Prognostic factors of hepatocellular carcinoma in patients undergoing hepatic resection

GASTROENTEROLOGY

Prognostic Factors of Hepatocellular Undergoing Hepatic Resection RYOHEI IZUMI,* KOHICHI SHIMIZU,* TOHRU AKITAKA NONOMURA,” and ITSUO MIYAZAKI*

II,*

1994:106:720-727

Carcinoma in Patients

YAGI,*

MASAO

OSAMU

MATSUI,?

Second Department of *Surgery, ?Radiology, and 5Pathology, School of Medicine, Kanazawa University, Kanazawa, Japan

Background/Aims: Prognostic analysis on hepatocellular carcinoma (HCC) in patients undergoing hepatectomy is necessary to determine the clinical value of hepatectomy on prognosis. Methods: Survival and disease-free survival were analyzed in 104 HCC patients undergoing hepatectomy using clinicopathologic factors by univariate and multivariate analyses. The value of the International Union Against Cancer (UICC) TNM classification on prognosis was assessed in the patients. Results: In multivariate analysis, portal vein invasion was the most influential factor. The difference between stage 1 and 2 or stage 3 and 4A using UICC’s TNM classification was not significant with respect to survival or disease-free survival. The UICC’s classification was modified as follows; stage 1, solitary tumor without vascular invasion: stage 2, solitary or multiple tumor(s) involving adjacent to vessel branch; stage 3, tumor(s) involving major vessel branch or with regional lymph nodal metastasis; and stage 4, tumor(s) with distant metastasis. The differences between each stage in the modified classification were significant with respect to disease-free survival. Conclusions: The UICC’s TNM classification was not of prognostic significance. Further studies on survival in patients with HCC are necessary to evaluate the value of the UICC’s TNM classification; some modification may be necessary.

I

ment modality for HCC must be carefully assessed by evaluating long-term survival. A clinically useful staging system

based on proven

plasms’-3

cinoma

in diagnostic

imaging

and clinical screening

(HCC)

in high-risk

for hepatic

neo-

for hepatocellular

car-

patient

populations

have

made it possible to diagnose resectable small asymptomatic HCCs.‘-’ Improvements in preoperative estimation of operative risks of liver surgery in patients with impaired hepatic function”37 and technical advances in hepatic resection’-” have made hepatic resection safer. The clinical value of hepatectomy as a treatment modality for HCC has been reported from Eastern and Western countries. 12-15 Long-term survival after transcatheter arterial embolization” or percutaneous intratumoral ethanol injection therapy” Therefore, the clinical

has also been recently reported. value of hepatectomy as a treat-

variables

would

be

patients

hepatic resection. Many prognostic

variables47133’s2’9 have been reported

in patients many

undergoing

criteria

cially tumor of HCC

most likely to benefit from

hepatic

resection;

unfortunately,

were used to define these variables, size.52’0 Therefore,

is necessary

hepatic resection gested

that

(UICC)

staging

a uniform

to evaluate

International

Union

classification

UICC hepatic

factors.

staging

The

clinical

classification23

resection

value

of

have sug-

Against

Cancer

for HCC is useful for this

purpose.2”22 In the present study, survival free survival (DFS) were analyzed using pathological

espe-

classification

the clinical

for HCC. Recent publications

the

value

and diseaseclinical and

of the current

regarding

prognosis

after

was evaluated.

Patients and Methods Clinical records, actual operative specimens, and histological slides of all patients who had undergone hepatic resection for HCC at the Department of Surgery of Kanazawa University Hospital retrospective

were reviewed.

Patients

study when complete

were selected

resection

One hundred

tumor

and nontumorous

ten patients

went hepatic resection

without

1976 to May 1990; 1 patient hepatic

resection

without

documentation

liver was available.

distant

at the Department

metastasis

under-

of Surgery from May

with bone metastasis

resection

for this

of all macroscopic

HCC of the liver was achieved and histological of the resected

mprovements

prognostic

useful in identifying

of distant

underwent

metastases

fol-

lowed by systemic chemotherapy, and 1 patient with remote lymph nodal metastasis underwent hepatic resection with lymph

nodal dissection.

