Abstract

Objectives. To determine the recurrence rate and risk factors for urinary tract infection (UTI) in children who present with a febrile UTI and have a negative radiologic evaluation. Febrile UTIs with no urinary tract abnormalities are a common cause of morbidity in children.

Methods. We performed a retrospective review of all children referred to our medical center after a febrile UTI.

Results. We reviewed 850 charts. Of 850 children, 78 had had a febrile UTI and normal ultrasound and voiding cystourography findings. Of the 78 children, 25 had had a recurrent UTI (3 boys and 22 girls). Forty-five percent of the girls with a febrile UTI developed a recurrent UTI and 14% of the boys had a recurrent UTI (P = 0.02). Three boys (two younger than 1 year of age) were uncircumcised and had one recurrent febrile UTI. Eleven (39%) of 28 girls who first presented at younger than 1 year of age and 7 (58%) of 12 girls who presented at 5 years of age or older had recurrent UTIs. The recurrence rate in the 2 to 5-year-old age group was 24% (4 of 17). Seven of the older girls exhibited symptoms of dysfunctional elimination syndrome.

Conclusions. In children with a febrile UTI and a negative radiologic evaluation, recurrence was more common in girls. Boys who were uncircumcised may be at an increased risk of infection during the first year of life. In girls, the age at the time of the first infection was not predictive of recurrence. Although dysfunctional voiding and elimination may contribute to recurrent febrile UTIs in young children, an association seems to be present in children 5 years old and older.

Febrile urinary tract infections (UTIs) affect an estimated 1.0% of boys and 3.1% of girls[1] and cause significant morbidity. In the general pediatric population the recurrence rates for UTI are very high. Within 1 year of a first infection, approximately 30% of boys and 40% of girls will develop a repeat UTI.[2] After the first episode, children can expect a recurrence rate of 30%. This rate will double for each subsequent infection.[3] Anatomic obstruction (posterior urethral valves, ureteropelvic junction obstruction, ureterovesical obstruction, and ureterocele) as an etiology for UTI is seen in 2% to 10%,[4] and 30% to 50% will have vesicoureteral reflux.[5] More recent studies have suggested that less than 1% of children with a first-time UTI will have obstruction, reflecting earlier detection with ultrasonography before the onset of a UTI.[6] Young children who develop a febrile UTI frequently undergo radiologic evaluations to identify any anatomic abnormalities. Often, these tests fail to demonstrate any urinary tract anomalies such as obstruction or vesicoureteral reflux. The natural history of children who initially present with a febrile UTI and who have no identifiable anatomic abnormality demonstrated on the radiologic evaluation remains undefined, including the likelihood of recurrence and the risk factors for repeated infection. The purpose of this study was to determine the recurrence rate of febrile infections in children who present with a febrile UTI and have a negative radio-logic evaluation (defined as the absence of hydronephrosis with normal renal parenchyma on renal ultrasonography, as well as the absence of reflux on voiding cystourethrography) and to identify the risk factors for UTI recurrence.

Material and Methods

A retrospective evaluation of children referred to our medical center (all services) for radiologic evaluation after a febrile UTI was performed. Renal/bladder ultrasonography and voiding cystourethrography were performed on all patients to rule out hydronephrosis and reflux, respectively. We reviewed the medical records for symptoms, method of specimen collection, type of organism, oral/intravenous antibiotic administration, toilet training, evidence of voiding/elimination dysfunction, and circumcision status. Follow-up to determine recurrence was done by telephone interview with the patient's parents or pediatrician, for which the institutional review board gave approval. UTI recurrence was defined by the presence of a fever and positive urine culture (more than 100,000 organisms per voided specimen in older children and per catheterized specimen in most children younger than 5 years). The culture results were confirmed when performed at our institution. Children with a positive culture in the absence of fever were excluded. Statistical significance was analyzed using the Student t test.

