Recommended Guideline for Uniform Reporting of Neonatal Resuscitation: The Neonatal Utstein Style

We based our approach on previous Utstein-style guidelines, in particular the pediatric Utstein style published in 1995⁸  and the 2019 update for in-hospital cardiac arrest.⁹  We started with key domains identified in previous Utstein-style guidelines: setting, patient, preevent, resuscitation process, postresuscitation process, and outcomes.⁹  The preevent domain provoked much discussion, because the event could be construed to start at the time of birth or at the initiation of resuscitation interventions. Therefore, this domain was replaced with 2 separate domains: antepartum and birth/preresuscitation. The antepartum domain includes relevant factors during pregnancy up to the time of birth, and birth/preresuscitation includes factors from the time of birth until the initiation of resuscitation, defined here as initiation of positive pressure ventilation (PPV). Neonatal interventions typically characterized as “initial steps” are included in the birth/preresuscitation domain.

The group identified candidate data elements within each domain from Utstein-style guidelines for other populations,^8,9  previous NLS systematic reviews, and the published ILCOR NLS consensus outcome rating.¹⁰  These candidate elements were subsequently endorsed, modified, or rejected on the basis of open discussion within the Neonatal Utstein Working Group. Additional data elements were included on the basis of open group discussion. When consensus was not achieved for a given data element, a vote was held with majority rule.

All data elements were classified as core versus supplemental. Core data elements should be collected and reported for all neonatal resuscitation studies. These elements were felt to be both important for comparative analyses and likely to be routinely and reliably captured. Supplemental data elements may be collected and reported using standard definitions when possible. These data elements were those thought to be important for more detailed analyses but may not be routinely available or reliable in all settings.

The Neonatal Utstein Working Group met virtually approximately every 3 weeks over a 14-month period and employed an iterative consensus-based process similar to that used for development of other Utstein-style guidelines.^6,8  The draft guideline was presented to a joint meeting of the ILCOR NLS Task Force and an invited group of neonatal content experts on December 3, 2021. In addition, because the Neonatal Utstein Working Group members work in highly resourced settings, the draft guideline was circulated to providers and investigators working in lower resource settings to ensure its relevance in a global setting. The Neonatal Utstein Working Group revised the guideline to reflect solicited feedback. The ILCOR NLS Task Force approved the final reporting guideline, and the ILCOR board endorsed the guideline as an ILCOR statement.

The Neonatal Utstein template (Fig 1) displays core and supplemental data elements across all 7 domains. The Data Table (Table 1) provides detailed information for each data element, including the recommended format for collecting each data field for individual patients. In the following text, we provide the rationale for selecting specific data elements, expand on data definitions as applicable, and provide recommendations for reporting data elements in aggregate.

This reporting guideline focuses solely on resuscitation of newborns immediately after birth, regardless of gestational age or birth location. This is because the physiology, techniques, and outcomes of resuscitation for newborns immediately after birth differ from newborns in other settings (ie, NICU) who have already transitioned to the extrauterine environment. Because neonatal resuscitation almost always indicates an inability to establish effective ventilation,¹¹  the group considered it important to include events related to both respiratory failure and cardiac failure. Therefore, these reporting standards pertain to the following delivery room resuscitative events:

1. Respiratory failure: apnea or insufficient respirations from any cause leading to a clinical decision to provide PPV

2. Bradycardia: heart rate <100 beats per minute leading to a clinical decision to provide PPV

3. Severe bradycardia: heart rate <60 beats per minute not responding to PPV leading to a clinical decision to provide chest compressions

4. Cardiac arrest: cessation of cardiac mechanical activity, determined by absent heart sounds and a pulseless state, leading to a clinical decision to provide chest compressions

We did not specify a minimum duration of time an intervention should be performed to classify an event as eligible. Resuscitation events with even brief episodes of PPV or chest compressions should be included. Newborns with apnea, insufficient respirations, or bradycardia who respond to stimulation or other initial steps and do not receive PPV or chest compressions should not be included. If a newborn receives continuous positive airway pressure (CPAP) without PPV or chest compressions, that resuscitation would not be considered an event per this reporting guideline. However, for newborns who receive CPAP in addition to PPV and/or chest compressions, core data will include information regarding the use of CPAP during resuscitation.

Two core elements relate to setting. The first is country, which is important to contextualize available resources and policies around the limits of viability. The second core element is in-hospital or out- of-hospital birth, with hospital defined as any licensed health facility that provides care for newborns at birth. Births that occur within the walls of hospitals that provide maternity and newborn care are considered in-hospital, regardless of where in the hospital the birth occurs.

