Symptomatic Urinary Tract Infection in Children: Experience in a Regional Hospital in Hong Kong
We performed a retrospective review of the demographic, clinical, imaging and outcome data of 94 children diagnosed in our hospital to have symptomatic urinary tract infection (UTI) between January 1994 to June 1996. Our findings were compared with previous local reports to look for changing epidemiology, and with overseas studies to look for ethnic differences. We found that in this local Chinese cohort, UTI mainly affected boys (78%) and infants (85% < 1 year), in whom it presented with fever and non-specific symptoms. The diagnosis was established by demonstrating pyuria and significant positive bacterial culture from either suprapubic aspirated (42.5%), catheter (34%) or clean catch urine (14.9%) samples, but 8 patients required DMSA scan for retrospective diagnosis because of mixed growth in bag urine or no growth due to prior antibiotic therapy. The causative organisms were E. coli in 78% of cases, but could also be Klebsiella, Proteus and Enterococci. Complete imaging studies were performed in 86% of our patients and revealed hydronephrosis in 12%, VUR in 23%, persistent renal scarring in 16% of patients. No abnormalities were found in 56%. There was a strong association between VUR, recurrent UTI and renal scarring. On medium term follow up, 16% of patients had recurrence of UTI. The prognosis was favourable as 77% of the 31 refluxing units and 43% of hydronephrosis detected resolved spontaneously. One patient required partial nephrectomy and excision of ureterocele and two patients required circumcision. An overall of 21 patients (22%) had "complicated" UTI and they required long term follow up management, preferably by paediatric nephrologists in consultation with urologists.
本文從人口統計學、臨床、影像學和治療效果等方面，對自 1994 年 1 月至 1996 年 6 月期間我院診斷的 94 例兒童症狀性尿道感染進行回顧性總結。把我們的發現與本地以前的報告進行比較以掌握變化中的流行病學資料，並與海外的有關研究進行比較以探尋種族差異。我們發現，在此組本地華人病例，尿道感染主要影響男童（ 78%）和嬰幼兒（85%<1 歲），他們表現為發熱和非特異性症狀。確診以表現為膿尿和經恥骨上膀胱穿刺（ 42.5%）、導管(34%)取尿或清潔中斷尿液(14.9%)細菌培養呈明顯陽性為準，但是，其中 8 例病人由於留取的尿液內細菌呈混合生長或因為應用抗生素治療造成細菌不能生長而行 DMSA 掃描。78% 病例的病源體是大腸杆菌，但也可能是克雷白杆菌、變形杆菌和腸球菌。 86% 的病人接受完整的影像學檢查，並提示 12% 病例為腎盂積水、23% 為膀胱輸尿管返流、16% 表現為永久性腎臟斑痕化。56% 病例無異常表現。膀胱輸尿管返流、復發性尿道感染和腎臟斑痕化之間存在緊密關聯。 31 個返流性單位中的 77% 預後良好，43% 腎盂積水病例自發緩解。1 例病人需要部分性腎臟切除和輸尿管疝切除術，2 例病人需要行包皮環切術。共 21 例（22%）病人因存在複雜的尿道感染而尤其需要兒童腎科醫生和泌尿科醫生的長期隨訪治療。
Keyword : Children; Chinese; Renal scarring; Urinary tract infection; Vesicoureteral reflux
Urinary tract infection (UTI) is a common bacterial infection in children. Studies in Western countries suggested that it accounted for 5% of febrile illnesses in young children.1 It is especially common in infants, female and the White race.2 Imaging studies following UTI revealed a high incidence of abnormalities in the renal tract, with vesicoureteric reflux (VUR) in 30-50%1 and obstructive uropathies in 1-4%.3 Evidence of renal parenchymal damage was present in 1.6-15% as seen on intravenous urography3 and 59% as seen on 99mTechnetium-dimercaptosuccinic acids (DMSA) scans.4 The risk of renal scarring was positively associated with the severity of VUR and number of recurrent febrile UTI. The relationship of such radiographic findings to long term sequalae such as hypertension and chronic renal failure have been challenged. Nevertheless four of the 52 patients with VUR reported by Smellie et al developed end stage renal failure after 10 years.5 A review of the literature also found that end stage renal disease developed in 3-10% of patients with extensive scarring.1 Ethnic difference has also been reported in the incidence of UTI and VUR.6
Despite controversies in its management, it is important to recognise UTI in young children and be familiar with its clinical significance. In the local Chinese population, case series of childhood UTI were reported in the mid-1980's.7,8 In the present study we aimed to review the epidemiology, clinical findings, imaging studies and outcome of a more recent cohort of UTI patients and highlight any secular changes by comparing with previous reports, or any significant differences from the Western populations.