The 8 patients

who died within

30

days were excluded from the study. A total of 104 patients (88 men and 16 women) were included. The mean age (*SD) Abbreviations used in this paper: e-stage, extracted stage of TNM classification of UICC; DFS, disease-free survival; m-stage, modified stage of TNM classification of UICC; UICC, International Union Against Cancer. 0 1994 by the American Gastroenterological Association 0016-5065/94/$3.00

PROGNOSTIC FACTORS OF HCC AFTER HEPATIC RESECTION

March 1994

was 60.9 ? 11.2 years (range, resections,

major hepatic

12-80

resections

tients (23%). Of these 24 patients, proven

cirrhosis.

80 patients

Minor

hepatic

resections

had repeated

had repeated

hepatic

sus-host

in 24 pa-

11 (46%) had histologically

(77%). Of these 80 patients,

sis. Six patients patient

years). As initial hepatic were performed

were performed

resection

resection

of bleeding of 8.9%

in

73 (91%) had cirrho-

hepatic

1

once, and

twice because of recur-

into intrahepatic

HCC was analyzed without or extrahepatic

and risk of postoperative

recurrence

spect to nine clinicopathologic margin, tumors,

presence

degree

mondson

features:

of a biloma

positive

Disease-free

when noncancerous

the survival and DFS of patients and DFS analysis

surgical of por-

number

of

to Ed-

in the nontumor-

surgical

including

remote

margin

was con-

liver tissue was <5 histologically.

were studied

mm

hepatic

two subgroups. resection

computed

teriography,

examination

included

tant metastases, patients

Generalized

104 patients whereas

without

distant

was also analyzed

Wil-

from the date of

to the date when recurrent

or histological

disease was diag-

tomography,

including

hepatic

metastases. with

ar-

of the tumor.

Survival

2 patients

with dis-

DFS analysis was performed

in patients

by

the survival and DFS be-

nosed by ultrasonography, analysis

as described

DFS was measured

Survival

on 102

after recurrence

recurrence

The

bleeding,

one with formation

margin,

and one with a bile

3-year,

and

5-year,

7-year

rates in the patients

who underwent

were 65.4%,

58.8%,

and 40.6%,

year, 5-year,

and 7-year

and 13.4%,

respectively

were diagnosed without

cirrhosis.

overall

survival

hepatic

resection

respectively.

The

DFS rates were 28.9%, (Figure

in 74 patients: Recurrence

the most common

1). Recurrences was diagnosed

3-

20%, of HCC

60 with cirrhosis

in 30 patients

site for recurrence,

were other frequent

recurrent

rences were diagnosed

and I4 within

1

with cirrhosis

and

and bone and lung

sites. Extrahepatic

in 5 of 12 patients

recur-

(42%) without

cirrhosis but in only 4 of 60 patients (7%) with cirrhosis. In 2 patients with distant metastasis, hepatic recurrence developed

within

1 year after operation

Prognostic

to the

with results of log rank tests

into subgroups.

coxon test was used in comparing tween

two

leak.

Thereafter,

according

was performed

Kaplan and Meier and compared of patients

rate

failure,

7 without cirrhosis and after 3 years in 5 patients with cirrhosis and 1 without cirrhosis. The remnant liver was

re-

UICC TNM classification.

after division

liver

year after the operation

according

metastasis

from rhe liver cut end to the tumor

Survival

at the surgical

with

with

presence

of cirrhosis

of distant

lymph nodal metastasis.

ulcer. The morbidity

six patients

with postoperative

and one

Survival

disease-free

differentiation

and Steiner,*” presence

ous liver, and presence sidered

were evaluated

of hepatic vein invasion,

of cellular

strati-

recurrence.

tumor size, presence of encapsulation,

tal vein invasion,

from a duodenal

bleeding,

Survival and Recurrence

Postoperative follow-up consisted of a combination of serial a-fetoprotein sampling, ultrasound examination, and computed tomographic examination of the hepatic remnant, with hepatic angiography performed when recurrence was suspected. Investigations were performed at bimonthly intervals for the first postoperative year and every 3 months thereafter in 104 patients. fication

one of intracerebral

included

patients

rence in the hepatic remnant.