Results

We reviewed the charts of 850 children who presented to our institution from 1993 to 2000 with a febrile UTI. Of the 850 children, 78 of these children had a febrile UTI and normal ultrasound and voiding cystourethrography findings (Table I). Of the 78 children, 21 were boys (mean age 21 months, range 1 to 168) and 57 were girls (mean age 31 months, range 1 to 120). Thirteen of the boys were younger than 1 year of age and seven were between 2 and 5 years of age. Twenty-eight girls were younger than 1 year of age, 17 were between 2 and 5 years of age, and 12 were older than 5 years. Of the 13 boys who were younger than 1 year of age, 12 (92%) were uncircumcised at the time of their first infection. Of the 7 boys 2 to 5 years of age, 6 were uncircumcised at the time of the initial UTI. None of the boys younger than 1 year old were toilet trained at the time of the initial infection, and all 7 (100%) of the 2 to 5-year-old boys were toilet trained. None of the girls younger than 1 year old, 14 (82%) of the 17 girls 2 to 5 years old, and all 12 (100%) of the girls older than 5 years were toilet trained at the time of the initial UTI. Follow-up was available on 69 (88%) of the 78 patients (mean follow-up 3.5 years, range 1 to 7). Of the 78 children, 25 had recurrent febrile UTIs (3 boys and 22 girls). Of the 78 children, 45% of the girls developed a recurrence and 14% of the boys did so (P = 0.02). All 3 boys were uncircumcised and had one episode of recurrent febrile UTI. Two (15%) of the 13 boys younger than 1 year of age and 1 (14%) of the 7 boys between 2 and 5 years old had a recurrence. Of the 22 girls who had a recur-rent febrile UTI, the average number of recurrent episodes was 2.1 (range 1 to 5). The patient age at the time of first infection was not predictive of recurrence. Eleven (39%) of the 28 girls who first presented with a febrile UTI at younger than 1 year of age had a recurrent febrile UTI, and 7 (58%) of the 12 who presented at 5 years of age or older also had a recurrent febrile UTI. The recurrence rate in the 2 to 5-year-old age group was 24% (4 of 17). One half of the girls with a recurrent UTI were toilet trained at the time of their second infection. Seven of the older girls (5 years of age and older) had symptoms of dysfunctional elimination syndrome. No differences were seen between those children who received intravenous antibiotics (and were hospitalized) and those who received oral antibiotics.

Table I. Children with negative radiologic workup and recurrent UTIs

Age (yr)	< 1	2-5	>5
Males (n=21)	13	7	1
Uncircumcised at first UTI	12/13 (92)	6/7 (86)	0/1 (0)
Toilet trained at first UTI	0/13 (0)	7/7 (100)	1/1 (100)
Children with recurrent UTI (n=3, 14%)	2/13 (15)	1/7 (14)	0/1 (0)
Females (n=57)	28	17	12
Toilet trained at first UTI	0/28 (0)	14/17 (82)	12/12 (100)
Children with recurrent UTI (n=22, 45%)	11/28 (39)	4/17 (24)	7/12 (58)

Key: UTI = urinary tract infection
Numbers in parentheses are percentages.

Comment

Our study sought to determine the natural history and risk factors for recurrent UTIs in children with a normal anatomic workup. We observed that girls had a statismically significant risk of recurrent febrile UTI compared with boys, even those who were uncircumcised. We found that the age at the time of the first UTI was not predictive of recurrence. It also appeared that UTIs in girls aged 5 years and older were more often associated with dysfunctional elimination.

An inherent weakness of this study was the inability to confirm the results of those urine samples that were bagged specimens. It is possible that some of these specimens had been contaminated. However, all these children were febrile and if not seen at our institution at the time of illness were evaluated by their pediatrician.

In children with symptomatic UTIs, approximately one half to two thirds have normal radiographic studies.[7] Our findings are comparable with those studies that evaluated recurrence in children with an anatomically normal urinary tract. Siegel et al.[8] identified 52 of 811 infants and 660 preschool children who presented with a symptomatic UTI and had a normal radiologic evaluation. They reported that these children had recurrent infections that primarily occurred within the first year. How-ever, the infections did not occur during the next 3 years.[8] Jodal and Winberg[9] similarly reported a 50% recurrence rate in girls irrespective of the character of the first infection. In boys, recurrence was less common and was rarely seen after 2 years of age.[9]

The treatment of these patients with normal anatomy and recurrent UTI is best determined by the sex of the patient In our study, most (92%) boys who presented with a febrile UTI were uncircumcised. This finding is in agreement with several investigators who have shown that the number of uropathogenic organisms surrounding the urethra is increased in uncircumcised boys during the first 6 months of life.[10] In a recent retrospective cohort study, Schoen et al.[11] demonstrated that circumcision is associated with a 10-fold reduction in the incidence of having a UTI during the first year of life. Although the foreskin may be a risk factor for recurrence, the rate of recurrence in our study was small (14%). Thus, we recommend treatment for phimosis in boys with multiple UTIs, for whom the morbidity of infection outweighs the risks of circumcision.

