Previous investigations into metallo-β-lactamase (MBL)-harbouring Enterobacterales suggest that susceptibility testing in zinc-limited media may be more appropriate in predicting β-lactam in vivo activity. There are limited data with MBL-harbouring Pseudomonas aeruginosa. Forty-three MBL-harbouring P. aeruginosa isolates (IMP, n = 11; VIM, n = 12; NDM, n = 10; SPM, n = 10) and two P. aeruginosa control isolates (KPC, n = 1; WT, n = 1) were evaluated. Meropenem activity was evaluated in the murine neutropenic thigh model using humanized exposures. Susceptibility testing was conducted in conventional CAMHB, EDTA-supplemented CAMHB (3-300 mg/L EDTA) and Chelex-treated CAMHB (0-1.0 mg/L re-supplemented zinc), resulting in a range of meropenem MIC values for each isolate. A sigmoidal Emax model was fitted to fT>MIC versus change in log10 cfu/thigh to estimate the goodness of fit (R2). Increasing EDTA concentrations or limiting the amount of zinc in broth resulted in several-fold reductions in MIC among the majority of the MBL-harbouring P. aeruginosa while the MICs for the KPC and WT isolates were unchanged. Bacterial killing in vivo was variable, with the range of killing spanning -3.29 to +4.81 log10 change in cfu/thigh. Addition of 30 mg/L EDTA and Chelex-treated CAMHB (with no zinc supplementation) provided broth conditions for susceptibility testing that best predicted in vivo efficacy (R2 > 0.7). Among MBL-harbouring P. aeruginosa, meropenem in vivo efficacy is best represented by the pharmacodynamic profile generated using MICs determined in EDTA-supplemented or zinc-limited broth. In addition to previous data with Enterobacterales, antibiotic susceptibility testing in media that approximates physiological conditions makes it possible to uncover potential and existing therapeutic agents.