Tech Coloproctol DOI 10.1007/s10151-013-1112-x

ORIGINAL ARTICLE

Management of angiogram-negative acute colonic hemorrhage: safety and efficacy of colonoscopy-guided superselective embolization J. Heianna • T. Miyauchi • H. Yamano • K. Yoshikawa • M. Hashimoto • S. Murayama

Received: 23 May 2013 / Accepted: 20 December 2013 Ó Springer-Verlag Italia 2014

Abstract Background We evaluated the efficacy and safety of superselective embolization with assistance of colonoscopy for acute colonic hemorrhage. Methods Of 92 cases of acute colonic hemorrhage requiring colonoscopic intervention, 11 (12 %) could not be successfully treated. Of these, 10 patients (9 men, mean age 65.5 years, range 39–75 years) underwent superselective embolization. Hemorrhage was caused by diverticular disease (n = 8), polypectomy (n = 1), and vascular malformation (n = 1). In all 10 cases, the radiopaque clips were placed at the bleeding point via colonoscopy. Microcatheters were used in all procedures, and embolization was performed at the level of the vasa recta leading to or near the clips with Gelfoam particles, microcoils, or both. Results Immediate hemostasis was achieved in all patients. In 6 of 10 patients (60 %), selective angiograms showed no active extravasation at the time of the procedure and the embolization was performed using clips as a landmark. In the remaining four patients, selective angiograms showed active extravasation from the vasa recta

leading to the clips. The mean number of embolized vessels with no active extravasation and with active extravasation was 1.83 (range 1–3) and 1.25 (range 1–2), respectively. The mean duration of clinical follow-up was 11.6 months (range 1–29 months). One patient (10 %) bled from a different site than the treated site a month after embolization, but the bleeding ceased after endoscopic intervention. All the patients (100 %) were evaluated for objective evidence of ischemia by colonoscopy. Four of the 10 patients (40 %) were found endoscopically to have small areas of ischemia involving only the mucosa, but they remained asymptomatic. There was no bowel infarction or stricture. Conclusions Colonoscopy-assisted superselective embolization may be a safe and useful procedure for acute colonic hemorrhage without active extravasation on angiogram. Keywords Angiogram-negative  Colonic hemorrhage  Embolization  Colonoscopy

Introduction J. Heianna (&)  S. Murayama Department of Radiology, Ryukyu University of Medicine, Okinawa, Japan e-mail: [email protected] T. Miyauchi Diagnostic Radiology, Akita Red Cross Hospital, Akita, Japan H. Yamano  K. Yoshikawa Division of Gastroenterology, Akita Red Cross Hospital, Akita, Japan M. Hashimoto Department of Radiology, Akita University of Medicine, Akita, Japan

Although more than 80 % of cases of colonic hemorrhage stop with conservative treatment, refractory bleeding requires intervention beyond medical management [1, 2]. The therapeutic options include surgery, endoscopy, and transcatheter arterial embolization (TAE). Generally, TAE of acute colonic hemorrhage is the therapy of choice in patients who are poor surgical risks and in patients who cannot be successfully treated by endoscopic intervention. Advances in microcatheter technology have enabled superselective catheterization and embolization of more distal arteries. Recently, many reports have suggested that

123

Tech Coloproctol

superselective embolization for colonic hemorrhage rapidly arrests bleeding while minimizing the risk of ischemia [3–19]. However, the most important limitation of TAE is the inability to treat patients whose angiogram does not show active extravasation, necessitating emergent operations for recurrent or persistent bleeding [14]. Until now, in the case of colonic hemorrhage that could not be treated by colonoscopy, we placed radiopaque clips at the bleeding site colonoscopically and then performed embolization of the vasa recta leading to or near the clips. The purpose of this study was to review our small retrospective series of patients who underwent superselective embolization with the assistance of colonoscopy for colonic hemorrhage that could not treated endoscopically. The ultimate goal of this study is to propose a clinical approach for the management of angiogram-negative acute colonic hemorrhage.

