World Journal of Surgery

, Volume 33, Issue 3, pp 597–602

Clinical Experience with Infantile Hepatic Hemangioendothelioma

Authors

    • Department of Pediatric SurgerySeoul National University Children’s Hospital
  • Hyoung-Joo Kwon
    • Department of Pediatric SurgerySeoul National University Children’s Hospital
    • Department of Pediatric SurgerySeoul National University Children’s Hospital
  • Woong-Jae Yun
    • Department of PathologySeoul National University Children’s Hospital
  • Sung-Eun Jung
    • Department of Pediatric SurgerySeoul National University Children’s Hospital
Article

DOI: 10.1007/s00268-008-9882-4

Cite this article as:
Moon, S., Kwon, H., Park, K. et al. World J Surg (2009) 33: 597. doi:10.1007/s00268-008-9882-4
  • 223 Views

Abstract

Background

Infantile hepatic hemangioendothelioma (IHHE) is a rare disorder with only a few series reported in the medical literature. We reviewed our treatment experience with IHHE over 17 years.

Methods

A retrospective analysis of patients with IHHE between 1991 and 2008 was performed.

Results

Sixteen patients (median age 30 days) with IHHE were identified. A palpable abdominal mass was the most common presentation. All except two cases could be diagnosed radiologically. Thirteen patients had a unilobar single tumor and three patients had bilobar disease. Nine patients with symptoms and a resectable tumor underwent complete resection. Three patients with symptoms and unresectable tumor underwent medical treatment with steroids and interferon. Four asymptomatic patients were closely observed. Overall, 14 patients were cured and 1 patient died of postoperative bleeding. One patient is still on medication, and the tumor has greatly decreased in size. Two patients with bilobar disease showed elevated levels of serum α-fetoprotein at presentation. Histopathology confirmed type 1 IHHE in all of the 10 specimens.

Conclusions

The presence of clinical symptoms is a key element determining the treatment options. In symptomatic patients, primary surgical resection should be considered whenever feasible.

Introduction

Infantile hepatic hemangioendothelioma (IHHE) is, although rare, the most common benign hepatic tumor in children. IHHE is also the third most common hepatic tumor in children [1]. Altogether, 85% percent of patients with IHHE are diagnosed before the age of 2 months, and most cases present before the age of 6 months. IHHE is rarely diagnosed beyond the age of 3 years [2]. The clinical spectrum of IHHE is wide, from spontaneous regression without any treatment to life-threatening high-output congestive heart failure with respiratory compromise. IHHE is also known to be associated with the Kasabach-Merritt syndrome, which is a hematologic emergency [3].

Many options exist for the treatment of IHHE. Asymptomatic lesions may undergo spontaneous regression without any treatment within a year, whereas symptomatic lesions require medical and/or interventional treatment modalities. Steroids and interferon-α are two commonly used agents for medical treatment [4, 5]. Interventional modalities include hepatic artery ligation or embolization, hepatic resection, and orthotopic liver transplantation [6]. There are numerous case reports presenting individual treatment outcomes. However, due to the rarity of this disease, there are only a few series of patients with the clinical features and the treatment outcomes of IHHE reported.

The purpose of this study was to present the clinical experience and treatment outcome of IHHE at our center over 17 years.

Patients and methods

We reviewed retrospectively the medical records of 16 patients (7 males, 9 females) who were treated at the Department of Pediatric Surgery, Seoul National University Children’s Hospital, Seoul, Korea, between January 1991 and July 2008. The information collected was as follows: age at presentation and diagnosis, presenting symptoms, diagnostic procedures, location of the tumor, level of serum α-fetoprotein (AFP) at presentation, therapeutic modalities employed, histopathologic type, and treatment outcome. Age at diagnosis was defined as either when the IHHE was diagnosed radiologically or diagnosed pathologically (percutaneous needle biopsy or surgical resection) when the diagnosis could not be confirmed radiologically. Serum AFP was measured in all patients at initial presentation, and the level was interpreted to be either within normal limits or elevated considering the reference value for the patient’s age. The reference values for serum AFP in normal infants have been previously described [7].

All patients initially underwent abdominal ultrasonography (US) for diagnosis and proceeded to computed tomography (CT) and/or magnetic resonance imaging (MRI) according to the diagnostic impression gained from the US. Percutaneous needle biopsy or open biopsy was attempted when the diagnosis could not be confirmed by the radiological findings.

