- Email: [email protected]
Transcatheter Arterial Embolization with N-Butyl Cyanoacrylate for Acute Life-threatening Gastroduodenal Bleeding Uncontrolled by Endoscopic Hemostasis
CLINICAL STUDY
Transcatheter Arterial Embolization with N-Butyl Cyanoacrylate for Acute Life-threatening Gastroduodenal Bleeding Uncontrolled by Endoscopic Hemostasis Hiroyuki Morishita, PhD, MD, Takuji Yamagami, PhD, MD, Tomohiro Matsumoto, PhD, MD, Shunsuke Asai, MD, Koji Masui, MD, Hideki Sato, PhD, MD, Atsushi Majima, MD, and Osamu Sato, PhD, MD
ABSTRACT Purpose: To investigate the feasibility, efficacy, and safety of transcatheter arterial embolization with N-butyl cyanoacrylate (NBCA) for gastroduodenal nonvariceal bleeding uncontrolled by endoscopic hemostasis. Materials and Methods: Between January 2006 and December 2011, a total of 317 patients underwent emergent endoscopic therapy for nonvariceal gastroduodenal bleeding, but hemostasis was not achieved in 20 cases. Emergent surgery was performed immediately following endoscopy in two patients. Arteriography was performed in the remaining 18 patients, and embolization with NBCA was performed in 15 patients (mean age, 71.3 y) in whom the bleeding site was detected on arteriography. For embolization, NBCA was mixed with iodized oil at a ratio of 1:1.5–1:4, and no other embolic material was used in the procedure. Technical and clinical success rates, recurrent bleeding, procedural time, complications, and clinical outcomes were determined for each procedure. Results: Embolization with NBCA was technically and clinically successful in all procedures, without major complications. No patient receiving embolization with NBCA experienced recurrent bleeding or required further treatment after the one-session procedure. All patients were discharged after clinical improvement. The time between puncture of the femoral artery and completion of embolization ranged from 25 to 240 minutes (mean, 66 min), and the time between the microcatheter reaching the ultimate catheter location selected for embolization and hemostasis ranged from 142 to 550 seconds (mean, 322 s). Conclusions: In this limited series, embolization with NBCA was found to be a safe, feasible, and effective treatment for gastroduodenal arterial bleeding when endoscopic hemostasis had failed.
ABBREVIATIONS GDA = gastroduodenal artery, NBCA = N-butyl cyanoacrylate, PVA = polyvinyl alcohol
Vascular embolization is widely considered as the first-line alternative treatment for severe gastroduodenal bleeding
From the Departments of Diagnostic Radiology, (H.M., T.M., S.A., O.S.) and Gastroenterology (H.S., A.M.), Japan Red Cross Kyoto Daiichi Hospital; and Department of Radiology (T.Y., K.M.), Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan. Received September 25, 2012; final revision received December 17, 2012; accepted December 18, 2012. Address correspondence to H.M., Department of Diagnostic Radiology, Japanese Red Cross Kyoto Daiichi Hospital, 15-749 Honmachi, Higashiyama, Kyoto 605-0981, Japan; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2013 J Vasc Interv Radiol 2013; 24:432–438 http://dx.doi.org/10.1016/j.jvir.2012.12.017
refractory to endoscopic treatment (1,2). According to many reports, technical and clinical success rates of endovascular treatment are high, ranging from 69% to 100% and from 63% to 97% (2), respectively. The embolic agents described in the literature are various, and the choice of the best embolic agent remains a matter of debate (2). Of course, coils are considered the first-choice embolic material, as supported by a great deal of experience with embolization with coils. However, coil embolization is not always successful, depending on anatomy, the size of arterial lesions, technical ability, and coagulopathy related to the administration of anticoagulant or antiplatelet drugs and disseminated intravascular coagulation (3). In addition, failure of coil embolization results in a lifethreatening situation, especially when the patient is in critical condition (3).
Volume 24
’
Number 3
’
March
’
2013
Loffroy et al (4) reported that the use of coils alone was significantly associated with early recurrent bleeding. To achieve hemostasis in a critical situation, it is favorable to select the embolic material that can embolize the bleeding vessel completely and permanently in a short time. N-butyl cyanoacrylate (NBCA; Histoacryl; B. Braun, Tuttlingen, Germany) is compatible with these requirements: (i) NBCA can embolize bleeding vessels not occluded with coils (3); (ii) NBCA is a liquid and permanent embolic material (3); and (iii) the polymerization time for NBCA/iodized oil mixtures at ratios of 1:1–1:4 is 1–4 seconds (5). Indeed, NBCA, a rapidly hardening liquid adhesive agent, has been used as an effective polymerizing liquid embolic agent for gastroduodenal artery (GDA) hemorrhage (6–8), but its applications have been limited because of technical difficulties and the risk of organ ischemia (2). Our purpose in the present study was to investigate if transcatheter arterial embolization with NBCA alone is a feasible, effective, and safe method for the treatment of nonvariceal gastroduodenal bleeding uncontrolled by endoscopic hemostasis.
MATERIALS AND METHODS This was a retrospective study of patients treated for nonvariceal gastroduodenal bleeding treated by transcatheter arterial embolization with NBCA. All procedures were performed according to the ethical standards of the World Medical Association Declaration of Helsinki. Our institutional review board does not require approval for this type of retrospective study. Written informed consent was obtained from all patients.
Patients Between January 2006 and December 2011, a total of 317 emergent endoscopic therapies were performed for patients with life-threatening nonvariceal gastroduodenal bleeding (209 men and 108 women; mean age, 67.6 y) at our institution. Emergent endoscopic therapy was attempted in all cases, but endoscopic hemostasis was not achieved in a total of 20 cases. Emergent surgery was performed immediately following endoscopy in two patients who had experienced cardiopulmonary arrest during endoscopy. Soon after endoscopic hemostasis failed, arteriography was performed for the remaining 18 cases, and embolization with NBCA was attempted in 15 cases in which the bleeding site was detected on arteriography. These included 12 men and three women, with a mean age of 71.3 years. The characteristics of each case are shown in Table 1. Before endoscopy, all patients were in unstable hemodynamic condition. The mean hemoglobin level before the procedure was 8.0 g/dL (range, 5.3–10.9 g/ dL), and 13 patients received blood transfusions (mean, 6.3 U packed red blood cells; range, 2–12 U). In the three cases in which the bleeding site was not identified on arteriography, empiric embolization with coils was attempted in
433
one case and no interventional therapy was attempted in the other two. Incidentally, the patient treated with coil embolization experienced recurrent bleeding 170 days after the procedure, and hemostasis was finally achieved by embolization with NBCA (patient 12; Table 1). The underlying diseases were gastric ulcer (n ¼ 7), duodenal ulcer (n ¼ 5), gastric cancer (n ¼ 1), duodenal tumor (n ¼ 1), and pseudoaneurysm of the pancreaticoduodenal artery (n ¼ 1). The embolized arteries were the left gastric artery (n ¼ 6; gastric branch, n ¼ 5; main branch, n ¼ 1), duodenal branch of anterior superior pancreaticoduodenal artery (n ¼ 2), duodenal branch of inferior pancreaticoduodenal artery (n ¼ 2), duodenal branch of posterior superior pancreaticoduodenal artery (n ¼ 1), duodenal branch of posterior inferior pancreaticoduodenal artery (n ¼ 1), gastric branch of right gastric artery (n ¼ 1), main branch of splenic artery (n ¼ 1), and main branch of GDA (n ¼ 1). Ten cases had additional coagulopathic factors that further complicated the achievement of hemostasis: administration of antiplatelet drug (n ¼ 5; aspirin, n ¼ 4; cilostazol, n ¼ 1), administration of anticoagulant and antiplatelet drugs (n ¼ 2, including one patient undergoing dialysis; warfarin potassium, n ¼ 2; sarpogrelate HCl, n ¼ 1; and ticlopidine HCl, n ¼ 1), disseminated intravascular coagulation (n ¼ 2), and dialysis alone (n ¼ 1).
