Antone A. Medeiros

A classification scheme for b-lactamases based on functional characteristics is presented. Three major groups of enzymes are defined by their substrate and inhibitor profiles: group 1 cephalosporinases that are not well inhibited by clavulanic acid; group 2 penicillinases, cephalosporinases, and broadspectrum b-lactamases that are generally inhibited by active site-directed b-lactamase inhibitors; and the group 3 metallob-lactamases that hydrolyze penicillins, cephalosporins, and carbapenems and that are poorly inhibited by almost all b-lactam-containing molecules. Functional characteristics have been correlated with molecular structure in a dendrogram for those enzymes with known amino acid sequences. b-Lactamases (EC 3.5.2.6) have been designated by the Nomenclature Committee of the International Union of Biochemistry as ‘‘enzymes hydrolysing amides, amidines and other CON bonds . . . separated on the basis of the substrate: . . . cyclic amides’’ (323). These enzymes are the major cause of bacterial resistance to b-lactam antibiotics and have been the subject of extensive microbiological, biochemical, and genetic investigations. Investigators have described more than 190 unique bacterial proteins with the ability to interact with the variety of b-lactam-containing molecules that can serve as sub-

beta-Lactamases are the principal mechanism of bacterial resistance to beta-lactam antibiotics. In recent years the number and variety of new beta-lactamases detected has risen at an alarming rate, apparently in response to the clinical use of novel classes of beta-lactam antibiotics. This paper reviews the structure and evolution of beta-lactamases in an attempt to understand the pressures that have contributed to their emergence.

Enterococcus faecium strains resistant to ampicillin, high levels of gentamicin, and vancomycin but susceptible to teicoplanin (vanB class vancomycin resistance) were recovered from 37 patients during an outbreak involving a 250-bed university-affiliated hospital. Three isolates with vancomycin MICs ranging from 8 to 256 micrograms/ml all hybridized with a vanB probe. Restriction endonuclease analysis of chromosomal and plasmid DNA suggested that all isolates tested were derived from a single clone. Vancomycin resistance was shown to be transferable. Risk factors for acquiring the epidemic strain included proximity to another case patient (P, 0.0005) and exposure to a nurse who cared for another case patient (P, 0.007). Contamination of the environment by the epidemic strain occurred significantly more often when case patients had diarrhea (P, 0.001). Placing patients in private rooms and requiring the use of gowns as well as gloves by personnel controlled the outbreak. These findings suggest that multidrug-resistant E. faecium strains with transferable vanB class vancomycin resistance will emerge as important nosocomial pathogens. Because extensive environmental contamination may occur when affected patients develop diarrhea, barrier precautions, including the use of both gowns and gloves, should be implemented as soon as these pathogens are encountered.

The rate of macrolide resistance among Streptococcus pneumoniae is increasing, but some investigators have questioned its clinical relevance. We conducted a matched case-control study of patients with bacteremic pneumococcal infection at 4 hospitals to determine whether development of breakthrough bacteremia during macrolide treatment was related to macrolide susceptibility of the pneumococcal isolate. Case patients (n=86) were patients who had pneumococcal bacteremia and an isolate that was either resistant or intermediately resistant to erythromycin. Controls (n=141) were patients matched for age, sex, location, and year that bacteremia developed who had an erythromycin-susceptible pneumococcus isolated. Excluding patients with meningitis, 18 (24%) of 76 case patients and none of 136 matched controls were taking a macrolide when blood was obtained for culture (P=.00000012). Moreover, 5 (24%) of 21 case patients with the low-level-resistant M phenotype and none of 40 controls were taking a macrolide (P=.00157). These data show that development of breakthrough bacteremia during macrolide or azalide therapy is more likely to occur among patients infected with an erythromycin-resistant pneumococcus, and they also indicate that in vitro macrolide resistance resulting from both the efflux and methylase mechanisms is clinically relevant.