Safe and Permissible Limits of Hepatectomy in Obstructive Jaundice Patients

Tetsuya Takahashi, M.D., Shinji Togo, M.D., Ph.D., Kuniya Tanaka, M.D., Ph.D., Itaru Endo, M.D.,
Yoshiro Fujii, M.D., Hiroshi Shimada, M.D., Ph.D.
Second Department of Surgery, Yokohama City University School of Medicine, 3-9 Fukuura, Kanazawa-ku, 236-0004 Yokohama, Japan
Published Online: April 19, 2004
Abstract. The safe and permissible limits of hepatectomy in obstructive
jaundice patients and the usefulness of preoperative portal embolization
(PE) for increasing the limit for safe hepatectomy were examined. We clas-
sified 416 patients with hepatectomy performed over 9 years under the fol-
lowing headings: normal liver function (n = 242); chronic hepatitis (n =
71); liver cirrhosis (n = 64); and liver after relief of obstructive jaundice
(n = 39). Hepatectomy was done after the total bilirubin level was reduced
below 3 mg/dl by preoperative biliary drainage. Factors influencing the
maximum total bilirubin level measured within 2 weeks after hepatectomy
were investigated, and this level was taken to reflect the degree of surgical
stress. PE was carried out in 18 patients with obstructive jaundice. The
maximum total bilirubin, expressed as a logarithm, was significantly cor-
related with the percent of liver resected in all groups. Hepatectomy fol-
lowed by a maximum total bilirubin of less than 8.5 mg/dl was accepted as
safe, and hepatectomy followed by a bilirubin level of 14.4 mg/dl was
deemed the maximum permissible resection. On the basis of these results,
the safe and permissible limits of hepatectomy in patients with obstructive
jaundice were 48.7% and 71.6%, respectively. PE decreased the maximum
total bilirubin from 8.5 mg/dl to 3.9 mg/dl when 48.7% of the liver (a safe
proportion in all cases) was resected; PE increased the safe limit of hepa-
tectomy from 48.7% to 67.4% when a maximum posthepatectomy total bil-
irubin level of 8.5 mg/dl was accepted as safe.
Hepatectomy has become a safe operative procedure owing to re-
cent improvements in the assessment of liver functional reserve,
surgical technique, and perioperative treatment. However, ex-
tended hepatectomy, especially in patients with liver cirrhosis or
obstructive jaundice, has sometimes been accompanied by fatal
liver failure.
Many articles have referred to the risk factors associated with
major hepatectomy, but few have examined the safe or permissible
limits of the hepatectomy in patients with obstructive jaundice [1].
In this study, we employed the hypothesis that the maximum to-
tal bilirubin level during the first 2 weeks after hepatectomy reflects
the grade of surgical stress. First, we attempted to determine the
safe and permissible limits of hepatectomy in obstructive jaundice
patients retrospectively based on the postoperative maximum total
bilirubin level. Then we examined the usefulness of preoperative
portal embolization (PE) for increasing the permissible hepatic re-
section volume.
Materials and Methods
Patients
Enrolled in this study were 416 consecutive patients who under-
went hepatectomy during the 9 years from May 1992 to April 2001
in our institution The diseases necessitating hepatectomy were he-
patocellular carcinoma in 133 cases, cholangiocellular carcinoma
in 19, metastatic liver tumor in 190, gallbladder carcinoma in 21,
hilar bile duct carcinoma in 29, benign liver tumor in 18, and other
diseases in 6 (Table 1). The hepatectomies included 67 subsegmen-
tectomies, 88 segmentectomies, 3 bisegmentectomies, 8 caudate lo-
bectomies, 98 lobectomies, 35 extended lobectomies, 12 trisegmen-
tectomies, and 105 nonanatomic resections (Table 2).
The patients were classified, on the basis of histologic and hema-
tologic findings in the resected liver parenchyma, into four groups:
(1) normal liver function (group N, n = 242); (2) chronic hepatitis
(CH, n = 71); (3) liver cirrhosis (LC, n = 64); and (4) relief from
obstructive jaundice (OJ, n = 39).
