Infant UTI: A Mountain or a Molehill?

Ricardo A. Quinonez, MD


August 31, 2015


Is a UTI in an infant or young child a serious bacterial infection?

Ricardo A. Quinonez, MD
Associate Professor of Pediatrics, Baylor College of Medicine; Chief, Division of Pediatric Hospital Medicine, Children's Hospital of San Antonio, San Antonio, Texas

For years and since the publication of landmark studies[1,2] of fever in children, urinary tract infections (UTIs), particularly in infants, have been classified right alongside bacteremia and meningitis as "serious bacterial infections," or SBIs. A better understanding of long-term outcomes of children who have experienced UTIs, the continuous debate over UTI prophylaxis, and other recent studies have called into question the need to aggressively pursue the diagnosis of UTI in children. A comprehensive review[3] of the literature, published in 2013, analyzed studies regarding prevalence of UTI; risk for sepsis in infants and young children with UTI; association of UTIs with renal scarring; and the subsequent, related issues of whether prevention of UTI and possible renal scarring mitigates any potential long-term negative sequelae such as hypertension or renal failure.

First is the question of prevalence. In children younger than 18 months of age evaluated for unexplained fever, UTIs are relatively common, documented in between 2% and 7.5% of cases.[4]

How common is sepsis in children with a UTI? A population-based review[5] found that UTI or pyelonephritis was the documented source in less than 1 in 20 cases of sepsis; the majority (over 70%) of these children were infants and about half had urinary tract anomalies. The fatality rate for these cases of urosepsis was 3.7%, which is significantly lower than the rate for other causes of sepsis in children.

The American Academy of Pediatrics (AAP) guideline[6] recommends early treatment of documented and suspected UTI in order to reduce the likelihood of renal scarring and subsequent chronic kidney disease (CKD). However, a number of randomized controlled trials (RCTs) and observational studies have demonstrated that treatment does not lessen the development of renal scarring.[3] Notably, a prospective RCT[7] involving 287 children reported that scarring occurred in 30% of children with UTI, regardless of whether antibiotics were started early or late in the course of the febrile illness. In contrast, an observational study[8] involving 230 children with a first febrile UTI found that delays of more than 4 days between presentation of fever and initiation of treatment did increase scarring in children with both refluxing and nonrefluxing UTI.

From available evidence, it is safe to conclude that UTIs are associated with subsequent development of renal scarring. However, does that scarring increase the risk for long-term negative outcomes such as hypertension and CKD? Two prospective RCTs and a systematic review have examined that question. The first RCT,[9] published in 1996, followed 111 females for a mean of 15 years following a UTI, and found no clinically significant differences in glomerular filtration rate (GFR) or incidence of hypertension between those with moderate or severe scarring and healthy controls. A second RCT,[10] involving over 1200 children with symptomatic UTI, followed these patients for an average of 25 years. The overwhelming majority (93%) of the 57 participants in the original sample who were available for long-term follow-up had evidence of persistent renal scarring on repeated dimercaptosuccinic acid (DMSA) scan. There was no statistically significant difference, however, in GFR or incidence of hypertension between these participants and healthy controls. A systematic review[11] involving over 1500 patients with CKD examined the association between childhood UTI and subsequent CKD and found no cases directly related to UTI. These same authors also examined a regional database in Finland and reviewed the records of patients for whom the cause of CKD was undetermined; they found no patients for whom UTIs appeared to be causal.

Another reason given to aggressively pursue UTI diagnosis is that doing so may identify vesicourethral reflux (VUR), which can be treated either medically or surgically. The recent publication of the Randomized Intervention for Children with Vesicoureteral Reflux (RIVUR) study,[12] as well as the prior publication of the Prevention of Recurrent Urinary Tract Infection in Children with Vesicoureteric Reflux and Normal Renal Tracts (PRIVENT) trial,[13] has demonstrated that treating recurrent UTIs with prophylactic antibiotics indeed can prevent the recurrence of UTIs (number needed to treat [NNT] in RIVUR of 8, NNT in PRIVENT of 14). However, the important outcome of renal scarring was not prevented in either trial. Also, these trials put into question the importance of the presence of VUR on the ability of antibiotics to prevent UTI recurrence. In the PRIVENT trial, for example, which included children with and without VUR, there was no difference in the efficacy of antibiotics in either of these two groups. In both of these trials, the incidence of resistant bacteria to common antibiotics was significantly higher in the prophylaxis groups.

Therefore, knowing that a child has VUR makes no difference in renal scarring and apparently it makes no difference in our ability to prevent UTIs, because in children with and without VUR, prophylaxis prevents UTIs. However, surely for the most severe cases at least surgery could be offered if we know that a child has VUR. In at least one Cochrane review,[14] which included 20 RCTs, the NNT for surgery to prevent one UTI at 5 years of age was 8. There is also no current evidence that surgical correction prevents important outcomes such as scarring or CKD. Taken together, all of these facts, along with the one that most cases of VUR self-resolve, not only put into question the importance of prophylaxis for UTIs, but they also (and more importantly) question the necessity of detecting VUR at all.

A final question involves the possible association of UTI with other conditions. Previous studies have suggested a link between UTI and apparent life-threatening events (ALTE) and respiratory syncytial virus (RSV) bronchiolitis. Studies suggesting a link with RSV were conducted prior to the publication of the 2011 AAP guideline[6] diagnostic criteria which require both a positive urine culture and a urinalysis with evidence of pyuria in order to diagnose a UTI. A subsequent study[15] calculated UTI rates in children with RSV infection using both the older and the more recent diagnostic criteria, and found that the previous association disappeared. Similarly, all of the studies that have shown an association of UTI with ALTE used the older criteria to define a UTI, which required only a positive urine culture, rather than a culture and pyuria, to establish the diagnosis. In those earlier studies linking UTI to other conditions, the rate of asymptomatic bacteriuria in children is similar to the rate of UTI. Thus, it is probably a safe assumption that this association is a mirage and that what the authors were actually detecting were cases of colonization.

What are the conclusions from these studies?

  • UTIs rarely progress to sepsis.

  • Treating UTIs early does not seem to affect outcomes.

  • UTIs are not a significant predisposing factor of CKD.

  • Treating VUR either medically or surgically does not improve meaningful outcomes.

  • The association of UTIs with other clinical entities is often a mirage.

To return to the original question: Are UTIs a serious bacterial infection? The landmark study by Newman and colleagues[3] and other subsequent publications would suggest that the answer is a resounding no. It’s hard to disagree that treating a true UTI probably relieves symptoms and makes a patient better faster, and in many cases that may have value to individual patients or families. However, knowing that UTIs are more akin to otitis than to bacteremia or meningitis really brings into question the need to aggressively pursue the diagnosis.

The lesson with this information? The next time one is faced with an infant with fever, shared decision-making with a family should include all of these considerations with a pause before ordering that next cath.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: