Abstract
Background
Little is known about the clinical outcomes of patients admitted to the intensive care unit (ICU) for severe asthma exacerbation requiring ventilatory support.
Research question
Our objectives were (i) to describe the initial presentation, therapeutic management and clinical outcomes of adults admitted to the ICU for asthma exacerbation requiring ventilatory support and (ii) to identify risk factors for a prolonged course and/or fatal outcome.
Study design and methods
We conducted a 10-year multicentre (n=21) retrospective study in France. All adult patients with asthma admitted to the ICU for an exacerbation requiring invasive (IMV) and/or non-invasive (NIV) mechanical ventilation were included. The primary endpoint was ‘complicated course’, a composite criterion of death during the ICU stay and/or the need for IMV for ≥7 days.
Results
From 2010 to 2019, 349 patients were included (women 57%, age 54 (40–65) years). Acute respiratory failure was the primary reason for ICU admission. However, 12% of patients were admitted following a cardiorespiratory arrest. Respiratory tract infections were documented in 36% of patients, with Haemophilus influenzae, Streptococcus pneumoniae and Rhinovirus as the predominant pathogens. Three-quarters of the patients required IMV, with a median duration of 5 (3–12) days. One-third underwent NIV as initial ventilatory support, but NIV failed in 23% of them. In-hospital mortality was 8%. The composite primary endpoint was observed in 34% of patients. Multivariable analysis identified Sepsis-related Organ Failure Assessment score, pH≤7.3 and a documented respiratory tract infection on ICU admission as independent risk factors for a complicated course.
Conclusions
Exacerbation requiring ventilatory support was associated with a substantial burden in adult patients with asthma. Respiratory infection was the predominant suspected triggering factor. Respiratory tract infections were documented in 36% of patients and were independently associated with a prolonged course and/or fatal outcome in the ICU.
Keywords: Asthma, Critical Care, Assisted Ventilation
WHAT IS ALREADY KNOWN ON THIS TOPIC
Severe asthma exacerbations requiring ventilatory support are uncommon but associated with significant morbidity and mortality. Evidence regarding intensive care unit (ICU) management, triggering factors and predictors of poor outcomes in this population remains limited.
WHAT THIS STUDY ADDS
This study provides the most comprehensive contemporary description of the presentation, ventilatory management, complications and outcomes of adults with severe asthma exacerbations requiring ventilatory support. It confirms the substantial burden of morbidity and mortality in this population and demonstrates that respiratory tract infections are common and independently associated with prolonged invasive ventilation or death.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The study identifies early bedside prognostic markers—including respiratory infection, acidosis and organ failure severity—that may help clinicians rapidly stratify risk and optimise early management in patients admitted in the ICU with severe asthma exacerbations.
Introduction
Asthma is a common chronic disorder affecting nearly 400 million people worldwide.1 Its natural history is characterised by acute exacerbations, defined as a sudden worsening of symptoms and lung function from the patient’s usual status.2 Despite prompt management using mainly oxygen, bronchodilators and steroids, up to a third of patients presenting to the emergency department for acute exacerbation of asthma fail to improve and require hospitalisation.3–5 A small proportion of these patients (2%–20%) are admitted to the intensive care unit (ICU),3 5–7 and approximately 15%–60% of them require endotracheal intubation for invasive mechanical ventilation (IMV).6–10 Respiratory support may also be provided through non-invasive ventilation (NIV). Despite limited supporting evidence11 12 and the absence of recommendations from scientific organisations,2 13 the use of NIV has increased among physicians in Western countries over the past decade.3 4 14 In this population of severely ill patients, morbidity and mortality remain considerable. For patients requiring IMV, the median duration of mechanical ventilation ranges from 3 to 4 days,9 15–17 and the ICU length of stay ranges from 3 to 7 days.4 6–8 15–17 Death occurs in approximately 2%–15% of patients requiring ventilation (IMV and/or NIV).3–8 15–17 There is limited knowledge regarding risk factors for poor clinical outcomes in patients admitted in the ICU with asthma exacerbation. Studies with small sample sizes have identified intrinsic risk factors for hospital mortality, such as age and immunocompromised status.7 8 Database studies have identified risk factors mainly related to the initial presentation, such as neurological disorders,7 8 occurrence of cardiorespiratory arrest during prehospital management,7 10 15 acidemia or hypercapnia,7–9 presence of organ failure8–10 15 and necessity of IMV.8–10 However, these studies presented several limitations. First, no large-sample study has specifically examined the most severely ill population (requiring mechanical ventilation). Second, previous studies have focused exclusively on hospital mortality; however, given the predictably low mortality rate, a composite endpoint of morbidity would be more informative for clinicians. Third, most studies used large-scale databases, resulting in limited data on the triggering factors of exacerbation and their impact on the ICU course and clinical outcomes. Finally, only a few studies have focused on the therapeutic management within the ICU, especially regarding ventilatory support. To gain insights into this issue, we designed the 10-year retrospective multicentre ESAVA study. Our objective was to describe the initial presentation, therapeutic management and clinical outcomes of adults admitted to the ICU with acute exacerbation of asthma requiring ventilatory support. Furthermore, we sought to identify risk factors associated with a prolonged course and/or fatal outcome in the ICU.
Methods
Study design and patient selection
We conducted a 10-year retrospective multicentre cohort study in 21 ICUs in France. From 1 January 2010 to 31 December 2019, all adult patients (aged ≥18 years) with asthma (as defined by the Global Initiative for Asthma (GINA)2), admitted to the participating ICUs with an acute exacerbation of asthma requiring IMV and/or NIV during the first 48 hours of their ICU stay were screened. Patients who met the eligibility criteria were included (for details regarding patient selection and inclusion and exclusion criteria, see online supplemental material).