Preoperative Management
Percutaneous or endoscopic biliary drainage was carried out in all
patients with obstructive jaundice. Percutaneous biliary drainage
was performed with ultrasonic guidance. To select the site for a skin
puncture, the bile duct, portal vein, hepatic vein, and hepatic artery
were clearly visualized by abdominal ultrasonography. After local
anesthesia with 1% lidocaine (Xylocaine) had been induced in the
skin at the planned site of puncture, the bile duct puncture using a
22-gauge percutaneous transhepatic cholangiography (PTC)
needle was then performed. A guide wire was next inserted under
fluoroscopic guidance to ascertain whether there was any reflux of
bile, and the puncture tract was dilated. An 8F pigtail catheter was
finally inserted into the bile duct cavity [2]. Endoscopic biliary
drainage was performed following endoscopic retrograde cholan-
giography (ERC). A long guide wire was inserted deep into the
biliary systemunder fluoroscopic guidance, and the endoscope was
then withdrawn slowly. A biliary tube was placed in the dilated bile
duct transnasally [3]. Hepatectomy was carried out after the total
bilirubin concentration fell to 3 mg/dl or less.
Correspondence to: Tetsuya Takahashi, M.D., e-mail: ex7t-tkhs@asahi-
net.or.jp
WORLD
Journal of
SURGERY
2004 by the Societe
Internationale de Chirurgie
World J. Surg. 28, 475481, 2004
DOI: 10.1007/s00268-004-7128-7
Portal embolization was applied in 18 cases in which more than
the right lobe was later resected in obstructive jaundice groups. We
used a transileocolic PE technique. A 7F polyethylene catheter
with a balloon was inserted into the portal vein at laparotomy
through the ileocolic vein. After portography had defined the in-
trahepatic portal anatomy, the portal vein branch of the lobe to be
resected was embolized under fluoroscopic control. The emboliz-
ing material consisted of a mixture of gelatin (Spongel; Yamanou-
chi, Tokyo, Japan), monoethanolamine oleate (Oldamin; Grelan,
Tokyo, Japan), diatrizoate sodiummeglumine (60%Urografin; Ni-
hon Schering, Osaka, Japan), and gentamicin (40 mg) [4]. Hepa-
tectomy was usually performed 3 to 4 weeks after PE.
Surgical Technique
An upper midline skin incision with a right subcostal transverse
incision or a bilateral transverse epigastric subcostal incision was
employed for hepatectomy. Intraoperative ultrasonography was
performed routinely to determine the locations of lesions and to
assist in selecting the plane of resection. The lesion site and the
resection line were marked with a coagulator on the liver surface.
When anatomic resections were performed, the hepatic artery,
portal vein, and hepatic duct were dissected at the liver hilum after
cholecystectomy. If the caudate lobe needed to be preserved, the
portal vein and hepatic duct were cut distal to the bifurcation of the
caudate branches.
Parenchymal dissection was carried out routinely using ultra-
sonic dissection. The intermittent Pringle maneuver was per-
formed during parenchymal transection, with clamping of the por-
tal vein and the hepatic artery for 15 minutes and then declamping
for 5 minutes or hemihepatic clamping. If a tumor involved the hi-
lar portal vein or hepatic artery, the vessel was partially resected
and reconstructed.
After hepatectomy, intraoperative cholangiography and the in-
digo injection test were performed to reveal any bile leakage. If bile
leakage was present, closure was made with 5-0 sutures. The cut
surface was covered with fibrin glue.
Lymph node dissection from the around the head of pancreas,
including the celiac region, to the hepatoduodenal ligament was
carried out in patients with hilar bile duct carcinoma, gallbladder
carcinoma, large metastatic liver carcinoma, or cholangiocellular
carcinoma. The bile duct was anastomosed with the jejunum using
4-0 absorbable sutures with a transhepatic or transjejunal biliary
drainage tube.
Variables
A total of 20 perioperative variables were studied. Fourteen pre-
operative variables were determined just before hepatectomy: age,
gender, percent resection, plasma retention rate of indocyanine
green at 15 minutes (normal <10%), serumtotal bilirubin (0.41.8
mg/dl), alanine aminotransferase (1145 U/l), serumalkaline phos-
phatase (109312 U/l), serum albumin (4.35.4 g/dl), lymphocyte
count (12003000/l), hemoglobin (13.817.2 g/dl), platelet count
(1839 x 10
4
/l), prothrombin time ratio (0.871.15), serumcholin-
esterase (255501 U/l), and any evidence of diabetes mellitus.
