Evaluation of a biofragmentable anastomosis ring for small intestinal anastomosis in ponies.

Abstract

Six ponies divided into two groups of three were used in a double crossover study design. Group 1 ponies had a small intestinal resection and anastomosis performed using a biofragmentable anastomosis ring (BAR); group 2 ponies had a hand-sewn small intestinal resection and anastomosis using a Gambee suture pattern. Approximately 30 days later, all ponies had a second celiotomy and anastomosed segments were removed. Group 1 ponies had a hand-sewn anastomosis performed and group 2 had a BAR. The anastomotic sites were collected at necropsy approximately 30 days later. Anastomosed intestinal segments were evaluated with ultrasound to determine lumen diameter, area, circumference, and wall thickness. Gross descriptions of adhesions were recorded and sections of the anastomotic site were taken for histological evaluation. Time to perform the BAR anastomosis was significantly less (P = .0004) than for the hand-sewn Gambee anastomosis. Ponies with handsewn anastomoses had no signs of colic, whereas five of six ponies with BAR anastomoses had several episodes of abdominal discomfort, between day 16 to 18, corresponding to the time of BAR disintegration as determined by abdominal radiographs. Hand-sewn anastomoses had a tendency to have more adhesion formation than BAR anastomoses, but all anastomoses, except one BAR anastomosis, were graded as having a low obstructive potential. BAR anastomoses had a significantly larger mean index of stenosis for intraluminal diameter (76% +/- 13.6), area (93.7% +/- 6.01) and circumference (75.8% +/- 14.0) than the hand-sewn anastomoses (26.6% +/- 11.9; 44.6% +/- 19.5; 26.8% +/- 12.7). The BAR anastomoses also had a significantly smaller mean intraluminal diameter (0.96 cm +/- 0.49), area (0.838 cm2 +/- 0.65) and circumference (3.28 cm +/- 1.63) than the hand sewn anastomoses (3.11 cm +/- 0.73; 7.99 cm2 +/- 3.9; 10.3 cm +/- 2.47). In addition, the BAR anastomoses had a significantly larger (P = .0069) bowel wall thickness at the anastomoses and a significantly larger (P = .047) wall thickness proximal to the anastomosis than the hand-sewn anastomoses, indicating some degree of hypertrophy because of chronic obstruction. No significant difference was found in the diameter, area, or circumference between bowel proximal and distal to the anastomosis for either the BAR or Gambee techniques, or between the BAR and Gambee anastomosis as a measure of chronic obstruction. There was a significantly higher (P = .0043) histological score (worse healing) for mucosal healing and continuity for the BAR, as well as a tendency to score higher for inflammation, anastomotic alignment, and anastomotic fibrosis. The BAR technique had a significantly higher (P = .0043) total histological score than the Gambee technique. Although the BAR was advantageous in many respects, results of this study suggest that it should not be used for equine small intestinal anastomosis because of the potential for stricture formation.