Endovascular Coil Migration to Duodenal Lumen After Gastroduodenal Artery Embolization in a 90-Year-Old-Woman, Complicating Upper-GI Bleed Management: A Case Report

Introduction

The most common cause of upper-gastrointestinal (GI) bleeds is peptic ulcer disease, usually due to an infection with Helicobacter pylori or the long-term use of NSAIDs [1]. Differential diagnosis of upper-gastrointestinal bleeding should include trauma, malignancy, and other erosive lesions. Clinical features of a bleeding peptic ulcer include hematemesis, melena, clinical features of anemia, and hypotension. Upper-gastrointestinal bleeding in the United States has an annual incidence of approximately 80 to 150 per 100 000 population, with an estimated mortality rate of 2% to 10% [2]. Upper esophagogastroduodenoscopy (EGD) is the standard method for diagnosing upper-GI bleeding [2]. First-line treatments for upper-GI bleeds typically include endoscopic placement of hemoclips, thermal coagulation, and epinephrine injections, although interventional radiology therapy may be indicated when endoscopy is unable to control the bleeding or there is rapid recurrence after endoscopic therapy [3]. Endovascular arterial coil embolization involves catheter directed placement of permanent stainless steel or platinum coils into the vessel to induce mechanical occlusion, promote platelet activation, and trigger the clotting cascade. Coil embolization has become a widely accepted treatment option for acute upper-gastrointestinal bleeds non-responsive to endoscopic therapeutic techniques. Technical success rates of endovascular coil embolization of upper-gastrointestinal bleeds range from 95% to 100%, and 50% to 80% achieve hemostasis [4]. Since the first transcatheter embolization for upper-GI tract bleeding in 1972, many technological advances have been made, particularly in endovascular management through coil embolization. Ischemia is the most common complication (7–16%), but other complications such as hemorrhage or reperfusion via retrograde filling or collaterals can also occur [4]. With a reported incidence of only 1% to 2%, endovascular coil migration for upper-GI bleeds from the placement site is an uncommon complication.

Migration can lead to recurrent bleeding, ulcer formation, or erosion into adjacent areas, underscoring its importance in post-procedural monitoring and awareness. Reported cases of endovascular coil migration are limited, and its clinical course and management strategies remain unclear. Herein, we present a case of a 90-year-old woman with an endovascular coil migration 13 days after undergoing coil embolization of the gastroduodenal artery (GDA), highlighting the potential for delayed complications even after technical successful embolization.

Case Report

INITIAL PRESENTATION (DAY ZERO):

A 90-year-old woman with a history of osteoporosis presented to the ED with acute epigastric abdominal pain, dizziness, and diarrhea. Initial laboratory assessment revealed a hemoglobin of 10.4 g/dl, which fell to 6.7 g/dl over the course of 2 days, and she received 3 units of packed red blood cells. Computed tomography angiography (CTA) revealed a duodenal bulb ulcer with surrounding phlegmon. There was no evidence of pneumoperitoneum or abscess, and the findings were characteristic of a contained perforation. Within the ulceration, there was focal contrast extravasation, raising concern for a pseudoaneurysm supplied by the gastroduodenal artery.

EMBOLIZATION PROCEDURE (DAY 2):

The patient underwent endovascular coil embolization of the gastroduodenal artery with platinum-based microcoils. No active extravasation was identified at the time. However, a pseudoaneurysm involving the middle to distal gastroduodenal artery was identified. The right femoral artery was accessed with a 21-gauge micropuncture needle, and a micro-guidewire was used to place a 5-Fr sheath. A 5-Fr Sos Omni catheter was advanced over a Bentson guidewire into the abdominal aorta, the celiac artery was selectively catheterized, and angiography was performed. The gastroduodenal artery was then selectively catheterized using a microcatheter and Fathom micro-guidewire. The microcatheter was then advanced to the gastroduodenal artery and coil embolization was performed.