There are 2 supplemental elements in the setting domain. For out- of-hospital births, an indicator for whether out-of-hospital birth was planned is meant to distinguish between unexpected births and births planned in homes or other nonhealth care settings. The second variable pertains to in-hospital births and refers to the definition of the hospital as a high-risk pregnancy center. The characteristics and resources of a high-risk pregnancy center are likely to differ widely by location, so reporting of this variable should be relevant to the local context. Finally, the Neonatal Utstein Working Group recognizes that resuscitation training for providers is an important aspect of the resuscitation setting. Given that there is no standard classification for resuscitation training, this element is not included in the neonatal Utstein template. Instead, the nature and frequency of resuscitation training for providers in the setting should be described by authors in the Methods of neonatal resuscitation publications.

This domain includes demographic information about the newborn and the mother. Most important is the gestational age at birth, defined as completed weeks of gestation using the best obstetrical estimate available. To facilitate subgroup analyses across studies, consistent gestational age subgroups should be reported, following World Health Organization definitions¹² : extremely preterm (<28 weeks), very preterm (28 to <32 weeks), moderate preterm (32 to <34 weeks), late preterm (34 to <37 weeks), and term (≥37 weeks). Additional core patient elements include presence of multiple gestation (regardless of birth order), birth weight, biological sex, and presence of major congenital anomalies or syndromes that were known prenatally or at the time of birth. Finally, the core element “prenatal care” indicates whether any prenatal care was provided at any point during the pregnancy.

Maternal characteristics identified as supplemental elements in the patient domain are specified in the Data Table. These include maternal age at delivery, cumulative years of education, hypertensive disease, diabetes, and tobacco smoking. In addition, for newborns with a major anomaly or syndrome indicated in the core patient domain, the supplemental element should include a description or listing of the specific anomalies and/or syndrome. Finally, we included 1 supplemental element to indicate whether there were any prenatal plans to limit the intensity or duration of resuscitation to better contextualize the resuscitation process and outcome.

Antepartum factors thought to impact the resuscitation process or outcomes were included. We did not include a comprehensive set of data elements that might increase the risk of resuscitation because the Neonatal Utstein reporting guideline only pertains to patients who receive resuscitation. Thus, information on process and outcomes of newborns who do not receive resuscitation will not be available. Three core elements in this domain are any maternal exposure to antenatal corticosteroids (regardless of type of corticosteroid or timing of exposure), presence of meconium-stained amniotic fluid, and mode of delivery.

Supplemental elements considered to impact the newborn’s physiologic status at the time of birth were included. Three of these pertain to the delivery. Assisted delivery should be indicated if any forceps, vacuum, or any other assistive devices were used at any time in the delivery (regardless of final mode of delivery). Two elements pertain to newborns born via cesarean delivery. The first is whether any labor occurred before delivery, because this may trigger fetal lung fluid absorption and affect pulmonary function.¹³  The second is whether the cesarean delivery was considered an emergency. Because definitions for emergency cesarean delivery vary,¹⁴  local classifications should be defined and used. If there was fetal exposure to antenatal corticosteroids, a supplemental element indicates whether at least 1 course was complete (defined as at least 48 hours of fetal exposure before delivery). Antepartum hemorrhage should be indicated as “yes” if there was any report of antepartum hemorrhage, regardless of amount or acuity, or if a placental abruption was diagnosed after 20 weeks’ gestation.

This category describes the newborn’s status and interventions performed between birth (defined as the time the infant’s body is fully exteriorized from the mother)¹⁵  and initiation of PPV or chest compressions. For interventions in the birth/preresuscitation, resuscitation, and postresuscitation domains, elements should reflect documentation of the given intervention. If an intervention is not documented as having been performed, the element is expected to be coded as “no.” Therefore, “unknown” response options are not provided in the Data Table.

Because cord management is associated with neonatal outcomes, core data include the timing of umbilical cord clamping and performance of umbilical cord milking. Consistent with previous ILCOR NLS systematic reviews, cord clamping is categorized as immediate (<30 seconds) versus later or delayed (≥30 seconds).^16,17  The core cord milking element includes any milking or stripping of the cord, regardless of whether the cord is intact or cut at the time of milking. Additional core elements include the newborn’s condition (heart rate and respiratory efforts at 1 minute after birth) and Apgar scores at 1 and 5 minutes. Respiratory effort is reported as adequate, inadequate, or absent. A practical guide for scoring this element is to use the Apgar score classification (strong cry, weak cry/hypoventilation, absent).¹⁸ 

Supplemental elements include the date of birth (month and year) and time of birth (in completed hours). Recording month and year of birth is necessary to assess temporal trends in resuscitation practice. Time of birth may be important to contextualize resuscitation performance and outcomes.¹⁹  Therefore, although the actual time of birth is collected for each patient, we recommend this element should be reported as a dichotomous indicator for overnight birth, with the hours constituting overnight defined in the Methods section by each investigator. Additional supplemental elements include umbilical cord pH (ideally from an arterial sample), more granular information around cord management, and the individual initial steps (stimulation, suction, thermoregulation) performed by the neonatal resuscitation team.

