Selective Spleen Embolization of Splenomegaly to Improve Thrombocytopenia Facilitating Open Aortic Aneurysm Repair

Introduction

Abdominal aortic aneurysm (AAA) above 5.5 cm mandates elective repair in order to avoid the disastrous complications of rupture. ^{1 –3} Elective open surgical repair can be associated with mortality rates of 3% to 5% as well as a need for blood transfusion and lengthy intensive care unit and hospitalization. ⁴ Endovascular aneurysm repair (EVAR) is less invasive and has been shown to have a reduced perioperative mortality and morbidity compared to open surgery. ^4,5 Importantly, EVAR requires anatomic criteria to be met in order to achieve an adequate seal. Some studies report that currently only 65% of patients may be suitable for EVAR based on their anatomy. ^6,7 Other conditions may also play a significant role in deciding type of surgical intervention. Hematologic diseases, although only rarely encountered in patients in need of aortic surgery, may have a major impact in surgical outcomes since they often cause severe anemia and predispose to infection and/or bleeding disorders due to thrombocytopenia and neutropenia. ^8,9 Certainly, EVAR would be preferable in cases of patients with AAA and concomitant hematologic disorders, although open surgery must be considered when patient anatomy is not suitable for endovascular treatment.

Hairy cell leukemia (HCL) represents a kind of lymphoid neoplasm where hypersplenism impairs the body’s ability to produce antibodies against encapsulated microorganisms and predisposes to sepsis. ^{8 –11} Splenomegaly is frequently encountered in these patients and may cause thrombocytopenia which in turn increases the risk of bleeding, especially if the patient is in need of major surgery. These patients often need an open surgical splenectomy or splenic artery embolization to achieve an increase of platelet count. Superselective embolization of splenic arteries to reduce the spleen size, rather than removing or completely embolizing it, seems especially attractive as the immune system is not compromised.

We present a patient with an AAA measuring 11 cm in maximum diameter, which was unsuitable for endovascular treatment. He was also having HCL and subsequent splenomegaly as well as severe thrombocytopenia. We report the successful treatment of this patient with supraselective intra-arterial embolization of 2 splenic artery branches, which resulted in normalization of platelet count and enabled successful open surgical repair of the AAA.

Case Description

An 82-year-old man having an 11-cm AAA presented for an elective aneurysm repair procedure (Figure 1A). The patient did not have any other comorbidities. Routine preoperative blood tests revealed a thrombocytopenia (20 000 platelets (PLT)/μL), white blood count of 3800 /L (Neut: 77.5%; Lymp: 12.3%), and hematocrite of 41.9% with a hemoglobin of 13.6 g/dL. Computed tomography (CT) imaging showed the spleen measured to 22.7 cm in length.

OPEN IN VIEWER

Figure 1.

Computer tomography revealed a gigantic abdominal aortic aneurysm (AAA) and the presence of a severely angulated and short AAA (hostile neck anatomy) contradicting endovascular treatment.

The patient had been hospitalized 1 year previously for treatment of HCL of variant type (CD25, CD103) and received 8 cycles of chemotherapy with pentostatin (5 mg once every 15 days). The last cycle was administered 6 months ago. At that time, the aneurysm had a diameter of 8.7 cm. No recurrence of disease was noted during follow-up. The spleen then measured 19 cm in length at that time.

Due to challenging aneurysm anatomy, endovascular repair was excluded (Figure 1B). Specifically, following 3-dimensional reconstruction of the CT images, a short (5 mm) and angulated (70°) aneurysm neck was obtained, which also presented intraluminal thrombus across its circumference and precluded the treatment of this AAA by endovascular means available to us at our institution. Due to the marked thrombocytopenia and the associated splenomegaly, priority was given to the treatment of hypersplenism. Following obtained consent, the patient underwent 2 sessions of supraselective intra-arterial embolization of 2 splenic artery branches with contour embolization particles, 250 to 335 μm in size (in penicillin solution). This resulted in isolation of the lower half of the spleen (Figure 2). Before each session and for the following 3 days, the patient received antibiotics (Netromycin 100 mg and Penicillin 1 000 000 IU). In order to prevent pneumococcal infection, a polyvalent pneumococcal vaccine was administered before the procedures. After each embolization, pain control was achieved with nonopioid analgesics. Three weeks later, a new abdominal ultrasonogram demonstrated a diminished spleen size now measuring 12.5 cm length, and a blood work revealed a normalization of platelet count (280 000 PLT/μL) allowing us to proceed with open surgical repair.

OPEN IN VIEWER

Figure 2.

Supraselective splenic artery embolization. A, Before embolization. B, After embolization, where a decrease in splenic vessel opacification is observed.

A conventional midline incision was performed, and transperitoneal access to the aneurysm was obtained. The infrarenal aorta was clamped, and aneurysm repair with aneurysmatectomy and infrarenal interposition of tube graft and remodeling of the orifice of the common iliac arteries were followed using a Dacron tube graft.

The patient’s postoperative recovery was uneventful with no signs of splenic abscess, septicemia, splenic vein thrombosis, or unremitting bronchopneumonia. The patient was discharged on the 10th postoperative day. A follow-up blood cell count at 3 months, 6 months, and a year after showed a normal platelet count (≥200 000 PLT/μL; Figure 3).

OPEN IN VIEWER

Figure 3.

Follow-up computed tomography angiography (CTA) revealing good graft patency.

