Defining an Anastomotic Stricture Following Oesophageal Atresia Repair

Area of Science:

• Pediatric Surgery
• Biostatistics
• Gastroenterology

Background:

• Babies with esophageal atresia often require multiple anastomotic dilatations post-repair.
• The number of dilatations is a discrete random variable.
• Characterizing this variable's probability distribution is crucial for understanding patient outcomes.

Purpose of the Study:

• To characterize the probability distribution of anastomotic dilatations in infants following esophageal atresia repair.
• To determine population-level probabilities of dilatation.
• To propose a statistical definition for anastomotic stricture.

Main Methods:

• Collected data on dilatations from a consecutive institutional series.
• Fitted the data to Poisson and negative binomial distributions using maximum likelihood estimators.
• Analyzed clinical variables including complexity of atresia, anastomotic leak, salvage surgery, and fundoplication.

Main Results:

• The data fit a negative binomial distribution (μˆ: 4.41, kˆ: 0.56) better than a Poisson distribution.
• The 0.9 quantile indicated 12 dilatations.
• 12 dilatations were significantly associated with salvage surgery, anastomosis resection, and fundoplication, supporting construct validity. The institutional stricture rate was 9%.

Conclusions:

• The negative binomial distribution accurately models the number of dilatations, enabling statistical inference for the broader population.
• A proposed definition of anastomotic stricture is the 0.9 quantile, corresponding to 12 dilatations.
• This framework allows for better understanding and management of anastomotic strictures in infants post-esophageal atresia repair.