Seminars in Surgical Oncology 6:3641 (1990)

Surgical Strategy for Primary Hepatocellular Carcinoma Associated With Cirrhosis TAKASHI KANEMATSU, MD, KEN SHIMBE, MD,AND KElZO SUGIMACHI, MD,FACS From the Department of Surgery 11, Faculty of Medicine, Kyushu University, Fukuoka, Japan ~




We evaluated the applicability of limited hepatic resection in cases of hepatocellular carcinoma (HCC) in cirrhotic patients. According to the seventy of impaired hepatocellular function, 37 patients underwent limited resection, and for 13, standard major hepatic resection was done. There were no significant differences in the mortality and survival rates between the two groups. This limited resection can be selectively used to treat cancer associated with cirrhosis and encapsulated tumors. For further evaluation of this limited procedure, an attempt was made to elucidate the manner in which the surgical margin is linked to a recurrence after curative resection for hepatocellular carcinoma in patients with cirrhosis. Forty patients were divided into two groups; those in whom the margin from the cut surface to HCC in the fresh specimen was less than 10 mm wide [TW( + ) ] and patients in whom the surgical margin was equal to or exceeded 10 mm [TW( -)I. There were 24 and 16 patients in the TW( + ) and TW( - ) groups, respectively. There was no statistically significant difference in the rate of recurrence between the two groups. In patients with a tumor less than 4 cm in diameter, the extent of TW is not linked to an early recurrence. However, when the tumor size exceeds 4 cm, 10 mm of TW is inadequate to achieve curability. When a wide resection is not feasible, then adjuvant therapy should be aggressive. KEYWORDS:limited hepatic resection, complication, actuarial survival

INTRODUCTION The management of primary hepatocellular carcinoma (HCC) is a difficult task, particularly in the presence of coexisting cirrhosis. The cirrhotic patient is most vulnerable to risk regarding complications of the surgery [ 11. A poor prognosis associated with major hepatic resection in the presence of cirrhosis and due to poor regeneration of residual liver is also well known [2]. Thus, cirrhosis is considered to be a limiting factor to massive excision and extensive surgery may be unduly hazardous and lifethreatening in such patients. Recent studies have paved the way for development of techniques and indications for liver resection for HCC associated cirrhosis [3-51. In 1984, we advocated “limited hepatic resection” for HCC in patients with cirrhosis. This terminology related to a technique involving enuckation of the tumor in which less than 10 mm of liver tissue surrounding the lesion is removed [6]. This procedure led to a comparably longer survival compared 0 1990 Wiley-Liss, Inc.

with the results achieved by standard hepatic lobectomy or segmental resection. And, it has been the policy of our surgical team to limited resection in cirrhotic patients with HCC in whom hepatocellular reserve was low. In this procedure, surgical margin from the tumor edge to the cut surface should be important for curability. We searched for possible correlations between the surgical margin in patients with HCC associated with cirrhosis and curative hepatectomies and recurrence.

CLINICAL MATERIALS The Validity of Limited Hepatic Resection in Cirrhotic Patients with HCC We studied 50 Japanese patients with a less than 8 cm HCC associated with cirrhosis and who underwent curAddress reprint requests to Dr. Keizo Sugimachi, Department of Surgery 11, Faculty of Medicine, Kyushu University, Fukuoka 812, Japan.