For each recruited patient, the investigator collected pre-defined clinical, biological, microbiological and radiological data on a case report form. The primary reason for ICU admission and the suspected triggering factor for exacerbation were extracted from the medical charts (for details and definitions, see online supplemental material). Recruited patients were followed until hospital discharge.
The study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines. For the detailed STROBE checklist, see online supplemental material.
Definitions
The primary ventilatory support indicated the first type of ventilatory support (either NIV or IMV) used after ICU referral. NIV was considered exclusive if the patient did not undergo IMV during their ICU stay. NIV failure was defined as the need for IMV after primary ventilatory support with NIV, irrespective of the duration of NIV before intubation. Patients were considered to have a respiratory tract infection at ICU admission if at least one pathogen was documented within the respiratory tract from day-1 to day+2 of ICU referral. Chronic immunosuppression was defined by at least one of the following conditions: congenital immune deficiency, splenectomy, neutropenia (<0.5 G/L), HIV infection, long-term corticosteroid therapy (>5 mg of prednisone equivalent daily dose for at least 30 days), other long-term immunosuppressive therapy, solid-organ or bone marrow transplantation, malignant hemopathy or cancer and chemotherapy in the last 30 days. The Simplified Acute Physiology Score II (SAPS II)18 and the Sepsis-related Organ Failure Assessment (SOFA) score19 were calculated using the worst values recorded during the first 24 hours in the ICU.
The primary endpoint was ‘complicated course’, which was a composite criterion of death during the ICU stay and/or the need for IMV for ≥7 days (for details, see online supplemental material).
Statistical analysis
Continuous variables were expressed as median (first and third quartiles), and their distribution was compared using the non-parametric Mann–Whitney U test (for independent samples) or the Wilcoxon signed-rank test (for paired samples). Categorical variables were presented as counts (percentages), and their distribution was compared using Fisher’s exact test.
Comparisons were conducted based on the primary endpoint, focusing on demographics, comorbid conditions, disease history, clinical and paraclinical findings and hospital course. Additional analyses evaluated the evolution of blood gas and ventilatory parameters during the first 3 days of mechanical ventilation. For some variables, such as airway pressure, relative change (percentage) was used instead of absolute change.
A multivariable logistic regression model was developed to identify variables independently associated with the primary endpoint. All clinically relevant variables were included without univariable screening. To reduce the risk of overfitting, the number of covariates in the model was limited to 11, ensuring approximately 10 events per variable.
Estimates were made after multiple imputation of all predictors with at most 3% missing data. These variables were imputed by chained equations on 30 imputed datasets, and the estimated coefficients on each of these datasets were aggregated according to Rubin’s rule to obtain the coefficients of the final model. The associations between variables and the primary composite endpoint were reported as ORs with their 95% CI.
Comparative analyses between the first and the second half of the inclusion window (2010–2014 vs 2015–2019) were made to describe the temporal trends of in-ICU management and outcomes over the 10-year period.
All tests were two-tailed at the 0.05 significance threshold. Analyses were carried out using R V.4.0.3.
Patient and public involvement
Patients and/or members of the public were not involved in the design, conduct, reporting or dissemination plans of this research. The study was based on retrospective analysis of routinely collected clinical data.
Results
Population and exacerbations characteristics
From January 2010 to December 2019, 378 patients were screened (figure 1). The final study group consisted of 349 patients (women 57%, age 54 (40–65) years, with body mass index of 26 (23–30) kg/m2, 18% had chronic immunosuppression, 29% were current smokers and the SAPS II score was 36 (26–50)) (table 1).
Figure 1. Flow chart of patients. ICU, intensive care unit.
Table 1. Baseline characteristics of adult patients admitted to the ICU for asthma exacerbation requiring ventilatory support.
| All patients | Missing data | Complicated course | P value | ||
|---|---|---|---|---|---|
| No | Yes | ||||
| n=349 | n=231 | n=118 | |||
| Age, median (IQR), years | 54 (40–65) | 0 | 54 (39–68) | 53 (41–64) | 0.51 |
| Female sex, No. (%) | 200 (57) | 0 | 133 (58) | 67 (57) | 0.91 |
| Body mass index, median (IQR), kg/m2 | 26 (23–30) | 23 | 26 (23–30) | 28 (23–31) | 0.56 |
| Current smoking, No. (%) | 97 (29) | 11 | 57 (26) | 40 (34) | 0.13 |
| Chronic immunosuppression, No. (%) | 64 (18) | 0 | 43 (19) | 21 (18) | 0.89 |
| History of asthma | |||||
| Allergic asthma, No. (%) | 151 (43) | 0 | 102 (44) | 49 (42) | 0.65 |
| Long-term inhaled CS, No. (%) | 264 (76) | 0 | 178 (77) | 86 (73) | 0.43 |
| Previous intubation, No. (%) | 65 (19) | 4 | 49 (21) | 16 (14) | 0.11 |
| Asthma exacerbation | |||||
| Time from symptom onset to ICU admission, median (IQR), days | 2((0–4) | 1 | 2((0–5) | 2 (1–4) | 0.90 |
