Survival of patients with skeletal metastases from hepatocellular carcinoma after surgical management

Transcription

1 Survival of patients with skeletal metastases from hepatocellular carcinoma after surgical management H. S. Cho, J. H. Oh, I. Han, H.-S. Kim From Seoul National University College of Medicine, Seoul, Korea Skeletal metastases from hepatocellular carcinoma are highly destructive vascular lesions which severely reduce the quality of life. Pre-existing liver cirrhosis presents unique challenges during the surgical management of such lesions. We carried out a retrospective study of 42 patients who had been managed surgically for skeletal metastases from hepatocellular carcinoma affecting the appendicular skeleton between January 2 and December 26. There were 38 men and four women with a mean age of 6.2 years (46 to 77). Surgery for a pathological fracture was undertaken in 3 patients and because of a high risk of fracture in 2. An intralesional surgical margin was achieved in 36 and a wide margin in six. Factors influencing survival were determined by univariate and multivariate analyses. The survival rates at one, two and three years after surgery were 42.2%, 25.8% and 9.8%, respectively. The median survival time was ten months (95% confidence interval 6.29 to 3.7). The number of skeletal metastases and the Child-Pugh grade were identified as independent prognostic factors by Cox regression analysis. The method of management of the hepatocellular carcinoma, its status in the liver, the surgical margin for skeletal metastases, the presence of a pathological fracture and adjuvant radiotherapy were not found to be significantly related to the survival of the patient, which was affected by hepatic function, as represented by the Child-Pugh grade. H. S. Cho, MD, Assistant Professor Department of Orthopaedic Surgery Kyungpook National University College of Medicine, 2 Dongduk-Ro Jung-Gu, Daegu, Korea. J. H. Oh, MD, Associate Professor I. Han, MD, Assistant Professor H.-S. Kim, MD, Associate Professor Department of Orthopaedic Surgery Seoul National University College of Medicine, Daehang-Ro Jongno-Gu, Seoul, Korea. Correspondence should be sent to Dr H.-S. Kim; mdchs@snu.ac.kr 29 British Editorial Society of Bone and Joint Surgery doi:.32/3-62x.9b $2. J Bone Joint Surg [Br] 29;9-B:55-2. Received 3 October 28; Accepted after revision June 29 Hepatocellular carcinoma is one of the most common cancers in Asia, and its incidence is increasing in the United States. 2 It has a dismal prognosis. In the past, metastases from hepatocellular carcinoma have not attracted much interest because most are not symptomatic and patients are not expected to survive for long. 3,4 However, skeletal metastases from hepatocarcinoma are so highly destructive that without appropriate management they severely affect the quality of life. 5 The incidence of skeletal metastases from hepatocellular carcinoma has been reported to range from 6% to 2%, 4,6 and, recently, it has increased because of the greater survival rates achieved by improvement in diagnosis and treatment. 7 Although favourable results have been recorded with radiotherapy, 8,9 surgical management is inevitable in the presence of a pathological fracture. However, the preexisting cirrhosis of the liver in most hepatocellular carcinomas presents unique challenges for the surgical management of metastases to bone. There have been several favourable accounts of the surgical management of extrahepatic metastases in hepatocellular carcinoma, -2 indicating that an aggressive surgical approach in selected cases may improve survival. Katagiri et al 3 proposed a scoring system for patients with skeletal metastases in order to create a prognostic model. However, other factors, especially liver function, are also involved in addition to the oncological prognostic factors in hepatocellular carcinoma patients. We have carried out a retrospective analysis to evaluate the overall survival rates and the factors affecting survival in patients with skeletal metastases from hepatocellular carcinoma which have been treated surgically. Patients and Methods We reviewed the data on 42 patients who had been managed surgically between January 2 and December 26 for skeletal metastases from hepatocellular carcinoma affecting the appendicular skeleton or pelvis. We obtained a list of patients from the orthopaedic surgical database at our institute and examined the radiological data and pathological and medical records, including the operation notes and laboratory results. We collected clinicopathological data including the clinical VOL. 9-B, No., NOVEMBER 29 55

