Hepatic metastasectomy in children




There are little data regarding the safety and efficacy of hepatic metastasectomy for solid tumors in childhood. We reviewed our institutional experience to assess operative mortality and morbidity, technique of resection, local control, and survival in pediatric patients undergoing liver resection for metastases.


All pediatric patients who underwent hepatic resection for metastatic disease from August 1988 to July 2005 were retrospectively identified and clinical data were collected.


Fifteen patients were identified during this period and primary malignancies included neuroblastoma (7), Wilms tumor (3), osteogenic sarcoma (2), malignant gastric epithelial tumor (1), and desmoplastic small round cell tumor (2). Twelve patients underwent anatomical hepatic resections and 3 had wedge resections. There were no intraoperative or postoperative deaths. The 2 postoperative complications included 1 wound infection and 1 bile collection. The median follow-up after hepatic resection was 1.6 years (0.2–7 years). Three patients remain alive. Eleven patients died of progressive disease; 4 patients suffered local recurrence. One patient died from enterocolitis and sepsis and was without evidence of malignancy at the time of death.


Hepatic metastasectomy in children is feasible and is associated with a low operative mortality and morbidity. In this small group of patients anatomic hepatectomy was associated with better local control compared with wedge resection. Overall prognosis in these patients remains poor and the decision to perform hepatic metastasectomy should be highly selective. Cancer 2007. © 2007 American Cancer Society.

Hepatic resection for metastatic disease has been described extensively in the adult literature. Hepatectomy for metastatic lesions from colorectal adenocarcinoma, renal adenocarcinoma, breast cancer, testicular cancer, and neuroendocrine tumors are feasible and have demonstrated therapeutic benefits in select patients.1–6 Specific employment of hepatic metastasectomy in pediatric malignancies is less common, however, with few reports describing its use in the treatment of nephroblastoma (Wilms tumor), pancreaticoblastoma, and choriocarcinoma.6–9 This report describes our institutional experience with hepatic metastasectomy in the pediatric population.


Data were retrospectively collected from August 1988 to July 2005; all pediatric patients with hepatic resection for metastatic disease were identified. Patients with hepatic resection for primary hepatic malignancy were excluded. Institutional Review Board waiver was obtained in accordance with the Health Insurance Portability and Accountability Act regulations. The inpatient and outpatient charts as well as operative reports were reviewed. Techniques of major hepatic resection were performed as described from previous reports.10, 11


Fifteen patients underwent hepatic metastasectomy and their clinical characteristics are summarized in Table 1. The median age at diagnosis was 5.5 years (range, 0.3–18.8 years). The median age at the time of hepatic resection was 6.2 years (range, 0.6–22.1 years). The average interval from diagnosis to hepatic resection was 1.5 years (range, 0.3–5.5). The distribution of primary tumors included 7 neuroblastomas, 3 Wilms tumors, 2 osteogenic sarcomas, 2 desmoplastic small round cell tumors, and 1 malignant epithelial neoplasm of the stomach.

Table 1. Clinical Characteristics and Surgical Technique
PatientAge at surgery, yTumor (Stage)ProcedureSegments removedMargins
  1. OS indicates osteosarcoma; NB, neuroblastoma; WT, Wilms tumor/nephroblastoma; DSRCT, desmoplastic small round cell tumor.

119.3OS (IV)R. Lobe7Negative
221.16OS (IV)Swiss CheesenonanatomicPositive
31.3NB (IV)R. Lobe5,8Negative
44.6NB (IV)R. Lobe6,8Negative
57.6NB (IV)Ext. Left2,3,4,5,8Negative
614.75NB (IV)Nonanatomic7Positive
710.8NB (IV)R. Lobe5,6,7,8Negative
84.6NB (IV)L. lateral segmentectomy2,3,4,6,8Negative
9.58NB (IV)L. segmentectomy2,3,5Negative
106WT (IV)Nonanatomic7,8Negative
117.3WT (II)Ext. Right4,5,6,7,8Negative
126.16WT (I)Ext. Right4,5,6,7,8Negative
1419.1DSRCT (IV)R. Lobe with bile duct repair5,6,7,8Negative
1522.1DSRCT (IV)R. posterior seg5,6,7Negative

