Initial surgery for spontaneous intestinal perforation in extremely low birth weight infants is not associated with mortality or in-hospital morbidities.

OBJECTIVE: Determine short-term outcomes following peritoneal drain (PD), laparotomy (LAP) after PD (PD-LAP), and LAP in extremely low birth weight (ELBW) infants with spontaneous intestinal perforation (SIP). STUDY DESIGN: ELBW infants with SIP were identified using the Children's Hospitals Neonatal Database. Mortality and length of stay (LOS) were compared among groups. RESULTS: Of 729 SIP infants from 6/2010-12/2016, 383(53%) received PD, 61(8%) PD-LAP, and 285(39%) LAP. PD infants had lower GA at birth, at SIP diagnosis and upon admission than PD-LAP or LAP; and higher sepsis rates than LAP. Bivariate analysis and Kaplan-Meier survival estimates suggested PD had increased mortality vs. PD-LAP and LAP (27%, 11.5%, and 15.8% respectively, p < 0.001). However, surgical approach was not significantly associated with mortality in multivariable analysis accounting for GA and illness severity. LOS did not differ by surgical approach. CONCLUSIONS: In ELBW infants with SIP, mortality, and LOS are independent of the initial surgical approach.

A quality improvement initiative standardizing the antibiotic treatment and feeding practices in patients with medical necrotizing enterocolitis.

OBJECTIVE: Evaluate the impact of a multidisciplinary guideline standardizing antibiotic duration and enteral feeding practices following medical necrotizing enterocolitis (mNEC). STUDY DESIGN: For preterm infants with Bell Stage 2 A mNEC and negative blood culture, antibiotic treatment was standardized to 7 days. Trophic feeds of unfortified human milk began 72 h after resolution of pneumatosis. Feeds were advanced by 20 cc/kg/day starting on the last day of antibiotics. Primary outcomes were antibiotic days and days to full feeds, defined as 120 cc/kg/day of enteral nutrition. Secondary outcomes included central line days and length of stay (LOS). RESULTS: Antibiotic duration decreased 23%. Time to start trophic feeds and time to full feeds decreased 33 and 16% respectively. Central line use dropped (98 to 72% of infants) and central line days were reduced by 59%. CONCLUSION: Implementation of a mNEC QI package reduced antibiotic duration, time to full feeds, central line use and CL days.

A Multicenter Collaborative to Improve Postoperative Pain Management in the NICU.

OBJECTIVES: This quality improvement initiative aimed to decrease unrelieved postoperative pain and improve family satisfaction with pain management. METHODS: NICUs within the Children's Hospitals Neonatal Consortium that care for infants with complex surgical problems participated in this collaborative. Each of these centers formed multidisciplinary teams to develop aims, interventions, and measurement strategies to test in multiple Plan-Do-Study-Act cycles. Centers were encouraged to adopt evidence-based interventions from the Clinical Practice Recommendations, which included pain assessment tools, pain score documentation, nonpharmacologic treatment measures, pain management guidelines, communication of a pain treatment plan, routine discussion of pain scores during team rounds, and parental involvement in pain management. Teams submitted data on a minimum of 10 surgeries per month, spanning from January to July 2019 (baseline), August 2019 to June 2021 (improvement work period), and July 2021 to December 2021 (sustain period). RESULTS: The percentage of patients with unrelieved pain in the 24-hour postoperative period decreased by 35% from 19.5% to 12.6%. Family satisfaction with pain management measured on a 3-point Likert scale with positive responses ≥2 increased from 93% to 96%. Compliance with appropriate pain assessment and numeric documentation of postoperative pain scores according to local NICU policy increased from 53% to 66%. The balancing measure of the percentage of patients with any consecutive sedation scores showed a decrease from 20.8% at baseline to 13.3%. All improvements were maintained during the sustain period. CONCLUSIONS: Standardization of pain management and workflow in the postoperative period across disciplines can improve pain control in infants.

Eliminating Central Line Associated Bloodstream Infections in Pediatric Oncology Patients: A Quality Improvement Effort.

