Comparison of expression of virulence factors by Escherichia coli causing cystitis and E. coli colonizing the periurethra of healthy girls

Relatedness of Escherichia coli colonizing women longitudinally

PURPOSE

The longitudinal colonization patterns by Escherichia coli of the vaginal introitus and urinary tract were investigated.

MATERIALS AND METHODS

Cultures of the vaginal introitus and midstream urine were collected once a week for 12 consecutive weeks from five women with (patients) and five without (controls) a history of urinary tract infection (UTI).

RESULTS

A total of 63 E. coli isolates was obtained from the 10 women, 26 from controls and 37 from patients. The bacterial counts of E. coli present in control individuals were uniformly low, < or = 200 E. coli/mL. The numbers in patients were higher and more variable, reaching > 10(5)/mL in urine and vaginal specimens. In 16 instances, E. coli was present in the urine and the vaginal introitus concurrently (matched isolates). Random amplified polymorphic DNA (RAPD) fingerprinting was used to characterize all matched E. coli isolates. Concurrent vaginal and urinary tract colonization was more common in the patient population, and usually, the same E. coli strain was present at both sites; only 15% of the matched isolates represented different strains. The RAPD fingerprinting was also carried out on selected isolates recovered from four patients and three control individuals over the 12-week study period. Colonization of the vaginal introitus and urinary tract in these individuals varied over time. Generally, however, a predominant E. coli strain was present in the vaginal milieu, urinary tract, or both, either continuously (for as long as 9 consecutive weeks in one patient) or intermittently.

CONCLUSION

The results support the concept that the vaginal mucosa acts as reservoir of E. coli which may enter the urinary tract.

Expression of virulence factors among Escherichia coli isolated from the periurethra and urine of children with neurogenic bladder on intermittent catheterization

BACKGROUND

Patients with neurogenic bladder caused by spinal cord injury or myelomeningocele empty their bladder several times a day by intermittent catheterization. Bacteriuria without symptoms of infection is frequently present in these patients. Occasionally a clone of Escherichia coli that has been carried for weeks without symptoms causes a symptomatic urinary tract infection. Virulence factors are commonly expressed among E. coli causing infection in patients with normal urinary tracts. However, it is unknown whether expression of virulence factors by an E. coli clone colonizing the neurogenic bladder increases the risk of subsequent infection. In this study we examined the prevalence of virulence factor expression among E. coli isolated from the periurethra and urine of patients with neurogenic bladder.

METHODS

The prevalence of virulence factors was examined among E. coli isolated from the periurethra and urine in patients with neurogenic bladder who received intermittent catheterization and were followed for 6 months. Representative isolates from the 37 clonal types of E. coli detected in the periurethra and urine of children with neurogenic bladder were assessed for O antigen, hemolysin, aerobactin, serum resistance and type I and P-adhesin.

RESULTS

All clones were serum-resistant and expressed type I adhesin, none expressed aerobactin and two expressed hemolysin. The presence of P-adhesin was not unique to clones associated with symptomatic infection. The presence of P-adhesin carried for weeks in a clone did not predict subsequent infection in the neurogenic bladder.

CONCLUSION

Bacterial virulence factors did not predict infection of the neurogenic bladder.

The ability of periurethral Escherichia coli to grow in a voiding system is a key for the dominance of E. coli cystitis

The generally accepted concept of the pathogenesis of cystitis is that bacteria on the periurethra migrate via the urethra to the bladder urine. To explore why Escherichia coli is the dominant organism isolated from cystitis, we examined two potential mechanisms that might account for the dominance of E. coli as the etiologic agent in cystitis in prepubertal, non-sexually active girls. First, the frequency of carriage of aerobic bacteria on the periurethra of two populations-healthy girls and girls after their first episode of cystitis was determined. Second, the survival capability of periurethral bacteria in urine under voiding conditions was examined. Two voids were simulated over 6 h by 10(2) dilution. A control was run in broth instead of urine. We found that Gram-positive bacteria, particularly coagulase-negative staphylococci, were much more common on the periurethra than E. coli. E. coli did grow well in urine in an in vitro voiding system whereas most Gram-positive bacteria did not. Growth in urine despite voiding may be the key reason for the dominance of E. coli in cystitis.