Table 1. Diseases treated with hepatectomy.
Disease All cases (n = 416) N (n = 242) CH (n = 71) LC (n = 64) OJ (n = 39)
Primary liver tumor
HCC 133 15 60 58
Cholangiocellular Ca 19 9 3 1 6
Metastatic liver tumor 190 179 6 5
Gallbladder Ca 21 16 5
Hilar bile duct Ca 29 5 24
Benign liver tumors
Hemangioma 8 6 2
Intrahepatic lithiasis 2 2
Hepatic cyst 3 3
Liver abscess 2 2
Granuloma 2 2
Focal nodular hyperplasia 1 1
Primary sclerosing cholangitis 2 2
Benign hilar stricture 1 1
Others 3 2 1
HCC: hepatocellular carcinoma; Ca: carcinoma; N: normal liver; CH: chronic hepatitis; LC: liver cirrhosis; OJ: obstructive jaundice.
Table 2. Type of hepatectomy, by disease.
Type All cases (n = 416) N (n = 242) CH (n = 71) LC (n = 64) OJ (n = 39)
Nonanatomic resection 105 58 16 31
Anatomic resection
Subsegmentectomy 67 39 17 10 1
Segmentectomy 88 56 14 14 4
Bisegmentectomy 3 1 1 1
Caudate lobectomy 8 1 1 3 3
Lobectomy 98 58 16 6 18
Extended lobectomy 35 23 6 6
Trisegmentectomy 12 6 6
476 World J. Surg. Vol. 28, No. 5, May 2004
Liver volumes were calculated from preoperative computed to-
mography (CT) scans. Serial abdominal transverse CT scans per-
formed at 1 cm intervals, including the entire liver, were used to
calculate the liver volume using the method of Heymsfield et al.
(with minor modifications) [5]. The total liver and resected liver
were traced on each slice with a cursor and their areas calculated.
The volume was obtained from the sum of all slices. The percent
resection was defined as the resected liver volume expressed as a
percentage of the total liver volume, excluding the tumor.
There were four intraoperative variables: the need for additional
resection, duration of operation, volume of transfusion, and dura-
tion of ischemia in the remnant liver. Additional resections in-
cluded resection of primary tumors or organs other than the liver,
removal of a tumor thrombus in the portal vein or the inferior vena
cava, lymph node dissection in the hepatoduodenal ligament, re-
construction of the biliary tract, and reconstruction of the hepatic
artery or portal vein leading to ischemic-reperfusion injury.
The two postoperative variables examined were the maximum
total bilirubin level during the 2 weeks after hepatectomy and mor-
bidity, defined as any presence of postoperative hemorrhage, por-
tal thrombosis, or septic infection with abnormal clinical symptoms.
For the patients with PE, the data obtained after PE were used.
Analysis
All values were expressed as the mean SD. The coefficient of
correlation was calculated, and values of 0.4 or more were consid-
ered statistically significant. Statistical analyses were performed us-
ing the
2
test for independence (Fishers exact probability test)
and Students t-test for comparison of various pairs of factors. A p
value of less than 0.05 was considered statistically significant.
Results
Surgical Results of Overall Cases
The mean percent of the liver resected in all hepatectomized pa-
tients was 28.3%19.6%. The mean maximal total bilirubin for all
patients measured during the first 2 weeks after hepatectomy was
3.4 4.3 mg/dl; that for patients with a normally functioning liver
was 2.7 1.8 mg/dl, for chronic hepatitis patients 3.1 5.7 mg/dl,
for those with liver cirrhosis 4.3 6.9 mg/dl, and for those with
obstructive jaundice 7.3 5.4 mg/dl. The morbidity rate was 15.6%
(65/416) and the mortality rate 2.4% (10/416) (Table 3).
The maximum total bilirubin and mortality rate had a positive
relation with Childs classification. There was no relation with the
percent of the liver resected or the morbidity rate. We had no Child
C case in this study (Table 4).
Fatal Liver Failure Cases
There were 10 deaths due to liver failure. One of these patients had
chronic hepatitis C, three had liver cirrhosis, and six had obstructive
jaundice. Liver failure resulted from excessive volume resection in
two cases and from postoperative complications such as hemor-
rhage, portal thrombosis, or septic infection in eight.