2 56 H. S. CHO, J. H. OH, I. HAN, H.-S. KIM Table I. Stages of the American Joint Committee on cancer staging system Stage TX TO T T2 T3 T4 Definition Primary tumour cannot be assessed No evidence of primary tumour Solitary tumour without vascular invasion Solitary tumour with vascular invasion or multiple tumours none more than 5 cm in size Multiple tumours more than 5 cm in size or tumour involving a major branch of the portal or hepatic vein(s) Tumour(s) with direct invasion of adjacent organs other than gallbladder or with perforation of the visceral peritoneum Table II. The Child-Pugh classification Points Parameter 2 3 Total bilirubin (mg/dl) < 2 2 to 3 > 3 Serum albumin (g/l) > to 35 < 28 INR * <.7.7 to 2.2 > 2.2 Ascites None, suppressed with medication Refractory Hepatic encephalopathy None Grade I to grade II (or suppressed with medication) Grade III to grade IV (or refractory) *INR, international normalised ratio features, the aetiology of the hepatocellular carcinoma, the initial treatment, the status of the tumour in the liver at the time of surgery for skeletal metastases, vascular invasion of the portal or hepatic vein, the level of alphafetoprotein, the Child-Pugh classification, 4,5 the presence of metastases to internal organs, the number and location of skeletal metastases, the presence of a pathological fracture, blood loss, the surgical margin, adjuvant radiotherapy, the recurrence of a bone lesion and the final status. The initial treatment had included transarterial chemoembolisation, liver resection, radiofrequency ablation, radiotherapy or conservative management. The status of primary lesions in the liver was based on dynamic CT. This was checked before surgery for skeletal metastases and graded using the T-stage system of the American Joint Committee on Cancer 6 (Table I). The Child-Pugh classification has five clinical measures of liver disease. Each was awarded a score of from to 3, where 3 indicated the most severe form of derangement (Table II). All the patients were classified as Child-Pugh class A to C, and evaluated by medical hepatologists before surgery. There were 38 men and four women with a mean age of 6.2 years (46 to 77). The mean duration of follow-up was months (2 days to 48 months). Surgery was undertaken for a pathological fracture in 3 patients and because of a high risk of fracture in 2. The rating system of Mirels 7 was used to predict the risk of a pathological fracture. The mean Mirels score in the 2 patients was.8 (9 to 2). There were 5 metastases in the upper limb with 2 in the humerus and three in the clavicle, and 27 in the lower limb with 24 in the femur and three in the pelvis. We obtained plain radiographs in patients with bone pain, and if metastases were suspected bone scans were undertaken. A single bone metastasis was seen in 4 patients and multiple metastases in 28. There was only one case of a solitary bone metastasis without any other tumour in any part of the body, including the primary site. In all cases the metastatic lesions were excised by curettage or wide resection followed by internal fixation by an intramedullary nail, a long plate, or a prosthesis. An intramedullary nail was usually used for a diaphyseal fracture, plate fixation for a fracture involving the metaphysis and arthroplasty with a prosthesis for a fracture involving the joint or the neck of the femur. Polymethylmethacrylate bone cement was used to fill defects. An intralesional surgical margin was obtained in 36 and a wide margin in six. The indications for wide resection included a single bone metastasis with a well-controlled hepatocellular carcinoma and no other internal metastases. The functional evaluation of patients was performed using the classification system of the International Society of Limb Salvage. 