Superselective Microcoil Embolization for the Treatment of Lower Gastrointestinal Hemorrhage

Clinical Studies

Superselective Microcoil Embolization for the Treatment of Lower Gastrointestinal Hemorrhage William T. Kuo, MD, David E. Lee, MD, Wael E.A. Saad, MBBCh, Nikhil Patel, MD, Lawrence G. Sahler, MD, and David L. Waldman, MD, PhD PURPOSE: To evaluate the safety and effectiveness of superselective microcoil embolization for the treatment of lower gastrointestinal (LGI) hemorrhage. MATERIALS AND METHODS: A retrospective review of LGI superselective microcoil embolization data for a 10-year period was performed. During this period, twenty-two patients with evidence on angiography of LGI bleeding underwent superselective microcoil embolization. Hemorrhage was treated in the colon (n ⴝ 19) and jejunum (n ⴝ 3). Ivalon was used adjunctively in two patients and gelfoam was used as a secondary agent in two additional patients. Postembolization ischemia was evaluated objectively in 14 patients by colonoscopy (n ⴝ 10), surgical specimen (n ⴝ 3), and barium enema (n ⴝ 1). All patients were followed for clinical evidence of bowel ischemia. Four patients died before further follow-up could be performed. Additionally, 122 cases of LGI hemorrhage treated with superselective microcoil embolization were identified in a review of the literature. A meta-analysis was then performed, combining the data in this study and the data from the literature, to estimate the rate of major and minor ischemic complications on a total of 144 superselective microcoil embolizations. RESULTS: Immediate hemostasis was achieved in all 22 patients in this study. Complete clinical success was achieved in 86% of patients (19 of 22 patients). Rebleeding occurred in 14% of patients (3 of 22 patients) and each underwent colonoscopic intervention with success. Postembolization objective follow-up was performed in 64% of patients (14 of 22 patients). Ten patients underwent follow-up colonoscopy; one patient received a follow-up barium enema; and three patients underwent subsequent surgery. Colonic resection (one partial and one total) was performed in two patients. The partial colectomy was performed in a patient who had been diagnosed with colonic polyps and dysplasia. The total colectomy was performed on a patient with history of chronic LGI bleeding complicated by long-term anticoagulation therapy and a history of tubular adenoma resection. The third surgical patient (16 months old) underwent a follow-up exploratory laparotomy after embolization of a proximal jejunal branch of the superior mesenteric artery. None of the three patients who underwent surgery were found to have postembolic ischemic changes in the bowel specimen. Four patients in this study died, for reasons unrelated to hemorrhage or embolization, before further follow-up could be performed. The last four patients were followed clinically and experienced no symptoms of intestinal ischemia. A minor ischemic complication was reported in 4.5% of patients (1 of 22 patients), and there were no major ischemic complications (0%) in this series. A review of the data from 122 cases of LGI superselective microcoil embolization in the literature is also presented. Combined with the data in this study, the minor complication rate was 9% (13 of 144 patients), and the major complication rate was 0% (0 of 144 patients). CONCLUSION: Superselective microcoil embolization is a safe and effective treatment for LGI hemorrhage. J Vasc Interv Radiol 2003; 14:1503–1509 Abbreviations:

LGI ⫽ lower gastrointestinal, PVA ⫽ polyvinyl alcohol

ACUTE lower gastrointestinal (LGI) hemorrhage is a potentially life threatFrom the Department of Diagnostic Radiology, Division of Vascular and Interventional Radiology, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY 14642. Received June 21, 2003; sent for revision August 5; final revision received August 11; accepted August 25. Presented at the 2003 SIR Annual Meeting. Address correspondence to W.T.K.; E-mail: [email protected] None of the authors has identified a potential conflict of interest. © SIR, 2003 DOI: 10.1097/01.RVI.0000099780.23569.E6

ening condition. Although most cases can be managed medically, 10%–15% require intervention to control the hemorrhage (1). The therapeutic options include surgery, vasopressin infusion, endoscopy, and transcatheter embolization. Surgery is associated with mortality rates as high as 15%– 30% in emergent operations (1–3), and vasopressin is associated with high rates of rebleeding and multiple side effects (1). There is still ongoing controversy regarding the preferential use of transcatheter embolotherapy versus transcatheter vasopressin infusion and