Limited Hepatic Resection for HCC


ative hepatic resection. In these 50 patients, more than 1 l0OL cm of liver parenchyma surrounding the tumor was removed with formal major hepatic resection when clinical status and hepatocellular reserve were adequate. In cases of a small nodule, insufficient liver function, and/or poor clinical status, limited hepatic resection was usually performed. Documentation of adequate liver function is 5 1 mandatory for hepatic surgery. We mainly used the dye retention test utilizing indocyanine green (ICG). Blood -0-0-0 chemistry was also carefully monitored. In general, patients with an ICG over 30% (normal 0-10%), serum albumin below 3.5 g/dl, and total bilirubin over 1.5 mg/ dl were considered poor risks for major hepatic resec0 tion. In these patients with severely impared liver func0 1 2 3 4 5 tion, limited hepatic resection was appraised, with refYEARS AFTER SURGERY erence to the degree of hepatocellular reserve. After 1. A comparison of actuarial survival of patients undergoing surgery, the choice of adjuvant chemotherapy was de- Fig. limited hepatic resection with those subjected to the standard major pended on the clinical status. Some patients were ran- resection. There was no statistical significance in the postoperative domized and prescribed daily, oral 5-fluorouracil 600- survival rate. Reprinted with permission from Annals of Surgery 199: 51-56, copyright 1984 by J.B. Lippincott Company. 800 mg, as long as tolerated.

Correlations Between the Surgical Margin, and Curative Resections and Recurrence For entry into this study, patients had to have a histologically verified cirrhosis of the liver and a solitary HCC of less than 10 cm in diameter, without evidence of metastases. Patients were followed closely, and those who died of causes other than cancer were excluded from the study. Forty patients met all the above-mentioned criteria. The resected liver specimen was minutely examined by certified pathologists. The specimen was cut at a thickness of 1.O cm. According to the minimal width of the surgical margin in the fresh resected specimen, the patients were divided in to two groups. When there was TABLE I. Complications and Operative Death Complicatiodoperative death Complications Pleural effusion Intraperitoneal infection Upper GI bleeding Wound infection Intraabdominal bleeding Hyperbilirubinemia Intractable ascites Operative Death Peritonitis Liver failure Upper GI bleeding Sepsis Total Mortality rate (W)

Limited (n = 37)

Standard (n = 13)

5 3 3 3 2 2

1 1 1 0 1 0 0


2 1 1 0

0 1 0 1

4 10.8

2 15.4

Reprinted with permission from Annals of Surgery 199: 51-56, copyright 1984 by J.B. Lippincott Company.

less than a 10 mm surgical margin, this was defined as tumor wedge positive [TW( +)I. On the contrary, when a surgical margin equal to or more than 10 mm was left, this was expressed as tumor wedge negative [TW( -)I. This classification has been advocated by the Liver CanTABLE 11. Category Score Categories Age 55 > 55 I Liver function BSP 30% > BSP 30% 5 Tumor size 4cm > 4cm 5 Nuber of nodules Single Multiple Histologic classification I, I1 111, IV Chemotherapy Yes No Capsulea

(+ +) (+I (-)

Number of patients

Number of recurrences

Category score

10 16

4 6

-0'08971 0.0897 1



13 13

3 7

-0'14180 +0.07507


9 17

5 5

-0'12773 0.33209 + 0.20436

16 10

7 3

- 0.12628

17 9

7 3

+0.01718 o,07443 - 0.05725





9 7 10

0 3 7





0.11 145

- 1.23482

+O. 19112 2.21237 +0.97755

"Capsule: ( + + ), the mass is firmly encapsulated with fibrous tissue. (+), the mass is surrounded by fibrous casule, but incomplete. (-), no capsule. Reprinted from Annals of Surgery 199: 51-56, copyright 1984 by J.B. Lippincott Company.


Kanematsu and Sugimachi

cer Study Group of Japan. In addition, the presence of a capsule surrounding the tumor was carefully documented. After macroscopic examinations, the specimens were fixed in 10%formalin and then cut and prepared for microscopic examination with hematoxylin and eosin stain. Histologically, the classification of Edmondson and Steiner was used [7]. The patients were seen regularly every 1 or 2 months during the first 3 postoperative years, and at 6-month intervals thereafter.

dysfunction, tumor size, number of tumor nodules, histologic classification by Edmondson and Steiner, chemotherapy, and capsule of the tumor.

