Idiopathic esophageal submucosal hematoma after flow diverter stent placement for an unruptured intracranial aneurysm: illustrative case

Abstract

BACKGROUND

Flow diverter stent placement for unruptured intracranial aneurysms requires intensive antithrombotic therapy, which can lead to unexpected hemorrhagic complications. However, standard protocols for managing such conditions have not been established. A rare case of idiopathic esophageal submucosal hematoma after flow diverter stent placement, successfully managed with the continuation of antithrombotic therapy, is reported.

OBSERVATIONS

A 76-year-old woman had undergone coil embolization 12 years earlier for a ruptured right paraclinoid internal carotid artery aneurysm. Follow-up imaging showed recurrence of the aneurysm, and she was admitted for flow diverter stent placement. She received dual antiplatelet therapy (aspirin and prasugrel) for 2 weeks prior to treatment, along with appropriate systemic heparinization during the procedure. One hour after treatment, she developed severe chest pain. Contrast-enhanced CT imaging showed an esophageal submucosal hematoma without active bleeding. Conservative treatment was started without discontinuing antiplatelet therapy, resulting in the near-complete resolution of the hematoma on CT imaging by day 11, and she was discharged on day 21.

LESSONS

Although rare, idiopathic esophageal submucosal hematoma should be considered a potential complication when a patient complains of chest pain after endovascular treatment. When antiplatelet therapy is required and no active bleeding is present, continuation of antiplatelet therapy can be a viable option under careful observation.

https://thejns.org/doi/10.3171/CASE25495

Flow diverter stent placement for unruptured intracranial aneurysms requires potent antithrombotic therapy during the perioperative period. In contrast, a spontaneous esophageal submucosal hematoma is often caused by the rupture of submucosal vessels due to a sudden increase in intraluminal esophageal pressure.¹ Antithrombotic therapy is a well-recognized risk factor for such hematomas.²

Whereas esophageal submucosal hematomas associated with endovascular procedures requiring antithrombotic therapy have been sporadically reported, their management remains challenging, since it requires balancing the risks of hematoma enlargement and thrombotic complications.^3–9 Currently, no standardized treatment protocol exists for these cases.

A case of esophageal submucosal hematoma occurring after flow diverter stent placement for a recurrent unruptured intracranial aneurysm, successfully managed with the continuation of antithrombotic therapy, is reported. In addition, the relevant literature is reviewed to provide insights into this rare but significant complication, with a particular focus on management strategies.

Illustrative Case

A 76-year-old woman presented with a history of subarachnoid hemorrhage 12 years earlier. Imaging showed aneurysms in the right paraclinoid internal carotid artery (ICA) and right middle cerebral artery (MCA), with the rupture attributed to the paraclinoid ICA aneurysm (Fig. 1). Coil embolization was performed for both aneurysms, resulting in a neck remnant in the ruptured ICA aneurysm and complete occlusion in the MCA aneurysm. She had slight hemiparesis on the left side as a sequela and was discharged home (modified Rankin Scale [mRS] score of 3).

FIG. 1.

A: At the time of the initial treatment 12 years ago, 3D digital subtraction angiography (DSA) revealed aneurysms in the right paraclinoid ICA and right MCA. B: Coil embolization was subsequently performed on both aneurysms. The ruptured right paraclinoid ICA aneurysm was treated with a neck remnant, while the right MCA aneurysm was completely occluded. C: During follow-up, 3D-DSA demonstrated a recurrence of the right paraclinoid ICA aneurysm. D:A flow diverter stent was placed. The stent demonstrated good apposition, and the procedure was successfully completed.

Follow-up MRI showed recurrence of the previously ruptured paraclinoid ICA aneurysm. The patient was subsequently admitted for flow diverter stent placement. Two weeks prior to the procedure, dual antiplatelet therapy (DAPT) with aspirin (100 mg) and prasugrel (3.75 mg) was started. On admission, she showed no significant changes in her condition. Preoperative platelet function tests confirmed optimal antiplatelet effects, with a P2Y12 reaction unit (PRU) of 115 and aspirin reaction unit (ARU) of 383.

The procedure was performed under general anesthesia. At the beginning of the procedure, the baseline activated clotting time (ACT) was 120 seconds, and the target ACT was set at twice the baseline value (approximately 240 seconds). A bolus of 3000 units of heparin was administered, resulting in an increase in ACT to 247 seconds. Thereafter, 1000 units of heparin was administered every hour to maintain the ACT between 240 and 260 seconds throughout the procedure. Right femoral artery cannulation was done, followed by insertion of a 5-Fr Axcel Guide (Medikit). A 5-Fr Phenom Plus catheter (Medtronic), 2.7-Fr Phenom microcatheter (Medtronic), and Chikai 0.014-inch microguidewire (Asahi Inteccn) were used to construct a triaxial system, which was navigated to the appropriate position. A 4.25 × 20–mm Pipeline embolization device (Medtronic) was chosen for placement at the terminus of the right ICA. Cone-beam CT imaging confirmed good apposition of the stent to the parent artery. Heparin reversal was not performed. During the perioperative period, tracheal intubation was carefully performed under direct visualization using a video laryngoscope, and extubation was uneventful. No vomiting or coughing was observed, and no findings suggestive of bleeding or mucosal injury in the oral cavity or pharynx were noted during airway management.

