Machine Learning-Based Critical Congenital Heart Disease Screening Using Dual-Site Pulse Oximetry Measurements.

BACKGROUND: Oxygen saturation (Spo2) screening has not led to earlier detection of critical congenital heart disease (CCHD). Adding pulse oximetry features (ie, perfusion data and radiofemoral pulse delay) may improve CCHD detection, especially coarctation of the aorta (CoA). We developed and tested a machine learning (ML) pulse oximetry algorithm to enhance CCHD detection. METHODS AND RESULTS: Six sites prospectively enrolled newborns with and without CCHD and recorded simultaneous pre- and postductal pulse oximetry. We focused on models at 1 versus 2 time points and with/without pulse delay for our ML algorithms. The sensitivity, specificity, and area under the receiver operating characteristic curve were compared between the Spo2-alone and ML algorithms. A total of 523 newborns were enrolled (no CHD, 317; CHD, 74; CCHD, 132, of whom 21 had isolated CoA). When applying the Spo2-alone algorithm to all patients, 26.2% of CCHD would be missed. We narrowed the sample to patients with both 2 time point measurements and pulse-delay data (no CHD, 65; CCHD, 14) to compare ML performance. Among these patients, sensitivity for CCHD detection increased with both the addition of pulse delay and a second time point. All ML models had 100% specificity. With a 2-time-points+pulse-delay model, CCHD sensitivity increased to 92.86% (P=0.25) compared with Spo2 alone (71.43%), and CoA increased to 66.67% (P=0.5) from 0. The area under the receiver operating characteristic curve for CCHD and CoA detection significantly improved (0.96 versus 0.83 for CCHD, 0.83 versus 0.48 for CoA; both P=0.03) using the 2-time-points+pulse-delay model compared with Spo2 alone. CONCLUSIONS: ML pulse oximetry that combines oxygenation, perfusion data, and pulse delay at 2 time points may improve detection of CCHD and CoA within 48 hours after birth. REGISTRATION: URL: https://www.clinicaltrials.gov/study/NCT04056104?term=NCT04056104&rank=1; Unique identifier: NCT04056104.

Novel scoring tool of hypoxemic respiratory failure and pulmonary hypertension for defining severity of persistent pulmonary hypertension of newborn.

OBJECTIVE: To obtain preliminary validity data for a hypoxemic respiratory failure/pulmonary hypertension (HRF/PH) score for classifying persistent pulmonary hypertension of the newborn (PPHN). STUDY DESIGN: Retrospective chart review of 100 consecutive neonates admitted to a Children's hospital from 2016-2021 with PPHN, gestational age ≥34 weeks, and echocardiograms in the first week. We assessed the correlation between HRF/PH score and short-term outcomes using linear and logistic regressions. RESULTS: HRF/PH scores ranged 2-12 (mean 8.5), and were classified mild (0-5), moderate (6-10), and severe (11-15), with 6%, 77% and 17% infants in respective categories. HRF/PH score category correlated with invasive ventilation, nitric oxide, high frequency ventilation, vasoactive infusions, extracorporeal life support and death. HRF/PH score category did not correlate with duration of support or length of stay. CONCLUSION: The HRF/PH score offers a promising representation of disease severity for PPHN. The tool requires further validation in prospective studies and evaluation for long-term outcomes.

Protocol-Driven Initiation and Weaning of High-Flow Nasal Cannula for Patients With Bronchiolitis: A Quality Improvement Initiative.

OBJECTIVES: Bronchiolitis is the most common cause for nonelective infant hospitalization in the United States with increasing utilization of high-flow nasal cannula (HFNC). We standardized initiation and weaning of HFNC for bronchiolitis and quantified the impact on outcomes. Our specific aim was to reduce hospital and ICU length of stay (LOS) by 10% between two bronchiolitis seasons after implementation. DESIGN: A quality improvement (QI) project using statistical process control methodology. SETTING: Tertiary-care children's hospital with 24 PICU and 48 acute care pediatric beds. PATIENTS: Children less than 24 months old with bronchiolitis without other respiratory diagnoses or underlying cardiac, respiratory, or neuromuscular disorders between December 2017 and November 2018 (baseline), and December 2018 and February 2020 (postintervention). INTERVENTIONS: Interventions included development of an HFNC protocol with initiation and weaning guidelines, modification of protocol and respiratory assessment classification, education, and QI rounds with a focus on efficient HFNC weaning, transfer, and/or discharge. MEASUREMENTS AND MAIN RESULTS: A total of 223 children were included (96 baseline and 127 postintervention). The primary outcome metric, average LOS per patient, decreased from 4.0 to 2.8 days, and the average ICU LOS per patient decreased from 2.8 to 1.9 days. The secondary outcome metric, average HFNC treatment hours per patient, decreased from 44.0 to 36.3 hours. The primary and secondary outcomes met criteria for special cause variation. Balancing measures included ICU readmission rates, 30-day readmission rates, and adverse events, which were not different between the two periods. CONCLUSIONS: A standardized protocol for HFNC management for patients with bronchiolitis was associated with decreased hospital and ICU LOS, less time on HFNC, and no difference in readmissions or adverse events.

