Inhaled anticholinergics and short‐acting beta2‐agonists versus short‐acting beta2‐agonists alone for children with acute asthma in hospital
Abstract
Background
Inhaled anticholinergics given in addition to β2‐agonists are effective in reducing hospital admissions in children presenting to the emergency department with a moderate to severe asthma exacerbation. It seems logical to assume a similar beneficial effect in children hospitalised for an acute asthma exacerbation.
Objectives
To assess the efficacy and safety of anticholinergics added to β2‐agonists as inhaled or nebulised therapy in children hospitalised for an acute asthma exacerbation. To investigate the characteristics of patients or therapy, if any, that would influence the magnitude of response attributable to the addition of anticholinergics.
Search methods
We identified trials from the Cochrane Airways Group Specialised Register of trials (CAGR), which is derived through systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED and PsycINFO and through handsearching of respiratory journals and meeting abstracts. The search is current to November 2013.
Selection criteria
Randomised trials comparing the combination of inhaled or nebulised anticholinergics and short‐acting β2‐agonists versus short‐acting β2‐agonists alone in children one to 18 years of age hospitalised for an acute asthma exacerbation were eligible.
Data collection and analysis
Two review authors independently assessed the methodological quality of trials and extracted data; disagreement was resolved by consensus or with the input of a third review author, when needed. Primary outcomes were duration of hospital stay and serious adverse events. Secondary outcomes included admission and duration of stay in the intensive care unit (ICU), ventilation assistance, time to short‐acting β2‐agonists spaced at four hours or longer, supplemental asthma therapy, duration of supplemental oxygen, change from baseline in asthma severity, relapse after discharge, adverse health effects and withdrawals.
Main results
Seven randomised trials were included, four of which reported usable data on 472 children with asthma one to 18 years of age who were admitted to paediatric wards. No trials included patients admitted to the ICU. The anticholinergic used, ipratropium bromide 250 μg, was given every one to eight hours over a period from four hours to the entire length of the hospital stay. Two of four trials (50%) contributing data were deemed of high methodological quality. The addition of anticholinergics to β2‐agonists showed no evidence of effect on the duration of hospital admission (mean difference (MD) ‐0.28 hours, 95% confidence interval (CI) ‐5.07 to 4.52, 3 studies, 327 participants, moderate quality evidence) and no serious or non‐serious adverse events were reported in any included trials. As a result of the similarity of trials, we could not explore the influence of age, admission site, intensity of anticholinergic treatment and co‐interventions on primary outcomes. No statistically significant group difference was noted in other secondary outcomes, including the need for supplemental asthma therapy, time to short‐acting β2‐agonists spaced at four hours or longer, asthma clinical scores, lung function and overall withdrawals for any reason.
Authors' conclusions
In children hospitalised for an acute asthma exacerbation, no evidence of benefit for length of hospital stay and other markers of response to therapy was noted when nebulised anticholinergics were added to short‐acting β2‐agonists. No adverse health effects were reported, yet the small number of trials combined with inadequate reporting prevent firm reassurance regarding the safety of anticholinergics. In the absence of trials conducted in ICUs, no conclusion can be drawn regarding children with impending respiratory failure. These findings support current national and international recommendations indicating that healthcare practitioners should refrain from using anticholinergics in children hospitalised for acute asthma.
PICOs
Plain language summary
Are inhaled anticholinergics added to β2‐agonists beneficial in children hospitalised with acute asthma?
Background: Anticholinergics (e.g. ipratropium bromide, atropine sulfate) are inhaled drugs. They relax the airway muscles and decrease secretions. Anticholinergics are sometimes used in addition to beta2‐agonists (such as salbutamol and terbutaline), which are potent drugs given to relax smooth muscles in the airways in children with acute asthma. We do not know whether the addition of inhaled anticholinergics to beta2‐agonists is beneficial for children hospitalised with acute asthma.
Review question: We wished to examine the efficacy and safety of inhaled or nebulised (mist inhaled into the lungs) anticholinergics added to beta2‐agonists compared with beta2‐agonists alone in children one to 18 years of age hospitalised for an acute asthma exacerbation.
Study characteristics: In reviewing evidence available until November 2013, we found seven eligible studies of children hospitalised with acute asthma; four of these studies (472 children one to 18 years of age) contributed data to the review. Four studies compared the combination of anticholinergics (ipratropium bromide) and beta2‐agonists versus the same dose of beta2‐agonists alone. Included studies enrolled both girls and boys, with a gender ratio ranging from 59% to 73% males.
