INDIAN J o u Vol. 42
RNAL OF P E D I A T R I C S April, 1975
No. 327
S C R E E N I N G T E C H N I Q U E S IN T H E D I A G N O S I S OF U R I N A R Y T R A C T I N F E C T I O N IN C H I L D R E N PERVIN AHMAD**, N.L. SHARMA AND A . M . KHAN Lucknow
Urinary tract infections are quite common in paediattic practice and are responsible for significant morbidity. I f untreated, they often lead to irreversible kidney damage. It is important to appreciate that the presence of symptoms in infants is a poor guide to infection since many infections are asymptomatic. It becomes imperative, therefore, to diagnose and treat them early. U r i n a r y tract infec-tions are usually diagnosed on the basis of positive urine cultures (colony counts above 100,000 per ml. of urine), which is a time consuming procedure and requires a well equipped laboratory. Attempts to simplify the diagnosis of urinary tract infections, specially for the busy practitioner and laboratories, have led to the introduction of various screening procedures, viz. triphenyl tetrazolium chloride ( T T C ) t e s t ; nitrate test, microscopic examination of urine/'or pus cells and smears stained by Gram's method. Chemical tests have been proved to be unreliable (Eliot and Pryles 1964, KinclaidSmith et al. 1967). In this study an *From the Departments of Paediatrics and Pathology
and Bacteriology, K.G. Medical College, Lucknow. **Present address: Department of Paediatrics, J.N. Medical College, Aligarh Muslim University, Allgarh. Received on January 25. 1975.
attempt has been m a d e to correlate stained smears and microscopic examination o f urine for pus cells with bacterial colony counts and to evaluate these tests as reliable screening procedures. Material and Methods
A total of 150 children were investigated. Out of these, 50 represented the control group, while the remaining 100 cases were those patients who had any one or a combination of the following criteria: (i) Clinical manifestations suggestive of urinary tract infection, ( i i ) p y u r i a , (iii) pyrexia o f obscure origin, (iv) suspicion of congenital abnormalities o f the urinary tract. Collection of the urine in most of the cases was by the clean catch midstream tech= nique in a sterile tube. In all the cases the genitalia were cleaned with soap and water before collecting the urine specimen. All the samples were examined within one hour of collection. When the examination was delayed the specimens were stored at 4~ in a refrigerator. T h e following investigations were carried out: I. Routine urine examination, which included specfic gravity, reaction and chemical examination. 2. Microscopic examination of the urine which comprised (i) pus cell count]
98
VOL. 42, 327
INDIAN JOURNAL O F PEDIATRICS
In 20 cases, the bacterial count was below 100,000/ml. while 75 cultures were sterile.
H . P . F . o f the centrifuged specimen at the rate of 1000 R P . M . for 10 minutes. T h e deposit was e x a m i n e d microscopically a n d at least 10 to 15 high power fields were examined. (ii) G r a m ' s staining of freshly passed uncentrifuged specimen of urine. 3.
Samples with a bacterial colony count above 100,000/ml. showed m o r e than 5 pus cells in 51 out of 55 eases (92.7%), 33 (60.6%) revealed m o r e than 10 pus cells (Table 1). All the cultures showed significant bacteriuria where the pus cell count] H.P.F. was m o r e t h a n 6, whereas only 44.4% were positive when the pus cells were less than 5]H.P.F.
Quantitative urine culture : The n u m b e r of bacteria per ml. o f urine were d e t e r m i n e d by the technique of Schneierson (1962).
Slides o f uncentrifuged urine revealed organisms in 51 out of 55(92.7%) eases with colony counts above 100,000/ml. of urine. No child with a u r i n a r y colony count below 100,000 showed any organism (Table 2),
Observations
Fifty-five out o f 150 cases turned out to h a v e definite u r i n a r y tract infection (bacterial colony count a b o v e 100,000/ml.
Table
1.
Correlation of bacterial counts with pus cells. Pus cells/H.P.F.
No. of cases
8
Bacterial count (number/ml.)
1-5
Below 10,000
.
6-10 .
11-20 .
M o r e t h a n 20
.
12
10,000 to 100,000
5
--
--
--
55
Above I t 0,000
4
18
26
7
T a b l e 2.
Correlation of bacterial counts with Gram's stained smear. Bacteria in G r a m ' s stained smear
Bacterial count/ml.
