High-flow nasal cannula

The use of HFNC in pediatric acute hypoxemic respiratory failure has increased over the last decade, with nearly one-quarter of all children admitted to the PICU receiving this form of respiratory support. The popularity of HFNC is likely related to its ease of use, portability, tolerability, and its success in treating perinatal lung disease and viral bronchiolitis. ^, Recent randomized controlled trials including children with bronchiolitis suggest that HFNC may be superior to the standard low-flow nasal cannula ^, and equivalent to CPAP in prevention of treatment failure requiring escalation of care. Clinicians may also choose to use HFNC in children at risk for development of PARDS, although data are limited in this subset of patients.

Although incompletely understood, the benefits of HFNC are undoubtedly multifactorial. In a tachypneic patient with diminished tidal volumes, the bulk movement of oxygen-rich gas past the nasopharyngeal dead space results in improved alveolar ventilation. The increased flow rates provided by HFNC may also provide enough low-level positive pressure to overcome subtle upper airway obstruction. Additionally, provision of conditioned inspiratory gas at high flow rates likely reduces inspiratory resistance through the nasal passages, improves mucociliary clearance, and reduces metabolic work associated with heating and humidifying the inspiratory gas. Together, these mechanisms often improve respiratory mechanics and gas exchange in children with acute respiratory failure ( Fig. 55.1 ).

Commonly used devices for delivery of heated humidified gas mixtures via high-flow nasal cannula (HFNC). (A) HFNC system assembled using a blender heater/humidifier and wire heated circuit. (B) Airvo-2 HFNC system (Fisher & Paykel Healthcare); (C) Precision Flow high-velocity nasal insufflation HFNC system (Vapotherm, Inc.).

The HFNC system includes the following basic elements: (1) a source of pressurized and blended oxygen and air, (2) a water reservoir attached to a heated humidifier, (3) a heated circuit that maintains gas temperature and humidity, and (4) a nonocclusive cannula interface. With initiation of HFNC, the clinician sets the gas temperature, fraction of inspired oxygen (Fio ₂ ), and flow rate. We recommend an initial gas temperature 1°C to 2°C below body temperature for comfort. The initial HFNC Fio ₂ should be chosen based on patient need and physiology and adjusted to target a desired peripheral capillary oxygen saturation (Spo ₂ ). While there is no consensus regarding the ideal gas flow rate, there is evidence to support weight-based dosing, at least in infants. Modest respiratory support is provided with flow rates between 0.5 and 1.0 L/kg per minute, while increasing the flow to 1.5 to 2.0 L/kg per minute further attenuates intrathoracic pressure swings associated with work of breathing. Flows greater than 2 L/kg per minute may not provide additional clinical benefit.

Noninvasive positive-pressure ventilation

Modes of NIPPV, such as CPAP and BiPAP, are beneficial in patients with upper airway obstruction, neuromuscular disease, diseases of increased airway resistance (i.e., asthma or viral bronchiolitis), and in restrictive diseases, such as pneumonia and PARDS. ^, In acute bronchospasm, expiratory resistance is greater than inspiratory resistance. The increased lower airway resistance prolongs the expiratory time constant, leading to carbon dioxide retention and tachypnea. A vicious cycle ensues whereby the increased respiratory rate precludes complete exhalation, causing air trapping, increased functional residual capacity (FRC), and intrinsic positive end-expiratory pressure (PEEP). Intrinsic PEEP is an additional load that the fatigued patient with respiratory embarrassment must work against. This may be particularly problematic in small children, whose smaller airway caliber and more compliant chest wall make them vulnerable to an obstructive process, such as bronchiolitis and asthma. For these infants and children, NIPPV can unload fatigued respiratory muscles and prevent collapse of peripheral small airways to allow for a more complete exhalation and reduced work of breathing.