The liver was the only site of metastatic disease in 9 patients at the time of hepatic metastasectomy. Six patients had synchronous liver metastasis noted at the time of their primary resection, which were biopsied and treated before definitive hepatic resections. One patient had concurrent lung metastases from Wilms tumor and the other 5 patients had widespread neuroblastoma in the bone and the bone marrow. The technique of hepatic resection included 5 right lobectomies, 2 extended right hepatectomies, 1 right posterior segmentectomy, 1 extended left lobectomy, 2 left segmentectomies, 1 central hepatic resection, and 3 nonanatomic resections (Table 1).

There were no intraoperative or postoperative deaths; the median estimated blood loss was 27 mL per kg (range, 7.5–100 mL/kg). Two postoperative complications included 1 wound infection and 1 bile collection. The local wound infection occurred in a malnourished patient with neuroblastoma approximately 2 weeks postoperatively. Wound culture revealed multiple enteric organisms; the infection was treated with local wound care. The postoperative bile collection occurred on postoperative day 5 in a patient with osteosarcoma. This was treated with percutaneous drainage for 3 weeks.

Hepatic function after surgery was measured via serum total bilirubin (TB) levels and serum alanine transaminase (ALT) levels. Measurements were obtained before surgical resection, during the postoperative period, and during follow-up. All patients had normal to high-normal levels of TB (range, 0.2–1.0 mg/dL) preoperatively except 1 patient who had biliary obstruction secondary to metastatic lesions that required drainage via a percutaneous biliary tube. ALT levels were also normal to slightly elevated (range, 8–66 units) going into surgery. Postoperatively, the majority of patients continued to have normal TB and ALT levels except for those patients who had a local recurrence of disease. None of the patients experienced hepatic insufficiency within the immediate postoperative period. No patients required biliary drainage procedures after surgery. Median time to discharge home was 6 days from the time of surgery (range, 5–17 days). All patients were fully ambulatory and taking a normal diet at the time of discharge and Karnofsky scores were above 80 in each.

Patient outcome and follow-up is listed in Table 2. Follow-up for all patients consisted of interval radiographic imaging with CT scan to determine the presence of reoccurrence and/or metastatic diseases in addition to regular clinic follow-up. Median survival for all patients was 9 months (95% confidence interval [CI]: 6.2–11.8 months). Four patients suffered hepatic recurrence: 3 had widespread neuroblastoma, 1 at 2 months after a nonanatomic hepatic resection with a positive margin and the other 2 approximately 1 month after anatomic resections with negative margins. The last recurrence was in a patient with osteogenic sarcoma 3.8 years after a right hepatic lobectomy also with negative margins. Two patients in this study died from sepsis/necrotizing enterocolitis secondary to chemotherapy 4 months after hepatic resection and were without evidence of recurrence at the time of death. Three patients remain alive after hepatic resection; 1 patient with stage IV neuroblastoma, 1 patient with desmoplastic small round cell tumor, and 1 with malignant epithelial neoplasm of the stomach—this patient has had no evidence of disease 9.5 years after the resection of liver metastases (Table 2).

Table 2. Patient Follow-up and Outcomes
PatientHepatic recurrenceFollow-up* moOutcome
  • DOD indicates died of disease; NED, no evidence of disease; AWD, alive with disease.

  • *

    Follow-up time is from the time of hepatic resection to either time of death or the last clinic visit.

  • Patient died of sepsis/enterocolitis secondary to chemotherapy with no evidence of liver disease.