IMMEDIATE FOLLOW-UP:

A follow-up computed tomography (CT) scan of the abdomen and pelvis with contrast was performed the same day after the embolization procedure and again showed phlegmon surrounding the distal stomach and duodenal bulb, with duodenal bulb ulceration, consistent with a contained perforation.

OUTPATIENT ENDOSCOPY (DAY 13):

Thirteen days later, the patient underwent an outpatient esophagogastroduodenoscopy (EGD), which revealed a coiled metallic object at the junction of the duodenal bulb and second portion of duodenum (Figure 1). After consulting another physician intra-procedurally, both physicians agreed to not remove endoscopically at the time due to risk of perforation and uncontrolled bleeding. The endovascular coil remained partially embedded in the duodenal wall (Figure 2). Surgical management in a hospital setting was considered as an option.

RE-PRESENTATION (6 WEEKS):

Six weeks later the patient re-presented to the ED with progressively worsening, sharp, moderate-to-severe right flank and right lower-quadrant abdominal pain, ongoing for the past 10 to 12 hours, with a 3-month history of intermittent symptoms. A CT abdomen and pelvis with intravenous (IV) contrast showed prominent accentuation and thickening of the gastric wall distally near the gastroduodenal junction. High-density material with associated artifact in the right upper quadrant was consistent with embolization material, given the patient’s prior gastroduodenal artery coil embolization.

A gastrointestinal (GI) upper series with small-bowel follow-through demonstrated passage of contrast material into the stomach and proximal duodenum without evidence of leakage. A surgeon assessed the condition as a contained perforated duodenal ulcer. With only localized findings, the patient was managed conservatively with a recommended nasogastric tube, nothing by mouth, IV fluids, antibiotics, and a proton pump inhibitor (PPI). The patient’s initial perforated duodenal ulcer was evident on imaging and required endovascular coil embolization. However, follow-up endoscopy did not provide a description of peptic ulcer disease, limiting direct endoscopic correlation with her recurrent symptoms.

FOUR-MONTH FOLLOW-UP:

At a 4-month follow-up, the patient was asymptomatic. Watchful waiting was recommended to allow for possible spontaneous endovascular coil migration. She had stopped taking the PPI and subsequently had nausea and vomiting, then resumed taking the PPI and the symptoms improved. She denied fevers or chills but had occasional mid-epigastrium pain, relieved with meals, and constipation. She was prescribed pantoprazole, a PPI, 40 mg 1 tablet orally (po) twice a day (bid) and an osmotic laxative, polyethylene glycol 3350, 17 g, ½ to 1 capful daily to every other day.

SUBSEQUENT FOLLOW-UP:

At further follow-up, she remained asymptomatic and the embolization coil had still not passed.

LATE PRESENTATION (1 YEAR 9 MONTHS):

A year and 9 months later, the patient presented to the ED with epigastric abdominal pain. Her symptoms had persisted for the last 2 months and were progressively getting worse. She was evaluated previously by her primary care provider as she was having severe epigastric abdominal discomfort. There was concern for a duodenal perforation and possible dislodgment of the intra-abdominal endovascular coil. She denied any shortness of breath, tachycardia, or palpitations, but did have nausea and vomiting. In the ED, she was given 2 mg of morphine, fluids, and ondansetron. EKG showed a normal sinus rhythm with a widened QRS. A chest X-ray showed an enlarged cardiac silhouette, and CT of the abdomen/pelvis showed no evidence of intra-abdominal endovascular coil dislodgement. Her vital signs included a blood pressure (BP) of 197/76, for which she was given amlodipine 10 mg. She was admitted overnight and showed no signs of peritoneal symptoms, hematochezia, hematemesis, nausea, vomiting, or constipation on re-evaluation the following morning. There was concern for worsening peptic ulcer disease despite compliance with PPI therapy, and an outpatient EGD was recommended. She was told to continue taking the PPI (pantoprazole) at the current dose and was medically assessed as stable for discharge home. To date, the endovascular coil has yet to pass.