RESULTS The Validity of Limited Hepatic Resection in Cirrhotic Patients with HCC

Among the 50 cirrhotic patients, 37 underwent limited hepatic resection and, for 13, the standard major hepatic resection was done, including six right lobectomies and seven segmental resections. There were 29 men and STATISTICAL ANALYSES eight women with an average age of 55.2 years (range of Statistical evaluation was made using Student’s t test 40-74) in the limited group, and ten men and three for unpaired observations. Chi square analysis with women with an average age of 52.4 years (range of 34Yates’ correlation factor was also made. Survival was 61) in the standard group. Impairment of hepatocellular analyzed by Kaplan and Meier’s actuarial method. Prob- function was more severe in the limited group, particuability values for comparing two sets of life-table data larly with regard to serum albumin levels (3.4 2 0.5 vs were determined using the generalized Wilcoxon test. 3.8 k 0.5 g/dl; P < 0.05). Although there was no staThe discriminant analysis method was applied to calcu- tistically significant difference, ICG was 30.7 11.8% late the score that would quantitatively express the re- vs. 25.1 12.7%, and total bilirubin was 1.3 0.5 vs. currence risk. For this calculation of the score, the Quan- 1.0 0.4 mg/dl in the limited and standard groups, retification Methods I1 proposed by Hayashi was used [8]. spectively. The risk factors considered were age, degree of liver Table I lists complications after each type of hepatic



Fig. 2. Resection using the limited procedure. A fm fibrous capsule is present around the mass. Note the severe cirrhotic pattern in the liver parenchyma. Reprinted from Annals of Surgery 199: 51-56, copyright 1984 by J.B. Lippincott Company.

* *

Limited Hepatic Resection for HCC


Fig. 3. Microscopy of the hepatocellular carcinoma (HCC), which is surrounded by a firm fibrous capsule (C) (H&E, X 12). Reprinted from Annals of Surgery 199: 51-56, copyright 1984 by J.B. Lippincott Company.

resection. The frequency of pleural effusion was relatively high. Regarding postoperative mortality in the two groups, intra-abdominal infection was a contributing factor in the death of two patients with limited hepatic resection. Upper gastrointestinal bleeding from peptic ulcers occurred in one patient with limited resection, and another died of liver failure. In the standard group, an extensive sepsis led to death in one patient, and another died of liver failure. The mortality rate in the limited group was 10.8%, and in cases of major hepatic resection was 15.4%, with no statistically significant difference (x2 = 0.00354). The Kaplan-Meier life-table estimate of cumulative survival at 1 , 2, and 5 years for those who underwent hepatic resection with limited type was 79.9, 60.3, and 32.6%, respectively. However, rates for individuals undergoing standard operation were 78.7, 67.5, and 22.5%, respectively. There were no statistically significant differences in the survival rates following resection of primary liver cancer with each procedure, as illustrated in Figure 1. There was a recurrence in nine in the limited group in

the interval between 8 and 69 months, and four patients who underwent major hepatic resection died of recurrent disease from 8 to 39 months after the surgery. One patient treated with the limited procedure has survived, but a recurrence is evident. In order to clarify factors related to the recurrence risk in the limited group, discriminant analysis was performed. The results obtained are shown in Table 11. The score of the respective factor varies with zero as the midpoint: the lower the score, the higher the discriminant power which indicates no recurrence, and conversely, the higher the score, the higher the discriminant power indicating recurrence. The range of each score means that the category with a wider range is a more important discriminant one in predicting the likelihood of recurrence. The factor that has the highest power to discriminate recurrence is the capsule ( - ) category. Contrarily, the factor with a higher power to discriminate no recurrence is the capsule ( + + ) category. The wide range in the category of capsule means that the presence of a tumor capsule enables only a limited discrimination with regard to recurrence and nonrecurrence. Macro- and mi-

Kanematsu and Sugimachi



cro-scopic findings of fibrous capsule are shown in Figures 2 and 3.

Correlation Between the Surgical Margin and Curative Hepatic Resection and Recurrences



" %.


0 8 .




0 0 0

4 .

2 .

0 0


coo* 0 0 .