further studies are warranted to elucidate this issue (4). Although colonoscopic intervention has become a first line method to investigate and to treat LGI hemorrhage, endoscopy is limited in cases of massive hemorrhage because the presence of copious blood and stool may prevent visualization of a bleeding source. Given a positive dynamic bleeding scan, mesenteric angiography becomes important not only because it localizes the active bleeding, but also because it offers a pathway for potential treatment. Modern transcatheter embolotherapy has

1503

1504

•

Superselective Embolotherapy for Lower Gastrointestinal Hemorrhage

become a viable option in the algorithm for managing LGI hemorrhage, and microcoils have become the preferred agent for many interventionalists (5– 8). The weaker anastamotic blood supply of the LGI tract, compared with the upper gastrointestinal system, is well known and predisposes the colon to a theoretically increased risk of postembolic ischemia. Although many cases of LGI embolization have been reported in the literature, the safety of this procedure has not been clearly defined. Several cases of postembolic colonic infarction were reported in the 1980s (9,10); but since then, newer catheter-based techniques have become available allowing more precise—superselective— hemostasis while preserving collateral blood flow to bowel mucosa (11). Therefore, it is necessary to redefine the postembolic infarction rate in terms of data from modern embolotherapy. This study presents the authors’ experience with superselective LGI microcoil embolization for LGI hemorrhage with the goal of evaluating the safety and effectiveness of this treatment. Furthermore, a review of the literature for similar cases of LGI microcoil embolization is presented in an attempt to estimate the rate of major and minor ischemic complications from this modern transcatheter technique.

MATERIALS AND METHODS Patient Population A retrospective review of all microcoil embolizations for LGI hemorrhage for a 10 year period (1992–2002) was performed. An institutional review board exemption for the study was obtained. Twenty-two patients undergoing microcoil embolization were identified. In this consecutive group, there were 11 men and 11 women with age ranging from 16 months to 84 years with a mean age of 62 years. Nuclear scintigraphy (tagged red blood cell scan) was used to identify active hemorrhage in 11 patients before angiography. Other modalities were used to diagnose hemorrhage in five additional patients: colonoscopy was performed on three patients, abdominal computed tomography (CT) was diagnostic in one patient (active mesenteric bleeding), and abdominal ultrasound (US) was used to evaluate

a 16-month-old patient (intraperitoneal hematoma) before angiography. One patient, after rectocele repair, was taken directly to angiography because a large postoperative hematoma was diagnosed from physical examination of the cul-de-sac. The five remaining patients underwent emergent angiography without a screening examination. The average pre-embolization transfusion requirement per patient was 6.8 U of packed red blood cells. One patient, with an international normalized ratio of 3.5 from warfarin therapy, was transfused with 8 U of fresh frozen plasma resulting in an international normalized ratio of 1.4 before angiography. Three other patients on warfarin did not require fresh frozen plasma because the medication had been withheld earlier resulting in international normalized ratio values ranging from 1.0 to 1.2 at the time of embolization. Heparin infusion was discontinued in a separate patient, with an activated partial thromboplastin time of 76.8 seconds, resulting in an activated partial thromboplastin time of 33.3 seconds at the time of angiography. Embolization Technique In this case series, mesenteric angiography was performed in each patient with use of digital subtraction imaging and selective arterial contrast injections with standard 5-F catheters. All patients exhibited contrast extravasation on angiography. When the bleeding site was identified, superselective embolization was performed with use of a 3-F coaxial microcatheter system with either the Tracker-18 microcatheter (Target Therapeutics, San Jose, CA) or Renegade microcatheter (Boston Scientific, Natick, MA) with a 0.014-inch guide wire. The microcatheter was positioned as close as possible proximal to the bleeding site. The target artery was the vasa recta; although, some embolizations were performed within the marginal artery of Drummond or the distal intestinal arcades. The primary embolic agent in each case was microcoils (Cook, Bloomington, IN) and a secondary agent, either Gelfoam pledgets (Upjohn, Kalamazoo, MI) or polyvinyl alcohol (PVA) particles as 250 –1000 ␮m particles of Ivalon (Unipoint Indus-

