| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 All authors: Department of Diagnostic Radiology, Research Institute of Radiological Science, Yonsei University College of Medicine, Yonsei University, Yongdong Severance Hospital #146-92, Dogok-Dong, Kangnam-Gu, Seoul 135-270, South Korea.
Received June 6, 2003; accepted after revision September 30, 2003.
Address correspondence to K.-H. Lee
(doctorlkh{at}yumc.yonsei.ac.kr).
Percutaneous transcatheter arterial embolization has also become an accepted therapeutic option for the treatment of lower gastrointestinal bleeding, particularly in patients who are considered poor surgical risks [1, 2]. However, evidence exists that repeated embolizations or surgical interventions may prove necessary because bleeding may recur after an initially successful arterial embolization; this problem may affect as many as 33% of the patients who undergo embolization [2]. When gastrointestinal bleeding recurs after a successful initial treatment, preexisting collateral arteries are believed to be largely responsible. To our knowledge, the middle sacral artery has not previously been implicated as a collateral circulation source for rectal bleeding. We present a case of recurrent rectal bleeding after successful transcatheter arterial embolization that was caused by the development of collateral circulation from the middle sacral artery and discuss the significance of the middle sacral artery in the management of rectal bleeding.
Case History
A 61-year-old woman was hospitalized for intracranial hemorrhage. She had intermittent hematochezia the week before presentation, but no bleeding focus was identified on colonoscopy, small-bowel series, or technetium-99m RBC scan. However, on the 18th hospital day, a massive episode of hematochezia occurred that was immediately evaluated on angiography. The patient was hemodynamically unstable, with blood pressure of 80 over 40 mm Hg and tachycardia of 120 beats per minute. Laboratory findings were as follows: hemoglobin, 5.7 g/dL; hematocrit, 16.4%; platelet count, 56 x 103/µL; prothrombin time, 25.6 sec; and activated prothrombin time, 64.5 sec. She received a transfusion of 6 U of packed RBC. On initial angiography, extravasation of contrast material was noted in the distal rectum from the right superior hemorrhoidal artery, a branch of the inferior mesenteric artery (Fig. 1A). Selective transcatheter arterial embolization was performed using a mixture of 0.5 mL of N-butyl-2-cyanoacrylate (Histoacryl, B. Braun) and 1 mL of iodized oil (Lipiodol, Guerbet) via a 3-French coaxial Renegade microcatheter (Boston Scientific), and no other bleeding focus was identified (Fig. 1B).
|
|
Three hours after the first embolization, the patient underwent another angiography examination because of recurrent bleeding. The second angiogram showed recurrent contrast medium extravasation at the site of the previous bleeding that was being fed from the left superior hemorrhoidal artery and from both sides of the middle and inferior hemorrhoidal arteries. These feeders were successfully embolized with approximately 1 mm3 of Gelfoam gelatin sponge particles (Upjohn) as embolizing material. However, the patient experienced further rectal bleeding. A new angiogram of the pelvic vasculature revealed another contrast medium extravasation, this time from the middle sacral artery, arising at the aortic bifurcation (Fig. 1C). Accordingly, the middle sacral artery was successfully embolized with gelatin sponge particles and five pieces of microcoils (Cook) 2–3 mm in diameter and 2–4 cm in length, which finally resolved the hematochezia (Fig. 1D). No sign of bowel ischemia or recurrent bleeding was observed at follow-up. A benign healed ulcer at the rectum was seen on follow-up colonoscopy. The patient was discharged after she recovered from her intracranial hemorrhage.
|
|
Discussion
In a large series of colonic embolizations, the overall technical success rate was 93%, and prolonged clinical success was 81%, although several variations of embolization were performed as a result of the different embolizing materials used.
Embolization for lower gastrointestinal bleeding raises two main concerns. The risk of ischemia and infarction is the most significant, but technical advances such as the coaxial microcatheter and the presence of a collateral vascular supply have reduced the risk of ischemia, and studies have supported the safety of superselective transcatheter arterial embolization in lower gastrointestinal bleeding [1–3].