Patients and Methods
All children admitted to the Department of Paediatrics, Tuen Mun Hospital from 1st January 1994 to 30th June 1995, and diagnosed to have symptomatic UTI were included and their records were retrieved for analysis.
UTI was diagnosed by the following criteria: (1) presence of symptoms such as fever, irritability, vomiting or dysuria or frequency, and (2) any bacterial growth in the suprapubic aspirated urine, or a pure bacterial growth of ?05 in midstream, catheterised or clean catch urine samples, or (3) any inflammatory change demonstrated in the early DMSA scans. Treatment of the acute UTI consisted of 7-10 days of cefuroxime or netromycin depending on clinical responses. Follow up imaging studies (USG, VCUG and DMSA) were recommended for all patients below 5 years old.
The ultrasonographic (USG) examination was performed by the radiologists using adapted equipment with high-resolution curvilinear and linear transducers. Voiding cystourethrography (VCUG) was performed by filling the bladder by urethral catheterisation. VUR was graded according to the International Reflux Study Committee classification. DMSA scans were performed by intravenous injection of 99mTc-DMSA. Images were acquired by a gamma camera with patients in supine position. The fractional left and right renal activity was calculated for each kidney after background correction.
Patients were given prophylactic antibiotics until imaging investigations were completed. Those found to have VUR or other urological abnormalities were given long term prophylactic antibiotics (usually trimethoprim, co-trimoxazole or nitrofurantoin), as were those who developed recurrent UTIs. Parents were advised that when their children developed unexplained fever, they should immediately consult medical advice to have UTI diagnosed or excluded. During follow-up visits, any recurrence of UTI was recorded. Patient's body weight and height, blood pressure, urinalysis and urine cultures were done. The age and sex, presenting symptoms, culture results, the investigation results, duration of follow up and outcome were analysed.
The data were analysed by Chi-square test or Fisher's exact test where appropriate. A P value <0.05 was considered significant.
Demographic Characteristics (Table 1)
Ninety-four patients were studied, including 73 boys (78%) and 21 girls (22%). The mean age of presentation was 9.5 months (median 3 months, range one day to 7.9 years). Eight patients (8.5%) were neonates (M:F = 7:1). Seventy-two (76.6%) were between 1-12 months old (M:F = 57:15), and 14 were older children (M:F = 9:5).
Symptoms and Signs (Table 1)
Fever (>=38.0°C) was the major symptom in 87.5% of the neonates (<1 month) and in 94.4% of infants. Other symptoms in these age groups were also non-specific. In older children, only 50% had fever, but 70% presented with dysuria. One neonate (12%) and 6 infants (8%) presented as febrile seizure. One patient had Escherichia coli septicemia secondary to UTI.
UTI was documented by significant bacterial growth in the suprapubic aspirated urine in 40 patients (42.5%), catheterised urine in 32 patients (34%) and midstream urine in 14 (14.9%). In 8 patients UTI was diagnosed by significant growths in bag urine samples and confirmed by evidence of acute pyelonephritis in DMSA scan done within 2 months.
Microbiologic Findings (Table 2)
Escherichia coli was obtained in 74 patients (78%), followed by Enterococci (6%) and Klebsiella sp. (4%). Mixed organisms or no growth were obtained in 4% of cases and diagnosis was confirmed by positive findings of acute pyelonephritis in the early DMSA scans.
Only a minority of patients did not undergo full investigation either because of older age at presentation or parental refusal. Eighty-three of the 94 (86%) had undergone all 3 investigations (USG, VCUG & DMSA), with normal findings in 53 (56%).
USG was performed on 90 patients and showed abnormalities in 11 (12%): Six patients had unilateral hydronephrosis; two had bilateral hydronephrosis; and one had a duplex left kidney with ectopic ureterocele.
VCUG was performed in 87 patients. Twenty patients (23%) had VUR which was bilateral in 11 (8 boys and 3 girls) and unilateral in 9 (6 boys and 3 girls). Of the 31 refluxing kidneys in these 20 patients, 23 were mild (IRSC Grade I-III) while 8 were severe (Grade IV-V). Table 3 shows the relationship of the kidneys in relation to sex, laterality, VUR, grading of VUR and DMSA scan abnormalities.
