Urinary tract infections (UTIs) are a common and important clinical problem in childhood and may lead to systemic illness and renal injury in the short term; with repeated infections, renal scarring, hypertension, and end-stage renal dysfunction that may develop.

The overall prevalence of UTI is estimated at 5% in febrile infants but varies widely by race and gender.^1,2 The highest prevalence rates of childhood UTI occur in uncircumcised male infants under 3 months of age (prevalence ~20%), and among females (prevalence ~8%). Circumcised older male children have the lowest prevalence of UTI (~1%) (Table 104-1).

Most UTIs beyond the newborn period represent an ascending infection. Colonization of the periurethral area by uropathogenic enteric organisms is the first step. The most common bacterial species is Escherichia coli, which accounts for about 80% of UTIs in children. Other bacteria include both gram-negative species (Klebsiella, Proteus, Enterobacter, and Citrobacter) and gram-positive species (Staphylococcus saprophyticus, Enterococcus, and rarely, Staphylococcus aureus). Attachment of bacteria to uroepithelial cells is an active process mediated by specific bacterial adhesions and specific receptor sites on the epithelial cells. This process allows bacteria to ascend into the kidney, even in children without vesicoureteral reflux (VUR). In the kidney, the bacterial inoculum can produce an infection with an intense inflammatory response that may ultimately lead to renal scarring.

Many host factors influence the predisposition that children may have to UTI, including familial predisposition, genitourinary anatomy and function, instrumentation, and sexual activity as well as periurethral flora. Determining risk factors in a child presenting with UTI is important in preventing further recurrences.

First-degree relatives of children with UTIs are more likely to have UTIs,³ and adherence of bacteria may, at least in part, be genetically determined.

Uncircumcised febrile male infants have a four- to tenfold higher prevalence of UTIs than circumcised males do.⁴ Although uncircumcised males are at increased risk for the development of a UTI, it is important to point out that UTIs do not develop in most uncircumcised boys.⁵ It is estimated that 195 circumcisions would be needed to prevent one hospital admission for UTI in the first year of life.⁴

Bladder and bowel dysfunction (BBD) refers to a functional disorder of unknown etiology characterized by the features delineated in Table 104-2. BBD, also known as dysfunctional elimination syndrome or voiding dysfunction, usually appears in healthy school-aged children and may persist for months to years. Although BBD is a relatively common condition in the pediatric population, with a prevalence estimated at 15%,⁶ it is often underdiagnosed and undertreated by primary care physicians.⁷ Approximately 40% of toilet-trained children with their first UTI^8-10 and 80% of children with recurrent UTI¹¹ report symptoms of BBD. In a study of 141 girls older than 3 years with recurrent (more than three) UTIs, 108 had BBD.¹¹ This syndrome is also a risk factor for persistence of VUR^12-14 and for renal scarring.^9,14

VUR is the most frequently occurring urologic abnormality in children, with an overall prevalence of 1% and a prevalence of 40% in young children with febrile UTIs. A strong genetic predisposition for VUR exists and has been identified in up to 40% of siblings of children with VUR. The incidence of VUR is markedly lower in the African American population. VUR is graded according to the international classification from grade I to grade V (Figure 104-1). In children presenting with their first UTI to a primary care physician, 95% of cases of VUR have been found to be grades I to III.¹⁵ Most VUR improves or resolves as the child ages, with resolution occurring more frequently in children with low-grade and unilateral disease.^16,17 The likelihood of resolution of VUR, 5 years after detection, as derived from a large meta-analysis, is summarized in Figure 104-2.¹⁸

Current evidence implies that VUR is neither necessary nor sufficient to cause renal scarring and that exclusive focus on VUR, without a search for other modifiable risk factors (such as BBD), may be inadequate. Nevertheless, the risk of renal scarring increases in children with VUR compared with children without VUR (41% versus 17% in a systematic review) and increases with increasing grades of VUR.¹⁹

Children with obstructive abnormalities, whether anatomic (e.g. posterior urethral valves, ureteropelvic junction obstruction, constipation), neurologic (e.g. myelomeningocele with neurogenic bladder), or functional (e.g. BBD), are at increased risk for the development of UTIs. Stagnant urine is an excellent culture medium for most uropathogens. However, obstructive anatomic abnormalities in children presenting with their first UTI are infrequent (1% to 4%) in nonsyndromic children.^15,20-22 Obstruction should be suspected when other family members have had urologic abnormalities, when dysmorphic features are detected on physical examination, or when symptoms do not respond to appropriate therapy.