Results: No additional benefit was noted by adding anticholinergics to β2‐agonists in terms of duration of hospital stay in patients compared to those who received beta2‐agonists alone. Two of four trials (50%) contributing data were deemed of high methodological quality. No trial reported information on serious adverse events. No statistically significant group difference was noted in other markers of response to therapy, that is, the need for supplemental asthma therapy, time to short‐acting beta2‐agonists spaced at four hours or longer, asthma clinical scores, lung function and overall withdrawals for any reason.
Conclusion: No apparent benefit is derived from adding anticholinergics to beta2‐agonists in children hospitalised for an acute asthma exacerbation, that is, beyond initial treatment in the emergency department. No adverse health effects were reported, yet the small number of trials combined with inadequate reporting prevents firm reassurance regarding the safety of anticholinergics. In the absence of trials conducted in the intensive care unit (ICU), no conclusion can be drawn regarding children with very severe exacerbations who are admitted to the ICU. Our findings support the ongoing recommendations provided by national and international guidelines.
Quality of the results: This review is based on a small number of identified trials conducted in children with acute asthma. All trials contributing to the primary outcome are of high methodological quality, but they are few. As the addition of new trials may change the conclusion, the quality of evidence was downgraded from high to moderate. Additional and larger trials are needed.
Authors' conclusions
Summary of findings
| Combination of AC + beta2‐agonists versus beta2‐agonists alone | ||||||
| Patient or population: children hospitalised with an acute asthma exacerbation | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Beta2‐agonistsalone | Combination of AC + beta2‐agonists | |||||
| Duration of the hospital stay (hours) | The mean duration of the hospital stay (hours) in the control groups was | The mean reduction in duration of hospital stay (hours) was 0.28 hours in the intervention groups Mean 43.8 (39 to 49) hours | ‐0.28 hours | 327 | ⊕⊕⊕⊝ | |
| Time to regular dose of short‐acting beta2‐agonists (hours)2 Follow‐up: 4 hours or longer | The mean time to regular short‐acting beta2‐agonists in the control groups was 33 hours | The mean reduction in time to regular short‐acting beta2‐agonists in the intervention groups was 2.17 hours Mean 31 (26 to 36) hours | ‐2.17 hours | 269 | ⊕⊕⊕⊝ | |
| Asthma clinical scores Follow‐up: 8 to 36 hours after initial treatment | See comment | See comment | 0.02 SMD (‐0.34 to 0.38) | 117 | ⊕⊕⊕⊝ | Scores were measured on different scales. |
| Admission to the intensive care unit | See comment | See comment | Not estimable | 210 | ⊕⊕⊕⊝ | Single trial reported admission to the intensive care unit. No events were reported |
| Overall withdrawals | 12 per 100 | 7 per 100 | OR 0.59 | 294 | ⊕⊕⊕⊝ | |
| Adverse health events | See comment | See comment | Not estimable | 290 | ⊕⊕⊕⊝ | No adverse events were reported in either trial |
| *The basis for the assumed risk is the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence. | ||||||
| 1Downgraded because of wide confidence intervals. 2This is the time at which emergency dosing with the trial intervention was replaced with regular 4‐hourly treatment with short‐acting beta2‐agonists. 3No events were reported in either arm of the studies. | ||||||
Background
Description of the condition
Asthma is a chronic inflammatory disorder of the airways that is associated with airway hyperresponsiveness (Ortiz‐Alvarez 2012). Globally the prevalence of severe asthma symptoms, defined as four or more attacks of wheeze or one or more night per week of sleep disturbance from wheeze or wheeze affecting speech in the past 12 months ranged from 0% to 20% in children (Asher 2010; Lai 2009). Asthma exacerbation is defined as an acute or subacute worsening of symptoms and lung function from the individual's usual status or, in some cases, the initial presentation of asthma (GINA 2014). One of the main goals in the management of acute asthma exacerbations in children is to achieve a rapid reversal of airflow obstruction (NAEPP 2011). This is achieved by using inhaled or nebulised short‐acting β2‐agonists (Camargo 2003; Karpel 1997), which are the most effective bronchodilators because of their rapid onset of action and the magnitude of bronchodilation that they achieve (Sears 1992; Svedmyr 1985). Systemic corticosteroids should be added early in the course of treatment for patients who have moderate or severe exacerbations or for children who fail to respond promptly and completely to bronchodilators (Rechelefsky 2003; Rowe 2004).