8 This included the six parameters of pain, function, emotional acceptance, the use of walking aids, walking ability and gait; each was scored from to 5, for a maximum score of 3. Data from the fracture and non-fracture groups were compared. Statistical analysis. The Kaplan-Meier method and the logrank test were used to calculate overall survival rates and differences in survival with respect to variables. Multivariate analysis was conducted using Cox s regression model to determine the most significant variables related to survival. In order to compare the functional data between the fracture and non-fracture groups, Student s unpaired t-test was used. All statistical analyses were performed using SPSS version 2. (SPSS Inc., Chicago, Illinois) with a p-value of.5 deemed to be significant. THE JOURNAL OF BONE AND JOINT SURGERY

3 SURVIVAL OF PATIENTS WITH SKELETAL METASTASES FROM HEPATOCELLULAR CARCINOMA AFTER SURGICAL MANAGEMENT 57 Table III. Details of the patients and the Kaplan-Meier analysis according to clinical variables Factor * Number Median survival (yrs) 95% CI p-value Gender Male to Female 4 6. to 9.7 Age (yrs) < to > to Viral agent HBV to HCV to 5.4 Alcohol to 7.2 Location Upper limb to Lower limb to 25.5 Pathological fracture to Impending fracture to 7.7 Number of lesions One to Multiple to.97 Pulmonary metastases Present to Absent to 5.22 Liver management Conservative 5 2 NA.39 Non-operative to 2.38 Operative to 8.47 T-stage (AJCC) 4 9 NA to to to to.8 AFP (mg/ml) < to > to.6 Portal or hepatic vein Invasion(-) to Invasion(+) to 5.39 Child-Pugh classification A to B or C 3 5. to 2.7 Estimated blood loss (cc) < to to to 3.3 > to 8.4 Surgical margin Intralesional to Wide to Radiation (-) to (+) to 7.28 * HBV, hepatitis B virus; HCV, hepatitis C virus; AJCC, American Joint Committee on Cancer; AFP, serum alphafetoprotein 95% CI, 95% confidence interval NA, not available VOL. 9-B, No., NOVEMBER 29

4 58 H. S. CHO, J. H. OH, I. HAN, H.-S. KIM Child-Pugh class A Child-Pugh class B or C Fig. Kaplan-Meier overall survival curve for all patients Fig. 2 Kaplan-Meier overall survival curves based on the Child-Pugh classification. Table IV. Independent prognostic factors determined by multivariate analysis (Cox regression analysis) Independent prognostic factors Relative risk (95% CI) p-value Number of skeletal metastases Single Multiple 3.76 (.36 to.42). Child-Pugh class A B or C 2.7 (.6 to 6.35) Single Multiple.2 Results The details of the 42 patients are given in Table III. Complications. Post-operative complications included two deep infections and four cases of delayed wound healing. The two deep infections occurred after hip replacement. One patient who had undergone curettage and reconstruction with a reinforcement cage for a periacetabular metastasis was managed by open drainage and debridement. The other with a bipolar hemiarthroplasty was treated by irrigation and debridement. None of the four patients with delayed wound healing required major surgical management. Survival of patients. The one-, two- and three-year survival rates after surgery for skeletal metastases from hepatocellular carcinoma were 42.2%, 25.8% and 9.8% respectively (Fig. ). The median survival time after operation was ten months (95% confidence interval (CI) 6.29 to 3.7). No patient lived for more than four years. We applied the Kaplan-Meier method and the log-rank test to each clinical variable to determine its impact on the survival rate. The anatomical location of the metastases, Fig. 3 Kaplan-Meier overall survival curves based on the number of skeletal lesions. the number of skeletal lesions, the level of alphafetoprotein, the Child-Pugh grade and the surgical margin were found to be associated with survival (Table III). Cox regression analysis identified the Child-Pugh grade and the number of skeletal lesions as independent prognostic factors (Table IV). Patients with Child-Pugh class-a disease were found to have a better prognosis than those with Child-Pugh class B or class C (Fig. 2). The survival rates of class-a patients at one and three years after surgery were 59.% and 27.8%, while that of class-b or class-c patients was 9.2% at one THE JOURNAL OF BONE AND JOINT SURGERY