December 2003

JVIR

tries, High Point, NC), was used in four of the earliest patients. The decision to use a secondary embolic agent was operator dependent. Treatment sites included the colon (n ⫽ 19) and jejunum (n ⫽ 3). Postembolization angiography was immediately performed to confirm control of hemorrhage. Definitions and Data Analysis Data were analyzed according to definitions and guidelines for percutaneous transcatheter embolization determined by the Society of Interventional Radiology (12). Technical success was defined as immediate cessation of hemorrhage evaluated by completion angiography. Complete clinical success was defined as the resolution of signs or symptoms that prompted embolization until the end of the study observation period. Rebleeding was assessed by checking clinical parameters (passage of blood per rectum) and by hematologic evaluation (hematocrit, packed red blood cell transfusions). Ischemic complications were defined based on outcome. A major ischemic complication was defined as bowel infarction that required surgery. Minor ischemic complications were defined as conditions that resulted in no sequelae and required nominal therapy or no further treatment (12). An objective incidence of ischemia was defined as the number of ischemic complications (minor or major) divided by the number of patients with postembolization objective follow-up. Major and minor ischemic complication rates were determined for the 22 patients in this study. After this, a review of the literature (1992–2002) regarding superselective embolotherapy for LGI hemorrhage identified 122 cases of microcoil embolization for analysis. The data from this study was combined with the data from the literature and the minor and major ischemic complication rates from 144 cases of superselective microcoil embolization were estimated. Study Endpoints In this study, 14 patients underwent objective evaluation for bowel ischemia by colonoscopy (n ⫽ 10), surgery (n ⫽ 3), or barium enema (n ⫽ 1)

Volume 14

Number 12

Kuo et al

•

1505

formed in subsequent healthcare visits (mean, 5 months).

RESULTS

Figure 1. (a) Superior mesenteric angiography in a 38-year-old man with diverticular disease and acute LGI hemorrhage showed contrast extravasation in the hepatic flexure. (b) Superselective embolization with two microcoils was performed within the vasa recta achieving successful hemostasis.

at a mean of 41 days, 4 days, and 1 day, respectively. All patients were monitored daily for clinical signs/ symptoms of intestinal ischemia (abdominal pain/tenderness, peritoneal signs, nausea, diarrhea, fever) and with laboratory assessment (elevated

lactate and white blood count) until discharge or death (mean, 26 days). Four patients died, for reasons unrelated to hemorrhage or embolization, before further follow-up could be performed. In the remaining 18 patients, additional clinical follow-up was per-

Superselective transcatheter microcoil embolization was performed in 22 patients with LGI hemorrhage and technical success (immediate hemostasis) was achieved in all patients. Eleven superior mesenteric artery branch catheterizations were performed with the following embolization distributions: jejunal branch (n ⫽ 3), ileocolic artery (n ⫽ 2), branch of right colic (n ⫽ 4), branch of middle colic (n ⫽ 1), and cecal branch (n ⫽ 1). Nine inferior mesenteric artery branch embolizations were performed with the following distributions: branch of left colic artery (n ⫽ 2) and branch of superior hemorrhoidal artery (n ⫽ 7). Two internal iliac artery catheterizations were performed with embolization of the middle rectal artery (n ⫽ 2). Figures 1 and 2 demonstrate two cases of microcoil embolization in this series. The etiologies of LGI hemorrhage in this study are described in Table 1. Diagnoses were made by colonoscopy in 13 patients (three pre-embolization and 10 postembolization), barium enema in one patient, abdominal CT in one patient, postsurgical examination in one patient (after rectocele repair), and surgical specimen in 3 patients. It was assumed that anticoagulation therapy contributed to hemorrhage in patients taking such medication. One patient with rectal ulcers was diagnosed with idiopathic thrombocytopenic purpura by bone marrow aspirate. Etiologies of hemorrhage were unknown in three patients—two died before scheduled colonoscopy and one refused follow-up endoscopy. Postembolization rebleeding occurred in 14% of patients (3 of 22 patients) within 1 day of embolization. The first patient, diagnosed with rectal ulcers and ITP, developed hemorrhage from a branch of the superior rectal artery which was initially embolized with two microcoils. After embolotherapy, the patient required 9 U of packed red blood cells to maintain a stable hematocrit level. Colonoscopy was performed the next day and revealed a rectal ulcer 10 –13 cm from the anal verge with evidence of active hemorrhage. The rebleeding was suc-