Eighty-eight patients had undergone DMSA scan at a mean time of 2.7 months post-UTI (86 cases within 6 months post-UTI). Decreased isotope uptakes were found in 36 kidneys (20%) of 35 patients (40%). After second scan at a mean time of 12.1 months post-UTI (range: 2.3-27 months), 22 kidneys became normal (hence having transient acute pyelonephritic changes only). Persistent abnormalities were found in the 14 kidneys of 13 patients (16%). As shown in Table 3, 7 kidneys among the 67 patients without VUR had persistent abnormalities (1 renal agenesis, 1 small contracted or dysplastic kidney, 5 focal cortical scars), while 7 kidneys among the 20 patients with VUR had persistent abnormalities (4 small contracted kidneys and 3 focal renal scars). As shown in Table 4, renal scarring was significantly associated with presence of VUR especially high grade VUR, but not with sex or side of VUR, or infecting organism.
The mean duration of follow-up was 30.7 months (median 21 months, range 1-83 months). Most patients were either discharged or defaulted themselves after investigations showed no renal tract abnormalities and there were no UTI recurrences after 1-2 years. Only 3 patients had defaulted follow up despite significant renal abnormalities: small cortical scar (1), bilateral VUR (1) and right renal agenesis and left grade V VUR with focal scar (1).
Fifteen patients (16%) had recurrences of UTI despite prophylactic antibiotics therapy. The majority (80%) occurred within the first 3 months, even before full imaging studies could be arranged. Five patients had two or more relapses. Associated renal abnormalities were found in 11 of the 15 patients (73%). These included: 8 patients with VUR (with small contracted kidneys in 2, focal scar in one and renal agenesis in one), 2 with dilated renal pelvis, and one with no VUR but a small contracted kidney. E. coli was the commonest microorganism in the recurrent UTI being involved in half of the episodes. Klebsiella and Enterococci was the second and third common causative microorganism respectively. As shown in Table 5, there was strong association of UTI recurrence with VUR and persistent DMSA abnormalities.
Spontaneous resolution of VUR occurred in 24 kidneys of 16 patients (77% of the 31 refluxing units). This was detected at a mean of 31 months after the UTI. However, this was not the true time of resolution but just reflected the average interval we would repeat voiding cystograms. As shown in Table 6, there was no relationship to the sex, laterality, grading of VUR or presence of renal scarring.
Four out of the 7 hydronephrotic kidneys (43%) resolved spontaneously. Surgical operations were required for 3 patients: one patient with duplex left kidney and ectopic ureterocele underwent partial nephrectomy and ureterectomy, and two other patients underwent circumcision because of multiple recurrences of UTI.
No patients had hypertension and renal impairment. However the length of follow up was probably inadequate for such complications to develop.
This study describes the demographic, clinical and imaging features of an unselected cohort of children with first time UTI presenting to a regional hospital in Hong Kong in 1994-95. Compared with previous reports by Chow et al7 (cohort from 1978-85) and So et al8 (cohort from 1984-86), this is a more recent cohort with more comprehensive imaging studies performed, especially with the availability of DMSA scan. However we should be aware that this is a hospital based study and some patients with UTI may have been missed because they were treated by private practitioners and not referred to our hospital.
Our data showed that children with UTI commonly first presented in infancy (80% of our cases). This was in agreement with So and Davies8 (86% of boys and 60% of girls presenting in the first year) but differed from the report by Chow et al7 (only 45.5% in the first year). Since the inclusion criteria were similar, the difference may reflect genuine decrease in incidence of UTI in older children because they have been picked up at an earlier age or because of improvement in other risk factors such as constipation and poor toilet habits.
In contrast, Hoberman et al reported a female predominance (276 girls versus 33 boys) in their cohort of children in the USA who had UTI below 2 years of age.9 Similar female predominance were also noted in the meta-analysis by Downs, which showed a prevalence of 3% and 2% in febrile boys below and above 1 year old respectively while the prevalence was 7% and 8% for febrile girls below and above 1 year old respectivley.1 This is probably because neonatal circumcision was commonly practised in the USA while it is seldom performed in Hong Kong. Previous studies have shown that the prevalence of UTI could be increased 10 times in uncircumcised boys compared to circumcised boys.10,11
The presenting symptoms of our patients were similar to those of previous reports. Indeed they have not changed since the early description by Smellie12 that young children presented with fever and non-specific symptoms while older children would have increasing symptoms of cystitis. Thus current guidelines mandates testing urine for all febrile children below 2 years of age.13,14
Escherichia coli accounted for 78% of UTI episodes in our study, followed much less commonly by Klebsiella, Proteus and Enterococci. In contrast to previous impression,15 non-E.coli UTI were not significantly associated with complicated UTI such as the presence of VUR or renal scarring. However this may be due to the small sample size of our cohort.
