Budesonide; Glycopyrrolate; Formoterol

What it is

This is a pressurised metered-dose inhaler containing three separate drugs in a co-suspension formulation: budesonide, an inhaled corticosteroid; glycopyrrolate, an anticholinergic; and formoterol fumarate, a long-acting beta-2 agonist. In the US it comes in two strengths per actuation, 160/9/4.8 mcg and 160/18/4.8 mcg, with two inhalations taken twice daily, so the delivered daily pairs differ between the COPD and asthma indications. It was first approved in the US in 2020 for COPD, and the asthma indication was added later. It is a maintenance inhaler and is not indicated for relief of acute bronchospasm.

What the research says

The question this drug was built to answer is whether adding an inhaled steroid to two long-acting bronchodilators cuts flare-ups, and in the large trials it does, by a modest relative amount with a real pneumonia cost. ETHOS, with 8,509 patients over 52 weeks, brought the annual rate of moderate or severe COPD exacerbations from 1.42 on the two bronchodilators down to 1.08, a 24% reduction. Across the whole triple-therapy literature, a meta-analysis put the trade-off in absolute terms for people with a prior exacerbation: 230 fewer exacerbations and 16 more pneumonias per 1,000 patients. The mortality claim is genuinely contested. ETHOS found fewer deaths than on the bronchodilator-only arm but no significant difference against the steroid-containing dual arm, and a large real-world cohort found the opposite direction, which its authors attributed to the trials withdrawing inhaled steroids at randomisation. In asthma, the twin KALOS and LOGOS trials showed better lung function and a 14% reduction in severe exacerbations against the two-drug comparison.

Evidence grade: Well established.

How it works

Drug class: Fixed-dose triple inhaled combination: an inhaled corticosteroid (budesonide), a long-acting muscarinic antagonist (glycopyrrolate), and a long-acting beta-2 agonist (formoterol fumarate), delivered from one pressurised metered-dose inhaler

Three drugs with three different jobs come out of the same inhaler. Formoterol stimulates beta-2 receptors on airway muscle so the muscle relaxes and the airway widens, and it keeps doing so for about twelve hours. Glycopyrrolate blocks muscarinic receptors, removing the nerve signal that keeps airway muscle contracted, so the airway widens by a second, independent route. Budesonide is a corticosteroid that damps the inflammation in the airway wall, which is the part aimed at reducing flare-ups rather than at opening the airway in the moment. None of the three works fast enough to be a rescue inhaler, and the label says so explicitly. (Source 1)

What it is used for

• Two large randomised trials and several meta-analyses show fewer moderate or severe exacerbations and better lung function than the matching two-drug inhalers, with pneumonia occurring more often. Benefit concentrates in people who have already had an exacerbation. Evidence: established. (Source 2)
• ETHOS found a lower risk of death on the higher-steroid triple therapy than on the two bronchodilators, but no significant difference against budesonide-formoterol, and no benefit for the lower-steroid triple dose. A large real-world cohort found higher mortality on triple therapy. Mortality is not an approved claim in the US label. Evidence: disputed. (Source 3)
• The twin phase 3 KALOS and LOGOS trials in 4,311 treated participants with asthma inadequately controlled on medium or high-dose inhaled steroid plus a long-acting beta agonist met their primary lung function end points and reduced severe exacerbations against the pooled two-drug comparator, though not significantly against the co-suspension two-drug comparator alone. Evidence: established. (Source 4)
• Explicitly excluded. The label states it is not indicated for relief of acute bronchospasm and should not be started in acutely deteriorating COPD or asthma. Evidence: not-supported. (Source 1)
• One 19-patient open-label crossover pilot reported better inspiratory capacity on triple therapy than after run-in, with no significant change in exhaled nitric oxide or symptom questionnaires. Nineteen men, open label, no placebo: this is a signal, not an answer. Evidence: limited. (Source 5)