5 SURVIVAL OF PATIENTS WITH SKELETAL METASTASES FROM HEPATOCELLULAR CARCINOMA AFTER SURGICAL MANAGEMENT 59 Child-Pugh A + single lesion Overall Curettage En bloc resection Fig. 4 Fig. 5a Kaplan-Meier survival of patients with a single bone lesion and of Child- Pugh class A against the overall survival of all patients. Curettage En bloc resection year, with only one patient living for more than one year after surgery. In terms of the numbers of skeletal metastases, patients with a single bone lesion had better survival than those with multiple lesions (Fig. 3). A subgroup with a single bone lesion and Child-Pugh class A had a much better survival rate of % and 69.7% at one and three years after surgery as compared with the overall survival (Fig. 4). The margin of surgical resection did not have any impact on survival. Even although en bloc resection seemed to confer longer survival in those with a single bone lesion, it had no advantage over curettage when taking the Child-Pugh classification into consideration (Fig. 5). Furthermore, the status of hepatocellular carcinoma and the presence of a pathological fracture were not found to be significantly associated with patient survival (p =.87 and p =.428 respectively). Post-operative mortality at 3 days. The 3-day post-operative mortality was.9% (5 of 42 patients, Table V). Two patients with American Joint Committee on Cancer T4-stage disease died from hepatic rupture and two from bleeding from cerebral metastases. The mean survival time of the four patients with cerebral metastasis was 3.25 months. One of the three Child-Pugh class-c patients died from hepatic failure three weeks after surgery. Local recurrence. There were three local recurrences (7.%), and all three patients had an intralesional resection of the metastasis (Table VI). One had a local recurrence at nine months after operation despite adjuvant radiotherapy and died one month later. Another local recurrence was seen ten months after surgery. This patient had not had post-operative radiotherapy, and therefore radiotherapy with helical tomography was undertaken. He was alive when last seen 6 months later and was able to perform Fig. 5b Kaplan-Meier overall survival curves based on the margin of surgical resection. For those with a single bone lesion, en bloc resection tended to confer a survival advantage over curettage (a), but showed no advantage over curettage when taking the Child-Pugh classification into consideration (b). daily activities. Another patient who had not received postoperative radiotherapy had a re-operation for local recurrence 5 months after the initial surgery. An en bloc resection was carried out and he was alive 38 months after the initial surgery without further local recurrence. Functional score. Functional evaluation was undertaken in the 36 patients who survived for more than two months after surgery using the classification system of the International Society of Limb Salvage. 8 There were 24 patients in the fracture and 2 in the non-fracture group. The mean score was 23.4 (7 to 28) in the fracture group and 22.8 (6 to 27) in the non-fracture group (p =.5). VOL. 9-B, No., NOVEMBER 29

6 5 H. S. CHO, J. H. OH, I. HAN, H.