Asymptomatic bacteriuria in school childrenPV Badami, Leena P Deodhar
Department of Pathology and Microbiology, Seth G. S. Medical College, Parel, Bombay-400 012., India
1000 school children between the ages of 5 and 10 years were screened for asymptomatic bacteriuria. These were healthy children attending the school regularly. There were 120 girls and 810 boys. Dip slide method was used for urine cultures and Griess nitrite test was also carried out on all urine samples. Four out of 190 girls and 7 out of 810 boys showed asymptomatic bacteriuria giving an incidence of 2.1% for school girls and 0.8% for school boys and an overall incidence of 1.1% for all school children. E. colt was the commonest organism isolated followed by Klebsiella aerogenes. Results with the Griess nitrite test were discouraging. Plain X-rays of abdomen did not reveal any abnormality of urinary tract. Four out o f 11 cases were followed up for a period o f 6 months. None of these developed overt renal disease in the follow up period.
Infections of the urinary tract, overt and occult, are probably the commonest bacterial infections in our civilisation surpassed in frequency only by the viral infections of the respiratory tract. In contrast to the viral infection of the respiratory tract which is self limited, urinary tract infections are often not so, and tend to persist in an asymptomatic form to be followed weeks, months or even years later by symptomatic bouts of cystitis and pyelonephritis.
Incidence of asymptomatic bacteriuria in school children is reported to be between 1.5 to 2.5 %,  the figures being slightly higher in school girls. These girls who are shown to have asymptomatic or isolated episodes of bacteriuria in the past, always remain at high risk of developing symptomatic infection at a later stage.
Thus, asymptomatic bacteriuria in school children is a significant problem since this is not associated with burnt out pyelonephritis but with active disease process in kidney. Hence, if left undetected and untreated, these children can develop bouts of cystitis and pyelonephritis and many may go into renal failure at a later age.
A screening of 1000 healthy school children was carried out for asymptomatic bacteriuria. This included 525 school children attending K.E.M. Hospital Clinic for regular checkup and 475 boys studying in the private school located near the K.E.M. Hospital. All the children were between the ages 5 and 10 years, apparently healthy and attending the school regularly. Details of past history of any illness like fever, burning micturition, lumbar pain suggesting a urinary tract infection were noted down.
A clean-catch midstream specimen of urine was collected after cleaning the external genitals with detergent solution first and then with distilled water. The urine samples were transferred to the laboratory within 15 minutes of collection and were subjected to following procedures.
1. Within 30 minutes of collection, samples were inoculated on to dip slides (The method used was the one devised by Guttman and Naylor  ). The results were read next morning by comparing colony density on dip slides, with standard photographs of dip slides showing growths corresponding to 10 3 , 10 4 , 10 5 and 10 6 colonies/ml. of sample.
2. Griess nitrite test was performed on all urine samples (Method used was similar to that used by Thysell 9 ).
3. Urine samples were preserved at 4°C. Whenever dip slides or nitrite test yielded a positive result, the corresponding sample was subjected for colony count by pourplate method. 
4. The pH of the urine samples and the microscopic findings of the centrifuged deposit were noted down.
Final identification of the organisms isolated from urine samples was based on morphology of the organisms in gram's staining, motility of the organisms, colony characters on MacConkey's medium and various biochemical reactions routinely used for identification of gram negative and gram positive organisms.
Antibiotic sensitivity testing of all these isolates was carried out by disc diffusion method.
To prepare photographic standards a urinary strain of E. coli was chosen and the dip slides (covered with the medium -nutrient agar and MacConkey's agar) innoculated with dilutions containing 10 2 to 10 5 and above of E. coli/ml were photographed [Figure 1].
5. Urine samples of the above children showing significant bacteriuria  (i.e. presence of one lac or more pathogenic bacteria per mililitre of freshly voided midstream urine) were repeated at weekly intervals for colony counts.
6. Children showing significant bacteriuria were subjected to plain X-ray examination of the abdomen.
Out of 1000 school children, there were 190 school girls and 810 boys. Of the 190 school girls 4 showed significant colony counts. None of these four girls gave any past history of fever, burning micturition or lumbar pain suggestive of overt urinary infection. In all these cases, quantitative urine cultures were done at weekly intervals and each time the sample showed significant colony counts and the same organism was isolated each time.
Of the 810 boys screened for bacteriuria, only 7 showed significant counts. Quantitative culture was repeated in three of these cases and thrice in remaining four cases. On second and further cultures also, significant counts were obtained and precisely the same organism was isolated.
