Comparative efficacy and safety of phosphodiesterase inhibitors in chronic obstructive pulmonary disease: a systematic review and network meta-analysis

Phosphodiesterase (PDE) inhibitors have been investigated as potential therapies for chronic obstructive pulmonary disease (COPD), and combined PDE3 and PDE4 inhibition has been proposed to provide complementary anti-inflammatory and bronchodilatory effects. This network meta-analysis compared the efficacy and safety of oral roflumilast, inhaled tanimilast, and inhaled ensifentrine. A systematic search (PubMed, Embase, Web of Science, Cochrane Library) through January 2026 identified randomized controlled trials (RCTs) comparing PDE inhibitors with placebo or active controls. The endpoints comprised forced expiratory volume in one second (FEV₁), St. George’s Respiratory Questionnaire (SGRQ), annualized acute exacerbation rate (AAER), and adverse events (AEs). A frequentist random-effects network meta-analysis (NMA) was conducted. We included 17 reports describing 20 distinct RCTs (18,672 patients). Ensifentrine (mean difference [MD] = 75.82 mL) and roflumilast (MD = 55.60 mL) significantly improved FEV₁ compared with placebo, whereas tanimilast did not. The FEV₁ and SGRQ networks were anchored by the common placebo comparator, with no active-treatment head-to-head comparisons available. Ensifentrine had the highest surface under the cumulative ranking curve (SUCRA) value for FEV₁ improvement (0.9381). No intervention significantly improved SGRQ scores or reached the 4-point minimal clinically important difference. Ensifentrine and roflumilast significantly reduced the annualized acute exacerbation rate. Oral roflumilast was associated with higher risks of total adverse events and adverse-event-related treatment discontinuation versus placebo, whereas no statistically significant differences from placebo were observed for the evaluated safety outcomes of inhaled ensifentrine and tanimilast. Ensifentrine had the largest FEV₁ point estimate, the lowest exacerbation rate ratio, and the highest SUCRA value for FEV₁ in the primary analysis. These findings support further evaluation of inhaled dual PDE3/4 inhibition as an add-on maintenance strategy for COPD. Not applicable.