Percentage Resection and Maximal Total Bilirubin
The percent of liver resected was significantly correlated with the
maximum total bilirubin levels in all groups, expressed logarithmi-
cally, without preoperative PE. The slopes of the correlation curves
became steeper in the following order: normal liver, chronic hepa-
titis, obstructive jaundice, and liver cirrhosis (Fig. 1).
Maximum Total Bilirubin and Outcome after Hepatectomy
There were no deaths due to liver failure in patients with a maxi-
mumtotal bilirubin of 8.5 mg/dl or less. The highest maximumtotal
bilirubin level in a surviving patient was 14.4 mg/dl.
Safe and Permissible Limits of Hepatectomy
When hepatectomy in patients with a maximum total bilirubin of
8.5 mg/dl or less was accepted as safe, the safe limit of hepatectomy
in obstructive jaundice patients was 48.7%. However, when hepa-
tectomy with a maximum total bilirubin of 14.4 mg/dl or less was
considered safe, the permissible limit of hepatectomy was 71.6%
(Fig. 2).
Risk Factors for Liver Failure in Obstructive Jaundice
Other than a high percentage resection, the only risk factor for
postoperative liver failure in obstructive jaundice patients was a
postoperative complication (Table 5).
Efficacy of Preoperative Portal Embolization
The percentage resection before PE was 66.9% 8.7%, and that
after PE was 59.6% 6.7%. The percentage resection had de-
creased in 16 of 18 cases after PE (Fig. 3).
Portal embolization decreased the maximumtotal bilirubin level
from 8.5 mg/dl to 3.9 mg/dl in 48.7% of hepatectomy patients with
obstructive jaundice. PE extended the safety range of hepatectomy
Table 3. Hepatectomy and its outcome, by disease.
Parameter All cases (n = 416) N (n = 242) CH (n = 71) LC (n = 64) OJ (n = 39)
Percentage resection (%) 28.3 19.6 29.8 20.0 23.2 12.6 17.1 14.9 43.1 20.1
Maximal total bilirubin (mg/dl) 3.4 4.3 2.7 1.8 3.1 5.7 4.3 6.9 7.3 5.4
Morbidity rate (%) 15.6 11.6 16.9 10.9 46.2
Mortality rate (%) 2.4 0.0 1.4 4.7 15.4
Table 4. Relation with Childs classification.
Childs classification
Parameter A (n = 311) B (n = 105) p
Percentage resection (%) 27.5 19.3 30.7 20.5 0.1449
Maximal total bilirubin (mg/dl) 3.0 3.6 4.7 6.0 < 0.001
Morbidity rate (%) 14.1 20.0 0.2032
Mortality rate (%) 0.96 6.67 < 0.01
477 Takahashi et al.: Hepatectomy for Obstructive Jaundice
from 48.7% to 67.4% when a maximum total bilirubin level after
hepatectomy of less than 8.5 mg/dl was accepted as indicating a safe
hepatectomy (Fig. 4).
Discussion
Hepatectomy has become a safer operative procedure owing to ad-
vances in surgical technique and to the possibility of precise assess-
ment of preoperative liver functional reserve. On the other hand,
this progress has expanded the range of preoperative indications
for hepatectomy and increased the resectable volume. Conse-
quently, posthepatectomy liver failure remains a controversial
problem, especially in patients with liver cirrhosis or obstructive
jaundice.
When discussing surgical stress after hepatectomy, the loss of
hepatic volume and unpredictable intraoperative and postopera-
tive stress due, for example, to prolonged operations, massive
blood loss, ischemia during vessel reconstruction, postoperative
hemorrhage, and infectious complications should be considered. In
other words, posthepatectomy liver failure is caused by insufficient
liver volume following excessive resection, whatever the degree to
which the remnant liver has been subjected to surgical stress [6].
Table 5. Risk factors for mortality associated with obstructive jaundice.
