EBNEO Commentary: Impact of attending neonatologist presence on neonatal intubation success and adverse events: a cohort study

MANUSCRIPT CITATION

Trinh C, Hodgson KA, Downes M, Manley BJ, Thio M, Assad M, Bibl K, Chabra S, DeMartino C, DeMeo S, Glass K, Herrick H, Quek BH, Iben S, Jung P, Kim J, Mehrem AA, Moussa A, Narvey M, O’Shea J, Pouppirt N, Puia-Dumitrescu M, Rumpel J, Shay R, Tingay D, Tyler M, Unrau J, Wagner M, Wildenhain P, Nishisaki A, Foglia EE. Impact of attending neonatologist presence on neonatal intubation success and adverse events: a cohort study. Journal of Perinatology 2025; PMID: 41282162

REVIEWED BY:

Dr Christopher Angus, Neonatal Research Fellow
Institution: Neonatal Unit, James Cook University Hospital, South Tees Hospitals NHS foundation Trust, Middlesbrough, UK
Email: Christopher.angus2@nhs.net

Prof Prakash Kannan Loganathan, Consultant Neonatologist
Email: pkannanloganathan@nhs.net
Telephone: 01642 850850 Ext 53517

CORRESPONDING AUTHOR

Name: Prof Prakash Kannan Loganathan, Consultant Neonatologist
Email: pkannanloganathan@nhs.net
Telephone: 01642 850850 Ext 53517

TYPE OF INVESTIGATION

Choose from one of the following: prevention, treatment, diagnosis, prognosis, etiology, quality improvement or continuing medical education, economics of healthcare or interventions, clinical prediction guides, differential diagnosis, or systematic review.

QUESTION

In neonates underdoing endotracheal intubation in the Neonatal Intensive Care Unit (NICU) or Delivery Room (P), does the presence of an attending Neonatologist (I) compared to absence (C), improve first attempt intubation success and reduces adverse tracheal intubation-associated events (TIAEs) (O), during the procedure (T).

METHODS

Design: Retrospective cohort study
Allocation: Not applicable
Blinding: Not applicable
Follow-up period: Not applicable
Setting: 19 neonatal intensive care units (NICUs), across the USA, Canada, Australia, Germany and Austria.
Patients: Neonates requiring endotracheal intubation. Inclusion: intubation occurred between October 1st, 2014, and December 31st, 2022 (16,010 endotracheal intubations). Exclusion: Intubations where an attending Neonatologist was the initial intubation operator. Intubations that had conflicting or missing data for the primary outcome (first attempt success), exposure of interest (attending Neonatologist presence) or covariates. After exclusion, there were total of 12,652 intubation encounters.
Intervention: Presence of attending neonatologist during intubation

Outcomes:
Primary outcome: First attempt intubation success, defined as successful intubation on the first attempt by the first operator.
Secondary outcomes: Severe and non-severe tracheal intubation adverse events (TIAEs), severe oxygen desaturation (≥20% absolute decrease in peripheral oxygen saturation), and multiple attempts (defined as ≥3 attempts)
Non-severe TIAEs: oesophageal intubation with immediate recognition, mainstem intubation, lip trauma, pain or agitation requiring additional sedation delaying intubation, epistaxis, emesis without aspiration, and dysrhythmia, including bradycardia with heart rate.
Severe TIAEs: direct airway injury, oesophageal intubation with delayed recognition, emesis with aspiration, laryngospasm, pneumothorax, gum or dental trauma, hypotension requiring intervention, need for cardiac compressions and cardiac arrest.

Analysis and Sample Size: 12,652 intubation encounters analysed. Directed Acyclic Graph (DAG) identified the following confounders: prematurity (<37 weeks), known difficult airway, operator grade (years), premedication, use of videolaryngoscope, nighttime intubation. Univariate logistic regression modelling was performed to estimate the unadjusted association between attending presence and each individual primary and secondary outcomes. Multivariable modelling using generalized estimating equations with covariate selection guided by the (DAG) and site clustering accounted for by use of an exchangeable correlation structure was then performed to estimate the adjusted causal effect of attending presence on each outcome. Patient follow-up: 100% of intubations meeting inclusion and exclusion criteria. MAIN RESULTS Of the 12,652 included intubations, 8391 (66.3%) had an attending present. Neonatologists were more likely to be present during intubation of infants who were of greater weight and age, had a known history of difficult airway, occurred during the day, and were in the delivery room. Intubations with an attending neonatologist present were also more likely to use a video laryngoscope. Univariate analysis found attending neonatologist presence was positively associated with first attempt intubation success (OR 1.11, 95% CI 1.04–1.20). However, on multivariate analysis, when controlling for confounders, attending presence was found to be negatively associated with first attempt intubation success (adjusted OR (aOR) 0.78, 95% CI 0.70–0.86). Similar negative association between attending presence and other outcomes are shown in table 1.