The location of the tumor was defined as either unilobar (right or left hemilobar) when a single mass was confined to the hemiliver or bilobar when a single mass involved both the hemiliver and multiple masses were scattered throughout the whole liver.

We used steroids and interferon-α (IFN) for medical treatment, and hepatic artery embolization and hepatic resection for interventional treatment. Digoxin and diuretics were given for symptomatic congestive heart failure. When the patient had clinical symptoms and the tumor was considered to be readily resectable (i.e., unilobar solitary tumor), primary surgical resection was attempted. The goal of the surgery was to remove the tumor completely by anatomic resection. For the patients without symptoms, the tumor was left alone without any treatment regardless of its extent. In cases that were symptomatic but the tumor was not resectable or bilobar disease was present, medical treatment with or without preceding hepatic artery embolization was planned. The steroid dose was started at 2.5 mg/kg/day and tapered to 1.0 mg/kg/day over 6 to 8 weeks. The dose of IFN was adjusted between 750,000 and 1,500,000 IU/day, and maintained for up to 1 year. Radiation therapy and orthotopic liver transplantation was not performed during the study period.

The treatment outcomes were classified as cure, regression, or death. A cure was defined as complete disappearance of the tumor regardless of the treatment modality employed. When the tumor was regressing, with or without medical treatment, the outcome was defined as “in regression.” In cases with complete regression, the time interval between the diagnosis and complete regression was recorded.

Results

The mean age at presentation was 49 days (range 1–182 days; median 14 days) and the mean age at diagnosis was 60 days (range 1–186 days; median 30 days). In all, 5 patients were found to have a hepatic tumor prenatally, and 12 patients (including these 5) with prenatally detected tumors were diagnosed within 3 months of age. Four patients were diagnosed between the age of 3 and 12 months. The median follow-up duration after diagnosis was 236 days (range 3 days to 8.9 years).

The most common clinical presentation was a palpable abdominal mass (n = 6), abdominal distension (n = 5), and congestive heart failure (n = 5). One patient (no. 5) had associated Kasabach-Merritt syndrome. Five patients had no clinical symptoms.

All of the patients underwent abdominal US as the initial diagnostic evaluation. Based on the US results, nine patients underwent CT scanning and five patients MRI. Two patients underwent both CT and MRI examinations. After radiologic examination, 14 patients were diagnosed as having IHHE. In the remaining two cases, a hepatoblastoma could not be determined by the radiological examination, and percutaneous needle biopsy was attempted. The needle biopsy results confirmed the diagnosis of IHHE in one patient, but an adequate tissue sample was not obtained from the other patient. This patient eventually underwent hepatic resection and was confirmed to have IHHE (Table 1).
Table 1

Clinical characteristics and treatment outcome of 16 patients with IHHE

Patient no.

Sex

Age at diagnosis (days)

Disease extent

AFP (ng/ml)

Treatment

Time to complete regression (days)

Follow-up duration after diagnosis (days)

Outcome

1a

F

41

Left

1,460

Sectionectomy

22

Cure

2b

M

1

Right

30,872

Observation

846

1,045

Cure

3

F

186

Left

13

Tumorectomy

1,396

Cure

4

M

152

Right

27

Observation

3058

3,200

Cure

5b

F

70

Left

180

Steroid + sectionectomy

62

Cure

6

M

59

Bilobar, multiple

143

Steroid + DGX + DUT

914

2,605

Cure

7

F

19

Right

4,840

Bisegmentectomy

1,555

Cure

8a

M

90

Right

138

Observation

257

257

Cure

9b

M

5

Left

13,200

Tumorectomy

3

Dead

10b

M

1

Right

10,600

Observation

784

1,900

Cure

11

F

158

Bilobar, multiple

130

Steroid + IFN

180

522

Cure

12

F

82

Left

92

Sectionectomy

215

Cure

13

F

1

Bilobar, single

1,800

HAE + steroid + IFN + DGX + DUT

569

In regression

14

M

14

Right

16,040

Bisegmentectomy

270

Cure

15

F

123

Left

98

Sectionectomy

141

Cure

16b

F

1

Right

48,400

Hemihepatectomy

57

Cure

DGX, Digoxin; DUT, Diuretics; IFN, Interferon-α; HAE, Hepatic artery embolization

