BackgroundPrior investigations evaluating the predictive value of zinc-depleted media for MBL-susceptibility testing have focused on Enterobacterales. Therein, bacterial killing observed with meropenem (MEM) in vivo was concordant with its pharmacodynamic profile using MIC values determined in zinc-depleted media compared with conventional cation-adjusted Mueller-Hinton broth (CAMHB). This study aims to evaluate the exposure-response relationship of MEM against VIM- and NDM-harboring P. aeruginosa (PSA) using the murine thigh infection model and zinc-depleted MICs.MethodsMBL-harboring PSA isolates (VIM n=11; NDM n=10) were tested both in vivo (neutropenic murine thigh infection model) and in vitro (broth microdilution). The 24h murine thigh study was conducted with treatment groups receiving a humanized MEM 2g q8h (3h infusion) dose. Six different zinc-limited media were prepared by the addition of EDTA at concentrations ranging from 3 to 300 mg/L to CAMHB. MEM MICs were determined in triplicate in conventional CAMHB and zinc-limited media. Time > MIC values (generated in each zinc-depleted media) were then plotted against the change in 24h bacterial density count in an Emax model.ResultsAverage 0 h bacterial densities were 5.21 ± 0.40 and 5.13 ± 0.81 log10 CFU/thigh for NDM and VIM isolates, respectively. MEM resulted in -0.09 CFU reduction to +3.69 CFU growth against NDM isolates. MEM resulted in -2.59 CFU reduction to +4.81 CFU growth against VIM isolates. All MEM MICs in conventional CAMHB were >64 µg/mL for NDM and ranged from 8 to >64 µg/mL for VIM isolates. Increasing EDTA concentrations resulted in several-fold MIC reductions and on average, a larger magnitude of reduction was observed among VIM- (6-fold) compared with NDM-harboring PSA (4-fold) in CAMHB-EDTA 300 mg/L relative to CAMHB. For both NDM- and VIM-harboring PSA, an Emax model with MICs generated in CAMHB+EDTA 30 mg/L (r2 = 0.88) provided the highest correlation with MEM in vivo activity compared with CAMHB (r2 = 0.55).ConclusionResults indicate that MIC values generated in conventional CAMHB do not appropriately characterize the in vivo efficacy of meropenem against MBL-harboring PSA, and addition of EDTA (30 mg/L) to CAMHB appears to be a viable option for in vitro testing of these organisms.Disclosures David P. Nicolau, PharmD, Abbvie, Cepheid, Merck, Paratek, Pfizer, Wockhardt, Shionogi, Tetraphase (Other Financial or Material Support, I have been a consultant, speakers bureau member, or have received research funding from the above listed companies.)