Interactions

• Strong CYP3A4 inhibitors: ritonavir, ketoconazole, itraconazole, clarithromycin and others (label): Budesonide is broken down by the CYP3A4 enzyme. Drugs that block that enzyme let more budesonide build up in the blood, which raises the chance of whole-body steroid effects such as adrenal suppression. (Source 6)
• Grapefruit juice (pharmacokinetic study): Grapefruit juice blocks CYP3A in the gut wall. A crossover study in eight healthy men showed four days of regular-strength grapefruit juice about doubled the bioavailability of oral budesonide. The important caveat is that the study used swallowed budesonide capsules, not an inhaler, and most inhaled budesonide bypasses gut-wall metabolism, so the size of any effect on an inhaler is unknown. (Source 7)
• St John's wort and other CYP3A4 inducers (theoretical): This is theory, not measurement. The label establishes that CYP3A4 is budesonide's main metabolic route, and St John's wort is a known inducer of that enzyme, so it would be expected to lower budesonide exposure and could blunt the steroid's effect. We found no study of St John's wort with inhaled or oral budesonide, and the label names only inhibitors, not inducers. (Source 6)
• Caffeine, theophylline and other xanthines (label): Xanthines, the family that includes caffeine and theophylline, can deepen the drop in blood potassium that a long-acting beta-2 agonist causes. The label names xanthine derivatives alongside steroids and diuretics for this effect. (Source 6)
• Loop and thiazide diuretics (and by the same mechanism, anything that lowers potassium) (label): Potassium can fall and the ECG can change when these diuretics are combined with a beta-2 agonist, and the label says this is acutely worsened if the beta-2 agonist dose is exceeded. This is the main reason potassium is worth watching in people on both. (Source 8)
• MAO inhibitors, tricyclic antidepressants and other QTc-prolonging drugs (label): These drugs amplify formoterol's effect on the heart, and prolonging the QTc interval raises the risk of ventricular arrhythmia. (Source 8)
• Beta-blockers, including timolol eye drops for glaucoma (label): Beta-blockers work against formoterol and can trigger severe airway narrowing in people with COPD or asthma. The label says people with these conditions should not normally be given beta-blockers, and that where there is no alternative a cardioselective one may be considered with caution. This matters for anyone also using a non-selective beta-blocker eye drop. (Source 9)
• Other anticholinergic medicines (including over-the-counter antihistamines and bladder drugs) (label): Glycopyrrolate's anticholinergic effects add to those of other anticholinergic drugs, which can worsen dry mouth, constipation, urinary retention and glaucoma symptoms. The label advises avoiding the combination. (Source 9)
• Alcohol (label): No alcohol interaction is documented. We searched the whole US label and found no mention of alcohol anywhere in it, and the label states plainly that no formal interaction studies were performed with this product at all, which is the honest ceiling on what is known. (Source 10)

Stopping it

• The one large randomised withdrawal trial of inhaled steroids on top of two long-acting bronchodilators, WISDOM, stepped fluticasone down over 12 weeks in 2,485 people with a history of COPD exacerbation. Time to the first moderate or severe exacerbation met the prespecified non-inferiority criterion, so stopping the steroid did not clearly increase flare-ups. (Source 11)
• Lung function did fall when the steroid came out, by about 38-43 mL of trough FEV1, with no change in breathlessness and only minor changes in health status. So the trade-off on stopping the steroid component is measurable lung function against no clear change in flare-ups or symptoms. (Source 11)
• The mortality debate turns on exactly this point. The real-world cohort author argues that the trials' apparent mortality benefit came from patients who had been on inhaled steroids being randomised to a non-steroid arm, which is a withdrawal effect rather than a benefit of the triple inhaler. Anyone reading the mortality numbers should know that stopping an established steroid at randomisation is the alternative explanation on the table. (Source 12)
• The label describes one situation where stopping needs care in the opposite direction: a person moving off systemic steroid tablets onto this inhaler. It instructs a slow taper, because the inhaled dose does not replace suppressed adrenal function. There is no withdrawal syndrome or dependence described for the inhaler itself. (Source 13)

What goes wrong

In the same trial, confirmed pneumonia was about twice as common in the arms containing an inhaled steroid as in the bronchodilator-only arm. (Source 2)

• Randomized trial, High certainty.
• Size: 8,509 patients.
• Who: Adults with moderate-to-very-severe COPD and a prior exacerbation.
• How long: 52 weeks.
• Result: Confirmed pneumonia 3.5-4.5% in steroid-containing arms versus 2.3% in glycopyrrolate-formoterol; any adverse event 61.7-64.5% across arms. Absolute excess roughly 1.2-2.2 percentage points over 52 weeks.
• Funding: industry-funded (AstraZeneca)

The incidence of any adverse event was similar across the treatment groups (range, 61.7 to 64.5%); the incidence of confirmed pneumonia ranged from 3.5 to 4.5% in the groups that included inhaled glucocorticoid use and was 2.3% in the glycopyrrolate-formoterol group.

A real-world cohort of 144,000 COPD patients found higher, not lower, mortality on triple therapy than on dual bronchodilators, and attributed the trial mortality findings to inhaled steroids being withdrawn at randomisation. (Source 12)

• Cohort study, Low certainty.
• Size: 117,729 new users of LAMA-LABA-ICS and 26,666 of LAMA-LABA.
• Who: COPD patients aged 50 or older in the UK Clinical Practice Research Datalink, treated 2002-2018, weighted by fine stratification of propensity scores.
• How long: One year of follow-up.
• Result: All-cause mortality hazard ratio 1.17 (95% CI 1.04-1.31); severe exacerbation 1.19 (1.08-1.32); pneumonia 1.29 (1.16-1.45). No excess mortality in subgroups with an asthma diagnosis (HR 0.99, 0.74-1.34) or two or more prior exacerbations (HR 0.88, 0.70-1.11). Observational: confounding by indication cannot be excluded and this is an association, not a demonstrated effect.
• Funding: not stated in the abstract; the author has published extensively on bias in COPD pharmacoepidemiology.

Limit of this finding: This is an observational cohort, not a trial, so treatment was not randomised and sicker patients are more likely to have been put on triple therapy; the author's own explanation for the gap with the trials is inhaled-steroid withdrawal at randomisation. It should be read alongside the ETHOS trial results (martinez-2021-ethos-mortality), which point the other way, rather than as a refutation of them.

Randomized trials of triple therapy including an inhaled corticosteroid (ICS) for chronic obstructive pulmonary disease (COPD) reported remarkable benefits on mortality compared with dual bronchodilators, likely resulting from ICS withdrawal at randomization.

