reported intubation rate in children admitted to PICU with AHRF and treated with conventional flow-cycled NIV-PS is 40%-60%.

Considering an alfa -error=0.05 and power=80%, the study would have needed 20 patients in each group to detect a 50% reduction in the primary end point, i.e. the intubation rate. Data distribution are tested with Kolmogorov-Smirnov analysis and analyzed with parametric or non-parametric statistics, according to. Normally distributed variables are expressed as mean (SD) while median and interquartile range are used to report non-normally dsitributed variables. Categorical variables are compared with chi-square or Fisher exact test, as appropriate. A p value <0.05 was considered as significant. Sample size calculation was performed with GPower 3.1.2 software (Kiel University, Germany). Other statistics were performed with SPSS 15.0 (SPSS inc, Chicago, IL, USA) and p<0.05 was considered as statistically significant

Study Population

All consecutive children with infectious AHRF meeting the pediatric criteria for moderate ARDS definition admitted to PICU between the first January 2015 and first January 2017. Eligible patients received Noninvasive Respiratory Support as a first line respiratory treatment and were divided in two groups: 1) patients receiving flow-cycled NIV-PS as a first line treatment (control group) and b) patients receiving NIV NAVA as a first line treatment (treatment group).

Experimental protocol.

The routine management of children wth AHRF treated with noninvasive respiratory support did not vary in the study period and included: 1) standard medical therapy including oxygen therapy, antibiotics, steroids, inhaled beta 2 agonists if needed 2) maintaining semirecumbent position 2) sedation provided according to PICU protocol (dexmedetomidine 0.5-0.7mcg/Kg/hour) to maintain a COMFORT score between 17 and 26; To achieve the "best ventilatory assistance" during NIV-PS, the PS level (i.e. the inspiratory pressure support above PEEP) and NAVA gain were set to obtain an expiratory tidal volume 6-9 ml/Kg (while the child was on active inspiration, as shown by a positive inflection on the pressure trace), a reduction in RR and in chest retractions, as reported previous studies.

In both trials Positive End-Expiratory Pressure (PEEP) was titrated from 4 to 8 cmH2O (in step of 2 cmH2O) to obtain SaO2> 94% with FiO2 <0.6. During NIV-PS the inspiratory flow-trigger was set at maximum sensitivity level not causing autotriggering. The expiratory cycling-off was adjusted by the attending physicians to obtain the best synchronization, according to flow/pressure tracings. During NIV-NAVA, Edi trigger was set at 0.5 μV above the resting Edi to assist respiratory effort without responding to electrical noise.

NIV NAVA was delivered with ServoI ventilator equipped with Edi module (Maquet, Solna, Sweden).

NIV PS flow cycled was delivered with different ventilators equipped with a specific module for lueaks compesaton and pediatric /neonatal inspiratory trigger module (ServoI Maquet Solna Sweden; Evita VN 500 Lubeck Germany). Pediatric Interfaces includes full face masks Performax, in different size (neonatal and pediatric) (Respironics, Murrysville).

Inclusion Criteria

Age > 1month and < 5 years; paO2: FIO2 <300 with oxygen mask (FiO2=0.4) for 15 minutes Respiratory Rate (RR)>2SD according to age Bilateral Chest X ray infiltrates Absence of cardiopathy diagnosed by ecochardiography .

Exclusion Criteria:

Emergency need for intubation Glasgow Coma Scale <12 pH<7.25 Cough reflex impairment Upper-airway obstruction Hemodynamic instability (need for vasopressor or inotropes) Evidence of pneumothorax on lung echo or chest x ray Esophageal surgery, neuromuscular, mitochondrial, metabolic or chromosomal diseases with baby hypotonia and lesions of medulla.

Were excluded even patients whose parents denied the consent to manage personal data at admission and patient lacking data on medical charts report

Consecutive children with infectious Acute Respiratory Failure meeting the pediatric criteria for moderate ARDS Definition admitted to PICU of Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy between the first January 2015 and first January 2017.