There is indirect evidence that alteration of the normal periurethral flora in females (Lactobacillus and Corynebacterium spp.) promotes attachment of pathogenic bacteria. Sexual activity is also associated with UTI in women. In a prospective study of young sexually active women, recent intercourse was independently associated with the development of UTIs.²³ The use of spermicidal condoms and spermicidal jelly with diaphragms has been independently associated with E. coli bacteriuria, thus suggesting that these agents predispose to UTI by altering the vaginal flora.²³

The risk for UTI increases with increasing duration of catheterization. In a 7-year retrospective study, nosocomial UTIs were found to be the fifth most common infection in hospitalized children. However, only 50% of children with nosocomial UTIs were catheterized, and nosocomial UTIs occurred with a disproportionately high rate in newborns.²⁴

In general, the younger the child, the less specific the presenting signs and symptoms of UTI; many infants with UTI are described as “well appearing.”

Several prospective studies have shown that infants and young children can present with fever as the sole manifestation of UTI.^1,2 Furthermore, the presence of another potential source for fever (upper respiratory tract infection, acute otitis media, acute gastroenteritis, etc.) does not rule out the possibility of UTI. In one study, the prevalence of UTI in infants and young girls with fever was 6% if no other potential source of fever was identified, but the prevalence was 3% even if another potential source of fever was identified.² This finding highlights the importance of considering urine cultures in all febrile infants without a definite source for fever.

In a meta-analysis of the diagnostic accuracy of the symptoms and signs of UTI in children younger than 2 years, the following symptoms and signs were the most helpful in identifying children with UTI:²⁵

• 
  

  History of UTI (likelihood ratio 2.3)

  
  
• 
  

  Temperature >40°C (LR 3.2)

  
  
• 
  

  Suprapubic tenderness (summary LR 4.4)

  
  
• 
  

  Lack of circumcision (summary LR 2.8)

Combinations of signs and symptoms were more useful than individual signs for identifying children with and without UTI:²⁵

• 
  

  Temperature >39°C for ≥48 hours in absence of another source for fever (LR 4.0)

  
  
• 
  

  Temperature <39°C, presence of another source for fever (LR 0.37)

In verbal children, the following symptoms and signs were the most helpful in identifying children with UTI:²⁵

• 
  

  Abdominal pain (LR 6.3)

  
  
• 
  

  Back pain (LR 3.6)

  
  
• 
  

  Dysuria, frequency, or both (LR 2.2)

  
  
• 
  

  New-onset urinary incontinence (LR 4.6)

The decision to obtain a urine sample for culture is best made on a case-by-case basis, taking into consideration the child’s age, gender, circumcision status, and the presenting signs and symptoms (Figure 104-3).

In general, urine samples for urinalysis (dipstick and microscopic examination) and culture are indicated in the following patients:

• 
  

  Girls and uncircumcised boys younger than 2 years with at least one risk factor for UTI (history of UTI, temperature >39°C, fever without apparent source [particularly if the child will be treated with antibiotics], ill appearance, suprapubic tenderness, fever >24 hours, or nonblack race).

  
  
• 
  

  Circumcised boys younger than 2 years with suprapubic tenderness or at least two risk factors for UTI (history of UTI, temperature >39°C, fever without apparent source [particularly if the child will be treated with antibiotics], ill appearance, suprapubic tenderness, fever >24 hours, or nonblack race).

  
  
• 
  

  Girls and uncircumcised boys older than 2 years with any of the following urinary symptoms: abdominal pain, back pain, dysuria, frequency, high fever, or new-onset incontinence.

  
  
• 
  

  Circumcised boys older than 2 years with multiple urinary symptoms (abdominal pain, back pain, dysuria, frequency, high fever, or new-onset incontinence).

  
  
• 
  

  Febrile infants and children with abnormalities of the urinary tract or family history of urinary tract disease.

The differential diagnosis of a well-appearing infant or young child with “fever without a definite source” is extensive but most commonly includes UTI, occult bacteremia, and viral infections. In children vaccinated against Haemophilus influenzae and Streptococcus pneumoniae, the odds of UTI are much higher than the odds of occult bacteremia. The differential diagnosis of an older child presenting with urinary symptoms and bacteriuria includes nonspecific vulvovaginitis, an abdominal process such as appendicitis, urinary calculi, urethritis secondary to sexually transmitted disease, and a vaginal foreign body. Patients with group A streptococcal infection, appendicitis, and Kawasaki disease may present with fever, abdominal pain, and pyuria.