Predictive Ability of 10-Minute Apgar Scores for Mortality and Neurodevelopmental Disability

The study is a cohort study of participants from the Optimizing Cooling Trial with an 18- to 22-month outcome assessment.¹⁴  This trial enrolled participants from 18 NRN-affiliated tertiary US academic centers from October 2010 to November 2013. Follow-up was completed by January 2016. The inclusion criteria for the trial were infants with a gestational age of ≥36 weeks, age at admission of <6 hours after birth, and who fulfilled biochemical and clinical criteria for diagnosis of moderate or severe hypoxic-ischemic encephalopathy.¹⁴  Biochemical criteria included umbilical cord blood gas or any blood gas during the first hour after birth with a pH of ≤ 7.0 or a base deficit of ≥ 16 mmol/L. If pH was between 7.01 and 7.15, a base deficit was between 10 mmol/L and 15.9 mmol/L, or a blood gas was not available, additional criteria were required. These included an acute perinatal event (eg, late or variable decelerations, cord prolapse, cord rupture, uterine rupture, maternal trauma, hemorrhage, or acute cardiorespiratory arrest), and either a 10-minute Apgar score of ≤5 or assisted ventilation initiated at birth and continued for at least 10 minutes.¹⁵  Once these criteria were met, all infants underwent a standardized neurologic examination to identify infants with moderate or severe hypoxic-ischemic encephalopathy for inclusion in the trial.¹⁵  The trial was stopped for in-hospital mortality and futility concerns by the data safety and monitoring committee after enrollment of 364 infants.¹⁵  All 347 infants with primary outcome data were included in the current study as the follow-up analysis of the Optimizing Cooling Trial found no difference in death or moderate or severe disability between hypothermia treatment groups of longer versus usual duration of cooling and deeper versus usual depth of cooling. The trial had random assignment stratified by the center and level of encephalopathy in a 2 × 2 factorial design to 33.5°C or 32.0°C for 72 hours or 120 hours. The institutional review board approved the trial at each site. Written informed consent was obtained from the parents of study participants.

As death is a competing outcome for disability, the composite outcome of death or moderate or severe disability at 18 to 22 months of age was selected as the primary outcome similar to the Optimizing Cooling Trial. Follow-up evaluations were completed in 95% (347 of 364) of the infants at 18 to 22 months of age. Neurologic and developmental assessments were performed by trained and annually certified examiners masked to the intervention, and data on vision and hearing were obtained.¹⁴  Each center had one or more certified examiners, including neonatologists or developmental pediatricians.¹⁶  Severe disability was defined as any of the following: a Bayley III cognitive score of <70, a Gross Motor Function Classification System (GMFCS)¹⁷  level of 3 to 5, blindness, or profound hearing loss (inability to understand commands despite amplification). Moderate disability was defined as a Bayley III cognitive score of 70 to 84 and either a GMFCS level of 2, a seizure disorder, or a hearing deficit requiring amplification to understand commands. Mild disability was defined as a cognitive score of 70 to 84, or a cognitive score of ≥85 and any of the following: presence of a GMFCS level 1 or 2, seizure disorder, or hearing loss not requiring amplification. Normal neurocognitive status was defined as a cognitive score of ≥85 in the absence of any neurosensory deficits or seizures after NICU discharge.¹⁴ 

The primary analysis was performed by using all study participants with an Apgar score at 10 minutes and outcomes irrespective of the depth or duration of therapeutic hypothermia received. Bivariate analyses (χ² tests and t tests) were performed to identify maternal and fetal variables and neonatal variables available during resuscitation associated with outcomes. The variables considered for bivariate analyses included maternal demographic characteristics (race, education, insurance status, age, and gravidity), prenatal variables (maternal hypertension and diabetes), peripartum variables (fetal heart rate decelerations, antepartum hemorrhage, cord prolapse, uterine rupture, maternal pyrexia, shoulder dystocia, timing of rupture of membranes, and emergency cesarean delivery), and neonatal variables (birth weight, gestational age, sex, Apgar scores, resuscitation at birth and at 10 minutes, including individual interventions and medications, time to spontaneous respiration during resuscitation, cord or first postnatal blood-gas-base deficit, pH, Pco₂, inotropic support requirement before enrollment, seizures before enrollment, and trial group assignment). Maternal hypertension was defined as the presence of chronic hypertension, preeclampsia, or eclampsia. Differences in overall Apgar scores by the outcome were compared by using median tests. Because the number of infants with missing 15- and 20-minute Apgar score with a last nonmissing Apgar score of <7 were high (153 of 347, 44.0% and 178 of 347, 51.3%, respectively), analyses were limited to Apgar scores ≤10 minutes to avoid underpowered exploratory analysis. Generalized estimating equation models were used to examine relationships between Apgar scores and outcomes, controlling for hypothermia treatment assignment and hypoxic-ischemic encephalopathy (HIE) level and accounting for clustering by the center. In addition, an interaction between Apgar score and hypothermia treatment was tested by using a similar generalized estimating equation model to determine if the hypothermia treatment effect differed across Apgar scores. Associations were expressed as relative risk (RR) and 95% confidence intervals (CI). Receiver operator characteristic curve area under the curve (AUC) analysis was conducted to assess the predictive ability of the models for death or moderate or severe disability.