Procedure In our daily clinical work as interventional radiologists in an emergency unit, NBCA has been used for many years to embolize bleeding that is uncontrollable with coil embolization. There might be advantages to the use of NBCA because it can achieve embolization more quickly than coils (9), and NBCA can be used in cases complicated by coagulopathy because NBCA polymerization is not affected by the coagulation process (10). Indeed, all the patients described in the present study were in a critical situation, and 10 of 15 showed characteristics of coagulopathy. The use of NBCA was approved by a multidisciplinary conference involving emergency physicians, surgeons, endoscopists, and interventional radiologists, and patients or their families agreed to treatment with NBCA embolization. The procedure was performed under local anesthesia via a right femoral approach. First, a 5-F shepherd hook– shaped catheter (Terumo Clinical Supply, Gifu, Japan) was inserted, and diagnostic celiac and superior mesenteric arteriography was performed in all 18 cases. In five cases, extravasation of contrast medium was identified on celiac arteriography (n ¼ 3) or superior mesenteric angiography (n ¼ 2). Then, arteriography via the splanchnic arteries was performed after the coaxial advancement of a 2.2-F microcatheter (Progreat b; Terumo Clinical Supply). In 15 cases, including the five cases mentioned earlier, the extravasation of contrast medium or a pseudoaneurysm was identified near the clip, which was used for endoscopic
434
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
Morishita et al
’
JVIR
Table 1 . Characteristics of Patients in whom Endoscopic Hemostasis Was Not Achieved Underlying Pt. No./Sex/Age (y)
Risk Factor
Disease
Hb (g/dL)/Hct (%)
Platelets (/lL)
INR
PRBCs (U)
for Hemorrhage
Gastric ulcer
6.6/20.5
65,000
1.8
12
Warfarin potassium, sarpogrelate HCl, dialysis
2/M/52
Gastric ulcer
9.4/31.2
187,000
1.1
8
Aspirin
3/M/57 4/F/83
Duodenal ulcer Gastric ulcer
8.9/26.6 5.3/17.9
139,000 112,000
1.3 1.2
8 8
– Aspirin
5/M/68
Gastric cancer
9.5/28.2
146,000
1
4
Cilostazol
6/M/84 7/M/68
Gastric ulcer Gastric ulcer
7.9/24.0 8.5/25.3
145,000 75,000
1 1.3
6 10
– Aspirin
8/M/78
Duodenal tumor
9.8/29.4
125,000
3.8
8
Warfarin potassium,
9/F/89
Duodenal ulcer
7.3/22.2
70,000
1.3
4
ticlopidine HCl DIC
10/M/97
Duodenal ulcer
8.1/25.3
129,000
1.2
NA
–
11/M/58 12/M/45
Gastric ulcer PDA pseudoaneurysm
10.4/31.4 5.3/15.8
327,000 202,000
1 1
NA 4
– Dialysis
13/F/77
Duodenal ulcer
6.1/17.8
40,000
1.3
6
DIC
14/M/62 15/M/83
Gastric ulcer Duodenal ulcer
10.9/33.1 6.3/19.1
218,000 248,000
0.9 1.2
2 2
– Aspirin
1/M/68
DIC ¼ disseminated intravascular coagulation, Hb ¼ hemoglobin, Hct ¼ hematocrit, INR ¼ International Normalized Ratio, NA ¼ not attempted, PDA ¼ pancreaticoduodenal artery, PRBC ¼ packed red blood cell.
hemostasis. In the remaining three cases, the bleeding site was not detected at all. We performed embolization with NBCA for all cases with an identified bleeding site. The microcatheter was advanced as close as possible to the bleeding site to wedge into the bleeding artery and to restrict blood flow. NBCA was mixed with Lipiodol (Laboratoire Andre Guerbet, Aulnay-sous-Bois, France) at a ratio of 1:1.5–1:4 (ie, NBCA to Lipiodol). NBCA and Lipiodol concentrations used were based on a report by Pollak and White (5). Before the injection of NBCA, the catheter lumen was flushed with a 5% glucose solution to avoid premature polymerization. NBCA was injected under continuous fluoroscopic monitoring. The endpoint of NBCA injection was extravasation of NBCA from the bleeding site and/or filling of the pseudoaneurysm and the parent artery with NBCA. The microcatheter was pulled back immediately after the injection to avoid inadvertent adherence to the vascular wall. Then, the inner lumen of the angiographic catheter was aspirated, and postembolic angiography was performed to confirm hemostasis. The first follow-up endoscopy was attempted in all cases within 24 hours after the procedure. After embolization, all patients were hospitalized in an intensive care unit or high care unit, and continuous vital sign monitoring, complete blood count, and checking of clinical status were carried out to detect any signs of recurrent bleeding. A second follow-up endoscopy and complete blood counts were carried out to confirm hemostasis. We evaluated the technical and clinical success rate, incidence of recurrent bleeding, complications associated with embolization, clinical outcomes, and time required to perform embolization, and achievement of hemostasis. Technical success was defined by the confirmation of
hemostasis on arteriography at the end of the procedure and endoscopy performed within 24 hours after the procedure. Recurrent bleeding was defined by identification of bleeding on endoscopy or significant decrease of hemoglobin levels. Clinical success was defined as a clinical improvement and no need for additional invasive treatment until complete hemostasis was confirmed by the second follow-up endoscopy and complete blood counts. Time required for the achievement of hemostasis was defined as the duration between the puncture of the femoral artery and the end of embolization. Time required to perform embolization was defined as the duration between the microcatheter reaching the ultimate catheter location selected for embolization and the end of the injection with NBCA. To check these times, we reviewed the time record in the nursing chart and the angiographic images on our picture archiving and communication system.
RESULTS Embolization with NBCA was performed in a single session in all patients (Table 2). Embolization with NBCA was successful in all procedures (NBCA/Lipiodol volume range, 0.3–1 mL; mean, 0.53 mL). In all cases, the feeding artery and/or pseudoaneurysm was embolized completely with a single injection (Figs 1, 2). No patient experienced recurrent bleeding or required further treatment after the embolization. The second follow-up endoscopy performed after embolization (mean, 14.1 d; range, 4–25 d) showed the achievement of hemostasis, no significant ischemic changes in the embolized areas, and
Volume 24
’
Number 3
’
March
’
2013
435
Table 2 . Results of Embolization with NBCA for Upper Gastrointestinal Bleeding Embolized
NBCA:Lipiodol
NBCA/Lipiodol
Time Required for
Time Required for
Pt. No.
Artery
Ratio
Volume (mL)
Hemostasis (min)
Embolization (s)
1 2
RGA gastric branch LGA gastric branch
1:1.5 1:2
1 0.5
80 25
313 378
3
IPDA duodenal branch
1:3
1
70
235
4 5
SPA main branch LGA branch
1:3 1:2
0.5 0.3
70 45
160 548
6
LGA branch
1:4
0.3
30
477
7 8
LGA IPDA duodenal branch
1:5 1:4
0.4 1
35 90
65 550
9
ASPDA duodenal branch
1:1.5
0.3
35
142
10 11
ASPDA duodenal branch LGA branch
1:2 1:1.5
0.3 0.3
60 25
343 152
12
PIPDA duodenal branch
1:3
0.5
240
332
13 14
PSPDA duodenal branch LGA branch
1:4 1:3
0.5 1
65 60
222 500
15
GDA trunk
1:1.5
0.7
60
415
ASPDA ¼ anterior superior pancreaticoduodenal artery, GDA ¼ gastroduodenal artery, IPDA ¼ inferior pancreaticoduodenal artery, LGA ¼ left gastric artery, NBCA ¼ N-butyl cyanoacrylate, PIPDA ¼ posterior inferior pancreaticoduodenal artery, PSPDA ¼ posterior superior pancreaticoduodenal artery, RGA ¼ right gastric artery, SPA ¼ splenic artery.
Figure 1. Embolization in an 83-year-old man with bleeding from a duodenal ulcer at endoscopy. (a) Selective arteriogram of the GDA (black arrow) before embolization shows no evidence of active bleeding. The GDA near the metallic clips (white arrows) used for endoscopic hemostasis is narrowing. (b) Incidentally, the tip of the microcatheter (white arrow) reached into the duodenum through the perforation of the GDA (black arrows), and extravasation of contrast medium is seen in the lumen of the descending duodenum (white arrowheads). (c) After pulling back the tip of the microcatheter near the perforated hole, NBCA/Lipiodol mixture (1:1.5) was injected. NBCA embolized the proximal and distal sides to the bleeding site in the GDA in a one-session procedure (black arrows). (d) Selective arteriogram of the inferior pancreaticoduodenal artery (white arrow) shows that retrograde arterial flow stops on the distal side of the bleeding site of the GDA (black arrow). No ischemic complications were reported.
no worsening of endoscopic findings in all patients compared with the findings before embolization. Hemoglobin levels stabilized after the procedure. Complete hemostasis was achieved technically and clinically in all 15 patients, including 10 with risk factor(s) for hemorrhage. Therefore, the technical and clinical success rates were each 100% (15 of 15). There were no major complications related to embolization (ie, an unplanned increase in the level of care, prolonged hospitalization, permanent adverse sequelae, or death). However, all patients had minor complications of epigastric pain, nausea, and occasionally vomiting—the same as before endoscopy. With the administration of proton pump inhibitors or H2 blockers, their symptoms were relieved in a few days. To improve anemia, 10 of 15
patients received blood transfusion (mean, 2.6 U packed red blood cells; range, 2–4 U) within 1 week after the procedure. All cases showed clinical improvement without recurrent bleeding, and all patients were discharged from the hospital. Two patients (patients 3 and 5; Table 1) died of associated malignancy (pancreatic cancer and gastric cancer) 98 and 183 days after the procedure, respectively. Time required for the achievement of hemostasis ranged from 25 to 240 minutes (mean, 66 min), and time required to perform embolization ranged from 142 to 550 seconds (mean, 322 s). In patient 12, in whom empiric embolization with microcoils was attempted, a long time (240 min) was required to achieve hemostasis compared with the other 14 cases because the microcoils interfered with the access to the parent artery of the bleeding site.
