Urinary tract infections (UTIs) are a frequent cause of fever in infants and young children, accounting for more than 1.1 million visits annually, and occurring in 2.4% to 2.8% of all children.^1,2 The prevalence of UTIs varies with age and gender. The prevalence of UTIs in febrile children less than 2 years of age is 7%, and in children less than 5 years of age is 3.4%.^3,4 Urinary tract infections are divided into two overlapping categories: lower UTIs, which are limited to cystitis and urethritis, and upper UTIs, which include ureteritis, pyelitis, and pyelonephritis.⁵ Pyelonephritis can cause renal scarring, which is thought to lead to hypertension and renal failure later in life; the early recognition and treatment of an upper tract infection is thought to reduce the risk of scarring.^5–8 However, the recent literature supporting this is controversial.^9–13

Concomitant bacteremia may complicate a febrile UTI.¹⁴ In the age of effective vaccines against Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis, a UTI is the most common serious bacterial illness (SBI) of young infants.¹⁴ However, the natural history of a UTI is different than other causes of SBI, since children generally defervesce after 24 hours and have a benign course.¹⁵ A young age is the most significant risk factor for developing associated bacteremia. In infants less than 3 months of age, the rate of bacteremia with UTI has been reported between 1.5% and 9.8%. After 3 months of age, the risk of associated bacteremia decreases to less than 5%.^14,16,17

There are risk factors for developing a UTI that are important to recognize in children. Children presenting with a high fever (>39°C) or those who have greater than 24 to 48 hours of symptoms are more likely to have a UTI, particularly in the absence of an alternate source of infection.⁸ Accordingly, children without these features may be identified as being low risk for a UTI (Table 87-1).¹⁸ A prior history of a febrile UTI or a known urogenital abnormality increases the risk of developing a UTI. In children without such history, certain demographic factors may increase the risk of developing a UTI: gender, circumcision status, and race. In the first 3 months of life, boys make up about 75% of patients with UTIs, while in children older than 2 to 3 months, females predominate.¹⁸ Uncircumcised male infants less than 3 months of age have 2 to 3 times the risk of developing a UTI as do females of the same age, and 10 times the risk of circumcised males of the same age.¹⁹ After 3 months of age, febrile female infants have a higher prevalence of UTIs (2%–10%) compared with their male counterparts. Circumcised males have less than a 0.5% chance of developing a UTI after 6 months of age.³ Females have a trimodal age distribution for UTIs, with highest rates occurring in the first year of life, at the time of toilet training, and when sexual activity begins.⁵ Among sexually active adolescent girls, there are 0.5 UTIs per person-year.²⁰ Race may also play a role in the incidence of UTI; black children have a much lower incidence of UTIs when compared with children of Caucasian, Latino, or Asian descent.³

In its most recent guidelines in 2016, the American Academy of Pediatrics (AAP) makes specific recommendations regarding the testing, treatment, and follow-up of UTIs in children.⁸ These recommendations are highlighted throughout this chapter.

PATHOPHYSIOLOGY

Urinary tract infections may be caused by colonization by viruses, parasites, fungi, or bacteria. Viruses, namely adenovirus and BK virus, usually cause lower UTIs or hemorrhagic cystitis. Fungal UTIs tend to occur in immunocompromised patients or those on long-term antibiotic therapy. Bacterial UTIs are the most common in patients of all ages.

Bacterial pathogens may enter the urinary tract by one of three mechanisms: retrograde ascent from the urethra, hematogenous seeding, or by direct instrumentation or the presence of an indwelling catheter. The most common route is fecal–perineal; therefore, gram-negative organisms are the most common cause of UTIs in children. Escherichia coli is responsible for more than 80% of UTIs in children.¹⁴ Bacteria that cause UTIs, such as E. coli species, often exhibit distinctive virulence factors that overcome the normal host defenses. Gram-positive pathogens may also cause UTIs and include group B streptococci, Enterococcus spp., and Staphylococcus aureus.^3,21,22 The presence of S. aureus suggests a more invasive infection due to hematogenous spread from other sites such as abscess, osteomyelitis, endocarditis, or bacteremia. In otherwise healthy children, Lactobacillus spp., coagulase-negative Staphylococci spp., and Corynebacterium spp. are generally not considered to be pathogens. Nosocomial UTIs associated with indwelling catheters are most often caused by E. coli, Candida albicans, and Pseudomonas aeruginosa.²¹ Of note, an infection with an organism other than E. coli is associated with a higher risk of ultimately developing renal scarring.²³

