Beta-lactam resistance profiles in urinary tract infection among Escherichia coli isolates in Bloemfontein

English: This study was designed to elucidate the epidemiology, nature and extent of β-lactam resistance in urinary tract infections caused by Escherichia coli isolates in Bloemfontein hospitals. To reach this goal it was necessary to phenotypically characterise and re-identify the E. coli isolates by the Mastascan identification system. Pure cultures were obtained by streaking single colonies onto MacConkey agar containing 50 µg / ml ampicillin. Single colonies were then picked off and inoculated into Mueller-Hinton broth, grown overnight at 37°C and re-streaked onto MacConkey agar containing 50 µg / ml ampicillin. Three colonies were then picked and stored at -20°C in a freeze mixture of 10 % proteose and 10 % glycerol. E. coli isolates were mated to a universal recipient strain (162) in pre-warmed Mueller- Hinton broth. Transconjugants were selected on MacConkey agar supplemented with 50 µg / ml ampicillin and 50 µg / ml nalidixic acid. Lactose-negative colonies resistant to nalidixic acid and ampicillin were inoculated into Mueller-Hinton broth, incubated for six hours and restreaked onto MacConkey agar containing ampicillin. Lactose-negative colonies were picked and considered to be transconjugants. It was found that fifty four (45 %) out of 120 ampicillin-resistant isolates transferred ampicillin-resistance determinants to an E. coli recipient (162) by conjugation. Seventy-five percent of ampicillin-resistant isolates were from female patients, indicating that urinary tract infections are more prevalent in females than in males. The National Committee for Clinical Laboratory Standards* agar dilution method was used to determine minimum inhibitory concentration (MIC) distributions of 12 β-lactam antimicrobial agents (ampicillin, amoxycillin, piperacillin, augmentin, cefoxitin, cefotaxime, cefepime, ceftazidime, cephazolin, ceftriaxone, cefuroxime and imipenem). Strains found to be resistant had MICs that overlapped the range where susceptibility was normally assumed. This was due to inducible β-lactamase producer strains, which go undetected by the Kirby- Bauer disk diffusion method but can be identified by using the Jarlier double-disk method. MIC frequency distributions for the penicillins showed that elevated doses should be administered in order to maximise antibiotic concentration at the site of infection or that a second antibiotic agent or inhibitor should be used in combination therapy. Beta-laetam susceptibility profiles were determined by the Kirby-Bauer disk diffusion method. This method was also used to determine the correlation of MIC values with the inhibition zone diameters in order to predict treatment outcomes from inhibition zone diameters. The Jarlier double-disk technique was used to detect extended-spectrum β-lactamaseproducing organisms and the frequencies at which they occurred. There was no difference between the ratios of ESBL-producers in hospitalised and non-hospitalised patients, although the absolute numbers were different. This was probably due to the 48 hour cut-off point used to define hospitalisation. Samples taken before 48 hours were considered to be nonhospitalised. There were many more female than male patients with urinary tract infections in the Bloemfontein hospitals during the period of the study. The extent of joint resistance to β-lactam antibiotics among E. coli isolates was assessed by comparing two agents at a time. The observed incidence of joint resistance was compared to the rate of double resistance expected if it had been acquired as two independent events. It was found that even amongst two antibiotics that are biologically cross-resistant (ampicillin and augmentin) a close correlation exists between the concentrations of the two agents required to inhibit individual E. coli strains.