Background: Systemic corticosteroids improve the outcomes of COPD exacerbations at an increased risk of side effects. Inhaled corticosteroids may exert a better benefit to risk ratio.

Aims: We aimed to assess the safety and efficacy of inhaled compared to systemic corticosteroids in COPD exacerbations.

Methods: We conducted a meta-analysis of randomised controlled trials (RCTs; protocol ID: CRD42021284297). We assessed the outcomes outlined in the ERS COPD Exacerbations Core Outcome Set and rated the certainty of evidence according to GRADE.

Results: We included 20 eligible RCTs with 2,160 participants comparing inhaled or nebulised budesonide to systemic corticosteroids in moderate or severe COPD exacerbations. All RCTs entailed high risk of bias. We found with low certainty that both corticosteroid routes had similar impact on treatment failure (relative risk, RR 1.75, 95% confidence intervals [0.76, 4.03], I²=0%, 565 participants), breathlessness (standardised mean difference 0.0 [-0.19, 0.19], I² =0%, n=442), serious adverse events (RR 1.47 [0.56, 3.88], I²=0%, n=246), health-related quality of life, length of hospitalisation, future exacerbations and mortality. Moderate-certainty evidence implied a tendency for less adverse events with inhaled compared to systemic corticosteroids (RR 0.80 [0.64, 1.0], I²=0%, n=480). There was no evidence of higher risk of hospital or critical care admission with inhaled steroids. Data on the subgroup of patients with eosinophilic exacerbations were not reported.

Conclusions: Our findings suggest potential non-inferiority of inhaled to systemic corticosteroids in unselected COPD exacerbations. Large and rigorous RCTs assessing eosinophilic exacerbations are warranted.