The main defense mechanism against developing a UTI is the constant anterograde flow of urine from the kidneys with intermittent complete emptying of the bladder. The urine and the mucosal wall of the collecting system also have antimicrobial qualities.²⁴ Any interference with normal voiding increases the risk of developing a UTI. Chronic bowel and bladder dysfunction is an often-overlooked risk factor for developing a UTI, though it is associated with recurrent UTIs and renal scarring.²⁵

SIGNS AND SYMPTOMS

The classic symptoms of a UTI such as dysuria, urgency, increased urinary frequency, and flank pain are less frequently reported in children than in adults.³ Young children less than 2 years of age may present with fever, vomiting, or anorexia. Infants are more likely to present with nonspecific symptoms such as fussiness, lethargy, feeding intolerance, or failure to thrive. Neonates may even present without a fever and only with jaundice.¹⁵ In the older child, symptoms may become more specific, including dysuria, frequency, urgency, suprapubic discomfort, hematuria, and flank or back pain.²⁴ In some children, constipation may represent an equivalent to dysuria.⁵ Patients with abdominal or flank pain, high fever, vomiting, or other systemic signs must be evaluated for pyelonephritis or an upper tract infection.⁵ Young children with fever and UTI are presumed to have an upper tract infection due to the unreliability of signs and symptoms that distinguish these conditions in older children and adults.⁵

A thorough history and physical examination should be performed in any child suspected of having a UTI. In the ill-appearing or toxic child, a focused history should be conducted simultaneously with the physical examination, and stabilization as necessary. In addition to asking about presenting symptoms, it is important to inquire about prior episodes of febrile UTIs, known or suspected colonization with resistant pathogens, known urinary obstruction (especially renal stones), recent urinary tract instrumentation, and recent antibiotic use. The clinician should also specifically inquire about bowel and bladder emptying habits. Such symptoms include infrequent or incomplete bladder emptying, withholding maneuvers, daytime urgency–frequency syndrome, and enuresis or incontinence. In adolescent patients, consideration should be given to complications associated with sexual activity, including sexually transmitted infections and pregnancy. Adolescents should be interviewed without the presence of family or friends, to inquire about these specific risk factors.

A thorough physical examination should also be performed. The physical examination should begin by assessing the child’s overall appearance and vital signs. Overall signs of toxicity such as irritability or lethargy should prompt immediate resuscitative intervention. Particular attention should be made to the capillary refill and color, which can provide information about the child’s hydration status and perfusion. Next, the abdominal examination may reveal mild suprapubic tenderness. Any lateralizing abdominal tenderness or peritoneal signs should prompt a search for other etiologies of the patient’s symptoms. Percussion of the costovertebral angles may reveal tenderness suggestive of upper urinary tract infection; however, the absence of this sign does not rule out pyelonephritis in the younger child. In addition, careful pulmonary and ear nose and throat exams may reveal other alternate sources of the fever. Finally, all children should have a genital examination, looking for vaginal or penile discharge or foreign bodies, epididymitis, orchitis, or anatomic abnormalities.

DIAGNOSTIC EVALUATION

The decision to test for a UTI is generally straightforward in older children presenting with urinary symptoms; it may not be as straightforward in younger children. In stable children, the expected benefit of identifying and treating a UTI should be combined with pre-test probability, and must be weighed against the risk and discomfort of invasive maneuvers necessary to test for a UTI (Table 87-1). All ill-appearing toxic children, irrespective of age, should be tested for the presence of a UTI. Those children identified as being low-risk for UTI may be followed clinically, with urine obtained later in the course of their illness.⁸