Description of the intervention
Anticholinergic agents, such as ipratropium bromide and atropine sulfate, have a slower onset of action and a weaker bronchodilating effect but may specifically relieve cholinergic bronchomotor tone and decrease mucosal edema and secretions (Chapman 1996; Gross 1988; Silverman 1990). Thus, with their slower onset but prolonged duration of action, anticholinergics (AC) can work as complementary therapy to β2‐agonists, thereby enhancing bronchodilation.
Several trials have explored the role of ipratropium bromide in the emergency department setting. One systematic review of randomised controlled trials (RCTs) concluded that multiple doses of inhaled anticholinergic agents added to β2‐agonist therapy in the initial treatment of children with acute asthma exacerbations were safe and efficacious, with most of the effect observed in those with severe asthma exacerbations and no apparent benefit noted in children presenting with mild to moderate asthma exacerbations (Plotnick 2000). An updated Cochrane review suggests that the benefit extends to children with moderate and severe exacerbations (Griffiths 2013). Moreover, multiple doses of inhaled anticholinergic agents added to β2‐agonists have been shown to be cost‐effective (Lord 1999).
How the intervention might work
To our knowledge, two trials have assessed the efficacy of ipratropium bromide added to β2‐agonists after the emergency department treatment period, that is, in children hospitalised for an acute asthma exacerbation (Craven 2001; Goggin 2001). Both trials independently demonstrated no benefit conferred by the addition of anticholinergics. Consequently, national and international asthma guidelines currently recommend that inhaled anticholinergics should not be used in hospitalised children with acute asthma (GINA 2014; NAEPP 2011). Yet, in several institutions, anticholinergics are used for a specified period of time after children are admitted, particularly, but not only, those admitted to the intensive care unit (ICU).
Why it is important to do this review
The difference in practice between institutions, termed practice variation, underlines a gap between recommended and observed care, or a care gap. In the absence of an identified systematic review, we believe that a Cochrane review would clarify the evidence accumulated to date regarding the role of anticholinergics in the treatment of children hospitalised for an acute asthma exacerbation.
Objectives
To assess the efficacy and safety of anticholinergics added to β2‐agonists as inhaled or nebulised therapy in children hospitalised for an acute asthma exacerbation. To investigate the characteristics of patients or therapy, if any, that would influence the magnitude of response attributable to the addition of anticholinergics.
Methods
Criteria for considering studies for this review
Types of studies
We included all RCTs.
Types of participants
We included children one to 18 years of age who were hospitalised for an acute asthma exacerbation.
Types of interventions
Intervention: nebulised or inhaled anticholinergics with short‐acting β2‐agonists.
Comparison: nebulised or inhaled short‐acting β2‐agonists alone.
Co‐interventions: Systemic corticosteroids were anticipated and permitted, provided they were similar in the two groups.
Types of outcome measures
Primary outcomes
-
Duration of hospital stay.
-
Serious adverse events.
Secondary outcomes
-
Duration of stay in the ICU (for those admitted to the ICU).
-
Admission to the ICU (for those admitted on the wards).
-
Ventilation assistance.
-
Need for supplemental asthma therapy (e.g. aminophylline).
-
Duration of need for supplemental oxygen.
-
Time to short‐acting β2‐agonists spaced at four hours or longer.
-
Change from baseline in asthma severity measured as lung function, symptoms or clinical scores.
-
Relapse within 72 hours of discharge from hospital.
-
Adverse health events.
-
Withdrawals, namely, overall withdrawals, withdrawals due to poor control of symptoms or deterioration and withdrawals due to adverse effects.
Search methods for identification of studies
Electronic searches
We identified trials from the Cochrane Airways Group Specialised Register of trials (CAGR), which is derived through systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED and PsycINFO, and through handsearching of respiratory journals and meeting abstracts (see Appendix 1 for further details). We searched all records in the CAGR coded as "asthma," using the strategy provided in Appendix 2. The literature search covered all years from inception until November 2013.
No restriction on type or language of publication was applied.
Searching other resources
We checked the reference lists of all identified RCTs to identify potentially relevant citations. We checked the websites of international headquarters of pharmaceutical companies producing anticholinergics for reports of relevant completed trials. We also explored the ClinicalTrials.gov website for relevant clinical trials.
Data collection and analysis
Two review authors (KV and BFC) independently extracted data, and disagreement was resolved by consensus or through the input of a third review author (FMD). When necessary, we contacted authors of included studies to request missing data.
Selection of studies
One review author (KV) independently reviewed each abstract and annotated each as (1) RCT; (2) clearly not an RCT; or (3) unclear. The full‐text publications of citations identified as included RCTs or unclear were reviewed by two review authors independently.
