Key messages
• Adding a long-acting beta2-agonist (LABA) or a long-acting muscarinic antagonist (LAMA) to medium-dose inhaled corticosteroids (ICS) likely reduces asthma attacks requiring treatment with oral steroids and increases the odds of satisfactory symptom control compared to ICS alone, whereas doubling the dose of ICS probably does not. The amount of data we found was much larger for LABAs than for LAMAs.
• We need to learn more about the long-term side effects of high-dose ICS because the average duration of the included studies was only six months. Using the lowest effective ICS doses is encouraged to minimise corticosteroid-associated side effects.
What is asthma, and how is it treated?
Asthma is a chronic respiratory condition characterised by inflammation and narrowing of the airways that causes symptoms such as wheezing, coughing, chest tightness, and shortness of breath. Treatment involves the use of inhalers, which are relievers (e.g. short-acting bronchodilators) and, if needed, preventers (e.g. ICS), as well as avoiding triggers and maintaining a healthy lifestyle.
What did we want to find out?
In this study, we wanted to find the best way to improve asthma control in teenagers and adults who were not managing their asthma well with medium-dose ICS. We compared three different treatments.
The three treatments we looked at were:
• adding a LABA inhaler;
• adding a LAMA inhaler;
• doubling the dose of the current ICS inhaler.
The main goal was to see which treatment was the most effective in improving asthma control and which one was the safest in terms of side effects.
What did we do?
We collected and analysed data from 35 studies, which included a total of 38,276 people with uncontrolled asthma while on medium-dose ICS, using a special method called network meta-analysis. This enabled us to simultaneously compare multiple inhaler groups. We compared adding a LABA or a LAMA to medium-dose ICS, versus doubling the dose of ICS or using medium-dose ICS alone.
What did we find?
Adding a LABA or a LAMA to medium-dose ICS likely reduces asthma attacks requiring treatment with oral steroids. It also increases the odds of satisfactory symptom control compared to ICS alone whereas doubling the dose of ICS probably does not. The amount of data we found was much larger for LABAs than for LAMAs.
Adding a LABA or LAMA to medium-dose ICS or doubling the dose of ICS is unlikely to reduce asthma-related hospitalisations or serious side effects. The addition of a LAMA to ICS possibly reduces side effects and treatment discontinuation. However, the combination of ICS/LAMA therapy requires two separate inhalers whereas ICS/LABA combinations are available in a single inhaler.
What are the limitations of the evidence?
We need to learn more about the long-term side effects of high-dose ICS because the average duration of the included studies was only six months. The study results may not be applicable to people who smoke or individuals who experience side effects from LAMAs because those individuals were either not included or were very few in this review.
How up-to-date is this evidence?
This review is up-to-date to December 2022.
The review findings suggest that MD- or HD-ICS/LABA and MD-ICS/LAMA reduce moderate to severe asthma exacerbations and increase the odds of ACQ responders compared to MD-ICS whereas HD-ICS probably does not. The evidence is generally stronger for MD- and HD-ICS/LABA than for MD-ICS/LAMA primarily due to a larger evidence base. There is no evidence to suggest that ICS/LABA, ICS/LAMA, or HD-ICS/LABA reduces severe asthma exacerbations or SAEs compared to MD-ICS. MD-ICS/LAMA likely reduces all-cause AEs and results in a slight reduction in treatment discontinuation due to AEs compared to MD-ICS.
The above findings may assist in deciding on a treatment option during the stepwise approach of asthma management. Longer-term safety of higher than medium-dose ICS needs to be addressed in phase 4 or observational studies given that the median duration of included studies was six months.
Inhaled corticosteroids (ICS) are the mainstay treatment for persistent asthma. Escalating treatment is required when asthma is not controlled with ICS therapy alone, which would include, but is not limited to, adding a long-acting beta2-agonist (LABA) or a long-acting muscarinic antagonist (LAMA) or doubling the dose of ICS.