1506

•

Superselective Embolotherapy for Lower Gastrointestinal Hemorrhage

December 2003

JVIR

Table 1 Etiologies of Lower Gastrointestinal Hemorrhage in 22 Patients Etiology

n

Diverticular disease Diverticular disease on coumadin Rectal villous adenoma Rectal adenocarcinoma Metastatic colorectal carcinoma Rectal ulcer on coumadin Rectal ulcers with ITP Hemorrhoids with coagulopathy Postpolypectomy Postrectocele repair Anticoagulation Unknown

6 1 2 1 2 1 1 1 1 1 2 3

Note.—ITP ⫽ idiopathic thrombocytopenic purpura.

Figure 2. (a) Inferior mesenteric angiography in a 66-year-old man on anticoagulation therapy with acute LGI bleeding. Contrast extravasation was noted in the sigmoid colon. (b) Magnified view of the sigmoid after placement of a single microcoil in the vasa recta demonstrated complete cessation of hemorrhage.

cessfully treated with heater probe and epinephrine injection with no further need for blood transfusions. The second patient with rebleeding was diagnosed with coagulopathy secondary to underlying liver disease. Extravasation from the right superior

rectal artery was initially identified on angiography and embolized with microcoils. Four hours later, the patient developed recurrent bleeding requiring an additional 4 U of packed red blood cells to maintain a stable hematocrit level. Proctoscopy was immedi-

ately performed and showed steady bleeding from a hemorrhoid along the right lateral rectal wall. Hemostasis was achieved with endoscopic cautery and sutures with no further need for blood transfusions. The third patient with recurrent hemorrhage was a 53-year-old woman with history of synthetic mitral valve replacement requiring long-term anticoagulation therapy. The patient had developed bright red blood per rectum, which was originally identified on a tagged red blood cell scan but was not seen on initial angiography or colonoscopy. To prevent further bleeding, the patient underwent subtotal colectomy (bowel specimen unremarkable). One week later, the patient presented with recurrent episodes of bright red blood per rectum. Repeat angiography demonstrated brisk bleeding from the superior hemorrhoidal artery, and this was successfully embolized with two microcoils. Another week later, the patient continued to bleed with hematocrit level falling to 25 from 30. A colonoscopy was performed showing a 2-cm rectal ulcer with active hemorrhage. The ulcer was excised endoscopically achieving successful hemostasis (no further bright red blood per rectum and stable hematocrit level). Three patients underwent postembolization surgery. The first patient had a partial colonic resection secondary to persistent abdominal pain and colonic polyps with dysplasia. In the second patient, a total colectomy was

Volume 14

Number 12

Kuo et al

•

1507

Table 2 Ischemic Complications Following Superselective LGI Embolization Involving Microcoils Year/Author

Minor Ischemia

Infarction

Objective Follow up

Cases

1996 Curzon et al. (13) 1997 Gordon et al. (5) 1998 Peck et al. (14) 1998 Lederman et al. (8) 1998 Nicholson et al. (15) 1999 Dobson and Nicholson (16) 2000 Evangelista and Hallisey (6) 2001 Funaki et al. (17) 2001 Bandi et al. (11) 2001 Defreyne et al. (18) 2002 Debarros et al. (19) 2003 Current study Total

2 0 0 0 3 1 2 3 ? 0 1 1 13

0 ? 0 0 0 0 0 0 0 0 0 0 0

? 5 3 2 ? 3 12 7 ? 0 16 14 62

9 13 6 10 14 4 17 24 8 1 16 22 144

Note.—? ⫽ data unspecified or indeterminate. The objective incidence of minor ischemia is 21% (13/62) with overall incidence of 9% (13/144). The estimated incidence of infarction is 0% (0/144).