0 0


0 3 3 0

Tumor Size

B Tw (mm)

50. 0






20. 0 10








DISCUSSION There was no significant difference in the incidence of operative mortality when less extensive resection was


6 -

0 0 0

4 -

2. 0 .


Of these 40 patients, there were 24 in the TW( + ) group and 16 in the TW( - ) group. Mean value of TW was 3.3 mm and 18.0 mm in the T W ( + ) and TW(-) groups, respectively. Until February 1988, recurrence was evident in ten of 24 patients (41.7%) in the TW( + ) group, and in eight of 16 patients (50.9%) in the TW(-) group, with no statistical significance. The relationship among tumor size, value of TW, and recurrence was examined according to the disease-free interval: the disease-free interval was either within 24 months after the first operation (group 1) or exceeded 24 months (group 2). Figures 4A and B show the results for all cases and for group 1, respectively. In group 1, for those patients with a tumor less than 4 cm, the recurrence rate was 23% (three of 13 patients) and 0% (none of three patients) in TW( +) and TW( -) groups, respectively. However, in those patients with a tumor equal to or more than 4 crn, the recurrence rate was 100% (three of three patients) and 80% (four of five patients) in the TW( + ) and TW( - ) groups. In group 2, there was no particular related tendency. Of the 18 patients with recurrence in both groups, there was a recurrence arising from the liver remnant adjacent to the segment where the resected tumor had been located in three patients of the TW( +) group. These tumor sizes were 3, 5, and 7 cm, and the margin from the cut surface to HCC were 1, 4, and 5 mm, respectively. The recurrences occurred at 7, 7, and 86 months after the hepatic resection, respectively. Conversely, recurrence arising from the adjacent segments related to the primary tumor did not occur in the TW( - ) group. In the remaining 15 patients, there was no recurrence in the first resected stump; rather it was in different segments. In one patient with a tumor 4.3 cm in diameter, a recurrence was detected 20 months later. The surgical margin in this patient was 48 mm.





c o o

T u m o r Size


Fig. 4 (A and B). Relationship between tumor size and the surgical margin (TW) from the cut surface to HCC with regard to recurrence. Open circles indicate recurrence-free cases and closed circles indicate recurrent cases. Asterisks show recurrence from the resected hepatic stump. A, all cases. B, group 1. Reprinted from Annals of Surgery 209: 297-301, copyright 1989 by J.B. Lippincott Company.