In another 22 cases, the first culture yielded a colony count between 10 4 -10 5 /ml. of urine sample. However a repeat colony count after one week in these cases did not yield significant counts in any of these samples.
Griess nitrite test was positive only in 5 of the 11 cases.
Routine urine examination showed trace of albumin and pyuria (more than 8-10 pus cells per high power field) in only 2 of the 11 cases. Urinary pH was acidic in all except one of the 11 cases. In this case, where urinary pH was alkaline the organism isolated was Proteus mirabilis.
Distribution of organisms isolated was as follows:[Table 1]
Antibiotic sensitivity testing showed results as shown in [Table 2].
Only 4 cases were available for follow up. All these 4 boys were given a course of antibiotic as guided by antibiotic sensitivity tests and quantitative culure was repeated after the antibiotic course. Three of the four samples were sterile; and in one, colony counts of 10m organisms/ml of the same pathogen was obtained. In 6 months follow up period none of these cases developed any sympttoms of overt urinary tract infection.
Plain X-rays of kidney-ureter-bladder region were taken in 4 cases but no abnormality was detected in any of these cases.
The contribution of significant bacteriuria to the pathogenesis of pyelonephritis in childhood has been under intense investigation during the past 15 years. A distinct pattern has now begun to emerge which should prove useful in providing a framework for assessing the magnitude of the problem and a rational approach to the management. The ultimate aim is both to reduce the considerable morbidity associated with urinary infections and to prevent renal disease.
In the present study, the criterion of bacteriuria of 1 lac or more organisms per ml. of urine was applied to distinguish between true bacteriuria and contamination. For the quantitative culture of all the urine samples that were screened, the dipslide technique devised by Guttman and Naylor  was employed. This is a very simple, inexpensive and reliable method which has been extensively tested by investigators like Cohen and Kass  and Boothman et al  and has been found to give very good correlation with pourplate counts. It was also easy to differenciate between true bacteriuria and contamination since both nutrient agar and MacConkey's agar media were used. Also the isolation of the organisms and their identification was rendered more easy.
In the present study, all urine samples were also subjected to Griess nitrite test. Only 5 out of 11 samples showing significant colony counts gave a positive test. Results of the nitrite test as reported by other authors like Thysell  also point out that this test showed low sensitivity and only 27 out of 74 urine specimens in his study with more than 10 bacteria per ml gave positive results with nitrite test. There are number of other chemical tests for detection of bacteriuria like triphenyl tetrazolium chloride test, catalase test and subnormal urinary glucose concentraion; but none of the chemical tests for bacteriuria appear reliable enough to be employed in mass screening of population for bacteriuria.
Thus, in our series overall incidence of bacteriuria in school children was 1.1%. Out of 190 girls there were 4 cases of asymptomatic bacteriuria giving an incidence of 2.1% for school girls. None of the 4 girls were available for follow up. Kunin  has reported an incidence of 1.2% in school girls in his series.
Our figure of 2.1% compares well with that of Newcastle Asymptomatic Bacteriuria Research Group  ; their figure being 2.5%. Mathur et al  reported 4% incidence of bacteriuria in 50 children attending Pediatric outpatient for complaints other than those referable to urinary tract.
Of the 810 boys screened for bacteriuria 7 showed consistently high colony counts on all three cultures, giving an incidence of 0.8%. Our figure is higher as compared to that reported by Kunin  (0.03%) and Newcastle Asymptomatic Bacteriuria Research Group  (0.2%).
In our series there was no significant difference in the incidence of bacteriuria in the two socio-economic groups. Similar are the findings of other workers. ,
Only 2 out of 11 bacteriuric children showed pyuria, with more than 10 leucocytes per high power field. Thus only 20% of the bacteriuric children had pyuria: Newcastle Asymptomatic Bacteriuria Research Group  found a 44% incidence of pyuria in 254 bacteriuric children.
Radiologic findings did not reveal any abnormality in the present series though the studies were limited to only plain X-ray of abdomen.
E. coli was the commonest pathogen (45.5%) that was isolated followed by Klebsiella in 27.2% of cases. In Kunin's  series E. coli was isolated in 71% of cases and in Newcastle Asymptomatic bacteriuria Group series  in 91.6% of cases. In both the series Klebsiella was the next commonest pathogen followed by Staphylococcus, Proteus and Pseudomonas.
We are grateful to the Dean, Seth G. S. Medical College and K.E.M. Hospital, Bombay and also the Principal, R. M. Bhat High School, Parel, for allowing us to use the material.
[Table 1], [Table 2]