Central Line-Associated Bloodstream Infections (CLABSI) are the largest contributor to harm across the Children's Hospital's Solutions for Patient Safety network. Pediatric hematology/oncology (PHO) patients are at increased risk for CLABSI due to multiple factors. Consequently, traditional CLABSI prevention strategies are insufficient to eliminate CLABSI in this high-risk population. Methods: Our SMART aim was to reduce the CLABSI rate by 50% from a baseline of 1.89/1000 central line days to less than 0.9/1000 central line days by December 31, 2021. We created a multidisciplinary team being mindful to identify roles and responsibilities upfront. We developed a key driver diagram and designed and implemented interventions to influence our primary outcome. Results: We implemented interventions and conducted Plan-Do-Study-Act cycles concurrently. We found that performing audits by directly observing tasks rather than auditing documentation resulted in more accurate compliance assessments. As a result, our CLABSI rate improved from 1.89/1000 central line days in 2020 with 11 primary CLABSI to 0.73/1000 central line days in 2021 with four primary CLABSI. Average days between events improved from 30 days in 2020 to 73 days in 2021, and we achieved an unprecedented 542 days CLABSI-free, extending into 2022. Conclusions: Through a multimodal approach and utilizing characteristics of high-reliability organizations, we significantly reduced primary CLABSI, approaching zero in our PHO population and doubling the average days between events. Future efforts will focus on the sustained engagement of all stakeholders and improving our safety culture.

Variability in antibiotic duration for necrotizing enterocolitis and outcomes in a large multicenter cohort.

OBJECTIVES: To evaluate variability in antibiotic duration for necrotizing enterocolitis (NEC) and associated clinical outcomes. STUDY DESIGN: Five-hundred ninety-one infants with NEC (315 medical; 276 surgical) were included from 22 centers participating in Children's Hospitals Neonatal Consortium (CHNC). Multivariable analyses were used to determine predictors of variability in time to full feeds (TFF) and length of stay (LOS). RESULTS: Median (IQR) antibiotic duration was 12 (9, 17) days for medical and 17 (14, 21) days for surgical NEC. Wide variability in antibiotic use existed both within and among centers. Duration of antibiotic therapy was associated with longer TFF in both medical (OR 1.04, 95% CI [1.01, 1.05], p < 0.001) and surgical NEC (OR 1.02 [1, 1.03] p = 0.046); and with longer LOS in medical (OR 1.03 [1.02, 1.04], p < 0.001) and surgical NEC (OR 1.01 [1.01, 1.02], p = 0.002). CONCLUSION: Antibiotic duration for both medical and surgical NEC remains variable within and among high level NICUs.

Central Line Utilization and Complications in Infants with Congenital Diaphragmatic Hernia.

OBJECTIVE: Infants with congenital diaphragmatic hernia (CDH) require multiple invasive interventions carrying inherent risks, including central venous and arterial line placement. We hypothesized that specific clinical or catheter characteristics are associated with higher risk of nonelective removal (NER) due to complications and may be amenable to efforts to reduce patient harm. STUDY DESIGN: Infants with CDH were identified in the Children's Hospital's Neonatal Database (CHND) from 2010 to 2016. Central line use, duration, and complications resulting in NER are described and analyzed by extracorporeal membrane oxygenation (ECMO) use. RESULTS: A total of 1,106 CDH infants were included; nearly all (98%) had a central line placed, (average of three central lines) with a total dwell time of 22 days (interquartile range [IQR]: 14-39). Umbilical arterial and venous lines were most common, followed by extremity peripherally inserted central catheters (PICCs); 12% (361/3,027 central lines) were removed secondary to complications. Malposition was the most frequent indication for NER and was twice as likely in infants with intrathoracic liver position. One quarter of central lines in those receiving ECMO was placed while receiving this therapy. CONCLUSION: Central lines are an important component of intensive care for infants with CDH. Careful selection of line type and location and understanding of common complications may attenuate the need for early removal and reduce risk of infection, obstruction, and malposition in this high-risk group of patients. KEY POINTS: · Central line placement near universal in congenital diaphragmatic hernia infants.. · Mean of three lines placed per patient; total duration 22 days.. · Clinical patient characteristics affect risk..

Predicting treatment of pulmonary hypertension at discharge in infants with congenital diaphragmatic hernia.