Discussion

Since there is limited literature on endovascular coil complications, there are no standard recommendations on management of coil migrations after upper-GI bleeding. Our case contributes to the limited literature by illustrating a rare but known complication of endovascular coil migration into the duodenum that occurred within 2 weeks of embolization. This patient’s presentation underscores the importance of vigilance during follow-up, especially in patients with recurrent GI bleeding and signs and symptoms of GI bleeding. With no clear consensus on management, asymptomatic patients may be observed, while those with obstruction or bleeding may require multidisciplinary intervention.

The first-line treatment for upper-gastrointestinal bleeding is endoscopic hemostasis. However, severe upper-GI bleeding requires surgical intervention in 10% to 15% of patients. According to NICE guidelines, percutaneous embolization as salvage therapy can be performed [6]. Surgery can be associated with high operative mortality rates. Arterial embolization in the upper-GI tract is generally considered very safe due to the collateral blood supply to the stomach and duodenum [1]. The most notable complication is rebleeding, which can be managed using repeat angiography and embolization. Coagulopathy is a significant risk factor for rebleeding and mortality [7]. There is a very limited literature on endovascular coil migration following embolization of upper-GI bleeds. The pathogenesis of endovascular coil erosion is not fully understood, but it is believed to involve persistent inflammation or infection at the coil site, possibly combined with high arterial pressure or ulcer recurrence, which can weaken local tissue integrity and predispose to erosion [8,9]. Additionally, mechanical stress from arterial pulsations may contribute to endovascular coil displacement, particularly in the setting of pseudoaneurysms, where the vascular wall is already compromised [8]. Endovascular coil migration can also be influenced by the complexity of the vascular anatomy and the arterial supply to the ulcer or lesion. A coil may migrate into a pseudoaneurysm, presenting as rebleeding or other complications.

Case reports of migrated endovascular coils have demonstrated a wide range in symptom severity and outcomes such as life-threatening duodenal ulcer hemorrhage successfully managed, as well as fatal outcomes [10,11]. Unlike prior cases with migration, our case is unique in its early asymptomatic presentation, with erosion occurring within 2 weeks of coil embolization, whereas most cases have been reported months later [12]. Some case reports have described removal of the endovascular coils in symptomatic patients when initial placement was insufficient [12]. Patients that are asymptomatic have been advised against prophylactic removal of coils, although no clear guideline has been established for symptomatic patients. Given that fewer than 20 cases have been documented worldwide, the literature is sparse regarding the optimal management of coil migration following upper-gastrointestinal bleeding embolization, particularly in asymptomatic cases such as ours [13]. With the limited literature demonstrating a wide variety of case presentation symptoms, each case should be individually addressed in collaborations with gastroenterology, interventional radiology, and surgery. Early multidisciplinary involvement may improve outcomes, and awareness is needed to prevent fatal outcomes [8].

Our case highlights the clinical dilemma between surgical and conservative approaches. Although endoscopic removal was considered, the risk of perforation and uncontrolled bleeding outweighed the potential benefit. The patient remained stable, which supports a conservative strategy with close follow-up. Practical recommendations can be drawn from this, such as monitoring, multidisciplinary evaluation, and a clear standard for when to intervene. Surgical or endoscopic removal should be reserved for patients who develop obstruction, perforation, or ongoing bleeding, whereas stable patients can be safely managed conservatively.

Conclusions

Endovascular coil embolization is a therapeutic option for upper-GI bleeding refractory to endoscopic intervention, and coil migration is a rare complication. Future studies should explore endovascular coil design modifications and predictive factors for migration. There is no established standard approach for managing endovascular coil migration. Therefore, clinicians should closely monitor symptomatic patients using serial imaging and repeat esophagogastroduodenoscopy (EGD). In hemodynamically stable patients, coil migration can often be managed conservatively with observation rather than immediate intervention. In unstable patients, surgery and interventional radiology should be consulted.