Asymptomatic bacteriuria in school girls. IV Difficulties of level diagnosis and the possible relation to the character of infecting bacteria.

As no method for localization of urinary tract infection has been shown to be absolutely reliable, six methods have been run parallelly in a study of 60 school girls with asymptomatic bacteriuria caused by E. coli. A poor correlation was obtained between "high infection" indicated by bladder washout test and abnormal findings of C-reactive protein, sedimentation rate, antibody titres or renal concentrating capacity, while findings of parenchymal reduction on the pyelogram and reflux on the urethrocystogram were often found within this group. The low frequency of abnormal findings of C-reactive protein, sedimentation rate and renal concentrating capacity in girls with a "high infection" may be explained by the finding that the ABU patients were infected with changed E. coli strains probably not able to cause the usual inflammatory reactions. The patients' bacteria were found to be less efficient in providing antigen for antibody determinations than were standard strains of same O group, indicating a difference between strains from ABU patients and standard strains. From the battery of tests used it seemed that most of the girls had a bladder infection. All tests were normal in 48% of the patients while at least three of the methods were abnormal in 12%.