In patients with restrictive lung disease, compliance is decreased and chest wall expansion is limited. Acute processes (including infection, effusion, alveolar or interstitial edema) and chronic processes (such as neuromuscular dysfunction or thoracic cage abnormalities) can all contribute to restrictive lung disease. The resultant decreased FRC and decreased tidal volume lead to a compensatory increase in respiratory rate necessary to maintain minute ventilation. The decreased FRC can also lead to further alveolar collapse and progressively worsening lung compliance from atelectasis. The application of positive pressure in these patients decreases the inspiratory work of breathing, allowing generation of higher tidal volumes, subsequently increasing FRC. CPAP raises expiratory pressures above atmospheric pressure, making it appropriate for use in hypoxic respiratory failure in the acute setting. CPAP does not generate inspiratory flow, nor does it directly increase tidal volume. It improves FRC by overcoming inspiratory work produced by elastic forces. ^, Improved FRC can, in turn, decrease atelectasis and improve ventilation/perfusion (V/Q) matching. Furthermore, the reduction in inspiratory work provided by CPAP allows the respiratory muscles to generate higher tidal volumes, indirectly improving ventilation. In the CPAP mode, devices provide a constant flow with resultant constant positive pressure throughout the respiratory cycle while the patient breathes spontaneously. CPAP is usually well tolerated in children, ^{, ,} but some patients may need sedation at initiation or throughout the implementation to improve tolerance of the device-patient interface.

With BiPAP, the clinician sets an expiratory positive airway pressure (EPAP), an inspiratory positive airway pressure (IPAP), inspiratory time, back-up respiratory rate, and Fio ₂ . The IPAP assists with inspiration, while the EPAP maintains airway patency throughout expiration. The difference between these two pressures assists in the generation of tidal volume; thus, BiPAP is an appropriate mode of ventilation for both hypoxic and hypercarbic respiratory failure.

A well-fitted patient interface is essential for effective CPAP and BiPAP use ( Fig. 55.2 ). Air leaks around an ill-fitting mask will prevent generation of adequate mean airway pressure. However, masks that are too tight can lead to skin breakdown and pressure ulcers that will preclude prolonged use. There are multiple patient interfaces available, including nasal pillows, nasal masks, oronasal masks, total face masks, helmets, and mouthpieces. ^, The nasal, oronasal, and total face masks are most commonly used in the PICU. With an appropriately fitting mask and patient-device synchrony, BiPAP can effectively provide mean airway pressure, improve oxygenation, and unload fatigued respiratory muscles in children with acute respiratory failure. BiPAP is generally well tolerated in patients, and its low-risk profile makes it an attractive first-line support therapy. However, care should be taken in patients with persistent hypoxemia and elevated respiratory rates, as both have been associated with failure of BiPAP. ^, Goals of NIPPV are given in Box 55.1 .

Various interfaces for delivery of noninvasive ventilation support. (A) Nasal pillows (Medical Innovations PedFlow and ResMed Swift FX). (B) Nasal (Sleepnet MiniMe 2) and oronasal (Respironics AF531) masks. (C) Full face mask (Respironics FitLife) and helmet (arol NIV10301/X).

(Images courtesy of the manufacturers.)

Negative-pressure ventilation

The original negative-pressure iron lung ventilator used a large chamber enclosing the entire body save for the head ( Fig. 55.3 ). Advances in intubation and mechanical ventilation combined with the cumbersome nature of the iron lung led to the widespread use of invasive PPV as the main support modality in patients with respiratory failure. While invasive PPV can improve outcomes in these patients, it is not free of morbidity and potential complications, particularly those related to airway injury, ventilator-associated lung injury, use of sedation and muscle relaxation, and increased hospital stay. This has resulted in a renewed interest in NPV devices (e.g., RTX Biphasic Cuirass Ventilator, Hayek Medical) with new technology that overcomes some of the inadequacies of the traditional iron lung ( Fig. 55.4 ).

Tank respirator (“iron lung”). During inspiration (A), a negative pressure is created in the chamber using a bellows system (shown in figure) or a piston. Expiration (B) can be either passive or assisted with positive pressure (so-called biphasic ventilation ).

Hayek RTX cuirass ventilator. The chest cuirass is made of flexible plastic. This ventilator is capable of conventional negative pressure ventilation as well as high-frequency chest wall oscillations.

The cuirass interface is a plastic shell that covers the anterior chest wall to deliver either continuous negative pressure or biphasic ventilation ( Fig. 55.5 ). In continuous negative extrathoracic pressure (CNEP) mode, negative pressure is maintained at a constant level throughout the respiratory cycle while the patient breathes spontaneously. Conversely, during biphasic cuirass ventilation (BCV), inspiratory and expiratory phases are fully controlled (control mode) by modifying the negativity of the air pressure applied to the cuirass.

Cuirass ventilator. During inspiration, negative pressure is created in the cuirass shell. Expiration can be either passive or assisted with positive pressure (so-called biphasic ventilation ). This can be applied in both the supine (A) and nonsupine (B) positions.