performed because of a history of tubular adenoma (previously resected) and chronic LGI bleeding complicated by long-term anticoagulation therapy. The third patient, a 16-month-old girl, underwent a follow-up exploratory laparotomy status post embolization of a proximal jejunal branch of the superior mesenteric artery. At surgery, none of the three patients was found to have ischemic changes in the embolized bowel specimen. In this study, complete clinical success was achieved in 19 of 22 patients (86%). Among all 22 patients, there were no clinical signs or symptoms of intestinal ischemia and no cases of postembolic infarction (0%). A minor ischemic complication was reported in 1 of 22 patients (4.5%). This patient had undergone successful microcoil embolization of an ileocolic branch hemorrhage. Colonoscopy was performed the next day showing a probable postembolization ulcer in the ascending colon. The endoscopy report suggested that an arterial venous malformation or a Dieulafoy submucosal artery was obliterated by the therapeutic angiography procedure. The patient was asymptomatic and required no further treatment. The objective incidence of minor ischemia in this series was 7% (1 of 14 patients) with an overall incidence of 4.5% (1 of 22 patients). The estimated incidence of infarction was 0% (0 of 22 patients). When this data was combined with 122 cases in the literature, the objective incidence of minor isch-

emia was 21% (13 of 62 patients) with an overall incidence of 9% (13 of 144 patients). In 144 cases, the estimated incidence of infarction was 0% (Table 2).

DISCUSSION In 1974, Bookstein used autogenous clot formed on oxidized cellulose to arrest LGI hemorrhage in three patients (20). Gianturco described the first coil embolization in 1975 (21). In 1977, Goldberger and Bookstein (22) described transcatheter embolization for acute diverticular hemorrhage as an alternative to surgery ; and in 1978, Bookstein (23) further described embolotherapy for LGI hemorrhage reporting no evidence of postembolic infarction. Between 1981 to 1984, postembolic ischemic complications, including infarction, were reported in the literature in five studies (9,10,24 – 26). However, because of the small number of cases, the risk of bowel infarction was still not clearly defined. Until recently, objective postembolization follow-up data has been sparse in the literature (Table 2). Furthermore, most published series have combined therapeutic embolization data from a variety of embolysates (including PVA, gelfoam, and microcoils) making it difficult to define the risk of ischemic complication in terms of a single embolic agent. Bandi (11) reviewed the literature for two decades and estimated a 10%– 20% incidence of postembolization co-

lonic infarction in the 1980s with no cases of infarction after 1992. Postembolic infarction in the 1980s was described in subselective catheterizations with gelfoam embolizations, which are no longer widely used (9,10,26). The recent decrease in postembolic infarction has been attributed to modern superselective techniques (11). Therefore, there is a need to redefine the incidence of infarction with use of the most recent data from modern embolotherapy and specifically in terms of microcoils. The safety and effectiveness of superselective microcoil embolotherapy is supported in the literature. Funaki (17) achieved 96% hemostasis in a series of 25 patients. Debarros (19) demonstrated the safety of superselective microcoil embolization in 16 patients with LGI hemorrhage (each with objective follow-up) and showed no postembolic infarction. This series of 22 patients also demonstrates no major ischemic complications, and these results agree with other embolization data (Table 2). During the last decade (1992–2002), there have been several case series of microcoil embolization for LGI hemorrhage. One instance of postembolic infarction was reported by Funaki (17) in 2001, where superselective embolization was not technically possible and emergency embolotherapy was performed in branches of both the inferior mesenteric artery and superior mesenteric artery as a lifesaving maneuver. Because this procedure was not superselective, it is not in-