Risk factor Surviving patients Fatal cases p
Percentage resection (%) 39.9 20.1 60.6 7.8 < 0.05
Morbidity rate (%) 39.4 83.3 < 0.05
Fig. 1. Relation between percentage
resection (percent of the liver resected)
and maximum total bilirubin. The
percentage resection was significantly
correlated with the maximum total
bilirubin levels in all groups, expressed
logarithmically, without preoperative
portal embolization. N (normal liver):
y = 0.006x + 0.214 (r = 0.472); CH
(chronic hepatitis): y = 0.009x + 0.131
(r = 0.420); OJ (obstructive jaundice):
y = 0.010x + 0.442 (r = 0.513); LC (liver
cirrhosis): y = 0.013x + 0.250 (r = 0.637).
Fig. 2. Safe and permissible limits of
hepatectomy in patients with obstructive
jaundice. The safe limit of hepatectomy
was 48.7% and the permissible limit of
hepatectomy 71.6% in obstructive
jaundice patients. Circles: surviving
patient; triangle: deceased patient; SL:
safe limit; PL: permissible limit. OJ
(obstructive jaundice): y = 0.010x +
0.442 (r = 0.513).
478 World J. Surg. Vol. 28, No. 5, May 2004
Few cases of liver failure are due to the physical shortage of liver
volume alone, as liver functional reserve of the normal or cirrhotic
liver can nowbe assessed. However, extended resection of a liver in
a patient with obstructive jaundice is sometimes accompanied by
fatal liver failure.
Liver failure can be explained by a short supply of adenosine tri-
phosphate (ATP) resulting from inadequate mitochondrial oxida-
tive phosphorylation [7]. A shortage of ATP due to increased ATP
consumption to maintain other basic metabolic processes leads to
insufficient liver regeneration and thus to insufficient relief of the
various surgery-related stresses after hepatectomy. If the consump-
tion of ATP exceeds its production, bilirubin transport is also dis-
turbed; and hyperbilirubinemia, which is sometimes followed by
fatal liver failure, results. In short, hyperbilirubinemia after hepa-
tectomy indicates an energy crisis and a need for liver regeneration.
LaMont and Isselbacher [8] reported three basic pathophysi-
ologic mechanisms of postoperative jaundice: an increased pig-
ment load from overproduction of bilirubin in cases of transfusion
or hematoma, impaired excretion of bilirubin due to hepatocellular
damage, and obstruction of the extrahepatic bile duct. In hepatec-
tomy cases, in addition to these factors, insufficient remnant liver,
massive transfusion, or heavy postoperative hemorrhage may help
explain the overproduction of bilirubin. Septic infections increase
the production of inflammatory cytokines such as tumor necrosis
Fig. 3. Changes in percentage resection
before and after portal embolization.
Percentage resection was decreased by
portal embolization in 16 of 18 cases.
Fig. 4. Efficacy of preoperative portal
embolization in patients with obstructive
jaundice. Preoperative portal
embolization (PE) decreased the
maximal total bilirubin from 8.5 mg/dl to
3.9 mg/dl in 48.7% of patients with
hepatectomy for obstructive jaundice. PE
extended the safety range of hepatectomy
from 48.7% to 67.4% when a maximum
total bilirubin after hepatectomy of less
than 8.5 mg/dl was taken as indicating a
safe hepatectomy. Circles: PE case; SL*:
safe limit without PE; SL**: safe limit
with PE. A (without PE, n = 21): y =
0.010x + 0.442 (r = 0.513). B (with PE,
n = 19): y = 0.018x - 0.285 (r = 0.429).
479 Takahashi et al.: Hepatectomy for Obstructive Jaundice
factor- (TNF- ) or interleukin-6 from Kupffer cells via stimula-
tion by lipopolysaccharides, a process that causes cholestasis [9].
Hyperbilirubinemia due to bile stasis after infection and surgery is
caused by disturbed transport of conjugated bilirubin to the bile
canaliculi, a process for which ATP energy is required [7]. Thus
hyperbilirubinemia is closely related to surgical stress. Moreover,
the total bilirubin level has become a predictive index for liver fail-
ure after hepatectomy, and we therefore suggest that maximum to-
tal bilirubin reflects surgical stress.
