Pseudomonas aeruginosa forms biofilms in the cystic fibrosis lung. Quorum sensing (QS) controls biofilm maturation, immune evasion, antibiotic tolerance and virulence factor production. Garlic shows QS inhibitory activity in vitro and in animal models. We report the first clinical trial in man of a QS inhibitor.We randomized 34 patients to garlic or olive oil capsules (both 656 mg daily). Clinical outcomes and safety bloods were measured at baseline and after 8 weeks treatment. In this exploratory study, analysis was per protocol.Eight patients withdrew, leaving 26 for analysis (13 garlic). With placebo, there was a greater decline in mean (SD) percentage change from baseline FEV(1) [-3.6% (11.3)] than with garlic [-2.0% (12.3)]. This was not significant (mean difference = 1.6, 95% CI -12.7 to 15.9, P = 0.8). The mean (SD) increase in weight was greater with garlic [1.0% (2.0)] than with placebo [0.6% (2.0)]--non-significant (mean difference = 0.4%, 95% CI -1.3 to 2.0, P = 0.6). The median (range) change in clinical score with garlic was -1 (-3 to 5) and 1 (-1 to 4) with placebo (negative score means improvement). This was non-significant [median difference = -1 (-3 to 0), P = 0.16]. In the garlic group, seven patients had IV antibiotics versus five placebo. There was a highly significant correlation between plasma and sputum measurements of the QS molecule 3-oxo-C12-HSL (Pearson correlation coefficient = 0.914, P = 0.004). At the end of treatment five patients in each group had abnormal liver function or triglycerides and five garlic patients (one placebo) reported minor adverse effects.Garlic capsules were well tolerated. Although there was no significant effect of garlic compared to placebo in this pilot study, there was a suggestion of improvement with garlic which should be investigated in a larger trial.