1508

•

Superselective Embolotherapy for Lower Gastrointestinal Hemorrhage

cluded in Table 2. In 1997, Gordon (5) described a case of bowel necrosis, likely multifactorial, in a patient with history of two liver transplants and ulcerative colitis and who later developed a stricture and bacterial sepsis 2 months after embolization. Because this episode of bowel ischemia was not directly attributable to coil embolization, this case was not included in the estimation of a postembolic infarction rate. Advancements in microcatheter technology that allow superselective catheterization have resulted in vast improvements in transcatheter embolotherapy. The development of finer torqueable guide wires and coaxial microcatheters coupled with advances in digital fluoroscopic imaging allows more precise vascular intervention. This is particularly important in the colon because of the relative lack of collateral arterial flow. Superselective embolization allows theoretical preservation of blood flow adjacent to the hemorrhagic site, thus minimizing ischemia to the bowel mucosa (11). Microcoils have become the agent of choice for many interventionalists (5– 8). The wide range of available coil sizes allows custom selection of coil to match the vessel diameter. Because these coils are highly radiopaque, they can more accurately be deployed. This is in contrast to PVA particles, which are not radiopaque and less easily controlled during deployment. Although Guy (27) was successful with the use of PVA particles (100 –590 ␮m) in the superselective treatment of nine of 10 patients with no evidence of bowel infarction and Defreyne (18) was similarly successful with PVA particle embolization in nine patients with LGI hemorrhage, there is a theoretically increased risk of ischemia with the use of PVA particles. The occlusion of smaller vessels with impairment of intramural collateral microcirculation was shown by Kusano (28), who produced gangrenous bowel in three dogs injected with small (149 –250 ␮m) PVA particles; however, there was no necrosis when medium to large particles (420 –1000 ␮m) were deployed. In a subsequent clinical study that used larger particles (590 –1000 ␮m) in five patients with LGI bleeding, there were no cases of infarction (28); however, it is believed that large PVA particles (420 –1000 ␮m) may potentially frag-

ment and occlude the submucosal plexus (28). Further evidence that extremely distal embolization causes infarction has been reported by Görich (29) who described four bowel infarctions with Ethibloc (Ethicon GmbH, Norderstedt, Germany). In a study by Debarros (19), PVA particles were deployed in the superselective treatment of 10 cases of LGI hemorrhage and one patient was clinically diagnosed with postembolization ischemia that required exploratory laparotomy and a right hemicolectomy. In the same study, 16 patients were treated with superselective embolization with microcoils and there were no major ischemic complications. Among this subgroup, one patient developed asymptomatic mucosal ischemia that required only follow-up endoscopy and no further treatment (19). Based on these findings, future studies are warranted to compare the safety of large PVA particles against microcoils. Although Gelfoam was used effectively in the 1980s, it is no longer widely used for LGI hemorrhage. Gelfoam has the disadvantage of being only a temporary embolic agent and it cannot easily be deployed superselectively. Similarly, liquid embolic agents such as glue are suboptimal because of difficulties in assessing adequate dilution and in controlling vascular penetration (18). The complications of vasopressin infusion (myocardial infarction, arrhythmia, hypertension, volume overload) are well known and significant in patients with LGI hemorrhage who are typically older with multiple medical problems. Furthermore, the effectiveness of vasopressin is limited secondary to a rebleeding rate as high as 50% (1). Although further studies are needed to clarify the controversy regarding transcatheter embolization versus vasopressin infusion, a comprehensive review of the literature by Darcy (4) concluded that embolotherapy will likely be used preferentially over vasopressin infusion because embolization appears to pose fewer risks and can be completed more rapidly than a vasopressin infusion protocol. The treatment of acute LGI hemorrhage has traditionally been surgical and the high morbidity and mortality from bowel resection is well-documented (1–3). Patients with acute LGI