| Pre-hospital systemic CS, No. (%) | 159 (46) | 0 | 108 (47) | 51 (43) | 0.57 |
| Suspected triggering factor | 1 | ||||
| Respiratory tract infection, No. (%) | 122 (35) | – | 73 (32) | 49 (42) | 0.08 |
| Allergen/toxic exposure, No. (%) | 39 (11) | – | 31 (14) | 8 (7) | 0.07 |
| Therapy non-adherence/shift, No. (%) | 45 (13) | – | 36 (16) | 9 (8) | 0.04 |
| Other, No. (%) | 17 (5) | – | 11 (5) | 6 (5) | 1.00 |
| None, No. (%) | 131 (38) | – | 82 (36) | 49 (42) | 0.30 |
| Reason for ICU admission | 0 | 0.01 | |||
| Acute respiratory failure, No. (%) | 284 (81) | – | 196 (85) | 88 (75) | – |
| Cardiorespiratory arrest, No. (%) | 42 (12) | – | 19 (8) | 23 (20) | – |
| Coma, No. (%) | 21 (6) | – | 15 (7) | 6 (5) | – |
| Characteristics on ICU admission | |||||
| Fever (>38.0 °C), No. (%) | 65 (19) | 0 | 43 (19) | 22 (19) | 1.00 |
| Wheezing, No. (%) | 265 (76) | 0 | 174 (75) | 91 (77) | 0.79 |
| Agitation or coma, No. (%) | 161 (46) | 0 | 96 (42) | 65 (55) | 0.018 |
| Arterial pH, median (IQR) | 7.23 (7.14–7.32) | 4 | 7.26 (7.17–7.34) | 7.17 (7.08–7.24) | <0.001 |
| Arterial PaCO2, median (IQR), mmHg | 59 (46–76) | 4 | 56 (43–72) | 67 (54–81) | <0.001 |
| PaO2/FiO2 ratio, median (IQR) | 270 (188–355) | 72 | 286 (205-370) | 233 (152–332) | 0.005 |
| Severity on ICU admission | |||||
| SAPS II, median (IQR) | 36 (26–50) | 0 | 32 (24–45) | 44 (32–58) | <0.001 |
| SOFA score, median (IQR) | 4 (2–7) | 0 | 3 (2–6) | 7 (5–9) | <0.001 |
| Respiratory tract infection on ICU admission | |||||
| Bacteria, No. (%) | 87 (25) | 0 | 45 (20) | 42 (36) | 0.002 |
| Virus, No. (%) | 52 (15) | 0 | 24 (10) | 28 (24) | 0.001 |
| Bacteria and/or virus, No. (%) | 126 (36) | 0 | 64 (28) | 62 (53) | <0.001 |
The primary endpoint was ‘complicated course’, a composite criterion of death during the ICU stay and/or the need for invasive mechanical ventilation for ≥7 days.
CS, corticosteroids; FiO2, fraction of inspired O2; ICU, intensive care unit; IQR, interquartile range (first through third quartiles); No, number; PaCO2, partial pressure of CO2; PaO2, partial pressure of O2; SAPS II, Simplified Severity Index II; SOFA, Sequential Organ Failure Assessment.
Three-quarters of the population were treated with long-term inhaled corticosteroids, and 9% of patients had already received asthma biotherapy. Approximately one in five patients had already been intubated for an asthma exacerbation. The median time from symptom onset to ICU admission was 2 (0–4) days, and almost half of the patients received systemic corticosteroid therapy before hospital referral. A suspected triggering factor for exacerbation was identified in only 60% of patients, with respiratory tract infection being the predominant cause.
Acute respiratory failure was the primary reason for ICU admission; however, 42 (12%) patients were admitted following a cardiorespiratory arrest that occurred in the emergency room or during prehospital management. Complete clinical, biological and radiological findings on ICU admission are depicted in online supplemental table S1. Wheezing and cough were the most frequent respiratory symptoms, present in 76% and 35% of patients, respectively. Conversely, symptoms suggestive of respiratory tract infection such as fever, expectoration, rhinorrhea or myalgia were rarely reported. Only 32% of patients exhibited pulmonary radiological abnormalities. Hypercapnia (>45 mmHg) and blood acidosis (≤7.3) were present in 77% and 70% of patients, respectively. Blood eosinophil counts were low; however, this finding should be interpreted with caution, considering that approximately half of the patients received systemic corticosteroids prior to ICU referral. Details on positive microbiological results are depicted in online supplemental table S2. A respiratory tract infection was documented in 36% of patients. The predominant pathogens were Haemophilus influenzae, Streptococcus pneumoniae and Rhinovirus, which were identified in 26, 25 and 20 patients, respectively. Mixed bacterial–viral co-infection was documented in 13 patients.
Management, hospital course and outcomes
Ventilatory support and pharmacological treatments are presented in table 2. Of the 115 patients who received primary NIV, 26 experienced NIV failure. Finally, three-quarters of patients were intubated, for a median duration of IMV of 5 (3–12) days. Among them, 176 (68%) received a continuous infusion of curare for a mean duration of 3 (1–6) days, 112 (43%) and 19 (7%) were treated with ketamine and inhaled halogenated anaesthetics, respectively, and 16 (5%) required extracorporeal life support. Only three patients underwent tracheostomy. 41 patients received oxygen therapy with high-flow nasal cannula at some point during their stay. Hospital course and outcomes are depicted in table 3.20 63 (18%) patients experienced at least one episode of ICU-acquired pneumonia. Barotraumatic complications, including pneumothorax or pneumomediastinum, were rare, diagnosed in 9 (3%) patients, all of whom were receiving IMV. 26 (8%) patients died during their hospital stay. In patients receiving IMV, in-hospital mortality reached 10%. Among the 42 patients admitted to the ICU following a cardiorespiratory arrest, 14 (33%) died during their hospital stay. Of the 307 patients admitted to the ICU without a previous cardiorespiratory arrest, 4% died in hospital. Among survivors, three-quarters were discharged home, while the remainder were transferred to short- or medium-stay hospitals.
