Arterioureteral fistula (AUF) is a rare yet life-threatening condition associated with a mortality rate of approximately 20% two decades ago [1]. Recently, advanced endovascular interventions have provided less invasive treatment options, significantly decreasing mortality rates. However, the difficulty in diagnosing AUF can result in delayed treatment, which may lead to catastrophic outcomes [2,3].

Predisposing factors for developing AUF include prolonged use of a double-J (DJ) stent, prior radiation therapy, and pelvic and vascular surgery. Due to these factors, adhesion develops between the artery and the ureter. Over time, the continuous pressure from the adjacent artery weakens the ureteral wall, leading to the formation of a fistula [1,4]. Therefore, in patients with multiple risk factors for AUF, a high degree of suspicion is necessary to diagnose AUF early and facilitate effective treatment. Here in, we present a case from which we can learn valuable lessons about the diagnosis and treatment of AUF.

This study was approved by the Institutional Review Board at Seoul National University Boramae Medical Center (20-2023-44). Informed consent was waived for this study by the Board.

An 84-year-old female presented to our emergency department complaining of lower abdominal pain that began the previous day. She had a history of intracerebral hemorrhage and cerebrovascular infarct and severe cognitive disorder. Abdominal computed tomography (CT) revealed bilateral internal iliac artery (IIA) aneurysms (diameter: left 55 mm, right 49 mm) with hydronephrosis and evidence of rupture on the left side (Fig. 1). The patient and their caregiver refused open repair considering the patient’s advanced age and poor general condition; therefore, the left IIA aneurysm was embolized, and a percutaneous nephrostomy catheter was inserted, which was later replaced by a DJ ureteral stent.

Figure 1. Computed tomography showed bilateral internal iliac artery aneurysms (white arrow) with rupture (dashed arrow) on the left side (A) and dilatation of the left ureter (white arrow) (B). Angiography after the embolization of left internal iliac artery aneurysm with coils (black arrow) (C).

Six weeks later, a retrograde pyelogram was performed to check the ureteral stent function and a small dye leakage adjacent to the left IIA aneurysm was observed (Fig. 2A). We recommended open repair for the ureter perforation, which was believed to have developed while manipulating a wire to pass through the stenotic site of the ureter; however, the patient again declined surgery.

Figure 2. Retrograde pyelogram demonstrated contrast leakage (black arrow) near the embolic material (A). Computed tomography showed the enlarged left internal iliac artery aneurysm (white arrow) (B) with arterial inflow (dashed arrow) from the common iliac artery (C).

The patient returned for a scheduled DJ change three months later, and while advancing the guide wire, active bleeding occurred. It was only then that AUF was clinically diagnosed for the first time. Computed tomography revealed that the left IIA aneurysm increased from 55 mm to 63 mm with the continuation of arterial inflow to the aneurysm from the common iliac artery (Fig. 2B, C). We performed standard endovascular aneurysm repair (EVAR) with right IIA embolization to mitigate the inflow issue to the left IIA aneurysm because of the short neck of the IIA aneurysm, a significant difference in diameter between the common and external iliac artery, and the presence of a contralateral IIA aneurysm. The patient was stable for 3 months following the EVAR. Although massive hematuria ceased after EVAR, per the recommendation of the urologist, we opted to insert a percutaneous nephrostomy tube instead of changing the DJ stent regularly, considering the bleeding risk. However, the patient cut the nephrostomy tube at home using scissors due to a severe cognitive disorder and then came to the hospital approximately three days later. We checked the tubogram, and the left renal vein was visualized, which was presumed to be injured when the patient pulled out the nephrostomy tube (Fig. 3A).

Figure 3. Tubogram via resected percutaneous nephrostomy showed the left renal vein (white arrow) (A). Computed tomography revealed the infected aortic stent graft (B) with increased fluid collection (white arrow) around the stent graft (C).

Two weeks later, the patient complained of fever, back pain, and hematuria. Computed tomography revealed complicated fluid collection around the stent graft near the renal arteries, which is indicative of infection, and the blood culture was positive for methicillin-resistant Staphylococcus aureus (Fig. 3B). Because the patient was hemodynamically stable except for fever, she was treated with vancomycin and rifampin for 2 weeks. However, the complicated fluid collection continued to evolve (Fig. 3C), necessitating the decision to convert to open repair.

