Abstract
IMPORTANCE: The Apgar score, the first clinical assessment to direct measures to stabilize newborn infants, is also used for risk assessment. Its accuracy in estimating outcomes remains poor among very preterm (VPT) infants. OBJECTIVE: To assess the utility of the combined 5-minute Apgar score and umbilical artery pH (UA-pH) for estimating risks of mortality and severe neonatal morbidity among VPT infants. DESIGN, SETTING, AND PARTICIPANTS: This cohort study (Effective Perinatal Intensive Care in Europe [EPICE]) analyzed infants born at less than 32 weeks' gestation between April 2011 and September 2012 across 11 European countries. All liveborn VPT infants with Apgar scores and UA-pH data were included. Data were analyzed between February and December 2025. EXPOSURES: Apgar score at 5 minutes and UA-pH. The Apgar score was classified as lower than 7 and 7 or higher, and the UA-pH values were categorized as low (<7.20) and normal (≥7.20). Four groups that combined these 2 measures were defined: Apgar score lower than 7 and low UA-pH; Apgar score lower than 7 and normal UA-pH; Apgar score 7 or higher and low UA-pH; and Apgar score 7 or higher and normal UA-pH. MAIN OUTCOMES AND MEASURES: Combined outcome of mortality and/or any adverse morbidity (intraventricular hemorrhage [IVH] >grade 2, cystic periventricular leukomalacia, moderate or severe bronchopulmonary dysplasia [BPD], retinopathy of prematurity ≥stage 2, and necrotizing enterocolitis). Modified Poisson regression was used to estimate relative risks (RRs) between the exposure and the combined mortality and morbidity outcome and 3 individual components: mortality, IVH, and BPD. Models were adjusted for perinatal variables associated with Apgar score and UA-pH and adverse neonatal outcomes. RESULTS: Of 7900 liveborn infants in the EPICE cohort, 4174 (52.8%) had information on Apgar score and UA-pH. These infants included 2249 males (53.9%) and had a median [IQR] gestational age of 29.9 [27.9-31.0] weeks and median [IQR] birth weight of 1240 [960-1520] g. A total of 367 infants (8.8%) had an Apgar score 7 or higher but a low UA-pH, 558 (13.4%) had an Apgar score lower than 7 but a normal UA-pH, and 196 (4.7%) had an Apgar score lower than 7 and a low UA-pH. Infants with an Apgar score lower than 7 had a higher frequency of the combined outcome among those with a normal UA-pH (270 [48.4%] vs 596 [19.5%]) and a low UA-pH (108 [55.1%] vs 596 [19.5%]), with similar adjusted RRs (ARRs; low: 1.4 [95% CI, 1.2-1.7]; normal: 1.4 [95% CI, 1.3-1.6]). For mortality risk, associations were robust for an Apgar score lower than 7 and a low UA-pH (ARR, 2.4; 95% CI, 1.7-3.3) and absent with an Apgar score of 7 or higher and a low UA-pH (ARR, 1.2; 95% CI, 0.8-1.8). IVH risk was increased in all 3 subcategories, including an Apgar score of 7 or higher with a low UA-pH (ARR, 2.0; 95% CI, 1.3-3.0). BPD risk was associated only with an Apgar score lower than 7 and a normal UA-pH (ARR, 1.4; 95% CI, 1.2-1.7). CONCLUSIONS AND RELEVANCE: In this cohort study of VPT infants, combining information on UA-pH with the 5-minute Apgar score was associated with improved accuracy in estimating the risk of some adverse outcomes-notably mortality and IVH, which occurred soon after birth. These results highlight the importance of exploring the associations of early markers of risk with neonatal mortality and key neonatal morbidities separately.