Materials and methods From January 2002 to April 2010, 1,457 cases of acute colonic hemorrhage were treated at Akita Red Cross Hospital. Endoscopy was performed in all cases by a gastroenterologist. Out of 92 cases of colonic hemorrhage requiring hemostasis, bleeding could be stopped in 81 cases by endoscopy. Bowel preparation was performed with oral bowel cleansers (1,098/1,457; 75 %), lukewarm enema (169/1,457; 12 %), and glycerin enema (145/1,457; 10 %). Forty-five patients (3 %) could not undergo bowel preparation. No patients received sedation during the procedure. The endoscopes used were as follows: CF-H260AZI, CFQ240ZI, PCF-Q240ZI, CF-200Z (OLYMPUS, Tokyo, Japan). Among the 11 patients with refractory colonic bleeding, one patient with no comorbidities, not on any medications, and with no abnormal blood test findings was selected for surgery because of multiple diverticular hemorrhage. The remaining 10 patients (nine men, one woman; mean age 65.5 years, range 39–75 years) who underwent attempted TAE for recurrent colonic hemorrhage were retrospectively reviewed. No patients underwent CT scanning before endoscopy in this study. Patients with repeated bleeding from the same site who were not treated endoscopically underwent TAE. Patients who underwent TAE had a large number of comorbidities (hypertension; n = 4, diabetes; n = 3, liver cirrhosis; n = 1, coronary artery disease; n = 2, chronic renal failure; n = 1, and gastric cancer; n = 1). All 10 patients underwent endoscopic evaluation and treatment of colonic hemorrhage, but endoscopic treatment did not stop the bleeding since clips could not be placed well on the exposed vessel. Instead, radiopaque clips were placed at the bleeding site via colonoscopy to mark the embolization site in all cases. All 10 patients underwent angiography on the day when clips

123

were placed at the bleeding site. The decision to attempt embolization rather than surgery was made by a gastroenterologist based on the clinical condition of the patient. The average pre-embolization transfusion requirement per patient was 5.8 units of packed red blood cells. After written informed consent was obtained from each patient, digital subtraction angiography was performed, using standard angiographic catheters, to provide selective diagnostic mesenteric arteriograms. Colonoscopic studies were used to guide catheterization. If the bleeding source appeared to be in the distal colon, the inferior mesenteric artery was catheterized initially and if bleeding appeared to be in the proximal colon, the superior mesenteric artery was treated first. Scopolamine butylbromide (10 mg) was administered at the discretion of the radiologist to decrease bowel peristalsis. The volume of the contrast medium was 25 ml/5 s in the case of superior mesenteric artery and 12 ml/3 s in the case of inferior mesenteric artery. In peripheral vessels, contrast medium was injected by hand pressure. The fluoroscopy duration was 15 s per examination. Regardless of whether or not active extravasation was identified on angiogram, we superselectively embolized the vasa recta leading to or near the clips with use of a 3-French coaxial microcatheter system consisting of Renegade microcatheter (Boston Scientific, Natick, MA, USA) and 0.016 inch guide wire (TERUMO, Tokyo, Japan) (Figs. 1, 2). Embolization materials included Gelfoam particles (Upjohn, Don Mills, Ontario, Canada) (n = 2), microcoils (Boston Scientific, Cork, Ireland) (n = 5), or both (n = 3) (Table 1). The selection of embolic material was at the discretion of the operator performing the procedure. Although criteria for selection of embolic material were not provided initially, in the later part of our study, we used microcoils only since we judged that Gelfoam particles were not suitable as an embolic material. Embolization was performed until no further extravasation was seen or the vasa recta near the clip did not appear. Colonoscopy for evaluation of objective signs of bowel infarction was attempted within 2 weeks after embolization in all patients. Minor ischemic complications were defined as events that required no further therapy. Major ischemic complications were defined as bowel infarction and stricture that required surgery. The outcome after embolization was assessed by review of medical records including postembolization endoscopy.