Data extraction and management
Two review authors (KV and BFC) independently extracted data, and disagreement was resolved by consensus or with the input of a third review author (FMD), when needed.
Assessment of risk of bias in included studies
We assessed the methodological quality of included trials by using the 'Risk of bias' tool of The Cochrane Collaboration, which is based on:
-
random sequence generation;
-
allocation concealment;
-
blinding of participants and personnel;
-
blinding of outcome assessment;
-
incomplete outcome data;
-
selective outcome reporting; and
-
other bias.
Quality assessment was performed independently by two review authors. We resolved disagreement by consensus or with the input of a third review author. The study was deemed to have high methodological quality if reported randomisation procedures and blinding were adequate and there was a low and balanced group attrition, supporting a low risk of bias.
Measures of treatment effect
We calculated treatment effects for dichotomous variables as risk ratio (RR) or risk difference (RD), or both, with 95% confidence intervals (CIs). We assumed equivalence if the RR estimate and its 95% CI fell between 0.9 and 1.1. For continuous outcomes, such as lung function tests, we calculated pooled statistics as mean difference (MD) or standardised mean difference (SMD) with 95% CI.
Unit of analysis issues
The unit of analysis was the participant. If a trial had more than one intervention or control group, we considered additional comparisons, when appropriate.
Dealing with missing data
When possible, we contacted investigators or study sponsors to obtain missing numerical outcome data or data in the required format to allow aggregation in the review. We did not impute missing data.
Assessment of heterogeneity
Homogeneity between included studies for which results were pooled will be tested with the DerSimonian and Laird method, and I2 > 40% was to be used as the cutoff level for significance. In cases of statistically significant heterogeneity, a random‐effects model was applied to the summary estimates. Unless specified otherwise, the fixed‐effect model was used.
Assessment of reporting biases
If missing or incomplete outcome data were identified, we attempted to contact study authors to obtain the missing data.
Data synthesis
Summary estimates were reported with their 95% CIs. We performed meta‐analysis using RevMan 5.2.
Subgroup analysis and investigation of heterogeneity
Subgroup analyses were planned to explore possible reasons for heterogeneity of study results for primary outcomes. A priori defined subgroups were based on:
-
age (preschool children vs school‐aged children);
-
admission site (hospital ward vs intensive care unit);
-
intensity of anticholinergic treatment; and
-
co‐intervention with systemic corticosteroids (yes or no).
Sensitivity analysis
For the primary outcomes, we planned to perform sensitivity analyses for publication status by removing the unpublished data, and for methodological quality by removing trials that did not meet the following criteria: random sequence generation, double‐blinding and low and balanced attrition in both groups.
Summary of findings table
We created a 'Summary of findings' table using the primary and secondary outcomes. We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias) to assess the quality of a body of evidence as it relates to the studies contributing data to the meta‐analyses for prespecified outcomes. We used methods and recommendations as described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and GRADE pro software. We justified all decisions to downgrade or upgrade the quality of studies by using footnotes and made comments to aid readers' understanding of the review, when necessary.
Results
Description of studies
See Characteristics of included studies; Characteristics of excluded studies and Characteristics of ongoing studies
Results of the search
The search, conducted until November 2013, yielded a total of 127 citations (Figure 1). Of these, 120 citations were excluded for the following non–mutually exclusive reasons: (1) duplicate references (N = 3), (2) ongoing trials (N = 1), (3) studies not randomised trials (N = 8), (4) participants not asthmatic (N = 7), (5) participants not exclusively children (N = 38), (6) participants not hospitalised for an acute asthma exacerbation or not receiving treatment beyond initial treatment in the emergency department (N = 48), (7) intervention not a combination of anticholinergics and β2‐agonists (N = 11) and (8) control intervention not β2‐agonist alone (N = 4).
Study flow diagram.
Included studies
Seven trials met the inclusion criteria for this review. Of these, three eligible clinical trials did not contribute data to the review because reports were incomplete (Lew 1990; Mirkinson 2000; Ozdemir 2003). The data presented hereafter pertain to only four eligible trials, representing a total of 472 children hospitalised for an acute asthma exacerbation that contributed data to this meta‐analysis (Craven 2001; Goggin 2001; Rayner 1987; Storr 1986). All four trials were published in full text. We describe below the characteristics of trials that contributed data for analysis for this review.
Design
All included trials had a parallel‐group design (Craven 2001; Goggin 2001; Rayner 1987; Storr 1986), and the data on lung function as presented in one trial (Lew 1990) could not be aggregated because of its cross‐over design.