Classification and regression tree (CART) analyses¹⁸  were done to determine cut points for Apgar scores and identify combinations of predictive variables (available during resuscitation and those found to be associated with outcomes on bivariate analyses) that were most predictive for the composite outcome and death. To develop these models, the 10-minute Apgar score was introduced as the first variable, allowing the variable to determine the cut point for the score that best discriminated between infants with versus without the outcome, and then the model selected the subsequent variables with the best discrimination and their cut points.

On the basis of the incidence of the primary outcome in the Optimizing Cooling trial, the available sample size was sufficient to determine if the 10-minute Apgar score was predictive of the primary outcome with the receiver operator characteristic AUC values of ≥0.65 with a Type-I (α) error of 0.01 and Type-II (β) error of 0.05.

The study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology statement for reporting observational studies.¹⁹  The Optimizing Cooling Trial was registered with ClinicalTrials.gov (NCT01192776).

A total of 354 infants had data on survival, and among these infants, 347 had 18- to 22-month outcomes. Of those with 18- to 22-month outcomes, 307 infants had a documented 10-minute Apgar score and 225 infants had been transferred to the study centers for hypothermia treatment. There were 26 infants with an Apgar score of 0 at 10 minutes. Median 10-minute Apgar scores were significantly lower among infants with the composite outcome (P value < .001; Table 1). In addition, several variables available during resuscitation were found to be associated with outcomes (P value < .1) by bivariate analysis (Table 1).

A 10-minute Apgar score of 0 was independently predictive of death or moderate/severe disability (adjusted relative risk = 1.72, 95% CI 1.11–2.68, P value = .016); however, the AUC was low (AUC = 0.56, Table 2). Half of the infants (50%, 13 of 26) with a 10-minute Apgar score of 0 survived. Of the survivors with a 10-minute Apgar score of 0, 46% (6 of 13) had no disability, 16% (2 of 13) had mild disability, and 38% (5 of 13) had moderate or severe disability (Fig 1, Supplemental Table 3). There was no significant interaction between hypothermia treatment and 10-minute Apgar score for death and/or moderate or severe disability (Wald χ² (3) = 3.86, P value = .277) and death (Wald χ² (3) = 1.71, P value = .635).

CART analysis-based prediction models were built to predict the risk of death or moderate or severe disability and death at 10 minutes by using the variables identified by the bivariate analysis and the Apgar score as the first variable. The CART analysis-based prediction model to predict the risk of death or moderate or severe disability identified a 10-minute Apgar score <2 and no maternal hypertension as the most predictive variables (Fig 2). The AUC of the 10-minute model to predict the risk of death or moderate or severe disability was improved to 0.66 compared with the AUC of the 10-minute Apgar score alone of 0.56. The CART analysis-based prediction model to predict the risk of death identified a 10-minute Apgar score <2 and cord blood pH <6.9 as the most predictive variables (Fig 3). On the basis of the variables and cutoffs identified by the CART analysis, infants were grouped and compared. This analysis is consistent with the CART analysis identification of high-risk infants (Supplemental Tables 4 and 5).

The current study aimed to evaluate the predictive performance of Apgar scores for death or moderate or severe disability at 18 to 22 months of age using data from the most recent and larger sample-sized Optimizing Cooling Trial. This study reveals that, although the 10-minute Apgar score is predictive of death or moderate or severe disability, the predictive accuracy is low. The predictive accuracy of the 10-minute Apgar score improves when combined with other CART-analysis-identified predictive variables. The majority of surviving infants with a 10-minute Apgar score of 0 did not have moderate or severe disability at 18 to 22 months of age, indicating that prediction of the outcome at 10 minutes may be too early.

