Surgery for Obesity and Related Diseases ] (2014) 00–00
Case report
Transcatheter arterial vasopressin infusion for gastrojejunostomy hemorrhage after laparoscopic Roux-en-Y gastric bypass: a report of 3 cases Po-Chih Chang, M.D.a,b, Chih-Kun Huang, M.D.a,b,*, Kirubakaran Malapan, M.D.b a b
Keywords:
Division of General Surgery, Department of Surgery, E-DA Hospital/I-Shou University, Kaohsiung City, Taiwan Bariatric and Metabolic International Surgery Center, E-DA Hospital/I-Shou University, Kaohsiung City, Taiwan Received August 20, 2013; accepted September 12, 2013
Angiography; Endoscopy; Gastrojejunostomy hemorrhage; Laparoscopic Roux-en-Y gastric bypass; Vasopressin
Gastrojejunostomy (GJ) hemorrhage, occurring as a consequence of a marginal ulcer (MU) or staple-line bleeding, is not infrequent after laparoscopic Roux-en-Y gastric bypass (LRYGB). Most cases respond well to conservative treatment such as endoscopic management or acid reduction medication [1–3]. Surgery remains the primary management after failed conservative treatment, but little has been reported about the possibility of angiographic management [2,4]. Herein, we report 3 patients who presented with GJ hemorrhage, 2 hours, 3 hours, and 50 days after LRYGB, who were successfully treated with continuous transcatheter arterial vasopressin infusion. Case 1 A 51-year-old obese male with type II diabetes mellitus and hypertension with a body mass index (BMI) of 38.6 kg/m2 underwent LRYGB. The GJ anastomosis was constructed in an antecolic/antegastric fashion using a 35-mm linearstapled anastomosis (Ethicon Endo-Surgery, Cincinnati, OH). The anastomotic defect was closed in a single layer using intracorporeal, hand-sewn, absorbable sutures. The patient experienced persistent hematemesis 2 hours after surgery despite treatment with high-dose proton pump inhibitors (PPIs) and octreotide. He was administered a * Correspondence: Chih-Kun Huang, M.D., Division of General Surgery, Department of Surgery, Bariatric and Metabolic International Surgery Center, No. 1, Yi-Da Road, Jiao-Su Village, Yan-Chao Distinct, Kaohsiung City, 824, Taiwan. E-mail:
[email protected]
transfusion of 4 units of packed red blood cells (PRBCs). Esophagogastroduodenoscopy (EGD) showed fresh blood in the gastric pouch; however, the staple line could not be visualized (Fig. 1A). In spite of aggressive resuscitation, his condition deteriorated and his hemoglobin (Hgb) dropped to 8.2 g/dL. Emergent celiac angiography showed increased vascularity on the right wall of the gastric pouch without gross contrast leakage from either the left gastric artery (LGA) or superior mesenteric artery (SMA) branches (Fig. 1B). A Progreat microcatheter (Terumo, Tokyo, Japan) was placed into the branch of the LGA and vasopressin at the rate of 2.8 IU/hr was infused to achieve hemostasis. Within 4 hours, there was no further hematemesis, and the vasopressin infusion was gradually tapered and terminated after 24 hours. EGD on the fifth day showed a MU with no active bleeding (Fig. 1C). The patient was discharged on the sixth day and was treated with oral PPIs for 3 months. An EGD 1 year later showed no ulcer at the GJ anastomotic site (Fig. 1D).
Case 2 A 26-year-old male with a BMI of 50.4 kg/m2 underwent an uneventful antecolic/antegastric LRYGB. A 25-mL gastric pouch was created, and GJ anastomosis was constructed using a 35-mm linear stapler (Covidien, Mansfield, MA). The patient recovered and was discharged on postoperative day 2 taking a liquid diet. He presented 7 weeks later with complaints of persistent black stools, epigastric pain, and intermittent hematemesis after taking nonsteroidal
1550-7289/13/$ – see front matter r 2014 American Society for Metabolic and Bariatric Surgery. All rights reserved. http://dx.doi.org/10.1016/j.soard.2013.09.007
2
P.-C. Chang et al. / Surgery for Obesity and Related Diseases ] (2014) 00–00
Fig. 1. Case 1: (A) Upper endoscopy showed blood in the gastric pouch preventing further intervention. (B) Angiography via the left gastric artery (LGA) showed increased vascularity, but no contrast extravasation. Vasopressin infusion into LGA was performed through a microcatheter. (C) On the fifth postoperative day, upper endoscopy found no active bleeding. A marginal ulcer was seen. (D) One year later, normal appearance of the gastric pouch and gastrojejunostomy was noted on upper endoscopy.
