Successful Treatment of Hypersplenism in Wilson's Disease by Partial Splenic Embolization

ABSTRACT

Aim: Hypersplenism can occur in patients with Wilson's disease (WD). Surgical splenectomy is a conventional treatment for this condition; however, emotional and neurological deterioration may follow splenectomy. In recent years, partial splenic embolization (PSE) has been increasingly performed as a nonsurgical alternative treatment for hypersplenism. The aim of this study was to evaluate the effectiveness and safety of PSE compared with splenectomy in the treatment of hypersplenism in WD patients. Methods: Fifty WD patients with hypersplenism were randomly divided into two groups (group A and group B), each including 25 patients. Patients in groups A and B were treated with PSE and splenectomy, respectively. Data were collected on the clinical efficacy of each procedure, adverse reactions, hematologic and blood chemistry test results, and abdominal computed tomography (CT) scan findings (group A only). Results: Marked improvements in the platelet and leukocyte counts after PSE and splenectomy were observed in all patients. PSE was associated with improved liver function without severe complications, and no significant changes in emotional and neurological symptoms were observed. In contrast, seven WD patients suffered neurological deterioration after splenectomy. Conclusions: Hypersplenism in WD patients was successfully treated by PSE, which appears to be a safe and effective alternative treatment for WD-induced hypersplenism.

Keywords::

• partial splenic embolization
• splenectomy
• hypersplenism
• Wilson's disease
• neurological deterioration
• sodium 2,3-dimercapto-1-propane sulfonate (DMPS)

Introduction

Wilson's disease (WD) is an autosomal recessive disorder affecting copper metabolism. Toxic copper accumulation in various tissues and organs may contribute to neurological, psychiatric, and hepatic symptoms [1–4]. Hypersplenism, resulting from cirrhosis, characterized by splenomegaly, thrombocytopenia, leukopenia, and anemia, is frequent in WD patients. Thrombocytopenia and leukopenia increase the risk of spontaneous bleeding and bacterial infections. However, some anticopper drugs, such as penicillamine and trientine, occasionally cause bone marrow depression, including anemia/leukopenia, and can even paradoxically worsen symptoms [2,4]. Therefore, anticopper therapy has often to be stopped or reduced in WD patients, which invariably leads to disease progression.

Surgical splenectomy is a traditional treatment for hypersplenism in WD patients; however, splenectomy is associated with significant postoperative complications [5,6]. Additionally, splenectomy can lead to serious emotional and neurological deterioration in WD patients [7]. Partial splenic embolization (PSE) has been^{FIGURE 1} widely used as an alternative to splenectomy for the treatment of hypersplenism because it is minimally invasive and associated with fewer complications [5,6,8,9].

However, the role of PSE in the management of hypersplenism in WD patients has not yet been described. This prospective study was performed to evaluate the effectiveness and safety of PSE compared with splenectomy in the treatment of hypersplenism in WD patients.

Methods

Patients

The study was approved by the Ethics Committee of Anhui University of Traditional Chinese Medicine. In this study, 50 hypersplenic WD patients were recruited from the Neurology Department of the First Affiliated Hospital of Anhui University of Traditional Chinese Medicine between June 2010 and October 2013. The 50 WD patients recruited in our study had a mix of hepatic and neurological disease (Figure 1) and were randomly divided into groups A and B, each containing 25 patients. Group A comprised 16 males and 9 females aged 14–46 years (average 21.480 ± 8.496 years) who were treated with percutaneous transcatheter PSE. Group B included 14 males and 11 females aged 13–46 years (average 25.840 ± 10.266 years) who were treated with splenectomy. The neurological manifestations of the 50 WD patients included dysarthria, tremor, drooling, dystonia, incoordination, dysphagia, spasticity, and psychiatric symptoms. All of the patients had received chelation therapy, including D-penicillamine and zinc compounds; the duration of chelation therapy ranged from 1 to 23 years. At the time of study recruitment, all of the patients were treated with anticopper therapy using a combination of zinc and intravenous sodium 2, 3-dimercapto-1-propane sulfonate (DMPS) for 2 weeks before PSE or splenectomy. The daily DMPS dose was 5–15 mg/kg. Simultaneously, the patients were put on a low-copper diet and received good perioperative management. Table 1 shows the clinical characteristics of hypersplenic WD patients before PSE or splenectomy (means ± SDs). Anticopper therapy was stopped on the day before surgery. Once the patients' physical conditions had stabilized, anticopper therapy was reinitiated; early anticopper therapy was planned to continue indefinitely. The time until stabilization was approximately 3 days to 1 week after surgery.