The same cohort's numerical results run opposite to the trial signal for death, severe exacerbation and pneumonia alike. (Source 12)

• Cohort study, Low certainty.
• Size: 144,395 new users in total.
• Who: UK primary-care COPD cohort.
• How long: One year.
• Result: Mortality HR 1.17 (1.04-1.31); severe exacerbation HR 1.19 (1.08-1.32); pneumonia HR 1.29 (1.16-1.45)
• Funding: not stated in the abstract.

Limit of this finding: This is an observational cohort, not a trial, so treatment was not randomised and sicker patients are more likely to have been put on triple therapy; the author's own explanation for the gap with the trials is inhaled-steroid withdrawal at randomisation. It should be read alongside the ETHOS trial results (martinez-2021-ethos-mortality), which point the other way, rather than as a refutation of them.

The adjusted hazard ratio (HR) of all-cause mortality with LAMA-LABA-ICS compared with LAMA-LABA was 1.17 (95% CI: 1.04-1.31) while for severe exacerbation and pneumonia it was 1.19 (1.08-1.32) and 1.29 (1.16-1.45) respectively.

The label's own safety section reports confirmed pneumonia in 4.2% of people on the triple inhaler over 52 weeks against 2.3% on the two bronchodilators alone, with a similar rate (4.5%) on the steroid-containing dual inhaler. (Source 14)

• Official position, Certainty not rated.
• Size: ETHOS safety population: 2,144 on triple, 2,125 on glycopyrrolate-formoterol, 2,136 on budesonide-formoterol.
• Who: Adults with COPD, mean age 64.7 years, 59.7% male.
• How long: 52 weeks.
• Result: 52-week trial: confirmed pneumonia 4.2% on BREZTRI 320/18/9.6 (n = 2144), 3.5% on BGF 160/18/9.6 (n = 2124), 2.3% on GFF 18/9.6 (n = 2125), 4.5% on BFF 320/9.6 (n = 2136); fatal pneumonia in 2 subjects on BGF 160 and 3 on GFF, none on BREZTRI 320. 24-week trial: 1.9%, 1.6% and 1.9%, no fatal cases.
• Funding: manufacturer label (AstraZeneca), SPL published 2026-06-23.

Limit of this finding: The label's own arithmetic does not add up in the second trial it describes: it gives the 24-week trial a total of 1,896 subjects, but the three arms it lists (639, 625 and 320) come to 1,584. The quote is reproduced exactly as the FDA label prints it and has not been corrected. Read the pneumonia percentages for the 52-week trial, which are internally consistent, as the usable figures; do not rely on the 24-week arm sizes or that total.

In a 52-week trial of subjects with COPD (n = 8,529), the incidence of confirmed pneumonia was 4.2% for BREZTRI AEROSPHERE 320 mcg/18 mcg/9.6 mcg (n = 2144), 3.5% for budesonide, glycopyrrolate and formoterol fumarate [BGF MDI 160 mcg/18 mcg/9.6 mcg] (n = 2124), 2.3% for GFF MDI 18 mcg/9.6 mcg (n = 2125) and 4.5% for BFF MDI 320 mcg/9.6 mcg (n = 2136).

The label names pneumonia among the commonest adverse reactions for both the COPD and the asthma indication. (Source 15)

• Official position, Certainty not rated.
• Size: Pooled trial populations.
• Who: Adults with COPD; adults and adolescents 12 and over with asthma.
• How long: Up to 52 weeks.
• Result: COPD reactions at 2% or more: upper respiratory tract infection, pneumonia, back pain, oral candidiasis, influenza, muscle spasm, urinary tract infection, cough, sinusitis, diarrhoea. Asthma reactions at 2% or more: nasopharyngitis, pneumonia, headache.
• Funding: manufacturer label (AstraZeneca), SPL published 2026-06-23.

•COPD: Most common adverse reactions (incidence ≥ 2%) are upper respiratory tract infection, pneumonia, back pain, oral candidiasis, influenza, muscle spasm, urinary tract infection, cough, sinusitis and diarrhea. (6.1)

What the evidence supports

Over 52 weeks, triple therapy cut the annual rate of moderate or severe COPD exacerbations by about a quarter compared with the two long-acting bronchodilators. (Source 2)

• Randomized trial, High certainty.
• Size: 8,509 patients in the modified intention-to-treat population.
• Who: Adults with moderate-to-very-severe COPD and at least one exacerbation in the previous year.
• How long: 52 weeks.
• Result: Annual exacerbation rates 1.08 (triple, budesonide 320 mcg), 1.07 (triple, budesonide 160 mcg), 1.42 (glycopyrrolate-formoterol), 1.24 (budesonide-formoterol). Triple 320 versus glycopyrrolate-formoterol rate ratio 0.76 (95% CI 0.69-0.83), P<0.001; versus budesonide-formoterol 0.87 (0.79-0.95), P = 0.003. In absolute terms about 0.34 fewer exacerbations per patient-year against the bronchodilator pair.
• Funding: industry-funded (AstraZeneca); several authors are AstraZeneca employees.