OBJECTIVE: To predict pulmonary hypertension (PH) therapy at discharge in a large multicenter cohort of infants with congenital diaphragmatic hernia (CDH). STUDY DESIGN: Six-year linked records from Children's Hospitals Neonatal Database and Pediatric Health Information System were used; patients whose diaphragmatic hernia was repaired before admission or referral, who were previously home before admission or referral, and non-survivors were excluded. The primary outcome was the use of PH medications at discharge and the secondary outcome was an inter-center variation of therapies during inpatient utilization. Clinical factors were used to develop a multivariable equation randomly applied to 80% cohort; validated in the remaining 20% infants. RESULTS: A total of 831 infants with CDH from 23 centers were analyzed. Overall, 11.6% of survivors were discharged on PH medication. Center, duration of mechanical ventilation, and duration of inhaled nitric oxide were associated with the use of PH medication at discharge. This model performed well in the validation cohort area under the receiver operating characteristic curve of 0.9, goodness-of-fit χ2, p = 0.17. CONCLUSIONS: Clinical variables can predict the need for long-term PH medication after NICU hospitalization in surviving infants with CDH. This information may be useful to educate families and guide the development of clinical guidelines.

STEPP IN: A Multicenter Quality Improvement Collaborative Standardizing Postoperative Handoffs.

OBJECTIVE: To reduce care failures by 30% through implementation of standardized communication processes for postoperative handoff in NICU patients undergoing surgery over 12 months and sustained over 6 months. METHODS: Nineteen Children's Hospitals Neonatal Consortium centers collaborated in a quality improvement initiative to reduce postoperative care failures in a surgical neonatal setting by decreasing respiratory care failures and all other communication failures. Evidence-based clinical practice recommendations and a collaborative framework supported local teams' implementation of standardized postoperative handoff communication. Process measures included compliance with center-defined handoff staff presence, use of center-defined handoff tool, and the proportion of handoffs with interruptions. Participant handoff satisfaction was the balancing measure. Baseline data were collected for 8 months, followed by a 12-month action phase and 7-month sustain phase. RESULTS: On average, 181 postoperative handoffs per month were monitored across sites, and 320 respondents per month assessed the handoff process. Communication failures specific to respiratory care decreased by 73.2% (8.2% to 4.6% and with a second special cause signal to 2.2%). All other communication care failures decreased by 49.4% (17% to 8.6%). Eighty-four percent of participants reported high satisfaction. Compliance with use of the handoff tool and required staff attendance increased whereas interruptions decreased over the project time line. CONCLUSIONS: Team engagement within a quality improvement framework had a positive impact on the perioperative handoff process for high-risk surgical neonates. We improved care as demonstrated by a decrease in postoperative care failures while maintaining high provider satisfaction.

Effects of Guitar Accompaniment Patterns on Hospitalized Infants: A Randomized Controlled Trial.

A neonatal intensive care unit (NICU) can be an extremely stressful environment for infants receiving complex medical care at a pediatric facility. Music therapy can help address the stressful environment by increasing comfort and relaxation as well as decreasing a patient's physiological response of heart rate and respiratory rate. A randomized controlled trial was completed, examining the effects of 2 guitar accompaniment patterns on infants (an arpeggiated pattern and a bass/chord pattern) in a NICU. There were 180 infants enrolled with mean chronological ages of 7.4 ± 6.1 weeks and postmenstrual ages (gestational age at birth + their chronological age) of 39.8 ± 7.9 weeks. All subject enrollees participated in a 12-minute initial music therapy session. Outcome measures included heart rate, respirations, and comfort responses. The results of the research study demonstrated that the arpeggiated guitar pattern had a lower mean heart rate and respiratory across all 3 data points (pre, during, and post intervention); however, the difference in means between the 2 accompaniment pattern groups was not statistically significant. Using the Neonatal Infant Pain Scale (NIPS) to measure comfort response, the results demonstrated no difference between the 2 group accompaniment patterns. Although the results show no significant differences among accompaniment pattern groups, the researchers describe the clinical significance that supports the use of both accompaniment patterns as suitable interventions for infants in the NICU receiving music therapy intervention.

Prevalence and patterns of gestational parent's own milk feeds among infants with major congenital surgical anomalies in the NICU.

OBJECTIVE: To describe the prevalence and patterns of gestational parent's own milk (GPOM) feedings among infants undergoing major surgery during their neonatal intensive care unit (NICU) admission. STUDY DESIGN: We analyzed de-identified electronic medical records of all infants admitted to a regional NICU 2014-2015 who underwent surgery for a gastrointestinal, cardiac, or other major organ system defect(s). RESULTS: Of 79 infants, 85% received any GPOM during the NICU hospitalization. The median proportion of GPOM feeds was 66%. There was a trend toward decreassing proportions of GPOM with progressive months in NICU. The rate of any and exclusive GPOM feeds at NICU discharge was 49% and 29%, respectively. Infants who had a GI anomaly were more likely than infants with a cardiac anomaly to be discharged from NICU receiving GPOM. CONCLUSION: Barriers to the exclusive and continued provision of GPOM in this population require further study and intervention.