December 2003

JVIR

hemorrhage, usually an older population, tend to be high surgical risks. Thus, the advantages of embolotherapy versus surgery include a high effectiveness (17) and an inherently less invasive procedure. The initial cessation of bleeding with use of a transcatheter approach may also serve to stabilize the patient for later surgical resection if necessary (9). In many institutions, embolotherapy has progressed to supercede vasopressin infusion and surgery for the treatment of massive LGI hemorrhage, which cannot be managed endoscopically. Few cases of small bowel LGI embolization have been reported in the literature because bleeding typically involves the colon. None of the three patients in this study with LGI small bowel hemorrhage suffered detectable postembolic ischemic injury (one surgical follow-up, two clinical followups). To the authors’ knowledge, small bowel infarction after embolotherapy has not been reported and the results in this study agree with the literature (14,25). The increased risk of ischemia in the LGI tract stems from a lack of abundant collateral vessels, particularly in the colon. Ross (30), with use of celloidin corrosion injection, showed very few anastamoses between the vasa recta of the colon compared with the small intestine. However, subsequent anatomic studies have demonstrated rich intramural networks and collaterals that preserve perfusion at the level of the colonic mucosa (31,32). This may help to explain the decrease in ischemic complications after superselective embolotherapy. The authors of this study agree with Evangelista (5), Gordon (6), and Lederman (33) that a distal site at the level of the vasa recta or marginal artery should be chosen for embolization in order to minimize the region at risk for ischemia. For further safety, the authors believe that all patients should undergo objective follow-up (preferably colonoscopy) not only to identify possible ischemic complications but also to determine the etiology of hemorrhage and the potential need for surgical resection. A limitation of transcatheter therapy is the technical difficulty in reaching the target site in patients with atherosclerotic occlusive disease or vasospasm. This study was not designed

Volume 14

Number 12

to assess the rate of superselective LGI catheterization success. Funaki (17) estimated a catheterization success rate of 93% in 25 patients, and Debarros (19) reported 100% catheter placement success in 16 patients who underwent superselective LGI microcoil embolization. Precise catheter placement may partially depend on operator skill; however, based on this data, the authors believe there is a high success rate when attempting superselective mesenteric catheterization.

Kuo et al

7.

8.

9.

CONCLUSION Modern superselective microcoil embolization is a safe and effective treatment for LGI hemorrhage. The data in this study agrees with the literature, which collectively shows a minimal risk of significant ischemic complication when a modern transcatheter technique is used. When this series was combined with several other studies, the postembolic infarction rate was estimated as 0%. In light of these results and until further studies demonstrate a superior alternative treatment, massive LGI hemorrhage detected by mesenteric angiography should be treated with attempted superselective microcoil embolization to control bleeding. This data presents modern microcoil embolotherapy as a safe treatment option in an algorithm for the management of LGI hemorrhage. References 1. Billingham RP. The conundrum of lower gastrointestinal bleeding. Surg Clin North Am 1997; 77:241–252. 2. Behringer GE, Albright NL. Diverticular disease of the colon: a frequent cause of massive rectal bleeding. Am J Surg 1973; 125:419 – 423. 3. Giacchino JL, Geis WP, Pickleman JR, Dado DV, Hadcock WE, Freeark RJ. Changing perspectives in massive lower intestinal hemorrhage. Surgery 1979; 86:368- 376. 4. Darcy M. Treatment of lower gastrointestinal bleeding: vasopressin infusion versus embolization. J Vasc Interv Radiol 2003; 14:535–543. 5. Gordon RL, Ahl KL, Kerlan RK, et al. Selective arterial embolization for the control of lower gastrointestinal bleeding. Am J Surg 1997; 174:24 –28. 6. Evangelista PT, Hallisey MJ. Transcatheter embolization for acute lower

