Spontaneous aortic thrombosis, without atherosclerosis, aneurysm, or congenital coagulopathy, is a rare phenomenon due to the large diameter and rapid flow velocity of the abdominal aorta [1]. Given its uncommon occurrence, the incidence remains poorly established, and its presentation varies widely. This phenomenon ranges from asymptomatic chronic thrombosis to acute, catastrophic events causing end-organ ischemia, including limb, bowel, and renal ischemia.

Instances of spontaneous aortic thrombosis in the literature are exceedingly rare. However, existing case reports have identified malignancy and hematological disorders as provoking factors for aortic thrombosis [2,3]. Notably, several cases of aortic thrombosis have been reported in patients undergoing cisplatin-based chemotherapy [4].

Given the scarcity of literature on aortic thrombosis, there is currently no consensus regarding its optimal treatment. Reported treatment options include anticoagulation, thrombectomy, thrombolysis, and endovascular grafting [5,6].

We present a case of a patient with urothelial carcinoma treated with cisplatin and gemcitabine, who developed a spontaneous aortic thrombus at the level of the renal arteries. The patient underwent treatment with catheter-directed thrombolysis and subsequent thrombectomy. In this report, we also discuss the treatment considerations for this rare entity. Informed consent was obtained from the patient for this publication, and Institutional Review Board approval was waived due to the retrospective nature of the case report.

The patient was a 58-year-old Caucasian male with advanced upper tract urothelial cancer involving the left ureter, para-aortic, and pelvic lymph nodes, receiving gemcitabine/cisplatin chemotherapy. The chemotherapy regime consisted of gemcitabine (1,600 mg) and cisplatin (100 mg). The patient was a non-smoker with no known previous personal or familial thrombotic tendencies. Given the lack of concerning familial history, tests for hypercoagulability were not performed on this patient. He presented to the emergency department with an acute onset of abdominal pain five days after his last dose of chemotherapy. Physical examination revealed mild tenderness in the lower abdomen post-analgesia administration, with no signs of peritonitis. Bilateral lower limbs were warm with a capillary refill time of <3 seconds. Bilateral femoral, popliteal, posterior tibial, and dorsalis pedis pulses were well felt and graded as 3+.

Initial computed tomography (CT) mesenteric angiography, performed to investigate his acute-onset pain, revealed an abdominal aortic thrombus measuring 1.6×1.4×2.5 cm at the level of the renal arteries, extending into the origin of the left renal artery (Fig. 1). Both kidneys were generally well perfused, but patchy hypodesities were observed in the left kidney, possibly representing segmental infarcts. Notably, this thrombus was absent on a staging CT of the thorax, abdomen, and pelvis, performed just 13 days prior.

Figure 1. Initial axial and coronal computed tomography scan showing an abdominal aortic thrombus (arrows) at the renal artery level.

Intravenous (IV) heparin was initially administered. Following discussions with the urology team, interventional radiologist, hematologist, and the patient, a decision was made for catheter-directed thrombolysis, acknowledging the risk of distal embolic phenomena. Thrombolysis was chosen over surgery initially to attempt targeted treatment of the thrombus while avoiding the excessive morbidity associated with open surgery. A Cragg McNamara infusion catheter (Medtronic) was positioned across the length of the clot, from the distal suprarenal to proximal infrarenal abdominal aorta, via superficial femoral artery access (Fig. 2). Thrombolysis was performed with an alteplase infusion of 50 mg in 500 mL at 2 mg/hour. A repeat CT aortogram performed 10 hours later demonstrated only a marginal decrease in the size of the aortic thrombus to 1.4×1.1×1.9 cm (Fig. 3).

Figure 2. Cragg McNamara infusion catheter positioned from the distal suprarenal to proximal infrarenal abdominal aorta, across the length of the clot.

Figure 3. Repeat axial and coronal computed tomography scans were conducted after thrombolytic therapy. The scan revealed a smaller abdominal aortic thrombus (arrows) at the renal artery level.

Given the minimal improvement after thrombolysis and concerns regarding the risk of embolism from a shrinking thrombus, the institution’s aortic multidisciplinary team collectively recommended definitive surgical treatment, involving open femoral cutdown and aortic thrombectomy. This procedure was performed 2 days after the patient’s initial presentation and completion of thrombolysis.

During surgery, a left vertical groin incision was performed first, and the common femoral artery (CFA), superficial femoral, and deep femoral arteries were dissected out and slung. The left CFA was punctured with an 8Fr sheath, subsequently upsized to a 20Fr sheath to facilitate clot removal. Right groin access via ultrasound-guided puncture with a 6Fr sheath was also performed. Initial angiography revealed a thrombus between the right and left renal arteries in the aorta (Fig. 4). Both renal arteries were patent, but perfusion to the left kidney appeared poor. An 8×80 mm Mustang balloon (Boston Scientific) was passed up to the right iliac vessels and inflated to occlude the vessels, preventing distal embolization to the right limb during the trawling process. As the thrombus was assessed to be at the level of the renal arteries and unlikely to embolize into the left renal artery, they were left undisturbed. A Coda balloon (Cook Incorporated) was passed up the left groin into the aorta past both renal arteries, just above the aortic thrombus. The sheath was inserted entirely via the groin puncture, and its tip sat in the left common iliac artery, below the thrombus. The Coda balloon was inflated, and the clots were trawled to the position of the sheath, thereby “sandwiching” the clots before removal of the balloon and sheath together via the left groin. The left superficial and deep femoral arteries were clamped during the trawling process to prevent distal embolization. A completion angiogram demonstrated no further filling defects. The left CFA arteriotomy was closed, and the left groin was then closed in layers over a surgical drain. Bilateral dorsalis pedis and posterior tibial artery Doppler signals were good post-operatively.

