Recurrent Sphingomonas paucimobilis -bacteraemia associated with a multi-bacterial water-borne epidemic among neutropenic patients.

A cluster of septicaemias due to several water-related species occurred in a haematological unit of a university hospital. In recurrent septicaemias of a leukaemic patient caused by Sphingomonas paucimobilis, genotyping of the blood isolates by use of random amplified polymorphic DNA-analysis verified the presence of two distinct S. paucimobilis strains during two of the separate episodes. A strain of S. paucimobilis identical to one of the patient's was isolated from tap water collected in the haematological unit. Thus S. paucimobilis present in blood cultures was directly linked to bacterial colonization of the hospital water system. Heterogeneous finger-printing patterns among the clinical and environmental isolates indicated the distribution of a variety of S. paucimobilis clones in the hospital environment. This link also explained the multi-microbial nature of the outbreak.

Recurrence of urinary tract infections in adult patients with community-acquired pyelonephritis caused by E. coli: a 1-year follow-up.

In a prospective study, 42 women were followed for recurrence of urinary tract infections (UTIs) for 1 y after an index episode of community-acquired pyelonephritis caused by Escherichia coli. Altogether, 26 repeat episodes were detected. Of these, 20 occurred at least 1 month after the index episode and were regarded as recurrences. In all, 40%, (17 of 42) of the women had recurrences. An earlier history of UTI increased the risk of recurrence: 52%, of the 29 women with previous UTI had at least 1 recurrence, compared with 15%, of the 13 patients without previous UTI. E. coli caused the majority (73%) of the recurrences. Genotype comparisons by RAPD-PCR analysis between E. coli isolates from a patient showed that 75%. of the original and recurrent strains were genetically non-identical. Of the 54 E. coli strains, 42 were carrying genes coding for G adhesins of P fimbriae: 40 isolates carried class II, I class III and 1 carried both class II and III G adhesin genes. Each of the virulence-associated factors (genes for G adhesins, MRHA, haemolysin, type 1C fimbriae, and O and K antigens) was evenly distributed among E. coli isolates of index episodes, independent of the recurrences. The index isolates, however, had more virulence-associated factors than did the isolates from the recurrences which were mainly due to lower UTIs.

Low virulence of Escherichia coli strains causing urinary tract infection in renal disease patients.

The distribution of urinary bacterial species was determined and the virulence factors of Escherichia coli urinary strains analysed by molecular and phenotyping methods in episodes of urinary tract infection in renal disease patients (n =68) in comparison with other immunocompromised patients (n =59) and non-immunocompromised patients (n =21). Escherichia coli was isolated in 116 (78%) of the 148 patients, being the species most frequently isolated in all groups (75% of renal disease patients, 76% of other immunocompromised patients, 95% of non-immunocompromised patients). All other pathogens showed a similar distribution in the renal disease and other immunocompromised patient groups. All virulence factors of Escherichia coli tested for (genes for G adhesins, expression of MR adhesins, production of haemolysin, presence of certain O and K antigens) were found more often in non-immunocompromised than in immunocompromised patients. The factors allowing the highest degree of discrimination between immunocompromised and non-immunocompromised patients were the prevalence of genes for G adhesins (35% vs. 65%) and expression of MR adhesins (32% vs. 55%). It is concluded that there is a lower prevalence of G adhesins and MR adhesins in Escherichia coli strains from immunocompromised patients than non-immunocompromised patients, suggesting that less virulent Escherichia coli strains may cause urinary tract infections more frequently in renal disease patients and other immunocompromised patients. Moreover, the spectrum of urinary pathogens other than Escherichia coli is similar in both immunocompromised patient groups investigated.

Random amplified polymorphic DNA (RAPD) analysis of Escherichia coli strains: comparison of urinary and concomitant blood isolates of urosepsis patients.

The discriminatory power of random amplified polymorphic DNA (RAPD) analysis was assessed for detection of intraspecies variation in Escherichia coli strains of clinical origin. Three primers (OPF 5, OPF 7 and OPF 8) were preselected from commercial 10-mer primers by the number of distinct bands obtained. These primers were used in testing 26 urinary and 13 blood isolates from 26 patients and E. coli ATCC 25922, OPF 5, OPF 7 or OPF 8 alone separated the strains into 15 to 21 RAPD types. A combination of the results of the three primers gave 25 RAPD types. When blood and urine isolates of each patient were analysed in parallel, all blood-urine pairs were found identical, and with one exception they were also unique. RAPD analysis had a high discriminatory power. It separated the strains equally well or better than ribotyping, and obviously better than serotyping which grouped the urine strains into 8 serogroups leaving 18 strains untypable or incompletely typed. Thus, to verify the identity or non-identity of isolated E. coli strains, RAPD analysis was shown to be a sensitive and reproducible technique which is technically less demanding, more rapid and more economical than either serotyping or ribotyping. However, in its present application, this technique cannot fully replace determination of the serotype or virulence factors which may show correlations with different manifestations of infection.