The presence of recurrent

disease,

721

(Table

Factors Related to Survival

Tumor size was significant in terms but not with respect to DFS. The difference between tumor

1).

of survival in survival

tumor size > 5 cm and that 52 cm and between size

was significant,

>

5 cm

and

that

from

2 to

5 cm

but the difference between

tumor

size or

2 cm and that from 2 to 5 cm was not significant (Figure 2A). The difference in DFS between tumor size zz 2 cm and that >5 cm was significant (Figure 2B). Portal vein invasion was a significant indicator of both survival and DFS, and all patients

with tumor

portal branch

recurrence

experienced

involving within

a major

1 year after

who received

therapeutic treatment. Significant variables in univariate analysis were chosen for Cox’s multivariate regression analysis.‘5 x2 was used for comparison recurrence was defined

of therapeutic

using the TNM classification.

treatments

Statistical

after

significance

as a P value < 0.05.

Results Operative Mortality

and Morbidity

Thirty-day mortality rate was 7.1%. Death was directly related to hepatic resection in five patients, resulting from postoperative bleeding in one patient and from postoperative liver or multiorgan failure in four. One patient died of blood transfusion-related graft-ver-

Flgure 1. Survival rate and DFS rate of patients undergoing hepatic resection for HCC.

722

IZUMI ET AL.

GASTROENTEROLOGY Vol. 106. No. 3

Table1. Recurrence of HCC in Patients Undergoing Hepatic Resection

Recurrence Negative Positive Recurrence site Liver Liver and lung Liver and bone Liver, lung, and bone Liver and lymph node Bone and adrenal gland Bone and lung

the hepatic operation

resection (Figure

cant in terms

Patients with liver cirrhosis

Patients without liver cirrhosis

Total

24 (28.6%) 60 (71.4%)

6 (30%) 14 (70%)

30 (28.9%) 74 (71.1%)

56 2 1 0 0

9 1 0 1 1

65 3 1 1 1

0 0

1 1

1

0

1 1 1

and died within

3). Hepatic

of survival

tumors

was a significant

groups

for both

significant

survival

difference

2 years after the

vein invasion

was signifi-

but not for DFS. Number indicator and DFS,

between

among but

a unicentric

the

there

of

three

was no

tumor

and

Figure 3. (A) Survival rate and (6) DFS rate of patients undergoing hepatic resection for HCC in relation to portal vein invasion. VPO, tumor(s) without vascular invasion; VPl, tumor(s) involving an adjacent peripheral branch of the portal vein(s); VP2, tumor(s) involving a major branch of the portal vein(s).

multicentric

tumors

involving

better

survival. Also, there was no significant multicentric

unilobar

disease in survival sis was a significant liver cirrhosis,

disease

lobes in terms difference

of

between

and multicentric

bilobar

and DFS. Presence of distant

metasta-

indicator

of survival.

presence of encapsulation,

Association surgical

of

margin

of tumor, and differentiation of tumor by Edmondson and Steiner were not significant indicators of long survival and DFS.

Multivariate Analysis of Prognostic Factors Significant

_. 10

I-

1

Figure 2. (A) Survival rate and (6) DFS rate of patients undergoing hepatic resection for HCC in relation to size of tumor. Ts, tumor size in greatest dimension.

prognostic

factors

identified

by uni-

variate analysis were entered into multivariate analysis using a Cox proportional hazards model. Of the five clinicopathological variables influencing survival, portal vein invasion was the most significant. Tumor size was also significant. Other variables did not show significant prognostic influence (Table 2). Excluding the 37 patients with tumor involving portal vein, no variable showed significant prognostic influence. In the 37 patients with tumor involving portal vein, tumor size was the most and only significant variable. Of the three clinicopathological variables influencing DFS, portal vein invasion was the

March 1994

PROGNOSTIC FACTORS OF HCC AFTER HEPATIC RESECTION

Table 2. Independent Survival

Prognostic

Identified

Variables

723

Influencing

by Cox Proportional

Hazards

Model Regression coefficient

Variables Tumor size Portal vein invasion Hepatic vein invasion No. of tumors Distant metastasis

only

significant

2.365419

5.197

0.02479

1.335245

2.986042

19.995

0.00002

5.721283 6.028678 1.383257

3.952774 2.497586 8.162908

2.095 0.058 2.872

0.15097 0.80976 0.09333

variable

portal

P

5.392500

(Table

DFS did not show significant or without

F

SE

3). Other

influence

variables

in patients

in with

vein invasion.