The annual rates of moderate or severe exacerbations were 1.08 in the 320-μg-budesonide triple-therapy group (2137 patients), 1.07 in the 160-μg-budesonide triple-therapy group (2121 patients), 1.42 in the glycopyrrolate-formoterol group (2120 patients), and 1.24 in the budesonide-formoterol group (2131 patients).

After the trial retrieved almost all missing vital-status data, the higher-steroid triple therapy showed a lower risk of death than the two bronchodilators. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 8,509 randomised; week-52 vital status available for 99.6% of the intention-to-treat population after retrieval.
• Who: Adults with moderate-to-very-severe COPD and prior exacerbation history.
• How long: 52 weeks.
• Result: Hazard ratio for death 0.51 (95% CI 0.33-0.80), unadjusted P = 0.0035, triple 320 versus glycopyrrolate-formoterol. Cardiovascular deaths 0.5%, 0.8%, 1.4% and 0.5% in the triple 320, triple 160, glycopyrrolate-formoterol and budesonide-formoterol arms.
• Funding: industry-funded (AstraZeneca); mortality was a prespecified secondary end point and 384 of 8,509 patients were missing vital status in the original analysis.

Limit of this finding: This trial result does not stand alone. A real-world cohort of about 144,000 COPD patients (suissa-2022-realworld-triple) found the opposite direction for death on triple therapy (hazard ratio 1.17, 95% CI 1.04-1.31) and argues the trial benefit came from patients having their inhaled steroid withdrawn when they were randomised to a non-steroid arm. The two should be read together: the trial shows what happened inside the trial, and it is disputed whether that carries over to people already taking an inhaled steroid.

risk of death with BGF 320 was significantly lower than GFF (hazard ratio, 0.51; 95% confidence interval, 0.33-0.80; unadjusted P = 0.0035)

In the 24-week lung function trial, triple therapy improved the 4-hour FEV1 curve against the steroid-bronchodilator pair by about 104 mL. (Source 16)

• Randomized trial, Moderate certainty.
• Size: 1,902 randomised (640 triple, 627 glycopyrrolate-formoterol, 316 budesonide-formoterol, 319 open-label budesonide-formoterol dry powder)
• Who: Current or former smokers aged 40-80 with moderate-to-very-severe COPD, symptomatic on two or more inhaled maintenance therapies, no exacerbation history required.
• How long: 24 weeks, with a 28-week safety extension.
• Result: FEV1 AUC0-4 least squares mean difference 104 mL (95% CI 77-131), p<0.0001 versus budesonide-formoterol MDI; 91 mL (64-117) versus the dry-powder comparator; pre-dose trough FEV1 22 mL (4-39), p=0.0139 versus glycopyrrolate-formoterol.
• Funding: industry-funded: Pearl, a member of the AstraZeneca Group.

Over 24 weeks, BGF MDI significantly improved FEV1 AUC0-4 versus BFF MDI (least squares mean difference 104 mL, 95% CI 77 to 131; p<0·0001)

In two twin phase 3 asthma trials, triple therapy reduced severe exacerbations by 14% against the pooled two-drug comparator and improved trough FEV1 by 76 mL. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 8,820 recruited, 4,311 received treatment across KALOS and LOGOS.
• Who: People aged 12-80 with asthma inadequately controlled on daily medium or high-dose inhaled corticosteroid plus long-acting beta-2 agonist, 378 sites in 20 countries and 324 sites in 15 countries.
• How long: 24 to 52 weeks.
• Result: Severe exacerbation incidence rate ratio 0.86 (95% CI 0.76-0.97), p=0.012 versus the combined two-drug comparator and 0.82 (0.71-0.94), p=0.0043 versus the current-formulation comparator; trough FEV1 difference 76 mL (95% CI 57-94), p<0.0001; FEV1 AUC0-3 90 mL (72-108), p<0.0001.
• Funding: industry-funded (AstraZeneca)

BGF 28·8 reduced severe exacerbation rates versus BFFcombined (incidence rate ratio 0·86, 95% CI 0·76-0·97; p=0·012) and versus BFFS (0·82, 0·71-0·94; p=0·0043).

A systematic review of single-inhaler triple therapies put the number needed to treat at roughly 25 to 50 patients to keep one person free of a moderate or severe exacerbation. (Source 17)

• Systematic review, Low certainty.
• Size: Six RCTs from 15 reports, identified from 523 records.
• Who: Patients with COPD.
• How long: Varies; the reviewers flag important design differences between trials.
• Result: Annual moderate or severe exacerbation reduction 15-52% versus LAMA/LABA, 15-35% versus LABA/ICS, 20% versus LAMA; patient-based NNT roughly 25-50, event-based NNT roughly 3-11. Greater absolute benefit with higher eosinophil counts, more frequent prior exacerbations, and in ex-smokers.
• Funding: not stated in the abstract; PROSPERO CRD42018102125.

The patient-based number needed to treat for the moderate or severe exacerbation outcome ranged between approximately 25-50 (preventing one patient from having an event) and the event-based number needed to treat of around 3-11 (preventing one event).