-S. KIM Table V. Clinical details of the patients who had early post-operative death Case Gender Age (yrs) Lesions Child-Pugh classification T-stage Internal organ Resection margin Cause of death M 5 7 A 4 None Curettage Hepatic rupture 2 M 46 2 B Brain, lung Curettage Cerebral haemorrhage 3 M 58 3 C 4 Brain Curettage Cerebral haemorrhage 4 M 53 C 3 None Curettage Hepatic failure 5 M 5 5 B 4 Lung Curettage Hepatic rupture Table VI. Clinical details of the patients with local recurrence Case Gender Age (yrs) Child-Pugh classification Resection Radiotherapy Time to recurrence (mths) Management Follow-up for recurrence (mths) Final status * M 58 B Curettage Yes 9 None DOD 2 M 64 A Curettage No 5 Revision 38 Lost to follow-up 3 M 72 A Curettage No Tomotherapy 6 AWD * DOD, died of disease; AWD, alive with disease Discussion Hepatocellular carcinoma has a dismal prognosis, because of its aggressiveness and its effect on liver function. Associated chronic liver disease indicates a poorer prognosis, increased peri-operative morbidity and mortality, and a higher recurrence rate of the tumour. Post-mortem studies of hepatocellular metastases have shown that they are most common in the thorax followed by the abdomen and bone. 3 A recent study by Fukutomi et al 7 showed that the incidence of skeletal metastases is rising because of the increased longevity due to advances in diagnosis and treatment. Skeletal metastases from hepatocellular carcinoma are highly destructive vascular lesions and severely affect the quality of life. 5 Radiotherapy may have a more important role in the treatment of symptomatic skeletal metastases from hepatocellular carcinoma than in those from other primary sites because of the presence of cirrhosis of the liver and the likelihood that narcotic drugs will precipitate hepatic coma. 9 Several reports have addressed the favourable effects of radiotherapy on skeletal metastases from hepatocellular carcinoma. Seong et al 8 noted that 73% and Kaizu et al 9 84% of their patients had relief from pain after radiotherapy. However, no preferred treatment for skeletal metastases has been established, 9,2 and surgical management is inevitable in patients with a pathological fracture, paralysis by cord compression, or intractable pain. Scoring systems have been introduced to assess the prognosis of patients with skeletal metastases. Tokuhashi et al 2 and Tomita et al 27 devised a system for spinal metastases from general cancers based on the findings of patients treated by surgery. Katagiri et al 3 described a system based on the primary tumour, visceral metastases, performance status, previous chemotherapy and multiple skeletal metastases. Lin et al 23 noted the prognosis after surgery for skeletal metastases from renal-cell carcinoma. They found that local control and implant stability were important factors in patients with a likelihood of longterm survival. Studies have addressed survival in patients with bone metastasis from carcinoma of the lung, 24 prostate 25 and breast, 26 but the prognosis after skeletal metastases from hepatocellular carcinoma after surgery has not been recorded. In our study, the Child-Pugh grade and the number of metastases were found to be independent prognostic factors for survival in patients with skeletal metastases from hepatocellular carcinoma after surgical management. The Child-Pugh scores are used to predict mortality during surgery. This system was originally designed to predict the prognosis of patients with portal hypertension during shunt placement. 4,5 However, since then its usefulness has been recognised in predicting the peri-operative risks in many forms of extrahepatic and hepatic surgery. In relation to hepatocellular carcinoma, the Child-Pugh classification is one of the most important prognostic factors. Park et al 27 concluded that it is an independent prognostic factor in hepatocellular carcinoma in addition to the stage, the serum level of alpha-fetoprotein and portal-vein thrombosis. Although there has been some debate over whether aggressive management of a solitary metastasis in various cancers confers gain in survival, 23,28,29 most authors have recognised that patients with a solitary bone metastasis have better survival than those with multiple metastases, regardless of the aggressiveness of treatment. 24,3,3 Therefore durable fixation and local control for the patient with a solitary bone lesion are needed to avoid complicating future management. There was only one solitary skeletal metastasis, defined as a bone metastasis with no tumour in any other part of the body including the primary site or with a primary lesion of resectable status, in our study. Surgical resection THE JOURNAL OF BONE AND JOINT SURGERY