Therapeutic hypothermia is now an established intervention for reducing death or moderate or severe disability in term and late preterm infants with moderate or severe hypoxic-ischemic encephalopathy at birth,²⁰  including the infants with low 10-minute Apgar scores.^9,10  Few studies reported outcomes of infants with a low 10-minute Apgar score in term and late preterm infants born before therapeutic hypothermia became the standard of care for treating infants with moderate or severe hypoxic-ischemic encephalopathy.^6,21,22  However, as therapeutic hypothermia is proven to reduce death or moderate or severe disability in term and late preterm infants with moderate or severe hypoxic-ischemic encephalopathy and is widely accepted as the standard of care, the previous studies may not be clinically relevant for informing current practice.

In this study, 31% (8 of 26) of infants with a 10-minute Apgar score of 0 survived with mild or no disability at 18 to 22 months of age. This finding of 31% survival with mild or no disability at 18 to 22 months of age with a 10-minute Apgar score of 0 is important to remember during resuscitation scenarios in which the infant requires ongoing cardiopulmonary resuscitation at 10 minutes and highlights the need for continuing resuscitation for such infants. There have been a few single-center retrospective cohort studies that have reported outcomes of infants with a 10-minute Apgar score of 0 (N = 9–13) who received therapeutic hypothermia and have reported wide variations in survival (25% to 89%) and survival with mild or no disability (0% to 56%) at 9 to 24 months of age.^10,23–25  In a recently published metaanalysis by the ILCOR Neonatal Life Support Task Force on the duration of resuscitation at birth and outcomes, these studies were identified as having a high risk of bias.²⁶  In a study of infants who received therapeutic hypothermia in 117 NICUs from Japan,²⁷  11% (3 of 28) of infants with a 10-minute Apgar score of 0 survived with mild or no disability at 18 to 22 months of age. In the initial NRN therapeutic hypothermia trial, 24% (6 of 25) of the infants with a 10-minute Apgar score of 0 survived with mild or no disability at 18 to 22 months of age,⁷  and 21% (5 of 24) of the infants survived with mild or no disability at 6 to 7 years of age.⁸  In the TOBY trial,²⁸  37.5% (9 of 24) infants with a 10-minute Apgar score of 0 survived with mild or no disability at 18 months of age.¹⁰  In the ICE trial,²⁹  18.2% (2 of 11) infants with a 10-minute Apgar score of 0 survived with mild or no disability at 24 months of age.¹⁰  In the current trial, 31% of the survivors had no or mild disability. The earlier studies that reported outcomes from the hypothermia trials included control infants who did not receive therapeutic hypothermia, but all infants in the current report received hypothermia for the usual duration and usual depth and longer duration and greater depth.

The current study leverages the large clinical trial and follow-up infrastructure of the NRN and provides high-quality evidence for informing current practice from a recent therapeutic hypothermia trial from 18 major US academic centers. The current study provides evidence in support of the 2020 AHA/ILCOR recommendation for continuing resuscitative efforts for infants who have a 10-minute Apgar score of 0. The CART analysis-identified predictive variables, including the absence of maternal hypertension, low Apgar score, and higher cord gas base deficit, have also been reported from the secondary analysis of the initial NRN therapeutic hypothermia trial to predict the risk of death or moderate or severe disability.³⁰  Placental insufficiency seen in hypertensive disorders of pregnancy may be a potential preconditioning event triggering adaptive fetal mechanisms that may alleviate the cellular injuries associated with hypoxic-ischemic encephalopathy, possibly explaining the association of the absence of maternal hypertension with an increase in the risk of death or moderate or severe disability in infants with hypoxic-ischemic encephalopathy.³¹  CART analysis-based models using neurologic examination findings from the initial NRN therapeutic hypothermia trial and the Optimizing Cooling trial, including variables from the first 6 hours after birth have identified variables like decerebrate posture, lack of spontaneous activity, presence of seizures, and absence of suck as other important variables for prediction of death or moderate or severe disability after the immediate resuscitation period.^30,32  The 10-minute Apgar score has low predictive accuracy for clinical use. Apgar scores alone should not be used to make decisions regarding discontinuing resuscitation efforts. In infants without a heart rate after optimal resuscitation efforts, cessation of resuscitation efforts should be discussed with the team and the family.¹² 