10,000
No. o f c~ses
Present
Absent
2
~
2
10,000 to 100,000
12
~
12
above
55
51
4
100,000
99
AHMAD E T . A L . ~ S G R E E N I N O TEOHNIQUES IN THE DIAGNOSIS OF U R IN A R Y INFECTION
It is evident from Table 3, that the number o f pus cells per H P.F. correlated with the presence of organisms in Gram's stained smear. 49 out of 51 cases (96%) were positive for organisms when the pus cells per H P.F. were more than 5. Only 2 cases with a pus cell count of less than 5 per high power field were positive for organisms. Discussion T h e majority of workers agree that quantitative bacterial count on centrifuged, freshly voided urine permits delineation of contamination from true bacteriuria. Often facilities to carry out these procedures are not available. In such circumstances there is great need for developing screening tests which are simple and reliable. Among various tests, quantitative pyuria and Gram's staining of fresh uncentrifuged urine for organisms is considered to be one of the most important indices o f suspicion for the diagnosis of urinary tract infection. In 55 cases having a bacterial count of over 100,000, a very significant correlation with the n u m b e r of pus eells/H.P.F, has been found. In 51 (92.7%) cases the n u m b e r
Table 3,
of pus cells were more than 5/H.P.F., while in 33 (60.6%)cases these were more than 10/H.P.F. We have also observed an increasing incidence o f positive cultures with increasing n u m b e r of pus cells per H.P.F. which is reported by other workers as well (Philpot 1956, Kass 1956, Belapurkar 1970). Similarly, the presence of bacteria in Gram's stained smear is also considered to be significant. In our study 51 out o f 55 cases (92.7%) showed bacteria with Gram's staining. T h e r e were no bacteria in the Gram's stained smear in any o f the cases showing a colony count below 100,000/ml. of urine. Pryles and Steg (1959) found a similar correlation in 95% of their cases. L a m et al. (1967) observed 100% correlation in all specimens with colony counts above 100,000/ml. and only 2% of smears were positive for bacteria where the colony count was below 10,000/ml. of urine. Belapurkar (1970) and P u r w a r et al. (1972) have also reported similar observations. It is thus concluded that quantitative pyuria and G r a m ' s stained smear, alone or in combination, is a reliable and simple screening test in suspecting urinary tract infection and can be employed as a
Correlation of pus cells/H.P.F, with positwe bactedal smears. Bacteria in G r a m ' s stained
Pus cells/H.P.F.
No. of cases
Present
Absent
1-5
9
2
7
6-I0
18
16
2
I 1-20
26
26
,:>20
7
7
100
INDIAN JOURNAL
OF
PEDIATRICS
VOL. 42, N o . 327
d i a g n o s t i c tool w h e r e c u l t u r e f a c i l i t i e s a r e not available.
Summary In
150
cases
(50
controls
and
100
patients) urine culture, pus ceUs/H.P.F, and Gram's stain of uncentrifuged urine was d o n e . 55 p a t i e n t s count of more than
showed a bacterial 100,000 p e r m l . O u t
o f these, 51 (92.7%) cases h a d m o r e t h a n 5 p u s c e l l s / H . P . F , a n d 51 ( 9 2 . 7 % ) cases showed bacteria in Gram's
stained smears
of urine. By a c o m b i n e d b a c t e r i a l s m e a r a n d examination for p u s cells ( m o r e than 5 / H . P . F . ) 9 6 % o f cases w i t h s i g n i f i c a n t bacteriuria could be diagnosed.
References Belapurkar, K M . , Manjrekar, P.S. and Kaul, K.K. (1970). Urinary tract infection II. Indian Pedlar. 7, 449.
