Relationships between pharmacodynamic indices (PI), such as the area under the concentration-time curve (AUC)/MIC ratio and time > MIC (T(>MIC)), and efficacy have been described for antimicrobial drugs. The use of these quantitative relationships may increase the power to demonstrate significant effects of drugs, obviating the need to include large numbers in comparative trials. Patients with cystic fibrosis (CF) hospitalized for treatment of an infectious exacerbation due to Pseudomonas aeruginosa were eligible for the study. They received tobramycin 3.3 mg/kg tid as initial therapy in combination with ticarcillin 125 mg/kg qid. Blood samples were drawn at t = 0-8 h after infusion. Pharmocokinetic parameters and PI were calculated for every individual and correlated to the relative improvement in forced expiratory volume during the first second (FEV1) and forced vital capacity (FVC) between pretreatment and days 9-11 as a measure of efficacy. The 3 PI fAUC/MIC, f Peak/MIC, and T(>MIC) of tobramycin showed a significant correlation with effect and was the highest for the fAUC/MIC relationships with FEV1 and FVC as determined both by the Emax model as well as Spearman correlations (r = 0.77, P = 0.002 and 0.58, P = 0.036 for FEV1 and FVC). Pharmacokinetic parameters AUC and Peak as such showed no significant correlation with effect, nor did the MIC values. There is a significant relationship between PI of aminoglycosides and efficacy parameter (increase in FEV1 and FVC) in patients with CF. This study demonstrates the applicability of pharmacodynamic relationships in determining efficacy of antimicrobial therapy, by demonstrating a strong PI-effect relationship in a group of only 13 patients.