Background: Airway obstruction is the main trait of severe equine asthma that affects respiratory function and elicits detrimental effects on clinical presentation. Only few and underpowered clinical studies have investigated the impact of improvement in lung function induced by bronchodilators on the clinical signs of asthma-affected horses.

Objectives: To identify the minimal important difference (MID) in lung function elicited by bronchodilator leading to a meaningful improvement in clinical signs.

Study design: Pairwise meta-analysis and meta-regression analysis.

Methods: Literature searches were performed for studies that investigated the effect of bronchodilator therapy on lung function and clinical condition of asthmatic horses. The relationship between the change in lung function variables and clinical score was analysed via random-effect meta-regression. One-point change of the Improved clinically Detectable Equine Asthma Scoring System (IDEASS) score was used to identify the MID.

Results: A significant (P<0.05) relationship was found between the changes in IDEASS score and maximum change in transpulmonary pressure (?Ρ_max) or pulmonary resistance (R_L). Since only the model resulting for R_L passed through the origin (Y-intercept when X = 0: -0.31, 95% CI -0.75 to 0.14), this variable was used to identify the MID correlated with a meaningful improvement in clinical signs. The resulting MID value was a change in R_L of 0.63 cm H₂O/L/s (95% CI 0.33-0.94), representing the slope of meta-regression model (high quality of evidence).

Main limitations: No long-term studies investigated the effect of bronchodilator agents on both lung function and clinical signs in asthmatic horses.

Conclusions: In conclusion, bronchodilator pharmacotherapy in equine asthma elicits clinically meaningful effect when R_L increases >=1 cm H₂O/L/s, a value indicating the MID. Assessing the MID based on change in RL may improve the quality of evidence and the scientific impact of future clinical trials as it extends beyond the simple, and limiting, evaluation of statistical significance.