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

gastrointestinal hemorrhage. J Vasc Interv Radiol 2000; 11:601– 606. Nemcek AA. Transcatheter therapy for lower gastrointestinal bleeding. Tech Vasc Intervent Radiol 2000; 3:139 –146. Lederman 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. Rosenkrantz H, Bookstein JJ, Rosen RJ, Goff WB II, Healey JF. Postembolic colonic infarction. Radiology 1982; 142: 47–51. Uflacker R. Transcatheter embolization for treatment of acute lower gastrointestinal bleeding. Acta Radiol 1987; 28:425– 430. Bandi R, Shetty PC, Sharma RP, Burke TH, Burke MW, Kastan D. Superselective arterial embolization for the treatment of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 2001; 12:1399 –1405. Drooz AT, Lewis CA, Allen TE, et al. Quality improvement guidelines for percutaneous transcatheter embolization: SCVIR Standards of Practice Committee—Society of Cardiovascular and Interventional Radiology. J Vasc Interv Radiol 1997; 8:889 – 894. Curzon IL, Nicholson AA, Dyet JF, Hartley J. Transcatheter coil embolotherapy for major colonic hemorrhage (abstract). Cardiovasc Intervent Radiol 1996; 19:S83. Peck DJ, McLoughlin RF, Hughson MN, Rankin RN. Percutaneous embolotherapy of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 1998; 9:747–751. Nicholson AA, Ettles DF, Hartley JE, et al. Transcatheter coil embolotherapy: a safe and effective option for major colonic haemorrhage. Gut 1998; 43:79 – 84. Dobson CC, Nicholson AA. Treatment of rectal hemorrhage by coil embolization. Cardiovasc Intervent Radiol 1999; 22:143–146. Funaki B, Kostelic JK, Lorenz J, et al. Superselective microcoil embolization of colonic hemorrhage. AJR Am J Roentgenol 2001; 177:829 – 836. Defreyne L, Vanlangenhove P, De Vos M, et al. Embolization as a first approach with endoscopically unmanageable acute nonvariceal gastrointestinal hemorrhage. Radiology 2001; 218: 739 –748. Debarros J, Rosas L, Cohen J, Vignati P, Sardella W, Hallisey M. The changing paradigm for the treatment of co-

20.

21. 22.

23.

24.

25. 26.

27.

28.

29.

30. 31. 32. 33.

•

1509

lonic hemorrhage. Dis Colon Rectum 2002; 45:802– 808. Bookstein JJ, Chlosta EM, Foley D, Walter JF. Transcatheter hemostasis of gastrointestinal bleeding using modified autogenous clot. Radiology 1974; 113:277- 285. Gianturco C, Anderson JH, Wallace S. A mechanical device for arterial occlusion. AJR 1975; 124:428 – 435. Goldberger LE, Bookstein JJ. Transcatheter embolization for the treatment of diverticular hemorrhage. Radiology 1977; 122:613– 617. Bookstein JJ, Naderi MJ, Walter JF. Transcatheter embolization for lower gastrointestinal bleeding. Radiology 1978; 127:345–349. Gerlock AJ, Muhletaler CA, Berger JL, Halter SA, O’Leary JP, Avant GR. Infarction after embolization of the ileocolic artery. Cardiovasc Intervent Radiol 1981; 4:292–205. Palmaz JC, Walter JF, Cho KJ. Therapeutic embolization of the small-bowel arteries. Radiology 1984; 152:377–382. Walker WJ, Goldin AR, Shaff MI, Allibone GW. Per catheter control of hemorrhage from the superior and inferior mesenteric arteries. Clin Radiol 1980; 31:71– 80. Guy G, Shetty PC, Sharma R, Burke M, Burke T. Acute lower gastrointestinal hemorrhage: treatment by superselective embolization with polyvinyl alcohol particles. AJR Am J Roentgenol 1992; 159:521–526. Kusano S, Murata K, Ohuchi H, Motohashi O, Atari H. Low-dose particulate polyvinylalcohol embolization in massive small artery intestinal hemorrhage: experimental and clinical results. Invest Radiol 1987; 22:388 –392. Görich J, Brambs JH, Allmenröder C, et al. The role of embolization treatment of acute hemorrhage. Fortschritte auf dem Gebiete der Röntgenstrahlen und der Neuen Bildgebenden Verfahren 1993; 159:379 –387. Ross JA. Vascular patterns of small and large intestine compared. Br J Surg 1952; 39:330 –333. Griffiths JD. Extramural and intramural blood supply of colon. BMJ 1961; 1:323–326. Brockis JG, Moffat DB. The intrinsic blood vessels of the pelvic colon. J Anat 1958; 92:52–56. Lederman HP, Schoch E, Jost R, Zollikofer CL. Embolization of the vasa recta in acute lower gastrointestinal hemorrhage: a report of five cases. Cardiovasc Intervent Radiol 1999; 22:315– 320.