The other concern after successful transcatheter arterial embolization is recurrent bleeding. The rate of recurrent bleeding has been reported to be as high as 33%; several contributing factors have been proposed, including the choice of embolic material used and the presence of underlying disease [2]. Several reports have suggested that causes of recurrent bleeding may include vasospasm of feeder arteries, lesions bleeding at the time of interrogation, multifocal disease such as extensive diverticulosis or bowel metastases, collateral flow from adjacent vessels, and feeder vessels that were not identified on the initial angiography examination [4, 5]. Unfortunately, these causes have not yet been well evaluated. Most patients who experience recurrent bleeding after embolization are treated surgically, and imaging follow-up is not performed.
The rectum has a relatively abundant vascular supply that includes the superior hemorrhoidal arteries that branch from the inferior mesenteric artery and the middle and inferior hemorrhoidal arteries that branch from the internal iliac arteries. This abundance lowers the risk of ischemia or infarction after transarterial embolization, but it also raises the risk of recurrent bleeding after an initially successful procedure. Interventional radiologists must evaluate the feeding vessels thoroughly when treating rectal bleeding.
To date, the role of the middle sacral artery as a major contributor to recurrent rectal bleeding has received little attention. Anatomically, the middle sacral artery originates at the posterior surface of the aortic bifurcation and extends to the tip of the coccyx. In addition to the lowest lumbar and the right and left lateral sacral arteries, the middle sacral artery gives rise to a variable number of branches that pass forward beneath the peritoneum or sigmoid mesocolon to the rectum, where they anastomose with other rectal arteries [6]. The middle sacral artery is easily neglected because it is not routinely assessed in the management of rectal bleeding. As a result, little is known about its significance as a possible cause of rectal bleeding.
Various options are available for embolizing materials, which include coils, gelatin sponge particles, Histoacryl, and polyvinyl alcohol; the outcome also varies according to which material is chosen and how it is used [7]. However, no consensus exists in the literature about the choice of the embolizing agent. Histoacryl achieves rapid and complete embolization of the target artery, regardless of the vascular distribution and artery diameter, but its use is associated with a complication rate that is higher than that found with other agents because of the risk of embolizing nontargeted arteries [8]. A high rate of bleeding recurrence has been observed when polyvinyl alcohol particles or gelatin sponge particles are used alone. Gelatin sponge particles are used as temporary embolizing material and are relatively safe, but they are not conducive to rapid and complete embolization and the rate of recurrent bleeding is higher than with other materials. Use of several materials combined has been reported to be superior to the use of single materials [7].
We used Histoacryl for the initial embolization of the right superior hemorrhoidal artery because superselection of the bleeding focus appeared possible and we believed we could achieve rapid and complete embolization of bleeding without inadvertently embolizing any nontargeted arteries. After all, it was the middle sacral artery that was the collateral feeder and the source of rebleeding. However, numerous irregular small collateral routes existed between the middle sacral artery branch and the bleeding focus at the rectum (Fig. 1C), so superselection proved impossible. Therefore, we used microcoils to achieve hemostasis at the proximal part and gelatin sponges for the multiple distal branches.
Our strategy for angiography of lower gastrointestinal bleeding depends on the level of the suspected bleeding focus identified on colonoscopy or radioisotope scanning. For a bleeding focus in the left colon, we target the inferior mesenteric vasculature first and the superior mesenteric vasculature second. For a bleeding focus in the right colon or small bowel, we target the superior mesenteric vasculature first and the inferior mesenteric vasculature second. If we cannot determine the bleeding source, we target the inferior mesenteric artery first and the superior mesenteric artery second because bladder opacification and artifact may obscure a potential lesion. In rectal bleeding, our first target of angiography is the inferior mesenteric artery and our second target is the two internal iliac arteries.
The middle sacral artery usually is not well visualized on aortography and is difficult to catheterize. However, the middle sacral artery may be hypertrophied as a feeder when hypervascular tumors develop in the sacral or presacral space, or it may develop as a collateral route when atherosclerosis or a previous embolization causes circulation abnormalities in the pelvis or lower extremities. The middle sacral artery was hypertrophied in this patient, possibly because of atherosclerosis in the pelvic arteries resulting from a preexisting collateral route into the rectum. Therefore, we inferred that the middle sacral artery was acting as a collateral feeder or focus of recurrent rectal bleeding, and we easily catheterized it at the posterior wall of the aortic bifurcation.
In conclusion, we suggest that the middle sacral artery should be considered as a collateral feeder in cases of recurrent rectal bleeding.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