aNeedle biopsy was attempted for a diagnosis

bLiver mass was detected prenatally

Of the 16 patients, 13 (81%) had a unilobar single tumor (7 right hemiliver, 6 left hemiliver) (Fig. 1a, b) had bilobar disease (2 bilobar multiple, 1 single huge mass involving both lobes) (Fig. 1c). The level of serum AFP at presentation was within normal limits in 14 patients. In one patient (no. 16), however, the AFP level was elevated at the follow-up examinations (118,800 ng/ml at 2 weeks of age and 306,800 ng/ml at 1 month of age). Two patients with bilobar disease showed moderately elevated AFP levels at presentation (Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00268-008-9882-4/MediaObjects/268_2008_9882_Fig1_HTML.jpg
Fig. 1

Representative computed tomography (CT) scans of unilobar disease (a,b) and bilobar disease (c). a Patient 14, b Patient 15, c Patient 16. Patients 14 and 15 underwent bisegmentectomy (S5, S6) and left lateral sectionectomy, respectively

Overall nine patients underwent surgical resection. Complete tumor removal could be achieved in all patients. For the patient who had combined Kasabach-Merritt syndrome, surgical resection was attempted after 3 weeks of steroid therapy to control thrombocytopenia and coagulopathy. The three patients with bilobar disease underwent medical treatment. For a very large bilobar mass (patient 13), right hepatic artery embolization preceded the initiation of medical therapy. The remaining four patients who had unilobar tumors without clinical symptoms received no treatment and were closely followed in the outpatient setting (Table 1). The pathology examination confirmed the diagnosis of type 1 IHHE in all nine surgical specimens and in one needle biopsy specimen (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs00268-008-9882-4/MediaObjects/268_2008_9882_Fig2_HTML.jpg
Fig. 2

Type 1 infantile hepatic hemangioendothelioma (IHHE) in patient 16. The lesion is characterized by proliferation of well formed capillary spaces lined by a single layer of bland endothelial cells. (H&E, ×200)

Eight of nine patients were cured by the surgical resection alone. One patient died of postoperative bleeding on the day of the surgery. Of the three patients who underwent medical treatment, two patients (nos. 6 and 11) were cured after 914 days and 180 days, respectively. The other patient (no. 13) is still on treatment with IFN, and the tumor has markedly decreased with internal necrosis (Fig. 3a, b). All four patients who were closely followed without treatment were cured. The mean time to complete regression was 3.4 years (range 257 days to 8.5 years).
https://static-content.springer.com/image/art%3A10.1007%2Fs00268-008-9882-4/MediaObjects/268_2008_9882_Fig3_HTML.jpg
Fig. 3

a CT scan at initial presentation. A large, single mass was present involving both lobes of the liver. b The mass has markedly decreased in size with increased internal necrosis 16 months after the initiation of treatment

There was no tumor recurrence during the study period.

Discussion

IHHE is the most common benign hepatic tumor in children and undergoes spontaneous regression without any treatment in some cases. However, because of its large size at presentation (thus causing abdominal distension and a palpable mass) and occasional associated medical complications such as congestive heart failure or the Kasabach-Merritt syndrome, medical or surgical intervention is generally required. Prior studies have reported a 12% to 40% survival rate if symptoms alone are treated while awaiting tumor regression, whereas the survival rate improves to 70% if specific therapy such as steroid treatment or resection is undertaken [8]. Daller et al. reported a 62% cure rate in their study where every patient underwent medical or surgical treatment [9]. In our study, by contrast, 12 of 16 patients underwent specific therapy for IHHE and an overall 88% cure rate was achieved. This difference can be explained in two ways: there were (1) more operable cases (9/16 vs. 5/13) and (2) more asymptomatic patients (4/16 vs. 0/13) in our study. Most of our symptomatic patients had resectable lesions, even though IHHE is known to be diffuse in nature and therefore unsuitable for resection [10]. In addition, the lower success rate of medical treatment reported by Daller et al. reflects the wide disparity in disease severity and unpredictable treatment response of IHHE.

Zenge et al. recommended that resection be reserved for patients with life-threatening complications or for tumors indistinguishable from malignant tumors by radiologic studies [11]. Samuel and Spitz recommended primary resection in symptomatic patients with readily resectable solitary tumors [10]. Our indications for resection were similar to those of Samuel and Spitz, and we achieved good results. Although we had one postoperative death, there were no serious postoperative complications in the remaining cases.

