Healthcare-Associated Infections in the Neonate

Late-Onset Sepsis

Sepsis in the NICU population is difficult to define given diagnostic challenges, nonspecific clinical and laboratory signs of infection, and variability in presentation. No consensus definition for neonatal sepsis currently exists. Sepsis in the NICU population is instead categorized by timing. Early-onset sepsis is defined as a bloodstream or cerebrospinal fluid infection occurring in the first 72 hours of life and is primarily the result of exposure to organisms in the amniotic fluid or genital tract. The National Institute for Child Health and Human Development Neonatal Research Network (NICHD NRN) defines late-onset sepsis (LOS) as a bloodstream infection (BSI) occurring in neonates >72 hours of age. ^, These infections are acquired from the NICU environment or the infant’s own microbiome. Most late-onset infections therefore are categorized as HAIs.

LOS is most common among extremely preterm infants, occurring in approximately 38% of extremely low birth weight (ELBW) infants (<1000 g) who survive beyond 3 days of age. The risk for LOS increases with decreasing gestational age (GA), affecting approximately 61% of infants born at 22 weeks’ gestational age compared with 20% of infants born at 28 weeks’ gestation. Variable rates of LOS among very low birth weight (VLBW) infants (<1500 g) have been reported from Neonatal Networks in Japan (5%), Canada (17%), South America (21%), and Israel (29.5%). At the institutional level, the prevalence of LOS in VLBW infants varies even more: 11%–32% in the NICUs of the NICHD NRN and 7%–74% in the NICUs participating in the Canadian Neonatal Network. NICUs with higher rates of ELBW admissions or more aggressive policies in resuscitating neonates at the limit of viability may partly explain some of the reported variability in prevalence. ^,

BSIs are the most common HAI in the NICU, but other infections of sterile sites (e.g., urinary tract infection [UTI], meningitis, peritonitis) occur and can arise from various sources. ^, The most common organisms associated with HAI in the NICU often vary by site of infection ( Table 94.1 ). The National Healthcare Safety Network (NHSN) and the International Nosocomial Infection Control Consortium use discrete surveillance definitions to categorize HAIs into site-specific diseases, including primary BSI, UTI, ventilator-associated events (VAEs), and surgical site infections (SSIs). ^, These HAIs are stratified by five birth weight (BW) categories (≤750 g, 751–1000 g, 1001–1500 g, 1501–2500 g, and ≥2501 g). Individual HAI surveillance definitions are complex, frequently altered, and non-NICU-specific, creating discordance with clinical definitions. Close coordination between the infection control and prevention team and the clinical team is necessary for effective and accurate prospective surveillance within the NICU.

Central Line-Associated Bloodstream Infections

BSIs occur in approximately 34% of ELBW infants and are most commonly attributed to gram-positive organisms, such as coagulase-negative staphylococci (CoNS) and Staphylococcus aureus. ^, Many BSI are device-related, with the majority attributed to a central venous catheter (CVC) (e.g., umbilical venous catheter [UVC], peripherally inserted central catheter [PICC], or tunneled CVC). ^, The NHSN definition for a central-line-associated bloodstream (CLABSI) is a laboratory confirmed BSI occurring in a patient with a CVC that has been in place for at least 2 consecutive calendar days and that is not attributable to an infection at another site. Criteria for a laboratory confirmed BSI in an infant ≤1 year of age includes isolation of a traditional neonatal pathogen (e.g. Escherichia coli) from ≥1 blood culture or common commensal species (e.g. CoNS) from 2 or more blood cultures collected on separate occasions. In addition, the presence of ≥1 of the following clinical signs of infection is required: apnea, bradycardia, fever (>38°C), or hypothermia (<36°C). Data are stratified by BW categories and reported as infections per 1000 central line-days.

Individual NICUs in the US report CLABSI rates to the NHSN and, ideally, monitor and compare their rates with annual benchmarks. Values at the high end of the NHSN data should drive infection prevention efforts. In fact, CLABSIs have been a major focus of surveillance and prevention efforts in NICUs over the last decade. Data on CLABSI rates from NHSN participating level II/III NICUs showed a continued decrease in CLABSIs from 1.06 per 1000 central line-days in 2017 to 0.97 per 1000 line-days in 2018. Improved management practices that focus on the maintenance and timely removal of UVCs, CVCs, and PICCs, as well as standardized feeding practices aimed at minimizing catheter dwell times, have been shown to be effective strategies in reducing CLABSIs. ^,

Optimal management of a CLABSI constitutes removal of the intravascular catheter once the BSI is confirmed. Nevertheless, the vital importance of reliable intravascular access in a critically ill neonate must be acknowledged, especially as successful treatment of CLABSI in situ has become more common. There have been no randomized trials to guide management of CLABSIs in the NICU, but several large observational cohort studies have compared outcomes of LOS in neonates with CVCs treated with or without CVC removal, with data suggesting that management strategies depend largely on the pathogen and clinical condition of the infant. If treatment with the central line remaining in situ is attempted, antimicrobial agents should be administered through the catheter. The algorithm shown in Fig. 94.1 provides a framework for the management of CLABSIs in neonates.