What the evidence does not support

The same mortality analysis found no significant difference against the steroid-containing dual therapy, and no benefit at all for the lower steroid dose. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 8,509 randomised.
• Who: Adults with moderate-to-very-severe COPD.
• How long: 52 weeks.
• Result: Triple 320 versus budesonide-formoterol hazard ratio 0.72 (95% CI 0.44-1.16), P = 0.1721; triple 160 not significantly different from either dual comparator.
• Funding: industry-funded (AstraZeneca)

There were no significant differences in mortality when comparing BGF 320 with BFF (hazard ratio, 0.72; 95% confidence interval, 0.44-1.16; P = 0.1721), nor were significant differences observed when comparing BGF 160 against either dual comparator.

In that same trial the week-24 trough FEV1 gain over the two bronchodilators was not statistically significant, so the lung-function advantage depends on which measure you pick. (Source 16)

• Randomized trial, Moderate certainty.
• Size: 1,902 randomised.
• Who: Symptomatic moderate-to-very-severe COPD without a required exacerbation history.
• How long: 24 weeks.
• Result: Change from baseline in morning pre-dose trough FEV1 at week 24 versus glycopyrrolate-formoterol 13 mL (95% CI -9 to 36), p=0.2375.
• Funding: industry-funded (Pearl, AstraZeneca Group)

there was a non-significant improvement in the change from baseline in morning pre-dose trough FEV1 at week 24 versus GFF MDI (13 mL, -9 to 36 mL; p=0·2375)

In the same asthma programme, the exacerbation reduction against the matched co-suspension two-drug inhaler alone was not statistically significant. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 4,311 treated participants.
• Who: Asthma inadequately controlled on inhaled corticosteroid plus long-acting beta-2 agonist.
• How long: 24 to 52 weeks.
• Result: Exacerbation rate ratio 0.90 (95% CI 0.78-1.03), p=0.12 for triple versus the co-suspension budesonide-formoterol comparator. Adverse events occurred in 53.2% on the higher-glycopyrronium triple dose versus 55.2% and 58.4% on the two-drug comparators.
• Funding: industry-funded (AstraZeneca)

Exacerbation rate ratio for BGF 28·8 versus BFFA was 0·90 (95% CI 0·78-1·03; p=0·12).

That same meta-analysis found no significant benefit for breathlessness scores or hospitalisation. (Source 18)

• Meta-analysis, Moderate certainty.
• Size: 11 studies, 14,145 patients.
• Who: Stable COPD with dyspnoea and/or exercise intolerance.
• How long: Varies by trial.
• Result: Dyspnoea standardised mean difference 0.09 (95% CI -0.02 to 0.19), P = 0.09; hospitalisation rate ratio 0.78 (0.58-1.06), P = 0.11. Subgroup analysis by inhaler class found no significant difference in any critical outcome.
• Funding: not stated in the abstract.

No significant difference in dyspnea scores (standardized mean difference, 0.09; 95% CI, -0.02 to 0.19; P = 0.09) or risk of hospitalization (rate ratio, 0.78; 95% CI, 0.58-1.06; P = 0.11) was noted.

Where the evidence is mixed

Pneumonia was uncommon and similar across arms in the shorter 24-week trial, so the pneumonia excess seen over a year is not visible at six months. (Source 16)

• Randomized trial, Moderate certainty.
• Size: 1,902 randomised.
• Who: Moderate-to-very-severe COPD.
• How long: 24 weeks.
• Result: Pneumonia incidence under 2% and similar across treatments; two treatment-related deaths, both in the glycopyrrolate-formoterol group.
• Funding: industry-funded (Pearl, AstraZeneca Group)

Pneumonia incidence was low (<2%) and similar across treatments. There were two treatment-related deaths, both in the GFF MDI group.

Across 21 trials of inhaled triple therapy, exacerbation rates fell against every comparator, with pneumonia rising about 50% against dual bronchodilation. (Source 19)

• Meta-analysis, Moderate certainty.
• Size: 21 trials (19 publications)
• Who: Patients with COPD, mostly advanced.
• How long: Varies by trial.
• Result: Moderate or severe exacerbation rate ratios 0.71 (95% CI 0.60-0.85) versus LAMA monotherapy, 0.78 (0.70-0.88) versus LAMA+LABA, 0.77 (0.66-0.91) versus ICS+LABA; pneumonia relative risk 1.53 (1.25-1.87) versus LAMA+LABA. Quality was graded with GRADE.
• Funding: not stated in the abstract; published in The BMJ with a PROSPERO registration.

The overall safety profile of triple therapy is reassuring, but pneumonia was significantly higher with triple therapy than with dual therapy of LAMA and LABA (relative risk 1.53, 95% confidence interval 1.25 to 1.87).

A GRADE-rated meta-analysis stated the trade-off in absolute numbers: 230 fewer exacerbations and 16 more pneumonias per 1,000 patients with a prior exacerbation. (Source 18)

• Meta-analysis, Moderate certainty.
• Size: 11 studies, 14,145 patients.
• Who: Stable COPD with dyspnoea and/or exercise intolerance.
• How long: Varies by trial.
• Result: Pneumonia relative risk 1.47 (95% CI 1.20-1.80), P<0.001; exacerbation relative risk 0.75 (0.68-0.82), P<0.001; 230 fewer exacerbations and 16 more pneumonias per 1,000 patients with one or more exacerbations in the past year.
• Funding: not stated in the abstract; GRADE applied to rate certainty.

For patients with a history of one or more AECOPDs in the past year, triple therapy resulted in 230 fewer AECOPDs and 16 more cases of pneumonia per 1,000 patients.

The reviewers also warned that cross-trial comparison of these inhalers is unreliable because the trials differ in design and population. (Source 17)

• Systematic review, Low certainty.
• Size: Six RCTs.
• Who: COPD.
• How long: Varies.
• Result: No pooled estimate; a judgement about heterogeneity and about the higher pneumonia incidence in the largest trial.
• Funding: not stated in the abstract.

There were important differences in study designs and populations impacting the interpretation of the results and indicating there would be significant heterogeneity in cross-trial comparisons.

In 19 patients with asthma-COPD overlap, adding glycopyrrolate to budesonide-formoterol improved inspiratory capacity but changed no symptom or inflammation measure. (Source 5)

• Randomized trial, Very low certainty.
• Size: 19 patients, all male.
• Who: Stable asthma-COPD overlap, mean age 70.7 years, all ex-smokers with a mean 63.1 pack-year history.
• How long: 12-week open-label crossover with 4-week washout, run-in and treatment periods.
• Result: Inspiratory capacity 1.93 L after run-in, 1.85 L after dual therapy, 2.11 L after triple therapy (p < 0.02 versus run-in); exhaled nitric oxide, asthma control questionnaire and COPD assessment test scores not significantly different between periods.
• Funding: not stated in the abstract; open-label pilot with no placebo.

IC values after the BUD/GLY/FORM triple therapy were significantly higher than those after the run-in (p < 0.02). FeNO values, ACQ, and CAT scores were not significantly different among the run-in, wash-out, and triple-therapy periods.

Where the research disagrees

Whether single-inhaler triple therapy reduces death in COPD

• Martinez and the ETHOS investigators (2021), after retrieving nearly all missing vital-status data, Prespecified secondary end point of a 52-week randomised double-blind trial in 8,509 patients, industry-funded: "risk of death with BGF 320 was significantly lower than GFF (hazard ratio, 0.51; 95% confidence interval, 0.33-0.80; unadjusted P = 0.0035)" (Source 3)
• The same analysis, on the comparison that isolates the muscarinic antagonist rather than the steroid, Same randomised trial, different comparator arm: "There were no significant differences in mortality when comparing BGF 320 with BFF (hazard ratio, 0.72; 95% confidence interval, 0.44-1.16; P = 0.1721), nor were significant differences observed when comparing BGF 160 against either dual comparator." (Source 3)
• Suissa, real-world cohort (2022), Propensity-weighted new-user cohort of 144,395 UK primary-care COPD patients, one year of follow-up; observational and open to confounding by indication: "The adjusted hazard ratio (HR) of all-cause mortality with LAMA-LABA-ICS compared with LAMA-LABA was 1.17 (95% CI: 1.04-1.31) while for severe exacerbation and pneumonia it was 1.19 (1.08-1.32) and 1.29 (1.16-1.45) respectively." (Source 12)

Whether the exacerbation benefit justifies the pneumonia risk for everyone, or only for people who already exacerbate

• Zheng and colleagues, BMJ meta-analysis (2018), Meta-analysis of 21 randomised trials with GRADE assessment: "Use of triple therapy resulted in a lower rate of moderate or severe exacerbations of COPD, better lung function, and better health related quality of life than dual therapy or monotherapy in patients with advanced COPD." (Source 19)
• Mammen and colleagues, Annals of the American Thoracic Society (2020), keyed koarai-2020-annals-triple, GRADE-rated meta-analysis of 11 trials, 14,145 patients, with absolute event estimates per 1,000: "In patients with COPD who complain of dyspnea and/or exercise intolerance, triple therapy is not superior to maintenance long-acting bronchodilator therapy, except in patients with a history of one or more exacerbations in the past year, in whom the benefits of reduction in AECOPD outweigh the increased risk of pneumonia." (Source 18)

How much

• Reference intake: There is no dietary reference intake for an inhaler. Dosing is set by the prescriber. As a position, the US label (SPL published 2026-06-23) gives two inhalations of 160 mcg/9 mcg/4.8 mcg twice daily for COPD, and two inhalations of 160 mcg/18 mcg/4.8 mcg twice daily for asthma. (Source 20)
• Upper limit: No tolerable upper intake level exists for a prescription drug. The label's stated ceiling is the same as its dose: the preparation section instructs that no more than two inhalations twice daily are taken, and the warnings add that it must not be combined with another long-acting beta-2 agonist because of overdose risk. (Source 20)
• Studied: ETHOS gave twice-daily budesonide 320 mcg or 160 mcg with glycopyrrolate 18 mcg and formoterol 9.6 mcg, against glycopyrrolate 18 mcg plus formoterol 9.6 mcg or budesonide 320 mcg plus formoterol 9.6 mcg, in 8,509 patients for 52 weeks. (Source 2)
• Studied: KRONOS gave budesonide/glycopyrrolate/formoterol 320/18/9.6 mcg twice daily against glycopyrrolate/formoterol 18/9.6 mcg, budesonide/formoterol 320/9.6 mcg and open-label budesonide/formoterol dry powder 400/12 mcg in 1,902 patients for 24 weeks. (Source 16)
• Studied: KALOS and LOGOS in asthma gave budesonide-glycopyrronium-formoterol at 320/28.8/10 mcg or 320/14.4/10 mcg twice daily against budesonide-formoterol 320/10 mcg and 320/9 mcg, for 24 to 52 weeks. (Source 4)

A common belief, and what the research shows

The belief: A triple inhaler is a stronger reliever, so it is the right thing to reach for when breathing suddenly gets worse.

What the research shows: It is the opposite. Everything in this inhaler is long-acting maintenance treatment, and the label states it directly: "BREZTRI AEROSPHERE is not indicated for the relief of acute bronchospasm". A second common misreading is that the headline mortality result settles the question. It does not: the same analysis that found a lower risk of death against the two bronchodilators reported no significant difference against budesonide-formoterol and no benefit for the lower steroid dose, and a 144,000-patient real-world cohort found mortality running in the other direction, attributed by its author to inhaled steroids being withdrawn at randomisation in the trials.

Questions and answers

What is it?

It is one inhaler holding three different maintenance drugs: budesonide, an inhaled steroid; glycopyrrolate, an anticholinergic; and formoterol, a long-acting beta-2 agonist. Two puffs are taken twice daily. In the US it is approved for the maintenance treatment of COPD in adults and of asthma in people 12 and over, and it is not a rescue inhaler. (Source 1)

What does it do in the body?

Formoterol and glycopyrrolate both widen the airway, by two different routes: one stimulates the beta-2 receptors on airway muscle and the other blocks the muscarinic signal that keeps the muscle tight. Budesonide does not open the airway; it reduces the inflammation in the airway wall, and that is the part credited with cutting flare-ups. None of the three acts fast enough to break an acute attack, which is why the label carries a limitation of use. (Source 21)

Is it good or bad for you?

Good and bad in the same breath, and which dominates depends on who is taking it. For people with COPD who have already had a flare-up, a GRADE-rated meta-analysis put the balance at 230 fewer exacerbations and 16 more pneumonias per 1,000 patients. For people without that history, the same review found triple therapy was not superior to bronchodilator maintenance, so the pneumonia risk is taken on with less to show for it. The steroid also carries the usual class risks: oral thrush, reduced bone density, cataract and glaucoma with long-term use, and growth monitoring in adolescents. (Source 18)

How do you get more of it?

Not applicable in the usual sense. This is a prescription inhaler, not a nutrient, so no food, supplement or behaviour increases it, and the amount delivered is fixed by the device and the prescribed schedule. What does change how much reaches the lung is technique: the label requires priming the inhaler with four sprays before first use, re-priming after seven days unused, and rinsing the mouth after each dose. Dose decisions belong to the prescriber. (Source 22)

If it is harmful, what reduces it?

Stopping the inhaler is what reduces exposure, and the component that carries most of the harm signal is the steroid. The one large randomised withdrawal trial, WISDOM, removed the inhaled steroid in steps over 12 weeks from people on two long-acting bronchodilators and found no clear rise in exacerbations, with a trough FEV1 loss of about 38-43 mL. Rinsing the mouth without swallowing after each dose is the label's measure for reducing the local steroid effect in the throat. Changing or stopping a maintenance inhaler is a prescriber's decision. (Source 11)

Why might someone be low in it or missing it?

Does not apply: nobody is naturally deficient in a manufactured inhaler. The useful version of the question is why someone might not be getting the benefit the trials showed, and the honest answer from the literature is that benefit is concentrated in a subgroup. A systematic review found the absolute benefit was greater in people with higher blood eosinophil counts, a history of frequent exacerbations, and in ex-smokers, which means people outside those groups may be taking the inhaler and getting little from it. (Source 17)

Which whole foods contain it or feed it?

None. There is no whole food containing budesonide, glycopyrrolate or formoterol, and nothing in the diet substitutes for them. Food matters here only as interaction: grapefruit juice blocks the gut enzyme that breaks budesonide down, and a crossover study in eight healthy men found it about doubled the bioavailability of swallowed budesonide, though that study did not test an inhaler. Caffeine and other xanthines are named by the label as able to deepen the fall in blood potassium that formoterol can cause. (Source 7)

What happens if you do not have it?

For someone with COPD and a history of flare-ups, going without this class of treatment means a higher exacerbation rate. In ETHOS the annual rate of moderate or severe exacerbations was 1.42 on the two long-acting bronchodilators and 1.08 on triple therapy, a difference of roughly a third of an exacerbation per person per year. Across the wider literature the exacerbation rate ratio was about 0.75 to 0.78 against dual therapy. What goes away alongside the benefit is the extra pneumonia risk, which is why the absolute framing matters. (Source 2)

How can you test for it?

