Risk factors associated with extended-spectrum beta-lactamase-producing organisms at a tertiary care hospital.

BACKGROUND

In 1995, beta-lactam inhibitor combinations replaced third-generation cephalosporins as empirical therapy in an effort to manage extended-spectrum beta-lactamase (ESBL) resistance. This study investigated the relationship between antibiotic usage and ESBL organisms from 1994 through 2002 using epidemiological and molecular analysis.

METHODS

A case-control study of 119 patients with ESBL organisms and 132 patients with non-ESBL organisms was conducted. Demographics, co-morbidities, device utilization and antibiotic use were analysed for all patients and infected patients only (cases = 75, controls = 83). Both exposure and degree of exposure (in grams) to antibiotics were included. A dot blot hybridization technique was used to identify genes in plasmid extracts from the ESBL organisms.

RESULTS

Ventilator days OR 1.1 (1.06, 1.15) P < 0.001, adult respiratory distress syndrome (ARDS) OR 3.1 (1.0, 9.7) P = 0.05, prior aminoglycoside use OR 2.7 (1.2, 6.1) P = 0.02, prior third-generation cephalosporin use OR 7.2 (2.6, 20) P < 0.001, and prior trimethoprim/sulfamethoxazole use OR 8.8 (3.1, 26) P < 0.001 were significantly associated with ESBL organisms by multivariate analysis. All models were concordant with a significant association of ventilator days, third-generation cephalosporins and trimethoprim/sulfamethoxazole with ESBL organisms. beta-Lactamase inhibitor combinations were not associated with ESBL organisms. Hybridization of plasmid extracts demonstrated that 95% of the ESBL organisms carried intI1, a mobile DNA element with a sulphonamide-resistance (R) gene and a frequent carrier of other R factors. Genes for specific types of trimethoprim-R and aminoglycoside-R were present in 26% and 40% of the extracts, respectively.

CONCLUSIONS

These data indicate that, besides patient risk factors and third-generation cephalosporins, other antibiotics may provide selective pressures in maintaining ESBL organisms due to multiple resistance genes on plasmids. beta-Lactamase inhibitor combinations appear to be an acceptable substitute to third-generation cephalosporins in strategies to control ESBL organisms.