Airway management and postoperative respiratory recovery after intravitreal injection in preterm infants.

BACKGROUND: Preterm infants undergoing intravitreal ranibizumab injection for retinopathy of prematurity are vulnerable to perioperative respiratory instability. Whether airway strategy is associated with postoperative respiratory recovery remains unclear. METHODS: We conducted a retrospective single-center cohort study of 134 preterm infants undergoing intravitreal ranibizumab injection under general anesthesia. Infants were managed with a laryngeal mask airway (LMA, n = 83) or endotracheal intubation (ETI, n = 51). The primary outcome was respiratory support escalation within 24 hours. The primary analysis used inverse probability of treatment weighting. RESULTS: Respiratory support escalation occurred in 6/83 (7.2%) infants in the LMA group and 20/51 (39.2%) in the ETI group. In the weighted analysis, LMA use was associated with a lower risk than ETI (RR 0.36, 95% CI 0.15-0.88; P = 0.026). In unadjusted secondary analyses, the LMA group was more likely to return to baseline support by 24 hours (90.4% versus 66.7%, P = 0.012) and less likely to require unplanned reintubation (1.2% versus 9.8%, P = 0.044). CONCLUSION: In selected preterm infants undergoing intravitreal injection under general anesthesia in the operating room, LMA use was associated with better early postoperative respiratory recovery than ETI. The findings may apply when LMA placement is technically feasible and clinically appropriate. IMPACT: In preterm infants undergoing intravitreal injection under general anesthesia in the operating room, laryngeal mask airway use was associated with a lower risk of respiratory support escalation within 24 hours than endotracheal intubation. Using inverse probability of treatment weighting to address measured baseline differences, this study provides a confounding-adjusted estimate and extends observations from laser photocoagulation to the briefer setting of intravitreal injection. The findings are associative and hypothesis-generating and may inform airway-device selection when laryngeal mask placement is technically feasible and clinically considered an appropriate airway option. Prospective multicenter studies are warranted.