Lower pass threshold (≥93%) for critical congenital heart disease screening at high altitude prevents repeat screening and reduces false positives.

OBJECTIVE: We evaluated first screen pass rate for two pass thresholds for critical congenital heart disease (CCHD) oxygen saturation (SpO2) screening at higher altitude. STUDY DESIGN: A retrospective cohort of 948 newborns underwent CCHD screening near sea-level (n = 463) vs 6250 ft altitude (n = 485) over 3 years. Standard SpO2 pass threshold ≥95% and lower SpO2 pass threshold ≥93% (high-altitude screen) were applied to first measurements to compare pass frequencies. RESULTS: The median SpO2 was lower in high-altitude newborns (96% vs 99%-p < 0.001). The high-altitude newborns passed the AAP algorithm first screen less often (89.3% vs 99.6%-p < 0.001). With the high-altitude algorithm, 98% of high-altitude newborns passed the first screen. CONCLUSION: Lowering the SpO2 pass threshold by 2% at >6000 ft, significantly increased first screen pass rate. Adjustments for altitude may reduce nursing time to conduct repeat measurements and prevent transfers for echocardiograms. Larger studies are necessary to assess impact on false negatives.

Factors to Consider to Study Preductal Oxygen Saturation Targets in Neonatal Pulmonary Hypertension.

There are potential benefits and risks to the infant with higher and lower oxygen saturation (SpO2) targets, and the ideal range for infants with pulmonary hypertension (PH) remains unknown. Targeting high SpO2 can promote pulmonary vasodilation but cause oxygen toxicity. Targeting lower SpO2 may increase pulmonary vascular resistance, especially in the presence of acidosis and hypothermia. We will conduct a randomized pilot trial to compare two ranges of target preductal SpO2 in late-preterm and term infants with hypoxic respiratory failure (HRF) and acute pulmonary hypertension (aPH) of the newborn. We will assess the reliability of a newly created HRF/PH score that could be used in larger trials. We will assess trial feasibility and obtain preliminary estimates of outcomes. Our primary hypothesis is that in neonates with PH and HRF, targeting preductal SpO2 of 95-99% (intervention) will result in lower pulmonary vascular resistance and pulmonary arterial pressures, and lower the need for pulmonary vasodilators (inhaled nitric oxide-iNO, milrinone and sildenafil) compared to targeting SpO2 at 91-95% (standard). We also speculate that a higher SpO2 target can potentially induce oxidative stress and decrease response to iNO (oxygenation and pulmonary vasodilation) for those patients that still require iNO in this range. We present considerations in planning this trial as well as some of the details of the protocol design (Clinicaltrials.gov (NCT04938167)).

Enhanced Critical Congenital Cardiac Disease Screening by Combining Interpretable Machine Learning Algorithms.

Critical Congenital Heart Disease (CCHD) screening that only uses oxygen saturation (SpO2), measured by pulse oximetry, fails to detect an estimated 900 US newborns annually. The addition of other pulse oximetry features such as perfusion index (PIx), heart rate, pulse delay and photoplethysmography characteristics may improve detection of CCHD, especially those with systemic blood flow obstruction such as Coarctation of the Aorta (CoA). To comprehensively study the most relevant features associated with CCHD, we investigated interpretable machine learning (ML) algorithms by using Recursive Feature Elimination (RFE) to identify an optimal subset of features. We then incorporated the trained ML models into the current SpO2-alone screening algorithm. Our proposed enhanced CCHD screening system, which adds the ML model, improved sensitivity by approximately 10 percentage points compared to the current standard SpO2-alone method with minimal to no impact on specificity.Clinical relevance- This establishes proof of concept for a ML algorithm that combines pulse oximetry features to improve detection of CCHD with little impact on false positive rate.