Participants
All four trials involved children one to 18 years of age (Craven 2001; Goggin 2001; Rayner 1987; Storr 1986). The mean (or median) age of children in three studies was five years or younger (Craven 2001; Goggin 2001; Storr 1986) and 6.5 years in Rayner 1987. Most trials described a gender ratio ranging from 59% to 73% males. One study (Goggin 2001) enrolled children with moderate to severe asthma symptoms at baseline, defined as requiring inhaled β2‐agonists at a minimum of every two hours, having forced expiratory volume in one second (FEV1) of 25% to 80% of predicted value or having a clinical asthma score of three to nine on a scale of zero to 10 (higher indicating worse). The other trials did not report asthma severity at baseline (Craven 2001; Rayner 1987; Storr 1986), and no trials provided data that were stratified according to the severity of baseline airway obstruction.
Intervention drugs
Important variability must be noted in the proportion of participants who had received anticholinergics before randomisation. Indeed, in Goggin 2001, children had received intensive ipratropium bromide before randomisation, with a mean of 5.9 doses in the control group and 6.0 doses in the treatment group, but in Craven 2001, only three of 210 participants (all in the treatment group) had received anticholinergics in the emergency department before randomisation. The number of participants who received anticholinergics before enrolment was not reported in the remaining three studies (Lew 1990; Rayner 1987; Storr 1986)
The intervention drug and frequency (Table 1) were as follows: 250 μg ipratropium bromide every four hours, every six hours afterwards and then every eight hours, depending on the asthma care algorithm phase (Craven 2001); 250 μg ipratropium bromide every 30 to 60 minutes initially, progressing to every two hours and then to every four hours as the participant improved clinically (Goggin 2001); 250 μg ipratropium bromide every eight hours (Rayner 1987); or 250 μg ipratropium bromide given within set limits at the discretion of the nursing staff (Storr 1986).
| Studies | Anticholinergic treatment |
| 250 μg of ipratropium bromide by jet nebulisation every 4 hours during phase I, every 6 hours during phase II and every 8 hours during phase III of the ACA | |
| Nebulised ipratropium bromide inhalation solution 1.0 mL (250 μg) every half an hour to 1 hour at the beginning, progressing to 2 hours and then to 4 hours as the patient improves clinically | |
| An inhalation of 0.1% atropine sulfate (0.05 mg/kg up to 2 mg in total) at baseline or 4 hours later depending on randomisation | |
| Nebulised ipratropium 250 μg 30 minutes after the first dose of salbutamol and every 8 hours afterwards | |
| Nebuliser with 250 μg ipratropium bromide given within set limits at the discretion of the nursing staff |
Co‐intervention
The use of systemic corticosteroids was variable. All children in two trials received corticosteroids delivered intravenously or orally (Craven 2001; Goggin 2001). However, only 78% of participants in the control group and 74% of those in the treatment group were given oral corticosteroids in one study (Rayner 1987), and fewer participants received systemic steroids (26% and 28% in control and treatment groups, respectively) in the remaining study (Storr 1986). Intravenous aminophylline was used as a rescue therapy in two older studies (Rayner 1987; Storr 1986). If the clinical condition of a child was deteriorating in Craven 2001, subcutaneous epinephrine (0.01 mg/kg, maximum 0.5 mg) was added as part of the rescue therapy.
Outcomes
The primary efficacy outcome, that is, duration of hospital stay, was documented in three trials (Craven 2001; Goggin 2001; Rayner 1987) in a total of 327 children. None of these trials reported the primary safety outcome for our review, that is, serious adverse events. Other reported outcomes included admission to the ICU, need for supplemental asthma therapy, time to short‐acting β2‐agonists spaced at four hours or longer, relapse within 72 hours of discharge from the hospital and change from baseline in asthma severity measured as lung function, symptoms or clinical scores, adverse health effects and withdrawals.
Excluded studies
Risk of bias in included studies
Full details of risk of bias for all seven eligible trials are listed in the Characteristics of included studies table, and a graphical summary is presented in Figure 2. However, the following information pertains only to the four trials contributing data to the review.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
All four trials reported the method of randomisation in sufficient detail to confirm adequacy.
Allocation
Two of four trials contributing data reported details on allocation concealment, and both convincingly reported a valid allocation concealment (Craven 2001; Goggin 2001). Therefore, we judged these two trials to be at low risk of bias and the remaining trials to be at unclear risk of bias (Rayner 1987; Storr 1986).