Current guidelines recommend that young children with first symptomatic UTI should undergo USG and VCUG. The role of DMSA is more controversial.13,16 Compared with previous local reports, our data provided a more accurate incidence of urological abnormalities that would be found in children with first time UTI, because 86% of patients had undergone a "complete" investigation. We found that 44% of patients had abnormalities on either or all of USG, VCUG or DMSA. Twelve percent had abnormal USG findings, mostly being hydronephrosis but one patient having duplex left kidney with obstructed upper moiety and ectopic ureterocele. Twenty-three percent had VUR which was of severe grade (IV-V) in a quarter. This was lower than the incidence of 30-50% in Western studies.1 Forty percent of patients had abnormalities on the first DMSA scan, though only 16% had persistent abnormalities (i.e. renal dysplasia or focal scars) as shown by repeated DMSA scans. Such information would be useful to counsel parents in the local scenario. We did not do second DMSA scan for patients who had normal first scan. However, from a previous study where all children had undergone both acute and follow up DMSA scans, no patients with normal initial scan had abnormalities on second scan and 15% of all patients had renal scars.9
Hoberman et al reported that in his cohort of 309 US children, USG detected renal tract dilatation in 12%, VCUG detected VUR in 39% but only 1.9% were of grade IV.9 They concluded that USG was of limited value as they did not modify management. However their cohort was highly selective and children with known urological abnormalities were excluded. In their population, most urological abnormalities would have been detected by antenatal USG screening. This may not apply to our population where antenatal ultrasound screening was not a standard or universal practice. For our patients, a change in management plan was required in 21 patients as a result of the imaging studies: surgery in one, prophylactic antibiotic and monitoring VUR in 20 patients, monitoring renal tract dilatation in 8, long term follow up for the 13 patients with persistent DMSA abnormalities. We must point out that the long standing practice of prophylactic antibiotic for VUR was based on weak evidence from a few randomised trials and has become a controversial and undecided issue recently.17,18
Our study was limited by the short duration of follow up in most cases, and clinically important outcomes such as hypertension or renal impairment were not detected in our patients. Nevertheless concerning the prognosis of these patients, we showed that firstly, recurrence of UTI was an important problem. It recurred in 16% of our patients, most in the first 3 months. Recurrence was highly associated with VUR and renal scarring/dysplasia, and not with gender nor early DMSA abnormalities. We recommend that all young children with first UTI be closely observed for recurrence in the first 6 months and they should be covered with prophylactic antibiotics till results of imaging studies were available.
Secondly persistent DMSA abnormalities (renal scarring or dysplasia) were present in 14 kidneys (16%) of 13 patients (14%). It may be an important surrogate indicator for long term sequalae, though this is still controversial. We also showed significant associations between recurrence of UTI, VUR (especially high grade VUR) and renal scarring/dysplasia. This was in support of previous reports. A meta-analysis by AAP revealed that the risk of renal scarring in patients with high grade VUR was 8-10 times of those without VUR and 4-6 times of those with low grade VUR.1 Also the risk of renal scarring rose exponentially as the number of recurrent UTI increases.19 However VUR is not the only important factor for renal scarring/dysplasia. A separate meta-analysis by Gordon et al in 2003 showed that among seven studies involving 1062 kidneys, abnormal DMSA scan occurred in 44% of the kidneys with VUR and in 31% of kidneys without VUR. They concluded that a positive MCU only increased the likelihood ratio of abnormal DMSA scan by 2.34 times (CI 1.53 to 3.57), and a negative MCU reduced the likelihood ratio to 0.71 (CI 0.58 to 0.85).4
Thirdly VUR has a high chance of spontaneous resolution (77% in our patients), and can be managed medically. Previous reviews have shown that the chance of resolution decreases with advancing age, increased severity of VUR and bilateral VUR.20 This was not seen in our patients, probably because of the small sample size and short duration of follow up.
Lastly surgical correction was indeed required in one patient and circumcision was performed in two patients to stop multiple recurrences of UTI. Thus management of the minority of patients with complicated UTI should involve a joint approach of nephrologists and urologists.