436
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
Morishita et al
’
JVIR
Figure 2. Bleeding gastric ulcer in a 52-year-old man. (a) Selective arteriogram of the left gastric artery (LGA, black arrow) shows no evidence of active bleeding, but a gastric branch (white arrow) is stopped near the metallic clips (arrowhead) used for hemostasis. (b) It was very difficult to cannulate the microcatheter into the gastric branch because of its tortuosity (tip of the microcatheter, black arrow). Cannulation was successful, but the branch showed spasm. (c) The tip of a microcatheter (black arrow) has been advanced close to the metallic clips, and an arteriogram shows extravasation of contrast medium (white arrows) into the gastric lumen. (d) NBCA/Lipiodol mixture (1:2, 0.5 mL) was injected upon pulling out the microcatheter to fill the distal portion of the bleeding branch (black arrows). Small collateral vessels are also embolized (white arrows).
DISCUSSION As a first-line treatment for gastroduodenal bleeding, endoscopic therapy has achieved favorable bleeding control in as many as 98% of cases (11). However, when endoscopic hemostasis is not achieved, surgery or endovascular embolization is proposed as a second-line treatment. Surgery is associated with operative mortality rates as high as 20%– 40% (1), especially in high-risk cases. On the contrary, endovascular embolization has been performed for at least three decades, and has been shown to be effective at controlling hemorrhage and decreasing mortality (12–16). According to many reports, technical and clinical success rates of endovascular treatment are high, ranging from 69% to 100% and from 63% to 97% (2), respectively. Therefore, currently, endovascular embolization is considered as a firstline alternative for life-threatening gastroduodenal bleeding refractory to endoscopic treatment (14–20). Although the choice of embolic agent is an important predictor influencing clinical outcome, the choice of the best embolic agent remains a matter of debate (2,21). In most series, the choice of embolic agent depended on the discretion of each interventional radiologist, according to individual experience, material availability, angiographic findings, and ability to perform superselective catheterization of the bleeding vessel (2). The embolic agents described in the published literature are various: coils, NBCA, gelatin sponge, or polyvinyl alcohol (PVA). Coils are a popular and easy-to-use embolization material. The success of coil-only embolization in the GDA or pancreaticoduodenal arteries has been reported (22–24), but coil embolization is not always successful, and the use of coils alone has been reported to be significantly correlated with early recurrent bleeding (25). Gelatin sponge is absorbed a few weeks after embolization, so there still remains a risk of recanalization. In addition, Encarnacion et al (14) reported a low success rate of embolization chiefly with gelatin sponge alone (62%). Loffroy et al (26) reported a
high procedural success rate of embolization with coils and the combination of gelatin sponge and/or cyanoacrylate (95%). Aina et al (17) compared embolization with coils alone versus embolization with coils combined with PVA particles or Gelfoam. They reported that, by multivariate regression analysis, the use of coils alone was associated with recurrent bleeding in cases with coagulopathy, a finding that supports the use of PVA or Gelfoam in combination with coils in this subgroup. NBCA, a quick-hardening and permanent liquid adhesive material, can be injected through a catheter with a narrow lumen because of its low viscosity (27). Dotter et al (28) first reported the treatment of gastrointestinal bleeding with the use of cyanoacrylate as an adhesive material. Kish et al (29) reported a technical success rate of 75% with NBCA embolization for acute arterial hemorrhage. Yamakado et al (3) reported that embolization with NBCA for a ruptured pseudoaneurysm was technically successful in all nine of their cases, and that NBCA can be used to embolize bleeding vessels not occluded with coils. In their study (3), the technical and secondary clinical success rates were 100%, which are favorable compared with previous reports. NBCA has been used for GDA hemorrhage, and some favorable reports of this approach have been published (6–8). Toyoda et al (6) reported a technical success rate of 100% (N ¼ 7) and clinical success rate of 86% (six of seven), Jae et al (7) reported that the technical success rate was 100% (32 of 32) and clinical success rate was 91% (29 of 32), and Lee et al (8) reported that the technical success rate was 94% (15 of 16) and clinical success rate was 88% (14 of 16). Our results are comparable to or better than these results. In addition, these investigators (7,8) reported that embolization with NBCA achieved rapid, complete embolization to control massive bleeding from gastric or duodenal ulcers. However, in their studies, to avoid excessive and undesirable embolization or reflux of NBCA, combination with coil embolization is described as sometimes being necessary to slow blood flow and induce proximal occlusion of NBCA
Volume 24
’
Number 3
’
March
’
2013
(7,8). In the present study, combined NBCA/coil embolization was not necessary because we succeeded in advancing the tip of the microcatheter as close as possible to the bleeding site to wedge into the bleeding artery and restrict blood flow effectively and because we chose adequate concentrations of NBCA and Lipiodol in each case; these factors may result from our experience in embolization with NBCA. Of course, coil embolization before injection of NBCA is acceptable when it is not possible to wedge the microcatheter effectively or if the interventional radiologist is not accustomed to the use of NBCA (3,6–8). It is important to note that NBCA should not be used without careful consideration, and considerable experience is required to handle NBCA not only because it is difficult, but because excessive embolization or reflux of NBCA may cause undesirable complications (6–8). To achieve embolization with NBCA safely, there are various recommendations to keep in mind: (i) Pay careful attention to the specific vascular anatomy and the information obtained from test injections with contrast medium; (ii) use the wedged catheter technique with no pericatheter flow, which allows precise control of the extent of glue embolization; and (iii) perform a simulated injection with contrast medium in the same condition before the actual NBCA injection, with precise volume calculation (7). We believe NBCA should be used by experienced interventional radiologists with knowledge of the behavior of, and contraindications to, NBCA, and those who have never used NBCA should be trained and supervised by experienced interventional radiologists. NBCA polymerization is not affected by the coagulation process, and NBCA can be used in cases complicated by coagulopathy (7,10). Encarnacion et al (14) first identified that the presence or absence of coagulopathy is a major factor that affects embolization outcome. A strong correlation has been noted among coagulopathy, clinical failure, and mortality after embolization: patients with an impaired coagulation profile are three times more likely to experience recurrent bleeding after initially successful embolization and 10 times more likely to die as a result of bleeding compared with those with normal coagulation status (13,17,25). Indeed, cases in which endovascular hemostasis was not achieved often showed risks, eg, administration of an antiplatelet or anticoagulant drug or disseminated intravascular coagulation. In all patients in the present series—even though 10 of the 15 had coagulopathy—hemostasis was achieved with NBCA alone, and there were no cases of recurrent bleeding. We believe NBCA is the best embolic material for cases with a risk of hemorrhage. In the case of life-threatening upper gastrointestinal hemorrhage requiring urgent hemostasis, it is very important to achieve hemostasis as rapidly as possible. Surgery in the present cases would have been much more invasive and time-consuming, and poses a significantly higher risk of mortality and morbidity (1). Yonemitsu et al (9) reported that transcatheter arterial embolization with NBCA was
437
completed more quickly than transcatheter arterial embolization with microcoils or gelatin sponge particles. Toyoda et al (6) reported that transcatheter arterial embolization with NBCA with or without coils achieved hemostasis more rapidly than transcatheter arterial embolization with coils and/or Gelfoam in gastric or duodenal ulcer hemorrhage despite the difficulty of catheterization (which varied with the size and shape of the vessel) affecting the time needed to complete the procedure. They reported that the time required to complete embolization with NBCA in patients with bleeding gastric ulcer ranged from 45 to 115 minutes (mean, 72.5 min), and that for bleeding duodenal ulcer ranged from 50 to 70 minutes (mean, 61.7 min). In the present study, the time required to achieve hemostasis ranged from 25 to 240 minutes (mean, 66 min), which is comparable to their report. Most of the procedure time in the present study was spent identifying and approaching the bleeding site, and embolization was achieved with only a single injection in a short time (142–550 s; mean, 322 s). Ichiro et al (30) reported that the time of embolization in cases of unsuccessful initial endoscopic treatment or recurrent bleeding after endoscopic treatment ranged from 36 to 124 minutes (mean, 68 min), which does not seem significantly different than the procedure time in the present study. However, because empirical embolization with microcoils was attempted in more than half of their cases (30), we suppose that their time required to perform embolization was longer than ours, and their time spent identifying the bleeding site was shorter. Their technical success rate was 100%, but they reported an incidence of recurrent bleeding of 14%. If the same amount of time is needed, we believe it is desirable to spend time identifying the bleeding site to embolize rather than performing empiric embolization. NBCA may be considered as a primary embolic agent when superselective catheterization is not possible and standard microcoil embolization appears too difficult for technical or anatomic reasons. In particular, in the case of acute life-threatening upper gastroduodenal bleeding, we consider NBCA embolization suitable to achieve rapid hemostasis. There are several limitations to the present study, the first being its retrospective nature. Second, the number of cases is small because of the low incidence of the situation described. Third, NBCA alone was used as embolic material, and considerable experience is required to handle NBCA and achieve optimal results. In conclusion, embolization with NBCA is a safe, effective, and feasible treatment for life-threatening massive gastroduodenal bleeding when endoscopic hemostasis has failed. Further experience and investigation are required to define ideal indications and technical conditions.
