Abstract
A large multicenter retrospective study was conducted to examine practice patterns in the care of severe acute asthma in PICUs compared with adult ICUs. Eligible patients were 12–26 y old, admitted to a PICU or an adult ICU with severe acute asthma, and received noninvasive ventilation or mechanical ventilation. Multiple medication prescribing practice patterns were examined. 3,361 total patients were included: 583 (17.3%) were admitted to PICUs and 2,778 (82.7%) were admitted to adult ICUs. Significantly more patients admitted to PICUs received parenteral terbutaline (25.7% vs. 10.2%, standard mean differences = 0.412) than those admitted to adult ICUs. Admission to PICUs for asthma was associated with a trend toward increased mortality, cardiac arrhythmias and duration of mechanical ventilation compared with adult ICUs. Further studies are needed to assess the safety and efficacy of parenteral terbutaline in adolescents and young adults; guidelines for the management of severe acute asthma in the ICU are needed.
Keywords: asthma, intensive care, PICU, respiratory failure, terbutaline
KEY POINTS
Question: What is the practice pattern variation in the medical management of severe acute asthma in adolescents and young adults between PICUs and adult ICUs?
Findings: In this large claims-based retrospective cohort study, significantly more patients 12–26 y old with severe acute asthma undergoing mechanical ventilation received parenteral terbutaline (25.7% vs. 10.2%) if they were admitted to PICUs compared with adult ICUs. Admission to PICUs for asthma was associated with a trend toward increased mortality, cardiac arrhythmias and duration of mechanical ventilation compared with adult ICUs.
Meaning: More studies are needed on the safety and efficacy of parenteral terbutaline in the treatment of severe acute asthma among adolescents and young adults.
TO THE EDITOR:
Cases of severe acute asthma lead to almost 95,000 hospitalizations annually among children, adolescents, and adults in the United States (1). In the most severe cases, patients may require admission to the ICU for noninvasive ventilation (NIV) or invasive mechanical ventilation (IMV) (2). For many diseases associated with critical illness, the specific ICU that patients are admitted to is a strong driver of practices and outcomes; for example, adolescents admitted to adult SICUs rather than PICUs undergo higher rates of right heart catheterization and intubation (3, 4). In the case of asthma, adolescents and young adults may be admitted to a PICU or an adult ICU but it is unclear how medical practices differs between these two sites. In this study, we sought to characterize variation in the medical management of severe acute asthma between PICUs and adult ICUs among patients undergoing mechanical ventilation across U.S. hospitals. We hypothesized there would be significant differences in the practices between the two care sites given the lack of guidelines for critically ill patients with severe acute asthma.
METHODS
We used the Premier PINC AI Healthcare Database (2016–2022), an enhanced claims-based database that includes ~25% of all U.S. inpatient hospitalizations (5). We selected patients 12–26 years old with a diagnosis of asthma exacerbation (International Classification of Disease, 10th Revision, J45.21, J45.22, J45.31, J45.32, J45.41, J45.42, J45.51, J45.52, J45.901, and J45.902) or asthma (J45.x) plus respiratory failure (J96.0 and J96.2) present at admission admitted to a PICU or adult ICU and receiving steroids (prednisone, prednisolone, methylprednisolone, or dexamethasone) on hospital day 1 (6). We further restricted the cohort to patients who received NIV or IMV on hospital day 1 to enrich for patients with more severe exacerbations. We defined the lower limit of adolescence as 12 years old per Global Initiative for Asthma guidelines (7). We defined the upper limit of young adulthood as 26 years old as individuals can remain on parents’ insurance until 26 and prior literature has defined young adulthood lasting until age 26 (8). We excluded patients who received surgery on hospital day 1 or were admitted from an outside hospital.
We quantified the percentage of patients that received each of the following asthma treatments: inhaled corticosteroids, inhaled antimuscarinics, inhaled anesthetics, inhaled heliox, parenteral magnesium, parenteral paralytics, parenteral terbutaline, parenteral epinephrine, leukotriene inhibitors, theophylline/aminophylline and antibiotics during days 1 to 7 of hospitalization. We used standard mean differences (SMD) to assess the degree to which treatments differed between PICUs and adult ICUs (SMD > 0.2 was used to define practices that differed) (9, 10). Practices pattern differences were reassessed in a sensitivity analysis using a narrower cohort definition (patients 16–19 years old). In exploratory analyses, we then used hierarchical logistic and linear regression models to identify associations between admission to the PICU (vs. adult ICU) and patient outcomes of hospital mortality (primary outcome), mechanical ventilation duration, hospital length of stay, and diagnosis of a cardiac arrhythmia (using ICD-10 codes, among those without a cardiac arrythmia diagnosis present at admission) (11). Hierarchical models included fixed effect terms for demographics, hospital characteristics, COVID-19 diagnosis, asthma severity, use of IMV on hospital day 1, comorbidity burden, cardiac arrest present at admission (ICD-10 codes “I46.2,” “I46.8,” “I46.9”), and validated acute organ dysfunctions present at admission and a hospital indicator random intercept term to account for clustering (11, 12). This study was deemed not human subjects research by the Boston University Institutional Review Board (#H-41795).
