Studies of beta-lactam pharmacodynamics in infected patients are sparse. In this study, classification and regression tree (CART) and logistic regression analyses were used to identify which pharmacodynamic indices and magnitudes were significant predictors of meropenem efficacy for 101 adult patients with lower respiratory tract infections (LRTI). Using demographic data, a validated population pharmacokinetic model was employed to predict pharmacokinetic parameters and free serum concentrations in the studied patients. Pharmacodynamic indices [percentage of the dosing interval that free drug concentrations remain above the MIC (% fT > MIC), f(maximum concentration of drug in serum) (fC(max))/MIC, fC(min)/MIC, and f(area under the concentration-time curve) (fAUC)/MIC] were calculated based on the baseline pathogen with the highest drug MIC for each patient. The median (range) of percent fT > MIC, fC(max)/MIC, fC(min)/MIC, and fAUC/MIC were 100% (0 to 100%), 728.8 (0.8 to 15,777), 19.9 (0.01 to 278), and 3,605.4 (2.7 to 60,865.9), respectively. CART identified the following breakpoints as significant predictors for microbiological response: >54% fT > MIC, a fC(max)/MIC > 383, and a fC(min)/MIC > 5; fC(min)/MIC > 5 was the only significant predictor of clinical response. Due to 100% fT > MIC achieved in the majority of LRTI patients, fC(min)/MIC was the statistically significant parameter associated with meropenem clinical and microbiological response in the adults with LRTI. The findings for LRTI patients can be applied to optimize meropenem dose regimens to achieve clinical success and microbiological eradication in clinical practice.