FIGURE 1. Brain MRI of a patient with Wilson's disease. A, D: T1WI reveals a symmetric low signal in the bilateral basal ganglia; B, E: T2WI reveals a symmetric high signal in the bilateral basal ganglia; C, F: Axial fluid attenuated inversion recovery (FLAIR) MRI reveals a symmetric high signal in the bilateral basal ganglia.

TABLE 1. Clinical characteristics of hypersplenic patients with WD before PSE or splenectomy (means ± SDs)

Variables	Group A	Group B	P value
Patients (n)	25	25	> 0.05
Average age (years)	22.00±7.2	20.7±6.48	> 0.05
Gender (M/F)	16/9	14/11	> 0.05
Liver function
ALT (U/L)	75.79 ± 8.66	73.47 ± 9.33	> 0.05
AST (U/L)	71.52 ± 8.35	73.52 ± 8.44	> 0.05
TB (µmol/L)	24.42 ± 3.61	26.68 ± 2.18	> 0.05
A/G	0.96 ± 0.13	0.93 ± 0.19	> 0.05
WBC (×10^9/L)	2.17 ± 0.91	2.21 ± 0.87	> 0.05
PLT (×10^9/L)	57.57 ± 13.57	56.90 ± 14.38	> 0.05
Serum ceruloplasmin (g/L)	0.075 ± 0.032	0.077±0.038	> 0.05
24-hr urinary copper excretion (µg/24 h)	1032.769±783.970	935.124±571.877	> 0.05
Neurologic (n)	25	25	> 0.05
KF rings (n)	25	25	> 0.05
Brain MRI abnormalities of WD (n)	25	25	> 0.05

WBC, white blood cell; PLT, platelet; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TB, total bilirubin; WD, Wilson's disease.

The WD diagnosis was based on characteristic clinical manifestations, low serum ceruloplasmin (<0.2 g/L), elevated 24-hr urinary copper excretion (>100 µg/24 h), elevated liver copper levels (>250 µg/g dry weight), presence of a Kayser–Fleischer (K-F) ring, and elevated 24-hr urinary copper excretion following the administration of 2 × 500-mg doses of penicillamine (>1600 µg/24 h), as previously described. As our center was unable to perform copper quantification in liver tissue, liver copper dry weight >250 µg/g was not included in our diagnostic criteria. Signs of hypersplenism, including splenomegaly, thrombocytopenia, leukopenia, anemia, and bone marrow suppression, were evaluated by clinical laboratory testing, ultrasonography, and computed tomography. Indications for the treatment of hypersplenism in the patients were as follows: moderate to severe splenomegaly; an absence or a small amount of ascites; decreasing leukocyte and/or platelet (PLT) counts, particularly a PLT count of less than 60 × 10⁹/L and a leukocyte count of less than 3.0 × 10⁹/L; recurrent gastrointestinal bleeding; and a Child-Pugh liver function score of A or B. Patient gender, age, and disease severity were similar between the two groups.

Patients with severe jaundice, refractory ascites or other advanced liver disease, severe heart disease, renal failure, or allergies to the contrast medium were excluded from this study. Informed consent was obtained from all patients before the procedure.

Procedure for partial splenic embolization

PSE was performed as follows. Briefly, under strict aseptic conditions, the right femoral artery was accessed by a 5F-Cobra catheter via the Seldinger approach. The celiac axis was catheterized, and digital subtraction angiography (DSA) was performed to identify the vascular circulation routes. The tip of the catheter was then placed as distally as possible at the hilum of the spleen along a 0.038-inch guidewire. Embolization was achieved by injecting polyvinyl alcohol particles (10 mg) suspended in a saline solution containing antibiotics (80 mg of gentamicin) depending on the detailed anatomy of the splenic artery and its branches, as visualized by DSA. The injection was performed very slowly to avoid reflux. The aim of PSE was to occlude approximately 50–70% of the original splenic volume under angiographic control (Figure 2). Post-PSE supportive care included systemic prophylaxis with intravenous antibiotics (cefotiam 1 g every 12 h for 5 days) and individualized analgesic (0.1 g of ibuprofen or 5 mg of oxycodone for 2 to 7 days) treatment.

FIGURE 2. Splenic artery angiography before (A, B) and after (C, D) embolization.

Patients in group B underwent splenectomy under general anesthesia (general anesthesia was induced with 8–30 μg, 0.05–0.075 mg/kg midazolam, 0.3–0.6 mg/kg atracurium, and maintained with 4–12 mg/kg/h propofol, 5-10 μg/kg/min atracurium, and 0.1–2 μg/kg/min remifentanil; anesthesia time was approximately 90 to 120 min). Prophylactic systemic intravenous antibiotics (cefotiam 1 g every 12 hr for 5 days) were administered 1 week before the operation as well as several days postoperatively.