During inspiration, NPV creates subatmospheric (negative) pressures within the cuirass applied to the anterior chest wall, creating negative intra-alveolar pressures that allow air to flow into the lungs. During expiration, the NPV device delivers less negative pressure (set on the ventilator as a positive number), creating positive intra-alveolar pressure that facilitates air movement out of the lungs. This is physiologically very similar to spontaneous breathing, with the main exception that expiration in the NPV device is active, unlike restful spontaneous breathing in which expiration is passive and dependent on the elastic recoil of the chest wall.

The cuirass is available in seven pediatric and four adult sizes. The cuirass and connecting tubing to the ventilator are reusable, but the disposable foam liners are for individual use only. The foam liner is essential to create a seal between the cuirass and anterior chest wall that allows for generation of negative pressure.

CNEP is usually chosen as the initial support mode in patients undergoing NPV. It is the NPV equivalent of CPAP. The minimum CNEP support is −8 cm H ₂ O; this level is then adjusted until the work of breathing improves (in increments of 2 cm H ₂ O). A pressure of −14 cm H ₂ O generally is sufficient, but support can be escalated to higher negative pressures (e.g., −20 cm H ₂ O) as needed throughout the treatment course.

Escalation of support to control mode is warranted in cases of persistent respiratory distress or apnea unresponsive to CNEP. Control mode is the NPV equivalent of synchronized intermittent mandatory ventilation (SIMV) pressure control, in which the patient is able to take spontaneous breaths in between ventilator mandatory breaths. Inspiratory and expiratory pressures are set on the ventilator as negative and positive numbers, respectively, while maintaining inspiratory to expiratory pressures as a 3:1 ratio (e.g., −15, +5). The pressure differential, which is the mathematical number between inspiratory (I) and expiratory (E) pressures, determines the chest wall excursion and, ultimately, the tidal volume. The I:E ratio is usually set at 1:1 but can be changed as needed. The ventilator mandatory rate is set slightly above the patient respiratory rate but preferably not to exceed 60 breaths/min (e.g., set ventilator rate at 42 breaths/min if patient respiratory rate is 40 breaths/min).

NPV may improve cardiac output by lowering intrathoracic pressure and increasing venous return to the heart (right ventricular preload). It also improves alveolar recruitment and FRC optimization, which decreases the pulmonary vascular resistance and right ventricle afterload, thereby increasing pulmonary blood flow. These features make NPV a potentially advantageous modality in patients with right-sided heart failure, passive pulmonary circulation (Fontan circuit), or restrictive physiology (post–tetralogy of Fallot repair) and have been associated with increased pulmonary blood flow. ^{, ,}

To date, there is a paucity of data describing the use of NPV in pediatric acute respiratory failure. In 1996, Samuels et al. performed a prospective randomized controlled trial in 244 neonates comparing CNEP of −4 to −6 cm H ₂ O with standard therapy, which included CPAP of 4 cm H ₂ O. Despite not reaching statistical significance, the study showed that the need for intubation was lower in the NPV group compared with the group that received standard therapy (86% vs. 91%). The NPV group also received oxygen for fewer days compared with the control group (18.3 days vs. 33.6 days, respectively). Al-Bakhi retrospectively compared the use of NPV to invasive PPV in infants with recurrent apnea secondary to acute bronchiolitis. NPV was associated with a reduced rate of intubation, shorter PICU stay, and reduced sedation use. In an animal study of surfactant-depleted rabbits comparing NPV and PPV, Grasso and colleagues showed that NPV was associated with improved gas exchange, greater lung perfusion, better lung expansion, and decreased lung injury. A Cochrane review by Shah and colleagues suggested a decrease in need for intubation and shorter hospital stay with the use of NPV in children with acute respiratory failure.

In 2017, a large retrospective single-center study reported the use of NPV in pediatric patients with acute respiratory failure. Out of 233 patients supported with NPV, 163 (70%) had resolution of acute respiratory failure while receiving NPV, while 63 patients required change to PPV modalities, including intubation. NPV cuirass was removed from five patients owing to complications (gastroesophageal reflux, hypothermia, and skin bruising without sequelae) and from two other patients for transport. Viral bronchiolitis was the most common diagnosis (70% of the cases), and there was a 28% reduction in intubation rate during the study period compared with the prior 3 years.