The gold standard to diagnose a UTI is a urine culture from an adequate urine specimen. Other more rapid tests, such as the urine dipstick and urinalysis, may guide the clinician in real-time in the urgent or emergent care setting. There are various methods of obtaining an adequate urine specimen; the choice of method will depend on the child’s age and their risk for having a UTI (Fig. 87-1). The most accurate method of urine collection is the suprapubic aspirate (SPA).²⁶ It is the most sterile means of obtaining a urine sample, but is painful and is associated with risks such as bowel perforation.²⁷ Despite the use of ultrasound guidance, which improves accuracy and decreases associated complications, SPA is rarely performed.⁸ More commonly, a urinary catheterization is used to obtain a sterile urine sample. Success of catheterization can be improved by point-of-care ultrasound (POCUS), in order to ensure a full bladder prior to catheterization attempts (Fig. 87-2).²⁸ An alternative to catheterization is a urine bag specimen, in which a bag is attached to the perineum. This may be particularly useful in girls with labial adhesions or boys with phimosis, in whom catheterization is impossible, unsuccessful catheterization attempts have been made, or there is parental refusal of catheterization. However, it is important to note that a bagged urine specimen is the least reliable method of collection, since contamination with perineal and mucosal flora frequently leads to false positive cultures. A recent study showed that a two-step algorithm may reduce rates of urethral catheterization without prolonging the ED length of stay (Fig. 87-3).²⁹ Finally, in those children who are potty-trained, a clean-catch sample may be obtained. Parents should be instructed in the proper technique to decrease contamination, especially in girls.

Once a urine specimen is obtained, both urinalysis and urine culture should be performed. In ill-appearing children who are receiving empiric antibiotics, it is imperative that a sterile urine sample (by catheterization or SPA) be sent prior to treatment, since the urine may sterilize shortly after antibiotic administration.⁸ The urine dipstick may serve as a reliable screening tool, particularly in well-appearing febrile infants without an obvious source. In children younger than 2 years of age, the presence of both leukocyte esterase and nitrites on a urine dipstick has a positive likelihood ratio of 12.6; the absence of both carries a negative likelihood ratio of 0.13.³⁰ In children older than 2 years of age, the urine dipstick may be even more reliable, with a positive likelihood ratio of 38.54 when both are present.³¹ In addition, it is recommended that urinalysis and microscopy be performed in young children, in order to detect bacteriuria and pyuria. The results of the urinalysis should be carefully interpreted. The presence of leukocyte esterase and/or pyuria, defined as >5 white blood cells (WBCs) per high-power field (hpf) in a centrifuged specimen, or 10 WBCs per hpf in an uncentrifuged specimen, carries a sensitivity of greater than 90%; however, the specificity is poor. Many other entities can cause isolated pyuria including exercise, masturbation, appendicitis, gastroenteritis, acute glomerulonephritis, Kawasaki disease, vaginitis, and bubble bath soap. White blood cell casts on urinalysis are pathognomonic for pyelonephritis.¹⁹ The presence of urinary nitrites produced by gram-negative enteric bacteria, is highly specific (>90%) for a UTI. Yet caution should be taken in its absence, particularly in infants and children with frequent voiding habits, since this cannot be used to rule out a UTI.⁸ The presence of bacteria on Gram stain of an uncentrifuged urine sample is perhaps the single best rapid indicator of infection, with a positive likelihood ratio of 18.5 and a negative likelihood ratio of 0.07.³⁰ The gold standard for the diagnosis of UTI remains a urine culture obtained via a sterile technique, with growth of greater than 50,000 colony-forming units (CFUs) of a single uropathogenic organism.⁸

Additional laboratory studies may be indicated, especially in the ill-appearing or septic child. Blood cultures should be obtained in all children less than 2 months of age, and in ill-appearing children. Electrolytes and renal function tests should be considered on patients with signs of dehydration or toxicity. Recently, studies have investigated the use of inflammatory markers such as the C-reactive protein (CRP) and procalcitonin to rule in or rule out pyelonephritis. A low CRP makes an upper tract infection less likely; conversely, a high procalcitonin predicts upper tract involvement. However, studies have revealed heterogeneous results, and neither test can reliably predict or exclude pyelonephritis. As a result, there is insufficient data to recommend the routine use of inflammatory markers to distinguish upper from lower tract infections.³²

MANAGEMENT AND RADIOLOGIC EVALUATION OF THE URINARY TRACT

Empiric antibiotic therapy for a UTI is directed at the presumed infecting organism until results of the urine culture are obtained. The disposition of a child with a UTI depends on the patient’s age, compliance, follow-up, and the ability to tolerate oral liquids. Infants under 2 months of age should be admitted for IV antibiotics, since they have a high risk of associated bacteremia.¹⁴ Admission for IV antibiotics is also indicated in toxic-appearing children, those with evidence of sepsis, those who have failed outpatient therapy, immunocompromised patients, those at high risk for resistant organisms, children unable to tolerate oral fluids or medications, and those with inadequate follow-up or unreliable caretakers. The criteria for admission are listed in Table 87-2. Intravenous antibiotic therapy should continue until blood culture results are negative, and for a minimum of 24 hours after symptom resolution.³³