This domain captures key interventions performed during neonatal resuscitation. Resuscitation is considered to start when PPV or chest compressions is initiated. The Neonatal Utstein Working Group was unable to develop 1 single approach to define the end of neonatal resuscitation. This has been variably defined across studies using location (eg, all interventions in the delivery room), patient physiologic status (eg, heart rate >100 beats per minute), or a prespecified duration (eg, first 20 minutes after birth). Conceptually, the resuscitation process should capture interventions intended to support immediate neonatal transition and not interventions for infants who require ongoing support at the completion of resuscitation (such as continuous PPV during transfer to NICU). As such, investigators should define in the Methods section how the end of resuscitation is defined for the particular setting.

Respiratory support is key to the process of neonatal resuscitation.¹¹  Core data therefore capture the modes of respiratory support provided (CPAP and PPV) and the interface through which support is delivered (mask, prongs, supraglottic airway, endotracheal tube). Most current guidelines suggest commencing resuscitation of term newborns with air with addition and titration of supplemental oxygen under specific circumstances.^20,21  The provision of supplemental oxygen is therefore a core element. Cardiac compressions and epinephrine (adrenaline) are rarely required and indicate severe compromise^22–24 ; occurrence of these should be captured as core elements.

The number, professional group, and experience of resuscitation providers likely have an important impact on the resuscitation process. However, staffing models vary considerably across settings. For the first iteration of this consensus statement, the highest number of providers present and caring for the newborn at a single time during the resuscitation process is captured as a supplemental element. Devices to generate inflating pressure vary and may impact the quality of ventilation; the respiratory device is a supplemental element. The highest-pressure settings for each mode of respiratory support used throughout resuscitation (CPAP, positive end expiratory pressure, and peak inflation pressure) are also included as supplemental elements. The type of respiratory support (if any) in place at the completion of resuscitation is a further supplemental element.

Information on the timing and duration of resuscitative interventions was of high interest, but these are not consistently captured for all resuscitative interventions. Thus, limited data regarding initiation of key interventions (PPV through any interface, chest compressions, and intubation) are recorded as supplemental elements. These should be reported in minutes relative to time of birth. Duration of each intervention is defined as the interval from when the intervention was first initiated to last performed. If the intervention is ongoing at the end of resuscitation (such as ongoing PPV during transfer to the NICU), the endpoint for duration of the intervention is defined as the end of resuscitation. We discussed collecting data on whether chest compressions were coordinated with ventilation and the ratio used, but we ultimately decided this level of detail was too burdensome for most settings. However, these details should be collected and reported when human studies of cardiac compressions are conducted.

Finally, information about rare but important interventions will be captured as supplemental elements. These include the administration route, timing, dose, and total number of doses of epinephrine; the use and cumulative dose of fluid boluses (crystalloid, colloid, and blood products) for volume expansion; and other important procedures such as pleurocentesis and defibrillation.

Core elements in the care of the newborn after resuscitation include therapeutic hypothermia, the immediate disposition for the newborn after delivery, and temperature at 1 hour after birth. Regarding disposition, NICU refers to any area providing higher-level care (including a special care nursery), whereas routine care includes staying with parents and/or admission to newborn nursery. Therapeutic hypothermia is defined as a protocol-driven, time-limited controlled lowering of head or body temperature to mitigate the neurologic effects of hypoxic ischemic encephalopathy. This includes either selective head cooling or whole-body cooling and excludes processes related to “passive cooling.” Of note, therapeutic hypothermia performed with low-cost devices such as ice packs or phase-changing material is considered as “active cooling,” provided these devices are used in a protocol-driven method.²⁵ 

Supplemental elements include more details around therapeutic hypothermia (if performed) and transfer to a facility with higher level of care. Two elements related to glucose monitoring (action of monitoring and lowest measured value) were also included. Although need for resuscitative interventions may increase the risk of impaired glycemic control, there are insufficient data to identify the optimal timing for glucose monitoring or the correct targets that should prompt intervention.²⁰  Thus, these elements were included as supplemental variables and may help inform future studies.