Critical Congenital Heart Disease Detection in the Screening Era: Do Not Neglect the Examination!

Pulse oximetry oxygen saturation (SpO 2 )-based critical congenital heart disease (CCHD) screening is effective in detection of cyanotic heart lesions. We report a full-term male infant with normal perfusion who had passed the CCHD screening at approximately 24 hours after birth with preductal SpO 2 of 99% and postductal SpO 2 of 97%. Detection of a loud systolic cardiac murmur before discharge led to the diagnosis of pulmonary atresia (PA) with ventricular septal defect (PA-VSD) by echocardiogram. The infant was transferred to a tertiary care center after initiation of prostaglandin E1 (PGE1) therapy. Throughout the initial course, he was breathing comfortably without respiratory distress or desaturations on pulse oximetry. We believe that this is the first documented report of PA missed by CCHD screening. Thorough and serial clinical examinations of the newborn infant proved vital in the timely diagnosis of this critical disease. We review the hemodynamics and the recent literature evaluating utility of CCHD screening in the diagnosis of PA-VSD. Pulse oximetry-based CCHD screening should be considered a tool to enhance CCHD detection with an emphasis on detailed serial physical examinations in newborn infants.

Management of systemic hypotension in term infants with persistent pulmonary hypertension of the newborn: an illustrated review.

In persistent pulmonary hypertension of the newborn (PPHN), the ratio of pulmonary vascular resistance to systemic vascular resistance is increased. Extrapulmonary shunts (patent ductus arteriosus and patent foramen value) allow for right-to-left shunting and hypoxaemia. Systemic hypotension can occur in newborns with PPHN due to variety of reasons, such as enhanced peripheral vasodilation, impaired left ventricular function and decreased preload. Systemic hypotension can lead to end organ injury from poor perfusion and hypoxaemia in the newborn with PPHN. Thus, it must be managed swiftly. However, not all newborns with PPHN and systemic hypotension can be managed the same way. Individualised approach based on physiology and echocardiographic findings are necessary to improve perfusion to essential organs. Here we present a review of the physiology and mechanisms of systemic hypotension in PPHN, which can then guide treatment.

A Machine Learning Driven Pipeline for Automated Photoplethysmogram Signal Artifact Detection.

Recent advances in Critical Congenital Heart Disease (CCHD) research using Photoplethysmography (PPG) signals have yielded an Internet of Things (IoT) based enhanced screening method that performs CCHD detection comparable to SpO2 screening. The use of PPG signals, however, poses a challenge due to its measurements being prone to artifacts. To comprehensively study the most effective way to remove the artifact segments from PPG waveforms, we performed feature engineering and investigated both Machine Learning (ML) and rule based algorithms to identify the optimal method of artifact detection. Our proposed artifact detection system utilizes a 3-stage ML model that incorporates both Gradient Boosting (GB) and Random Forest (RF). The proposed system achieved 84.01% of Intersection over Union (IoU), which is competitive to state-of-the-art artifact detection methods tested on higher resolution PPG.

Inadequacies of hospital-level critical congenital heart disease screening data reports: implications for research and quality efforts.

OBJECTIVE: Assess the quality of critical congenital heart disease (CCHD) screening data reports in California, where CCHD screening is not mandatory but reporting is. STUDY DESIGN: Retrospective review of California hospital-level CCHD screening data to evaluate data reliability and adherence to state screening and reporting recommendations. Data were evaluated for internal consistency and compared to two databases. RESULTS: Over one-third of hospitals did not submit data. Only 70.7% of the Vital Records live births were reported in CCHD screening data. Only 46% of reporting hospitals submitted data with matching numbers of completed screens and results, and 22% matched their respective live births in a second database. CONCLUSION: CCHD data reporting in California is incomplete, which may miss 359 CCHD cases/year from non-reporting. Data inconsistencies may miss additional cases. Mandatory screening, reporting, and improvements in data reliability are urgently needed to inform screening modifications and enhance timely detection and disease surveillance.

Pediatric Critical Care Medicine Fellowship Simulation Use Survey.

