MICs of tigecycline determined by Etest were 4 to 12 times (three ATCC strains) and 2 to 8 times (50 clinical isolates) higher in Mueller-Hinton agar from Merck than in Mueller-Hinton agar from either Oxoid or Difco. This was related to a much higher concentration of manganese in the medium from Merck.Tigecycline is the first glycylcycline approved for clinical use (16). It can be oxidized when added to a broth medium that has been oxygenated during storage (13) or when tigecycline-containing broth is stored before inoculation. This can be solved using freshly autoclaved broth or recently prepared solid media (2) and is expected to be avoided in the disk diffusion or gradient diffusion (Etest) assays.During a nation-wide study in Spain on the activity of tigecycline against clinical isolates using Etest strips (AB Biodisk, Solna, Sweden), the results in our center were up to 16 to 20 times higher than those obtained in other centers (A. Pérez, Wyeth Farma S.A., Spain, personal communication). After excluding the possibility that some technical error occurred in our laboratory, we hypothesized that these results could be related to differences in the composition of the Mueller-Hinton agar used in different laboratories, as the commercial source for this medium in our center (Merck) was different from that in many other laboratories in Spain.In this study, the activity of tigecycline determined with Etest strips on two different lots of commercially available MuellerHinton agar from Oxoid (Basingstoke, Hampshire, England; lots 418599 and 456733), Difco (Difco Mueller-Hinton agar [BD, Sparks, MD]; lots 6093187 and 6177851), and Merck (Darmstadt, Germany; lots VL379337507 and VL546637607) was evaluated. We also determined the effect that differences in the concentrations of ions of the evaluated Mueller-Hinton agars may have on the in vitro activity of tigecycline.In a preliminary study, five clinical isolates (selected among those we already tested in the multicenter study in Spain) and four reference strains, namely, Escherichia coli ATTC 25922, Pseudomonas aeruginosa ATCC 27953, Staphylococcus aureus 29213, and Enterococcus faecalis 29212, were evaluated. MICs of tigecycline for these two groups of organisms were consistently 4 to 12 times higher in the medium from Merck than in the medium from Difco or Oxoid, except for the P. aeruginosa ATCC strain (MICs of 64 to 128 g/ml in the three media) (data not shown).In a second phase, MICs of tigecycline for 50 clinical isolates (one per patient; clonally unrelated as assessed by repetitive extragenic palindromic PCR and/or pulsed-field gel electrophoresis) were determined in duplicate experiments on different days. Extended-spectrum -lactamase (ESBL) production was confirmed by disk diffusion (4). Resistance to methicillin in S. aureus and to vancomycin in Enterococcus faecium was confirmed by PCR detection of the mecA (9) and vanA (12) genes, respectively. MICs of tigecycline were also determined by microdilution, according to CLSI guidelines (5) using fresh Mueller...