Let us consider at what total bilirubin level hyperbilirubinemia
becomes irreversible. No reports in the literature discuss the rela-
tion between maximumtotal bilirubin and the outcome of hepatec-
tomy. We believe that hepatectomy with a maximum total bilirubin
of 8.5 mg/dl or less (minimum level in fatal liver failure cases) dur-
ing the first 2 weeks after hepatectomy is safe, and that hepatec-
tomy with a maximumtotal bilirubin level of up to 14.4 mg/dl (maxi-
mum seen in surviving patients) is the permissible limit of the
hepatectomy. Hyperbilirubinemia becomes irreversible if the bili-
rubin level exceeds 14.4 mg/dl within 2 weeks after hepatectomy.
The accuracy of CT volumetry has already been demonstrated.
The actual resected liver volume was significantly correlated with
the estimated liver volume calculated by CT volumetry [10]. We
demonstrated that the percent of the liver resected (percentage re-
section) was significantly correlated with the maximum posthepa-
tectomy total bilirubin expressed logarithmically. For liver trans-
plantation from a living donor, a negative correlation between the
graft/standard volume ratio and the postoperative maximum total
bilirubin has already been demonstrated [11]. The rate of increase
in the bilirubin concentration accompanying an increase in the per-
centage resection rose in the following order: normal liver, chronic
hepatitis, obstructive jaundice, liver cirrhosis. This does not always
reflect liver functional reserve alone, as the surgical procedures
must differ in accordance with the underlying liver disease. There-
fore we examined retrospectively the relation between the maxi-
mum total bilirubin and the clinical outcome for each underlying
liver disease. Generally, the functional reserve of a liver in a patient
with obstructive jaundice who was relieved of the jaundice by bili-
ary drainage appears to be the same as that of a normal liver. From
a clinical viewpoint, however, we should recognize that insuffi-
ciency of liver functional reserve in patients with obstructive jaun-
dice must be second only to that in those with liver cirrhosis.
It is necessary to devise a way to decrease surgical stress nowthat
liver functional reserve can be assessed precisely [12]. On the other
hand, discussion is required on the question of the permissible limit
of hepatectomy when the degree of surgical stress is to be mini-
mized.
Yamanaka et al. proposed a prediction score for posthepatec-
tomy liver failure in patients with liver cirrhosis and reported that a
prediction score of 50 or less may be considered a safe zone [13].
We have used this prognostic score as an indication for hepatec-
tomy. In patients with normal liver function or in selected cirrhosis
patients, major hepatectomy can be performed safely with mini-
mum morbidity and mortality [14], and it is generally accepted that
trisegmentectomy is safe in a normal liver [15]. Hepatectomy is now
as safe a procedure in liver cirrhosis patients as it is in normal liver
patients, except in cases of intraabdominal hemorrhage [16]; and
resection of up to 40%of the liver is safe in a patient with a cirrhotic
liver [17, 18].
On the other hand, in obstructive jaundice patients, the maxi-
mum level of indocyanine green excreted in the bile [19] and the
daily output of bilirubin in drained bile [20] are thought to be useful
for evaluating liver functional reserve. Mitochondrial oxidative
phosphorylation in the liver is impaired by cholangitis [21], and sen-
sitivity to endotoxins increases following prolonged jaundice and
encephalopathy [22]. Endotoxemia stimulates the Kupffer cells to
release mediators that harm hepatocytes; this situation can be
caused by surgical manipulation (external factor) or by immune
systemdysfunction and bacterial translocation (internal factors) af-
ter hepatectomy [23].
Biliary reconstruction and lymph node dissection are recom-
mended with hepatectomy for patients with obstructive jaundice,
unlike for those with normal or cirrhotic livers. We have previously
reported a predictive formula for safe hepatectomy performed af-
ter obstructive jaundice has been relieved [1]. Except for this ar-
ticle, there have been few regarding the safe and permissible limits
of hepatectomy in patients with obstructive jaundice. For safe
hepatectomy, serum albumin levels of 3 g/dl or more are needed to
reduce postoperative mortality in obstructive jaundice patients
[24].
During this investigation we concluded that, in obstructive jaun-
dice patients, resection of up to 48.7%of the liver was safe and that
hepatectomy of up to 71.6% was the maximum permissible resec-
tion. Only postoperative complications were risk factors for liver
failure, even though there were only six deaths. Thus efforts must
be made to minimize postoperative complications in patients with
48.7% to 71.6% liver resections. Careful consideration is required
before a surgical treatment is selected when a resection of more
than 71.6% of the liver is indicated because of the high risk and
despite the lack of postoperative complications.