TNM Classification Pathological to classify

data at entry into the trial were used

IO2 patients

survival according

into DFS and 104 patients

to the UICC TNM classification.

of the patients had evidence of metastasis of lymph except 1 patient with regional and remote lymph metastasis.

In the UICC classification, metastasis.

1,

was performed

2, 3, and 4A. Significant

noted

in 74 patients

difference

who received

therapy

comparison

was noted except between

or 4A. Otherwise, was not significant, and median

difference

the difference

better

between

than stage 3 survival

stage 3. The difference

between

4A was not significant

in survival

In the TNM

classification

were classified further

stage 4B and stage 3 or (Figure

of UICC,

4).

stage 2, 3, or 4A

into two or three subgroups,

which

were subdivided according to factors such as tumor size, number of tumors, vascular invasion, and presence of distant

metastasis

(Table 4). Because portal vein invasion

time

DFS time of stage 2 were longer than those between

stage 3 and stage 4A

was also not significant, and both the median and DFS time of stage 4A were longer than

Table 3. Independent Disease-Free Proportional

Tumor size Portal vein invasion No. of tumors

and

stage 1 and 2

and both the median

of stage 1. The difference

Variables

in

stage 2 and stage

the stages. In both survival

DFS, stage 1 or 2 were significantly

was

after recur-

but no significant

Figure 4. (A) Survival rate and (6) DFS rate of patients undergoing hepatic resection for HCC in relation to the stages according to the UICC TNM classification. Results are expressed as mean 2 SD.

with stage

in comparison

four stages, between

to the UICC

on patients

rence with

4B in analysis

to the presence

DFS analysis according

TNM classification

nodes nodal

stage 4 are subdi-

vided to stage 4A and stage 4B according of distant

into None

Prognostic Survival Hazards

Regression coefficient

Variables Identified

survival those of

4. TNM Classification

Stage 1 Stage 2 Stage 3

Stage 4A Stage 48

Influencing

Tl T2 Tl T2 T3 T4 Any T

and Stage

Grouping NO NO Nl Nl NO, Nl NO, Nl Any N

of HCC MO MO MO MO MO MO Ml

by Cox

Model

SE

Table

F

P

1.750164

1.666536

1.103

0.29622

7.077904 2.337242

2.226387 1.810984

10.107 1.666

0.00198 0.19988

Tl, solitary, 52 cm, without vascular invasion. T2, solitary, 52 cm, with vascular invasion; multiple, one lobe, 52 cm, without vascular invasion; solitary, >2 cm, without vascular invasion. T3, solitary, >2 cm, with vascular invasion; multiple, one lobe, 52 cm, with vascular invasion; multiple, one lobe, >2 cm, with or without vascular invasion. T4, multiple, more than one lobe; invasion of major branch of portal or hepatic veins. Nl, regional. Ml, distant metastasis, including remote lymph nodal metastasis. NOTE. Classification according to UICC, 1987.

724

IZUMI ET AL.

GASTROENTEROLOGY Vol. 106, No. 3

without

vascular

categorized

invasion

into m-stage

sion in major

branch

nodal metastasis 4B including

in major hepatic with

was categorized

tases was categorized

were

2. Stage 4A with vascular inva-

and tumors

tumor(s)

branch

regional

into m-stage

with remote into m-stage

lymph

lymph 3. Stage

nodal metas-

4 (Table

5). Survival

and DFS were compared between stages according to the modified TNM classification. Because patients with distant

metastasis

modified Figure 5. Survival rate and DFS rate of patients undergoing hepatic resection. Stages according to the UICC classification: stage 1, solitary tumor 5 2 cm greatest dimension without vascular invasion; estage 2, solitary tumor > 2 cm in greatest dimension without vascular invasion.

TNM

tween m-stage different tients

1, m-stage

therapies

with

was the most significant 2 patients

invasion tients

1 or m-stage

between with

5). Three-year

invasion.

