Activity of tigecycline against ESBL-producing and AmpC-hyperproducing Enterobacteriaceae from South-East England

Sir, Escherichia coli and Klebsiella spp. with CTX-M extendedspectrum b-lactamases (ESBLs) and quinolone resistance are a rapidly increasing problem in the UK. These multiresistant bacteria pose a real therapeutic problem, with carbapenems and (potentially) temocillin the only intravenous antibiotics remaining consistently active for treatment in severe cases. Tigecycline (Wyeth, previously GAR-936) is a broad-spectrum glycylcycline derivative of minocycline and may provide a further treatment option. Tigecycline evades ribosomal protection [tet(M)] and efflux [tet(A-E)] mechanisms that compromise classical tetracyclines and has MICs 2 mg/L for Enterobacteriaceae except proteeae. We assessed tigecycline’s activity specifically versus ESBL and AmpC producers, most of them multiresistant to other antibiotics except carbapenems and temocillin. Consecutive Enterobacteriaceae isolates resistant to any or all of cefpodoxime, ceftazidime and cefotaxime were collected from 16 clinical diagnostic laboratories in London and SE England (see the Acknowledgements section) participating in a resistance survey. As previously described MICs were determined centrally by BSAC agar dilution methodology, and resistance mechanisms were inferred by interpretive reading of antibiograms. Genes for CTX-M ESBLs and acquired AmpC enzymes were sought by multiplex PCR. Members of a particularly prevalent CTX-Mproducing clone, E. coli strain A, with a characteristic blaCTX-M-15 linkage to IS26 were sought by specific PCR. The distribution of species and mechanism among the collected isolates were detailed previously. For the present investigation we used all those with a substantive cephalosporin resistance mechanism: i.e. hyperproduction of an AmpC or K1 b-lactamase or production of CTX-M or other ESBL. Among the ESBL-producing isolates, 67% were resistant to ciprofloxacin, 49% to gentamicin and 74% to trimethoprim. The MIC distribution for tetracycline was bimodal, with peaks at 2 and >128 mg/L and with 38% of the isolates in the latter resistant cluster; the distribution of minocycline MICs was strongly skewed, with 49% resistant with MICs of 8 mg/L. Tigecycline MIC distributions are shown by species and resistance group in Table 1. Among the whole collection (n = 846) 703 would count as susceptible at the EUCAST breakpoint of 1 mg/L; 108 as intermediate, with MICs of 2 mg/L and 35 as resistant, with MICs of >2 mg/L. In the case of E. coli, 99% (417/420) of the tigecycline MICs fell between 0.125 and 1 mg/L and only three isolates, all hyperproducers of AmpC, were intermediate, with none resistant. The isolates with intermediate resistance were also all resistant to minocycline and tetracycline, with MICs 8 mg/L. Modal MICs of tigecycline for E. coli groups with different cephalosporin mechanisms, including CTXM and non-CTX-M ESBLs, ranged from 0.12 to 0.5 mg/L. Among the 235 Klebsiella spp. isolates, 127 were susceptible (MIC range 0.125–1 mg/L) to tigecycline, 35% (n = 86) were intermediate and 9% (n = 22) were resistant, with MICs ranging from 4 to 16 mg/L. All the Klebsiella oxytoca isolates hyperproducing K1 enzyme were susceptible. The modal MIC for all these Klebsiella groups was 2 mg/L corresponding to intermediate. MICs for the Enterobacter spp. (n = 126) ranged from 0.125 to 16 mg/L with 85% (n = 108) indicating susceptibility and 46% (n = 58) falling on the modal value of 0.5 mg/L, which did not vary among groups with different cephalosporin resistance mechanisms: only 7 counted as intermediate and 11 as resistant. It is instructive to compare these distributions, for multiresistant isolates, with those for sequential Enterobacteriaceae from the 2004 BSAC bacteraemia surveillance. This survey found the same modal tigecycline MICs for E. coli and Enterobacter spp. as here (0.5 mg/L). However, the modal MIC for Klebsiella spp. was also 0.5 mg/L in the BSAC survey, two dilutions lower than found here. Moreover, none of the Klebsiella spp. (n = 241) or Enterobacter spp. (n = 206) from the BSAC surveillance was resistant to tigecycline, compared with 22/235 and 11/126 of those tested here (P < 0.05 in each case). Such data suggest that at least some of the factors accumulated by some multiresistant isolates have an effect against tigecycline. In summary, tigecycline had good activity against most ESBL-producing and AmpC-hyperproducing Enterobacteriaceae, and may be a therapeutic alternative to carbapenems in some infections caused by ESBLand AmpC-producing isolates, many of which are also multiresistant to quinolones, aminoglycosides and classical tetracyclines. Limits are, however, that many infections with ESBL producers occur in the urinary tract whereas tigecycline has largely biliary excretion with low urinary recovery and that, as found here, a substantial minority of ESBL-positive Klebsiella spp. and Enterobacter spp. (though very few E. coli) do have intermediate resistance to the compound.