The current study reports 18- to 22-month outcomes, but longer-term follow-up evaluation may provide more comprehensive evidence of neurodevelopmental and functional outcomes. Apgar scores at 15 and 20 minutes were missing in many infants. Because of missing Apgar scores, associations between 15- and 20-minute Apgar scores with death or disability could not be reliably analyzed. This study took place in a resource-rich setting with access to therapeutic hypothermia and the presence of skilled providers in the delivery room. Outcomes of the infants with hypoxic-ischemic encephalopathy treated at tertiary academic centers that are part of the NRN may differ from lower-level centers. Another limitation is that the data for infants who died in the delivery room or whose parents did not provide consent for the trial were not included in the analysis, which could affect the prediction models. Because the Optimizing Cooling Trial excluded infants with major congenital abnormality and severe growth restriction (birth weight <1800 g), these infants may have different outcomes than the outcomes seen in the current study. The strengths of the current study include collected data with multiple quality assurance checks from a well-conducted rigorous clinical trial. Because of the large sample size, we were able to conduct CART analysis and derive models with better predictive accuracy.

A 10-minute Apgar score of 0 alone does not predict the risk of death or moderate or severe disability well. The risk predictive accuracy of the 10-minute Apgar score is improved when combined with other variables but remains low, thus indicating that prediction of the outcome at 10 minutes may be too early. The current study provides evidence in support of the 2020 AHA/ILCOR recommendation for consideration of continuation resuscitative efforts for infants who need cardiopulmonary resuscitation at 10 to 20 minutes after birth instead of discontinuing resuscitation at 10 minutes after birth as suggested in the previous recommendation. The decision to discontinue resuscitation should be individualized, depending on many factors, including the timing of perinatal insult, available resources, and effectiveness of resuscitation.

We are indebted to our medical and nursing colleagues and the infants and their parents who agreed to take part in this study. The following investigators, in addition to those listed as authors, participated in this study:

NRN Steering Committee Chair: Richard A. Polin, MD, Division of Neonatology, College of Physicians and Surgeons, Columbia University, (2011–present).

Alpert Medical School of Brown University and Women & Infants Hospital of Rhode Island (U10 HD27904): Martin Keszler, MD, William Oh, MD, Betty R. Vohr, MD, Angelita M. Hensman, MS, RNC-NIC, Robert T. Burke, MD, MPH, Melinda Caskey, MD, Nicholas Guerina, MD, PhD, Andrea M. Knoll, Emilee Little, RN, BSN, Ross Sommers, MD, Birju A. Shah, MD, MPH, Elisa Vieira, RN, BSN, Barbara Alksninis, RNC, PNP, Mary Lenore Keszler, MD, Elisabeth C. McGowan, MD.

Case Western Reserve University, Rainbow Babies & Children's Hospital (U10 HD21364): Michele C. Walsh, MD, MS, Anna Maria Hibbs, MD, Nancy S. Newman, BA, RN, Bonnie S. Siner, RN, Eileen K. Stork, MD, Arlene Zadell, RN.

Children's Mercy Hospital, University of Missouri Kansas City School of Medicine (U10 HD68284): William E. Truog, MD, Eugenia K. Pallotto, MD, MSCE, Cheri Gauldin, RN, BSN, CCRC, Howard W. Kilbride, MD, Lisa Gaetano, MSN, RN, Anne M. Holmes, RN, MSN, MBA-HCM, CCRC, Kathy Johnson, RN, CCRC, Allison Knutson, BSN, RNC-NIC.

Cincinnati Children's Hospital Medical Center, University of Cincinnati Medical Center, and Good Samaritan Hospital (U10 HD27853, UL1 TR77): Brenda B. Poindexter, MD, MS, Kurt Schibler, MD, Cathy Grisby, BSN, CCRC, Suhas G. Kallapur, MD, Barbara Alexander, RN, Lenora Jackson, CRC, Kristin Kirker, CRC, Teresa L. Gratton, PA, Stephanie Merhar, MD, MS, Greg Muthig, BA, Sandra Wuertz, RN, BSN, CLC, Kimberly Yolton, PhD.