REFERENCES 1. Cheynel N, Peschaud F, Hagry O, Rat P, Ognois-Ausset P, Favre JP. Bleeding gastroduodenal ulcer: results of surgical management. Ann Chir 2001:232–235.
438
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
2. Loffroy R, Rao P, Ota S, De Lin M, Kwak BK, Geschwind JF. Embolization of acute nonvariceal upper gastrointestinal hemorrhage resistant to endoscopic treatment: results and predictors of recurrent bleeding. Cardiovasc Intervent Radiol 2010; 33:1088–1100. 3. Yamakado K, Nakatsuka A, Tanaka N, Takano K, Matsumura K, Takeda K. Transcatheter arterial embolization of ruptured pseudoaneurysms with coils and N-butyl cyanoacrylate. J Vasc Interv Radiol 2000; 11:66–72. 4. Loffroy R, Guiu B, D’Athis P, et al. Arterial embolotherapy for endoscopically unmanageable acute gastroduodenal hemorrhage: predictors of early rebleeding. Clin Gastroenterol Hepatol 2009; 7:515–523. 5. Pollak JS, White RI Jr. The use of cyanoacrylate adhesives in peripheral embolization. J Vasc Interv Radiol 2001; 12:907–913. 6. Toyoda H, Nakano S, Kumada T, et al. Estimation of usefulness of N-butyl-2-cyanoacrylate-lipiodol mixture in transcatheter arterial embolization for urgent control of life-threatening massive bleeding from gastric or duodenal ulcer. J Gastroenterol Hepatol 1996; 11:252–258. 7. Jae HJ, Chung JW, Jung AY, Lee W, Park JH. Transcatheter arterial embolization of nonvariceal upper gastrointestinal bleeding with N-butyl cyanoacrylate. Korean J Radiol 2007; 8:48–56. 8. Lee CW, Liu KL, Wang HP, Chen SJ, Tsang YM, Liu HM. Transcatheter arterial embolization of acute upper gastrointestinal tract bleeding with N-butyl-2-cyanoacrylate. J Vasc Interv Radiol 2007; 18:209–216. 9. Yonemitsu T, Kawai N, Sato M, et al. Evaluation of transcatheter arterial embolization with gelatin sponge particles, microcoils, and N-butyl cyanoacrylate for acute arterial bleeding in a coagulopathic condition. J Vasc Interv Radiol 2009; 20:1176–1187. 10. Yonemitsu T, Kawai N, Sato M, et al. Comparison of hemostatic durability between N-butyl cyanoacrylate and gelatin sponge particles in transcatheter arterial embolization for acute arterial hemorrhage in a coagulopathic condition in a swine model. Cardiovasc Intervent Radiol 2010; 33:1192–1197. 11. Liou TC, Lin SC, Wang HY, Chang WH. Optimal injection volume of epinephrine for endoscopic treatment of peptic ulcer bleeding. World J Gastroenterol 2006; 12:3108–3113. 12. Rosch J, Dotter CT, Brown MJ. Selective arterial embolization. A new method for control of acute gastrointestinal bleeding. Radiology 1972; 102:303–306. 13. Funaki B. Endovascular intervention for the treatment of acute arterial gastrointestinal hemorrhage. Gastroenterol Clin North Am 2002; 31:701–713. 14. Encarnacion CE, Kadir S, Beam CA, Payne CS. Gastrointestinal bleeding: treatment with gastrointestinal arterial embolization. Radiology 1992; 183:505–508. 15. Ljungdahl M, Eriksson LG, Nyman R, Gustavsson S. Arterial embolisation in management of massive bleeding from gastric and duodenal ulcers. Eur J Surg 2002; 168:384–390. 16. Holme JB, Nielsen DT, Funch-Jensen P, Mortensen FV. Transcatheter arterial embolization in patients with bleeding duodenal ulcer: an alternative to surgery. Acta Radiol 2006; 47:244–247.
Morishita et al
’
JVIR
17. Aina R, Oliva VL, Therasse E, et al. Arterial embolotherapy for upper gastrointestinal hemorrhage: outcome assessment. J Vasc Interv Radiol 2001; 12:195–200. 18. Loffroy R, Guiu B, Cercueil JP, et al. Refractory bleeding from gastroduodenal ulcers: arterial embolization in high-operative-risk patients. J Clin Gastroenterol 2008; 42:361–367. 19. Poultsides GA, Kim CJ, Orlando R III, Peros G, Hallisey MJ, Vignati PV. Angiographic embolization for gastroduodenal hemorrhage: safety, efficacy, and predictors of outcome. Arch Surg 2008; 143: 457–461. 20. Larssen L, Moger T, Bjornbeth BA, Lygren I, Klow NE. Transcatheter arterial embolization in the management of bleeding duodenal ulcers: a 5.5-year retrospective study of treatment and outcome. Scand J Gastroenterol 2008; 43:217–222. 21. Loffroy R, Guiu B. Role of transcatheter arterial embolization for massive bleeding from gastroduodenal ulcers. World J Gastroenterol 2009; 15:5889–5897. 22. Toyoda H, Nakano S, Takeda I, et al. Transcatheter arterial embolization for massive bleeding from duodenal ulcers not controlled by endoscopic hemostasis. Endoscopy 1995; 27:304–307. 23. BK R, van Munster IP, de Jager CP, Rutten MJ. Embolization as treatment of choice for bleeding peptic ulcers in high-risk patients. Dig Surg 2009; 26:37–42. 24. Ledermann HP, Schoch E, Jost R, Decurtins M, Zollikofer CL. Superselective coil embolization in acute gastrointestinal hemorrhage: personal experience in 10 patients and review of the literature. J Vasc Interv Radiol 1998; 9:753–760. 25. Loffroy R, Guiu B, D’Athis P, et al. Arterial embolotherapy for endoscopically unmanageable acute gastroduodenal hemorrhage: predictors of early rebleeding. Clin Gastroenterol Hepatol 2009; 7: 515–523. 26. Loffroy R, Guiu B, Mezzetta L, et al. Short- and long-term results of transcatheter embolization for massive arterial hemorrhage from gastroduodenal ulcers not controlled by endoscopic hemostasis. Can J Gastroenterol 2009; 23:115–120. 27. Yamagami T, Kato T, Iida S, Tanaka O, Nishimura T. Value of transcatheter arterial embolization with coils and N-butyl cyanoacrylate for long-term hepatic arterial infusion chemotherapy. Radiology 2004; 230:792–802. 28. Dotter CT, Goldman ML, Rosch J. Instant selective arterial occlusion with isobutyl 2-cyanoacrylate. Radiology 1975; 114:227–230. 29. Kish JW, Katz MD, Marx MV, Harrell DS, Hanks SE. N-butyl cyanoacrylate embolization for control of acute arterial hemorrhage. J Vasc Interv Radiol 2004; 15:689–695. 30. Ichiro I, Shushi H, Akihiko I, Yasuhiko I, Yasuyuki Y. Empiric transcatheter arterial embolization for massive bleeding from duodenal ulcers: efficacy and complications. J Vasc Interv Radiol 2011; 22:911–916.
