Search

There is urgent need to develop novel treatment strategies to reduce antimicrobial resistance. Collateral sensitivity (CS), where resistance to one antimicrobial increases susceptibility to other drugs, might enable selection against resistance during treatment. However, the success of this approach would depend on the conservation of CS networks across genetically diverse bacterial strains. Here, we examine CS conservation across diverse Escherichia coli strains isolated from urinary tract infections. We determine collateral susceptibilities of mutants resistant to relevant antimicrobials against 16 antibiotics. Multivariate statistical analyses show that resistance mechanisms, in particular efflux-related mutations, as well as the relative fitness of resistant strains, are principal contributors to collateral responses. Moreover, collateral responses shift the mutant selection window, suggesting that CS-informed therapies may affect evolutionary trajectories of antimicrobial resistance. Our data allow optimism for CS-informed therapy and further suggest that rapid detection of resistance mechanisms is important to accurately predict collateral responses.

The degree of correspondence between urinary and faecal Escherichia coli O-groups has been assessed in non-pregnant women with symptomatic urinary-tract infection or asymptomatic bacteriuria. In 20 of 26 patients with symptomatic urinary tract infection E. coli of the same O-groups as that of the urinary infecting strain was also present in the patient's faecal flora of only five of 25 patients with asymptomatic bacteriuria. This finding indicates that the majority of episodes of symptomatic urinary tract infection in non-pregnant women are not preceded by a significant period of asymptomatic bacteriuria. E. coli O6 showed correspondence between urinary and faecal isolates more frequently than did other O-groups, but it had a relatively low prevalence in the faecal flora of patients with urinary-tract infection caused by E. coli of other O-groups. This finding lends support to previous suggestions that E. coli O6 may be especially pathogenic for the urinary tract.

The distribution of K1, K5 and O antigens was examined in 500 clinical strains of Escherichia coli. Of 400 strains from urine, 52% belonged to serogroups O1, O2, O4, O6, O7, O8, O9, O18, O25 and O75; 34% were non-typable (NT) and 14% were autoagglutinable (AA). Antigen K1 was carried by 17% of these strains, and K5 by 15%. The numbers of O-serogroupable, NT and AA strains among 100 strains from blood were 62, 29 and 9, respectively. K1 antigen was detected on 20% of isolates from blood and K5 on 13%. There was no statistically significant difference in the distribution of K1, K5 or O antigens between strains from blood compared with those from urine. The occurrence of K1 and K5 antigens among smooth and AA strains suggested that AA strains were derived primarily from the common O-serogroups.

The incidence of multiple-antibiotic resistance among Escherichia coli isolated from urinary tract infections at Charing Cross Hospital, London, increased over the last 10 years, and its distribution was related to O-type. Among strains of the eight O-types most frequently causing such infections at this hospital, O4, O9, and O18 had a high incidence of multiple resistance (35, 22, and 19%, respectively); O2 and O6 had a intermediate incidence (14 and 11%, respectively); and O7, O1, and O75 had a low incidence (8, 6, and less than 3%, respectively). This nonrandom distribution appears to be a consequence of unequal plasmid recipient ability. After overnight mating with antibiotic-resistant donors, R-plasmid infection frequencies among antibiotic-sensitive urinary tract isolates differed by up to 10(5)-fold, and such differences were correlated with the variation in the incidence of antibiotic resistance among the O-types. The inherent differences in the ability to achieve significant R-plasmid spread, which appear to be determined by the host, not the plasmid, may be compounded in some cases by the inhibition of potential mating partners by colicin production.

One thousand bacterial isolates from consecutive suprapublic aspirations of urine in two groups of patients with symptomatic urinary tract infection and two groups with asymptomatic bacteriuria were studied. Staphylococci accounted for 13 and 27% of the isolates from symptomatic patients but for only 3 and 4% of strains from asymptomatic bacteriuria. Conversely, 18 and 30% of Escherichia coli isolates from asymptomatic bacteriuria were auto-agglutinable when tested with antisera to the common urinary O-serogroups, whereas such strains accounted for only 5 and 7% of the E. coli from symptomatic patients. The high prevalence of auto-agglutinable E. coli in asymptomatic bacteriuria was accompanied by a reduction in the frequency of the common uninary O-serogroups rather than other smooth strains.

The sensitivities to cefuroxime and cephaloridine were compared in 500 bacterial isolates from consecutive cases of urinary tract infection. In the first 200 organisms there was generally good correlation between the diameters of the zone of inhibition around a 30 microgram cefuroxime disc and the MICs. Correlation was poorest with streptococci. From the preliminary observations, criteria were derived to relate the disc sensitivities to MICs for all 500 strains. Neither cephalosporin had significant activity against Pseudomonas spp. but, apart from one strain of Klebsiella and one methicillin-resistant staphylococcus, the only bacteria sensitive to cephaloridine but resistant to cefuroxime were Str. faecalis. Of the 500 organisms, 45% were beta-lactamase-producing Gram-negative bacilli and, overall, cefuroxime was more active than cephaloridine against them. The greater stability of cefuroxime to the beta-lactamases from these bacilli were also supported by a smaller increase in MIC when the inoculum size was increased. On the basis of these results, cefuroxime should prove an effective choice for the treatment of most urinary tract infections.

Gram-positive, catalase-positive cocci (Micrococcaceae) have been obtained from the bladder by suprapubic aspiration of urine in antenatal and postnatal patients. These bacteria have been compared with similar organisms isolated from midstream specimens of urine when the suprapubic specimen was sterile. Fourteen (70%) of 20 organisms isolated from the bladder were micrococci and six (30%) were staphylococci. In contrast 17 (85%) of 20 similar organisms contaminating midstream urine specimens were staphylococci and only three (15%) were micrococci.