Noninvasive Respiratory Support for Pediatric Critical Asthma: A Multicenter Cohort Study

Abstract

BACKGROUND:

Noninvasive respiratory support (NRS) for pediatric critical asthma includes CPAP; bi-level positive airway pressure (BPAP); and heated, humidified, high-flow nasal cannula (HFNC). We used the Virtual Pediatric System database to estimate NRS by prescribing rates for pediatric critical asthma and characterize patient clinical features and in-patient outcomes by the initial NRS device applied.

METHODS:

We performed a retrospective cohort study from 125 participating pediatric ICUs among children 2–17 years of age hospitalized for critical asthma and prescribed NRS from 2017 through 2021. The primary outcomes were NRS modality prescribing rates and trends. Secondary outcomes were descriptive and included demographics, comorbidities, severity of illness indices, and NRS failure rates (defined as escalation from the initial NRS modality to invasive ventilation, HFNC to BPAP or CPAP, or CPAP to BPAP).

RESULTS:

Of the 10,083 encounters studied, the initial NRS modalities prescribed varied widely by hospital center (HFNC: 69.7 ± 29.6%; BPAP: 27.2 ± 7.1%; CPAP: 3.1 ± 5.9%). The mean rates of HFNC use increased from 59.7% in 2017 to 71.9% in 2021 (+2.5%/y). In contrast, BPAP (–1.6%/y) and CPAP (–0.8%/y) utilization declined throughout the study period. Older children who were obese and with a higher Pediatric Risk of Mortality III–Probability of Mortality score were more frequently prescribed BPAP and CPAP compared with HFNC. Those children on HFNC experienced higher noninvasive respiratory support failure rates versus BPAP (7.3% vs 2.4%; P < .001) but a lower subsequent invasive ventilation rate versus BPAP (0.8% vs 2.4%; P < .001).

CONCLUSIONS:

In this multi-center cohort study, we observed that children with critical asthma are increasingly exposed to HFNC compared with BPAP and CPAP. Rates of HFNC failure were greater than those of BPAP failure, but a majority were transitioned to BPAP without subsequent invasive ventilation. The next steps include prospective trials, including practical end points such as patient comfort and optimal delivery of nebulized treatments to distinguish device superiority and suitable NRS utilization.

Introduction

Each year, nearly 2 million children will experience an asthma exacerbation, which results in ∼800,000 emergency department visits and ∼65,000 in-patient encounters.¹ Cases of patients with severe exacerbation, referred to as critical asthma,² require cardiopulmonary monitoring in the pediatric ICU (PICU) for the application of adjunct treatments,³ including (NRS).^4,5 NRS devices, such as bi-level positive airway pressure (BPAP), CPAP, and, in recent years, heated, humidified, high-flow nasal cannula (HFNC), aim to rest a patient’s fatigued cardiopulmonary system; enhance the delivery of nebulized medications; and ultimately, decrease the frequency of higher-risk interventions, such as intubation and invasive ventilation.^6-9 Literature in support of NRS for critical asthma includes observational study designs^10,11 and single-center pilot trials^12-14 that assessed safety outcomes or uncommon clinical efficacy end points (eg, mortality and invasive ventilation). At this time, updated epidemiologic data are needed to inform prospective trial methodology that seeks to evaluate NRS efficacy and optimize adjunct critical asthma strategies in children.

To address these knowledge gaps, we used a multi-center database, the Virtual Pediatrics Systems (VPS) registry, to estimate NRS device prescribing rates among children hospitalized for critical asthma and characterize clinical features and in-patient outcomes by the initial NRS device applied in the PICU setting. We hypothesized that HFNC prescribing rates are increasing compared with BPAP and CPAP. Further, we postulated that, although NRS failure rates would be higher for those initially prescribed HFNC compared with BPAP, those patients receiving BPAP would experience higher rates of endotracheal intubation.

Quick Look.

Current Knowledge

NRS is prescribed for pediatric critical asthma to optimize pulmonary mechanics, rest a patient who is fatigued, and concurrently deliver inhaled bronchodilators. Devices include bi-level positive airway pressure (BPAP), CPAP, and high-flow nasal cannula (HFNC). Although commonly applied, there are inadequate descriptive and comparative data that characterize the application in critical asthma or consider the potential advantages of one modality over another.