7 SURVIVAL OF PATIENTS WITH SKELETAL METASTASES FROM HEPATOCELLULAR CARCINOMA AFTER SURGICAL MANAGEMENT 5 for a lesion in the liver is indicated in a small proportion of patients to preserve reasonable liver function. However, even after surgical resection, the rate of recurrence is high in the remaining cirrhotic liver and a true solitary bone metastasis is rare. We refer to a single bone metastasis with a well-controlled liver status as a solitary bone metastasis. In patients with a single bone metastasis, wide resection showed no advantage in survival over curettage in our study. However, considering that the rate of survival of patients with Child-Pugh class-a disease and a single bone metastasis was about 7% at three years after operation and that all local recurrences followed curettage, wide resection appears to be preferable in patients with a single bone lesion. Although the post-operative mortality at 3 days was.9%, only one patient died from deterioration in hepatic function related to the surgery, but four died from mass-related haemorrhage. Spontaneous rupture of hepatocellular carcinoma is sometimes encountered by clinicians and has a high rate of mortality whether rupture occurs in the liver or in metastatic sites such as the brain. This high rate of mortality is due to the hypervascularity of hepatocellular carcinoma and the presence of a coagulopathy which might promote excessive bleeding. Liu et al 35 assessed the risk factors for spontaneous rupture in hepatocellular carcinoma and found that a large tumour, a high level of serum alpha-fetoprotein, a low level of haemoglobin, a poor clotting profile and more advanced disease to be significant. These increase the risk of death after surgery. The aim of the management of patients with skeletal metastases is to improve their quality of life. Although surgery is indicated for a fracture, or if there is a risk of fracture, and for acute spinal paralysis, there are no standard guidelines for surgical management in other circumstances. In patients whose prognosis is poor, surgical intervention to relieve pain may not improve the overall quality of life and may lessen survival. Operation should be considered on an individual basis with a multidisciplinary approach after careful assessment. The limitations of our study are its retrospective nature and the limited number of patients enrolled. The numbers were limited because most symptomatic skeletal metastases are managed non-operatively. The median survival time of patients with skeletal metastases from hepatocellular carcinoma after surgical management was ten months. No patient lived for more than four years although those with a single bone metastasis and Child- Pugh class-a disease had a survival rate of 7% three years after surgery. Patients with risk factors for spontaneous rupture, with cerebral metastases or inadequate hepatic function have a poor post-operative prognosis and surgical aggressiveness to achieve local control and durable implantation should be considered carefully. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References. Tanaka H, Tsukuma H, Oshima A. Epidemiology of primary liver cancer in Japan. Gan To Kagaku Ryoho 2;28:5-4 (in Japanese). 2. El-Serag HB, Mason AC. Rising incidence of hepatocellular carcinoma in the United States. N Engl J Med 999;34: Lee YT, Geer DA. Primary liver cancer: pattern of metastasis. J Surg Oncol 987;36: Nakashima T, Okuda K, Kojiro M, et al. Pathology of hepatocellular carcinoma in Japan: 232 consecutive cases autopsied in ten years. Cancer 983;5: Loaw CC, Ng KT, Chen TJ, Liaw YF. Hepatocellular carcinoma presenting as bone metastasis. Cancer 989;64: No authors listed. Primary liver cancer in Japan: The Liver Cancer Study Group of Japan. Cancer 984;54: Fukutomi M, Yokota M, Chuman H, et al. Increased incidence of bone metastases in hepatocellular carcinoma. Eur J Gastroenterol Hepatol 2;3: Seong J, Koom WS, Park HC. Radiotherapy for painful bone metastases from hepatocellular carcinoma. Liver Int 25;25: Kaizu T, Karasawa K, Tanaka Y, et al. Radiotherapy