Clear question with an unclear answer: What is urinary tract infection in a preterm infant?

Urinary tract infections (UTIs) are a common cause of late-onset sepsis (LOS) in preterm infants.1, 2 While their true incidence is difficult to ascertain, clinicians have been encouraged to look for UTIs in the workup of LOS. There are, however, no clear guidelines on the diagnosis and management of UTIs in these patients which may lead to significant variations in diagnostic practices. The most glaring issue is the lack of a consensus definition of UTIs in preterm infants. Several challenges unique to preterm infants make this gap especially problematic. For instance, the most common manifestations of LOS in preterm neonates are increased respiratory support requirements, apnoea, bradycardia, and temperature instability, often with hypothermia. These symptoms are difficult to interpret because, while they can be the result of an infectious process, they can also be attributed to diseases caused by prematurity itself such as apnoea of prematurity and chronic lung disease of prematurity. This differs from the term newborn where the most common manifestation of LOS is fever, a reliable sign of an underlying infectious process. Since clinical signs do not help to distinguish an infectious from a non-infectious aetiology, microbiological criteria are paramount to making a precise diagnosis. However, contrary to term newborns, these criteria are not well established in preterm neonates. Most paediatric societies agree on the case definition for UTI in children 2 months of age and older, namely the growth of a significant amount of a single uropathogen—either ≥50 000 colony-forming units (CFU)/mL in a sample obtained via urine catheterisation (UC), ≥1000 CFU/mL for suprapubic aspiration (SPA) or ≥100 000 CFU/mL for clean catch samples—in the presence of compatible symptoms. Whether this applies to preterm infants is unclear as the studies from which this case definition was derived did not include these patients. The onus is therefore placed on the clinician to extrapolate concepts from the current body of evidence and make a case-by-case decision. Unfortunately, the literature on UTIs in preterm infants is difficult to interpret as many studies include a small number of patients with variable case definitions, often somewhat different from that of paediatric societies, and recent large-scale retrospective studies provide little information on their case definitions.3 We will examine some key elements of the diagnosis of UTIs to illustrate some of the unique challenges in preterm newborns. A look at three definitions used in studies on UTIs in preterm infants highlights important questions regarding the notion of ‘significant growth’ in preterm infants, a key aspect of the diagnosis of UTIs. Mohseny et al. used various definitions based on the method of collection: ≥10 000 CFU/mL of a single organism for UC samples, ≥100 000 CFU/mL for bagged samples (which are not accepted by most paediatric societies) and any growth for SPA.2 Ruangkit et al. defined UTIs as the growth of ≥1000 CFU/mL of up to two organisms in a sample obtained by SPA or UC.4 Walawender et al. relied exclusively on UC and defined UTIs as ≥50 000 CFU/mL of a single organism or ≥10 000 CFU/mL in the presence of leukocyte esterase or nitrites on urinalysis.5 The main questions here are (i) the ideal sampling method in preterm infants, (ii) whether dual microbial growth represents a true UTI, (iii) what CFU cut-off should be used and (iv) the role of the urinalysis in the diagnosis of UTIs in preterm infants. The diagnosis of UTI is generally made using a urine sample obtained under sterile conditions. In neonates, clinicians have traditionally relied on UC and SPA. Occasionally, some authors have accepted collector bags and midstream urine samples. A LOS workup is usually initiated in the presence of significant clinical symptoms and empiric antibiotics are often administered following prompt completion of the workup. As such, considering the high likelihood of contamination with both collector bag and midstream urine samples in preterm infants, these methods should be avoided. UC and SPA remain the sampling methods of choice. UC may be preferable to SPA as an initial sampling method as it is more likely to be successful and requires less technical expertise although with slightly higher contamination rates.6 Ultrasound guidance has been cited as a method to optimise SPA success rates and may make SPA preferable to UC in the hands of a skilled operator.7 Regardless of which method is chosen between UC and SPA, consideration for the alternate technique should be made if initial sampling is unsuccessful. Dual growth is not accepted in traditional consensus case definitions and is usually thought to be due to sample contamination. While obtaining sterile urine samples in older children has its challenges, obtaining them in preterm neonates is notoriously difficult. Contamination of UC samples may be as high as 50% in preterm newborns compared to only 10% in children and adults.8 In this context, clinicians may feel that it is a reasonable trade-off to expose some infants to unnecessary antibiotics in an effort not to miss a true invasive bacterial infection. Although the trade-off may seem reasonable, clinicians should remain sceptical of urine cultures with dual microbial growth, irrespective of the organisms isolated and their concentration. If the clinical context is highly suggestive of sepsis, a repeat urine culture should ideally be obtained by a reliable collection method or additional efforts to identify an alternative focus of infection should be made. The American Academy of Paediatrics states in its 2011 guideline on the diagnosis of UTIs: ‘Definitions of positive and negative culture results are operational and not absolute’. In essence, CFU cut offs were established to strike a reasonable balance between sensitivity and specificity. This balance was determined by a study by Hoberman et al. which showed that in children less than 2 years of age using a cut off of 50 000 CFU/mL for a catheterised sample significantly reduced the rate of dual growth and growth of non-uropathogens, especially in the setting of a positive urinalysis.9 The 100 000 CFU/mL cut off used for midstream samples was based on studies in adult women comparing colony counts in women with and without clinical symptoms compatible with acute pyelonephritis.10 It is difficult to know whether these cut offs can strike the same statistical balance in preterm infants. Considering that obtaining a urine sample under sterile conditions is difficult in these patients, it may be reasonable to consider the growth of a single uropathogen with a CFU count of >10 000/mL via UC as significant and provide treatment for a suspected UTI. This is also supported by a growing body of research that has shown that lower cut offs seem to have marginal effects on the sensitivity and specificity of our tests.10 The presence of leucocyte esterase, nitrites or pyuria certainly strengthens the diagnosis as this suggests the presence of bacteria in the urinary tree in sufficient quantities to produce a local inflammatory reaction. The main issue is that there are known differences between organisms in their propensity to yield a positive urinalysis. For example, while E. coli are frequently associated with pyuria and nitrites, organisms that are thought to frequently cause UTIs in preterm infants, such as Enterococcus and Pseudomonas, seem to cause pyuria in a smaller proportion of cases and do not produce nitrites.1, 2, 5 Furthermore, criteria for positive urinalysis were based in part on cut offs of 50 000 CFU/mL and may not be as accurate if lower thresholds are used. All things considered, a positive urinalysis increases the likelihood of a true UTI but a negative urinalysis does not rule it out in this population. This discussion highlights how little guidance neonatal clinicians have regarding the diagnosis of UTIs in preterm infants and the many pitfalls research on this topic has shown. Unnecessary and prolonged antibiotic use in preterm infants has been associated with increased risk of complications such as NEC and can have a significant impact on the microbiome and potential long-term health issues. Units should seek to establish local standardised approaches to UTI management to limit antibiotic use based on the limited available evidence while further research is conducted on this topic. Local initiatives should involve neonatal and infectious disease clinicians as well as medical microbiologists. We hope that the elements in this commentary will help guide these discussions and steer future research aimed at establishing a more universal consensus definition in the NICU population. Assil Abda: Conceptualization; writing – original draft; writing – review and editing. Christian Lachance: Conceptualization; supervision; writing – review and editing. Caroline Quach: Conceptualization; supervision; writing – review and editing. The authors declare no conflicts of interest. Assil Abda Christian Lachance Caroline Quach

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.