Results Colonic hemorrhage was caused by diverticular disease (n = 8), vascular malformation (n = 1), and polypectomy (n = 1) (Table 1). There were no patients who underwent TAE without undergoing a colonoscopy. All 10 patients

Tech Coloproctol Fig. 1 Superselective embolization of diverticular hemorrhage in a 71-year-old man (patient # 1). Radiopaque clips were placed at the bleeding point on the proximal side of the ascending colon via colonoscopy before angiography. a Superior mesenteric arteriogram shows no active extravasation. The bleeding point was on the clip nearest the oral side (arrow). b Selective ileocolic arteriogram shows no active extravasation. c Superselective angiogram shows active extravasation from the vasa recta leading to the clip (arrow). d After embolization of the vasa recta with Gelfoam particles, no extravasation is seen on angiogram

Fig. 2 Superselective embolization of diverticular hemorrhage in a 60-year-old man (patient # 8). a Radiopaque clips were placed at the bleeding site (arrow) on the distal side of the ascending colon via colonoscopy before angiography. Selective right colic arteriogram shows no extravasation. b Superselective arteriogram of the vasa recta does not show active extravasation. c After embolization of the vasa

recta associated with anastomosis, right colic arteriogram shows disappearance of the vasa recta near a clip. d Colonoscopy 13 days after transcatheter arterial embolization shows a mild ulcer and edema at the site of embolization. The patient has remained asymptomatic and has not required treatment

123

Tech Coloproctol Table 1 Summary of treatment results in 10 patients Patient #

Age/sex

Etiology

Extravasation

Embolic agent

Embolized vasa recta (#)

Ischemia

Rebleed No

1

71/M

Diverticulum

Yes

GP, S 9 1

2

No

2

75/M

After polypectomy

No

GP

1

Yes

No

3

74/M

Vascular malformation

No

GP

1

Yes

No

4

39/M

Diverticulum

No

GP, S 9 2

2

No

No

5

68/F

Diverticulum

No

GP, S 9 6

3

Yes

Yes

6

66/M

Diverticulum

No

S92

1

No

No

7

55/M

Diverticulum

Yes

S 9 2, C 9 2

1

No

No

8

60/M

Diverticulum

No

S 9 2, C 9 2

3

Yes

No

9

72/M

Diverticulum

Yes

S93

1

No

No

10

75/M

Diverticulum

Yes

S92

1

No

No

GP Gelfoam particles, S straight fibered platinum coil 5 mm length, C complex fibered platinum coil 2 mm/10 mm 9 number of microcoils used

underwent bowel preparation. All except for a post-polypectomy patient had confirmed adherent clots and prominent vessels on colonoscopy. Clips could not be placed well on the vessels endoscopically in any cases of diverticular hemorrhage since there were exposed vessels in the bottom of the diverticulum. Angiography detected an active extravasation of contrast in 4 of the 10 patients (40 %). Four active extravasations were seen from the vasa recta leading to the clips (Fig. 1c). In the remaining six patients (60 %), TAE was performed using clips as a landmark since active no extravasations were demonstrated even on the superselective angiogram. There were no cases of complications on angiography itself. There were 8 superior mesenteric artery branch catheterizations and two inferior mesenteric artery branch catheterizations. The superior mesenteric embolizations consisted of embolization of the ileocolic artery (n = 2) and a branch of the right colic artery (n = 6). The inferior mesenteric embolizations consisted of embolization of a branch of the left colic artery (n = 2). The mean number of embolized vasa recta was as follows: 1.25 (range 1–2) in cases with active extravasation and 1.83 (range 1–3) in cases without active extravasation (p = 0.1570 NS). The mean duration of clinical follow-up was 11.6 months (range 1–29 months). Post-embolization rebleeding occurred in one patient (10 %) at 1 month after embolization. This patient was diagnosed with diverticular hemorrhage and underwent TAE with no active extravasation. Colonoscopy revealed active bleeding from a different site that had been embolized previously. The bleeding was successfully treated with colonoscopic intervention. Recurrent bleeding was not seen in other patients during the clinical follow-up. All 10 patients (100 %) were evaluated for bowel ischemia with colonoscopy within 2 weeks after embolization. Minor ischemic complications included mild mucosal ischemia (small ulcers and fibrinopurulent debris

123

without necrosis) (Fig. 2d) in 4 of 10 patients (40 %). None of these 4 patients had active extravasation on angiography. Each of them was and has remained asymptomatic. Three of four cases (75 %) with mild ischemia were patients who underwent embolization using Gelfoam particles. In the remaining six patients, there were no mucosal abnormalities. No major ischemic complications including bowel infarction and stricture occurred. We were able to avoid surgery for persistent or recurrent hemorrhage and postembolic bowel infarction in all 10 patients.