Girls are at a greater risk of recurrent UTIs. In younger girls, treatment is complicated by the fact that girls younger than 1 year of age who present with a febrile UTI are as likely to have a recurrent UTI as those who present at 2 mo 5 years of age and are toilet trained. It can be argued that in younger boys and girls, voiding dysfunction plays a role in the recurrence of UTIs. Voiding dysfunction in in-fancy has been described. Several studies have related the occurrence of detrusor-sphincter dyssynergia and bladder instability as being a normal part of bladder maturation.[12-15] In our instimution, we do not routinely evaluate infants for voiding dysfunction. Severe voiding dysfunction (non-neurogenic neurogenic bladder) is a rare enmity. Jayanthi et al.[16] reported the occurrence of non-neurogenic neurogenic bladder in only 7 infants ranging in age from newborn to 30 months during a 16-year period. In addition, all these infants had radiographic findings suggestive of bladder dysfunction, including a thickened bladder wall, hydronephrosis, or vesicoureteral reflux. None of our patients were found to have any radiographic evidence of voiding dysfunction and, as such, this diagnosis was not pursued in children who were not yet toilet trained.

However, we agree that in those very few infants who are neurologically normal and have the above radiographic findings, anticholinergic medications and intermittent catheterization may be required. This subgroup of younger patients may also benefit from low-dose antibiotic prophylaxis through the time of toilet training. However, it should be noted that good long-term evidence supporting the efficacy of prophylactic antibiotic use in the prevention of recurrent UTIs is not available.[17]

In older girls, we recommend a careful inquiry into their voiding habits. In our study, 7 of the 12 girls who were 5 years or older when they presented with a febrile UTI demonstrated dysfunctional voiding and elimination. Almost all cases were demonstrated clinically without the use of urodynamic studies. Dysfunctional voiding is due to bladder instability, infrequent voiding, or, in its most severe form, non-neurogenic neurogenic bladder. Often this type of behavior will lead to an in-crease in bladder pressure. Published reports support an association between voiding dysfunction and UTI recurrence.[18] Constipation has also been linked to voiding dysfunction and UTI recurrence. O'Regan et al.[19] studied 47 children with recurrent UTIs and an-atomically normal urinary tracts. All the children studied showed signs of constipation. These investigators demonstrated that treatment of the constipation and voiding dysfunction appeared mo result in a decrease in the infection rate.

To diagnose dysfunctional elimination, we recommend the use of a simple 48-hour voiding diary. The time and the amount of urine voided should be recorded, as well as the number of wet episodes in between voids. From this information, we have been able to differentiate among the various etiologies without resorting to urodynamic studies. As a part of our routine workup, we inquire about the child's bowel habits, in particular the frequency and consistency of the stool. Treatment is based on the results of the voiding diary and includes timed voiding, double voiding, relaxation, leg abducmion with voiding, and anticholinergics and antibiotics when appropriate. In addition, constipation is aggressively treated with a regimen consisting of oral laxatives and fiber. Of note, in this study we did not evaluate children younger than 5 years of age for constipation. It is possible that the identification and treatment of fecal retention in this age group would have some impact on the recurrence of UTI.

In a recent evaluation of 140 patients seen at our institution for dysfunctional voiding or elimination, most were treated with the above methods. Thirty-eight patients had UTIs and 27 had known constipation. After 1 year of follow-up, 95% showed improvement in the incidence of UTIs, and 86% reported an improvement in constipation (personal communication, A. Hinds). In our experience, the identification and aggressive treatment of voiding dysfunction and dysfunctional elimination leads to a decreased incidence of UTI recurrence in girls 5 years of age and older.

