Hypoxemia on Emergence in Neonates and Young Infants: Incidence and Risk Factors in Patients < 2 Months Undergoing Inguinal Herniorrhaphy or Pyloromyotomy

Enhanced Recovery After Surgery (ERAS) guidelines for neonatal perioperative care recommend early postoperative extubation [1]. Emergence and extubation are particularly critical phases of anesthesia management and may be associated with significant morbidity [2, 3]. Yet there remains a paucity of literature describing risk factors for extubation-related complications in neonates. Most prior work has focused on factors associated with successful intubation and strategies to minimize airway-related complications, including multiple laryngoscopy attempts, hypoxemia, and cardiac arrest during induction of general anesthesia [4, 5]. We present a cohort of neonates and young infants who underwent general anesthesia with tracheal intubation for inguinal herniorrhaphy or pyloromyotomy and were subsequently extubated at case completion. We chose these surgical procedures as they have a high rate of intraoperative extubation and would include both preterm neonates with medical comorbidities and otherwise healthy infants. We hypothesized that patient factors including prematurity and medical comorbidities would be associated with hypoxemia on emergence. Our primary outcome was the incidence of clinically significant hypoxemia on emergence; secondary outcomes included incidence of reintubation and identification of factors associated with hypoxemia on emergence. Following IRB approval, we retrospectively reviewed all anesthetics in neonates and infants < 2 months of age (< 48 weeks postmenstrual age for preterm infants) who underwent general anesthesia with tracheal intubation and subsequent extubation in the operating room at our institution from August 2012 to May 2024. Demographic data, medical comorbidities, airway management variables, and vital sign trends were extracted and manually reviewed. Clinically significant hypoxemia was defined as ≥ 10% decrease in SpO2 from baseline and lasting ≥ 2 min. The authors manually reviewed charts with possible hypoxemia and confirmed the hypoxemia and classified the timing during the anesthetic (induction, maintenance, or emergence). Point estimates were calculated with 95% binomial confidence intervals (CI). Logistic regression was used to identify factors independently associated with hypoxemia on emergence. Statistical significance was defined as p < 0.05. A total of 979 cases in 960 unique patients met inclusion criteria. Median age was 43.4 days (IQR 29.9–60.4), median weight was 3.9 kg (IQR 3.3–4.5), and 34.2% (335/979) were born preterm (Table 1). The proportion of preterm and medically complex infants increased over time. From 2018 to 2024, 41% (213/519) were born preterm and 16.8% (87/519) diagnosed with respiratory distress syndrome (RDS) or bronchopulmonary dysplasia (BPD) compared with 26.5% (122/460) born prematurely and 10.0% (46/460) with RDS or BPD from 2012 to 2017. Infants who experienced hypoxemia on emergence had an older median chronologic age (87.5 days; IQR 54–98.3), reflecting that these infants were frequently born prematurely. Clinically significant hypoxemia on induction occurred in 53 cases (5.4%, 95% CI 4.1%–7.0%). 1.21 per 5 days (1.04, 1.41) 1.05 per week (1.01, 1.09) Clinically significant hypoxemia at induction Clinically significant hypoxemia on emergence occurred in 30 cases (3.1%; 95% CI 2.1%–4.4%). Five cases (0.5%; 95% CI 0.1%–1.2%) required reintubation. On multivariable analysis, prematurity (aOR 4.74; 95% CI 1.5–15; p = 0.008), a history of BPD or RDS (aOR 4.71; 95% CI 1.89–11.7; p = 0.001), and significant hypoxemia during induction (aOR 5.97; 95% CI 2.1–17; p = 0.001) were independently associated with hypoxemia on emergence. In this cohort, approximately 3% of neonates and young infants who were extubated following inguinal herniorrhaphy or pyloromyotomy experienced clinically significant hypoxemia on emergence and 0.5% required reintubation in the operating room. Neonates and young infants who were born prematurely, those with underlying pulmonary disease, and those who experienced hypoxemia during anesthesia induction were more likely to experience hypoxemia on emergence. Our findings align with prior literature demonstrating that preterm infants, particularly those with a history of BPD, are at an increased risk for postoperative respiratory events [6, 7]. Our findings build on this by demonstrating that hypoxemia during induction may be associated with reduced or compromised respiratory reserve, prompting anesthesiologists to anticipate a higher risk for airway-related complications at emergence. Similarly, clinicians may anticipate that former preterm neonates and infants with RDS or BPD require individualized extubation planning, with additional respiratory support readily available for this vulnerable subset of infants. This study is limited due to its retrospective design and reliance on electronic documentation. However, the large sample size, rigorous review of physiologic data, and clinically meaningful outcome definition strengthen our findings. Given that prolonged intubation in neonates carries well-recognized risks including subglottic inflammation and stenosis, ventilator-associated pneumonia, and feeding or swallowing difficulties; emergence and extubation in the operating room have been recommended as part of neonatal perioperative ERAS guidelines [1]. Still, emergence in high-risk infants warrants a cautious approach, particularly when intraoperative respiratory events are encountered, as a subset of patients may require additional respiratory support or reintubation. Understanding factors associated with clinically significant hypoxemia on emergence can improve perioperative risk stratification, guide extubation planning, and inform resource allocation. Future studies could evaluate the role of targeted noninvasive support strategies in improving extubation outcomes among high-risk neonates. The authors gratefully acknowledge James Peyton, Raymond Park, Pete Kovatsis, Amira Toivonen, and Gabor Asztalos for their assistance with this work. The authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.