Table 2. In-ICU management of adult patients admitted for asthma exacerbation requiring ventilatory support.
| All patients | Missing data | Complicated course | P value | ||
|---|---|---|---|---|---|
| No | Yes | ||||
| n=349 | n=231 | n=118 | |||
| Pharmacological treatments* | |||||
| Systemic CS therapy, No. (%)† | 332 (95) | 1 | 219 (95) | 113 (97) | 0.59 |
| β2-adrenergic agonist aerosols | |||||
| No. over the first 24 hours, median (IQR) | 9 (6–12) | 15 | 8 (6–12) | 8 (6–13) | 0.27 |
| No. over the first 72 hours, median (IQR) | 20 (15–26) | 22 | 18 (13–25) | 22 (18–35) | 0.002 |
| Anticholinergic aerosols, No. (%) | 306 (89) | 6 | 204 (90) | 102 (88) | 0.59 |
| Magnesium sulfate, No. (%)‡ | 196 (57) | 2 | 122 (53) | 74 (63) | 0.11 |
| Antibiotics on ICU admission, No. (%) | 209 (60) | 0 | 112 (49) | 97 (82) | <0.001 |
| Ventilatory support | |||||
| Primary NIV, No. (%) | 115 (33) | 0 | 100 (43) | 15 (13) | <0.001 |
| Exclusive NIV, No. (%) | 89 (26) | 0 | 89 (39) | 0 (0) | <0.001 |
| NIV failure, No. (%) | 26 (7) | 0 | 11 (5) | 15 (13) | 0.010 |
| Invasive mechanical ventilation, No. (%) | 260 (75) | 0 | 142 (62) | 118 (100) | <0.001 |
| Length of IMV, median (IQR), days | 5 (3–12) | 0 | 3 (2–4) | 12 (9–19) | <0.001 |
| Extracorporeal life support | |||||
| ECMO, No. (%) | 9 (3) | 0 | 0 (0.0) | 9 (8) | <0.001 |
| Length of ECMO, median (IQR), days | 5 (4–6) | 0 | – | 5 (4–6) | – |
| ECCO2R, No. (%) | 7 (2) | 0 | 2 (1) | 5 (4) | 0.046 |
| Length of ECCO2R, median (IQR), days | 4 (3–11) | 0 | 3 (2–3) | 8 (4–14) | 0.24 |
The primary endpoint was ‘complicated course’, a composite criterion of the death during the ICU stay and/or the need for invasive mechanical ventilation for ≥7 days.
During the first 72 hours of the ICU stay (unless otherwise stated).
Systemic corticosteroid therapy was defined as >0.5 mg/kg/day of prednisone equivalent.
Intravenous or inhaled route.
CS, corticosteroids; ECCO2R, extracorporeal carbon dioxide removal; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; IMV, invasive mechanical ventilation; IQR, interquartile range (first through third quartiles); NIV, non-invasive ventilation.
Table 3. Hospital course and outcomes in adult patients admitted to the ICU for asthma exacerbation requiring ventilatory support.
| All patients | Missing data | Complicated course | P value | ||
|---|---|---|---|---|---|
| No | Yes | ||||
| n=349 | n=231 | n=118 | |||
| Organ failures and supports | |||||
| Acute respiratory distress syndrome, No. (%)* | 32 (9) | 0 | 2 (1) | 30 (25) | <0.001 |
| Vasopressors, No. (%)† | 148 (42) | 0 | 60 (26) | 88 (75) | <0.001 |
| Renal replacement therapy, No. (%)‡ | 13 (4) | 0 | 2 (1) | 11 (9) | <0.001 |
| Complications | |||||
| ICU-acquired pneumonia, No. (%) | 63 (18) | 0 | 4 (2) | 59 (50) | <0.001 |
| Supraventricular rhythm abnormality, No. (%) | 29 (8) | 0 | 17 (7) | 12 (10) | 0.41 |
| Pneumothorax, No. (%) | 4 (1) | 0 | 0 (0) | 4 (3) | 0.013 |
| Pneumomediastinum, No. (%) | 8 (2) | 0 | 3 (1) | 5 (4) | 0.13 |
| Outcomes | |||||
| Length of ICU stay, median (IQR), days | 5 (3–11) | 1 | 4 (3–5) | 15 (11–21) | <0.001 |
| Length of hospital stay, median (IQR), days | 12 (7–19) | 2 | 9 (6–14) | 19 (13–33) | <0.001 |
| In-hospital mortality, No. (%) | 26 (8) | 0 | 1 (0.5) | 25 (21) | <0.001 |
The primary endpoint was ‘complicated course’, a composite criterion of death during the ICU stay and/or the need for invasive mechanical ventilation for ≥7 days.
According to the Berlin definition.
Adrenaline or norepinephrine.
Outside patients with chronic dialysis.
ICU, intensive care unit.
Analysis according to the primary endpoint
The primary endpoint could be evaluated in all patients. A complicated course was observed in 118 (34%) patients. Compared with their counterparts, they were more often admitted to the ICU due to cardiorespiratory arrest (20% vs 8%, p=0003) and had a higher SOFA score and SAPS II on ICU admission. They also experienced more frequent agitation or coma, lower pH, higher PaCO2 and lower PaO2/FiO2. Interestingly, they exhibited a higher prevalence of bilateral radiological opacities (15% vs 5%, p=0004) and were more frequently documented with a respiratory tract infection on ICU admission. Detailed ventilatory parameters during the first days of IMV are shown in figure 2. There were no between-group differences in respiratory rate, tidal volume and plateau pressure at the initiation of IMV (figure 2a). Peak pressure was slightly higher in patients with a complicated course (48 (38–66) vs 45 (35–56), p=0.03). Focusing on the clinical course during the first 24 and 72 hours of IMV, a marked decrease in peak and plateau pressure (figure 2b) and a progressive correction of hypercapnia and respiratory acidosis (online supplemental table S3) were observed in patients without a complicated course. In contrast, such mechanical and biological parameters remained remarkably stable in patients with a complicated course. The multivariable analysis revealed that the SOFA score, pH≤7.3 and a documented respiratory tract infection on ICU admission are independent risk factors for a complicated course (table 4).