Transcatheter Arterial Embolization with N-Butyl Cyanoacrylate for Acute Life-threatening Gastroduodenal Bleeding Uncontrolled by Endoscopic Hemostasis Hiroyuki Morishita, PhD, MD, Takuji Yamagami, PhD, MD, Tomohiro Matsumoto, PhD, MD, Shunsuke Asai, MD, Koji Masui, MD, Hideki Sato, PhD, MD, Atsushi Majima, MD, and Osamu Sato, PhD, MD
ABSTRACT Purpose: To investigate the feasibility, efficacy, and safety of transcatheter arterial embolization with N-butyl cyanoacrylate (NBCA) for gastroduodenal nonvariceal bleeding uncontrolled by endoscopic hemostasis. Materials and Methods: Between January 2006 and December 2011, a total of 317 patients underwent emergent endoscopic therapy for nonvariceal gastroduodenal bleeding, but hemostasis was not achieved in 20 cases. Emergent surgery was performed immediately following endoscopy in two patients. Arteriography was performed in the remaining 18 patients, and embolization with NBCA was performed in 15 patients (mean age, 71.3 y) in whom the bleeding site was detected on arteriography. For embolization, NBCA was mixed with iodized oil at a ratio of 1:1.5–1:4, and no other embolic material was used in the procedure. Technical and clinical success rates, recurrent bleeding, procedural time, complications, and clinical outcomes were determined for each procedure. Results: Embolization with NBCA was technically and clinically successful in all procedures, without major complications. No patient receiving embolization with NBCA experienced recurrent bleeding or required further treatment after the one-session procedure. All patients were discharged after clinical improvement. The time between puncture of the femoral artery and completion of embolization ranged from 25 to 240 minutes (mean, 66 min), and the time between the microcatheter reaching the ultimate catheter location selected for embolization and hemostasis ranged from 142 to 550 seconds (mean, 322 s). Conclusions: In this limited series, embolization with NBCA was found to be a safe, feasible, and effective treatment for gastroduodenal arterial bleeding when endoscopic hemostasis had failed.
ABBREVIATIONS GDA = gastroduodenal artery, NBCA = N-butyl cyanoacrylate, PVA = polyvinyl alcohol
Vascular embolization is widely considered as the first-line alternative treatment for severe gastroduodenal bleeding
From the Departments of Diagnostic Radiology, (H.M., T.M., S.A., O.S.) and Gastroenterology (H.S., A.M.), Japan Red Cross Kyoto Daiichi Hospital; and Department of Radiology (T.Y., K.M.), Graduate School of Medical Science, Kyoto Prefectural University of Medicine, Kyoto, Japan. Received September 25, 2012; final revision received December 17, 2012; accepted December 18, 2012. Address correspondence to H.M., Department of Diagnostic Radiology, Japanese Red Cross Kyoto Daiichi Hospital, 15-749 Honmachi, Higashiyama, Kyoto 605-0981, Japan; E-mail: [email protected] None of the authors have identified a conflict of interest. & SIR, 2013 J Vasc Interv Radiol 2013; 24:432–438 http://dx.doi.org/10.1016/j.jvir.2012.12.017
refractory to endoscopic treatment (1,2). According to many reports, technical and clinical success rates of endovascular treatment are high, ranging from 69% to 100% and from 63% to 97% (2), respectively. The embolic agents described in the literature are various, and the choice of the best embolic agent remains a matter of debate (2). Of course, coils are considered the first-choice embolic material, as supported by a great deal of experience with embolization with coils. However, coil embolization is not always successful, depending on anatomy, the size of arterial lesions, technical ability, and coagulopathy related to the administration of anticoagulant or antiplatelet drugs and disseminated intravascular coagulation (3). In addition, failure of coil embolization results in a lifethreatening situation, especially when the patient is in critical condition (3).
Volume 24
’
Number 3
’
March
’
2013
Loffroy et al (4) reported that the use of coils alone was significantly associated with early recurrent bleeding. To achieve hemostasis in a critical situation, it is favorable to select the embolic material that can embolize the bleeding vessel completely and permanently in a short time. N-butyl cyanoacrylate (NBCA; Histoacryl; B. Braun, Tuttlingen, Germany) is compatible with these requirements: (i) NBCA can embolize bleeding vessels not occluded with coils (3); (ii) NBCA is a liquid and permanent embolic material (3); and (iii) the polymerization time for NBCA/iodized oil mixtures at ratios of 1:1–1:4 is 1–4 seconds (5). Indeed, NBCA, a rapidly hardening liquid adhesive agent, has been used as an effective polymerizing liquid embolic agent for gastroduodenal artery (GDA) hemorrhage (6–8), but its applications have been limited because of technical difficulties and the risk of organ ischemia (2). Our purpose in the present study was to investigate if transcatheter arterial embolization with NBCA alone is a feasible, effective, and safe method for the treatment of nonvariceal gastroduodenal bleeding uncontrolled by endoscopic hemostasis.
MATERIALS AND METHODS This was a retrospective study of patients treated for nonvariceal gastroduodenal bleeding treated by transcatheter arterial embolization with NBCA. All procedures were performed according to the ethical standards of the World Medical Association Declaration of Helsinki. Our institutional review board does not require approval for this type of retrospective study. Written informed consent was obtained from all patients.
Patients Between January 2006 and December 2011, a total of 317 emergent endoscopic therapies were performed for patients with life-threatening nonvariceal gastroduodenal bleeding (209 men and 108 women; mean age, 67.6 y) at our institution. Emergent endoscopic therapy was attempted in all cases, but endoscopic hemostasis was not achieved in a total of 20 cases. Emergent surgery was performed immediately following endoscopy in two patients who had experienced cardiopulmonary arrest during endoscopy. Soon after endoscopic hemostasis failed, arteriography was performed for the remaining 18 cases, and embolization with NBCA was attempted in 15 cases in which the bleeding site was detected on arteriography. These included 12 men and three women, with a mean age of 71.3 years. The characteristics of each case are shown in Table 1. Before endoscopy, all patients were in unstable hemodynamic condition. The mean hemoglobin level before the procedure was 8.0 g/dL (range, 5.3–10.9 g/ dL), and 13 patients received blood transfusions (mean, 6.3 U packed red blood cells; range, 2–12 U). In the three cases in which the bleeding site was not identified on arteriography, empiric embolization with coils was attempted in
433
one case and no interventional therapy was attempted in the other two. Incidentally, the patient treated with coil embolization experienced recurrent bleeding 170 days after the procedure, and hemostasis was finally achieved by embolization with NBCA (patient 12; Table 1). The underlying diseases were gastric ulcer (n ¼ 7), duodenal ulcer (n ¼ 5), gastric cancer (n ¼ 1), duodenal tumor (n ¼ 1), and pseudoaneurysm of the pancreaticoduodenal artery (n ¼ 1). The embolized arteries were the left gastric artery (n ¼ 6; gastric branch, n ¼ 5; main branch, n ¼ 1), duodenal branch of anterior superior pancreaticoduodenal artery (n ¼ 2), duodenal branch of inferior pancreaticoduodenal artery (n ¼ 2), duodenal branch of posterior superior pancreaticoduodenal artery (n ¼ 1), duodenal branch of posterior inferior pancreaticoduodenal artery (n ¼ 1), gastric branch of right gastric artery (n ¼ 1), main branch of splenic artery (n ¼ 1), and main branch of GDA (n ¼ 1). Ten cases had additional coagulopathic factors that further complicated the achievement of hemostasis: administration of antiplatelet drug (n ¼ 5; aspirin, n ¼ 4; cilostazol, n ¼ 1), administration of anticoagulant and antiplatelet drugs (n ¼ 2, including one patient undergoing dialysis; warfarin potassium, n ¼ 2; sarpogrelate HCl, n ¼ 1; and ticlopidine HCl, n ¼ 1), disseminated intravascular coagulation (n ¼ 2), and dialysis alone (n ¼ 1).