What This Paper Contributes to Our Knowledge

In this multi-center, retrospective study that assessed data from 10,083 encounters from 125 children’s hospitals, we noted an increase in HFNC use by a mean of 2.5% annually, from 59.7% in 2017 to 71.9% in 2021. In contrast, BPAP use decreased by 1.6%/y and CPAP by 0.8%/y. Those receiving HFNC experienced greater device failure rates versus BPAP (7.3% vs 2.4%; P < .001) but a lower endotracheal intubation rate (0.8% vs 2.4%; P < .001).

Methods

Data Source

We conducted a multi-center retrospective cohort study by using in-patient encounter data from January 1, 2017, through December 31, 2021, from the VPS registry (Virtual Pediatric Systems, Los Angeles, California [http://www.myvps.org]). This registry represents data retrospectively collected through an international, voluntary collaborative, including de-identified data from more than 200 PICUs. The study period described above was chosen as NRS devices became mandatory fields in 2017. The study protocol was reviewed and approved by the Johns Hopkins All Children’s Hospital Institutional Review Board (IRB 00179595; initial approval on June 27, 2018).

Study Participants and Cohorts

Inclusion criteria were the following: (1) children 2 through 17 years of age, (2) admission to a participating center PICU located within the United States, (3) a principal diagnosis that corresponded with critical asthma (eg, status asthmaticus, asthma exacerbation) by using System Tracking for Audit and Reimbursement and the ICD-9,¹⁵ and the ICD-10 diagnostic codes,¹⁶ and (4) exposure to NRS during patients’ PICU stay. Exclusion criteria were the following: (1) admission tracheostomy dependence, (2) critical congenital heart disease, (3) pulmonary hypertension, (4) acute bronchiolitis, (4) acute chest syndrome, (5) acute laryngotracheobronchitis, and (6) if NRS was used only after endotracheal extubation. Children < 2 years of age were not included to avoid misclassification with diagnoses, for example, bronchiolitis, that may also present with bronchospasm. A comprehensive list of coding that corresponds to the study criteria are listed in Supplementary Table 1 (see the supplementary materials at http://www.rcjournal.com). The study sample was divided into cohorts defined by the initial NRS device applied on PICU admission. The VPS registry records NRS start and stop dates and times, but not unique NRS settings (eg, inspired oxygen, pressure, and flow parameters) or physical interfaces (ie, full face vs nasal masks).

Study Definitions and Outcomes

The primary outcomes were NRS device prescribing rates, including exposure to HFNC, BPAP, and CPAP by calendar year and hospital center. The secondary outcomes included patient descriptive characteristics and in-patient clinical trajectory summated and compared by the initial NRS device applied in the PICU setting. Patient descriptive features included demographics, anthropometrics, comorbidities, and severity of illness indices (ie, Pediatric Risk of Mortality III Probability of Mortality [PRISM III–POM] and Pediatric Index of Mortality III Risk of Mortality values). In-patient clinical outcomes included NRS failure rate; invasive ventilation data (including rate, timing, and duration of exposure); PICU stay (length of stay); NRS duration of exposure; and the frequency of pneumothoraces, cardiopulmonary resuscitation, and in-hospital mortality. We defined NRS failure as the following: (1) endotracheal intubation and invasive ventilation after NRS exposure, and (2) transition to an alternative NRS device (eg, HFNC transition to BPAP or CPAP, or CPAP transition to BPAP).

Statistical Analyses

Descriptive statistics are reported as proportion with percentage, mean ± SD, or median (interquartile range [IQR]). After assessment of data distribution and normalcy via Kolmogorov-Smirnov tests, continuous variables were compared among cohorts defined by initial NRS device applied in the PICU by using Kruskal Wallis and one-way analysis of variance. For categorical data, chi-square tests were used. Associations between the initial NRS device applied and subsequent NRS failure were assessed by using an exploratory logistic regression, yielding an adjusted odds ratio with 95% CI. A priori covariates were patient age, PRISM III–POM, and hospital center to account for severity of illness and potential practice variation at the study sites. All analyses were 2-sided, and type I error was set at 0.05. Encounters were used as the unit of analyses and assumed independent. Missing data were not imputed. Statistical analyses were completed by using Stata v15.1 (Stata Corp, College Station, Texas).

