The American Academy of Pediatrics (AAP) Practice Parameter on Urinary Tract Infections that is presented in this issue of American Family Physician1 is a timely one. Recent studies have brought into question some of the accepted standards in managing these patients, and many of these issues are addressed by the Practice Parameter.
Considering the diagnosis of urinary tract infection (UTI) in a child with an unexplained fever seems obvious, but confirming infection is problematic. Suprapubic aspiration is the recommended means of obtaining a culture specimen2,3; however, it is unrealistic to expect this practice to be widely accepted. The procedure is anxiety-provoking for the parent and the physician.
Urethral catheterization is a more realistic approach when an accurate diagnosis of UTI is essential, which is the case for most febrile infants. However, the “bag” specimen has its role in the diagnosis of pediatric UTIs. The parameter guidelines1 seem reasonable in this regard. They suggest that a bag specimen may be obtained in children who do not require immediate therapy. In this setting, if the urinalysis, culture and clinical parameters are taken together, the results can usually be interpreted in a reliable way. However, a bag specimen is not optimal in patients who are given antibiotics; the specimen may be contaminated and, after the administration of antibiotics, the opportunity to get another specimen is gone because the infection will be partially treated.
Urinalysis and culture should always be obtained. A microscopic analysis of the urine to quantify the degree of pyuria is particularly important in distinguishing infection from contamination when the culture is equivocal. Conversely, a negative dipstick urinalysis cannot be relied on to rule out a UTI.
The issue of “asymptomatic bacteriuria” is an important and often misunderstood one. It is well established that asymptomatic bacteriuria is common, difficult to eradicate and of no significance to the patient in the absence of vesicoureteral reflux or other urinary tract anomalies.4
The potentially harmful practice of “routine cultures” in asymptomatic patients with a history of UTI is widespread and should be discouraged. These cultures, if positive, will lead to the inappropriate treatment of bacteriuria. This exposes the child to unnecessary antibiotics and select resistant organisms, which will make future symptomatic infections more difficult to treat. Once a child's urinary tract has been shown to be normal by the appropriate radiographic studies, cultures should only be obtained if the child is symptomatic.
By far, the most controversial area in the management of pediatric UTIs is evaluation with imaging studies. Recently, the connection between vesicoureteral reflux and the long-term risks of hypertension and renal failure has been questioned.5
This has led some authors to suggest that vesicoureteral reflux is not an important entity and that voiding cystourethrograms (VCUGs) may be unnecessary in children with UTIs.5 The argument against the routine use of VCUGs in children with UTIs starts with the premise that many, if not all, of the renal changes seen in children with reflux are actually congenital areas of dysplasia rather than acquired scars resulting from infections. Furthermore, no strong evidence exists linking acquired renal scarring (as opposed to congenital dysplasia) and long-term risk of hypertension and renal failure. Also, evidence is lacking to suggest that medical or surgical management of reflux is any better at preventing long-term complications than simply treating UTIs aggressively when they occur. The argument concludes that VCUGs are unnecessary because there is no proven benefit to aggressive management of reflux.
While this argument raises some important issues, it does not tell the whole story. While some renal abnormalities in children with reflux are undoubtedly congenital, there is evidence that some of these changes are acquired. In 1992, researchers6 demonstrated in a prospective fashion that approximately 40 percent of patients with pyelonephritis confirmed by dimercaptosuccinic acid (DMSA) renal scan will develop a new scar in the same location as the area of inflammation on the original scan. Interestingly, although children with vesicoureteral reflux are more likely to get pyelonephritis, once pyelonephritis occurs, the risk of subsequent scarring is independent of the presence or absence of reflux.2
Thus, vesicoureteral reflux promotes renal scarring because it predisposes to pyelonephritis. The long-term significance of these acquired scars is indeed unclear and should be studied. However, to assume that this loss of normal renal tissue is a benign event in the absence of evidence seems unreasonable. Because reflux clearly puts a child at risk for renal injury, the burden is on those who would ignore the entity to prove that it is of no significance. Until such time, it seems reasonable to look for and treat reflux aggressively. Reflux occurs in 30 to 50 percent of children with a UTI.7 A renal ultrasound is normal in 75 percent of children with reflux.8 Therefore, children with a UTI should undergo a VCUG if reflux is thought to be an important entity.
The AAP Practice Parameter serves as a good start in developing a practical approach to evaluating and managing children from two months to two years of age with a febrile illness. These are children at highest risk for reflux and most likely to suffer renal scarring if pyelonephritis occurs. As children get older, the significance of reflux probably diminishes because new renal scarring appears to be less likely in older children. Furthermore, VCUGs appear to be more upsetting for older children. The obvious question is: “At what age does the morbidity of the VCUGs and subsequent treatment of reflux outweigh the risk of undetected reflux?” A reasonable approach, given current information, would be to obtain a VCUG in any child who has a UTI before toilet training. In older children, VCUG could be reserved for those with febrile or recurrent infections. However, studies on the morbidity associated with VCUGs are needed. The morbidity appears to be related to that of urethral catheterization. Thus, it is interesting that practitioners who would not hesitate to catheterize a child to obtain a reliable urine sample agonize over the issue of whether to obtain a VCUG.
The Practice Parameter of the AAP9 offers a sound approach to febrile infections in infants and young children. In the absence of clear-cut data regarding UTIs in children, we should be conservative and do our best to protect children from the possibility of renal injury. Because we know that pyelonephritis can lead to irreversible renal damage, it seems prudent for the time being to treat reflux aggressively. However, we should not hesitate to question established approaches to these problems in an appropriate fashion—with well-designed prospective studies. In this way, we can further refine our management of this relatively common problem.