Discussion Colonic hemorrhage is typically a disorder that affects the elderly with a wide range of clinical presentations. While more than 80 % of patients with colonic hemorrhage, who are characterized by minor bleeding, can be treated conservatively [1, 2], patients with continual active bleeding need emergency treatment because of their hypotensive state. However, the emergency treatment of patients with acute colonic hemorrhage remains a difficult clinical problem. Although surgery is still the foundation of treatment, the reported mortality rates after emergency colonic resection (hemicolectomy/subtotal colectomy) for colonic hemorrhage range from 15 to 30 % [20–22], and even limited bowel resection is associated with a mortality rate of approximately 5 % (0–25 %) in the best series [23]. Colonoscopic intervention has become a first-line method to treat colonic hemorrhage. However, if it is difficult to stop bleeding with colonoscopy, TAE can be considered as the next means of achieving hemostasis. Embolization for colonic hemorrhage was first performed by Goldberger and Bookstein in 1977 [24]. Advances in microcatheter technology in the 1990s that enabled superselective catheterization have resulted in vast improvements in transcatheter embolotherapy. The development of finer

Tech Coloproctol

torqueable guide wires and coaxial microcatheters coupled with advances in digital fluoroscopic imaging allows for more precise vascular intervention. As for embolic agents, most investigators have used Gelfoam particles, polyvinyl alcohol (PVA) particles, microcoils, or some combination of these. We initially used Gelfoam particles as the embolic material, but mucosal ischemia was seen in 3 of 5 (60 %) of the patients who underwent TAE with Gelfoam particles. Gelfoam and PVA particles are not radiopaque and are less easily controlled during deployment. We therefore hypothesize that the Gelfoam particles may have induced mucosal ischemia and changed our protocol to use microcoils since the embolic material is easy to visualize with fluoroscopy and deploy accurately. The safety and effectiveness of superselective microcoil embolotherapy are supported in the literature. Funaki et al. [9] achieved 96.0 % hemostasis in a series of 25 patients. Debarros et al. [12] demonstrated the safety of superselective microcoil embolization in 16 patients with lower gastrointestinal hemorrhage and showed no postembolic bowel infarction. This procedure should be the initial treatment of choice in any patient with colonic hemorrhage severe enough to require emergency treatment. However, even with superselective embolization and microcoils, the most important limitation of embolotherapy is the inability to treat patients who are not actively bleeding [3, 9, 10, 12]. If patients have no active bleeding at endoscopy or angiography for colonic hemorrhage, there is basically no problem with follow-up observation all on. However, if active bleeding is not confirmed at the time of the examination, we think that some form of treatment for colonic hemorrhage is necessary since the likelihood of recurrent bleeding occurring in the future is high. Burgess et al. [25] reported that patients who were treated with embolotherapy without active extravasation identified at the time of angiogram had a proven bowel infarction rate of 60 % and a 60 % mortality rate from continued bleeding or bowel infarction. They concluded that embolotherapy is not recommended if active extravasation is not identified angiographically. We hypothesized that the higher post-embolic bowel infarction and mortality rates in cases with no active extravasation are a result of embolization of uninvolved vessels or of extensive blind embolization. It is believed that as a rule more severe bleeding is best evaluated by angiography and recurrent intermittent bleeding is best evaluated with colonoscopy [10]. However, these two modalities are never effectively combined. In our institution, an attempt is first made to evaluate and treat all patients with acute colonic hemorrhage using colonoscopy. When endoscopic intervention fails to control hemorrhage, radiopaque clips are positioned as a guide at the bleeding source by colonoscopy, followed by TAE to attempt to stop the