Results

Study Sampling and Cohorts

We identified 10,083 unique encounters of patients hospitalized for critical asthma from 125 participating PICUs prescribed NRS and eligible for study (Fig. 1). Subgroups were classified by the specific NRS device applied at the time of PICU admission, including HFNC (n = 6,562 [65.1%]), BPAP (n = 3,164 [31.4%]), and CPAP (n = 357/10,083 [3.5%]). Aggregate annual NRS device prescribing rates are depicted in Figure 2. During the study period, HFNC prescribing increased annually by a mean of 2.5%, from 59.7% in 2017 to 71.9% in 2021. In contrast, BPAP use decreased by a mean of 1.6% annually, from 35.5% to 26.3%, and CPAP use decreased by a mean of .8% annually, from 5% to 1.8%. Although the mean ± SD initial NRS device selected varied widely by hospital center (69.7 ± 29.6% for HFNC, 27.2 ± 27.1% for BPAP, and 3.1 ± 5.9% for CPAP), prescribing rates did not correlate with center-specific critical asthma admission volumes.

Fig. 1.

Flow chart. VPS = Virtual Pediatric Systems, NRS = noninvasive respiratory support, HFNC = high-flow nasal cannula, BPAP = bi-level positive airway pressure.

Fig. 2.

Noninvasive respiratory support prescribing trends for pediatric critical asthma from 2017 through 2021 among 125 centers within the Virtual Pediatric Systems registry. HFNC = high-flow nasal cannula, BPAP = bi-level positive airway pressure.

Encounter Characteristics and In-Patient Clinical Outcomes

Encounter descriptive data and clinical outcomes can be found in Table 1. Compared with children prescribed BPAP or CPAP, those initially on HFNC were younger (mean ± SD HFNC, 6.9 ± 3.9 years; BPAP, 9 ± 4.5 years; and CPAP, 7.6 ± 4.4 years; P < .001), less frequently obese (HFNC, 6.7%; BPAP, 12.4%; and CPAP, 10.4%; P < .001), and had a median (IQR) lower severity of illness as indicated by PRISM III–POM scores (HFNC, 0.3% [0.3–0.7%]; BPAP, 0.6% [0.3–0.9%]; and CPAP, 0.5% [0.3–0.7%]; P < .001). Statistically significant, but non-clinically relevant, differences in median (IQR) PICU length of stay were observed (HFNC, 3 [3.1–4.4] d; BPAP, 3.4 [2.5–4.8] d; and CPAP, 3.2 [2–4.7] d; P < .001). The BPAP group had greater rates of in-patient cardiopulmonary resuscitation (0.3%) compared with those prescribed HFNC (0.1%) or CPAP (0%) (P = .030). There were no detectable differences in mortality or acquired pneumothoraces by NRS device groups.

Table 1.

General Sample and NRS Subgroup Clinical Characteristics Among Children Hospitalized for Critical Asthma in the PICU

The overall NRS-failure rate was 6.3% (635/10,083) and occurred at a median (IQR) of 5.8 (2.6–19.2) h after device initiation. The frequency of HFNC failure was 7.3% (483/6,562) with a majority transitioned to BPAP (74.3% [359/483]) or CPAP (20.7% [100/483]) without further escalation to invasive ventilation (4.9% [24/483]). In contrast, BPAP failure occurred in 2.4% (76/3,164), which represented a transition to invasive ventilation. The overall frequency of invasive ventilation was 1.4% (139/10,083) and was greater among the BPAP (2.4% [76/3,164]) and CPAP (2.2% [8/357]) groups compared with those initially on HFNC (0.8% [55/6,562]). In an exploratory adjusted logistic regression that accounted for patient age, severity of illness (PRISM III–POM), and participating hospital center, the initial trial of HFNC was associated with subsequent NRS-failure (adjusted odds ratio 2.4, 95% CI 1.9-2.9; P < .001).

Discussion

In this multi-center, retrospective cohort study, we observed an increasing trend of HFNC prescribing (+2.5%/y) compared with BPAP (–1.6%/y) and CPAP (–0.8%/y) for pediatric critical asthma. Older children who were obese and with a greater severity of illness (as indicated by PRISM III–POM data) were more frequently prescribed BPAP and CPAP compared with HFNC. Although data from the VPS registry cannot differentiate the provider intention with regard to HFNC use (eg, as NRS or simply to deliver nebulized therapies), the elevated rate of HFNC failure (7.3%) compared with BPAP failure (2.4%) emphasizes a need for a prospective controlled study to identify judicious and targeted NRS application. A majority of the patients who experienced HFNC failure were transitioned to BPAP or CPAP without further escalation to invasive ventilation. We speculate that this demonstrates prescriber selection bias rather than superiority of one NRS device over another. Although we await the development of anticipated critical asthma practice guidelines from the American Association for Respiratory Care and the Pediatric Acute Lung Injury and Sepsis Investigators Consortium, we speculate that the absence of trial-derived NRS comparative effectiveness data contributes to the variation observed in this report.