Limited Hepatic Resection for HCC


used, as compared to that of the standard group. Survival than 4 cm, the extent of TW is not linked to a relatively rates achieved by limited hepatic resection and major early recurrence within 24 months after curative surgical liver resection were almost equal. In terms of recurrence management. In those patients with a tumor equal to or in the limited group, the presence of a tumor capsule more than 4 cm, there was a recurrence even with a 10 appeared to be the most important factor. The frequency mm surgical margin. Thus, the factor “TW” was not a of this type of liver cancer is relatively high in Japan [9]. significant determinant of early recurrence within 2 It is still unclear why this type of liver cancer is more years. common in Japan. It seems that the thickness of the These results suggest that in cirrhotic patients with a capsule relative to the mass size is indicative of the slow- poor hepatocellular reserve, resection of small liver canness of the tumor growth. There are other types of HCC cer, with limited procedure, enhances the possibility of that do not have a thick capsule which probably grow cure. Providing that the patients for elective surgery are faster with little time for formation of a capsule [9]. carefully screened, the life span can probably be exThus, the limited hepatic resection seems to be the pro- tended, with limited hepatic resection. cedure of choice, especially when there is a welldemarcated liver cancer nodule. A wide excision should REFERENCES be done for noncapsulated mass. 1. Ong GB, Cham PKW: Primary carcinoma of the liver. Surg GyRegarding a curative hepatic resection for HCC, Okanecol Obstet 143:31-38, 1976. mot0 et al. [lo] stated that the surgical margin had to 2. Inouye AA, Whelan TJ, Jr: Primary liver cancer: A review of 205 cases in Hawaii. Am J Surg 13853-61, 1979. exceed 10 mm through the thick noncancerous liver parenchyma. Lee et al. [ l l ] suggested that the adequate 3. Makuuchi M, Hasegawa H, Yamazaki S: Ultrasonically guided subsegmentectomy. Surg Gynecol Obstet 161:34&350, 1985. margin was at least 20 mm. In another series, those same 4. Andrus CH, Kamiski DL: Segmental hepatic resection utilizing the ultrasonic dissector. Arch Surg 121:515-521, 1986. authors examined the “adequate margin” of tumor spec5. Traynor 0, Castaing D, Bismuth H: Peroperative ultrasonography imens based on a study of 103 patients with histologiin the surgery of hepatic tumours. Br J Surg 75:197-202, 1988. cally verified HCC, with or without liver cirrhosis, in- 6. Kanematsu T, Takenaka K, Matsumata T, et al: Limited hepatic resection effective for selected cirrhotic patients with primary cluding both single and two nodules [ 121. After defining the adequate and the inadequate margins as more than or 7. liver cancer. Ann Surg 199:51-56, 1984. Edmondson HA, Steiner PE: Primary carcinoma of the liver: A equal to 10 mm and less than 10 mm from cut surface to study of 100 cases among 48,900 necropsies. Cancer 4:462-503, 1954. the tumor margin, respectively, they found a signifi8. Hayashi C: On the predictions of phenomena from qualitative data cantly longer survival in patients with an adequate marand the quantifications of qualitative data from the mathematicogin. There is no consensus regarding an adequate surgistatistical point of view. Ann Inst Statist Math 3:69-98, 1952. cal margin for curative operation for HCC associated 9. Okuda K, Musha H, Nakajima Y, et al: Clinicopathologic features of encapsulated hepatocellular carcinoma: A study of 26 with cirrhosis. In patients with severely impaired hepacases. Cancer 40:124&1245, 1977. tocellular reserve or those with a tumor close to large 10. Okamoto E, Tanaka N, Yamanaka N, Toyosaka A: Results of surgical treatments of primary hepatocellular carcinoma: Some vascular structures, an adequate surgical margin may not aspects to improve long-term survival. World J Surg 8:360-366, be feasible. In cirrhotic patients, the balance between 1984. postoperative hepatocellular reserve and surgical curabil- 11. Lee CS, Chao CC, Lin TY: Partial hepatectomy on cirrhotic liver with a right lateral tumor. Surgery 98:942-948, 1985. ity will greatly influence the prognosis [13-151. The results achieved in our study showed that in the 12. Lee CS, Sung JL, Hwang LY, et al: Surgical treatment of 109 patients with symptomatic and asymptomatic hepatocellular cargroup with less than a 24-month disease-free interval, cinoma. Surgery 99:481490, 1986. there was little difference between TW( ) and TW( - ) 13. Bismuth H, Houssin D, Ornowski J, Meriggi F: Liver resections in cirrhotic patients: A Western experience. World J Surg 10:311groups in recurrence rate and disease-free period, regard317, 1986. less of size of the tumor. However, in this group with 14. Bismuth H, Houssin D, Mazmanian G: Postoperative liver insufficiency: Prevention and management. World J Surg 7:505-5 10, less than a 24-month disease-free interval, there is the 1983. possibility that a surgical margin of more than 5 mm was 15. Lin TY, Lee C S , Chen KM, Chen CC: Role of surgery in the enough to prevent recurrence in those patients with a treatment of primary carcinoma of the liver: A 31-year experience. Br J Surg 74:839-842, 1987. tumor less than 4 cm in diameter. When the tumor is less


Surgical strategy for primary hepatocellular carcinoma associated with cirrhosis.

We evaluated the applicability of limited hepatic resection in cases of hepatocellular carcinoma (HCC) in cirrhotic patients. According to the severit...
1MB Sizes 0 Downloads 0 Views