LAUSR

Endotracheal mechanical ventilation in preterm infants is a major risk factor for bronchopulmonary dysplasia (BPD) via ventilator-induced lung injury. Randomized clinical trials (RCT) have shown that the use of nasal continuous positive airway pressure (nCPAP) at birth decreases the need for endotracheal mechanical ventilation, as well as death or BPD. Most preterm infants are now managed by nCPAP at birth and are never intubated. Surfactant administration, if needed, is performed via transient tracheal intubation or through a small endotracheal catheter without intubation (LISA or MIST procedure). Unfortunately, studies have shown a high rate of nCPAP failure in the most premature newborns (22-26weeks’ gestational age), who have a distressing incidence of BPD. Consequently, in recent years, various non-invasive respiratory support modalities have been developedwith the hope of preventing endotracheal mechanical ventilation and BPD. As a result, heated humidified high-flow nasal cannula and various forms of synchronized intermittent positive pressure ventilation, among others, have drawn considerable attention. Non-invasive high-frequency oscillatory ventilation (NHFOV) is a new and exciting strategy that could redefine respiratory support of the preterm infant. Theoretical reasons for using NHFOV, as well as a number of case reports and observational studies reporting positive effects in human neonates and animal models, have been the focus of recent reviews. NHFOV is viewed as a gentler method of delivering mechanical ventilation. It has the advantages of both high-frequency ventilation, that is, absence of ventilator-patient asynchrony and high efficacy in removing CO2, and non-invasive CPAP, including among others, an improved oxygenation via an increased functional residual capacity. Recent surveys in Europe and Canada have shown that a significant number ofNICUs have already included this strategy in their therapeutic armamentarium, at times on a regular basis. The lack of RCT is however a big concern. The first ever (pilot) RCT on NHFOV was performed in full-term newborns with transient tachypnea. Results showed a reduced duration of respiratory distress and oxygen supplementation. Of probable greater clinical relevance is a recent parallel-arm, RCT comparing NHFOV versus biphasic nCPAP (control group) as a rescue therapy following failure of nCPAP in preterm infants born at a mean gestational age of 26 weeks. Mean postmenstrual age at the time of the study was around 29 weeks. The primary aim of this pilot RCT conducted in a low number of infants (17 in the NHFOV group and 23 in the biphasic nCPAP group) was to show the feasibility of a future, adequately powered RCT. While feasibility was shown, no positive effect on the need for endotracheal mechanical ventilation could nonetheless be found. In this issue of Pediatric Pulmonology, Zhu et al report the results of an unblinded RCT on NHFOV as the initial respiratory support for preterm infants. The authors compared the efficacy of nasal HFOV versus nasal CPAP (control group) in preventing the need for endotracheal intubation in preterm infants with moderate to severe respiratory distress syndrome at birth after surfactant administration via the INSURE (INtubation, SURfactant, Extubation) method. Eightyone preterm infants, with a gestational age of 28-34 weeks (mean gestational age∼ 32 weeks), were randomized into NHFOV and nCPAP groups. NHFOV was delivered by binasal prongs with a mean airway pressure of 6-10 cmH2O and an oscillation frequency of 6-12Hz, while nCPAP was “started at 6 cmH2O” in control infants. Results showed that the need for endotracheal mechanical ventilation was more than halved (24.3% vs 56.4%, P = 0.004) in the NHFOV group without increased adverse effects compared to nCPAP. The incidence of BPD (as defined by the NIH consensus in 2000), however, was not significantly decreased in the NHFOV group. The absence of a significant effect on BPD incidence may seem at odds with the positive effects of NHFOV previously reported in a preterm lambmodel of BPD. In these latter studies, NHFOV resulted in better gas exchange, reduced lung injury and improved alveolar formation and weight gain (another key factor for BPD prevention), compared to endotracheal pressure-limited, volume-targeted mechanical ventilation. Beyond potential species differences, other explanations could account for the absence of positive effect on BPD incidence in the study by Zhu et al. First, on top of a relatively small number of preterm newborns studied, the BPD incidence was low in the control group, as expected for preterm infants born at 32 weeks’ gestational age. This low incidence in control newborns greatly hinders the possibility of showing a positive effect of any intervention. In addition, while newborns in the control group were treated by noninvasive bubble CPAP, previous comparison in preterm lambs were performed between NHFOV and endotracheal ventilation, which in itself is a major risk factor for BPD. Moreover, the presence of high frequency oscillations (between 15 and 30 Hz), which is inherent to bubble CPAP, may have reduced the difference between the nHFOV and control groups in the study by Zhu et al. Further limitations of the study are significant. First and foremost, the study was performed in moderately preterm infants (mean gestational age of 32 weeks), leaving unanswered the question of a