Traditional NRS devices, for example, BPAP, optimally function by establishing a closed system to apply noninvasive ventilation. This approach generates improved minute ventilation and overcomes elevated intrinsic airway pressures present in critical asthma.¹⁷ A closed NRS system also permits the accurate titration of inspired gases and may optimize the delivery of nebulized medications. Systematic reviews and meta-analyses of pediatric trial data with regard to BPAP in critical asthma suggest sufficient evidence for its routine application in a monitored PICU setting but acknowledge the need for further inquiry.^7,18,19 Potential drawbacks of BPAP and CPAP include iatrogenic barometric airway trauma (ie, pneumothoraces and subcutaneous emphysema, reported in < 5%), mask-related pressure injuries (reported in 20–34%), gastric distention (reported in 5–10%), secondary aspiration pneumonitis (reported in < 5%), and the perception of patient discomfort or the requirement of sedation to optimize device adherence.^20,21 As an alternative, HFNC for critical asthma is gaining popularity given its simplicity, widespread availability, a perception of improved pediatric tolerability, and its function to deliver continuous nebulized therapies.^22-24

Several retrospective reports for children with asthma exacerbation have compared HFNC with standard oxygen therapy and observed improvements in breathing frequency, acidosis, and clinical severity scores.^25-28 The proposed mechanisms of action of HFNC include inspired gas conditioning (ie, applying heat and humidification), attenuation of nasopharyngeal inspiratory resistance, reduction of dead-space ventilation, and generation of a positive end-expiratory pressure.²⁹ Although an HFNC flow of 2 L/kg/min may generate 4–6 cm H₂O of pressure during exhalation,³⁰ flow > 40 L/min is rarely applied in pediatrics. In addition, HFNC flow > 6 L/min may compromise nebulized β₂ agonist delivery.^31-33 Determining when to apply HFNC to optimally deliver nebulized treatments at a lower flow versus applying a greater flow to enhance minute ventilation remains a critical knowledge gap. Because flow parameters are not recorded in the VPS registry, we could not account for the use intentions of providers who prescribed HFNC in this patient population.

Direct comparisons of BPAP to HFNC for pediatric critical asthma have been limited to single-center observational experiences. Pilar et al¹⁰ performed a retrospective cohort study of 42 children admitted for critical asthma, with 22 receiving HFNC and 20 receiving BPAP. They observed a 40% HFNC failure rate (n = 8) compared with 0% in the BPAP group. As with findings from this report, those in whom HFNC failed were simply transitioned to BPAP without further escalation. In 2022, our research group published a single-center retrospective cohort study of 39 children who received BPAP or HFNC for critical asthma.¹¹ Those initially receiving BPAP were older and had greater asthma severity, which suggests selection bias among providers using NRS in asthma. Findings from this report of the VPS registry corroborate these single-center experiences and highlight several opportunities for future research.

Limitations

The observational, retrospective nature of this registry-based report limits the ability to establish causality or delineate the rationale behind NRS device selection. The VPS registry does not include encounter-level granularity with regard to the severity of asthma exacerbation, the magnitude of NRS exposure (eg, flow and pressure parameters applied), and the presence and magnitude of other applied asthma treatments (eg, steroids, terbutaline, aminophylline, heliox, or magnesium). Although VPS data are quality assured at the institution level, there exists the possibility of errors related to misclassification as a result of erroneous coding that may account for our study findings. We assumed encounters were independent, but the subjects may represent repeated patients over multiple PICU admissions. Further, many VPS centers represent tertiary or quaternary facilities that may not reflect the care provided at community in-patient facilities or the emergency department environment where NRS is applied. The location of initiation of NRS is not discreetly stored within VPS; as a result, we could not determine if the type of NRS interface was a continuation of emergency department or general pediatric ward care versus the intention of a critical care provider team. Lastly, we excluded children < 2 years of age to improve subject homogeneity, but these findings cannot be generalized to this age group.

Conclusions

In this multi-center cohort study of children 2 to 17 years of age hospitalized in the PICU for critical asthma from 125 children’s hospitals, we observed an increasing use of HFNC (+2.5%/y) compared with BPAP (–1.6%/y) and CPAP (–0.8%/y). Although HFNC failure was more frequent than that of BPAP failure, most children transitioned to BPAP from HFNC without further escalation (ie, did not require endotracheal intubation). Future clinical trials should seek to determine the superiority of these NRS strategies, whereas evaluating pragmatic clinical end points such as patient comfort and the optimal delivery of nebulized therapies concurrent to NRS.