Procedure In our daily clinical work as interventional radiologists in an emergency unit, NBCA has been used for many years to embolize bleeding that is uncontrollable with coil embolization. There might be advantages to the use of NBCA because it can achieve embolization more quickly than coils (9), and NBCA can be used in cases complicated by coagulopathy because NBCA polymerization is not affected by the coagulation process (10). Indeed, all the patients described in the present study were in a critical situation, and 10 of 15 showed characteristics of coagulopathy. The use of NBCA was approved by a multidisciplinary conference involving emergency physicians, surgeons, endoscopists, and interventional radiologists, and patients or their families agreed to treatment with NBCA embolization. The procedure was performed under local anesthesia via a right femoral approach. First, a 5-F shepherd hook– shaped catheter (Terumo Clinical Supply, Gifu, Japan) was inserted, and diagnostic celiac and superior mesenteric arteriography was performed in all 18 cases. In five cases, extravasation of contrast medium was identified on celiac arteriography (n ¼ 3) or superior mesenteric angiography (n ¼ 2). Then, arteriography via the splanchnic arteries was performed after the coaxial advancement of a 2.2-F microcatheter (Progreat b; Terumo Clinical Supply). In 15 cases, including the five cases mentioned earlier, the extravasation of contrast medium or a pseudoaneurysm was identified near the clip, which was used for endoscopic
434
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
Morishita et al
’
JVIR
Table 1 . Characteristics of Patients in whom Endoscopic Hemostasis Was Not Achieved Underlying Pt. No./Sex/Age (y)
Risk Factor
Disease
Hb (g/dL)/Hct (%)
Platelets (/lL)
INR
PRBCs (U)
for Hemorrhage
Gastric ulcer
6.6/20.5
65,000
1.8
12
Warfarin potassium, sarpogrelate HCl, dialysis
2/M/52
Gastric ulcer
9.4/31.2
187,000
1.1
8
Aspirin
3/M/57 4/F/83
Duodenal ulcer Gastric ulcer
8.9/26.6 5.3/17.9
139,000 112,000
1.3 1.2
8 8
– Aspirin
5/M/68
Gastric cancer
9.5/28.2
146,000
1
4
Cilostazol
6/M/84 7/M/68
Gastric ulcer Gastric ulcer
7.9/24.0 8.5/25.3
145,000 75,000
1 1.3
6 10
– Aspirin
8/M/78
Duodenal tumor
9.8/29.4
125,000
3.8
8
Warfarin potassium,
9/F/89
Duodenal ulcer
7.3/22.2
70,000
1.3
4
ticlopidine HCl DIC
10/M/97
Duodenal ulcer
8.1/25.3
129,000
1.2
NA
–
11/M/58 12/M/45
Gastric ulcer PDA pseudoaneurysm
10.4/31.4 5.3/15.8
327,000 202,000
1 1
NA 4
– Dialysis
13/F/77
Duodenal ulcer
6.1/17.8
40,000
1.3
6
DIC
14/M/62 15/M/83
Gastric ulcer Duodenal ulcer
10.9/33.1 6.3/19.1
218,000 248,000
0.9 1.2
2 2
– Aspirin
1/M/68
DIC ¼ disseminated intravascular coagulation, Hb ¼ hemoglobin, Hct ¼ hematocrit, INR ¼ International Normalized Ratio, NA ¼ not attempted, PDA ¼ pancreaticoduodenal artery, PRBC ¼ packed red blood cell.
hemostasis. In the remaining three cases, the bleeding site was not detected at all. We performed embolization with NBCA for all cases with an identified bleeding site. The microcatheter was advanced as close as possible to the bleeding site to wedge into the bleeding artery and to restrict blood flow. NBCA was mixed with Lipiodol (Laboratoire Andre Guerbet, Aulnay-sous-Bois, France) at a ratio of 1:1.5–1:4 (ie, NBCA to Lipiodol). NBCA and Lipiodol concentrations used were based on a report by Pollak and White (5). Before the injection of NBCA, the catheter lumen was flushed with a 5% glucose solution to avoid premature polymerization. NBCA was injected under continuous fluoroscopic monitoring. The endpoint of NBCA injection was extravasation of NBCA from the bleeding site and/or filling of the pseudoaneurysm and the parent artery with NBCA. The microcatheter was pulled back immediately after the injection to avoid inadvertent adherence to the vascular wall. Then, the inner lumen of the angiographic catheter was aspirated, and postembolic angiography was performed to confirm hemostasis. The first follow-up endoscopy was attempted in all cases within 24 hours after the procedure. After embolization, all patients were hospitalized in an intensive care unit or high care unit, and continuous vital sign monitoring, complete blood count, and checking of clinical status were carried out to detect any signs of recurrent bleeding. A second follow-up endoscopy and complete blood counts were carried out to confirm hemostasis. We evaluated the technical and clinical success rate, incidence of recurrent bleeding, complications associated with embolization, clinical outcomes, and time required to perform embolization, and achievement of hemostasis. Technical success was defined by the confirmation of
hemostasis on arteriography at the end of the procedure and endoscopy performed within 24 hours after the procedure. Recurrent bleeding was defined by identification of bleeding on endoscopy or significant decrease of hemoglobin levels. Clinical success was defined as a clinical improvement and no need for additional invasive treatment until complete hemostasis was confirmed by the second follow-up endoscopy and complete blood counts. Time required for the achievement of hemostasis was defined as the duration between the puncture of the femoral artery and the end of embolization. Time required to perform embolization was defined as the duration between the microcatheter reaching the ultimate catheter location selected for embolization and the end of the injection with NBCA. To check these times, we reviewed the time record in the nursing chart and the angiographic images on our picture archiving and communication system.
RESULTS Embolization with NBCA was performed in a single session in all patients (Table 2). Embolization with NBCA was successful in all procedures (NBCA/Lipiodol volume range, 0.3–1 mL; mean, 0.53 mL). In all cases, the feeding artery and/or pseudoaneurysm was embolized completely with a single injection (Figs 1, 2). No patient experienced recurrent bleeding or required further treatment after the embolization. The second follow-up endoscopy performed after embolization (mean, 14.1 d; range, 4–25 d) showed the achievement of hemostasis, no significant ischemic changes in the embolized areas, and
Volume 24
’
Number 3
’
March
’
2013
435
Table 2 . Results of Embolization with NBCA for Upper Gastrointestinal Bleeding Embolized
NBCA:Lipiodol
NBCA/Lipiodol
Time Required for
Time Required for
Pt. No.
Artery
Ratio
Volume (mL)
Hemostasis (min)
Embolization (s)
1 2
RGA gastric branch LGA gastric branch
1:1.5 1:2
1 0.5
80 25
313 378
3
IPDA duodenal branch
1:3
1
70
235
4 5
SPA main branch LGA branch
1:3 1:2
0.5 0.3
70 45
160 548
6
LGA branch
1:4
0.3
30
477
7 8
LGA IPDA duodenal branch
1:5 1:4
0.4 1
35 90
65 550
9
ASPDA duodenal branch
1:1.5
0.3
35
142
10 11
ASPDA duodenal branch LGA branch
1:2 1:1.5
0.3 0.3
60 25
343 152
12
PIPDA duodenal branch
1:3
0.5
240
332
13 14
PSPDA duodenal branch LGA branch
1:4 1:3
0.5 1
65 60
222 500
15
GDA trunk
1:1.5
0.7
60
415
ASPDA ¼ anterior superior pancreaticoduodenal artery, GDA ¼ gastroduodenal artery, IPDA ¼ inferior pancreaticoduodenal artery, LGA ¼ left gastric artery, NBCA ¼ N-butyl cyanoacrylate, PIPDA ¼ posterior inferior pancreaticoduodenal artery, PSPDA ¼ posterior superior pancreaticoduodenal artery, RGA ¼ right gastric artery, SPA ¼ splenic artery.
Figure 1. Embolization in an 83-year-old man with bleeding from a duodenal ulcer at endoscopy. (a) Selective arteriogram of the GDA (black arrow) before embolization shows no evidence of active bleeding. The GDA near the metallic clips (white arrows) used for endoscopic hemostasis is narrowing. (b) Incidentally, the tip of the microcatheter (white arrow) reached into the duodenum through the perforation of the GDA (black arrows), and extravasation of contrast medium is seen in the lumen of the descending duodenum (white arrowheads). (c) After pulling back the tip of the microcatheter near the perforated hole, NBCA/Lipiodol mixture (1:1.5) was injected. NBCA embolized the proximal and distal sides to the bleeding site in the GDA in a one-session procedure (black arrows). (d) Selective arteriogram of the inferior pancreaticoduodenal artery (white arrow) shows that retrograde arterial flow stops on the distal side of the bleeding site of the GDA (black arrow). No ischemic complications were reported.
no worsening of endoscopic findings in all patients compared with the findings before embolization. Hemoglobin levels stabilized after the procedure. Complete hemostasis was achieved technically and clinically in all 15 patients, including 10 with risk factor(s) for hemorrhage. Therefore, the technical and clinical success rates were each 100% (15 of 15). There were no major complications related to embolization (ie, an unplanned increase in the level of care, prolonged hospitalization, permanent adverse sequelae, or death). However, all patients had minor complications of epigastric pain, nausea, and occasionally vomiting—the same as before endoscopy. With the administration of proton pump inhibitors or H2 blockers, their symptoms were relieved in a few days. To improve anemia, 10 of 15
patients received blood transfusion (mean, 2.6 U packed red blood cells; range, 2–4 U) within 1 week after the procedure. All cases showed clinical improvement without recurrent bleeding, and all patients were discharged from the hospital. Two patients (patients 3 and 5; Table 1) died of associated malignancy (pancreatic cancer and gastric cancer) 98 and 183 days after the procedure, respectively. Time required for the achievement of hemostasis ranged from 25 to 240 minutes (mean, 66 min), and time required to perform embolization ranged from 142 to 550 seconds (mean, 322 s). In patient 12, in whom empiric embolization with microcoils was attempted, a long time (240 min) was required to achieve hemostasis compared with the other 14 cases because the microcoils interfered with the access to the parent artery of the bleeding site.