RESULTS
We identified 3,361 patients with asthma exacerbations who received mechanical ventilation and steroids on hospital day 1 across 80 PICUs and 636 adult ICUs (PICU: 583 [17.3%]; adult ICU: 2,778 [82.7%]). The median number of included patients per PICU was 3.5 (interquartile range [IQR], 2–7.25) and per adult ICU was 3 (IQR, 2–5). The median age of patients admitted to PICUs was 15 y (IQR, 13–16) and in adult ICUs was 23 y (IQR, 20–25). Patients admitted to PICUs had lower rates of renal (5.0% vs. 10.5%) and neurologic (6.0% vs. 12.6%) organ dysfunction present at admission compared with adult ICU patients, were more likely to have an underlying diagnosis of severe asthma (29.7% vs. 23.4%) but were less likely to receive IMV on hospital day 1 (37.2% vs. 60.3%) (Table 1) (12).
TABLE 1.
Patient Characteristics and Covariates at Admission or Hospital Day 1
| Patient Characteristics | Overall (n = 3,361) | Adult ICU (n = 2,778) | PICU (n = 583) |
|---|---|---|---|
| Age, median [IQR] | 22 (19–24) | 23 (20–25) | 15 (13–16) |
| Sex, female, n (% female) | 1,819 (54.1) | 1,536 (55.3) | 283 (48.5) |
| Race, n (%) | |||
| Asian | 55 (1.6) | 36 (1.3) | 19 (3.3) |
| Black | 1,261 (37.5) | 998 (35.9) | 263 (45.1) |
| White | 1,578 (47.0) | 1,367 (49.2) | 211 (36.2) |
| Other | 373 (11.1) | 302 (10.9) | 71 (12.2) |
| Unknown | 94 (2.) | 75 (2.7) | 19 (3.3) |
| Ethnicity, n (%) | |||
| Hispanic | 400 (11.9) | 295 (10.6) | 105 (18.0) |
| Acute organ dysfunction, n (%) | |||
| Renal | 321 (9.6) | 292 (10.5) | 29 (5.0) |
| Neurologic | 384 (11.4) | 349 (12.6) | 35 (6.0) |
| Cardiac | 268 (8.0) | 229 (8.2) | 39 (6.7) |
| Hematologic | 70 (2.1) | 57 (2.1) | 13 (2.2) |
| Hepatic | 39 (1.2) | 36 (1.3) | 3 (0.5) |
| Cardiac arrest, n (%) | 181 (5.4) | 164 (5.9) | 17 (2.9) |
| Asthma diagnosis severity, n (%) | |||
| Unknown | 1,935 (57.6) | 1,739 (62.6) | 196 (33.6) |
| Mild | 237 (7.1) | 159 (5.7) | 78 (13.4) |
| Moderate | 365 (10.9) | 229 (8.2) | 136 (23.3) |
| Severe | 824 (24.5) | 651 (23.4) | 173 (29.7) |
| COVID-19 diagnosis, n (%) | 135 (4.0) | 100 (3.6) | 35 (6.0) |
| Gagne comorbidity index, median (IQR) | 2 (1–3) | 2 (1–3) | 2 (1–2) |
| Hospital characteristics, n (%) | |||
| Midwest | 805 (24.0) | 714 (25.7) | 91 (15.6) |
| Northeast | 500 (14.9) | 375 (13.5) | 125 (21.4) |
| South | 1,357 (40.4) | 1,071 (38.6) | 286 (49.1) |
| West | 699 (20.8) | 618 (22.2) | 81 (13.9) |
| Urban hospital | 3,078 (91.6) | 2,503 (90.1) | 575 (98.6) |
| Teaching hospital | 1,778 (52.9) | 1,313 (47.3) | 465 (79.8) |
IQR = interquartile range.