Blinding
All trials reported double‐blinding, with convincing details indicating low risk of bias.
Incomplete outcome data
All four trials contributing data (Craven 2001; Goggin 2001; Rayner 1987; Storr 1986) reported on missing data; the proportion of withdrawals or missing values was balanced in numbers across intervention groups with similar reasons for missing data across groups, or no missing data were noted at all; thus all four trials were considered at low risk of bias on this criterion.
Selective reporting
Two trials (Craven 2001; Goggin 2001) clearly specified primary and secondary outcomes and reported all outcomes and thus were considered at low risk of bias on this criterion.
Other potential sources of bias
We encountered no other significant sources of bias in the included trials contributing data to the review.
Effects of interventions
Primary efficacy outcome: duration of hospital stay
Three trials (Craven 2001; Goggin 2001; Rayner 1987) representing 327 children hospitalised with acute asthma contributed to the primary endpoint, that is, duration of hospital stay. The difference in hours of length of hospitalisation between participants treated with the combination of anticholinergics and short‐acting β2‐agonists versus β2‐agonists alone was not statistically significant, with an MD of ‐0.28 hours (95% CI ‐5.07 to 4.52; Analysis 1.1; Figure 3) and no apparent heterogeneity (I2 = 0%). Because of the similarity of trials contributing to the primary outcome with regards to the age of enrolled children, the absence of any trial conducted in the ICU, the use of similar doses of anticholinergics and co‐intervention with systemic corticosteroids in all participants in two trials (Craven 2001; Goggin 2001) and three‐quarters of participants in the remaining trial (Rayner 1987), we could not perform any of the a priori planned subgroup analyses on the primary efficacy outcome.
Forest plot: Combination of anticholinergics + β2‐agonists versus β2‐agonists alone, outcome: duration of the hospital stay (hours).
As all trials contributing data on the primary outcome were of high methodological quality and were published as full text, we did not perform sensitivity analyses on quality and publication status.
Primary safety outcome: serious adverse events
No trials reported information on serious adverse events.
Secondary outcomes
Duration of stay in ICU
No studies reported on this outcome.
Admission to the ICU
No admission to the ICU was described by the one study reporting on this outcome (Analysis 1.2).
Ventilation assistance
No studies reported on this outcome.
Need for supplemental asthma therapy
Four trials (Craven 2001; Goggin 2001; Rayner 1987; Storr 1986) representing 465 children contributed data on this outcome. No statistically significant impact of therapy on the need for supplemental therapy was described (RR 0.77, 95% CI 0.41 to 1.42; Analysis 1.3). The supplemental asthma therapy was intravenous aminophylline in two studies (Rayner 1987; Storr 1986), intravenous aminophylline or oral theophylline in one study (Goggin 2001) and an "intensification regimen," consisting of subcutaneous epinephrine (0.01 mg/kg, maximum 0.5 mg) and a one‐time 500‐μg dose of ipratropium bromide nebulised in combination with higher‐dose albuterol (5.0 mg) in another study (Craven 2001).
Time to short‐acting β2‐agonists spaced at four hours or longer
Two trials (Craven 2001; Goggin 2001) representing 290 participants contributed to this outcome. No statistically significant group difference (MD ‐2.17, 95% CI ‐7.01 to 2.66; Analysis 1.4) was reported.
Change from baseline in asthma severity measured as lung function, symptoms or clinical scores
Two trials reported on the change from baseline in the asthma clinical score eight to 36 hours after initial treatment (higher is worse). No statistically significant group difference was described (SMD 0.02, 95% CI ‐0.34 to 0.38; Analysis 1.5).
Only a single trial reported on peak expiratory flow rate (PEFR) and described no statistically significant group difference in any of these outcomes (Analysis 1.7). No pooling of data was thus possible.
Relapse
No relapse was reported in the one study reporting on these outcomes (Analysis 1.6).
Adverse health effects
Two trials (Craven 2001; Goggin 2001) totaling 290 children reported that no adverse health events were observed (Analysis 1.8).
Withdrawals
Two trials (Craven 2001; Goggin 2001) representing 294 children contributed to this outcome. No statistically significant group difference was noted in overall withdrawals (RR 0.80, 95% CI 0.38 to 1.70; Analysis 1.9) or in the proportion of withdrawals due to deterioration reported by a single trial (Analysis 1.10).