NPV use via cuirass may be suitable for pediatric patients with facial deformities, facial burns, claustrophobia, severe agitation, and with oronasal secretion burden ( Box 55.2 ). NPV is particularly useful in patients who failed extubation to avoid reintubation when there is difficulty weaning off the ventilator owing to chest wall muscle weakness. Other clinical conditions in which NPV has been successfully used include acute exacerbation of cystic fibrosis, air-leak syndrome, pneumatocele, and neuromuscular weakness. ^{, ,}

NPV has been used in patients with congenital hypoventilation syndrome (CHS). In 1994, Hartmann and colleagues described the safe and effective use of NPV in 9 patients with CHS with varying degrees of respiratory insufficiency. This modality of respiratory support was accepted by the parents and provided a better quality of life for patients in this case series. ^,

Anecdotally, NPV has been applied to patients concomitantly receiving PPV, both while intubated or on NIV. In intubated patients, the application of NPV may allow for faster and more successful weaning from PPV toward extubation. This is a promising area of clinical application that warrants exploration. Takeda et al. have shown that application of external biphasic cuirass ventilation in patients already receiving PPV facilitates secretion clearance and supports ventilation during temporary disconnections from PPV for endotracheal tube suctioning. Nocturnal use of NPV might have a role in long-term management of patients with neuromuscular weakness, but further studies are warranted.

Negative-pressure ventilation use in chest physiotherapy and secretion clearance

Clearance of airway secretions can be enhanced during cuirass ventilation by using the secretion clearance mode, which allows for mobilization and expectoration of secretions while maintaining lung inflation. There are two modes of NPV for airway clearance.

Neurally adjusted ventilatory assist

Neurally adjusted ventilatory assist (NAVA) is a pressure-assisted mode that uses the electrical activity of the diaphragm (EAdi) to trigger a spontaneous assisted breath and deliver inspiratory pressure in response to that activity. NAVA provides ventilatory support proportional to the EAdi. It also enables physiologic variations in tidal volume and inspiratory time from breath to breath. NAVA detects EAdi through eight electromyogram sensors located at the distal end of a special nasogastric or orogastric tube. These sensors are usually positioned near the end of the esophagus close to the gastroesophageal junction where the trunk of the phrenic nerve meets the diaphragmatic muscle. NAVA has been successfully used in acute respiratory failure in both children and adults undergoing mechanical ventilation. NAVA has been shown to improve ventilator synchrony in neonates and children. Lower peak inspiratory pressures and Fio ₂ requirements have been reported during NAVA in comparison with standard conventional ventilation, along with a reduced need for sedation and shorter PICU length of stay. ^,

In a prospective randomized crossover study, Vignaux et al. reported improved patient-ventilator synchrony in infants and children with acute respiratory failure receiving NAVA during NIV. In a more recent prospective study, Baudin et al. described the use of NIV-NAVA in 11 infants under 6 months of age with respiratory failure and showed a lower asynchrony index during NAVA compared with pressure assist control mode (3 ± 3% vs. 38 ± 21%, respectively).

Compared with pressure support ventilation, the use of NAVA improved patient synchrony during NIV in adults and low-birthweight infants with respiratory insufficiency. ^{, ,} In a recent systematic review, Beck et al. reported that the use of invasive and NIV NAVA and EAdi monitoring in children is feasible and safe. NAVA use improves patient-ventilator synchrony and results in lower peak inspiratory pressures when compared with conventional ventilation. This results in improved patient comfort, less sedation, and decreased length of stay. Ducharme-Crevier and colleagues reported a significantly lower total time spent in asynchrony during NIV NAVA (8%) compared with conventional NIV before (27%) and after (32%) NIV NAVA.

Bronchiolitis

While current practice guidelines for the management of bronchiolitis lack definitive recommendations on advanced levels of respiratory support, NIV use has been reasonably well studied in this patient population. Both CPAP and HFNC have been used to improve breathing effort and prevent intubation in infants and children with bronchiolitis, with recent multicenter studies suggesting that some centers preferentially use HFNC while others routinely chose CPAP.

The use of nasal CPAP for children with bronchiolitis was first reported in the early 1980s. CPAP improves ventilation and oxygenation by unloading the diaphragm, increasing FRC, maintaining airway patency, and supplementing oxygen. On the other hand, HFNC improves work of breathing through a variety of mechanisms, including overcoming some of the inspiratory resistance, improving mucociliary clearance, decreasing the energy cost associated with conditioning (heating and humidifying) the inspired gas, and replacing carbon dioxide–rich expiratory gas in the nasopharyngeal anatomic dead space with oxygen-rich gas. These effects are especially beneficial in bronchiolitis, given the increased dead space in infants and small children relative to adults.