Table 1: Association of attending neonatologist presence with primary and secondary intubation outcomes: unadjusted and adjusted odds ratios (95% CI)

Subgroup analyses (Table 2) showed a similar direction of effect across operator experience, gestational age, known difficult airway, and intubation location, with the strongest association among the most experienced operators (aOR 0.67, 95% CI 0.50–0.91).

Table 2. Subgroup analyses of the impact of attending neonatologist presence on first attempt intubation success.

CONCLUSION

After adjustment for measured confounders, attending neonatologist presence was associated with lower odds of first-attempt intubation success and higher odds of multiple attempts, severe desaturation, and tracheal intubation-associated adverse events. These findings likely reflect residual confounding, particularly selective attending presence during higher-risk or more complex intubations, although other factors such as team dynamics may also have contributed.

COMMENTARY

Neonatal intubation remains a critical yet high-risk procedure, with low first attempt success rates, frequent desaturation and adverse tracheal intubation-associated events (TIAEs) (1, 2). Factors including operator experience, premedication, video laryngoscopy and nasal high flow therapy use during intubation have been shown to enhance success (3-5). Intuitively, the presence of an attending neonatologist (consultant) should enhance success by providing immediate expertise, leadership and oversight, potentially reducing risks for junior operators during this stressful procedure. Trinh et al conducted a large multicentre cohort study to evaluate this assumption, which had previously not been examined. Surprisingly, after adjustments for confounders, an unexpected association between attending neonatologist presence and poorer outcomes was discovered; lower first attempt success rates, and higher rates of ≥ 3 attempts, severe desaturation and TIAEs.

 

This study has important strengths, including a large international multicentre registry, use of a DAG to guide confounder selection, and generalized estimating equations to account for clustering by site. Prespecified subgroup analyses also suggested a similar direction of effect across key clinical strata, including preterm infants and those with difficult airways.

 

Despite these strengths, several limitations should be acknowledged. As the authors note, the retrospective observational design is susceptible to residual confounding from factors not captured in the registry, including illness severity, anticipated procedural difficulty, and the reason for attending presence, which may bias causal effect estimates. The registry also lacked data on the timing of attending involvement, attending coaching expertise, operator training differences, and changes in practice over the study period. Operator experience was measured in years rather than procedural volume or recency, making this a relatively imprecise proxy for intubation expertise. Important unit-level and procedural details, such as video laryngoscopy type, specific premedication regimens including atropine use, and variation in supervisory/coaching style, were also unavailable and may have contributed to residual confounding. Exclusion of 1790 intubations with missing confounder data may have introduced selection bias. In addition, limited power for individual adverse events necessitated use of composite TIAE outcomes, combining events of heterogeneous clinical importance and complicating interpretation. Finally, the marked reversal from a positive unadjusted association (OR 1.11) to a negative adjusted association (aOR 0.78) suggests substantial confounding, most likely related to selective attending presence during higher-risk intubations.

 

This study suggests that attending neonatologist supervision alone may not mitigate the risks associated with neonatal intubation. Further research should prospectively evaluate operator-specific and supervisory factors influencing intubation success, including reasons for attending presence, coaching style and consistency, illness severity, and procedural dynamics (for example, through video review). With declining intubation opportunities due to advances in neonatal care, balancing trainee proficiency with patient safety is paramount. This study prompts reflection: attending neonatologist presence alone may be insufficient. It underscores the need to optimise every first intubation attempt (6) by matching operators to procedures and utilising adjuncts such as video laryngoscopy, adequate premedication and nasal high flow therapy. Importantly, these findings should not be interpreted as diminishing the importance of senior supervision, which remains fundamental to governance and patient safety during neonatal intubation.

 

REFERENCES

1. Foglia EE, Ades A, Sawyer T, Glass KM, Singh N, Jung P, et al. Neonatal Intubation Practice and Outcomes: An International Registry Study. Pediatrics 2019; 143 1.
2. O’Donnell CP, Kamlin CO, Davis PG, Morley CJ. Endotracheal intubation attempts during neonatal resuscitation: success rates, duration, and adverse effects. Pediatrics 2006; 117 1:e16-21.
3. Johnston L, Sawyer T, Ades A, Moussa A, Zenge J, Jung P, et al. Impact of Physician Training Level on Neonatal Tracheal Intubation Success Rates and Adverse Events: A Report from National Emergency Airway Registry for Neonates (NEAR4NEOS). Neonatology 2021; 118 4:434-42.
4. Lingappan K, Neveln N, Arnold JL, Fernandes CJ, Pammi M. Videolaryngoscopy versus direct laryngoscopy for tracheal intubation in neonates. Cochrane Database Syst Rev 2023; 5 5:Cd009975.
5. Hodgson KA, Owen LS, Kamlin COF, Roberts CT, Newman SE, Francis KL, et al. Nasal High-Flow Therapy during Neonatal Endotracheal Intubation. The New England journal of medicine 2022; 386 17:1627-37.
6. Bernhard M, Becker TK, Gries A, Knapp J, Wenzel V. The First Shot Is Often the Best Shot: First-Pass Intubation Success in Emergency Airway Management. Anesth Analg 2015; 121 5:1389-93.