Discussion
Summary of main results
In children hospitalised on hospital wards for an acute asthma exacerbation, the combination of nebulised anticholinergics with short‐acting β2‐agonists was associated with no statistically significant reduction in duration of hospital stay. We did not set a priori boundary for equivalence, but after reviewing the literature to determine the cut‐off for a clinically meaningful reduction in length of hospital stay, we proposed that a group difference of 8 hours be considered the minimally clinically important based on various reported length of hospital stay (Cunningham 2008; Smith 2003; Lim 2000). Indeed, the narrow confidence intervals rule out an effect of eight hours or larger of combination therapy over β2‐agonists alone. Because of the similarity of trial designs and participant characteristics (supported by the absence of significant heterogeneity) and incomplete reporting, it was not possible to explore whether characteristics of participants or therapy, such as age, admission site (ward or ICU) or intensity of anticholinergic treatment or co‐interventions, could influence the magnitude of response attributable to the addition of anticholinergics. All trials contributing to the primary outcome were of high methodological quality and were published in full text, thus no bias due to quality or publication status was apparent. Although power was severely limited by the small number of trials, precluding the use of any statistics (Higgins 2011), inspection of the funnel plot did not suggest bias.
This finding was supported by all secondary outcomes, which showed no statistically significant group differences in the need for supplemental asthma therapy, asthma clinical scores, time to short‐acting β2‐agonists spaced at four hours or longer and withdrawals. Although the remaining outcomes could not be pooled because they were reported in only one trial, none showed a statistically significant group difference or effect.
Our findings contrast with those of a prior systematic review of RCTs, which concluded on the efficacy of multiple doses of inhaled anticholinergics in combination with β2‐agonist therapy versus β2‐agonist alone in the emergency management of severe asthma exacerbations (Plotnick 2000) and, more recently, on the treatment of children with moderate and severe exacerbations (Griffiths 2013). Recognising that the onset of action of oral corticosteroids is within three to four hours (Rowe 2001), we hypothesise that the beneficial effect of anticholinergics is best observed before the onset of action of systemic corticosteroids, at which point the effect of the latter surpasses that of the former. As hospital admission typically occurs at least three to four hours after oral corticosteroids are administered in the emergency department, perhaps the relative beneficial effect of anticholinergics becomes negligible when compared with systemic corticosteroids.
Only two trials (Craven 2001; Goggin 2001) examined adverse health effects; in both trials, no adverse health effects were observed in either intervention group. In the absence of adverse events in 145 intervention participants in these two trials, the upper limit around this null estimate is 2%. No trial reported any serious adverse health event—our primary safety outcome—possibly because of the absence of events, although we cannot rule out sub‐optimal documentation or reporting. The paucity of data prevents any firm conclusions on the safety of either treatment strategy based on the proportion of adverse health effects or withdrawals, but the proportions would appear to be low. However, use of anticholinergics in the emergency department was not associated with an increase in serious adverse events or adverse events (Griffiths 2013; Plotnick 2000).
Overall completeness and applicability of evidence
The results of this review apply to children one to 18 years old, with a good representation of preschoolers and school‐aged children, as three studies (Craven 2001; Goggin 2001; Storr 1986) included children with a mean or median age younger than five years and one trial (Rayner 1987) included predominantly school‐aged children. Concomitant use of aminophylline and withholding of systemic corticosteroids in participants with acute asthma was noted in two older studies (Rayner 1987; Storr 1986); these approaches contrast to the current standard practice in which systemic corticosteroids was used systematically in all enrolled participants in the two recent studies (Craven 2001; Goggin 2001). Of note, one cannot extrapolate these data to children with impending respiratory failure admitted to the ICU, as the trial was conducted in the ICU.
Quality of the evidence
A very modest number of studies contributed data to the review. Trials contributing most of the data were of high methodological quality. Yet the four studies contributing data totaled only 472 children hospitalised for acute asthma. Clearly the large confidence intervals observed for all outcomes suggest that additional trials could change the conclusion. In addition, the paucity of trials prevented identification of subgroups (e.g. age, intensity of treatment, admission site) for whom the treatment may provide more or less effect. Finally, evidence regarding the safety profile of nebulised anticholinergics is insufficient.
Potential biases in the review process
No evidence of bias due to publication or methodological quality was shown by funnel plot. We recognise that, in view of the small number of studies, our power to identify bias was limited.
Agreements and disagreements with other studies or reviews
On the basis of efficacy, available results support the recommendation of current national and international guidelines (GINA 2014; NAEPP 2011) that anticholinergics should not be used in children hospitalised for an acute asthma exacerbation beyond initial treatment in the emergency department.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Forest plot: Combination of anticholinergics + β2‐agonists versus β2‐agonists alone, outcome: duration of the hospital stay (hours).