The difference

stage 1 patients

respect survival

to either

vascular

2) from the other pa-

and DFS of e-stage

survival

DFS (Figure

2 patients or stage 2

tween m-stage

was not

Survival

with

Five-year

stage 2 with vascular

who underwent

stage 3 patients,

but those of stage 2 patients

ple tumors or vascular invasion of stage 3 patients. There are two subgroups tumors

better

differences

3 or m-

between

the

and DFS of than those of with multi-

were not better than those

in more than one lobe (stage 4Aa) and the other

solitary or multiple tumors in one lobe involving a major branch of the portal or hepatic vein (stage 4Ab). Survival and DFS of patients with stage 4Aa were significantly better than those of stage 4Ab (Figure 6). The difference

bebe-

4, with respect to survival

of improvements survival

hepatic

resection

have

in diagnosis

and

rates in patients

resection

were

with HCC

33%-46%

in

Western

studies’8*26-28 and 25Yo-33% in Asian studies. 20,29-32 Whereas recurrences developed in many pa-

tients and the 5-year DFS was similar study,33 the survival

in stage 4A: one is multiple

differences

for the difference

rates of HCC after hepatic

treatment.

The survival

with m-stage

m-stage

Discussion because

invasion.

therapy

78).

increased

were significantly

with

except

3 and m-stage

vascular invasion were 60% and 0%, respectively, and survival of e-stage 2 was significantly better than that of e-stage 2 patients

in patients

were significant

the four m-stages,

(Figure

2

or DFS (Figure

and DFS of stage 2 patients

performed

2 than in patients

for pa-

Locoregional

7A). There were significant

tween

and with e-stage

on recurrence

stages.

be-

3. Significant

three stages, and differences between each stage were also significant in the modified classification with respect to

analysis,

without

with those of stage 1 patients

with vascular

significant

tumor

(e-stage

in stage 2. Survival

were compared

in multivariate

solitary

were extracted

patients patients

variable

with

were performed

was more frequently

into stage 4 in the

DFS was compared

2, and m-stage

four modified

stage 4. There stage

were categorized classification,

rate of patients

to that in another in the present

study

was better than in other reports. This was a result of the fact that patients who died within 30 days after the surgery were excluded in the present study, and that

was not significant between patients with stage 3 and patients with stage 4Aa in survival and DFS, but survival and DFS of patients with stage 3 were significantly better than those of patients with stage 4Ab. Because the stage according to the UICC TNM classification was not of clinical value in assessing the prognostic significance after resection in this study, modification of the TNM classification may be necessary to accurately predict the prognosis of HCC. In the modified TNM classification, stage 1 and stage 2 that included solitary tumors without vascular invasion were categorized into modified stage (m-stage) 1. Stage 2 with vascular invasion or intrahepatic metastases, stage 3, and stage 4A

Figure 6. Survival rate and DFS rate of patients undergoing hepatic resection. Stage according to the WCC classification: stage 4Aa, multiple tumors in more than one lobe; stage 4Ab, tumor(s) involving a major branch of the portal or hepatic vein(s).

March 1994

PROGNOSTIC FACTORS OF HCC AFTER HEPATIC RESECTION

Table 5. Modified

TNM Classification

and Stage

lated to the absence of significant

Grouping

of HCC Stage 1 Stage 2 Stage 3

Tl T2 T3 Any T Any T

Stage 4

NO NO NO Nl Any N

MO MO MO MO Ml

Tl, solitary without vascular invasion; T2, solitary or multiple with vascular invasion; T3, solitary or multiple, with invasion of major branch of portal or hepatic veins; Nl, regional; Ml, distant metastasis including remote lymph nodal metastasis.

vival and DFS between the present study.

out vascular

invasion

sion with vascular portal

(i.e., ethanol

gery, chemotherapy, ten performed

injection,

and arterial

on recurrence,

repeated

embolization)

resulting

sur-

were of-

recurrences

have

been reported, analyses based on both survival and DFS are better for identification of prognostic factors for HCC after hepatic survival

resection.

The

and recurrence analysis.

factor in many

reports,

survival invasion, encing

Tumor

or third

tumor

long-term

described

survival.

cance of a treatment Staging

influential

analysis.

Other

criteria

in

survival

to evaluate modality

for HCC

vein). Survival

factor

in

Yamanaka

et

le-

of patients

with tumor

the major hepatic

vessels was very poor. Fur-

in HCC associated

with cirrhosis,

with

may be unreasonable belonging

multicentric

was suggested to be present especially hepatitis virus infection. Therefore, it

to the

to consider same

these two subgroups

stage.

out the problems

Yamasaki

in defining

as

et a1.3” also

stage 4 in the UICC

classification.

The present study gives an account of an assessment of a modified TNM classification in some patients with cancer.

