Non-Intubated Spontaneous Ventilation Versus Endotracheal Intubation Anesthesia for Pediatric Thoracoscopic Lung Resection: A Retrospective Propensity-Score-Matched Study.

BACKGROUND

Non-intubated spontaneous ventilation anesthesia is increasingly used in adult thoracoscopic surgery, but pediatric evidence remains limited.

AIMS: To compare perioperative outcomes between two anesthetic regimens in children undergoing thoracoscopic anatomical lung lesion resection: a laryngeal mask airway-assisted non-intubated spontaneous ventilation anesthesia and a conventional endotracheal intubation anesthesia.

METHODS

This single-center retrospective cohort study included children undergoing thoracoscopic anatomical lung lesion resection for pulmonary sequestration or congenital pulmonary airway malformation between April 2024 and May 2025. Children received either a conventional endotracheal intubation anesthesia or a laryngeal mask airway-assisted non-intubated spontaneous ventilation anesthesia. One-to-one propensity score matching was performed. The primary outcome was the overall incidence of postoperative pulmonary complications before discharge. Secondary outcomes were conversion to endotracheal intubation, intraoperative lowest peripheral oxygen saturation, highest end-tidal carbon dioxide, postanesthesia care unit length of stay, chest tube indwelling time, and postoperative length of hospital stay. Other perioperative variables were exploratory.

RESULTS

Among 198 eligible children, 50 matched pairs were analyzed. Postoperative pulmonary complications occurred in 6 of 50 children (12.0%) in the endotracheal intubation group and in 0 of 50 children (0%) in the non-intubated spontaneous ventilation group (absolute risk reduction 12.0%, 95% confidence interval 3.0-21.0; p = 0.027). In a sensitivity analysis using Firth's logistic regression model adjusting for thoracic paravertebral block, the non-intubated regimen remained associated with a lower observed incidence of postoperative pulmonary complications (adjusted odds ratio 0.06, 95% confidence interval 0.0004-0.76; p = 0.029). No child in the non-intubated spontaneous ventilation group required conversion to endotracheal intubation. Secondary outcomes showed modest between-group differences in unadjusted analyses. However, after Holm-Bonferroni adjustment, adjusted p values for secondary outcomes ranged from 0.060 to 0.208.

CONCLUSION

In this retrospective propensity-score-matched pediatric cohort, a laryngeal mask airway-assisted non-intubated spontaneous ventilation anesthesia was feasible and was not associated with worse perioperative outcomes than a conventional endotracheal intubation anesthesia. A lower observed incidence of postoperative pulmonary complications was noted. Given regimen-level differences, prospective studies with standardized analgesia are needed to confirm safety and potential benefits.