Analytical assessment

White blood cell (WBC) and PLT counts were monitored before the procedure and at days 7, 14, 21, 28, and 35 after the procedure. Liver function tests (aspartate aminotransferase [AST], alanine aminotransferase [ALT], total bilirubin [TBIL], and the ratio of albumin and globulin [ALB/GLO] in the serum) were measured before and after the procedures. For the patients in group A, abdominal CT scans were performed before PSE and subsequently on days 7, 14, 21, 28, and 180 after PSE to determine the extent of splenic infarction (Figure 3). B-type ultrasonic of spleen was repeated routinely after PSE and splenectomy. Additionally, the frequency and type of complications associated with PSE and splenectomy were recorded.

FIGURE 3. Computed tomography (CT) of the abdomen after embolization. The low-density region represents the infarcted area of the spleen.

Statistical analysis

Data are expressed as means ± SD. To determine whether a statistically significant difference existed between the two groups, Student's t-test or the chi-square test was used. For normally distributed variables, an unpaired Student's t-test was used for comparisons between the two groups, and a paired Student's t-test was used to assess differences before and after PSE or splenectomy. For abnormally distributed variables, the rank sum test was used for between-group comparisons. Sample rates were compared using the chi-square test. Differences were considered significant at P < 0.05.

Results

Effects of PSE and splenectomy on blood cell counts

Patients who underwent^{FIGURE 4} PSE demonstrated^{FIGURE 5} marked improvements in WBC and PLT counts on days 7, 14, 21, 28, and 35 after the procedure. The WBC counts peaked on the seventh day and then gradually decreased during the follow-up period, whereas the PLT counts gradually increased, peaking on day 35. However, the post-PSE WBC and PLT counts remained significantly higher than the pre-PSE counts.

Following splenectomy, patients in group B also experienced marked improvements in WBC and PLT counts at days 7, 14, 21, 28, and 35. These values were all higher than those in group A, with significant differences between the two groups at postprocedure days 7, 14, and 21 (P < 0.05) (Figure 4).

FIGURE 4. Changes in the WBC and PLT counts before and after PSE (group A, n = 25) or splenectomy (group B, n = 25). Values are expressed as the means ± standard deviations. ^*Significant increase at the indicated time point (7, 14, 21, 28, or 35 days after the procedure) compared with the pre-PSE or pre-splenectomy values in each group (^*P < 0.05).

Effects of PSE and Splenectomy on Liver Function

Compared with pre-PSE levels, AST, ALT, and TBIL decreased significantly post-PSE (P < 0.05), whereas the ALB/GLO ratio significantly increased (P < 0.05).

In group B, AST, ALT, and TBIL also decreased postsplenectomy, whereas the ALB/GLO ratio increased. However, no significant differences in these values were found before and after splenectomy (P > 0.05) (Figure 5).

FIGURE 5. Changes in liver function parameters before and after PSE (group A, n = 20) or splenectomy (group B, n = 20). Values are expressed as the means ± standard deviations. Pre: before PSE or splenectomy; post: after PSE or splenectomy (^*P < 0.05).

Side effects and complications

Postembolization syndrome was the most frequent side effect in group A and was characterized by abdominal pain, fever, nausea, and vomiting. The incidence of postembolization syndrome was 84% (21/25). Five patients developed transient ascites. All these side effects were controlled with conservative therapy. Other severe complications of PSE, including splenic abscess, peritonitis, pancreatitis, and portal vein thrombosis, were not observed, and no patients in group A experienced neurological deterioration.

In contrast, all the patients in group B suffered postoperative complications. All the patients experienced fever and pain, and 11 patients developed portal vein thrombosis. Transient ascites and pleural effusion occurred in four patients and one patient, respectively. However, all postoperative complications were conservatively treated with good outcomes. Unfortunately, seven patients in group B suffered neurological deterioration postsplenectomy.