Figure 2. Values and course of ventilatory parameters in the subpopulation requiring invasive mechanical ventilation. (A) Plateau and peak pressures, respiratory rate and tidal volume on Day0 (start of invasive mechanical ventilation), Day+1 and Day+3. (B) Relative change in peak and plateau pressures between Day0 and Day+1 and between Day0 and Day+3. (A) The number of patients with a complicated course using volume-controlled ventilation was 106 at Day 0, 98 at Day+1 and 82 at Day+3. Between Day0 and Day+1, 1 patient died, 6 were now using pressure-targeted modes and 1 had missing information on the ventilatory mode. Between Day+1 and Day+3, 5 patients died, 10 were now using pressure-targeted modes and 1 had missing information on the ventilatory mode. The number of patients without a complicated course using volume-controlled ventilation was 123 at Day0, 72 at Day+1 and 13 at Day+3. Between Day0 and Day+1, 15 patients were extubated, 35 were now using pressure-targeted modes and 1 had missing information on the ventilatory mode. Between Day+1 and Day+3, 37 patients were extubated, and 22 were now using pressure-targeted modes. (B) The number of patients with a complicated course using volume-controlled ventilation was 98 for the change from Day0 to Day+1 and 82 for the change from Day0 to Day+3. Between Day+1 and Day+3, 5 patients died, 10 were now using pressure-targeted modes and 1 had missing information on the ventilatory mode. The number of patients without a complicated course using volume-controlled ventilation was 72 for the change from Day0 to Day+1 and 13 for the change from Day0 to Day+3. Between Day+1 and Day+3, 37 patients were extubated, and 22 were now using pressure-targeted modes. Relative changes in peak and plateau pressures mean the percentage of decrease from Day0 to Day+1 and from Day0 to Day+3 are labelled ‘Day+1 Day0 and ‘Day+3 Day0’ and are calculated as follows (for Day+1): (ValueDay+1 – ValueDay0)/ValueDay0. Unit of the vertical axis is %. For both panels, data are shown in box plots. The box indicates the interquartile range (Q1–Q3), and the solid line indicates the median. The upper and lower points indicate outliers. For each parameter and time point, groups were compared using the non-parametric Mann–Whitney U test. P-values are indicated when <0.05.
Table 4. Multivariable analysis of risk factors for a complicated course in adult patients admitted to the ICU for asthma exacerbation requiring ventilatory support.
| Multivariable analysis | ||
|---|---|---|
| OR (95% CI) | P value | |
| History of asthma | ||
| Long-term inhaled CS | 0.84 (0.46 to 1.54) | 0.58 |
| Previous intubation | 0.94 (0.47 to 1.89) | 0.86 |
| Asthma exacerbation | ||
| Therapy nonadherence/shift as a suspected triggering factor | 0.71 (0.29 to 1.71) | 0.44 |
| Cardiorespiratory arrest as reason for ICU admission | 1.16 (0.51 to 2.65) | 0.72 |
| Severity on ICU admission | ||
| SOFA score* | 1.33 (1.19 to 1.48) | <0.001 |
| Characteristics on ICU admission | ||
| Agitation or coma at ICU admission | 0.61 (0.33 to 1.12) | 0.11 |
| Arterial pH≤7.3 | 2.46 (1.01 to 6.02) | 0.049 |
| Arterial PaCO2>45 mmHg | 1.10 (0.44 to 2.72) | 0.84 |
| Urea† | 1.00 (0.92 to 1.10) | 0.97 |
| Normal chest X-ray | 0.70 (0.39 to 1.27) | 0.24 |
| Respiratory tract infection on ICU admission | ||
| Bacteria and/or virus | 2.35 (1.37 to 4.05) | 0.002 |
The primary endpoint was ‘complicated course’, a composite criterion of death during the ICU stay and/or the need for invasive mechanical ventilation for ≥7 days.
For an increase of 1 point in the SOFA score.
For an increase of 1 mmol/L of urea.
CS, corticosteroids; ICU, intensive care unit; PaCO2, partial pressure of CO2; SOFA SOFA, Sequential Organ Failure Assessment.
Temporal trends over the period
Online supplemental table S4 shows the temporal trends of in-ICU management and outcomes via the comparison of the first and second halves of the 10-year period. The use of primary NIV increased by 40% between the first and second halves of the study period, accompanied by a rise in NIV failure rates from 12% to 27% among patients receiving primary NIV. However, this increase was not associated with higher rates of IMV use, in-hospital mortality or complicated clinical courses.
Comparison to previous studies
Online supplemental table S5 shows our primary findings alongside those from previous studies on asthma exacerbation requiring ventilatory support in adults.
Discussion
This study examined the clinical spectrum and outcomes of asthma exacerbation requiring ventilatory support in 349 adults admitted to 21 ICU departments over 10 years. This is a rare population, representing 1.7 patients per year per centre. Respiratory tract infection was the predominant suspected triggering factor, documented in more than a third of patients. NIV was used as primary ventilatory support in 115 (33%) patients, but it failed in nearly a quarter of them. Barotraumatic complications occurred in only 9 (3%) patients, and 3 (1%) underwent tracheotomy. The primary composite endpoint of complicated course was observed in 34% of patients. This endpoint was independently associated with the SOFA score, blood acidosis and microbiologically documented respiratory tract infection at ICU admission.