Gestational Age-Specific Complete Blood Count Signatures in Necrotizing Enterocolitis.

Objective: Necrotizing enterocolitis (NEC) is characterized by peripheral cell abnormalities, yet few studies have analyzed the complete blood count (CBC) specifically by gestational age (GA). Our objective was to describe GA-specific immune abnormalities in NEC through a comprehensive analysis of the CBC differential. Methods: Using a cohort of 246 infants (177 cases, 69 controls) admitted to neonatal intensive care units at a single institution, we retrospectively analyzed CBCs around illness onset in NEC cases compared with controls. Cases included surgical NEC (S-NEC, 34.5%) and medical NEC (M-NEC, 65.5%). Infants were divided into those born at GA <33 and ≥33 weeks. Differences in CBC values were described as absolute and percent changes at NEC onset from baseline and at antibiotic completion after NEC. We used machine learning algorithms based on the CBC at NEC to generate predictive models for diagnosis. Results: At NEC onset, there was an acute drop in monocytes and lymphocytes along with a rise in bands in S-NEC infants born <33 weeks compared with M-NEC. In comparison, both M-NEC and S-NEC ≥33 weeks had a percent drop in neutrophils at diagnosis compared with controls. At antibiotic completion, monocytes in S-NEC <33 weeks significantly rose compared with M-NEC, yet for S-NEC ≥33 weeks, bands significantly dropped compared with M-NEC. Predictive modeling was able to accurately predict S-NEC from M-NEC and controls. Conclusion: There are discrete leukocyte patterns in NEC based on GA. The CBC at diagnosis may be useful in identifying patients who will require surgery.

Treatment of pulmonary hypertension during initial hospitalization in a multicenter cohort of infants with congenital diaphragmatic hernia (CDH).

OBJECTIVE: Describe inpatient pulmonary hypertension (PH) treatment and factors associated with therapy at discharge in a multicenter cohort of infants with CDH. METHODS: Six years linked records from Children's Hospitals Neonatal Database and Pediatric Health Information System were used to describe associations between prenatal/perinatal factors, clinical outcomes, echocardiographic findings and PH medications (PHM), during hospitalization and at discharge. RESULTS: Of 1106 CDH infants from 23 centers, 62.8% of infants received PHM, and 11.6% of survivors were discharged on PHM. Survivors discharged on PHM more frequently had intrathoracic liver, small for gestational age, and low 5 min APGARs compared with those discharged without PHM (p < 0.0001). Nearly one-third of infants discharged without PHM had PH on last inpatient echo. CONCLUSIONS: PH medication use is common in CDH. Identification of infants at risk for persistent PH may impact ongoing management. Post-discharge follow-up of all CDH infants with echocardiographic evidence of PH is warranted.

STEPP IN: Working Together to Keep Infants Warm in the Perioperative Period.

OBJECTIVES: Reduce postoperative hypothermia by up to 50% over a 12-month period in children's hospital NICUs and identify specific clinical practices that impact success. METHODS: Literature review, expert opinion, and benchmarking were used to develop clinical practice recommendations for maintaining perioperative euthermia that included the following: established euthermia before transport to the operating room (OR), standardized practice for maintaining euthermia on transport to and from the OR, and standardized practice to prevent intraoperative heat loss. Process measures were focused on maintaining euthermia during these time points. The outcome measure was the proportion of patients with postoperative hypothermia (temperature ≤36°C within 30 minutes of a return to the NICU or at the completion of a procedure in the NICU). Balancing measures were the proportion of patients with postoperative temperature >38°C or the presence of thermal burns. Multivariable logistic regression was used to identify key practices that improved outcome. RESULTS: Postoperative hypothermia decreased by 48%, from a baseline of 20.3% (January 2011 to September 2013) to 10.5% by June 2015. Strategies associated with decreased hypothermia include >90% compliance with patient euthermia (36.1-37.9°C) at times of OR arrival (odds ratio: 0.58; 95% confidence interval [CI]: 0.43-0.79; P < .001) and OR departure (odds ratio: 0.0.73; 95% CI: 0.56-0.95; P = .017) and prewarming the OR ambient temperature to >74°F (odds ratio: 0.78; 95% CI: 0.62-0.999; P = .05). Hyperthermia increased from a baseline of 1.1% to 2.2% during the project. No thermal burns were reported. CONCLUSIONS: Reducing postoperative hypothermia is possible. Key practices include prewarming the OR and compliance with strategies to maintain euthermia at select time points throughout the perioperative period.