OBJECTIVES: Simulation-based education is used in the U.S. Pediatric Critical Care Medicine fellowship programs, yet the prevalence and types of simulation used is unknown. A survey was developed to determine the prevalence, the perceived importance, and barriers associated with simulation-based education in these programs. DESIGN: A 43-item survey instrument was sent to all 66 U.S. Accreditation Council for Graduate Medical Education-accredited Pediatric Critical Care Medicine fellowship programs during the summer of 2018. We defined simulation broadly as "any type of simulation that involved mannequins, task trainers, standardized actors, team training, etc." SETTING: An online survey was used to obtain information regarding simulation used in Pediatric Critical Care Medicine fellowship programs. SUBJECTS: All sixty-six U.S. Accreditation Council for Graduate Medical Education-accredited Pediatric Critical Care Medicine fellowship programs were sent a survey request. MEASUREMENTS AND MAIN RESULTS: Forty-four of the 66 U.S. Accreditation Council for Graduate Medical Education-accredited Pediatric Critical Care Medicine fellowship programs (67%) responded to the survey. Ninety-eight percent of responding programs (n = 43) use simulation-based education in their Pediatric Critical Care Medicine fellowship curriculum. Most programs (56%) have incorporated simulation training into their Pediatric Critical Care Medicine fellowship curriculum in the last 4-10 years (range, <1 to >15 yr, median 4-6 yr). A variety of principles, concepts, and programs were reported as used in their simulation programs. The most commonly reported barriers to Pediatric Critical Care Medicine fellowship simulation-based education were lack of funding (56%) and lack of faculty with simulation experience (56%). The majority of programs (64%; N = 28) think simulation-based education is absolutely necessary to Pediatric Critical Care Medicine fellowship training. CONCLUSIONS: Nearly, all responding U.S. Accreditation Council for Graduate Medical Education-accredited Pediatric Critical Care Medicine fellowship programs use simulation-based education to train Pediatric Critical Care Medicine fellows with the majority perceiving simulation as absolutely necessary to Pediatric Critical Care Medicine fellow training. The reported types of simulation used in fellow training varied, as did training theories and concepts in the simulation programs. More research is needed to understand how to optimize and perhaps standardize parts of Pediatric Critical Care Medicine fellowship simulation training to improve the impact and outcomes of such training.

Spontaneous aortic thrombosis in neonates: a case report and review of literature.

Neonatal aortic thrombosis is a rare occurrence but can be life-threatening. Most aortic thrombosis in neonates is related to umbilical artery catheters. A case of a neonate with a spontaneous aortic thrombosis is described here along with a comprehensive review of the literature for cases of neonatal aortic thrombosis not related to any intravascular device or procedure. The aetiologies of these spontaneous thromboses and the relevance of hypercoagulable disorders are discussed. The cases were analysed for odds of death by treatment method adjusted for era. The reference treatment method was thrombolysis and anticoagulation. No other treatment modality had significantly lower odds than the reference. Surgery alone had higher odds for death than the reference, but this may be confounded by severity of case. The management recommendations for clinicians encountering neonates with spontaneous neonatal aortic thrombosis are discussed.

Oxygen Saturation and Perfusion Index-Based Enhanced Critical Congenital Heart Disease Screening.

OBJECTIVE: To determine if addition of perfusion index (PIx) to oxygen saturation (SpO2) screening improves detection of critical congenital heart disease (CCHD) with systemic outflow obstruction. STUDY DESIGN: We determined screening thresholds for PIx and applied these to a cohort of newborns with and without congenital heart disease (CHD). RESULTS: A total of 123 normal and 21 CHD newborns (including five with critical systemic outflow obstruction) were enrolled. Four of these five critical systemic obstruction subjects passed SpO2-based screen. Four out of these five subjects failed PIx-based screen. The sensitivity for detection of systemic obstruction CCHD when compared with healthy infants increased from 20% (95% confidence interval [CI]: 1-72%) with SpO2 screening alone to 80% (95% CI: 28-100%) with combined SpO2-PIx screen. However, 2.44% of normal infants failed PIx screen. CONCLUSION: Addition of PIx to SpO2 screening may detect additional cases of CCHD and further research is necessary to come up with optimal screening thresholds.

A novel system to collect dual pulse oximetry data for critical congenital heart disease screening research.