One hour postoperatively, she complained of severe chest pain. Her blood pressure was 160/90 mm Hg, with no other vital sign abnormalities. Considering her complaints, myocardial infarction, aortic dissection, pulmonary thromboembolism, pneumothorax, and esophageal rupture were suspected. Blood tests, an electrocardiogram, and echocardiography showed no abnormalities. Chest and abdominal contrast-enhanced CT angiography showed no signs of aortic dissection or pulmonary thromboembolism. However, significant wall thickening with poor contrast enhancement was observed along almost the entire length of the esophagus, resulting in luminal narrowing and displacement (Fig. 2). Extravasation was not observed. Based on the clinical and imaging findings, the patient was diagnosed with an esophageal submucosal hematoma, and she was managed with nothing by mouth. Since no evidence of active bleeding was observed on CT imaging, the risk of thrombotic complications associated with flow diverter stent placement was carefully assessed. Given the high risk of thrombosis, DAPT was continued without reduction or discontinuation, and the medications were administered orally in a crushed form with a small volume of water. Close imaging follow-up was also conducted.

FIG. 2.

Chest imaging tracked the progression of the esophageal submucosal hematoma. A: Contrast-enhanced CT scan obtained 1 hour postsurgery showing significant thickening of the esophageal wall (line) and narrowing of the lumen (dotted line), with no extravasation. B–D: Plain CT scans obtained at 3 hours (B), 1 day (C), and 4 days (D) postoperatively showing no hematoma enlargement. E: By day 11, the hematoma had decreased in size.

CT on postoperative day (POD) 4 showed no enlargement of the hematoma, and upper gastrointestinal endoscopy on POD 5 confirmed no active bleeding (Fig. 3). The patient resumed oral intake on POD 6, and her nutritional status stabilized. She was discharged home with an mRS score of 1 on POD 21 without episodes of neurological ischemic complications. Two months postoperatively, follow-up upper gastrointestinal endoscopy showed resolution of the hematoma. Six months after surgery, MRI demonstrated complete resolution of the aneurysm with no thrombotic complications.

FIG. 3.

Upper gastrointestinal endoscopy revealed the course of the esophageal hematoma. A: On the 5th day postsurgery, endoscopy showed a hematoma extending along the right side of the esophageal wall (arrowhead), with no signs of active bleeding. B: By 2 months postsurgery, endoscopy confirmed complete resolution of the hematoma.

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

The rare condition of esophageal submucosal hematoma was first described in 1957 by Williams, and it can be categorized into two types: traumatic, caused by foreign bodies or mechanical injury such as endoscopy; and spontaneous, resulting from vascular rupture in the mucosal layer due to increased intraluminal pressure from factors such as vomiting or eating.^1,10 These hematomas are more likely to occur in patients with a bleeding tendency, such as those undergoing antithrombotic therapy or dialysis.² Clinically, patients often present with chest pain or hematemesis, and the prognosis is generally good, with most cases resolving spontaneously. Although quite rare, there have been reports of cases progressing to hemorrhagic shock.³

In the literature reviewed, 10 cases of esophageal submucosal hematoma complicating cerebral endovascular treatment were identified, including the present case (Table 1).^3–9 All cases involved unruptured cerebral aneurysms, with 8 cases undergoing coiling and 2 cases undergoing flow diverter stent placement. Six cases presented with chest pain or epigastric pain, and hematemesis was observed in 6 cases. Although some cases required transfusion, most patients resumed oral intake within approximately 1 week, and all cases had good outcomes. Two cases were likely related to complications such as extubation or vomiting immediately after surgery, whereas the remaining 8 cases were likely related to the administration of antithrombotic therapy. In the present case, tracheal intubation was performed using a video laryngoscope, and no mechanical trauma related to extubation or nasogastric tube insertion/removal was observed. Additionally, no vomiting that could have increased intraluminal esophageal pressure was noted. Therefore, it is highly likely that antithrombotic therapy contributed to the development of the esophageal submucosal hematoma.

TABLE 1.

Reported cases of esophageal submucosal hematoma with neuroendovascular treatment

