Increased Amoxicillin–Clavulanic Acid Resistance in Escherichia coli Blood Isolates, Spain

To determine the evolution and trends of amoxicillin–clavulanic acid resistance among Escherichia coli isolates in Spain, we tested 9,090 blood isolates from 42 Spanish hospitals and compared resistance with trends in outpatient consumption. These isolates were collected by Spanish hospitals that participated in the European Antimicrobial Resistance Surveillance System network from April 2003 through December 2006.


The Study
The 42 participating Spanish hospitals were selected according to EARSS criteria (1,7). The total catchment population was ≈9 million people, or ≈22.5% of the Spanish population. The fi rst blood E. coli isolates obtained from each patient between 2003 and 2006 were included. Each laboratory identifi ed the strains and tested their susceptibilities according to standard microbiologic procedures; all used commercial microdilution systems. Susceptibility data were interpreted according to Clinical Laboratory Standards Institute criteria (8). For epidemiologic purposes, intermediate susceptibility to AMC was considered as resistance. Multidrug resistance was defi ned as resistance to >3 of the following antimicrobial agents: ciprofl oxacin, gentamicin, cotrimoxazole, and cefotaxime. To assess the comparability of susceptibility test results, an external quality assurance exercise (UK National External Quality Assessment Scheme) was performed yearly.
Hospital-acquired infections were defi ned as infections acquired at least 48 hours after hospital admission. Community-acquired infections were those in which E. coli-positive cultures were identifi ed at or within 48 hours of hospital admission.
Outpatient consumption of penicillin/β-lactamase inhibitors (World Health Organization code J01CR02) for the period 2002-2006 was assessed from the Especialidades Consumo de Medicamentos database, which showed retail pharmacy sales of all medicines acquired with National Health System prescriptions and covered nearly 100% of the Spanish population (5). The information was tabulated, and the number of units was converted into defi ned daily doses (DDD) of active drug ingredients according to WHO methodology (9). The number of DDD per 1,000 inhabitants per day (DIDs) was calculated for each active drug ingredient. were obtained from patients >15 and <64 years of age; and 5,909 (65%) were obtained from patients >64 years of age. There were 3,384 (37.9%) isolates implicated in hospitalacquired infections and 5,540 (62.1%) in community-acquired infections; information was missing for 166 cases.
The overall rate of invasive E. coli nonsusceptibility to AMC increased from 9.3% (2003) (Figure 1). Community-acquired infection probably included healthcare-associated infections, a recently described epidemiologic category distinct from both community-acquired and nosocomial status.
In this study, the number of blood isolates of E. coli producing extended-spectrum β-lactamase (ESBL) was 614 (6.7%); 188 of them (30.6%) were nonsusceptible to AMC. When ESBL-producing E. coli isolates were excluded from analysis, AMC nonsusceptibility increased from 8.4%

Conclusions
The increased AMC resistance of E. coli isolates from blood observed in this study is of serious concern from clinical and epidemiologic standpoints because AMC is the fi rst-choice antimicrobial treatment for many invasive E. coli infections. Increased AMC resistance coincided with growing AMC consumption at the community level. In urinary infections, previous treatment with AMC is a risk factor for the development of AMC resistance (10). AMC resistance mechanisms (β-lactamase overproduction, AmpC cephalosporinase hyperproduction, and inhibitor-resistant penicillinases) (11)  Dr Oteo is a specialist in medical microbiology. He works in the Centro Nacional de Microbiología of the Ministry of Health, Madrid, Spain. His primary research interest is the molecular basis and surveillance of bacterial resistance to antimicrobial agents.