for osseous metastases from hepatocellular carcinoma: a retrospective study of 57 patients. Am J Gastroenterol 998;93: Lam CM, Lo CM, Yuen WK, Liu CL, Fan ST. Prolonged survival in selective patients following surgical resection for pulmonary metastasis from hepatocellular carcinoma. Br J Surg 998;85: Arii S, Monden K, Niwano M, et al. Results of surgical treatment for recurrent hepatocellular carcinoma: comparison of outcome among patients with multicentric carcinogenesis, intrahepatic metastasis, and extrahepatic recurrence. J Hepatobiliary Pancreat Surg 998;5: Kurachi K, Suzuki S, Yokoi Y, et al. A 5-year survivor after resection of peritoneal metastases from pedunculated-type hepatocellular carcinoma. J Gastroenterol 22;37: Katagiri H, Takahashi M, Wakai K, et al. Prognostic factors and a scoring system for patients with skeletal metastasis. J Bone Joint Surg [Br] 25;87-B: Child CG, Turcotte JG. Surgery and portal hypertension. Major Probl Clin Surg 964;: Pugh RN, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 973;6: Green FL, Page DL, Fleming ID, et al. AJCC cancer staging manual. Sixth ed. New York: Springer, 22: Mirels H. Metastatic disease in long bones: a proposed scoring system for diagnosing impending pathologic fractures. Clin Orthop 989;249: Enneking WF, Dunham W, Gebhardt MC, Malawar M, Pritchard DJ. A system for the functional evaluation of reconstructive procedures after surgical treatment of tumors of the musculoskeletal system. Clin Orthop 993;286: Murakami R, Baba Y, Furusawa M, et al. Short communication: the value of embolization therapy in painful osseous metastases from hepatocellular carcinomas: comparative study with radiation therapy. Br J Radiol 996;69: Taki Y, Yamaoka Y, Takayasu T, et al. Bone metastases of hepatocellular carcinoma after liver resection. J Surg Oncol 992;5: Tokuhashi Y, Matsuzaki H, Toriyama S, Kawano H, Ohsaka S. Scoring system for the preoperative evaluation of metastatic spine tumor prognosis. Spine 99;5: Tomita K, Kawahara N, Kobayashi T, et al. Surgical strategy for spinal metastases. Spine 2;26: Lin PP, Mirza AN, Lewis VO, et al. Patient survival after surgery for osseous metastases from renal cell carcinoma. J Bone Joint Surg [Am] 27;89-A: Sugiura H, Yamada K, Sugiura T, Hida T, Mitsudomi T. Predictors of survival in patients with bone metastasis of lung cancer. Clin Orthop 28;466: Yamashita K. Denno K, Ueda T, et al. Prognostic significance of bone metastases in patients with metastatic prostate cancer. Cancer 993;7: Yamashita K, Ueda T, Komatsubara Y, et al. Breast cancer with bone-only metastases: visceral metastases-free rate in relation to anatomic distribution of bone metastases. Cancer 99;68: Park KW, Park JW, Cho SH, et al. Survival analysis for patients with hepatocellular carcinoma according to stage, liver function and treatment modalities. Korean J Hepatol 26;2:4-54 (in Korean). VOL. 9-B, No., NOVEMBER 29

8 52 H. S. CHO, J. H. OH, I. HAN, H.-S. KIM 28. Rubin P, Brasacchio R, Katz A. Solitary metastases: illusion versus reality. Semin Radiat Oncol 26;6: Jung ST, Ghert MA, Harrelson JM, Scully SP. Treatment of osseous metastases in patients with renal cell carcinoma. Clin Orthop 23;49: Koizumi M, Yoshimoto M, Kasumi F, Ogata E. Comparison between solitary and multiple skeletal metastatic lesions of breast cancer patients. Ann Oncol 23;4: Manabe J, Kawaguchi N, Matsumoto S, Tanizawa T. Surgical treatment of bone metastasis: indications and outcomes. Int J Clin Oncol 25;: Clarkston W, Inciardi M, Kirkpatrick S, et al. Acute hemoperitoneum from rupture of a hepatocellular carcinoma. J Clin Hastroenterol 988;: Chearanai O, Plengvanit U, Asavanich C, et al. Spontaneous rupture of primary hepatoma: report of 63 cases with particular reference to the pathogenesis and rationale treatment by hepatic artery ligation. Cancer 983;5: Kew MC, Dos Santos HA, Sherlock S. Diagnosis of primary cancer of the liver. Br Med J 97;4: Liu CL, Fan ST, Lo CM, et al. Management of spontaneous rupture of hepatocellular carcinoma: single-center experience. J Clin Oncol 2;9: THE JOURNAL OF BONE AND JOINT SURGERY