antiinflammatory drugs for gouty arthritis. The patient was hypotensive (blood pressure: 98/64 mm Hg), tachycardic (heart rate: 104 beats/min), and had a distended abdomen without rebound tenderness. His Hgb was 10.4 g/dL. After fluid resuscitation and blood transfusion, he underwent a diagnostic EGD, which showed a MU with an active bleeder at the GJ anastomosis (Fig. 2A). Epinephrine injection was attempted to achieve hemostasis, but failed. Subsequently, the patient underwent emergent celiac angiography with LGA superselective catheterization (Fig. 2B). A Progreat microcatheter (Terumo) was placed in the LGA, and no contrast extravasation was noted. A continuous vasopressin infusion was begun at a rate of 8 IU/hr, and gradually tapered as the patient improved hemodynamically. Concurrently, intravenous PPI treatment was initiated. There was no further evidence of hemorrhage, and he was begun on a liquid diet 3 days later (Fig. 2C). He was discharged from the hospital on the fifth postoperative day in good condition.
Case 3 A 32-year-old female with a BMI of 46.4 kg/m2 underwent an antecolic/antegastric LRYGB with creation of a 30-mL gastric pouch and GJ anastomosis using a 35-mm linear stapler (Covidien). Approximately 2 hours after surgery, she vomited 1300 mL of fresh blood. High-dose PPI treatment began, and she was given 4 units PRBCs and 4 units fresh frozen plasma; however, her Hgb dropped to 9.4 g/dL. Celiac angiography instead of EGD was performed, because there was a high suspicion of GJ bleeding, and to avoid possible anastomotic perforation during an endoscopic procedure. No obvious contrast leak was noted in either the LGA or SMA branches (Fig. 2D). A Progreat microcatheter (Terumo) was placed in the LGA and vasopressin was infused at a rate of 8 IU/hr. Two hours later, there was no further hematemesis, and the patient remained hemodynamically stable. She was discharged uneventfully, with a prescription for a PPI for 3 months.
Vasopressin for Gastrojejunostomy Hemorrhage / Surgery for Obesity and Related Diseases ] (2014) 00–00
3
Fig. 2. Case 2: (A) Upper endoscopy found blood filling within the gastric pouch. (B) Angiography via the left gastric artery showed no contrast extravasation. A microcatheter was placed for vasopressin infusion. (C) Three days later upper endoscopy found no bleeding from the marginal ulcer. Case 3: (D) Superselective angiography of the left gastric artery showed hypervascularity.
Discussion Upper gastrointestinal hemorrhage affects .4 to 3.2% of patients after LRYGB, and most bleeding occurs at the GJ anastomosis or gastric pouch in the early postoperative period [1,2]. EGD is the gold standard for identification of the bleeding source, and most cases can be successfully managed with an intravenous PPI and fluid resuscitation [1–3]. However, patients with active bleeding at the GJ anastomosis usually require endoscopic intervention with epinephrine injection, clipping, or heater probe cautery, followed by acid reduction medication. These maneuvers can usually achieve hemostasis [1,3]. Although infrequently required, surgical reexploration is the definitive option for cases of failed conservative and endoscopic treatment [2,4]. Very little is known about treatment modalities other than surgery for overt GJ hemorrhage after failed conservative treatment. Angiographic interventions with embolization or vasopressin infusion therapy are potential alternative therapies in cases of gastrointestinal hemorrhage [5,6]. Though
vasopressin infusion therapy for gastrointestinal hemorrhage has a relatively high incidence of rebleeding, the 3 patients presented herein demonstrated complete resolution of GJ hemorrhage without complications after vasopressin infusion therapy [6]. Although no contrast extravasation was found during superselective angiography in our patients, we still placed a microcatheter for continuous vasopressin infusion, instead of embolization, for the overt GJ hemorrhage. The gastric pouch created receives a limited blood supply from the first and/or second branches of the LGA. We hypothesized that the GJ hemorrhage could be successfully controlled by the vasoconstrictor effect of a direct arterial infusion of vasopressin. Furthermore, catheter-induced vasospasm also may have contributed to achieving hemostasis by temporary occlusion of the vessel during catheterization of the LGA [5]. Transcatheter vasopressin infusion is less invasive than surgical reexploration. Repetitive endoscopy for GJ hemorrhage may result in perforation of the newly-created GJ anastomosis [3,7]. Moreover, repetitive surgery will pose a
P.-C. Chang et al. / Surgery for Obesity and Related Diseases ] (2014) 00–00
4 Table 1 Summary of patient data Age (yr)/Sex//BMI (kg/m2)
Time to hemorrhage after LRYGB
Hgb level before LRYGB/after hemorrhage (g/dL)
Blood transfusion
Vasopressin infusion dosage/duration
Co-morbidities
Case 1
51/M/38.6
2 hours
15.5/8.2
140 units/24 hr
Case 2
26/M/50.38
50 days
14.0/10.7
PRBCs 8 U FFP 6 U PRBCs 8 U
Case 3
32/F/46.4
3 hours
15.2/9.4
PRBCs 8 U FFP 4 U
100 units/16 hr
T2 DM HTN GERD NASH Gout NASH
200 units/24 hr
BMI ¼ body mass index; F ¼ female; FFP ¼ fresh frozen plasma; GERD ¼ gastroesophageal reflux disease; Hgb ¼ hemoglobin; HTN ¼ hypertension; LRYGB ¼ laparoscopic Roux-en-Y gastric bypass; M ¼ male; NASH ¼ nonalcoholic steatohepatitis; PRBCs ¼ packed red blood cells; T2 DM ¼ type 2 diabetes mellitus.