The only medication to be used more frequently in the PICU was parenteral terbutaline (25.7% vs. 10.2%, SMD = 0.412). Medications used less frequently in the PICU included antibiotics (45.3% vs. 69.5%, SMD = 0.505), inhaled antimuscarinics (64.0% vs. 83.8%, SMD = 0.462), inhaled corticosteroids (34.1% vs. 45.8%, SMD = 0.239), and parenteral paralytics (7.2% vs. 16.6%, SMD = 0.293). Medications with no difference between PICU and adult ICU were inhaled heliox, inhaled epinephrine, leukotriene inhibitors, theophylline, parenteral epinephrine, parenteral magnesium, and ketamine (Table 2). Practices were similar in the sensitivity analysis among patients 16–19 years old.
TABLE 2.
Asthma Therapies Billed for on Hospital Days 1 to 7
| Full Cohort | Ages 16 to 19 | |||||
|---|---|---|---|---|---|---|
| Asthma Therapies | Adult ICU (n = 2,778) | PICU (n = 583) | SMD | Adult ICU (n = 463) | PICU (n = 194) | SMD |
| Inhaled therapies, n (%) | ||||||
| Inhaled antimuscarinics | 2,327 (83.8) | 373 (64.0) | 0.462 | 384 (82.9) | 118 (60.8) | 0.507 |
| Inhaled heliox | 56 (2.0) | 17 (2.9) | 0.058 | 13 (2.8) | 5 (2.6) | 0.014 |
| Inhaled corticosteroids | 1,271 (45.8) | 199 (34.1) | 0.239 | 220 (47.5) | 64 (33.0) | 0.300 |
| Inhaled epinephrine | 146 (5.3) | 20 (3.4) | 0.090 | 22 (4.8) | 8 (4.1) | 0.030 |
| Inhaled anesthetics | 8 (0.3) | 1 (0.2) | 0.024 | 3 (0.6) | 0 (0.0) | 0.114 |
| Oral therapies, n (%) | ||||||
| Leukotriene inhibitors | 794 (28.6) | 171 (29.3) | 0.017 | 139 (30.0) | 53 (27.3) | 0.060 |
| Theophylline/aminophylline | 41 (1.5) | 4 (0.7) | 0.076 | 6 (1.3) | 0 (0.0) | 0.162 |
| Parenteral therapies, n (%) | ||||||
| Parenteral terbutaline | 284 (10.2) | 150 (25.7) | 0.412 | 66 (14.3) | 51 (26.3) | 0.303 |
| Parenteral paralytics | 461 (16.6) | 42 (7.2) | 0.293 | 65 (14.0) | 14 (7.2) | 0.223 |
| Parenteral epinephrine | 629 (22.6) | 128 (22.0) | 0.016 | 99 (21.4) | 47 (24.2) | 0.068 |
| Parenteral ketamine | 610 (22.0) | 94 (16.1) | 0.149 | 91 (19.7) | 30 (15.5) | 0.110 |
| Parenteral magnesium | 1,844 (66.4) | 389 (66.7) | 0.007 | 311 (67.2) | 123 (63.4) | 0.079 |
| Antibiotics, n (%) | ||||||
| Antibiotics | 1,930 (69.5) | 264 (45.3) | 0.505 | 311 (67.2) | 88 (45.4) | 0.451 |
SMD = standard mean differences.
In the hierarchical multivariable-adjusted models, there was no difference between PICU admission and adult ICU admission for the primary outcome of hospital mortality (adjusted odds ratio 2.12 [95% CI, 0.85–5.26]). In secondary outcomes, there was an association between PICU admission and increased frequency of cardiac arrhythmias (adjusted odds ratio 2.48 [95% CI, 1.07–5.72]) and longer mechanical ventilation duration (beta 0.8 days [95% CI, 0.3–1.3]), but not hospital length of stay (0.4 d [95% CI –0.6 to 1.3]) (Table 3).
TABLE 3.
Exploratory Analyses
| Patient Outcomes | Overall (n = 3,361) | Adult ICU (n = 2,778) | PICU (n = 583) | Adjusted Effect Estimate (95% CI)a PICU vs. Adult ICU |
|---|---|---|---|---|
| Primary outcome | ||||
| Hospital mortality, n (%) | 166 (4.9) | 143 (5.1) | 23 (3.9) | 2.12 (0.85–5.26) |
| Secondary outcomes | ||||
| Mechanical ventilation duration, d—mean (sd) | 3.5 (4.0) | 3.5 (3.9) | 3.8 (4.5) | 0.8 (0.3–1.3) |
| Hospital length of stay, d—mean (sd) | 6.3 (7.2) | 6.2 (7.2) | 6.7 (6.8) | 0.4 (–0.6 to 1.3) |
| Cardiac arrhythmias, n (%)b | 113 (4.1) | 80 (3.5) | 33 (7.1) | 2.48 (1.07–5.72) |
Hierarchical linear and logistic regression model included demographics, hospital characteristics, COVID-19 diagnosis, asthma severity, comorbidity burden, use of invasive mechanical ventilation on hospital day 1, acute organ dysfunctions present at admission, and cardiac arrest present at admission. Odds ratio for hospital mortality and cardiac arrhythmias, mean difference in days for mechanical ventilation duration and hospital length of stay.