It has been reported that steroid therapy alone can induce complete tumor regression in 30% of patients and that 90% of patients have tumor regression when IFN is added to the regimen [12, 13]. In our study, there was no treatment failure with steroid and IFN therapy. However, the small number of patients who underwent medical treatment makes it difficult to draw conclusions on the use of steroid and IFN treatment for IHHE. The response to steroid therapy is often noted within 1 to 2 weeks, with approximately 5 months needed for complete resolution [14]. The IFN inhibits endothelial cell migration and proliferation, causing regression of a hemangioendothelioma resistant to steroids [15]. Our patients demonstrated rather delayed responses compared to prior published cases; however, the responses were much more delayed in those cases that were treated conservatively with close observation. Our findings support the effects of steroid and IFN treatment as causing early regression of IHHE. However, because long-term treatment with these medications can be accompanied by serious side effects, we do not advocate their use in asymptomatic patients to achieve early regression.

Hepatic artery embolization is usually effective in children with disease limited to only a few hepatic segments, but some success has been reported in children with diffuse IHHE as well [6]. We performed hepatic artery embolization in one patient (no. 16) along with steroid and IFN treatment; however, the therapeutic effect of the embolization was not definite, although the tumor is significantly reduced in size. In fact, there was no radiologic evidence that the tumor was supplied primarily by the right hepatic artery. However, the inoperable nature of the tumor with accompanying congestive heart failure compelled us to initiate treatment to obtain therapeutic effects immediately. Liver transplantation theoretically was an option for the patient, although practically impossible.

Some types of pediatric tumors, such as neuroblastomas, exhibit a more benign course in those cases that develop prenatally; however, little is known about the characteristic clinical features of prenatally identified IHHE. In our study, we found no significant differences in the clinical behavior between the antenatally detected cases compared to the others. However, because IHHE can be diagnosed by routine antenatal US [16], careful examination of the fetal liver mass can aid the postnatal diagnosis of IHHE and may help avoid the unnecessary irradiation of a CT scan.

Serum AFP levels are usually normal in patients with IHHE [17]; they have to be determined before a definitive diagnosis is made because IHHE is sometimes misdiagnosed radiologically as a necrotic hepatoblastoma [18]. In addition, the serum AFP level should be interpreted considering the infants’ age. Urbach et al. reported a neonate with a large liver mass and elevated serum AFP, initially misdiagnosed as a hepatoblastoma, which turned out to be IHHE after a biopsy; the authors realized that the AFP was actually within normal limits for a neonate [19]. In our study, the AFP levels were elevated in two patients but had returned to normal at the follow-up examination, and the diagnosis of IHHE was unchanged. For one patient (no. 16), however, a hepatoblastoma was considered as the initial diagnosis despite radiologic findings suggestive of IHHE. There are a few case reports of IHHE with elevated serum AFP levels [20]; however, our case showed the highest level of serum AFP reported in a patient with IHHE. The etiology of an elevated serum AFP level with IHHE is unclear. Although two of three patients with bilobar disease showed elevated levels of the serum AFP in this study, no correlation has been reported between the serum AFP level and the extent of the tumor [21].

Traditionally, two histologic types have been described for IHHE. The more common type 1 IHHE is known to be a benign tumor, and type 2 IHHE may resemble an angiosarcoma. In Dehner and Ishak’s series, the mortality rate was higher for patients with type 2 tumors [22]. Daller et al. reported a metastatic angiosarcoma in a patient with type 2 IHHE [9]. However, a recent series suggested that the histopathology of the tumor does not seem to correlate with the patient’s prognosis [6]. Although the histopathologic type was not determined in six of our patients, considering the disease course (five complete regression, one in regression), the likelihood of a disease relapse in these patients seems to be low.

Conclusions

Asymptomatic IHHE is predicted to regress spontaneously, but the time required for complete regression is unpredictable. For symptomatic and resectable IHHE, surgical resection can offer immediate, definitive cure; however, one of our patients died after surgery owing to uncontrollable bleeding. Symptomatic, unresectable IHHE requires a multimodal approach to induce tumor regression, and the outcome cannot be guaranteed. Serum AFP levels are usually within normal limits but are elevated in some patients with IHHE.

Copyright information

© Société Internationale de Chirurgie 2008