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 1 Duration of hospital stay (hours).
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 2 Admission to the intensive care unit.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 3 Need for supplemental asthma therapy.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 4 Time to short‐acting β2‐agonists spaced at 4 hours or longer (hours).
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 5 Asthma clinical scores 8 to 36 hours after initial treatment.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 6 Relapse within 72 hours of discharge from hospital.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 7 Percentages of predicted PEFR at 8 to 36 hours after initial treatment.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 8 Adverse health effects.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 9 Overall withdrawals.
Comparison 1 Combination of anticholinergics (AC) + β2‐agonists versus β2‐agonists alone, Outcome 10 Withdrawals due to deterioration.
| Combination of AC + beta2‐agonists versus beta2‐agonists alone | ||||||
| Patient or population: children hospitalised with an acute asthma exacerbation | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No. of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Beta2‐agonistsalone | Combination of AC + beta2‐agonists | |||||
| Duration of the hospital stay (hours) | The mean duration of the hospital stay (hours) in the control groups was | The mean reduction in duration of hospital stay (hours) was 0.28 hours in the intervention groups Mean 43.8 (39 to 49) hours | ‐0.28 hours | 327 | ⊕⊕⊕⊝ | |
| Time to regular dose of short‐acting beta2‐agonists (hours)2 Follow‐up: 4 hours or longer | The mean time to regular short‐acting beta2‐agonists in the control groups was 33 hours | The mean reduction in time to regular short‐acting beta2‐agonists in the intervention groups was 2.17 hours Mean 31 (26 to 36) hours | ‐2.17 hours | 269 | ⊕⊕⊕⊝ | |
| Asthma clinical scores Follow‐up: 8 to 36 hours after initial treatment | See comment | See comment | 0.02 SMD (‐0.34 to 0.38) | 117 | ⊕⊕⊕⊝ | Scores were measured on different scales. |
| Admission to the intensive care unit | See comment | See comment | Not estimable | 210 | ⊕⊕⊕⊝ | Single trial reported admission to the intensive care unit. No events were reported |
| Overall withdrawals | 12 per 100 | 7 per 100 | OR 0.59 | 294 | ⊕⊕⊕⊝ | |
| Adverse health events | See comment | See comment | Not estimable | 290 | ⊕⊕⊕⊝ | No adverse events were reported in either trial |
| *The basis for the assumed risk is the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence. | ||||||
| 1Downgraded because of wide confidence intervals. 2This is the time at which emergency dosing with the trial intervention was replaced with regular 4‐hourly treatment with short‐acting beta2‐agonists. 3No events were reported in either arm of the studies. | ||||||
| Studies | Anticholinergic treatment |
| 250 μg of ipratropium bromide by jet nebulisation every 4 hours during phase I, every 6 hours during phase II and every 8 hours during phase III of the ACA | |
| Nebulised ipratropium bromide inhalation solution 1.0 mL (250 μg) every half an hour to 1 hour at the beginning, progressing to 2 hours and then to 4 hours as the patient improves clinically | |
| An inhalation of 0.1% atropine sulfate (0.05 mg/kg up to 2 mg in total) at baseline or 4 hours later depending on randomisation | |
| Nebulised ipratropium 250 μg 30 minutes after the first dose of salbutamol and every 8 hours afterwards | |
| Nebuliser with 250 μg ipratropium bromide given within set limits at the discretion of the nursing staff |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Duration of hospital stay (hours) Show forest plot | 3 | 327 | Mean Difference (IV, Fixed, 95% CI) | ‐0.28 [‐5.07, 4.52] |
| 2 Admission to the intensive care unit Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 3 Need for supplemental asthma therapy Show forest plot | 4 | 465 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.77 [0.41, 1.42] |
| 4 Time to short‐acting β2‐agonists spaced at 4 hours or longer (hours) Show forest plot | 2 | 290 | Mean Difference (IV, Fixed, 95% CI) | ‐2.17 [‐7.01, 2.66] |
| 5 Asthma clinical scores 8 to 36 hours after initial treatment Show forest plot | 2 | 117 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.02 [‐0.34, 0.38] |
| 6 Relapse within 72 hours of discharge from hospital Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 7 Percentages of predicted PEFR at 8 to 36 hours after initial treatment Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 8 Adverse health effects Show forest plot | 2 | 290 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 9 Overall withdrawals Show forest plot | 2 | 294 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.59 [0.27, 1.30] |
| 10 Withdrawals due to deterioration Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