Discussion

PSE for the treatment of hypersplenism was first reported by Maddison in 1973. PSE is performed to occlude the arterial supply of the spleen, resulting in partial ischemic necrosis of the spleen, thereby decreasing splenic size and hypersplenism. Meanwhile, a portion of the splenic parenchyma is left viable to maintain splenic immunity, enabling surgery to be avoided in this high-risk patient cohort [10,11]. The increase in leukocyte counts after PSE was thought to result from a reduction in the platelet (PLT) pool due to reduced spleen volume, and the increased PLT count was thought to be associated with decreased splenic sequestration and thrombopoietin restoration [5,6,9,10]. As a minimally invasive, simple, and rapid procedure performed under local anesthesia, PSE also contributed to the improvement of liver function, which may be explained by decreasing liver congestion, enhancing the blood supply, and improving the capacity of hepatic protein synthesis [12–14]. Additionally, immunological mechanisms leading to the maintenance of splenic immunological function might be involved [12]. Splenectomy is characterized by surgical wounds; however, general anesthesia and postoperative infection may also exacerbate the existing liver damage in WD patients, enhancing the potential risk of hepatic encephalopathy and even of life-threatening liver failure [7,15–17]. Meanwhile, studies have ensured the long-term efficacy of PSE in alleviating hypersplenism and reducing complications in patients with hepatitis B and C virus-related liver cirrhosis, including increasing the blood cell counts and improving the liver function [6,18]. Anticopper therapy must often be stopped or reduced in WD patients with cirrhosis and hypersplenism, which invariably leads to disease progression because of some anticopper drugs, such as penicillamine and trientine, occasionally causing bone marrow depression, including anemia/leukopenia [2,4]. Thus, the elevated decoppering capacity due to the doses increase of anticopper drugs after PSE can exert long-term beneficial effect on WD patients.

Furthermore, splenectomy, which creates a large surgical wound, may also exacerbate serious neurological deterioration in some hypersplenic WD patients, whereas minimally invasive PSE appears to be relatively tolerable and safe. The reason for this difference is uncertain. Walshed observed that 8 of 10 WD patients with bleeding esophageal varices suffered serious neurological deterioration after splenectomy [7]. Since that study, few reports have been published on splenectomy for hypersplenism in WD patients. The lack of further reports of splenectomy in WD with cirrhosis and hypersplenism might be due to the neurological deterioration evoked by splenectomy as reported in early literature [7]. In fact, the phenomenon of neurological deterioration in WD patients is more commonly observed during the initial phase of treatment with different anticopper drugs [19–22]. The reason for this deterioration is uncertain. One theory is that the mechanism may involve mobilization and redistribution of hepatic copper [2,20]. The copper accumulation is initially in the liver and then mobilization of copper from the liver to the blood, brain, and other tissues, and the copper transported into the brain was mainly achieved through the blood-brain barrier (BBB) [23,24]. It was reported that the brain MRI abnormalities increased in WD patients after anticopper drugs induced neurological deterioration [19,20]. Meanwhile, based on the blood-brain barrier (BBB) study in WD patients, Stuerenburg proposed that the increase in BBB disturbance enables mobilization of copper from the liver to the brain more readily, causing the deterioration of neurological symptoms during anti-copper treatment [25]. Thus, we may speculate that the possible underlying mechanisms of neurological deterioration evoked by splenectomy might be similar to that in anticopper treatment. Splenectomy may promote the activation of copper mobilization and redistribution from the liver to the brain more readily, which may be associated with the BBB disturbance by large surgical wounds, general anesthesia, and postoperative infection due to splenectomy. Additionally, the deterioration of neurological symptom evoked by splenectomy in WD with cirrhosis and hypersplenism might be similar to D-penicillamine-induced neurological deterioration with sometimes irreversible damage [2]. Thus, splenectomy may cause the short-term and long-term deterioration of neurological symptom in WD patients with cirrhosis and hypersplenism.

At the time of study recruitment, all of the patients received anticopper therapy with a combination of zinc and intravenous DMPS for 2 weeks before PSE or splenectomy. DMPS is one of the best intravenous copper-chelating agents; it has low toxicity and only minor side effects [26]. In contrast, penicillamine has serious adverse effects, which can lead to bone marrow depression and, in some patients with hypersplenism and WD, drug intolerance. Moreover, the widespread use of trientine is difficult in China due to its high cost and lack of availability. Zinc monotherapy is unable to reduce high copper levels [27,28]. Therefore, the combination of zinc and short-term intravenous DMPS is the best approach for anticopper treatment in patients with hypersplenism and WD before PSE or splenectomy.

In conclusion, hypersplenism in WD patients was successfully treated with PSE. As a minimally invasive, simple, and rapid procedure performed under local anesthesia, PSE may be superior to surgical splenectomy as a treatment for WD with hypersplenism. A longer follow-up and a larger sample size would be useful to further estimate the clinical value of PSE for the management of patients with hypersplenism and WD.

Declaration of Interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Acknowledgments

The authors thank all the participants in this program for their help and cooperation, as well as their colleagues in the working groups. The research was supported by the National Natural Science Foundation of China (grant number: 81373599).