Owing to a paucity of high-quality randomized controlled trials, there is currently insufficient evidence to recommend the routine use of either CPAP or HFNC in children with bronchiolitis. ^{, ,} Several studies have compared the two respiratory support modalities in this form of respiratory illness. In one recent multicenter trial with a noninferiority design, 142 infants aged 6 months or less with a primary diagnosis of bronchiolitis were randomized to receive nasal CPAP at 7 cm H ₂ O versus HFNC at flow rates of 2 L/kg per minute. The primary outcome of treatment failure within 24 hours occurred in 31% of the infants in the CPAP group and 51% of infants in the HFNC group, with the authors rejecting their initial hypothesis of HFNC noninferiority. However, this finding was largely related to the chosen definition of failure, since both modalities performed similarly regarding the need to escalate support (intubation), and the majority of subjects who failed CPAP were rescued by HFNC following crossover. A subsequent study done by the same investigators showed an HFNC failure rate similar to the CPAP failure rate in the previous study (39% vs. 31%), suggesting that inexperience with HFNC may have biased the first study. A smaller study (n = 31) comparing HFNC to CPAP found no differences in outcomes and, again, HFNC was better tolerated by patients. Intubation rates were less than 10% for both trials and did not differ between the HFNC and CPAP groups. In a recent multicenter database study including over 6000 children with bronchiolitis, those receiving NIV as initial respiratory support had higher intubation rates compared with those receiving HFNC as initial therapy (20% vs. 11%, P < .001). While HFNC is likely better tolerated than CPAP, limited studies and concerns with trial design likely mean that more data are needed before definitive conclusions regarding the use of CPAP versus HFNC in these patients can be made.

Bronchiolitis was the most common diagnosis (70%) in a large retrospective study describing one center’s use of NPV in pediatric patients with acute respiratory failure. Of the 233 included patients, 170 (70%) had resolution of acute respiratory failure while receiving NPV, while 63 subjects required a change to PPV, including intubation. Complications related to the cuirass—including gastroesophageal reflux, hypothermia, and skin bruising—were rare (<5%), although enteral nutrition was delayed in one-third of patients and over half required intravenous sedation to facilitate patient-ventilator synchrony.

Asthma

Studies describing the use of NIV in children with critical asthma are limited, with most involving HFNC and BiPAP. ^{, ,} The proposed mechanisms of action of HFNC, including application of low levels of PEEP and washout of carbon dioxide–rich gas in the nasopharyngeal dead space, may make HFNC an attractive option in treatment of asthma. However, owing to concerns surrounding adequate drug deposition of aerosolized albuterol, clinicians should give careful consideration to initiating HFNC in a child with critical asthma who is otherwise stable or improving. An observational study comparing HFNC with standard oxygen suggested that children receiving HFNC had greater improvements in heart rate, respiratory rate, Spo ₂ /Fio ₂ ratio, pH, and partial pressure of carbon dioxide (pCO ₂ ). They also had a higher illness severity, more use of adjunctive medications, and a longer hospital length of stay than those children receiving simple oxygen delivery. A single-center randomized feasibility study comparing HFNC with standard oxygen therapy included 62 children in the emergency department with an acute asthma exacerbation. Children treated with HFNC for 2 hours were more likely to have improvement in the pulmonary score, an objective measure of respiratory distress, compared with those children receiving standard oxygen therapy. There were no differences in secondary outcomes—including need for hospitalization, need for PICU admission, or length of stay—between the groups.

There is significantly more evidence supporting the use of BiPAP for critical asthma. A single-center study comparing the two interventions showed a higher treatment failure and longer length of stay in the HFNC group compared with the BiPAP group. In children with an asthma exacerbation, acute inflammation of the lower airways causes increased airway resistance, a prolonged expiratory time constant, and premature airway closure during exhalation. The resultant dynamic hyperinflation causes positive static recoil pressure at the end of expiration, termed auto-PEEP . Because of auto-PEEP, a more forceful breath accompanied by a larger drop in pleural pressure is required to induce inspiration. In asthma, the application of BiPAP with appropriate levels of EPAP decreases the work of breathing associated with hyperinflation and auto-PEEP. One must select a level of EPAP that decreases the work of breathing while avoiding excessive pressures that could exacerbate hyperinflation.

