P2.P315: Thoracoscopy is Equivalent to Thoracotomy for Neonatal Procedures: A Propensity Score Matched Analysis using NSQIP-P

Background: Minimally invasive thoracic surgery in neonates is highly challenging, with improving, but unclear safety and efficacy in the short- and long-term. We aimed to investigate the safety of thoracic minimally invasive surgery (MIS) in neonates by comparing the 30-day outcomes of common thoracoscopic procedures to those of open procedures using the National Surgical Quality Improvement Program - Pediatric (NSQIP-P) database.

Methods: The NSQIP-P database was queried for patients < 28 days age who underwent an MIS or open thoracic procedure for Esophageal Atresia/Tracheoesophageal Fistula (EA/TEF) or Congenital Diaphragmatic Hernia (CDH) from 2017-2021. Variables which differed significantly (standardized mean difference > 0.10) between treatment groups were propensity score matched using nearest neighbor matching to adjust for variation in baseline characteristics amongst patients undergoing open and MIS procedures. Subsequent analysis was conducted on an intention-to-treat basis. Logistic regression was conducted on the matched data stratified for disease type (EA/TEF and CDH) to determine the association between procedures and 30-day adverse outcomes. Findings were presented as adjusted Odd’s ratios with 95% Confidence Intervals.

Results: A total of 2701 patients were included, of whom 1160 (42.9%) had EA/TEF and 1541 (57.1%) CDH. Of these, 160 (13.6%) and 167 (10.8%) patients with EA/TEF and CDH respectively underwent MIS procedures. Treatment groups were matched using propensity scores for the following clinical characteristics: prematurity, preoperative ventilation requirement, preoperative inotropic support, weight at surgery, ASA class and case type. On regression analysis, neonates with CDH undergoing open procedures had a higher Odd’s Ratio of mortality, postoperative sepsis, perioperative transfusion, and prolonged length of stay compared to those undergoing MIS procedures, whereas for neonates with EA/TEF, there was no significant difference in any outcomes between the two treatment groups [Table].

Conclusion: Despite the operative challenges of thoracic MIS procedures in neonates, they appear to be safe and efficacious in selected cases, with comparable or superior 30-day outcomes to open procedures. However, given that NSQIP-P does not capture several clinically relevant data points related to surgical approach and outcomes, such as preoperative ECMO, EA/TEF gap length, and baseline disease severity, it is likely that some confounders were not included in our propensity score matched analysis. Further studies adjusting for these variables as well as surgeon and institutional volume may help in validating these findings.