Portal embolization contributed to the decreased percentage re-
section and decreased the maximum total bilirubin level from 8.5
mg/dl to 3.9 mg/dl for safe resection of 48.7%of the liver in obstruc-
tive jaundice patients. When a maximum total bilirubin level of 8.5
mg/dl was regarded as safe for hepatectomy, PEincreased the limit
for safe hepatectomy from 48.7% to 67.4%. This increase is per-
haps attributable to a reduction in the degree of surgical stress on
the remnant liver as a result of a functional shift caused by PE,
which is regarded as atraumatic hepatectomy [4, 25].
Because the outcome of hepatectomy depends on the balance
between the liver functional reserve and the degree of surgical
stress, it is difficult to assess the degree of safety or risk of such a
procedure in advance. However, our findings from this retrospec-
tive study of 416 accumulated cases may be worthy of being used as
a standard index of the relative safety of hepatectomy or of PEwith
hepatectomy for obstructive jaundice. These new criteria may well
be of use in the treatment of individual cases, especially when ap-
plied in a flexible manner.
Re sume . On a examine les limites de se curite et dindication raisonnable
de lhe patectomie chez le patient cholestatique ainsi que lutilite de
lembolisation porte pre ope ratoire (EP) afin daugmenter la se curite de
lhe patectomie. Nous avons classe 416 cas dhe patectomie re alise es pendant
une pe riode de neuf ans sous les conditions suivantes: fonction he patique
normale (n = 242); he patite chronique (n = 71); cirrhose du foie (n = 64);
et re section du foie apre `s leve e dobstacle dans le cadre dun icte `re par
obstruction (n = 39). Lhe patectomie a e te re alise e seulement apre `s avoir
constate que la bilirubine totale e tait infe rieure a` 3 mg/dl par un drainage
biliaire pre ope ratoire. On a analyse les facteurs qui influenc aient le taux de
bilirubine totale maximale mesure e dans les deux semaines apre `s he patectomie
pour e valuer le degre de stress chirurgical. Une EP a e te re alise e dans 18
cas. Le taux maximal de bilirubine totale, exprime envaleur logarithmique,
corre lait de fac on significative avec le pourcentage de re section dans tous
les groupes. On a conside re quun taux de bilirubine de moins de 8.5 mg/dl
480 World J. Surg. Vol. 28, No. 5, May 2004
apre `s he patectomie constituait une intervention sure et quun taux de
bilirubine totale postope ratoire de 14.4 mg/dl e tait la limite dindication
raisonnable . Base sur ces re sultats, on a de termine les pourcentages de
re section sure et dindication raisonnable a`, respectivement, 48.7% et
71.6%. LEP a diminue la bilirubine totale de 8.5 a` 3.9 mg/dl lorsque 48.7%
du foie, une proportion conside re e sure dans tous les cas, a e te re se que e.
LEP a augmente la limite dhe patectomie sure de 48.7% a` 67.4%
lorsquun taux de la bilirubine totale maximale apre `s he patectomie de 8.5
mg/dl e tait pris comme limite sure.
Resumen. Se investigaron los l mites seguros y permisibles de la hepatectom a
en pacientes con ictericia obstructiva y tambie n la utilidad de la embolizacio n
portal preoperatoria (EP) con el fin de ampliar los l mites de la reseccion
segura. Clasificamos 416 casos de hepatectom a realizados a lo largo de 9
anos bajo las siguientes categor as: funcion hepatica normal (n = 242);
hepatitis cronica (n = 71); cirrosis hepatica (n = 64); e h gado luego
de correccion de ictericia obstructiva (n = 39). La hepatectom a fue
practicada una vez la bilirrubina total pudo ser reducida a niveles
inferiores a 3 mg/dl mediante drenaje biliar preoperatorio. Los factores
relacionados con el nivel maximo de bilirrubina total determiado a las 2
semanas luego de la hepatectom a fueron investigados, y este nivel fue
tomado como un indicador del grado de estre s quirurgico. Se realizo EP
en 18 casos de ictericia obstructiva. El maximo de la bilurribina total
expresado como logaritmo aparecio significativamente correlacionado con
el porcentaje de reseccion, en todos los grupos. La hepatectom a seguida de
un nivel maximo de bilirrubina total inferior a 8.5 mg/dl fue considerado
como seguro, y aquella seguida de unnivel de bilirrubina total de 14.4 mg/dl
fue considerado como correspondiente a la hepatectom a maxima permisibe.