Further elements considered but not included at this time were timing of therapeutic hypothermia initiation and intubation/invasive mechanical ventilation after resuscitation. These were excluded after discussion to maintain a parsimonious and clinically relevant core data set. Notably, information about respiratory interventions during and at the completion of resuscitation are captured within the resuscitation process domain.

The Neonatal Utstein Working Group recognizes that relevant outcomes vary for different populations of newborns on the basis of gestational age or other risk factors. In addition, resources to diagnose all important neonatal outcomes are not available globally. Thus, a discrete number of core outcome elements should be reported for all newborns. Outcomes that are gestational-age specific or specific to discrete populations based on risk factors are labeled as such in the Data Table.

For all outcomes, investigators should report whether outcome assessors were masked to relevant exposures and treatment protocols. For outcomes that require radiographic imaging or other diagnostic testing, the investigators should provide information on what method(s) were used to identify the outcome of interest, who was responsible for interpreting the results, and when the test or screening exam was performed. For outcomes that rely on standardized assessments, investigators should describe which standardized assessments were used and how outcome assessors were trained. For outcomes that require active screening, imaging, or standardized assessment, the data element should be coded unknown if no diagnostic examination or exam was performed. Similarly, outcome elements should be coded unknown if the patient was transferred or died before assessment. Summary results of these outcomes should be reported as the number of affected newborns among the number of newborns who were assessed with the relevant diagnostic or screening examination.

Clarification or justification for specific elements is provided below:

• Death in initial resuscitation area: For newborns who are transferred during ongoing resuscitation, this outcome should be recorded as yes if the resuscitative interventions are initiated in the location of birth, continued uninterrupted during transport, and the newborn subsequently dies during ongoing resuscitation. This does not pertain to ongoing interventions in place once the initial resuscitation is complete (such as continuous PPV during transfer to NICU).

• Death before last follow-up: This categorical data element is intended to document if the newborn died during the interval between the last point of contact (hospital discharge or last outpatient visit) and the last planned follow-up. This item should be scored unknown if the patient is lost to follow-up and survival status is unknown. Age at last follow-up should be reported in the methods of the paper.

• Neurodevelopmental impairment (NDI) at 18 to 24 months’ age corrected for gestation: The Neonatal Utstein Working Group recognizes that comprehensive neurodevelopmental follow-up is difficult and expensive, and tools used to assess NDI vary on the basis of language and local resources. As a result, this critical outcome was considered supplemental. The criteria for defining NDI should be clearly stated. This item should be scored unknown if the patient is lost to follow-up and records are insufficient to assign a category. In addition, individual components of the criteria used to define NDI should be reported. Where feasible, appropriate descriptive statistics (mean and SD, or median and interquartile range) should be reported for each measure.

• Brain injury on neuroimaging: This outcome was considered important to fully understand the pathophysiology of long-term NDI. However, because neuroimaging is not routinely collected on most newborns who receive resuscitation, particularly for newborns who receive PPV alone, this supplemental outcome is indicated for newborns who receive chest compressions during resuscitation at birth.

• Meconium aspiration syndrome: This should be reported for newborns born through meconium-stained amniotic fluid. The criteria used to define meconium aspiration syndrome should be described; a suggested practical definition is that used in Vain et al.²⁶ 

We employed a consensus-based approach to develop the first Utstein-style reporting guideline applicable to neonatal resuscitation. Given the global burden of neonatal resuscitation and the growing volume of clinical studies focused on the neonatal population, the lack of standardized reporting guidelines poses an important methodologic limitation to the field. A consistent reporting framework is anticipated to provide significant value for individual investigators and networks engaged in neonatal resuscitation research. In addition, this will facilitate data pooling in meta-analyses, ultimately enhancing the strength of treatment recommendations and neonatal resuscitation guidelines.

The Neonatal Utstein Working Group confronted multiple challenges in this process, and some issues remain unresolved. Foremost was the tension between what was desirable versus feasible to include. Neonatal resuscitation is performed in both high- and low-resource settings. In addition, resuscitation teams and resources available for high-risk patients, such as extremely preterm newborns or newborns with major anomalies, often differ from lower-risk patients who unexpectedly require resuscitation. In resuscitations with fewer resources, personnel are not typically available for real-time documentation, and physiologic monitor data (if used) are not typically stored or downloaded. Finally, many variables of potential interest before and after resuscitation are variably defined across settings. Thus, core data elements were those felt to be clinically relevant and most reliably captured for all newborns.