Duke University School of Medicine, University Hospital, University of North Carolina, and Duke Regional Hospital (U10 HD40492, UL1 TR1117): C. Michael Cotten, MD, MHS, Ronald N. Goldberg, MD, Kimberley A. Fisher, PhD, FNP-BC, IBCLC, Joanne Finkle, RN, JD, Sandra Grimes, RN, BSN, Ricki F. Goldstein, MD, Patricia L. Ashley, MD, PhD, William F. Malcolm, MD, Kathryn E. Gustafson, PhD, Matthew M. Laughon, MD, MPH, Carl L. Bose, MD, Janice Bernhardt, MS, RN, Cindy Clark, RN, Diane D. Warner, MD, MPH, Janice Wereszcsak, CPNP, Sofia Aliaga, MD, MPH.

Emory University, Children's Healthcare of Atlanta, Grady Memorial Hospital, and Emory University Hospital Midtown (U10 HD27851, UL1 TR454): Barbara J. Stoll, MD, David P. Carlton, MD, Shannon E. G. Hamrick, MD, Ellen C. Hale, RN, BS, CCRC, Yvonne Loggins, RN, Colleen Mackie, BS, RT, Diane I. Bottcher, MSN, RN.

Eunice Kennedy Shriver National Institute of Child Health and Human Development: Stephanie Wilson Archer, MA.

Indiana University, Riley Hospital for Children and Methodist Hospital at Indiana University Health (U10 HD27856): Gregory M. Sokol, MD, Brenda B. Poindexter, MD, MS, Heidi M. Harmon, MD, MS, Dianne E. Herron, RN, CCRC, Abbey C. Hines, PsyD, Susan Gunn, NNP-BC, CCRC, Lucy C. Smiley, CCRC.

McGovern Medical School at The University of Texas Health Science Center at Houston, Children's Memorial Hermann Hospital (U10 HD21373): Jon E. Tyson, MD, MPH, Amir M. Khan, MD, Kathleen A. Kennedy, MD, MPH, Claudia Pedroza, PhD, Julie Arldt-McAlister, MSN, APRN, Katrina Burson, RN, BSN, Allison G. Dempsey, PhD, Andrea F. Duncan, MD, MSClinRes, Carmen Garcia, RN, BSN, Janice John, CPNP, Patrick M. Jones, MD, MA, M. Layne Lillie, RN, BSN, Sara Martin, RN, Georgia E. McDavid, RN, Saba Siddiki, MD, Daniel K. Sperry, RN, Patti L. Pierce Tate, RCP, Sharon L. Wright, MT (ASCP).

Nationwide Children's Hospital and The Ohio State University Wexner Medical Center (U10 HD68278): Pablo J. Sánchez, MD, Leif D. Nelin, MD, Sudarshan R. Jadcherla, MD, Edward G. Shepherd, MD, Patricia Luzader, RN, Christine A. Fortney, PhD, RN, Nehal A. Parikh, MD.

RTI International (U10 HD36790): Abhik Das, PhD, Dennis Wallace, PhD, Marie G. Gantz, PhD, Carolyn M. Petrie Huitema, MS, CCRP, Jeanette O'Donnell Auman, BS, Margaret M. Crawford, BS, CCRP, Jenna Gabrio, BS, CCRP, Jamie E. Newman, PhD, MPH, James W. Pickett II, BS, Annie M. VonLehmden, BS, Kristin M. Zaterka-Baxter, RN, BSN.

Stanford University and Lucile Packard Children's Hospital (U10 HD27880, UL1 TR93): Krisa P. Van Meurs, MD, David K. Stevenson, MD, M. Bethany Ball, BS, CCRC, Lynne C. Huffman, MD, Anne M. DeBattista, RN, PNP, PhD, Hali E. Weiss, MD, Maria Elena DeAnda, PhD, Casey E. Krueger, PhD, Melinda S. Proud, RCP.

University of Alabama at Birmingham Health System and Children's Hospital of Alabama (U10 HD34216): Monica V. Collins, RN, BSN, MaEd, Shirley S. Cosby, RN, BSN.

University of California, Los Angeles, Mattel Children's Hospital, Santa Monica Hospital, Los Robles Hospital and Medical Center, and Olive View Medical Center (U10 HD68270): Uday Devaskar, MD, Meena Garg, MD, Teresa Chanlaw, MPH, Rachel Geller, RN, BSN.