bleeding. It is said that a weakness of colonoscopy is that it is relatively useless when performed without bowel preparation in patients with more significant bleeding [9, 10]; fortunately, all patients who underwent angiography also underwent bowel preparation, and radiopaque clips could be placed at the bleeding site via colonoscopy. Though a future study should address what to do when bowel preparation cannot be performed, it seems that CT angiography may be useful when bowel preparation cannot be performed. In this study, there were no patients who underwent CT scanning, but we think that this imaging modality will help to identify the bleeding site to some extent [26]. As in other reports [5, 8–19], in our series diverticular disease was the most common cause of bleeding. Diverticula develop at the site where the vasa recta penetrate the muscular wall of the colon [10] and diverticular hemorrhage generally has a single feeding vessel [8]. Actually, all active extravasation was seen from the vasa recta leading to the clips. Therefore, it is suggested that the vasa recta, which lead to the clips positioned at the bleeding source, may very likely correspond with the vessel responsible for the bleeding. Furthermore, this may also apply to bleeding other than diverticular hemorrhage. According to the literature, the outcomes of TAE for lower intestinal hemorrhage were estimated to be a 5.9–33.3 % incidence of rebleeding, 7.0–24 % incidence of ischemia with no cases of bowel infarction, and a 9.1–40 % Table 2 Summary of reported studies Studies

Years

#. of cases

Recurrent hemorrhage (%)

Evidence of ischemia (%)

Luchtefeld et al. [8]

2000

17

5.9

5.9

11.8

Fuaki et al. [9]

2001

27

11.1

7.4

11.1

Bandi et al. [11]

2001

35

34.3

24.0

22.9

DeBarros et al. [12]

2002

27

22.2

7.4

22.2

Gady et al. [13]

2003

10

30.0

10.0

40.0

Kuo et al. [14]

2003

22

13.6

4.5

9.1

d’Othee et al. [15]

2006

19

27.0

11.0

11.0

Hui-Chung et al. [18]

2013

22

18.0

9.0

23.0

Adusumilli et al. [19]

2013

71

15.0

2.0

7.0

Current study

2013

10

10.0

40.0

0.0

Patients requiring urgent colectomy (%)

123

Tech Coloproctol

incidence of emergency surgery (Table 2). However, Kuo et al. [14] reviewed the literature over the past decade and reported only a 43 % (62/144) incidence of objective follow-up, while our objective follow-up rate is 100 % (10/ 10). Therefore, the full extent of ischemic complications may be underestimated in the literature. Our outcomes were a rebleeding rate of 10 %, an ischemia rate of 40 % with no cases of bowel infarction, and a 0 % incidence of emergency surgery. The fact that our objective minor ischemia rate is higher than that in other studies is a problem that requires further clinical assessment, but we think that the ischemia rate could be reduced further with use of only microcoils as the embolic material. Although none of our patients with minor ischemic complications required further therapy and there were no major ischemic complications, we think that it is important to find a method of embolization capable of reducing the ischemic rate. A major problem with this study is that reliability is not yet sufficient since the sample size is small, i.e., six patients, and our method of treatment relies on the operator’s skill at endoscopic diagnosis and angiography.

Conclusions We show that TAE can be successfully performed without complications when bleeding cannot be stopped at colonoscopy provided a clip is placed at the site of bleeding to direct embolization. Continued research with a larger number of cases is required to demonstrate the safety and effectiveness of superselective TAE for colonic hemorrhage in cases without active extravasation. Conflict of interest

None.

References 1. Kubo A, Kagaya T, Nakagawa H (1986) Studies on complications of diverticular disease of the colon. Japan J Med 24:39–43 2. Bokhari M, Vernava AM, Ure T, Longo WE (1996) Diverticular hemorrhage in the elderly—is it well tolerated? Dis Colon Rectum 39:191–195 3. Peck DJ, Mcloughlin RF, Hughson MN, Rankin RN (1998) Percutaneous embolotherapy of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 9:747–751 4. Ledermann HP, Schoch E, Jost R, Zollikofer CL (1999) Embolization of the vasa recta in acute lower gastrointestinal hemorrhage: a report of five cases. Cardiovasc Intervent Radiol 22:315–320 5. Guy GE, Shetty PC, Sharma RP, Burke MW, Burke TH (1992) Acute lower gastrointestinal hemorrhage: treatment by superselective embolization with polyvinyl alcohol particles. Am J Roentgenol 159:521–526 6. Gordon RL, Ahl KL, Kerlan RK et al (1997) Selective arterial embolization for the control of lower gastrointestinal bleeding. Am J Surg 174:24–28