One important difference between this guideline and reporting guidelines for other resuscitation settings relates to defining the event of interest. For example, cardiac arrest is a clearly defined event with a distinct onset and endpoint (return of spontaneous circulation). In contrast, there is a continuum between stabilization and resuscitation after birth. Many newborns require support to successfully transition to the postnatal environment because of immaturity or congenital anomalies, whereas others require resuscitation for perinatal depression or asphyxia. This reporting guideline was designed to focus on resuscitation events, not stabilization events. However, there is no standard definition or method to distinguish between these. Thus, we employed a pragmatic approach of defining eligible events on the basis of the resuscitative interventions received (PPV or chest compressions), rather than defining resuscitation events on the basis of patients’ physiologic status, or need, for resuscitation. We acknowledge that it is often difficult to determine appropriateness of an intervention during newborn resuscitation on the basis of review of medical records. It is possible that some patients receive interventions such as PPV or chest compressions when they are not indicated. Likewise, some patients who need such interventions may not receive them.

The Neonatal Utstein reporting guideline provides a framework to standardize reporting key data elements in publications of neonatal clinical resuscitation. The guideline applies to both clinical trials and observational studies. Although we defined a parsimonious list of core data elements across 7 domains, these are not exclusive. This guideline is not meant to replace investigators’ autonomy and discretion when designing clinical studies and trials. Investigators are expected to identify and collect additional data that are appropriately informed by the intervention and study question of interest. However, it is the hope of the Neonatal Utstein Working Group that, at minimum, all specified core data elements will be reported using the standardized definitions provided here.

This first version of the Neonatal Utstein reporting guideline reflects the views and expertise of the ILCOR NLS; we acknowledge this methodologic limitation. We anticipate feedback from the neonatal resuscitation community will inform subsequent updates to this guideline. In addition, ongoing research will likely identify important clinical resuscitation data elements that are not included here. Future iterations of the Neonatal Utstein reporting guideline should account for emerging evidence to remain a useful resource for the neonatal resuscitation scientific community.

We thank the following individuals for their constructive review of the Neonatal Utstein template: Dr Hege Ersdal, Stavanger University Hospital and University of Stavanger, Stavanger, Norway; Dr Beena Kamath-Rayne, American Academy of Pediatrics, Itasca, IL; Dr Susan Niermeyer, University of Colorado Hospital and Children’s Hospital Colorado, Aurora, CO; and Dr Ju Lee Oei, the Royal Hospital for Women, Randwick NSW, Australia. In addition to the authors, the following ILCOR NLS Task Force Members provided input in the development of this guideline: Dr Daniela T. Costa-Nobre, Federal University of São Paulo, São Paulo, Brazil; Dr Maria F. de Almeida, Federal University of São Paulo, São Paulo, Brazil; Dr Walid El Naggar, Dalhousie University, Halifax, Nova Scotia, Canada; Dr Jorge G. Fabres, Universidad Catolica de Chile, Santiago, Chile; Dr Joe Fawke, Leicester Royal Infirmary, Leicester, United Kingdom; Dr Shigeharu Hosono, Jichi Medical University Saitama Medical Center, Saitama, Japan; Dr Tetsuya Isayama, National Center for Child Health and Development, Tokyo, Japan; Dr Mandira D. Kawakami, Federal University of São Paulo, São Paulo, Brazil; Dr Han-Suk Kim, College of Medicine, Seoul National University, Seoul, Korea; Dr Henry C. Lee, Stanford University School of Medicine, Palo Alto, California; Dr R. John Madar, University Hospitals Plymouth National Health Service Trust, Plymouth, United Kingdom; Dr Christopher J.D. McKinlay, Kidz First Neonatal Care, Auckland, New Zealand; Dr Firdose L. Nakwa, University of Witwatersrand, Johannesburg, South Africa; Dr Jeffrey M. Perlman, Weill Cornell Medical College, Cornell University, New York; Dr Bin Huey Quek, KK Women’s and Children’s Hospital, Singapore; Dr Yacov Rabi, University of Calgary and Alberta Children’s Hospital, Alberta, Canada; Dr Charles C. Roehr, Oxford University Hospitals, National Health Service Foundation Trust, United Kingdom; Dr Takahiro Sugiura, Toyohashi Municipal Hospital, Toyohashi, Aichi, Japan; and Dr Daniele Trevisanuto, University of Padua, Padua, Italy.

CPAP

continuous positive airway pressure

ILCOR

international liaison committee on resuscitation

NDI

neurodevelopmental impairment

NLS

neonatal life support

PPV

positive pressure ventilation