INTRODUCTION: Access to patient medical data is critical to building a real-time data analytic pipeline for improving care providers' ability to detect, diagnose, and prognosticate diseases. Critical congenital heart disease (CCHD) is a common group of neonatal life-threatening defects that must be promptly diagnosed to minimize morbidity and mortality. CCHD can be diagnosed both prenatally and postnatally. However, despite current screening practices involving oxygen saturation analysis, timely diagnosis is missed in approximately 900 infants with CCHD annually in the USA and can benefit from increased data processing capabilities. Adding non-invasive perfusion measurements to oxygen saturation data can improve the timeliness and fidelity of CCHD diagnostics. However, real-time monitoring and interpretation of non-invasive perfusion data are currently limited. METHODS: To address this challenge, we created a hardware and software architecture utilizing a Pi-top™ for collecting, visualizing, and storing dual oxygen saturation, perfusion indices, and photoplethysmography data. Data aggregation in our system is automated and all data files are coded with unique study identifiers to facilitate research purposes. RESULTS: Using this system, we have collected data from 190 neonates, 130 presumably without and 60 with congenital heart disease, in total comprising 1665 min of information. From these data, we are able to extract non-invasive perfusion features such as perfusion index, radiofemoral delay, and slope of systolic rise or diastolic fall. CONCLUSION: This data collection and waveform analysis is relatively inexpensive and can be used to enhance future CCHD screening algorithms.

Telemedicine for Interfacility Nurse Handoffs.

OBJECTIVE: To compare nurse preparedness and quality of patient handoff during interfacility transfers from a pretransfer emergency department to a PICU when conducted over telemedicine versus telephone. DESIGN: Cross-sectional nurse survey linked with patient electronic medical record data using multivariable, multilevel analysis. SETTING: Tertiary PICU within an academic children's hospital. PARTICIPANTS: PICU nurses who received a patient handoff between October 2017 and July 2018. INTERVENTIONS: None. MAIN RESULTS AND MEASUREMENTS: Among 239 eligible transfers, 106 surveys were completed by 55 nurses (44% survey response rate). Telemedicine was used for 30 handoffs (28%), and telephone was used for 76 handoffs (72%). Patients were comparable with respect to age, sex, race, primary spoken language, and insurance, but handoffs conducted over telemedicine involved patients with higher illness severity as measured by the Pediatric Risk of Mortality III score (4.4 vs 1.9; p = 0.05). After adjusting for Pediatric Risk of Mortality III score, survey recall time, and residual clustering by nurse, receiving nurses reported higher preparedness (measured on a five-point adjectival scale) following telemedicine handoffs compared with telephone handoffs (3.4 vs 3.1; p = 0.02). There were no statistically significant differences in both bivariable and multivariable analyses of handoff quality as measured by the Handoff Clinical Evaluation Exercise. Handoffs using telemedicine were associated with increased number of Illness severity, Patient summary, Action list, Situation awareness and contingency planning, Synthesis by receiver components (3.3 vs 2.8; p = 0.04), but this difference was not significant in the adjusted analysis (3.1 vs 2.9; p = 0.55). CONCLUSIONS: Telemedicine is feasible for nurse-to-nurse handoffs of critically ill patients between pretransfer and receiving facilities and may be associated with increased perceived and objective nurse preparedness upon patient arrival. Additional research is needed to demonstrate that telemedicine during nurse handoffs improves communication, decreases preventable adverse events, and impacts family and provider satisfaction.

Use of magnetic resonance imaging in severe pediatric traumatic brain injury: assessment of current practice.

OBJECTIVE: There is no consensus on the optimal timing and specific brain MRI sequences in the evaluation and management of severe pediatric traumatic brain injury (TBI), and information on current practices is lacking. The authors performed a survey of MRI practices among sites participating in a multicenter study of severe pediatric TBI to provide information for designing future clinical trials using MRI to assess brain injury after severe pediatric TBI. METHODS: Information on current imaging practices and resources was collected from 27 institutions participating in the Approaches and Decisions after Pediatric TBI Trial. Multiple-choice questions addressed the percentage of patients with TBI who have MRI studies, timing of MRI, MRI sequences used to investigate TBI, as well as the magnetic field strength of MR scanners used at the participating institutions and use of standardized MRI protocols for imaging after severe pediatric TBI. RESULTS: Overall, the reported use of MRI in pediatric patients with severe TBI at participating sites was high, with 40% of sites indicating that they obtain MRI studies in > 95% of this patient population. Differences were observed in the frequency of MRI use between US and international sites, with the US sites obtaining MRI in a higher proportion of their pediatric patients with severe TBI (94% of US vs 44% of international sites reported MRI in at least 70% of patients with severe TBI). The reported timing and composition of MRI studies was highly variable across sites. Sixty percent of sites reported typically obtaining an MRI study within the first 7 days postinjury, with the remainder of responses distributed throughout the first 30-day postinjury period. Responses indicated that MRI sequences sensitive for diffuse axonal injury and ischemia are frequently obtained in patients with TBI, whereas perfusion imaging and spectroscopy techniques are less common. CONCLUSIONS: Results from this survey suggest that despite the lack of consensus or guidelines, MRI is commonly obtained during the acute clinical setting after severe pediatric TBI. The variation in MRI practices highlights the need for additional studies to determine the utility, optimal timing, and composition of clinical MRI studies after TBI. The information in this survey describes current clinical MRI practices in children with severe TBI and identifies important challenges and objectives that should be considered when designing future studies.