123

7. Nicholson AA, Ettles DF, Hartley JE et al (1998) Transcatheter coil embolotherapy: a safe and effective option for major colonic hemorrhage. Gut 43:79–84 8. Luchtefeld MA, Senagore AJ, Szomstein M, Fedeson B, Erp JV, Rupp S (2000) Evaluation of transarterial embolization for lower gastrointestinal bleeding. Dis Colon Rectum 43:532–534 9. Funaki B, Kostelic JK, Lorenz J et al (2001) Superselective microcoil embolization of colonic hemorrhage. Am J Roentgenol 177:829–836 10. Funaki B (2004) Superselective embolization of lower gastrointestinal hemorrhage: a new paradigm. Abdom Imaging 29:434–438 11. Bandi R, Shetty PC, Sharma RP, Burke TH, Burke MW, Kastan D (2001) Superselective arterial embolization for the treatment of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 12:1399–1405 12. DeBarros J, Rosas L, Cohen J, Vignati P, Sardella W, Hallisey M (2002) The changing paradigm for the treatment of colonic hemorrhage. Dis Colon Rectum 45:802–808 13. Gady CJ, Reynolds CH, Blum A (2003) Selective arterial embolization for control of lower gastrointestinal bleeding: recommendations for a clinical management pathway. Curr Surg 60:344–347 14. Kuo WT, Lee DE, Saad WE, Patel N, Sahler LG, Waldman DL (2003) Superselective microcoil embolization for the treatment of lower gastrointestinal hemorrhage. J Vasc Interv Radiol 14:1503–1509 15. d’Othee BJ, Surapaneni P, Rabkin D, Nasser I, Clouse M (2006) Microcoil embolization for acute lower gastrointestinal bleeding. Cardiovasc Intervent Radiol 29:49–58 16. Ahmed TM, Cowley JB, Robinson G et al (2010) Long term follow-up of transcatheter coil embolotherapy for major colonic hemorrhage. Colorectal Dis 12:1013–1017 17. Mensel B, Kuhn JP, Kraft M et al (2012) Selective microcoil embolization of arterial gastrointestinal bleeding in the acute situation: outcome, complications, and factors affecting treatment success. Eur J Gastroenterol Hepatol 24:155–163 18. Teng HC, Liang HL, Lin YH et al (2013) The efficacy and longterm outcome of microcoil embolotherapy for acute lower gastrointestinal bleeding. Korean J Radiol 14:259–268 19. Adusumilli S, Gosselink MP, Ctercteko G et al (2013) The efficacy of selective arterial embolization in the management of colonic bleeding. Tech Coloproctol. doi:10.1007/s10151-0131088-6 20. Billingham RP (1997) The conundrum of lower gastrointestinal bleeding. Surg Clin North Am 77:241–252 21. Giacchino JL, Geis WP, Pickleman JR, Dando DV, Hadcock WE, Freeark RJ (1979) Changing perspective in massive lower intestinal hemorrhage. Surgery 86:368–376 22. Leitman IM, Paul AE, Shires GT (1989) Evaluation and management of massive lower gastrointestinal hemorrhage. Ann Sugr 209:175–180 23. Pennoyer WP, Vignati PV, Cohen JL (1996) Management of angiogram positive lower gastrointestinal hemorrhage: long term follow up of non operative treatments. Int J Colorectal Dis 11:279–282 24. Goldberger LE, Bookstein JJ (1977) Transcatheter embolization for treatment of diverticular haemorrhage. Radiology 122:613–617 25. Burgess AN, Evans PM (2004) Lower gastrointestinal hemorrhage and superselective angiographic embolization. ANZ J Surg 74:635–638 26. Duchat F, Soyer P, Boudiaf M et al (2010) Multi-detector row CT of patients with acute intestinal bleeding: a new perspective using multiplanar and MIP reformations from submillimeter isotropic voxels. Abdom Imaging 35:296–305

Management of angiogram-negative acute colonic hemorrhage: safety and efficacy of colonoscopy-guided superselective embolization.

We evaluated the efficacy and safety of superselective embolization with assistance of colonoscopy for acute colonic hemorrhage...
536KB Sizes 1 Downloads 0 Views