Among patients without cardiac arrhythmias present at admission (n = 2,742).
DISCUSSION
In adolescents and young adults with severe acute asthma undergoing invasive or NIV, admission to the PICU vs. the adult ICU was associated with many differences in asthma management, including large differences in the use of parenteral terbutaline, antimuscarinics, paralytics, inhaled corticosteroids and antibiotics. In addition, patients admitted to the PICU had longer mechanical ventilation durations and higher risks for cardiac arrhythmias despite lower baseline levels of IMV. These results suggest the long-term need to (a) quantify the safety and heterogeneous effects for potentially harmful therapies such as terbutaline across age groups and (b) establish guidelines for the management of critically ill patients with severe acute asthma.
There is limited evidence for parenteral terbutaline use and guidelines for non-mechanically ventilated patients only recommend IV terbutaline in acute asthma for children under 6, an age group excluded from our study (7, 10, 13). A randomized clinical trial comparing IV terbutaline to normal saline in pediatric patients (average age: 7 y) with severe acute asthma showed trends toward lower asthma severity after 24 h in the terbutaline treated group but also higher rates of cardiac arrhythmia and serum troponin elevation (14). Other small retrospective studies have shown a safe side effect profile of parenteral terbutaline (15, 16). We hypothesize that the large difference in terbutaline use in PICUs vs. adult ICUs could be due to adult ICU providers extrapolating concerns about cardiac arrhythmias with terbutaline use in older patients with cardiac comorbidities to younger, healthier patients. In our experience, providers in both the PICU and adult ICU use terbutaline to try and avoid IMV or to shorten its duration. Future studies examining the heterogeneity of treatment effect of parenteral terbutaline are needed, including understanding the population most likely to benefit and the population in which parenteral beta-agonists poses high enough risk to be contraindicated. Furthermore, this study highlights the absence of national consensus guidelines for the treatment of acute asthma in the ICU and we recommend a combined adult-pediatric task force be assembled to perform a systemic review and Delphi consensus process to help determine the best evidence-based recommendations and generation of additional high-quality efficacy studies to inform these future guidelines.
Our study has limitations. Patients admitted to adult ICUs generally had higher rates of organ failure than the PICU group. Thus, differences in severity of illness may explain higher uses of some therapies in the adult ICU. However, our finding of higher use of parenteral terbutaline in the PICU would not be explained by higher severity of illness in the adult ICU given that these treatments are considered “rescue” therapies when inhaled beta-agonists are ineffective. As we were unable to capture emergency department practices, pre-ICU practices are unclear and could have influenced rates of NIV and IMV in the ICU. We were unable to capture electrocardiogram data or laboratory serum troponin data between patients who did and did not receive IV terbutaline. Given the limitations of the database we were unable to examine variation in medication dosing, such as dosing of continuous albuterol. We also couldn’t analyze the site of intubation—pre-hospital, emergency department, or ICU—or ventilator management strategies given limitations of the database. We included mechanical ventilation type on day 1 (IMV vs. NIV) in models to account for baseline differences in asthma severity. However, it is possible that differences in the frequency of mechanical ventilation represent differences in PICU vs. adult ICU practices rather than differences in asthma severity. Finally, our observation that PICU patients had longer average durations of mechanical ventilation compared with adult ICU patients could potentially be due to differences inpatient selection for intubation in the pediatric vs. adult medical culture; we did not have granular patient-level data of patients” degree of respiratory compromise at time of intubation to perform such an analysis. Longer duration of mechanical ventilation in PICUs could also be due to the observed medical practices in our study including lower rates of administration of antibiotics, inhaled antimuscarinics, inhaled corticosteroids and parenteral paralytics in the PICU compared with the adult ICU.
In a large multicenter retrospective study, admission to the PICU (compared with the adult ICU) was associated with differences in severe asthma management including higher use of parenteral terbutaline, increased rates of cardiac arrhythmias, and increased duration of mechanical ventilation. More studies are needed to establish terbutaline’s safety profile in this age range. Guidelines for the treatment of asthma in the ICU are needed.
