Noninvasive Ventilation in Pediatric Intensive Care

Since the 2000s, the emergence of noninvasive ventilation (NIV) is certainly one of the major changes in the pediatric critical care field. Surprisingly, this evolution has initially been supported by limited evidence. Several physiological studies confirmed the potential of NIV in reducing the work of breathing and many authors reported their experience in observational studies (1). However, the impact of NIV on clinically important outcome in critically ill children has been studied in a single randomized controlled trial (RCT) including 50 patients (2), whereas three other RCT were restricted to short-term evaluation or specific conditions (3–5). In this issue of Pediatric Critical Care Medicine, Wolfler et al (6) provide new important data about NIV use in PICU. This cohort study is based on an Italian registry, which collected the ventilatory support provided to 5,741 patients admitted in 13 Italian PICU in 2011 and 2012. An historical group of 1,370 patients studied over 6 months in 2006 and 2007 is also reported for comparison over time. NIV was initiated in 164 patients prior to ICU admission, used in 585 children as a primary ventilatory support after ICU admission and in 85 cases after extubation, for a total of 834 patients treated with NIV (12% of the cohort). Among eligible patients with no respiratory support at PICU admission, NIV was more frequent than invasive ventilation, and its use increased from 12% in 2006 to 18% in 2012. In the mean time, the failure rate of NIV increased from 10% to 16%. The failure rate observed in similar PICU populations is close to 20% (7, 8). The authors should be commended for evaluating the implementation of this technology. This is the largest reported cohort of children treated with NIV. The high number of participating sites adds to the representativeness of the results and allows demonstration of the intercenter variability. Of note, several limitations of the study should be kept in mind. The observational study design and the absence of a comparable control group (randomly nonexposed to NIV) preclude the assessment of the specific benefits of NIV. The use of a registry limits the number and quality of variables for the evaluation of risk factors of NIV failure. The authors also isolated a population for an “intention-to-treat” analysis, excluding a relatively large proportion of the patients. In particular, 32% patients were excluded from further analysis because of ventilatory support initiated prior to ICU admission. Most of these patients were likely admitted in postoperative period, but this selection should be kept in mind. Another exclusion criterion was the presence of contraindications to NIV, met in 12% of children. The presence of cardiac failure among these contraindications may appear surprising, whereas NIV is commonly used in that context. Finally, the registry was implemented prior to the emergence of high-flow nasal cannula; therefore, the increasing use of this technology does not reflect in the results. Notwithstanding, the study by Wolfler et al (6) confirms that NIV has become a major ventilation option in PICU. In absence of RCT-based evidence, the clinical evidence of potential benefits has led to a wide acceptance of NIV. This clinical impression has been mostly confirmed by large historical studies showing that the dissemination of NIV as a primary ventilatory support in the management of bronchiolitis has been accompanied by a major decrease in intubation rate, length of stay, and health-related costs (9, 10). A randomization process regarding NIV use in cases of bronchiolitis would now raise ethical questions. However, we must emphasize that many NIV-related questions remain and should be prioritized in future research. The report by Wolfler et al (6) permits to highlight several research gaps of particular importance. Although NIV potential is great in bronchiolitis (10), its benefit in other pediatric pathologies has been much less studied. Wolfler et al (6) suggest that NIV is frequently used in many different diseases. This interesting result, combined with the relatively high success rates, opens the field for future evaluative work, especially in pneumonia, postoperative care, asthma, or acute decompensation of chronic diseases. Pediatric acute respiratory distress syndrome and hypoxemia severity have repeatedly been shown to be important risk factors of NIV failure (7, 8), which is again confirmed by Wolfler et al (6). The benefit of NIV in hypoxemic patients is, therefore, questionable. Awaiting further study for this indication, NIV use should be limited to the less severe patients and used with particular caution. This has been underlined by the recent Pediatric Acute Lung Injury Consensus Conference (11). The timing and duration of NIV periods vary among centers and have not been evaluated in the PICU setting. Wolfler et al (6) did not address this question but illustrated the complexity of patient trajectory, with various combination and order of noninvasive and invasive support periods. Moreover, the initiation of a support before the PICU admission in about one third of the patients raises the need to evaluate the NIV implementation during transport. Although such development seems achievable in areas with pediatric medical transport system (10), the simpler use of high flow nasal cannula could be an interesting alternative to evaluate (12). Of course, the potential benefit should be weighted with the risk of transporting a potentially unstable patient. The potential risks of NIV also merit further attention. Besides evident complications (skin abrasions and abdominal distension), which should be prevented, the most significant risk is the possible delay to take full control of the ventilation in a nonresponding patient. Although this delay has been shown to be deleterious in adults (13), pediatric data are scarce. The characterization of this risk should be a priority in additional studies, as clinicians always have to ponder the potential benefit of NIV with its risk. The timely reassessment of the clinical response after 1–2 hours of NIV seems of major importance in this context. Finally, the technical aspects of NIV should also be further evaluated. Wolfler et al (6) confirm that continuous positive airway pressure is the most frequently used mode during NIV, especially in infants. This preference is likely related to the simplicity of this mode and the absence of demonstrated benefit of bilevel support modes, in particular because of poor synchronization with the patient efforts. The recent neurally adjusted ventilatory assist mode now permits to achieve an efficient synchronization (14, 15). It will be important to evaluate the impact of this progress on NIV outcome. In conclusion, we agree with Wolfler et al (6) that NIV is beyond a promising therapy in the PICU and is rather an important ventilatory support option. This should not mask the fact that many questions related to the NIV use remain. In light of the pediatric critical care–specific barriers (limited enrolment capacity of each PICU, patient condition heterogeneity, and practice variability), international collaborative networks should be encouraged to address these questions, to permit the clinicians to provide an optimized, individualized, and safe ventilatory support to each critically ill child.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.