Esophageal atresia--a rabbit model to study anastomotic healing and the use of tissue adhesive fibrin sealant

The problem of bringing together two relatively widely separated, small, and fragile ends of a sick newborn baby's atretic esophagus remains a formidable surgical task, wherein the incidence of anastomotic leakage ranges from 10% to 27%. Recently, a multicomponent tissue adhesive fibrin sealant (Tisseel) has been licensed in Canada and declared useful for sealing gastrointestinal (GI) tract anastomoses. To study whether Tisseel might decrease the leakage rate of esophageal anastomoses in neonatal esophageal atresia and perhaps limit stricture formation, a rabbit model of esophageal atresia was developed. Twenty New Zealand white rabbits weighing 2.8 to 3.7 kg underwent thoracotomy and resection of a segment of esophagus with end-to-end, interrupted silk-sutured anastomosis under tension, to mimic the conditions found in newborn esophageal atresia. Four died immediately following operation. Ten rabbits had their anastomosis sealed with Tisseel, six control animals did not. All animals consumed variable amounts of water and food, starting 24 hours after surgery. Survival averaged 10.5 days (range, 5 to 20 days). Eight animals (five experimental, three control) were evaluated by means of barium esophagograms 1 week postoperatively, and all except one control animal demonstrated radiologic evidence of anastomotic leakage. Autopsy specimens revealed gross leakage in nine animals (seven experimental, two control). However, histology revealed leakage and periesophageal abscess formation in all experimental animals and in four control animals. The remaining two controls revealed only some degree of esophageal stenosis. This experiment showed no demonstrable benefit from the use of a fibrin sealant in preventing esophageal anastomotic leakage, such as that which occurs in repaired esophageal atresia.