Modified

without

vascular

the

original

stage

1 included

invasion

classification.

stage 2 with vascular

tumor

or multicentric

in a major vessel branch (i.e.,

Modified

invasion

1 and stage 2

Modified

stage

or intrahepatic

and stage 3 and stage 4A without major vessel branch.

stage

or intrahepatic

vascular

metastases

in

2 included metastases invasion

stage 3 was restricted

in a to

factor was vascular factor influ-

reports”.3”m3” have also

that the size of the lesion

is necessary

study

size not a significant

on prognosis. Analysis of long-term criteria

most

that the most influential

with

factor

size was an influential

but in the present

and DFS in multivariate

al.” reported

influential

of HCC was portal vein invasion

in multivariate

size was the second

most

and unicentric

thermore,

TNM

of the survival after the recurrence. Because improvements in treating

in

involving

pointed

in prolongation

stage 1 and stage 2 patients

invasion

or hepatic

in patients therapies

in the sur-

On the other hand, two subgroups were present in stage 4A, i.e., multicentric bilobar disease with or with-

carcinogenesis’” various

differences

725

has little

using

influence

uniform

the prognostic

between

different

staging signifistudies.

have been developed37X3” but

not yet fully introduced UICC proposed a staging

to clinical practice. Recently system for HCC composed of

variables

relating

to tumor

number

of tumors,

with

size, vascular

invasion,

each factor having

almost

and the

same significance in defining the stage. Few reports21322 are available regarding long-term survival for HCC after hepatic resection according to the UICC classification. In the present study, the UICC classification failed to assess the prognostic significance because of the absence of a significant difference between stage 1 and 2 or stage 3 and 4 in long-term survival; also, evaluations were performed to examine the reasons for the failure to predict the prognosis. The survival and DFS of solitary tumor > 2 cm in diameter without vascular invasion, which was defined as stage 2, were similar to those of stage I and significantly longer than those of the stage 2 with vascular invasion. These results were closely re-

Figure 7. (A) Survival rate and (B) DFS rate of patients undergoing hepatic resection for HCC in relation to the stages according to the modified TNM classification.

726

GASTROENTEROLOGY Vol. 106, No. 3

IZUMI ET AL.

far-advanced

tumors

involving

a major hepatic vessel and

tumors with regional lymph nodal metastasis. Stage 48 was categorized into modified stage 4. In the modified classification,

the

long-term

survival

stage 2, and stage 3 was significantly cant difference

was not noted between

4 in survival.

Because

with

with

stage 4 in the modified

small

distant

surgery

HCC

classification

1,

stage 3 and stage

the number TNM

in

of patients

classification

was

Stage 3 and stage 4 in modified

were restricted

to far-advanced

therapy was not effective for these advanced fore, the difference

stage

A signifi-

was contraindicated

metastasis,

in this study.

among different.

was not noted

HCCs, and

between

show that the UICC

for HCC may contain prognostic

influence

Therefore,

further

in patients

TNM

some problems of tumor

clinical

studies

classification

in evaluating

on long-term hepatic

the

invasion. survival

resection

fication

of UICC resection.

and in choosing

good candidates

1. Takashima T, Matsui 0. Infusion hepatic angiography in the

3.

4.

5.

6.

7.

8.

9. 10. 11.