In conclusion our review provided useful local data on childhood UTI, in terms of demographic data, imaging abnormalities, and short term outcome. We showed that there was an increasing trend of infantile UTI compared with reports 10 years ago. In contrast to overseas reports which have female preponderance, local data suggested a strong male predominance. The bacteriology was not different from previous or overseas reports. Because almost all our patients had undergone USG, VCUG and DMSA scans, our data provided more accurate information than previous local studies on the urological abnormalities in children presenting with first time UTI. Also we found a lower incidence of VUR than overseas figures (23% versus 30-50%). The incidence of persistent renal scarring were similar. Sixteen percent of patients had UTI recurrences. One patient required partial nephrectomy and excision of ureterocele and two patients required circumcision for multiple recurrences. There were strong associations between VUR, recurrent UTI and renal scarring. An overall of 21 patients (22%) with these various abnormalities required long term follow up management, preferably by paediatric nephrologists in consultation with the urologists.
1. Downs SM. Technical report: urinary tract infections in febrile infants and young children. The Urinary Tract Subcommittee of the American Academy of Pediatrics Committee on Quality Improvement. Pediatrics 1999;103:e54.
2. Shaw KN, Gorelick M, McGowan KL, Yakscoe NM, Schwartz JS. Prevalence of urinary tract infection in febrile young children in the emergency department. Pediatrics 1998;102:e16.
3. Dick PT, Feldman W. Routine diagnostic imaging for childhood urinary tract infections: a systematic overview. J Pediatr 1996;128:15-22.
4. Gordon I, Barkovics M, Pindoria S, Cole TJ, Woolf AS. Primary vesicoureteric reflux as a predictor of renal damage in children hospitalized with urinary tract infection: a systematic review and meta-analysis. J Am Soc Nephrol 2003;14:739-44.
5. Smellie JM, Barratt TM, Chantler C, et al. Medical versus surgical treatment in children with severe bilateral vesicoureteric reflux and bilateral nephropathy: a randomised trial. Lancet 2001;357:1329-33.
6. Melhem RE, Harpen MD. Ethnic factors in the variability of primary vesico-ureteral reflux with age. Pediatr Radiol 1997;27:750-1.
7. Chow CB, Yau FT, Leung NK. Symptomatic urinary tract infection in Hong Kong children. JHK Med Assoc 1988;40:276-80.
8. So LY, Davies DP. Urinary tract infection in childhood: a study of 137 cases. HK J Paediatr 1988;5:17-24.
9. Hoberman A, Charron M, Hickey RW, Baskin M, Kearney DH, Wald ER. Imaging studies after a first febrile urinary tract infection in young children. N Engl J Med 2003;348:195-202.
10. Craig JC, Knight JF, Sureshkumar P, Mantz E, Roy LP. Effect of circumcision on incidence of urinary tract infection in preschool boys. J Pediatr 1996;128:23-7.
11. To T, Agha M, Dick PT, Feldman W. Cohort study on circumcision of newborn boys and subsequent risk of urinary-tract infection. Lancet 1998;352:1813-6.
12. Smellie JM, Hodson CJ, Edwards D, Normand IC. Clinical and radiological features of urinary infection in childhood. Br Med J 1964;5419:1222-6.
13. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children. American Academy of Pediatrics. Committee on Quality Improvement. Subcommittee on Urinary Tract Infection. Pediatrics 1999;103(4 Pt 1):843-52.
14. Wong SN, Chiu W, Ho S, et al. Clinical Guideline on Management of Urinary Tract Infections in Children below 2 Years of Age (Part I): The Diagnosis and Initial Management. HK J Paediatr (new series) 2002;7:205-13.
15. Honkinen O, Lehtonen OP, Ruuskanen O, Huovinen P, Mertsola J. Cohort study of bacterial species causing urinary tract infection and urinary tract abnormalities in children. BMJ 1999;318:770-1.
16. Wong SN, Chiu W, Ho S, et al. Clinical guideline on management of urinary tract infections in children below 2 years of age (part II): investigations following a documented infection. HK J Paediatr (new series) 2003;8:47-54.
17. Mangiarotti P, Pizzini C, Fanos V. Antibiotic prophylaxis in children with relapsing urinary tract infections: review. J Chemother 2000;12:115-23.
18. Williams G, Lee A, Craig J. Antibiotics for the prevention of urinary tract infection in children: A systematic review of randomized controlled trials. J Pediatr 2001;138:868-74.
19. Jodal U. The natural history of bacteriuria in childhood. Infect Dis Clin North Am 1987;1:713-29.
20. Elder JS, Peters CA, Arant BS Jr, et al. Pediatric Vesicoureteral Reflux Guidelines Panel summary report on the management of primary vesicoureteral reflux in children. J Urol 1997;157:1846-51.