Figure 4. Angiogram demonstrating an aortic thrombus at the level of the left and right renal arteries.

Our patient was monitored in the high-dependency unit post-operatively. IV heparin was resumed and titrated to an activated partial thromboplastin time of 50-70 seconds. Subsequently, it was switched to oral apixaban with plans for long-term anticoagulation to prevent recurrence. His subsequent recovery was otherwise uneventful, and he was discharged on postoperative day 7. The patient was seen in the clinic 3 weeks post-operatively, showing good distal pulses bilaterally. An ultrasound scan revealed no residual thrombus. The final histology of the aortic thrombus was negative for malignancy. Postoperatively, our patient completed 6 cycles of dose-reduced gemcitabine and cisplatin chemotherapy, as well as 4 cycles of avelumab chemotherapy. He tolerated these treatments well without significant adverse effects.

Previous studies have identified early atherosclerotic plaque at mural attachment sites [7], suggesting that atherosclerosis is a risk factor for arterial thrombosis. However, spontaneous aortic thrombosis in a normal aorta is typically associated with several other risk factors, including malignancy, prior surgery, or genetically predisposed hypercoagulability [2,3]. In our patient, the cause of aortic thrombus formation was likely multifactorial. He had active urothelial carcinoma and underwent a cisplatin chemotherapy regimen, both recognized as risk factors for aortic thrombus formation. Additionally, mild atherosclerotic disease was also observed on his CT scans.

Venous and arterial thrombosis are well-established side effects of cisplatin, with approximately 10% of patients undergoing cisplatin chemotherapy reporting vascular toxicity [8]. However, these thrombi are usually venous, and chemotherapy-induced arterial thrombosis has rarely been reported [4]. The mechanisms through which cisplatin causes vascular incidents remains uncertain, but several theories have been postulated. Cisplatin may induce endothelial damage by enhancing the tissue factor activity of monocytes in the blood. It could also cause thrombosis by activating platelets or elevating plasma levels of von Willebrand factor [9]. Approximately 45% of cisplatin-induced vascular events occur early, during the first two courses of chemotherapy [10]. In our patient, the aortic thrombus was discovered after his third cycle of chemotherapy.

Thromboembolic events may contribute to mortality in patients with cancer, with reduced survival rates observed in this group [11]. Currently, no consensus exists regarding the most ideal treatment for aortic thrombi. Conservative management with anticoagulation has been the primary treatment for stable patients, supported by case reports demonstrating thrombus resolution with anticoagulation alone [12]. However, a systematic review and meta-analysis suggests that aortic surgery may be preferable to anticoagulation alone, particularly for patients at high risk of recurrence with no other major contraindications for surgery [2].

Systemic or catheter-directed thrombolysis are minimally invasive treatment methods that have shown some success in isolated case reports. However, they carry a significant risk of distal embolization and systemic bleeding during treatment and do not guarantee complete thrombus removal [13]. In our patient, thrombolysis was chosen initially over surgery, as we aimed for targeted treatment of the thrombus while avoiding excessive morbidity associated with open surgery.

Open aortic surgery can be considered for good surgical candidates who either have experienced embolic events or present a significant clot burden, offering the benefit of complete thrombus extraction. However, the procedure is not without risk, with a reported mortality of 2.6% [14] and complication rates ranging from 28.9% to 71% [14,15]. A caveat is that in many cases, the poor premorbid condition and prognosis of patients with cancer may be prohibitive for aggressive surgical management.

In our case, the patient was managed with a hybrid approach. Catheter-directed thrombolysis was initially performed with partial success, followed by a subsequent open thrombectomy as the definitive treatment. This was achieved with bilateral groin access, trawling the thrombus to the sheath, sandwiching and retrieving it, thereby minimizing the risk of embolization. An open cutdown was performed for the left groin for several reasons. Firstly, it allowed for direct visualization of the thrombus and enabled distal control by clamping of the left superficial and deep femoral arteries, thus preventing embolization to the lower limb. Secondly, an open cutdown allowed sufficient space for the clot to be removed with the balloon, which would not be possible through arterial puncture alone. Finally, it facilitated proper repair of the artery after clot removal, given the substantial sheath size used.

In conclusion, spontaneous aortic thrombosis is a rare condition, with a poor prognosis if not promptly diagnosed and treated. Currently, no guidelines exist to direct its management, and each option carries its own set of benefits and drawbacks. We presented a case of abdominal aortic thrombosis at the level of the renal arteries in a patient with urothelial carcinoma undergoing cisplatin-based chemotherapy. While the thrombosis was partially treated with catheter-directed thrombolysis, it ultimately required an open aortic thrombectomy. Further studies would be beneficial to establish guidelines for the optimal management of aortic thrombosis and future follow-up.

None.

The authors have nothing to disclose.

Concept and design: SPC, JKC. Analysis and interpretation: none. Data collection: YLL, VMZ. Writing the article: VMZ, YLL. Critical revision of the article: SPC, JKC, VMZ, YLL. Final approval of the article: all authors. Statistical analysis: none. Obtained funding: none. Overall responsibility: JKC, SPC.