Eliot, C.R. and Pryles, G.V. (1964). Observations on the use of triphenyl tetrazolium for the detection of bacteriuria. Pediatrics, 34, 424. Kass, E.M. (1956). Asymptomatic infection of the urinary tract. Trans. Ass. Amer. Physns. 69, 55. Kinclaid-Smith, P., Bullen, M., Mills, J., Russell, U,, Huoston, N. and Goon, F. (1964). The reliability of screening tests for bacteriuria in pregnancy. Lancet 2,61. Lain, C.N., Bremner, A.D., Maxwell, J.D., Murphy, A.V. and Low, W.J. (1967). Pyuria and bacteriuria. Arch. Dis. Childh. 42, 275. Philpot, V.B. (1956). Bacterial flora of urine specimen from normal adults. 07. Utol. 59,562. Pryles, C.V. and Steg, N L. (1959). Specimen of urine obtained from young girls by catheter versul voiding. Pediatrics, 32, 441. Purwar, V.N., Agrawal, S.P. and Dikshit, S.K. (1970). Gram's stained urine slides in the diagnosi~ of urinary tract infection in children. 07. Indian reed. Ass. 59, 3. Schneierson, S.S. (1962). Simplified proceaun for performing urinary bacterial counts. 07" Urol. 88; 2,
RNAL OF P E D I A T R I C S April, 1975
No. 327
S C R E E N I N G T E C H N I Q U E S IN T H E D I A G N O S I S OF U R I N A R Y T R A C T I N F E C T I O N IN C H I L D R E N PERVIN AHMAD**, N.L. SHARMA AND A . M . KHAN Lucknow
Urinary tract infections are quite common in paediattic practice and are responsible for significant morbidity. I f untreated, they often lead to irreversible kidney damage. It is important to appreciate that the presence of symptoms in infants is a poor guide to infection since many infections are asymptomatic. It becomes imperative, therefore, to diagnose and treat them early. U r i n a r y tract infec-tions are usually diagnosed on the basis of positive urine cultures (colony counts above 100,000 per ml. of urine), which is a time consuming procedure and requires a well equipped laboratory. Attempts to simplify the diagnosis of urinary tract infections, specially for the busy practitioner and laboratories, have led to the introduction of various screening procedures, viz. triphenyl tetrazolium chloride ( T T C ) t e s t ; nitrate test, microscopic examination of urine/'or pus cells and smears stained by Gram's method. Chemical tests have been proved to be unreliable (Eliot and Pryles 1964, KinclaidSmith et al. 1967). In this study an *From the Departments of Paediatrics and Pathology
and Bacteriology, K.G. Medical College, Lucknow. **Present address: Department of Paediatrics, J.N. Medical College, Aligarh Muslim University, Allgarh. Received on January 25. 1975.
attempt has been m a d e to correlate stained smears and microscopic examination o f urine for pus cells with bacterial colony counts and to evaluate these tests as reliable screening procedures. Material and Methods
A total of 150 children were investigated. Out of these, 50 represented the control group, while the remaining 100 cases were those patients who had any one or a combination of the following criteria: (i) Clinical manifestations suggestive of urinary tract infection, ( i i ) p y u r i a , (iii) pyrexia o f obscure origin, (iv) suspicion of congenital abnormalities o f the urinary tract. Collection of the urine in most of the cases was by the clean catch midstream tech= nique in a sterile tube. In all the cases the genitalia were cleaned with soap and water before collecting the urine specimen. All the samples were examined within one hour of collection. When the examination was delayed the specimens were stored at 4~ in a refrigerator. T h e following investigations were carried out: I. Routine urine examination, which included specfic gravity, reaction and chemical examination. 2. Microscopic examination of the urine which comprised (i) pus cell count]
98
VOL. 42, 327
INDIAN JOURNAL O F PEDIATRICS
In 20 cases, the bacterial count was below 100,000/ml. while 75 cultures were sterile.
H . P . F . o f the centrifuged specimen at the rate of 1000 R P . M . for 10 minutes. T h e deposit was e x a m i n e d microscopically a n d at least 10 to 15 high power fields were examined. (ii) G r a m ' s staining of freshly passed uncentrifuged specimen of urine. 3.
Samples with a bacterial colony count above 100,000/ml. showed m o r e than 5 pus cells in 51 out of 55 eases (92.7%), 33 (60.6%) revealed m o r e than 10 pus cells (Table 1). All the cultures showed significant bacteriuria where the pus cell count] H.P.F. was m o r e t h a n 6, whereas only 44.4% were positive when the pus cells were less than 5]H.P.F.
Quantitative urine culture : The n u m b e r of bacteria per ml. o f urine were d e t e r m i n e d by the technique of Schneierson (1962).
Slides o f uncentrifuged urine revealed organisms in 51 out of 55(92.7%) eases with colony counts above 100,000/ml. of urine. No child with a u r i n a r y colony count below 100,000 showed any organism (Table 2),
Observations
Fifty-five out o f 150 cases turned out to h a v e definite u r i n a r y tract infection (bacterial colony count a b o v e 100,000/ml.
Table
1.
Correlation of bacterial counts with pus cells. Pus cells/H.P.F.
No. of cases
8
Bacterial count (number/ml.)
1-5
Below 10,000
.
6-10 .
11-20 .
M o r e t h a n 20
.
12
10,000 to 100,000
5
--
--
--
55
Above I t 0,000
4
18
26
7
T a b l e 2.
Correlation of bacterial counts with Gram's stained smear. Bacteria in G r a m ' s stained smear
Bacterial count/ml.
10,000
No. o f c~ses
Present
Absent
2
~
2
10,000 to 100,000
12
~
12
above
55
51
4
100,000
99
AHMAD E T . A L . ~ S G R E E N I N O TEOHNIQUES IN THE DIAGNOSIS OF U R IN A R Y INFECTION
It is evident from Table 3, that the number o f pus cells per H P.F. correlated with the presence of organisms in Gram's stained smear. 49 out of 51 cases (96%) were positive for organisms when the pus cells per H P.F. were more than 5. Only 2 cases with a pus cell count of less than 5 per high power field were positive for organisms. Discussion T h e majority of workers agree that quantitative bacterial count on centrifuged, freshly voided urine permits delineation of contamination from true bacteriuria. Often facilities to carry out these procedures are not available. In such circumstances there is great need for developing screening tests which are simple and reliable. Among various tests, quantitative pyuria and Gram's staining of fresh uncentrifuged urine for organisms is considered to be one of the most important indices o f suspicion for the diagnosis of urinary tract infection. In 55 cases having a bacterial count of over 100,000, a very significant correlation with the n u m b e r of pus eells/H.P.F, has been found. In 51 (92.7%) cases the n u m b e r
Table 3,
of pus cells were more than 5/H.P.F., while in 33 (60.6%)cases these were more than 10/H.P.F. We have also observed an increasing incidence o f positive cultures with increasing n u m b e r of pus cells per H.P.F. which is reported by other workers as well (Philpot 1956, Kass 1956, Belapurkar 1970). Similarly, the presence of bacteria in Gram's stained smear is also considered to be significant. In our study 51 out o f 55 cases (92.7%) showed bacteria with Gram's staining. T h e r e were no bacteria in the Gram's stained smear in any o f the cases showing a colony count below 100,000/ml. of urine. Pryles and Steg (1959) found a similar correlation in 95% of their cases. L a m et al. (1967) observed 100% correlation in all specimens with colony counts above 100,000/ml. and only 2% of smears were positive for bacteria where the colony count was below 10,000/ml. of urine. Belapurkar (1970) and P u r w a r et al. (1972) have also reported similar observations. It is thus concluded that quantitative pyuria and G r a m ' s stained smear, alone or in combination, is a reliable and simple screening test in suspecting urinary tract infection and can be employed as a
Correlation of pus cells/H.P.F, with positwe bactedal smears. Bacteria in G r a m ' s stained
Pus cells/H.P.F.
No. of cases
Present
Absent
1-5
9
2
7
6-I0
18
16
2
I 1-20
26
26
,:>20
7
7
100
INDIAN JOURNAL
OF
PEDIATRICS
VOL. 42, N o . 327
d i a g n o s t i c tool w h e r e c u l t u r e f a c i l i t i e s a r e not available.
Summary In
150
cases
(50
controls
and
100
patients) urine culture, pus ceUs/H.P.F, and Gram's stain of uncentrifuged urine was d o n e . 55 p a t i e n t s count of more than
showed a bacterial 100,000 p e r m l . O u t
o f these, 51 (92.7%) cases h a d m o r e t h a n 5 p u s c e l l s / H . P . F , a n d 51 ( 9 2 . 7 % ) cases showed bacteria in Gram's
stained smears
of urine. By a c o m b i n e d b a c t e r i a l s m e a r a n d examination for p u s cells ( m o r e than 5 / H . P . F . ) 9 6 % o f cases w i t h s i g n i f i c a n t bacteriuria could be diagnosed.
References Belapurkar, K M . , Manjrekar, P.S. and Kaul, K.K. (1970). Urinary tract infection II. Indian Pedlar. 7, 449.
Eliot, C.R. and Pryles, G.V. (1964). Observations on the use of triphenyl tetrazolium for the detection of bacteriuria. Pediatrics, 34, 424. Kass, E.M. (1956). Asymptomatic infection of the urinary tract. Trans. Ass. Amer. Physns. 69, 55. Kinclaid-Smith, P., Bullen, M., Mills, J., Russell, U,, Huoston, N. and Goon, F. (1964). The reliability of screening tests for bacteriuria in pregnancy. Lancet 2,61. Lain, C.N., Bremner, A.D., Maxwell, J.D., Murphy, A.V. and Low, W.J. (1967). Pyuria and bacteriuria. Arch. Dis. Childh. 42, 275. Philpot, V.B. (1956). Bacterial flora of urine specimen from normal adults. 07. Utol. 59,562. Pryles, C.V. and Steg, N L. (1959). Specimen of urine obtained from young girls by catheter versul voiding. Pediatrics, 32, 441. Purwar, V.N., Agrawal, S.P. and Dikshit, S.K. (1970). Gram's stained urine slides in the diagnosi~ of urinary tract infection in children. 07. Indian reed. Ass. 59, 3. Schneierson, S.S. (1962). Simplified proceaun for performing urinary bacterial counts. 07" Urol. 88; 2,