Late-Onset Meningitis

Meningitis is a rare but serious manifestation of LOS, most commonly attributed to E. coli, Group B Streptococcus, S. aureus, CoNS, Enterococcus spp., and Candida albicans. ^, In a prospective study of 9641 VLBW infants who survived more than 3 days, late-onset meningitis occurred in 134 (1.4% of all infants; 5% of those who had a lumbar puncture [LP] performed). In a 2019 retrospective review of 94,603 infants (of all BWs) admitted to NICUs within the Canadian Neonatal Network from 2010–2016, 213 (0.2%) were diagnosed with late-onset meningitis. Compared with matched controls without meningitis, infants with meningitis had a significantly higher adjusted odds of seizures, bronchopulmonary dysplasia, and severe retinopathy of prematurity, as well as a longer duration of mechanical ventilation and hospitalization.

Because meningitis is relatively uncommon, the frequency of an LP as part of the standard evaluation for LOS varies widely between and within NICUs. However, reliance on blood culture alone to identify meningitis is insufficient, as more than one-third of infants with meningitis have sterile blood cultures. The presence of meningitis affects selection and duration of antibiotic therapy, informs long-term prognosis, and may help identify foci of infection, such as brain abscesses. Practice guidelines that aim to limit LPs for infants with suspected LOS have been associated with increased mortality. Therefore all infants with suspected LOS should have an LP as part of the initial diagnostic evaluation, unless the infant is too critically ill to tolerate the procedure. In this case, the LP should be performed when clinical stabilization is achieved. If cerebrospinal fluid (CSF) is successfully obtained, its interpretation can often be challenging as a result of several confounding variables including recent antibiotic exposure, traumatic lumbar punctures and sampling performed in the setting of a recent intraventricular hemorrhage, and variability in normative ranges for CSF parameters as a function of gestational and postnatal age. Every effort should therefore be made to obtain CSF cultures as soon as possible (i.e., prior to empiric antibiotic administration) when LOS and meningitis are suspected.

Urinary Tract Infection

UTIs are relatively common in the NICU, but unlike in the adult population, the vast majority are not urinary catheter–associated. ^, Retrospective studies have reported that UTIs among VLBW infants occurs at rates of 8%–11%. ^, Analysis of data from 1997–2010 from the Pediatrix Medical Group collected from 322 NICUs revealed that approximately 17% of infants were diagnosed with a UTI during their NICU stay. When stratified by BW category, the majority (47%) occurred in ELBW infants. Gram-negative organisms account for the majority of UTIs in the NICU, followed by gram-positives, and fungi. The most common pathogens include E. coli, Klebsiella spp., Enterobacter spp., Enterococcus spp., and Candida spp. ^,

There is considerable institutional variability in performing urine culture and analysis as part of the evaluation for LOS. This variability is due, in part, to difficulty in obtaining sterile urine specimens from extremely preterm neonates. Clean catch specimens are not possible, and there is high risk for contamination when specimens are obtained via bag collection. The only acceptable methods of urine collection for culture are suprapubic bladder aspiration and sterile urethral catheterization, both of which present challenges. Suprapubic aspiration may be more painful than urethral catheterization and is associated with rare but serious complications, such as bowel perforation. ^, Although a higher rate of success in obtaining urine via urethral catheterization compared with suprapubic bladder aspiration has been reported, technical challenges in ELBW infants increase the risk for false-positive results. ^, A retrospective review of positive urine cultures obtained in a NICU in Canada revealed that 53% of specimens obtained via bladder catheterization were deemed contaminants. Furthermore, the utilization of urinalysis to assist in the diagnosis of UTI in preterm neonates is challenging because lack of pyuria is common, making it difficult to distinguish between true infection and bacteriuria.

Examining paired blood and urine cultures for 189 VLBW infants undergoing an evaluation for LOS, among VLBW infants with a positive urine culture, 62% had a negative blood culture. In a case series of 60 UTIs (specimens primarily obtained via urethral catheterization) among NICU patients being evaluated for LOS, concurrent BSI with the same pathogen was detected in 52% of infections due to Candida and 8% of those due to bacteria. Data from the Pediatrix Medical Group revealed that 13% (127 of 976) of infants diagnosed with a UTI had a concordant positive blood culture. These data stress the need to include a urine culture of part of the evaluation for LOS as blood cultures are not a reliable method of screening for UTIs. However, continued efforts to improve collection strategies in an effort to minimize contamination and unnecessary antibiotic exposure are critical.