Seizure Treatment in Children Transported to Tertiary Care: Recommendation Adherence and Outcomes.

BACKGROUND AND OBJECTIVES: Convulsive seizures account for 15% of pediatric air transports. We evaluated seizure treatment received in community hospital emergency departments among transported patients for adherence to recommended management. METHODS: This study was a retrospective cohort study of children transported for an acute seizure to a tertiary pediatric hospital from 2010 to 2013. Seizure treatment was evaluated for adherence to recommended management. The primary outcome was intubation. RESULTS: Among 126 events, 61% did not receive recommended acute treatment. The most common deviation from recommended care was administration of >2 benzodiazepine doses. Lack of adherence to recommended care was associated with a greater than twofold increased risk of intubation (relative risk 2.4; 95% confidence interval, 1.4-4.13) and 1.5-fold increased risk of admission to the ICU (relative risk 1.65; 95% confidence interval, 1.24-2.16). Duration of ventilation was commonly <24 hours (87%) for patients who did or did not receive recommended acute seizure care. Among events treated initially with a benzodiazepine, only 32% received a recommended weight-based dosage, and underdosing was most common. CONCLUSIONS: Adherence to evidence-based recommended acute seizure treatment during initial care of pediatric patients using medical air transportation was poor. Intubation was more common when patients did not receive recommended acute seizure care. Educational efforts with a sustained quality focus should be directed to increase adherence to appropriate pediatric seizure treatment of children in community emergency departments.

Impact of educational video on critical congenital heart disease screening.

OBJECTIVE: To assess the status of pulse oximetry screening and barriers to implementing screening programs. METHODS: This was a prospective pre-post intervention survey of nurse managers and medical directors of hospital-based birthing centers in Oregon, Idaho, and Southern Washington. The intervention was a 7-minute video demonstrating and discussing pulse oximetry screening for critical congenital heart disease. RESULTS: Analysis of matched pairs showed a significant increase in the use of pulse oximetry screening during the study period from 52% to 73% (P < .0001). Following implementation of the video, the perception of all queried potential barriers decreased significantly among individuals from hospitals self-identified as nonscreening at baseline. Viewing the educational video was associated with an increase in the percentage of individuals from nonscreening hospitals that rated screening as "very beneficial" (45% vs 90%, P = .0001). CONCLUSIONS: An educational video was associated with improved opinions of pulse oximetry screening among hospitals not currently screening.

Risk factors for poor outcomes and therapeutic strategies for seasonal and pandemic influenza.

Seasonal influenza is a leading cause of morbidity and mortality worldwide annually while pandemic influenza, a unique entity, poses distinct challenges. The pediatric population is the primary vector for epidemics and the main focus of this article. While primary prevention with universal influenza vaccination is the best protection against significant illness, the antigenic shift and drift unique to influenza viruses leave a large population at risk even with universal vaccination. Early in an epidemic various diagnostic tests are available and discussed here. However, once an epidemic is established, testing is no longer necessary for diagnosis. Groups with particular vulnerability to serious illness include those <6 mo of age, children with underlying neuromuscular disease, pulmonary disorders, or other comorbid conditions. Early treatment with neuraminidase inhibitors is recommended for those with influenza infection requiring hospitalization. Respiratory failure and need for mechanical ventilation are the leading indications for intensive care unit admission among children. Complications of influenza such as pneumonia, empyema, myocarditis and neurologic involvement increase risk for intensive care unit admission and will be discussed as will the use of extracorporeal membrane support. An overview of the epidemiology of influenza with an emphasis on risk factors for critical illness and poor patient outcomes in the pediatric population as well as treatment strategies for critically ill children will be presented. Additionally, we will address some of the unique challenges posed by pandemic influenza and mitigation strategies.