risk to the viability of the GJ anastomosis, leading to possible catastrophic anastomotic leak or gastric pouch necrosis in the early postoperative period. In addition, the costs of surgical reexploration are considerably more than those of other less invasive treatments. Considering these issues, we adopted transcatheter arterial vasopressin infusion as a therapeutic option after failed conservative treatment in these patients with early GJ hemorrhage after LRYGB. Transcatheter arterial vasopressin infusion was effective even in the patient with a MU and massive hemorrhage that was unsuccessfully managed with endoscopy (Case 2). Because of the novelty of this technique, the definitive dose of arterial vasopressin infusion is unknown. It has been suggested that the initial dosage for controlling gastrointestinal hemorrhage should be 12 IU/hr, and then titrated according to the clinical response. When an arterial vasopressin infusion is administered, potential complications such as angina, myocardial infarction, or mesenteric ischemia should be monitored closely [5,6]. In the 3 patients presented herein with overt GJ hemorrhage, the initial dosage of vasopressin infused ranged between 2.8 and 8 IU/hr, and was gradually tapered (Table 1). There is also a risk of GJ necrosis using this technique. Fortunately, no vasopressin-related complications occurred during or after infusion therapy. We postulate that the LGA blood supply to the gastric pouch, which lacks collateral circulation, could have contributed to the effectiveness of a relatively low dosage of vasopressin infusion, thereby resulting in an uneventful course. In conclusion, GJ hemorrhage after LRYGB is a lethal condition and requires urgent intervention. Intravenous PPI with concomitant transcatheter arterial continuous vasopressin infusion may be an alternative treatment for patients
who have failed conservative and endoscopic treatment before surgical intervention becomes necessary. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. Acknowledgments The authors wish to acknowledge I-Chang Lin, M.D. and Po-Lin Sun, M.D., for assistance in the angiographic intervention. References [1] Jamil LH, Krause KR, Chengelis DL, et al. Endoscopic management of early upper gastrointestinal hemorrhage after laparoscopic Roux-en-Y gastric bypass. Am J Gastroenterol 2008;103:86–91. [2] Rabl C, Peeva S, Prado K, et al. Early and late abdominal bleeding after Roux-en-Y gastric bypass: Sources and tailored therapeutic strategies. Obes Surg 2011;21:413–20. [3] Gill RS, Whitlock KA, Mohamed R, Sarkhosh K, Birch DW, Karmali S. The role of upper gastrointestinal endoscopy in treating postoperative complications in bariatric surgery. J Interv Gastroenterol 2012;2:37–41. [4] Nguyen NT, Longoria M, Chalifoux S, Wilson SE. Gastrointestinal hemorrhage after laparoscopic gastric bypass. Obes Surg 2004;14: 1308–12. [5] Cherian MP, Mehta P, Kalyanpur TM, Hedgire SS, Narsinghpura KS. Arterial interventions in gastrointestinal bleeding. Semin Intervent Radiol 2009;26:184–96. [6] Walker TG, Salazar GM, Waltman AC. Angiographic evaluation and management of acute gastrointestinal hemorrhage. World J Gastroenterol 2012;18:1191–201. [7] Tang SJ, Rivas H, Tang L, Lara LF, Sreenarasimhaiah J, Rockey DC. Endoscopic hemostasis using endoclip in early gastrointestinal hemorrhage after gastric bypass surgery. Obes Surg 2007;17:1261–7.