What gets measured is the effect, not the drug. The trials used spirometry, specifically FEV1 measured as the area under the curve over the first hours after a dose and as morning pre-dose trough FEV1, plus counted exacerbations and the St George's Respiratory Questionnaire. Those two spirometry measures can disagree: in KRONOS the 4-hour FEV1 curve improved by 104 mL with high confidence while the week-24 trough gain of 13 mL was not statistically significant, so a single number can mislead. There is no routine blood test for these three drugs. (Source 16)

References

1. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL version 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - HIGHLIGHTS OF PRESCRIBING INFORMATION, Indications and Usage. 2026. Read the source
2. The New England journal of medicine. Triple Inhaled Therapy at Two Glucocorticoid Doses in Moderate-to-Very-Severe COPD.. 2020. PMID 32579807, DOI 10.1056/nejmoa1916046. Read the source
3. American journal of respiratory and critical care medicine. Reduced All-Cause Mortality in the ETHOS Trial of Budesonide/Glycopyrrolate/Formoterol for Chronic Obstructive Pulmonary Disease. A Randomized, Double-Blind, Multicenter, Parallel-Group Study.. 2021. PMID 33252985, DOI 10.1164/rccm.202006-2618oc. Read the source
4. The Lancet. Respiratory medicine. Papi A, Wise RA, Jackson DJ, et al; KALOS and LOGOS study investigators. Budesonide-glycopyrronium-formoterol fumarate dihydrate in uncontrolled asthma (KALOS and LOGOS): twin multicentre, double-blind, double-dummy, parallel-group, randomised, phase 3 trials. The Lancet Respiratory Medicine 2026. Cite as Papi et al. 2026; the key "israel-2026-kalos-logos" is a legacy key and names no author of this paper.. 2026. PMID 41692019, DOI 10.1016/s2213-2600(25)00457-6. Read the source
5. International journal of chronic obstructive pulmonary disease. Triple Therapy with Budesonide/Glycopyrrolate/Formoterol Fumarate Improves Inspiratory Capacity in Patients with Asthma-Chronic Obstructive Pulmonary Disease Overlap.. 2020. PMID 32103926, DOI 10.2147/copd.s231004. Read the source
6. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Drug Interactions 7.1-7.3. 2026. Read the source
7. Die Pharmazie. Grapefruit juice interaction with oral budesonide: equal effect on immediate-release and delayed-release formulations.. 2009. PMID 19694184. Read the source
8. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Drug Interactions 7.4-7.5. 2026. Read the source
9. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Drug Interactions 7.6-7.7. 2026. Read the source
10. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Full Prescribing Information, section 7 Drug Interactions introduction and 7.1. 2026. Read the source
11. The New England journal of medicine. Withdrawal of inhaled glucocorticoids and exacerbations of COPD.. 2014. PMID 25196117, DOI 10.1056/nejmoa1407154. Read the source
12. COPD. Triple Inhaler versus Dual Bronchodilator Therapy in COPD: Real-World Effectiveness on Mortality.. 2022. PMID 34544314, DOI 10.1080/15412555.2021.1977789. Read the source
13. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Warnings and Precautions highlights. 2026. Read the source
14. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL version 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - BREZTRI AEROSPHERE label, Warnings: Pneumonia. 2026. Read the source
15. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - HIGHLIGHTS OF PRESCRIBING INFORMATION, Adverse Reactions (most common reactions). 2026. Read the source
16. The Lancet. Respiratory medicine. Triple therapy with budesonide/glycopyrrolate/formoterol fumarate with co-suspension delivery technology versus dual therapies in chronic obstructive pulmonary disease (KRONOS): a double-blind, parallel-group, multicentre, phase 3 randomised controlled trial.. 2018. PMID 30232048, DOI 10.1016/s2213-2600(18)30327-8. Read the source
17. Respiratory research. Langham S, Lewis J, Pooley N, Embleton N, Langham J, Han MK, Chalmers JD. Single-inhaler triple therapy in patients with chronic obstructive pulmonary disease: a systematic review. Respiratory Research 2019;20:242. Cite as Langham et al. 2019; the key "halpin-2019-sitt-review" is a legacy key and names no author of this paper.. 2019. PMID 31684965, DOI 10.1186/s12931-019-1213-9. Read the source
18. Annals of the American Thoracic Society. Mammen MJ, Lloyd DR, Kumar S, Ahmed AS, Pai V, Kunadharaju R, Gupta S, Nici L, Aaron SD, Alexander PE. Triple Therapy versus Dual or Monotherapy with Long-Acting Bronchodilators for Chronic Obstructive Pulmonary Disease. A Systematic Review and Meta-analysis. Annals of the American Thoracic Society 2020;17(10):1308-1318, PMID 32692253, doi 10.1513/annalsats.202001-023oc. Cite as Mammen et al. 2020; the key "koarai-2020-annals-triple" is a legacy key and names no author of this paper. Not to be confused with the separate 2020 triple-therapy meta-analysis by Koarai et al. in Respiratory Investigation.. 2020. PMID 32692253, DOI 10.1513/annalsats.202001-023oc. Read the source
19. BMJ (Clinical research ed.). Triple therapy in the management of chronic obstructive pulmonary disease: systematic review and meta-analysis.. 2018. PMID 30401700, DOI 10.1136/bmj.k4388. Read the source
20. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL version 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - BREZTRI AEROSPHERE label, Recommended Dosage. 2026. Read the source
21. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL version 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Full Prescribing Information, section 1 Indications and Usage (1.1-1.3). 2026. Read the source
22. AstraZeneca Pharmaceuticals LP via DailyMed (FDA label, SPL published 2026-06-23). BREZTRI AEROSPHERE (budesonide, glycopyrrolate, and formoterol fumarate) inhalation aerosol - US prescribing information - Preparation and Administration Information 2.1. 2026. Read the source