Discussion

Hairy cell leukemia is a chronic lymphoid leukemia, initially described in 1958 by Bouroncle and colleagues. ¹⁰ It is a B-cell disease and has been named for the characteristic hair-like cytoplasmic projections on the malignant cell’s surface. ^10,11 The most common symptoms are related to bone marrow failure, as expected. Specifically, weakness and fatigue are most frequently reported resulting from anemia. Approximately, one-third of patients develop bleeding from thrombocytopenia, and another one-third have fever and infections from neutropenia. ^10,11 Splenomegaly has been reported to be present in as many as 96% of patients, although this is symptomatic in far fewer. ⁸ This increases the spleen’s mechanical filtering and subsequently, the destruction of erythrocytes and often that of white blood cells and platelets. As a result, compensatory bone marrow hyperplasia may occur in those cell lines that are reduced in the circulation. Moreover, massive splenomegaly has been suggested to be associated with a worse prognosis while several studies have observed that severely anemic patients and those with severe thrombocytopenia may also have a worse outcome. ⁸ Therefore, most authors would recommend treatment to be initiated when a declining trajectory predicts that the patient will reach a platelet count >100 000/L or an absolute granulocyte count consistently >1000/L. ⁹ The introduction of the purine nucleoside analogs transformed this disease into a highly treatable form of leukemia, and patients with the classic form of this rare leukemia currently present a near normal life expectancy. ⁸

Although sporadic cases of hematological malignancies associated with surgical pathologies in general and particularly with abdominal or thoracic aneurysms have been reported, ^{12 –14} there are no available data describing treatment strategies in cases in which a huge aortic aneurysm coexist with symptomatic splenomegaly. In our case, an 11-cm large AAA would be expected to present a >50% annual rupture risk, and therefore, its repair was absolutely indicated in a semiacute setting and despite the high operative risk. ¹⁵

There are several treatment options for patients presenting with splenomegaly and thrombocytopenia. Splenectomy is indicated in hematological diseases with resistant thrombocytopenia despite optimum medical therapy and has been reported to achieve 66% complete response rate. ¹⁶ On the other hand, this procedure is accompanied by 1% immediate postoperative mortality when performed via laparotomy, which is reduced to 0.2% after laparoscopic splenectomy. ¹⁶ In addition to perioperative complication rates of 13%, the long-term risk of fatal infection attributed to the absence of spleen may also be significant. ¹⁷ Others argue that a partial splenectomy may represent a better alternative since it achieves the same objectives without the adverse side effects of the total splenectomy. ^18,19 Furthermore, the benefits may be transient following this procedure because the spleen regenerates and the clinical/laboratory findings can relapse. In this regard, the subtotal splenectomy, preserving the superior splenic pole supplied only by the splenogastric vessels, has also been suggested. ^18,19 Moreover, endovascular options, such as total or selective splenic artery embolization, for reducing spleen reactivity also exist. A 51% response rate has been reported for selective splenic artery embolization with similar relapse rate to that seen after splenectomy. ²⁰ Although many reports strongly suggest that complete removal of functioning spleen is essential in hematologic diseases, there are data arguing that platelet response after selective embolization was not affected by the percentage of splenic infarction. Furthermore, avoidance of complete spleen infarction may reduce the complication rate. ^{20 –22} Indeed, complications have been shown to be related to the amount of spleen embolized, with volumes >70% associated with a higher risk of morbidity and mortality. Outcomes have improved with the use of partial rather than total embolization. ^{23 –26}

Taking into account technical considerations, one should bear in mind that branches of the splenic artery are end arteries, and therefore, occlusion beyond the hilum usually results in infarction of the respective part of the splenic parenchyma. ²⁷ We have used embolic particles instead of coils or plugs because this results only in occlusion of splenic branches in a “patchy” manner instead of causing complete obstruction of the prominent feeding artery. In this regard, we anticipated that the spleen would not be completely destroyed, but rather it would reduce in size and prevent or protect against platelet destruction. Others have also discouraged use of coils and advocate using Gelfoam in order to achieve selective spleen embolization. ^20,28 Moreover, we have targeted the lower pole vessels in order to avoid the development of a left pleural effusion as others have suggested. ²⁹ In our patient, an endovascular approach with the use of elective embolization was preferred due to its less invasive nature, also taking into account the patient’s severe thrombocytopenia. Additionally, with this technique, we attempted to preserve a postembolized splenic segment, which would continue to protect against serious infections, particularly in an immunocompromised patient.

Totally, percutaneous EVAR under local anesthesia has been proven advantageous in routine use with newer low-profile endografts. ³⁰ The EVAR would have been the treatment of choice in this patient as well since it would offer a reduced bleeding risk, especially if performed percutaneously. Unfortunately, EVAR was not indicated for this patient since it was a juxtarenal aneurysm with extensive neck angulation (>70°). Fenestrated EVAR would have been a reasonable solution, but due to lack of experience, this was avoided taking in mind the more extensive procedure which would be needed in case of technical failure. ³¹ The same holds true for other endovascular options for treatment of juxtarenal AAAs, such as the use of endostaples or snorkel/chimney techniques, where again the lack of sufficient experience with these modalities at that time favored the open surgical treatment option. It should be emphasized that apart of the patient’s age, there were no other relative or absolute contraindications to open or endovascular repair. As such, a staged open surgical repair was performed following selective splenic embolization and resultant normalization of platelet count.

Conclusion

Selective splenic embolization in patients with hypersplenism and subsequent thrombocytopenia who require major open aortic surgery may achieve significant rise in platelet count and optimize patient’s preoperative status in order to avoid bleeding complications and allow for safe and successful open aortic surgery.