University of Iowa and Mercy Medical Center (U10 HD53109, UL1 TR442): Edward F. Bell, MD, Tarah T. Colaizy, MD, MPH, Karen J. Johnson, RN, BSN, Jane E. Brumbaugh, MD, Diane L. Eastman, RN, CPNP, MA, Jacky R. Walker, RN, Claire A. Lindauer, RN, Jonathan M. Klein, MD, Jeffrey L. Segar, MD, John M. Dagle, MD, PhD, Julie B. Lindower, MD, MPH, Steven J. McElroy, MD, Glenda K. Rabe, MD, Robert D. Roghair, MD, Lauritz R. Meyer, MD, Cary R. Murphy, MD, Vipinchandra Bhavsar, MB, BS, Dan L. Ellsbury, MD, Donia B. Campbell, RNC-NIC, Cary R. Murphy, MD, Vipinchandra Bhavsar, MB, BS.

University of New Mexico Health Sciences Center (U10 HD53089, UL1TR41): Kristi L. Watterberg, MD, Robin K. Ohls, MD, Conra Backstrom Lacy, RN, Mary Ruffaner Hanson, RN, BSN, Sandra Sundquist Beauman, MSN, RNC, Carol H, Hartenberger, MPH, RN, Janell Fuller, MD, Andrea Freeman Duncan, MD, MScr, Jean R. Lowe, PhD.

University of Pennsylvania, Hospital of the University of Pennsylvania, Pennsylvania Hospital, and Children's Hospital of Philadelphia (U10 HD68244): Barbara Schmidt, MD, MSc, Haresh Kirpalani, MB, MSc, Sara B. DeMauro, MD, MSCE, Kevin C. Dysart, MD, Soraya Abbasi, MD, Aasma S. Chaudhary, BS, RRT, Toni Mancini, RN, BSN, CCRC, Dara M. Cucinotta, RN, Judy C. Bernbaum, MD, Marsha Gerdes, PhD, Hallam Hurt, MD.

University of Rochester Medical Center, Golisano Children's Hospital, and the University of Buffalo Women's and Children's Hospital of Buffalo (U10 HD68263, UL1 TR42): Carl D'Angio, MD, Ronnie Guillet, MD, PhD, Satyan Lakshminrusimha, MD, Michelle E. Hartley-McAndrew, MD, Nirupama Laroia, MD, Gary J. Myers, MD, Kelley Yost, PhD, Stephanie Guilford, BS, Rosemary Jensen, Karen Wynn, NNP, RN, Osman Farooq, MD, Anne Marie Reynolds, MD, MPH, Michael G. Sacilowski, MAT, Ann Marie Scorsone, MS, Holly I.M. Wadkins, MA, Ashley Williams, MS, Ed, Joan Merzbach, LMSW.

University of Texas Southwestern Medical Center, Parkland Health & Hospital System, and Children's Medical Center Dallas (U10 HD40689): Myra H. Wyckoff, MD, Luc P. Brion, MD, Lina F. Chalak, MD, MSCS, Roy J. Heyne, MD, Lijun Chen, PhD, RN, Diana M. Vasil, MSN, BSN, RNC-NIC, Sally S. Adams, MS, RN, CPNP, Catherine Twell Boatman, MS, CIMI, Alicia Guzman, Elizabeth T. Heyne, MS, MA, PA-C, PsyD, Lizette E. Lee, RN, Linda A. Madden, BSN, RN, CPNP, Emma Ramon, RNC-NIC, RN, BSN.

Wayne State University, University of Michigan, Hutzel Women's Hospital and Children's Hospital of Michigan (U10 HD21385): Athina Pappas, MD, Beena G. Sood, MD, MS, Rebecca Bara, RN, BSN, Kirsten Childs, RN, BSN, Mary E. Johnson, RN, BSN, Bogdan Panaitescu, MD, Sanjay Chawla, MD, Jeannette E. Prentice, MD, Lilia C. De Jesus, MD, Eunice Hinz Woldt, RN, MSN, Girija Natarajan, MD, Monika Bajaj, MD, John Barks, MD, Mary Christensen, RT, Stephanie A. Wiggins, MS.

Yale University (U10 HD27871): Richard A. Ehrenkranz, MD.

AHA

American Heart Association

AUC

area under the curve analysis

CART

classification and regression tree analysis

CI

confidence interval

DCC

data coordinating center

GMFCS

Gross Motor Function Classification System

ILCOR

International Liaison Committee on Resuscitation

NICHD

National Institute of Child Health and Human Development

NRN

National Institute of Child Health and Human Development Neonatal Research Network

RR

relative risk