To assess the efficacy and safety of adding a LABA or LAMA to ICS therapy versus doubling the dose of ICS in adolescents and adults whose asthma is not well controlled on medium-dose (MD)-ICS using a network meta-analysis (NMA), and to provide a ranking of these treatments according to their efficacy and safety.
We searched the Cochrane Airways Trials Register, CENTRAL, MEDLINE, Embase, Global Health, ClinicalTrials.gov, and the World Health Organization ICTRP for pre-registered randomised controlled trials (RCTs) from January 2008 to 19 December 2022.
We searched for studies including adolescents and adults with uncontrolled asthma who had been treated with or were eligible for MD-ICS, comparing it to high-dose (HD)-ICS, ICS/LAMA, or ICS/LABA. We excluded cluster- and cross-over RCTs. Studies were of at least 12 weeks duration.
We conducted a systematic review and network meta-analysis according to a previously published protocol. We used Cochrane’s Screen4ME workflow to assess search results. We used Grading of Recommendations Assessment, Development and Evaluation (GRADE) to assess the certainty of evidence. The primary outcome is asthma exacerbations (moderate and severe).
We included 38,276 participants from 35 studies (median duration 24 weeks (range 12 to 78); mean age 44.1; 38% male; 69% white; mean forced expiratory volume in one second 2.1 litres and 68% of predicted).
MD- and HD-ICS/LABA likely reduce and MD-ICS/LAMA possibly reduces moderate to severe asthma exacerbations compared to MD-ICS (hazard ratio (HR) 0.70, 95% credible interval (CrI) 0.59 to 0.82; moderate certainty; HR 0.59, 95% CrI 0.46 to 0.76; moderate certainty; and HR 0.56, 95% CrI 0.38 to 0.82; low certainty, respectively), whereas HD-ICS probably does not (HR 0.94, 95% CrI 0.70 to 1.24; moderate certainty). There is no clear evidence to suggest that any combination therapy or HD-ICS reduces severe asthma exacerbations compared to MD-ICS (low to moderate certainty).
This study suggests no clinically meaningful differences in the symptom or quality of life score between dual combinations and monotherapy (low to high certainty).
MD- and HD-ICS/LABA increase or likely increase the odds of Asthma Control Questionnaire (ACQ) responders at 6 and 12 months compared to MD-ICS (odds ratio (OR) 1.47, 95% CrI 1.23 to 1.76; high certainty; and OR 1.59, 95% CrI 1.31 to 1.94; high certainty at 6 months; and OR 1.61, 95% CrI 1.22 to 2.13; moderate certainty and OR 1.55, 95% CrI 1.20 to 2.00; high certainty at 12 months, respectively).
MD-ICS/LAMA probably increases the odds of ACQ responders at 6 months (OR 1.32, 95% CrI 1.11 to 1.57; moderate certainty). No data were available at 12 months. There is no clear evidence to suggest that HD-ICS increases the odds of ACQ responders or improves the symptom or qualify of life score compared to MD-ICS (very low to high certainty).
There is no evidence to suggest that ICS/LABA or ICS/LAMA reduces asthma-related or all-cause serious adverse events (SAEs) compared to MD-ICS (very low to high certainty). HD-ICS results in or likely results in little or no difference in the included safety outcomes compared to MD-ICS as well as HD-ICS/LABA compared to MD-ICS/LABA.
The pairwise meta-analysis shows that MD-ICS/LAMA likely reduces all-cause adverse events (AEs) and results in a slight reduction in treatment discontinuation due to AEs compared to MD-ICS (risk ratio (RR) 0.86, 95% confidence interval (CI) 0.77 to 0.96; 4 studies, 2238 participants; moderate certainty; and RR 0.51, 95% CI 0.26 to 0.99; 4 studies, 2239 participants; absolute risk reduction 10 fewer per 1000 participants; moderate certainty, respectively). The NMA evidence is in agreement with the pairwise evidence on treatment discontinuation due to AEs, but very uncertain on all-cause AEs, due to imprecision and heterogeneity.