Case No.	Authors & Year	Age (yrs), Sex	Antiplatelet	Surgery	Symptoms	Antiplatelet Therapy	Blood Transfusion	Cause	Resumption of Meals	Discharge	Outcome
1	Ito et al., 20173	75, F	A+C	Coil	Chest discomfort, hematemesis	Discontinued	Yes	Coughing on suction	POD7	POD21	Good
2	Yamashita et al., 20004	67, F	None	Coil	Nausea, hematemesis	Discontinued	No	Unknown	Unknown	POD16	Good
3	Yoshihara et al., 20135	70, F	A+C	Coil	Epigastric pain	Discontinued	No	Postop vomiting reflex	POD10	POD16	Good
4	Fujimoto et al., 20166	70, F	A	Coil	Epigastric pain, nausea	Discontinued	No	Unknown	Continuation	POD13	Good
5	Fujii et al., 20197	65, F	A+C	Coil	Hematemesis	Discontinued	Yes	Unknown	Unknown	Unknown	Good
6	Fujii et al., 20197	73, F	A+C	Stent-assisted coil	Epigastric pain hematemesis	Discontinued	Yes	Unknown	Unknown	Unknown	Good
7	Fujii et al., 20197	65, F	A+C	Stent-assisted coil	Epigastric pain	Continued (reduced dose)	No	Unknown	POD6	Unknown	Good
8	Oba et al., 20228	73, F	A+C	Stent-assisted coil	Chest discomfort, hematemesis	Discontinued	Yes	Unknown	POD10	POD33	Good
9	Takeyama et al., 20199	75, F	A+C	Flow diverter	Chest pain, hematemesis	Discontinued	Yes	Unknown	POD11	POD27	Good
10	Present case	76, F	A+P	Flow diverter	Chest pain	Continued (same dose)	No	Unknown	POD6	POD21	Good

A = aspirin; C = clopidogrel; P = prasugrel.

Antithrombotic therapy plays a critical role in cerebral endovascular treatment. Such procedures carry a risk of thrombotic complications, which may lead to severe sequelae.¹¹ To mitigate this risk, preoperative antiplatelet therapy is typically managed with DAPT due to drug resistance issues, and systemic heparinization is commonly performed during the procedure.¹² In this case, prasugrel was chosen instead of clopidogrel or ticagrelor. Clopidogrel was considered suboptimal due to the high prevalence of CYP2C19 loss-of-function polymorphisms among the Japanese population, which can reduce its metabolic activation and antiplatelet efficacy. Ticagrelor, although recognized internationally as an effective option, is not approved in Japan and therefore was not available for use. While prasugrel is generally associated with a higher risk of bleeding compared with clopidogrel, it was selected to prioritize the prevention of thrombotic complications in this patient.¹³ The usefulness of ARU and PRU in evaluating platelet inhibition has been well reported. Although there are no standardized cutoff values, an ARU > 550 is regarded as hyperresponsive, and a PRU within 95–208 is considered a normal response in clinical practice in Japan.¹⁴ For anticoagulation monitoring during procedures, the ACT is used. Although a universally accepted target range has not been established, most reports recommend maintaining the ACT at approximately twice the baseline value, or within a range of 250–300 seconds.^15–17 In the present case, the baseline ACT was 120 seconds and the target value was set at 240 seconds; throughout the procedure, the ACT was maintained between 240 and 260 seconds, allowing the operation to be completed under appropriate anticoagulated conditions. However, it should be noted that, even with appropriately controlled antithrombotic therapy within the abovementioned ranges, an esophageal submucosal hematoma, as seen in the present case, may still occur.

There are no clear guidelines regarding the optimal duration of antiplatelet therapy following an endovascular procedure, and none for the appropriate length of discontinuation in cases of bleeding complications. Traditionally, when bleeding complications occur, antithrombotic therapy is decreased or discontinued to reduce the risk of hematoma expansion. Indeed, in all reported cases of esophageal submucosal hematoma associated with neuroendovascular therapy, except for the present case, antiplatelet therapy was either withdrawn or decreased. However, the decision to continue or discontinue antithrombotic therapy must be made on a case-by-case basis, carefully balancing the risk of thrombosis associated with discontinuation against the risk of bleeding associated with continuation. In the present case, flow diverter stent placement had a high risk of perioperative thrombotic complications, making it difficult to discontinue DAPT management. Therefore, close imaging monitoring was used to assess for active bleeding or hematoma progression, and DAPT management was maintained, resulting in a good outcome, with no significant hematoma enlargement or ischemic complications. To the best of our knowledge, this is the first reported case of esophageal submucosal hematoma following cerebral endovascular treatment that was successfully managed without interruption of DAPT.

Lessons

The present report describes a rare case of esophageal submucosal hematoma following flow diverter stent placement. It is important to recognize that, even with optimal antithrombotic therapy within the appropriate range, esophageal submucosal hematomas can occur following cerebral endovascular treatment. Furthermore, in cases in which an esophageal submucosal hematoma is detected without evidence of active bleeding, continuation of DAPT under close imaging follow-up may be a viable management strategy.

Acknowledgments

We thank Dr. W. Luba from Forte Inc. for editing a draft of this paper.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Hashimoto, Tateoka, Murase, Kinouchi. Acquisition of data: Shimura, Kinouchi. Analysis and interpretation of data: Shimura, Horiuchi, Kinouchi. Drafting the article: Shimura, Murase, Kinouchi. Critically revising the article: Yoshioka, Tateoka, Horiuchi, Kinouchi. Reviewed submitted version of manuscript: Yoshioka, Tateoka, Horiuchi, Murase, Kinouchi. Approved the final version of the manuscript on behalf of all authors: Hashimoto. Administrative/technical/material support: Fujita. Study supervision: Hashimoto, Kinouchi.

Correspondence

Koji Hashimoto: Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Chuo, Yamanashi, Japan. hkouji@yamanashi.ac.jp.