Conclusions

Our results suggest that in children who present with a febrile UTI and have a normal radiologic evaluation, recurrence is more common in girls. However, uncircumcised boys are at a greater risk of infection during mhe first year of life. In the case of repeat UTI, these boys may benefit from treamment of phimosis. In girls, the age at the time of the first UTI was not predictive of recurrence. Recur-rent UTI in children 5 years of age and older was associated with dysfunctional elimination. These children can he evaluated with the use of a simple voiding diary, avoiding the need for urodynamic studies. Successful treatment includes timed voiding, anticholinergics, and adherence to a bowel regimen, without the need for biofeedback. Dysfunctional voiding may contribute to recurrent infection in younger children; however, no radiographic evidence supported this. Recurrent infection in this age group may respond to treatment with prophylactic antibiotics until the child is toilet trained.

References

1. Marild S: Aspects of pathogenesis and renal development in childhood pyelonephritis (Thesis). Goteborg, Vasastadens Bokbinderi AB. 1989.
2. Winberg J, Andersen HJ, and Bergstrom T, et al: Epidemiology of symptomatic urinary tract infection in childhood. Acta Paediatr Scand 63 (suppl 252): 1-20. 1974.
3. Schlager TA, and Loher JA: Urinary tract infection in outpatient febrile infants and children younger than 5 years of age. Pediatr Ann 22: 505-509. 1993.
4. Spenser JR, and Schaeffer AJ: Pediatric urinary tract infections. Urol Clin North Am 13: 661-672. 1986.
5. Weiss R, Tamminen-Mobius T, Koskinties O, et al: International reflux study in children. J Urol 148: 1644-1734. 1992.
6. Hoberman A, Charron M, Hickey R, et al: Imaging studies after a first febrile urinary tract infection in young children. N Engl J Med 348: 195-202, 2003.
7. Jodal U: The natural history of bacteriuria in childhood. Infect Dis Clin North Am 1: 713-729, 1987.
8. Siegel SR, Siegel B, Sokoloff BZ, et al: Urinary infections in infants and preschool children: five year follow-up. Am J Dis Child 134: 369-372, 1980.
9. Jodal U, and Winberg J: Management of children with unobstructed urinary tract infection. Pediatr Nephrol 1: 647-656, 1987.
10. Wiswell TE, Miller GM, and Gelston HM: Effect of circumcision status on periurethral bacterial flora during the first year of life. J Pediatr 113: 442-446, 1988.
11. Schoen EJ, Colby CJ, and Ray GT: Newborn circumcision decreases incidence and costs of urinary tract infections during the first year of life. Pediatrics 105: 789-793, 2000.
12. Yeung CK, Godley ML, Ho CK, et al: Some new insights into bladder function in infancy. Br J Urol 76: 235-240, 1995.
13. Sillen U, Bachelard M, Hansson S, et al: Video cystometric recording of dilating reflux in infancy. J Urol 155: 1711-1715, 1996.
14. Sillen U, Bachclard M, Hermanson G, et al: Gross bilateral reflux in infants: gradual decrease of detrusor hypercontractility. J Urol 155: 668-672. 1996.
15. Chandra M. Maddix H. and McVair M: Transient urodynamic dysfunction of infancy: relationship to urinary tract infections and vesicoureteral reflux. J Urol 155: 673–677. 1996.
16. Jayanthi VR, Khoury GA, McLorie GA, et al: The non-neurogenic neurogenic bladder of early infancy. J Urol 158: 1281-1285. 1997.
17. Le Saux N, Pham B, and Moller D: Evaluating the benefits of antimicrobial prophylaxis to prevent urinary tract infections in children: a systematic review. Can Med Assoc J 163: 523-529, 2000.
18. Koff S, Wagner TT, and Jayanthi VR: The relationship among dysfunctional elimination syndromes, primary vesicoureteral reflux and urinary tract infections in children. J Urol 160: 1019-1022, 1998.
19. O'Regan S, Yazbeck S, and Schick E: Constipation, bladder instability, urinary tract infection syndrome. Clin Nephrol 23: 152-154,1985.