Center, Gestational Age, and Race Impact End-of-Life Care Practices at Regional Neonatal Intensive Care Units.

OBJECTIVE: To assess the impact of intercenter variation and patient factors on end-of-life care practices for infants who die in regional neonatal intensive care units (NICUs). STUDY DESIGN: We conducted a retrospective cohort analysis using the Children's Hospital Neonatal Database during 2010-2016. A total of 6299 nonsurviving infants cared for in 32 participating regional NICUs were included to examine intercenter variation and the effects of gestational age, race, and cause of death on 3 end-of-life care practices: do not attempt resuscitation orders (DNR), cardiopulmonary resuscitation within 6 hours of death (CPR), and withdrawal of life-sustaining therapies (WLST). Factors associated with these practices were used to develop a multivariable equation. RESULTS: Dying infants in the cohort underwent DNR (55%), CPR (21%), and WLST (73%). Gestational age, cause of death, and race were significantly and differently associated with each practice: younger gestational age (<28 weeks) was associated with CPR (OR 1.7, 95% CI 1.5-2.1) but not with DNR or WLST, and central nervous system injury was associated with DNR (1.6, 1.3-1.9) and WLST (4.8, 3.7-6.2). Black race was associated with decreased odds of WLST (0.7, 0.6-0.8). Between centers, practices varied widely at different gestational ages, race, and causes of death. CONCLUSIONS: From the available data on end-of-life care practices for regional NICU patients, variability appears to be either individualized or without consistency.

Extracorporeal membrane oxygenation and bloodstream infection in congenital diaphragmatic hernia.

OBJECTIVE: To characterize the risk of bloodstream (BSI) and urinary tract infection (UTI) and describe antibiotic use in infants with congenital diaphragmatic hernia (CDH) requiring extracorporeal membrane oxygenation (ECMO). STUDY DESIGN: The Children's Hospitals Neonatal Database was queried for infants with CDH and ECMO treatment from 2010 to 2016. The outcomes included BSI, UTI, and antimicrobial medication. Member institutions completed a survey on infection practices. RESULT: Eighteen of the 338 patients identified (5.3%) had ≥1 BSI during their ECMO course. The likelihood of BSI increased with time: 1.2/1000 ECMO days; 0.6% (2/315) in the first week and rising to 14.6/1000; 8.6% (5/58) after 21 days (p = 0.002). More than 95% of patients received antibiotics each week on ECMO. CONCLUSIONS: Confirmed BSI is rare in infants with CDH treated with ECMO in the first week, but increases with the duration of ECMO. Use of antibiotics was extensive and did not correspond to infection frequency.

Does the initial surgery for necrotizing enterocolitis matter? Comparative outcomes for laparotomy vs. peritoneal drain as initial surgery for necrotizing enterocolitis in infants <1000 g birth weight.

PURPOSE: Quantify short-term outcomes associated with initial surgery [laparotomy (LAP) vs. peritoneal drain (PD)] for necrotizing enterocolitis (NEC) in extremely-low-birth-weight (ELBW) infants. METHODS: Using the Children's Hospitals Neonatal Database, we identified ELBW infants <32 weeks' gestation with surgical NEC (sNEC). Unadjusted and multivariable regression analyses were used to estimate the associations between LAP (or PD) and death/short bowel syndrome (SBS) and length of stay (LOS). RESULTS: LAP was the more common initial procedure for sNEC (n = 359/528, 68%). Infants receiving LAP were older and heavier. Initial procedure was unrelated to death/SBS in both bivariate (LAP: 43% vs PD: 46%, p = 0.573) and multivariable analyses (OR = 0.89, 95% CI = 0.57, 1.38, p = 0.6). LAP was inversely related to mortality (29% vs. 41%, p < 0.007) in bivariate analysis, but not significant in multivariable analysis accounting for markers of preoperative illness severity. However, the association between LAP and SBS (14% vs. 5%, p = 0.012) remained significant in multivariable analyses (adjusted OR = 2.25, p = 0.039). LOS among survivors was unrelated to the first surgical procedure in multivariable analysis. CONCLUSION: ELBW infants who undergo LAP as the initial operative procedure for sNEC may be at higher risk for SBS without a clear in-hospital survival advantage or shorter hospitalization. LEVEL OF EVIDENCE: Level II.