In the absence of definitive studies, there are currently insufficient data to recommend routine use of NIPPV in children with critical asthma. While not considered standard of care, NIPPV may be employed in an attempt to decrease the need for IMV. ^, Despite concerns regarding the safety of NIPPV use in asthma, including barotrauma, multiple observational studies suggest it to be safe and effective in improving the work of breathing and oxygenation. ^{, , ,}

Pediatric acute respiratory distress syndrome

While prospective data are limited, the use of NIV in all forms of pediatric acute respiratory failure, including PARDS, is increasing. ^{, , ,} Based on two multicenter cohort studies, the use of NIV for PARDS has increased from 8.5% in 2005 to 23% in 2016 to 2017. ^, An observational cohort study including 31 PICUs in the United Kingdom demonstrated decreased mortality, shorter duration of ventilation, and decreased length of PICU stay associated with successful use of NIV as first-line respiratory support for pediatric acute hypoxemic respiratory failure. While its association with favorable outcomes makes the use of NIV an attractive alternative to IMV, appropriate patient selection is paramount because failure of NIV has been associated with higher morbidity and mortality in PARDS. Khemani et al. described the incidence of PARDS in a prospective, cross-sectional, observational study including patients from 145 PICUs in 27 countries. Of the children diagnosed with PARDS, 22% were noninvasively ventilated (n = 160) and half required subsequent intubation. Patients who required IMV after NIV had significantly higher mortality and longer duration of ventilatory support than those children who were successfully managed with NIV. The degree of hypoxemia was strongly associated with need for subsequent intubation.

NIV is likely beneficial for children with mild PARDS when used early and judiciously. PARDS is a heterogenous lung disease caused by capillary and alveolar epithelial damage, surfactant dysfunction, and diffuse alveolar damage. These conditions lead to changes in respiratory resistance and compliance, causing hypoxic and hypercarbic respiratory failure. Implementation of NIV can prevent alveolar collapse and recruit atelectatic lung tissue, increase the FRC, and offload the inspiratory work of breathing to improve oxygenation and ventilation in these patients. Another potential advantage includes avoidance of ventilator-associated lung injury and need for sedation. Although the implementation of NIV in PARDS is physiologically sound, the data supporting its use are limited. Yañez et al. randomized 50 children aged 1 month to 15 years with acute respiratory failure to BiPAP or face mask oxygen. Nearly half of the children in this cohort had a diagnosis of bronchiolitis. Use of BiPAP was associated with a decrease in the need for subsequent intubation (28% vs. 60%, P = .045). Despite this promising outcome, children with more severe hypoxic respiratory failure are more likely to require intubation, with some studies describing a failure rate of over 50% in children with PARDS.

Immunocompromised patients

Children with malignancy comprise an estimated 6% of all intensive care admissions. While overall survival has improved, mortality associated with acute respiratory failure remains high. The unfavorable outcomes associated with IMV in immunocompromised patients make consideration of NIV as a first-line respiratory support modality an attractive alternative. ^{, , ,} Several small randomized studies in immunocompromised adult patients suggest that early use of NIV can decrease mortality and intubation rate when compared with oxygen therapy alone. Pediatric data are limited. In a single-center retrospective study including 23 immunocompromised children with acute respiratory failure, 13 (56%) children were successfully managed with NIV. These children had shorter ICU and hospital stays and fewer hospital-acquired infections than children managed with IMV. Another single-center study described clinical variables associated with NIV success and failure in a cohort of immunocompromised children. Among the 41 included subjects, 11 (27%) were successfully supported with NIV. In this small study, lower oxygen requirements, lower Spo ₂ /Fio ₂ ratio, and bacterial septicemia were predictive of NIV success, while fungal septicemia and culture-negative disease were predictive of NIV failure. In a more recent large retrospective cohort study, Pancera et al. reported an NIV success rate of 74.2% in 120 immunocompromised children with acute respiratory failure. The only randomized controlled trial to study NIV for acute respiratory failure in immunocompromised children found no evidence to support the early use of CPAP. Forty-two children were randomized to either CPAP in the PICU or low-flow supplemental oxygen on the general ward. There was no difference in need for intubation and mechanical ventilation between the groups, although the early CPAP group had higher mortality at 30 (33% vs. 5%, P = .041) and 90 (52% vs. 19%, P = .029) days.

Neuromuscular disease

Children presenting to the PICU with underlying neuromuscular disease and acute respiratory failure benefit from NIV. NIPPV is standard of care for children and adults with chronic respiratory failure associated with neuromuscular disease, including Duchenne muscular dystrophy and spinal muscular atrophy. Early application of NIV is also recommended during times of illness and acute worsening of their respiratory failure. ^, For those children who require IMV, NIV has been proposed to facilitate early weaning off IMV, most commonly by implementing it preemptively after extubation following major surgical procedures.