detection of small hepatocellular carcinoma. Radiology 1980; 136:321-321. Ohishi H, Uchida H, Yoshimura H, Ohne S, Ueda J, Katsuragi M, Matsuo N, Hosogi Y. Hepatocellular carcinoma detected by iodized oil: use of anticancer agents. Radiology 1985;154:2529. Matsui 0, Takashima T, Kadoya M, Suzuki M, Hirose J, Kameyama T, Choto S, Konishi H, Ida M, Yamaguchi A, lzumi R. Liver metastases from colorectal cancers: detection with CT during arterial portography. Radiology 1987; 165:65-69. Lai ECS, Ng IOL, You KT, Fan St, Mok FPT, Tan ESY, Wong J. Hepatic resection of small hepatocellular carcinoma: the Queen Mary Hospital experience. World J Surg 1991; 15:654-659. Tang ZY, Yu YQ, Hhou XD, Ma ZC, Yang R, Lu JZ, Lin ZY, Yang BH. Surgery of small hepatocellular carcinoma, analysis of 144 cases. Cancer 1989;64:536-541. Okamoto E, Kyo A, Yamanaka N, Tanaka N, Kuwata K. Prediction of the safe limits of hepatectomy by combined volumetric and functional measurements in patients with impaired hepatic function. Surgery 1984;95:586-591. Mizumoto R, Kawarada Y, Noguchi T. Pre-operative estimation of operative risk in liver surgery, with special reference to functional reserve of the remnant liver following major hepatic resection. Jpn J Surg 1979;9:343-349. lzumi R, Yabushita K, Shimizu K, Yagi M, Yamaguchi A, Konishi K, Nagakawa T, Miyazaki I. Hepatic resection using a water jet dissector. Surg Today 1993;23:31-35. Putnem CW. Techniques of ultrasonic dissection in resection of the liver. Surg Gynecol Obstet 1983; 157:475-478. Makuuchi M, Hasegawa H, Yamazaki S. Ultrasonically guided subsegmentectomy. Surg Gynecol Obstet 1985;161:346-350. Castaing D, Garden J, Bismuth H. Segmental liver resection using ultrasound-guided selective portal venous occlusion. Ann Surg 1989;210:20-23.

Franc0 D, Cappussotti L, Smadja C, Bouzari H, Meakins J, Kemeny F, Grange D, Dellepiane M. Resection of hepatocellular carcinomas: results in 72 European patients with cirrhosis. Gastroenterology 1990;98:733-738.

16. Yamashita Y, Takahashi M, Koga Y, Saito R, Nanakawa S, Hatanaka Y, Sato N, Nakashima K, Urata J, Toshizumi K, Ito K, Sumi S, Kan M. Prognostic factors in the treatment of hepatocellular carcinoma with transcatheter arterial embolization and arterial infusion. Cancer 1991;67:385-391. 17.

Dusheiko GM, Hobbs KEF, Dick R, Burroughs AK. Treatment of small hepatocellular carcinoma. Lancet 1992;340:285-288.

18.

Kemeny F, Vadrot J, Wu A, Smadja C, Meakins J, Franc0 D. Morphological and histological features of resected hepatocellular carcinoma in cirrhotic patients in the west. Hepatology 1989;9:253-257.

for

References

2.

15.

are

necessary to evaluate the clinical value of the TNM classihepatic

14. Gozzetti G, Mazziotti A, Cavallari A, Bellusci R, Bolondi L, Grigioni W, Bragaglia R, Grazi GL, Raffele ED. Clinical experience with hepatic resections for hepatocellular carcinoma in patients with cirrhosis. Surg Gynecol Obstet 1988;166:503-510.

these two

size and vascular

with HCC undergoing

13. Yamanaka N, Okamoto E, Toyosaka A, Mitunobu M, Fujihara S, Kato T, Fujimoto J, Oriyama T, Furukawa K, Kawamura E. Prognostic factors after hepatectomy for hepatocellular carcinomas, an univariate and multivariate analysis. Cancer 1991;65:11041110.

HCCs; there-

stages. The results

12. Okuda K, Ohtsuki T, Obata H, Tomimatsu M, Okazaki N, Hasegawa H, Nakajima Y, Ohnishi K. Natural history of hepatocellular carcinoma and prognosis in relation to treatment, study of 850 patients. Cancer 1985;56:918-928.

19. Hsu HC, Sheu JC, Lin YH, Chen DS, Lee CS, Hwang LY, Beasley P. Prognostic histologic features or resected small hepatocellular carcinoma (HCC) in Taiwan: a comparison with resected large HCC. Cancer 1985;56:672-680. 20. Suenaga M, Nakao A, Harada A, Nonami T, Okada Y, Sugiura H, Uehara S, Takagi H. Hepatic resection for hepatocellular carcinoma. World J Surg 1992;16:97-105. 21.

Ringe B, Pichlmayr R, Wittekind C, Tusch G. Surgical treatment of hepatocellular carcinoma: experience with liver resection and transplantation in 198 patients. World J Surg 1991;15:270285.

22.

lwatsuki S, Starzl TE, Sheahan DG, Yokoyama I, Demetris Al, Todo S, Tzakis AG, Van Thiel DH, Carr B, Selby R, Madariaga J. Hepatic resection versus transplantation for hepatocellular carcinoma. Ann Surg 1991; 214:221-229.

