Histopathologic and immunohistochemical sequelae of bariatric embolization in a porcine model

Abstract

Purpose: To evaluate the histopathologic sequelae of bariatric embolization on the gastric mucosa and to correlate with immunohistochemical evaluation of the gastric fundus, antrum, and duodenum.

Materials and methods: This study was performed on 12 swine stomach and duodenum specimens after necropsy. Of the 12 swine, 6 had previously undergone bariatric embolization of the gastric fundus, and the 6 control swine had undergone a sham procedure with saline. Gross pathologic, histopathologic, and immunohistochemical examinations of the stomach and duodenum were performed. Specifically, mucosal integrity, fibrosis, ghrelin-expressing cells, and gastrin-expressing cells were assessed.

Results: Gross and histopathologic evaluation of treatment animals showed healing or healed mucosal ulcers in 50% of animals, with gastritis in 100% of treatment animals and in five of six control animals. The ghrelin-immunoreactive mean cell density was significantly lower in the gastric fundus in the treated animals compared with control animals (15.3 vs 22.0, P < .01) but similar in the gastric antrum (9.3 vs 14.3, P = .08) and duodenum (8.5 vs 8.6, P = .89). The gastrin-expressing cell density was significantly lower in the antrum of treated animals compared with control animals (82.2 vs 126.4, P = .03). A trend toward increased fibrosis was suggested in the gastric fundus of treated animals compared with controls (P = .07).

Conclusions: Bariatric embolization resulted in a significant reduction in ghrelin-expressing cells in the gastric fundus without evidence of upregulation of ghrelin-expressing cells in the duodenum. Healing ulcerations in half of treated animals underscores the need for additional refinement of this procedure.

Figure 1

Selected images from the bariatric embolization procedure. (a) Initial celiac arteriogram shows the gastric fundus arterial supply, which include analogues of the main left gastric artery (long arrow), accessory left gastric artery (arrowhead), a gastric fundal branch arising from the splenic artery (short arrow), and a right gastric artery arising from the left hepatic artery (curved arrow). (b) Follow-up celiac arteriogram after superselective embolization was performed on all four fundal branches showing stasis of flow.

Figure 2

Section of gastric fundus (hematoxylin-eosin, ×100). (a) Control animal with preservation of the mucosa (M) and no evidence of ischemic damage. The submucosa (SM; bracket) and muscularis propria (MP) are normal. (b) Treatment animal showing well-preserved mucosa with ischemic damage as evidenced by dense confluent fibrosis involving the muscularis propria and submucosa (bracket) with fibrotic strands extending through the mucosa (arrowheads). Two microspheres (arrows) are present in the mucosa. (Available in color online at www.jvir.org.)

Figure 3

Immunohistochemical staining of the gastric fundus for ghrelin-expressing cells (×100). (a) Control animal shows multiple dark foci that represent ghrelin positivity. (b) Treatment animal shows decreased numbers of ghrelin-expressing cells. (Available in color online at www.jvir.org.)

Figure 4

Immunohistochemical staining of the gastric fundi (×100) and duodena confirms a significant reduction in the ghrelin-immunoreactive cell density in the gastric fundus after embolization. There is no compensatory upregulation of ghrelin-expressing cells in the duodena of treated animals.

Figure 5

Immunohistochemical staining of the gastric fundi for gastrin-immunoreactive cell density of the gastric antrum. (a) Control animal shows numerous punctate dark foci that represent stained gastrin. (b) Treated animal shows a substantial reduction in the gastrin-immunoreactive cell density. (Available in color online at www.jvir.org.)

Figure 6

Trichrome-stained sections from the gastric fundus show a trend toward an increase in fibrosis in the gastric fundus (6.58% ± 6.5 sham vs 18.34% ± 13.2 treatment, P = .07) primarily infiltrating the crypts.