436
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
Morishita et al
’
JVIR
Figure 2. Bleeding gastric ulcer in a 52-year-old man. (a) Selective arteriogram of the left gastric artery (LGA, black arrow) shows no evidence of active bleeding, but a gastric branch (white arrow) is stopped near the metallic clips (arrowhead) used for hemostasis. (b) It was very difficult to cannulate the microcatheter into the gastric branch because of its tortuosity (tip of the microcatheter, black arrow). Cannulation was successful, but the branch showed spasm. (c) The tip of a microcatheter (black arrow) has been advanced close to the metallic clips, and an arteriogram shows extravasation of contrast medium (white arrows) into the gastric lumen. (d) NBCA/Lipiodol mixture (1:2, 0.5 mL) was injected upon pulling out the microcatheter to fill the distal portion of the bleeding branch (black arrows). Small collateral vessels are also embolized (white arrows).
DISCUSSION As a first-line treatment for gastroduodenal bleeding, endoscopic therapy has achieved favorable bleeding control in as many as 98% of cases (11). However, when endoscopic hemostasis is not achieved, surgery or endovascular embolization is proposed as a second-line treatment. Surgery is associated with operative mortality rates as high as 20%– 40% (1), especially in high-risk cases. On the contrary, endovascular embolization has been performed for at least three decades, and has been shown to be effective at controlling hemorrhage and decreasing mortality (12–16). According to many reports, technical and clinical success rates of endovascular treatment are high, ranging from 69% to 100% and from 63% to 97% (2), respectively. Therefore, currently, endovascular embolization is considered as a firstline alternative for life-threatening gastroduodenal bleeding refractory to endoscopic treatment (14–20). Although the choice of embolic agent is an important predictor influencing clinical outcome, the choice of the best embolic agent remains a matter of debate (2,21). In most series, the choice of embolic agent depended on the discretion of each interventional radiologist, according to individual experience, material availability, angiographic findings, and ability to perform superselective catheterization of the bleeding vessel (2). The embolic agents described in the published literature are various: coils, NBCA, gelatin sponge, or polyvinyl alcohol (PVA). Coils are a popular and easy-to-use embolization material. The success of coil-only embolization in the GDA or pancreaticoduodenal arteries has been reported (22–24), but coil embolization is not always successful, and the use of coils alone has been reported to be significantly correlated with early recurrent bleeding (25). Gelatin sponge is absorbed a few weeks after embolization, so there still remains a risk of recanalization. In addition, Encarnacion et al (14) reported a low success rate of embolization chiefly with gelatin sponge alone (62%). Loffroy et al (26) reported a
high procedural success rate of embolization with coils and the combination of gelatin sponge and/or cyanoacrylate (95%). Aina et al (17) compared embolization with coils alone versus embolization with coils combined with PVA particles or Gelfoam. They reported that, by multivariate regression analysis, the use of coils alone was associated with recurrent bleeding in cases with coagulopathy, a finding that supports the use of PVA or Gelfoam in combination with coils in this subgroup. NBCA, a quick-hardening and permanent liquid adhesive material, can be injected through a catheter with a narrow lumen because of its low viscosity (27). Dotter et al (28) first reported the treatment of gastrointestinal bleeding with the use of cyanoacrylate as an adhesive material. Kish et al (29) reported a technical success rate of 75% with NBCA embolization for acute arterial hemorrhage. Yamakado et al (3) reported that embolization with NBCA for a ruptured pseudoaneurysm was technically successful in all nine of their cases, and that NBCA can be used to embolize bleeding vessels not occluded with coils. In their study (3), the technical and secondary clinical success rates were 100%, which are favorable compared with previous reports. NBCA has been used for GDA hemorrhage, and some favorable reports of this approach have been published (6–8). Toyoda et al (6) reported a technical success rate of 100% (N ¼ 7) and clinical success rate of 86% (six of seven), Jae et al (7) reported that the technical success rate was 100% (32 of 32) and clinical success rate was 91% (29 of 32), and Lee et al (8) reported that the technical success rate was 94% (15 of 16) and clinical success rate was 88% (14 of 16). Our results are comparable to or better than these results. In addition, these investigators (7,8) reported that embolization with NBCA achieved rapid, complete embolization to control massive bleeding from gastric or duodenal ulcers. However, in their studies, to avoid excessive and undesirable embolization or reflux of NBCA, combination with coil embolization is described as sometimes being necessary to slow blood flow and induce proximal occlusion of NBCA
Volume 24
’
Number 3
’
March
’
2013
(7,8). In the present study, combined NBCA/coil embolization was not necessary because we succeeded in advancing the tip of the microcatheter as close as possible to the bleeding site to wedge into the bleeding artery and restrict blood flow effectively and because we chose adequate concentrations of NBCA and Lipiodol in each case; these factors may result from our experience in embolization with NBCA. Of course, coil embolization before injection of NBCA is acceptable when it is not possible to wedge the microcatheter effectively or if the interventional radiologist is not accustomed to the use of NBCA (3,6–8). It is important to note that NBCA should not be used without careful consideration, and considerable experience is required to handle NBCA not only because it is difficult, but because excessive embolization or reflux of NBCA may cause undesirable complications (6–8). To achieve embolization with NBCA safely, there are various recommendations to keep in mind: (i) Pay careful attention to the specific vascular anatomy and the information obtained from test injections with contrast medium; (ii) use the wedged catheter technique with no pericatheter flow, which allows precise control of the extent of glue embolization; and (iii) perform a simulated injection with contrast medium in the same condition before the actual NBCA injection, with precise volume calculation (7). We believe NBCA should be used by experienced interventional radiologists with knowledge of the behavior of, and contraindications to, NBCA, and those who have never used NBCA should be trained and supervised by experienced interventional radiologists. NBCA polymerization is not affected by the coagulation process, and NBCA can be used in cases complicated by coagulopathy (7,10). Encarnacion et al (14) first identified that the presence or absence of coagulopathy is a major factor that affects embolization outcome. A strong correlation has been noted among coagulopathy, clinical failure, and mortality after embolization: patients with an impaired coagulation profile are three times more likely to experience recurrent bleeding after initially successful embolization and 10 times more likely to die as a result of bleeding compared with those with normal coagulation status (13,17,25). Indeed, cases in which endovascular hemostasis was not achieved often showed risks, eg, administration of an antiplatelet or anticoagulant drug or disseminated intravascular coagulation. In all patients in the present series—even though 10 of the 15 had coagulopathy—hemostasis was achieved with NBCA alone, and there were no cases of recurrent bleeding. We believe NBCA is the best embolic material for cases with a risk of hemorrhage. In the case of life-threatening upper gastrointestinal hemorrhage requiring urgent hemostasis, it is very important to achieve hemostasis as rapidly as possible. Surgery in the present cases would have been much more invasive and time-consuming, and poses a significantly higher risk of mortality and morbidity (1). Yonemitsu et al (9) reported that transcatheter arterial embolization with NBCA was
437
completed more quickly than transcatheter arterial embolization with microcoils or gelatin sponge particles. Toyoda et al (6) reported that transcatheter arterial embolization with NBCA with or without coils achieved hemostasis more rapidly than transcatheter arterial embolization with coils and/or Gelfoam in gastric or duodenal ulcer hemorrhage despite the difficulty of catheterization (which varied with the size and shape of the vessel) affecting the time needed to complete the procedure. They reported that the time required to complete embolization with NBCA in patients with bleeding gastric ulcer ranged from 45 to 115 minutes (mean, 72.5 min), and that for bleeding duodenal ulcer ranged from 50 to 70 minutes (mean, 61.7 min). In the present study, the time required to achieve hemostasis ranged from 25 to 240 minutes (mean, 66 min), which is comparable to their report. Most of the procedure time in the present study was spent identifying and approaching the bleeding site, and embolization was achieved with only a single injection in a short time (142–550 s; mean, 322 s). Ichiro et al (30) reported that the time of embolization in cases of unsuccessful initial endoscopic treatment or recurrent bleeding after endoscopic treatment ranged from 36 to 124 minutes (mean, 68 min), which does not seem significantly different than the procedure time in the present study. However, because empirical embolization with microcoils was attempted in more than half of their cases (30), we suppose that their time required to perform embolization was longer than ours, and their time spent identifying the bleeding site was shorter. Their technical success rate was 100%, but they reported an incidence of recurrent bleeding of 14%. If the same amount of time is needed, we believe it is desirable to spend time identifying the bleeding site to embolize rather than performing empiric embolization. NBCA may be considered as a primary embolic agent when superselective catheterization is not possible and standard microcoil embolization appears too difficult for technical or anatomic reasons. In particular, in the case of acute life-threatening upper gastroduodenal bleeding, we consider NBCA embolization suitable to achieve rapid hemostasis. There are several limitations to the present study, the first being its retrospective nature. Second, the number of cases is small because of the low incidence of the situation described. Third, NBCA alone was used as embolic material, and considerable experience is required to handle NBCA and achieve optimal results. In conclusion, embolization with NBCA is a safe, effective, and feasible treatment for life-threatening massive gastroduodenal bleeding when endoscopic hemostasis has failed. Further experience and investigation are required to define ideal indications and technical conditions.