In this study, clinicians used NIV as initial ventilatory support for nearly one-third of patients, which aligns with recent reports from North America.4 14 The high median age and prevalence of obesity in this population may have encouraged this approach, as clinicians may have been more inclined to suspect fixed obstructive syndrome mimicking chronic obstructive pulmonary disease (COPD) or overlap diseases such as obesity hypoventilation syndrome. No guidelines recommend this initial ventilatory management approach,2 13 and only one study has reported improved clinical outcomes in patients admitted in the ICU for asthma exacerbation receiving initial NIV.4 The NIV failure rate of 23% is consistent with the 5%–22% range reported in previous studies.4 14 However, the limited number of patients precluded assessment of the clinical implications of this event. Blood gas analysis in patients who were intubated revealed persistent hypercapnia throughout the first 24 hours of mechanical ventilation. These findings suggest that clinicians were likely adopting a permissive hypercapnia strategy, in which a certain degree of hypoventilation was tolerated.21–24 Regarding respiratory mechanics, median plateau pressures remained around 20 cmH2O, while peak pressures remained under 50 cmH2O throughout the entire period of IMV. This may have contributed to the low incidence of barotraumatic events, in the lower range of the 3–6% reported in previous studies.16 22 In-hospital mortality was 8%, and reached 10% in patients requiring IMV, in the 2%–15% range reported in cohorts of ICU patients with asthma exacerbations.3 4 6–8 15–17 Nearly one in eight patients was admitted following a cardiorespiratory arrest, a life-threatening event that may have significantly impacted the clinical outcome. For comparison, Binachon et al reported a 15% mortality rate in the overall population of mechanically ventilated asthmatics, but only 4% when those with a pre-hospital cardiorespiratory arrest were excluded.15
In this study, death and/or IMV for ≥7 days occurred in one-third of patients. This composite endpoint has not previously been applied in cohorts of patients with asthma requiring ventilation, making comparisons with previous data impossible. Notably, intrinsic factors such as age, smoking, previous intubation or immunocompromised status were not associated with the primary endpoint. This suggests that demographic characteristics and underlying diseases have a limited impact on clinical outcomes in this study. In addition to the SOFA score and blood acidosis at ICU admission, multivariable analysis identified microbiologically documented respiratory tract infection as independently associated with the primary endpoint. This affected over one-third of patients, whereas previous studies reported pneumonia in only 9 to 25% of patients with asthma requiring mechanical ventilation.4 15–17 The implementation of viral multiplex PCR into routine practice during the study period may have contributed to the high rate of viral detection. The substantial bacterial documentation may be partially attributed to the frequent collection of lower respiratory tract specimens, driven by the high prevalence of suspected infectious triggering factors at ICU admission. Regrettably, data on microbiological investigations were not systematically recorded, precluding definitive conclusions on this aspect. Consistent with our findings, Stefan et al identified pneumonia as a risk factor for NIV failure.4 This observation prompts several hypotheses regarding the relationship between respiratory tract infection and a complicated course in the ICU. The first hypothesis pertains to the systemic level: bacterial infections can lead to sepsis, a major contributor to multiple organ failure, potentially delaying weaning from mechanical ventilation and increasing mortality. The second hypothesis concerns the locoregional level. Respiratory tract infections may cause severe and prolonged airway obstruction unresponsive to bronchodilators and anti-inflammatory therapies. This contrasts with exacerbations triggered by allergen exposure or withdrawal of anti-asthmatic treatments. For instance, patients with asthma experiencing virus-induced (especially rhinovirus) exacerbations exhibit heightened inflammation, mucus production and airway hyperreactivity compared with those with allergen-induced asthma.25–28 Furthermore, their lung function recovery may be enhanced after prolonged steroid therapy, suggesting a mitigated response to corticosteroids.29 This phenotypic difference is supported by differences in biomarkers, with two previous studies showing an increase in sputum total cells and sputum neutrophils in virus-induced exacerbations compared with non-infectious exacerbations.30 31 Similarly, a study showed that patients with COPD with virus-induced exacerbations can experience prolonged symptom resolution.32 Our description of changes in respiratory mechanics during the initial phase of mechanical ventilation aligns with this second hypothesis. Among patients with a complicated course, over half of whom had a documented respiratory tract infection, we observed a remarkably stable airway pressure over the first 24–72 hours, suggesting prolonged and severe airway obstruction, which may have delayed weaning from mechanical ventilation. These findings may have practical implications for clinicians. Early identification of respiratory pathogens may contribute to risk stratification and help clinicians anticipate a potentially complicated course, including refractory airway obstruction. Therefore, respiratory pathogens should be actively investigated through comprehensive microbiological testing, including respiratory sampling. In patients with documented bacterial infection, early appropriate antimicrobial therapy should be considered. Whether early optimisation of asthma medical treatment, such as biotherapies, strategies or advanced supportive therapies, such as extracorporeal circulation, could improve outcomes in this high-risk subgroup deserves further investigation.
