Comparison of general endotracheal anesthesia versus sedation without endotracheal intubation during initial percutaneous endoscopic gastrostomy insertion for infants: A retrospective cohort study

Background: Critical airway incidents are a major cause of morbidity and mortality during anesthesia. Delayed management of airway obstruction quickly leads to severe complications due to the reduced apnea tolerance in infants and neonates. The decision of whether to intubate the trachea during anesthesia is therefore of great importance, particularly as an increasing number of procedures are performed outside of the operating room. Aim: In this retrospective cohort study, we evaluated airway management for infants below 6 months of age undergoing percutaneous endoscopic gastrostomy insertion. We compared demographic, procedural, and health outcome-related data for infants undergoing percutaneous endoscopic gastrostomy insertion under general endotracheal anesthesia (n = 105) to those receiving monitored anesthesia care (n = 44) without endotracheal intubation. Methods: A retrospective chart review was completed for all infants <6 months of age who underwent percutaneous endoscopic gastrostomy insertion in our institution's endoscopy suite between January 2002 and January 2017. Descriptive statistics summarized numeric variables using medians and corresponding ranges (minimum–maximum), and categorical variables using frequencies and percentages. Differences in study outcomes between patients undergoing general anesthesia or monitored anesthesia care were evaluated with univariate quantile or Firth logistic regression for numerical and categorical outcomes, respectively. Results are presented as β [95% confidence interval] or odds ratio [95% confidence interval] along with corresponding p-values. Results: Both groups were similar in distribution of age, race, and gender. However, patients selected for general anesthesia had lower median body weights (3.9 kg [range: 2.0–6.7] vs. 4.4 kg [range: 2.6–6.9]), higher percentages of cardiac (95.2% vs. 84.1%), and/or neurologic comorbidities (74.3% vs. 56.8%) and were more frequently given American Society of Anesthesiologists level IV classifications (41.9% vs. 29.6%) indicating that these infants may have had more severe disease than patients selected for monitored anesthesia care. Three monitored-anesthesia-care patients required intraoperative conversion to general anesthesia. General anesthesia patients experienced greater odds of intraoperative hypoxemia (45.2% vs. 29.0%; odds ratio: 2.0 [0.9–4.3], p-value:.09) and required postoperative airway intervention more frequently than monitored-anesthesia-care patients (13.03% vs. 2.3%; odds ratio: 4.6 [0.8–25.6], p-value:.08). Procedure times were identical in both groups (6 min), but general anesthesia resulted in longer median anesthesia times (44 min [range: 22–292] vs. 12 min [range:19–136]; β:13 [95% 6.9–19.1], p-value: <.001). Conclusion: Study results suggest that providers selected general anesthesia over monitored anesthesia care for infants and neonates with low body weights, cardiac comorbidities, and neurologic comorbidities. Increased rates of airway intervention, and increased length of stay may be at least partially related to more severe patient comorbidity, as indicated by higher American Society of Anesthesiologists classifications. However, due to the exploratory nature of these analyses, further confirmatory studies are needed to evaluate the impact of airway selection during PEG on postoperative patient outcomes.