23. Spiessl B, Beahrs OH, Hermanek P, Hutter RVP, Scheibe 0, Sobin LH, Wagner G, eds. Union lnternationale Contre le Cancer. TNM atlas illustrated guide to the TNM/pTNMclassification of malignant tumors. 3rd ed. Berlin: Springer-Verlag, 1989:98-105. 24. Edmondson HA, Steiner PE. Primary carcinoma of the liver: a study of 100 cases among 48900 necropsies. Cancer 1954; 7:462-503. 25.

Cox DR. Regression B1972;34:187-220.

26.

lwatsuki S, Shaw BW, Starzl TE. Experience with 150 liver resections. Ann Surg 1983; 197:247-253.

27.

Thompson HH, Tompkins RK, Longmire WP. Major hepatic resection, a 25-year experience. Ann Surg 1983;197:375-388.

models

and life tables.

J R Stat Sot

28. Sesto ME, Vogt DP, Hermann RE. Hepatic resection in 128 patients: a 24.year experience. Surgery 1987; 102:846-851. 29. Chen MF, Hwang TL, Jeng LBB, Jan YY, Wang CS, Chou FF. Hepatic resection in 120 patients with hepatocellular carcinoma. Arch Surg 1989; 124:1025-1028. 30.

Kanematsu T, Matsumata T, Takenaka K, Yoshida Y, Higashi H, Sugimachi K. Clinical management of recurrent hepatocellular carcinoma after primary resection. Br J Surg 1988; 75:203-206.

31. Tsuzuki T, Sugioka A, Ueda M, lida S, Kanai T, Yoshii H, Nakayasu

March 1994

32.

33.

34.

35.

36.

carcinoma. Surgery K. Hepatic resection for hepatocellular 1990; 107:511-520. Nagao T, Goto S, Kawano N, lnoue S, Mizuta T, Morioka Y, Omori Y. Hepatic resection for hepatocellular carcinoma, clinical features and long-term prognosis. Ann Surg 1987; 205:33-40. Arii S, Tanaka J, Yamazoe Y, Minematsu S, Morino T, Fujita K, Maetani S. Tobe T. Predictive factors for intrahepatic recurrence of hepatocellular carcinoma after partial hepatectomy. Cancer 1992;69:913-919. Nagasue N, Yukaya H. Ogawa Y, Sasaki Y, Chang YC, Niimi K. Clinical experience with 118 hepatic resections for hepatocellular carcinoma. Surgery 1986;99:694-701. Hsu HC, Wu ll, Wu MZ, Sheu JC, Lee CS, Chen D.S. Tumor invasiveness and prognosis in resected hepatocellular carcinoma. Cancer 1988;61:2095-2099. Kanematsu T, Takenaka K, Matsumata T, Furuta T, Sugimachi K. lnokuchi K. Limited hepatic resection effective for selected cirrhotic patients with primary liver cancer. Ann Surg 1984; 1995-56.

PROGNOSTIC FACTORS OF HCC AFTER HEPATIC RESECTION

727

37.

Funovica JM, Fritsch A, Herbst F, Piza F, Muhlbacher F, Langle F, Schemper M. Primaty hepatic cancer--the role of limited resection and total hepatectomy with orthotopic liver placement. Hepatogastroenterology 1988; 35:316-320. 38. Beahrs OH, Myers MH. Manual for staging of cancer. 3rd ed. Philadelphia: Lippincott, 1987:87-92. 39. Sakamoto M, Hirohashi S, Tsuda H, Shimosato Y, Makuuchi M, Hosoda Y. Multicentric independent development of hepatocellular carcinoma revealed by analysis of hepatitis B virus integration pattern. Am J Surg Pathol 1989;13:1064-1067. 40. Yamasaki S, Takayama T, Hasegawa H, Sugahara K, Makuuchi M. The staging system of primary liver cancer (in Japanese). Jpn J Gastroenterol Surg 1990;91:1337-1339.

Received February 16, 1993. Accepted October 19, 1993. Address requests for reprints to: Ryohei Izumi, M.D., Second De pattment of Surgery, School of Medicine, Kanazawa University, 131, Takara-Machi, Kanazawa, 920, Japan. Fax: (81) 762-32-6460.