This study has several limitations. First, although it was a multicentre study, it was conducted exclusively in France. Caution should be exercised when extrapolating the results to other countries, as numerous parameters may vary, including long-term asthma treatment, exacerbation management and the epidemiology of respiratory pathogens. Second, the study was retrospective, which limited our control over microbiological investigations, and only positive results were reported, with no distinction made between respiratory tract infection and colonisation. Furthermore, the retrospective design precluded the collection of certain data that are infrequently documented in medical records, such as vaccine history or symptoms before hospital referral. Although all included patients had a previously documented diagnosis of asthma according to GINA2 criteria and were admitted for an asthma exacerbation, we cannot exclude that some patients may have had undiagnosed COPD or asthma–COPD overlap, as a systematic assessment to rule out these conditions was not performed. Additionally, we did not collect post-exacerbation clinical data, such as pulmonary function tests, which could have provided valuable insights into the long-term consequences of the exacerbation in survivors. Moreover, even if this is one of the largest cohorts of patients with asthma requiring ventilation, the sample size should be taken into consideration when interpreting the results. Third, we selected a composite primary endpoint determined a priori. Given the predictable low hospital and ICU mortality, we did not choose death as the primary endpoint, as its low frequency would likely obscure differences between groups. Fourth, the retrospective design of the study precluded us from deciphering reasons for choosing NIV as primary support in a third of patients. For the same reason, NIV failure was defined without considering the dose and/or duration of NIV. Furthermore, criteria for intubation in patients primarily managed with NIV may vary between centres and over the study period. Finally, any viral detection via PCR on a respiratory tract specimen (including nasopharyngeal swab) was classified as a respiratory virus-associated infection. This approach may be questioned, as respiratory viruses can sometimes be detected in asymptomatic adults.33 34
Conclusion
Asthma exacerbation requiring ventilatory support was associated with a substantial burden in adults, with an in-hospital mortality of 8%. Respiratory tract infection was the predominant suspected triggering factor, and respiratory tract infection was documented in 36% of patients. These infections were independently associated with a prolonged course and/or fatal outcome in the ICU. These findings underscore the need for further research in this population, particularly to determine the most effective strategies for diagnosing and treating respiratory infections.
Supplementary material
Footnotes
Funding: This study has been supported by the European Research Network on Artificial Ventilation (REVA Network).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by the Institutional Review Board of the French Society of Intensive Care (Reference 19-35, 2019) and by the French Committee for Research and Evaluations in Health (CEREES) (TPS603984, 2019), in accordance with French regulations. All subjects (or their relatives) received written information about the study and provided their informed consent to participate before enrolment, in compliance with French regulations (Law no. 2012-300; 5 March 2012).
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.GBD 2015 Chronic Respiratory Disease Collaborators Global, regional, and national deaths, prevalence, disability-adjusted life years, and years lived with disability for chronic obstructive pulmonary disease and asthma, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Respir Med. 2015;5:691–706. doi: 10.1016/S2213-2600(17)30293-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Global strategy for asthma management and prevention: 2023 update. [11-Dec-2023]. https://ginasthma.org/wp-content/uploads/2023/07/GINA-2023-Full-report-23_07_06-WMS.pdf Available. Accessed.
- 3.Nanchal R, Kumar G, Majumdar T, et al. Utilization of mechanical ventilation for asthma exacerbations: analysis of a national database. Respir Care. 2014;59:644–53. doi: 10.4187/respcare.02505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Stefan MS, Nathanson BH, Lagu T, et al. Outcomes of Noninvasive and Invasive Ventilation in Patients Hospitalized with Asthma Exacerbation. Ann Am Thorac Soc. 2016;13:1096–104. doi: 10.1513/AnnalsATS.201510-701OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.McFadden ER. Acute severe asthma. Am J Respir Crit Care Med. 2003;168:740–59. doi: 10.1164/rccm.200208-902SO. [DOI] [PubMed] [Google Scholar]
- 6.Pendergraft TB, Stanford RH, Beasley R, et al. Rates and characteristics of intensive care unit admissions and intubations among asthma-related hospitalizations. Ann Allergy Asthma Immunol. 2004;93:29–35. doi: 10.1016/S1081-1206(10)61444-5. [DOI] [PubMed] [Google Scholar]
- 7.Gupta D, Keogh B, Chung KF, et al. Characteristics and outcome for admissions to adult, general critical care units with acute severe asthma: a secondary analysis of the ICNARC Case Mix Programme Database. Crit Care . 2004;8:10. doi: 10.1186/cc2835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stow PJ, Pilcher D, Wilson J, et al. Improved outcomes from acute severe asthma in Australian intensive care units (1996 2003) Thorax. 2007;62:842–7. doi: 10.1136/thx.2006.075317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Afessa B, Morales I, Cury JD. Clinical course and outcome of patients admitted to an ICU for status asthmaticus. Chest. 2001;120:1616–21. doi: 10.1378/chest.120.5.1616. [DOI] [PubMed] [Google Scholar]
