The incidence of small-bowel varices is low. Bleeding from ectopic varices is reported in 1–5 % of patients with liver cirrhosis and in 20–30 % of patients with extrahepatic portal hypertension [1, 4]. Of bleeding ectopic varices, 17 % each were found in the duodenum, jejunum, or ileum, 14 % in the colon, 8 % in the rectum, and 9 % in the peritoneum [5]. Roughly half of patients with ectopic variceal bleeding had a history of surgery [4]. In fact, 66 % of patients with small-bowel varices had undergone abdominal or pelvic surgery, suggesting that postoperative adhesions between an incision in the abdominal wall and the small bowel may result in the formation of collaterals drained by the systemic venous circulation [1]. In our patient, the mesenteric varices were attributable to portal hypertension and postoperative adhesions.
Therapeutic options for bleeding ectopic varices include local treatment for the varices and portal decompression. Local interventional radiology includes percutaneous transhepatic obliteration (PTO) and BRTO. Portal decompressive treatment includes surgical dissection and transjugular intrahepatic portosystemic shunt (TIPS). L’Hermine et al. [6] reported that PTO controlled bleeding from gastroesophageal varices in 83 % of their patients. However, the formation of new inflow routes may result in rebleeding after PTO [5, 7]. According to Ninoi et al. [8], BRTO more effectively controls gastric variceal bleeding than TIPS or PTO.
Ono et al. [9] reported the successful management of mesenteric varices with BRTO performed via an abdominal wall collateral vein; they accessed the vein percutaneously under real-time ultrasound guidance. Ikeda et al. [10] addressed mesenteric varices with RTO via the abdominal wall. They accessed the vein after surgical incision.
We identified a remarkably tortuous small draining vein functioning as a collateral vein from the mesenteric varices to an abdominal wall vein on contrast-enhanced CT images and in the venous phase of superior mesenteric arteriography. Therefore, we adopted RTO instead of BRTO by inserting an 18-gauge plastic needle in an abdominal wall vein after surgical incision.
Various sclerosants and tissue adhesives have been employed for the management of duodenal variceal bleeding. Hirota et al. [11] used an EOI mixture or endoscopic injection sclerotherapy at BRTO. EOI agglutinates platelets and destroys vascular endothelial cells; consequently, it functions as a sclerotic agent. To obtain sclerosis of mesenteric varices, the sclerosing agent must remain in situ for at least 1 h [11]. Ikeda et al. [10] reported that manual pressure applied to the abdominal wall keeps the injected EOI in the varices for 1 h. NBCA, a tissue adhesive that rapidly polymerizes upon contact with blood and embolizes the varices, has been used to achieve hemostasis in patients with gastrointestinal variceal bleeding [12]. In our patient, we observed remarkable extravasation into the small bowel as we performed retrograde venography through an abdominal wall vein by manually injecting the contrast medium. Therefore, we used the sclerosant with NBCA to obtain rapid embolization of the varices. This achieved a favorable outcome.
In the short-term follow-up period, our patient’s esophageal varices worsened and he underwent endoscopic variceal ligation 6 months after the initial procedure. He did not develop hemorrhages from ectopic varices, although the obliteration of a collateral vein can induce the development of other varices. Therefore, we will continue careful follow-up for the long-term.
In summary, we report a case of bleeding mesenteric varices controlled by RTO using NBCA delivered via an abdominal wall vein and found that this technique can be selected for treatment of postoperative mesenteric varices.