We planned to do an explantation of the aortic stent graft and debridement of the infected aortic wall via a trans-abdominal approach. However, the severe infection involving the aorta walls and inflammatory adhesions made the dissection of the targeted vessels almost impossible; subsequently, after dividing the renal vein, the fragile aortic wall below the renal vein ruptured. Therefore, we clamped the supraceliac aorta and managed to isolate the suprarenal aorta and performed proximal anastomosis. Nevertheless, hemostasis was not achieved due to the friable aortic wall and coagulopathy associated with the bleeding from rupture. We tried to dissect the more proximal part of the aorta, but we could not control the catastrophic bleeding. Eventually, cardiac arrest occurred during surgery, and the patient died.

Arterioureteral fistula is a rare but potentially life-threatening condition. The common risk factors for AUF development include a prolonged ureteral stent, prior genitourinary/pelvic/vascular surgery, and radiation therapy [3]. An arterial aneurysm is an uncommon risk factor for AUF, accounting for approximately 2.5% of all cases; however, primary fistulae are seen almost exclusively as a complication of arterial aneurysms [2]. In our patient, a huge IIA aneurysm caused compression of the ureter, leading to chronic inflammation. Additionally, a DJ stent to treat hydronephrosis increased the friction between the ureter and the artery, resulting in AUF.

Diagnosing AUF can be challenging because of the rarity of the disease and the initial presentation of AUF, often to be nonspecific intermittent hematuria. Although angiography, ureteropyelography, CT scans, and ureteroscopy can be used for diagnosis, the sensitivity of these tests is approximately 50% or even lower [5,6]. Provocative angiography, which involves the manipulation of ureteral stents or vascular catheters at the site of AUF, has a higher sensitivity of 60%-100%, but massive bleeding can occur during examination [7]. Above all, a high level of suspicion is crucial for rapid diagnosis when patients with the aforementioned risk factors present with hematuria. We could diagnose AUF because we performed an unintentional provocative maneuver during DJ stent change, which induced massive bleeding. In retrospect, the diagnosis of AUF could have been made when the patient visited the ER with hematuria 3 months prior to the massive hematuria event.

Various treatment options are available for AUF, including open procedures such as local reconstruction, ligature with or without extra-anatomic reconstruction, auto-transplantation, and endovascular treatments such as embolization and covered stents [2]. Over the past few years, the treatment of AUF has shifted from open surgical approaches to endovascular methods [8-10]. Endovascular treatment provides several benefits over open surgery, including significantly lower mortality rates (4% vs. 9%) and availability in patients with a hostile abdomen or unstable hemodynamics [1]. It offers the advantage of minimizing damage to surrounding tissues, such as the ureter or bowel, which may occur more frequently in open surgery. Furthermore, it is possible to occlude the AUF while preserving prograde flow, eliminating the necessity for revascularization [9]. In our case, the IIA aneurysm was the cause of AUF. We initially thought coil embolization would adequately block the inflow to the aneurysm, but later we resorted to EVAR. Unfortunately, this was after additional damage had occurred to the ureter. Our case underscores the significance of adequate inflow control in treating AUF, as its persistent pulsatile flow contributed to the enlargement of the aneurysm, worsening the situation.

Endovascular therapy has several advantages; however, it is not a flawless technique. There is a possibility of AUF recurrence, as well as stent-graft thrombosis or infection. Due to the rarity of the disease itself, there is currently no available data on the incidence of these complications. There is always a theoretical risk of graft infection because AUF is a potentially infected site. In our patient, infection had serious consequences because we treated the AUF with an aortic stent graft, which resulted in the infection extending almost up to the level of the superior mesenteric artery. Notably, the infection was more likely to have entered through the severed percutaneous nephrostomy, rather than originating directly from the AUF itself. Therefore, in cases where a stent graft is used to treat AUF, maintaining the nephrostomy may not be advisable. Instead, treatments such as primary closure of the defect or urinary diversion could be more suitable. Furthermore, it is crucial to contemplate long-term antibiotic therapy or open repair, including infected stent graft removal with an extraanatomical bypass.

In conclusion, endovascular treatment for AUF is widely used in hemodynamically unstable patients or those with a hostile abdomen. It is crucial to adequately block the inflow to the AUF and consider the potential complications, such as stent graft infection. Long-term antibiotic therapy or open repair tailored to patients should be considered as part of the management approach.

None.

The authors have nothing to disclose.

Concept and design: all authors. Analysis and interpretation: HM. Data collection: SL. Writing the article: SL, HM. Critical revision of the article: IMJ. Final approval of the article: all authors. Statistical analysis: none. Obtained funding: none. Overall responsibility: HM.