Footnotes
This study was supported by National Institute of Health (NIH) National Center for Advancing Translational Sciences (NCATS) (grant number 1KL2TR001411), National Heart, Lung, and Blood Institute (grant numbers K23HL153482 and 1UL1TR001430), and the Boston University Chobanian & Avedisian School of Medicine Department of Medicine Career Investment Award. This study’s contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or Boston University. The NIH and Boston University had no role in the design and conduct of the study, collection, management, analysis, and interpretation of the data, preparation, review, approval of the article, or decision to submit the article for publication.
The authors have disclosed that they do not have any potential conflicts of interest..
Contributor Information
Burton Shen, Email: burton.shen@bmc.org.
Collin Homer-Bouthiette, Email: collin.homer-bouthiette@bmc.org.
Divya Shankar, Email: divya.shankar2@bmc.or.
Brandon Pang, Email: brandon.pang@bmc.org.
Anica Law, Email: anicalaw@bu.edu.
Nicholas A. Bosch, Email: nicholas.bosch@bmc.org.
REFERENCES
- 1.CDC. Most recent national asthma data. Available at: https://www.cdc.gov/asthma/most_recent_national_asthma_data.htm. Accessed July 28, 2023
- 2.Leatherman J: Mechanical ventilation for severe asthma. Chest 2015; 147:1671–1680 [DOI] [PubMed] [Google Scholar]
- 3.Sanchez JL, MD, Lucas J, MD, Feustel PJ, PhD: Outcome of adolescent trauma admitted to an adult surgical intensive care unit versus a pediatric intensive care unit. J Trauma 2001; 51:478–480 [DOI] [PubMed] [Google Scholar]
- 4.Kurtz P, Fitts V, Sumer Z, et al. : How does care differ for neurological patients admitted to a neurocritical care unit versus a general ICU? Neurocrit Care 2011; 15:477–480 [DOI] [PubMed] [Google Scholar]
- 5.PINC AITM Applied Sciences, Premier Inc: PINC AITM healthcare database: Data that informs and performs (White Paper). 2024
- 6.Lindenauer PK, Stefan MS, Feemster LC, et al. : Use of antibiotics among patients hospitalized for exacerbations of asthma. JAMA Intern Med 2016; 176:1397–1400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Global Initiative for Asthma. Global Strategy for Asthma Management and Prevention, 2024. Available at: www.ginasthma.org. Accessed May 2024 [Google Scholar]
- 8.White PH, Cooley WC; Transitions Clinical Report Authoring Group, American Academy of Pediatrics, American Academy of Family Physicians, American College of Physicians: Supporting the health care transition from adolescence to adulthood in the medical home. Pediatrics 2018; 142:e20182587. [DOI] [PubMed] [Google Scholar]
- 9.Cohen J: Statistical Power Analysis for the Behavioral Sciences. Second Edition. Routledge,1988 [Google Scholar]
- 10.Gallardo-Gómez D, Richardson R, Dwan K: Standardized mean differences in meta-analysis: A tutorial. Cochrane Ev Synth 2024; 2:e12047 [Google Scholar]
- 11.Sun JW, Rogers JR, Her Q, et al. : Adaptation and validation of the combined comorbidity score for ICD-10-CM. Med Care 2017; 55:1046–1051 [DOI] [PubMed] [Google Scholar]
- 12.Bosch NA, Law AC, Rucci JM, et al. : Predictive validity of the sequential organ failure assessment score versus claims-based scores among critically Ill patients. Ann Am Thorac Soc. 2022; 19:1072–1076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Travers AH, Milan SJ, Jones AP, et al. : Addition of intravenous beta(2)-agonists to inhaled beta(2)-agonists for acute asthma. Cochrane Database Syst Rev 2012; 12:CD010179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bogie AL, Towne D, Luckett PM, et al. : Comparison of intravenous terbutaline versus normal saline in pediatric patients on continuous high-dose nebulized albuterol for status asthmaticus. Pediatr Emerg Care 2007; 23:355–361 [DOI] [PubMed] [Google Scholar]
- 15.Kambalapalli M, Nichani S, Upadhyayula S: Safety of intravenous terbutaline in acute severe asthma: A retrospective study. Acta Paediatr (Oslo, Norway : 1992) 2005; 94:1214–1217 [DOI] [PubMed] [Google Scholar]
- 16.Doymaz S, Schneider J: Safety of terbutaline for treatment of acute severe pediatric asthma. Pediatr Emerg Care 2018; 34:299–302 [DOI] [PubMed] [Google Scholar]