Predicting Risk of Infection in Infants with Congenital Diaphragmatic Hernia.

OBJECTIVE: To predict incident bloodstream infection and urinary tract infection (UTI) in infants with congenital diaphragmatic hernia (CDH). STUDY DESIGN: We conducted a retrospective analysis using the Children's Hospital Neonatal Database during 2010-2016. Infants with CDH admitted at 22 participating regional neonatal intensive care units were included; patients repaired or discharged to home prior to admission/referral were excluded. The primary outcome was death or the occurrence of bloodstream infection or UTI prior to discharge. Factors associated with this outcome were used to develop a multivariable equation using 80% of the cohort. Validation was performed in the remaining 20% of infants. RESULTS: Median gestation and postnatal age at referral in this cohort (n = 1085) were 38 weeks and 3.1 hours, respectively. The primary outcome occurred in 395 patients (36%); and was associated with low birth weight, low Apgar, low admission pH, renal and associated anomalies, patch repair, and extracorporeal membrane oxygenation (P < .001 for all; area under receiver operating curve = 0.824; goodness of fit χ2 = 0.52). After omitting death from the outcome measure, admission pH, patch repair of CDH, and duration of central line placement were significantly associated with incident bloodstream infection or UTI. CONCLUSIONS: Infants with CDH are at high risk of infection which was predicted by clinical factors. Early identification and low threshold for sepsis evaluations in high-risk infants may attenuate acquisition and the consequences of these infections.

Winging it: maternal perspectives and experiences of breastfeeding newborns with complex congenital surgical anomalies.

OBJECTIVE: To describe the experience of breastfeeding (inclusive of breast milk expression/pumping, provision of breast milk via devices, and at-breastfeeding) among mothers of newborns with complex congenital surgical anomalies and the contexts under which pro-breastfeeding behaviors and attitudes are facilitated or compromised. STUDY DESIGN: We used qualitative description to analyze 23 interviews conducted with 15 mothers of newborns undergoing surgery for gastrointestinal, cardiac, or neural tube defects. RESULTS: Breastfeeding experiences were characterized by naivety regarding the importance and rationale for exclusive breast milk feedings and best practices to facilitate milk supply maintenance and transition to at-breast feeds. Maternal breastfeeding views and behaviors were impacted by indeterminate prenatal plans to breastfeed/provide breast milk, limited prior breastfeeding exposure and knowledge, and gaps in postnatal lactation support. CONCLUSION: Future research should investigate methods to improve exclusive breast milk feeding and facilitate transitions to at-breast feeds among mothers of newborns with surgical congenital anomalies, with consideration of identified barriers.

Acquired Infection and Antimicrobial Utilization During Initial NICU Hospitalization in Infants With Congenital Diaphragmatic Hernia.

BACKGROUND: In addition to substantial medical and surgical intervention, neonates with congenital diaphragmatic hernia often have concurrent concerns for acquired infection. However, few studies focus on infection and corresponding antimicrobial utilization in this population. METHODS: The Children's Hospital Neonatal Database was queried for congenital diaphragmatic hernia infants hospitalized from January 2010 to February 2016. Patient charts were linked to the Pediatric Health Information Systems database. Descriptive clinical data including delivery history, cultures sent, diagnosed infection, antimicrobial use and outcomes were reported. RESULTS: A total of 1085 unique patients were identified after data linkages; 275 (25.3%) were born at <37 weeks' gestation. Bacteremia at delivery (2/1085) and in the first 7 days of life (8/1085) was less common than later infection, but 976 patients (89.9%) were treated with antibiotics. Median number of days on antibiotics was 6 [3,11] for those without a documented infection and 21 [13,36] for those with positive cultures. Incidence of urinary tract infection, bacteremia and pneumonia increased significantly over time and was most common after 28 days. Antibiotic use, conversely, decreased over time (92% of infants in week 1 to 44% in week 4 and beyond). CONCLUSIONS: Although culture positivity increased with age, risk of these selected infections was relatively low for a population in neonatal intensive care unit. An important mismatch is observed between culture negativity and high rates of antibiotic utilization. These data identify opportunities for antibiotic stewardship quality improvement programs.