REFERENCES 1. Cheynel N, Peschaud F, Hagry O, Rat P, Ognois-Ausset P, Favre JP. Bleeding gastroduodenal ulcer: results of surgical management. Ann Chir 2001:232–235.
438
’
NBCA Embolization for Life-threatening Gastroduodenal Bleeding
2. Loffroy R, Rao P, Ota S, De Lin M, Kwak BK, Geschwind JF. Embolization of acute nonvariceal upper gastrointestinal hemorrhage resistant to endoscopic treatment: results and predictors of recurrent bleeding. Cardiovasc Intervent Radiol 2010; 33:1088–1100. 3. Yamakado K, Nakatsuka A, Tanaka N, Takano K, Matsumura K, Takeda K. Transcatheter arterial embolization of ruptured pseudoaneurysms with coils and N-butyl cyanoacrylate. J Vasc Interv Radiol 2000; 11:66–72. 4. Loffroy R, Guiu B, D’Athis P, et al. Arterial embolotherapy for endoscopically unmanageable acute gastroduodenal hemorrhage: predictors of early rebleeding. Clin Gastroenterol Hepatol 2009; 7:515–523. 5. Pollak JS, White RI Jr. The use of cyanoacrylate adhesives in peripheral embolization. J Vasc Interv Radiol 2001; 12:907–913. 6. Toyoda H, Nakano S, Kumada T, et al. Estimation of usefulness of N-butyl-2-cyanoacrylate-lipiodol mixture in transcatheter arterial embolization for urgent control of life-threatening massive bleeding from gastric or duodenal ulcer. J Gastroenterol Hepatol 1996; 11:252–258. 7. Jae HJ, Chung JW, Jung AY, Lee W, Park JH. Transcatheter arterial embolization of nonvariceal upper gastrointestinal bleeding with N-butyl cyanoacrylate. Korean J Radiol 2007; 8:48–56. 8. Lee CW, Liu KL, Wang HP, Chen SJ, Tsang YM, Liu HM. Transcatheter arterial embolization of acute upper gastrointestinal tract bleeding with N-butyl-2-cyanoacrylate. J Vasc Interv Radiol 2007; 18:209–216. 9. Yonemitsu T, Kawai N, Sato M, et al. Evaluation of transcatheter arterial embolization with gelatin sponge particles, microcoils, and N-butyl cyanoacrylate for acute arterial bleeding in a coagulopathic condition. J Vasc Interv Radiol 2009; 20:1176–1187. 10. Yonemitsu T, Kawai N, Sato M, et al. Comparison of hemostatic durability between N-butyl cyanoacrylate and gelatin sponge particles in transcatheter arterial embolization for acute arterial hemorrhage in a coagulopathic condition in a swine model. Cardiovasc Intervent Radiol 2010; 33:1192–1197. 11. Liou TC, Lin SC, Wang HY, Chang WH. Optimal injection volume of epinephrine for endoscopic treatment of peptic ulcer bleeding. World J Gastroenterol 2006; 12:3108–3113. 12. Rosch J, Dotter CT, Brown MJ. Selective arterial embolization. A new method for control of acute gastrointestinal bleeding. Radiology 1972; 102:303–306. 13. Funaki B. Endovascular intervention for the treatment of acute arterial gastrointestinal hemorrhage. Gastroenterol Clin North Am 2002; 31:701–713. 14. Encarnacion CE, Kadir S, Beam CA, Payne CS. Gastrointestinal bleeding: treatment with gastrointestinal arterial embolization. Radiology 1992; 183:505–508. 15. Ljungdahl M, Eriksson LG, Nyman R, Gustavsson S. Arterial embolisation in management of massive bleeding from gastric and duodenal ulcers. Eur J Surg 2002; 168:384–390. 16. Holme JB, Nielsen DT, Funch-Jensen P, Mortensen FV. Transcatheter arterial embolization in patients with bleeding duodenal ulcer: an alternative to surgery. Acta Radiol 2006; 47:244–247.
Morishita et al
’
JVIR
17. Aina R, Oliva VL, Therasse E, et al. Arterial embolotherapy for upper gastrointestinal hemorrhage: outcome assessment. J Vasc Interv Radiol 2001; 12:195–200. 18. Loffroy R, Guiu B, Cercueil JP, et al. Refractory bleeding from gastroduodenal ulcers: arterial embolization in high-operative-risk patients. J Clin Gastroenterol 2008; 42:361–367. 19. Poultsides GA, Kim CJ, Orlando R III, Peros G, Hallisey MJ, Vignati PV. Angiographic embolization for gastroduodenal hemorrhage: safety, efficacy, and predictors of outcome. Arch Surg 2008; 143: 457–461. 20. Larssen L, Moger T, Bjornbeth BA, Lygren I, Klow NE. Transcatheter arterial embolization in the management of bleeding duodenal ulcers: a 5.5-year retrospective study of treatment and outcome. Scand J Gastroenterol 2008; 43:217–222. 21. Loffroy R, Guiu B. Role of transcatheter arterial embolization for massive bleeding from gastroduodenal ulcers. World J Gastroenterol 2009; 15:5889–5897. 22. Toyoda H, Nakano S, Takeda I, et al. Transcatheter arterial embolization for massive bleeding from duodenal ulcers not controlled by endoscopic hemostasis. Endoscopy 1995; 27:304–307. 23. BK R, van Munster IP, de Jager CP, Rutten MJ. Embolization as treatment of choice for bleeding peptic ulcers in high-risk patients. Dig Surg 2009; 26:37–42. 24. Ledermann HP, Schoch E, Jost R, Decurtins M, Zollikofer CL. Superselective coil embolization in acute gastrointestinal hemorrhage: personal experience in 10 patients and review of the literature. J Vasc Interv Radiol 1998; 9:753–760. 25. Loffroy R, Guiu B, D’Athis P, et al. Arterial embolotherapy for endoscopically unmanageable acute gastroduodenal hemorrhage: predictors of early rebleeding. Clin Gastroenterol Hepatol 2009; 7: 515–523. 26. Loffroy R, Guiu B, Mezzetta L, et al. Short- and long-term results of transcatheter embolization for massive arterial hemorrhage from gastroduodenal ulcers not controlled by endoscopic hemostasis. Can J Gastroenterol 2009; 23:115–120. 27. Yamagami T, Kato T, Iida S, Tanaka O, Nishimura T. Value of transcatheter arterial embolization with coils and N-butyl cyanoacrylate for long-term hepatic arterial infusion chemotherapy. Radiology 2004; 230:792–802. 28. Dotter CT, Goldman ML, Rosch J. Instant selective arterial occlusion with isobutyl 2-cyanoacrylate. Radiology 1975; 114:227–230. 29. Kish JW, Katz MD, Marx MV, Harrell DS, Hanks SE. N-butyl cyanoacrylate embolization for control of acute arterial hemorrhage. J Vasc Interv Radiol 2004; 15:689–695. 30. Ichiro I, Shushi H, Akihiko I, Yasuhiko I, Yasuyuki Y. Empiric transcatheter arterial embolization for massive bleeding from duodenal ulcers: efficacy and complications. J Vasc Interv Radiol 2011; 22:911–916.