With the advances in cancer treatment modalities, control of the primary site can be achieved in a majority of patients with solid tumors. In contrast, metastatic lesions continue to remain a treatment burden. It has been well established that patients with metastatic disease have a significantly worse prognosis than those with localized disease and require more effective treatments. For selected malignancies, metastasectomy has demonstrated survival benefit, best represented by hepatic resections for colorectal cancers and pulmonary resections for osteogenic sarcoma. However, there are few reports in the literature on hepatic metastasectomy in the pediatric population and its effectiveness.12–17

The most well-described use of hepatic metastasectomy is for Wilms tumors. Foster18 analyzed 15 cases of metastatic Wilms tumor from the literature and reported a 62% 2-year and 44% 5-year survival after hepatic metastasectomy. Morrow et al.19 described 4 Wilms tumor patients undergoing hepatic resection with a 5-year cumulative survival of 50%. Although none of our 3 patients with Wilms tumor survived more than 9 months, the literature still suggests that liver resection for Wilms tumor is indicated for isolated hematogenous metastases.

Stage 4 neuroblastoma patients may present with isolated liver metastases. Patients may benefit from resection of these lesions, resulting in prolonged survival and/or treatment reductions. Of the 7 neuroblastoma patients who underwent resection only 3 suffered a local recurrence within the liver. The remaining patients were able to avoid intensification of chemotherapy as well as radiation to the liver.

Desmoplastic small round cell tumor (DSRCT) is another solid tumor that frequently presents with aggressive disease, large primary tumors, and liver lesions. Optimal treatment of this tumor includes complete removal/debulking of all tumors present at the time of surgery. Two patients in this series were diagnosed with stage 4 DSRCTs and had evidence of isolated liver metastasis on imaging scans. Preoperatively, 1 patient had chronic bile duct obstruction secondary to a hilar tumor mass and required percutaneous biliary drainage. Resection of the hilar mass along with the liver metastases followed by bile duct reconstruction relieved the obstruction and allowed removal of a percutaneous biliary drainage catheter.

Three patients in our series are long-term survivors, having lived for greater than 2 years after hepatic metastasectomy. The first is a neuroblastoma patient who has completed all treatment and is disease-free at 59 months. The second patient had a primary diagnosis of osteogenic sarcoma. Whereas pulmonary metastasectomy for osteosarcoma has been shown to improve survival,7, 15, 16 hepatic metastasectomy for this disease has been less well characterized. This patient had isolated liver metastases resected with formal right lobectomy. This patient was disease-free for 2 years and then developed limb recurrence of the osteosarcoma treated with amputation. After this he remained disease-free for a total of 46 months before developing widespread reoccurrence including further hepatic metastases and succumbed to disease.

Our third long-term survivor in this series had a diagnosis of undifferentiated malignant epithelial neoplasm of the stomach. Gastric epithelial cancer is extremely rare in the pediatric population.17 This patient had 3 synchronous liver metastases that were not responsive to chemotherapy. Control of primary tumor, absence of extrahepatic disease, and lack of other effective therapy were confirmed before hepatic resection. Curative central hepatic resection was performed 5 months after the diagnosis of primary tumor. The patient has been followed regularly and is alive and well 9.5 years after hepatic metastasectomy with normal hepatic function.

Despite aggressive management in this series of patients, prognosis still remains poor for patients with metastatic liver disease. Benefits of metastasectomy, however, can include stabilization or improvement in patient liver functions, relief of biliary obstruction and removal of biliary drains, and a potential for long-term disease-free survival. Most patients are able to return home fairly quickly after surgery and continue with chemotherapy if necessary.

The decision to perform localized resection of liver metastases still remains a highly selective one. Prerequisites for hepatic metastasectomy include the control of the primary site, no evidence of extrahepatic disease, adequate hepatic reserve for resection, and a clear understanding of tumor biology. Whereas a careful anatomic resection should be the goal of surgical resection, it should not preclude a nonanatomic resection if negative margins can be obtained. Neuroblastoma and sarcoma patients may appear to benefit most from hepatic metastasectomy. This can result in 33% 3-year survival in a highly selected group of patients. We believe hepatic metastasectomy in children is feasible, with a low operative mortality and morbidity and the potential for prolonged survival.