Summary 1. The material consists of 5% consecutive cases of primary (first onset) urinary tract infections appearing from birth up to 16 years of age and which were examined and treated at the Childrens' Hospital in Goteborg. The infections occurred during a seven‐year period within a defined population. The circumstances under which the study was conducted suggest that most symptomatic infections occurring during the study period and for whom the parents sought medical advice, were 2. The total morbidity risk at I 1 years of age of symptomatic UTI was 3.0% for girlsand 1.1 % for boys. These are minimum figures. The morbidity risk is highest during the first month of life and then decreases, more rapidly in boys than in girls. Possible interpretations of the reason for decreasing risk with increasing age of falling ill with a first infection are suggested. The male/female ratio starts at 2.5: 1 during the first month and then successively changes to 1:20.There was no seasonal variation of the time of onset in either sex. 3. Presentation with fever was most common in the first year, after which it slowly decreased. Failure to thrive was a rare symptom. Certain pther age and sex differences in presenting symptoms were recorded. Most infections within the first year of life probablyinvolved the renal parenchyma. 4. The etiology varied with age and sex. If infections reach the urinary tract by theascending route, this could indicate differences in the environmental conditions in the periurethral region and may be a clue to a better understanding of the pathogenesis. 5. Obstructive malformations were found in 10% of boys and 1–2% of girls, and cannotexplain the high frequency of early infantile infections in either sex. 6. Narrowing of the bladder neck was common in males during the first year of life, the frequency declining with age. It disappeared spontaneously during follow‐up of individual cases, and was not regarded as an obstructive malformation. 7. Duplication of the collecting system was seen in 10 % of girls and in 5 % in boys, which is more than expected. The cause and nature of the association between infection and duplication are not known. 8. In 13 % of boys and 4.5 % of girls a reduction of the renal parenchyma was seen either at the first investigation or developed later, probably owing to infection. AlthoughUTI was more frequent in females than in males, the total number of patients with parenchymal damage was equal in both sexes, even during childhood. In boys, the kidneys might bemore vulnerable than in girls. 9. The immediate cure rate after 10 days' therapy was 97%. Recurrences were usually reinfections. 10. Recurrent infections were often difficult to diagnose. Pyuriaand symptoms of UTI were associated with insignificant bacteriuria in 10 % (30 of 300) of suspected recurrences. 11. Susceptibility to recurrence was studied in relation to various parameters. Girlswere at greater risk than boys, and the risk was in both sexes greatest during the first 2–3 months after a previous infection. Boys rare...

The role of bacterial adherence in the persistence of bacteria in the human urinary tract was analyzed. Women with chronic symptomatic urinary tract infections were subjected to deliberate colonization with nonvirulent Escherichia coli, after eradication of their current infections. E. coli organisms were instilled into the bladder through a catheter. The strain used for colonization, E. coli 83972, was isolated from a patient with stable bacteriuria. It lacked expressed adherence factors and did not belong to the uropathogenic O:K:H serotypes. Strain 83972 was transformed with the pap and pil DNA sequences encoding Galcl-4Galpand mannose-specific (type 1) adhesins. Patients were colonized with a mixture of the wild-type and the transformed strains. E. coli 83972 caused stable bacteriuria for >30 days in 7 of 12 individuals. In contrast, the Galal-4Galjp-recognizing or mannose-binding transformants were eliminated within 48 h. The consistent superiority of the wild-type strain in establishing stable bacteriuria when compared with the adhesive transformants did not appear to be due to differences in growth rates or to plasmid segregation. Rather, the transformants expressing the adhesin determinants were selectively eliminated by the host. This suggested that the acquisition of adherence factors is not sufficient to increase the fitness of E. coli for survival in the urinary tract of humans.

As no method for localization of urinary tract infection has been shown to be absolutely reliable, six procedures have been carried out simultaneously in 25 girls with acute symptomatic infections and the reliability of each method assessed. While clinical diagnosis of pyelonephritis or cystitis correlated well with the overall results of the battery of tests, the reliability of individual tests varied. Highest reliability was obtained with CRP determinations followed by antibody titration, sedimentation rate, and renal concentrating capacity. In the bladder washout test only 8 of the 14 patients with pyelonephritis had findings clearly indicating high infection. Intermittent or inadequate discharge of bacteria from the renal parenchyma is suggested as the major source for this inaccuracy. In fact, half of the 42 final washout specimens from girls with acute pyelonephritis contained less than 1000 bacteria per ml, indicating that low numbers of organisms in ureteric urine is common in childhood pyelonephritis.

The homogeneity of fecal coliform flora in 52 schoolgirls was studied by serotyping and biotyping 10 randomly selected colonies in one fecal culture from each child. Ninety-eight clones were identified and of these 52 were dominant and 46 were minor strains. The probability of including at least one isolate of the dominant clone in a small random sample of colonies was calculated to be 86% for one colony, 94% for 2, 97% for 3, 99% for 4, and 99.3% for 5 randomly selected colonies.

Objective-To investigate the effects of phenoxymethylpenicillin and erythromycin on urinary isolates from patients with untreated asymptomatic bacteriuria.Design-Retrospective study of subgroup of patients from cohort followed up till the end of 1986.Setting-Outpatient clinic for children with urinary tract infections. Patients-51 Girls aged under 15 with untreated asymptomatic bacteriuria.Interventions-Before 1982 intercurrent infections (mostly tonsillitis or otitis) were treated with phenoxymethylpenicillin; after 1982 erythromycin treatment was preferred.End points-Change of bacterial strain in urinary tract and symptomatic recurrences of disease.Measurements and main results -Bacteria identified by serotype and electrophoretic type and compared before and after antibiotic treatment. Bacteriuria eradicated and replaced by new strains in most patients treated with phenoxymethylpenicillin, leading to symptomatic recurrences in about 15%. Conversely, patients given erythromycin rarely showed change in bacteriuria and none suffered symptomatic recurrence.Conclusions-In girls with untreated asymptomatic bacteriuria the use of phenoxymethylpenicillin for intercurrent infections may lead to a change of urinary bacteria and leave them at substantial risk of acute pyelonephritis. With erythromycin this risk is small (2/20 courses in this series).