- 10.Younan R, Augy JL, Hermann B, et al. Severe asthma exacerbation: Changes in patient characteristics, management, and outcomes from 1997 to 2016 in 40 ICUs in the greater Paris area. J Intensive Med . 2024;4:209–15. doi: 10.1016/j.jointm.2023.08.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ganesh A, Shenoy S, Doshi V, et al. Use of noninvasive ventilation in adult patients with acute asthma exacerbation. Am J Ther. 2015;22:431–4. doi: 10.1097/MJT.0000000000000184. [DOI] [PubMed] [Google Scholar]
- 12.Lim WJ, Mohammed Akram R, Carson KV, et al. Non-invasive positive pressure ventilation for treatment of respiratory failure due to severe acute exacerbations of asthma. Cochrane Database Syst Rev. 2012;12:CD004360. doi: 10.1002/14651858.CD004360.pub4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Le Conte P, Terzi N, Mortamet G, et al. Management of severe asthma exacerbation: guidelines from the Société Française de Médecine d’Urgence, the Société de Réanimation de Langue Française and the French Group for Pediatric Intensive Care and Emergencies. Ann Intensive Care. 2019;9:115. doi: 10.1186/s13613-019-0584-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Althoff MD, Holguin F, Yang F, et al. Noninvasive Ventilation Use in Critically Ill Patients with Acute Asthma Exacerbations. Am J Respir Crit Care Med. 2020;202:1520–30. doi: 10.1164/rccm.201910-2021OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Binachon A, Grateau A, Allou N, et al. Acute severe asthma requiring invasive mechanical ventilation in the era of modern resuscitation techniques: A 10-year bicentric retrospective study. PLoS One. 2020;15:e0240063. doi: 10.1371/journal.pone.0240063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kao CC, Jain S, Guntupalli KK, et al. Mechanical ventilation for asthma: a 10-year experience. J Asthma. 2008;45:552–6. doi: 10.1080/02770900801999090. [DOI] [PubMed] [Google Scholar]
- 17.Zimmerman JL, Dellinger RP, Shah AN, et al. Endotracheal intubation and mechanical ventilation in severe asthma. Crit Care Med. 1993;21:1727–30. doi: 10.1097/00003246-199311000-00023. [DOI] [PubMed] [Google Scholar]
- 18.Le Gall J-R. A New Simplified Acute Physiology Score (SAPS II) Based on a European/North American Multicenter Study. JAMA. 1993;270:2957. doi: 10.1001/jama.1993.03510240069035. [DOI] [PubMed] [Google Scholar]
- 19.Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22:707–10. doi: 10.1007/BF01709751. [DOI] [PubMed] [Google Scholar]
- 20.Force ADT, Ranieri VM, Rubenfeld GD, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307:2526–33. doi: 10.1001/jama.2012.5669. [DOI] [PubMed] [Google Scholar]
- 21.Demoule A, Brochard L, Dres M, et al. How to ventilate obstructive and asthmatic patients. Intensive Care Med. 2020;46:2436–49. doi: 10.1007/s00134-020-06291-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Leatherman J. Mechanical ventilation for severe asthma. Chest. 2015;147:1671–80. doi: 10.1378/chest.14-1733. [DOI] [PubMed] [Google Scholar]
- 23.Oddo M, Feihl F, Schaller M-D, et al. Management of mechanical ventilation in acute severe asthma: practical aspects. Intensive Care Med. 2006;32:501–10. doi: 10.1007/s00134-005-0045-x. [DOI] [PubMed] [Google Scholar]
- 24.Darioli R, Perret C. Mechanical controlled hypoventilation in status asthmaticus. Am Rev Respir Dis. 1984;129:385–7. doi: 10.1164/arrd.1984.129.3.385. [DOI] [PubMed] [Google Scholar]
- 25.Miller EK, Hernandez JZ, Wimmenauer V, et al. A mechanistic role for type III IFN-λ1 in asthma exacerbations mediated by human rhinoviruses. Am J Respir Crit Care Med. 2012;185:508–16. doi: 10.1164/rccm.201108-1462OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Khoo S-K, Read J, Franks K, et al. Upper Airway Cell Transcriptomics Identify a Major New Immunological Phenotype with Strong Clinical Correlates in Young Children with Acute Wheezing. J Immunol. 2019;202:1845–58. doi: 10.4049/jimmunol.1800178. [DOI] [PubMed] [Google Scholar]
- 27.Turi KN, Ryckman KK, Shankar J. Metabolites in newborns are correlated with immune-response to acute respiratory syncytial virus infection and are associated with subsequent recurrent wheeze. Am J Respir Crit Care Med. 2018;197:A7366. doi: 10.1164/ajrccm-conference.2018.197.1_MeetingAbstracts.A7366. [DOI] [Google Scholar]
- 28.Jackson DJ, Johnston SL. The role of viruses in acute exacerbations of asthma. J Allergy Clin Immunol. 2010;125:1178–87. doi: 10.1016/j.jaci.2010.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kan-o K, Washio Y, Oki T, et al. Effects of treatment with corticosteroids on human rhinovirus-induced asthma exacerbations in pediatric inpatients: a prospective observational study. BMC Pulm Med. 2023;23:487. doi: 10.1186/s12890-023-02798-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wark PAB, Johnston SL, Moric I, et al. Neutrophil degranulation and cell lysis is associated with clinical severity in virus-induced asthma. Eur Respir J. 2002;19:68–75. doi: 10.1183/09031936.02.00226302. [DOI] [PubMed] [Google Scholar]
- 31.Bjerregaard A, Laing IA, Poulsen N, et al. Characteristics associated with clinical severity and inflammatory phenotype of naturally occurring virus-induced exacerbations of asthma in adults. Respir Med. 2017;123:34–41. doi: 10.1016/j.rmed.2016.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Seemungal T, Harper-Owen R, Bhowmik A, et al. Respiratory viruses, symptoms, and inflammatory markers in acute exacerbations and stable chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2001;164:1618–23. doi: 10.1164/ajrccm.164.9.2105011. [DOI] [PubMed] [Google Scholar]
- 33.Byington CL, Ampofo K, Stockmann C, et al. Community Surveillance of Respiratory Viruses Among Families in the Utah Better Identification of Germs-Longitudinal Viral Epidemiology (BIG-LoVE) Study. Clin Infect Dis. 2015;61:1217–24. doi: 10.1093/cid/civ486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Lieberman D, Shimoni A, Shemer-Avni Y, et al. Respiratory viruses in adults with community-acquired pneumonia. Chest. 2010;138:811–6. doi: 10.1378/chest.09-2717. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.