Con base en tales resultados, se considera que los niveles seguros y
permisibles de una hepatectom a en ictericia obstructiva son 48.7%y 71.6%
respectivamente. La EP disminuyo el nivel maximo de bilirrubina total de
8.5 a 3.9 mg/dl cuando se reseco un 48.7%del h gado, que es una proporcion
segura en todos los casos; y aumento el l mite de la hepatectom a de 48.7 a
67.4% cuando se considero se- guro un nivel de bilirrubina posthepatectom a
de 8.5 mg/dl.
Acknowledgments
The authors thanks Shinji Togo, M.D., Ph.D., Associate Professor
of the Second Department of Surgery and Hiroshi Shimada, M.D.,
Ph.D., Professor and Chairman of the Second Department of Sur-
gery for their advice and guidance. Our thanks are also due to Mr.
C.W.P. Reynolds for his linguistic assistance with this manuscript.
References
1. Takahashi T, Endo I, Shimada H, et al. Assessment of liver function by
prognostic score after relief of obstructive jaundice. J. Hepatobiliary.
Pancreat. Surg. 1997;4:269275
2. Takada T, Yasuda H, Hanyu F. Technique and management of percu-
taneous transhepatic cholangial drainage for treating an obstructive
jaundice. Hepatogastroenterology 1995;42:317322
3. Cotton PB, Burney PGJ, Mason RR. Transnasal bile duct catheterisa-
tion after endoscopic sphincterectomy. Gut 1979;20:285287
4. Makuuchi M, Thai BL, Takayasu K, et al. Preoperative portal emboli-
zation to increase safety of major hepatectomy for hilar bile duct carci-
noma: a preliminary report. Surgery 1990;107:521527
5. Heymsfield SB, Fulenwider T, Nordlinger B, et al. Accurate measure-
ment of liver, kidney and spleen volume and mass by computerized axial
tomography. Ann. Intern. Med. 1979;90:185187
6. Shimada H, Matsuba A, Nimoto S, et al. Liver failure after hepatic re-
section. Nippon Geka Gakkai Zasshi. 1989;4:566572
7. Shimada H, Endo I, Takahashi T, et al. Pathophysiology of major he-
patic resection and benefits of preoperative portal embolization in pre-
venting hepatic failure. J. Hepatobiliary. Pancreat. Surg. 1997;4:384
390
8. LaMont JT, Isselbacher KJ. Postoperative jaundice. N. Engl. J. Med.
1973;288:305307
9. Jansen PLM, Muller M. Early events in sepsis-associated cholestasis.
Gastroenterology 1999;116:486488
10. Kubota K, Makuuchi M, Kusaka K, et al. Measurement of liver volume
and hepatic functional reserve as a guide to decision-making in resec-
tional surgery for hepatic tumors. Hepatology 1997;26:11761181
11. Matsushita M, Shimamura T, Kamiyama T, et al. Living donor liver
transplantation and postoperative cholestasis from the aspect of rela-
tionship between degree of postoperative hyperbilirubinemia and graft
size. Tan Sui 2000;21:661666
12. Shimada M, Takenaka Y, Fujiwara Y, et al. Risk factors linked to post-
operative morbidity in patients with hepatocellular carcinoma. Br. J.
Surg. 1998;85:195198
13. Yamanaka N, Okamoto E, Kuwata K, et al. Amultiple regression equa-
tion for prediction of posthepatectomy liver failure. Ann. Surg. 1984;
200:658663
14. Capussotti L, Polastri R. Operative risks of major hepatic resections.
Hepatogastroenterology 1998;45:187190
15. Povoski SP, Fong Y, Blumgart LH. Extended left hepatectomy. World
J. Surg. 1999;23:12891293
16. Midorikawa Y, Kubota K, Takayama T, et al. A comparative study of
postoperative complications after hepatectomy in patients with and
without chronic liver disease. Surgery 1999;126:484491
17. Bozzetti F, Gennari L, Regalia E, et al. Morbidity and mortality after
surgical resection of liver tumors: analysis of 229 cases. Hepatogastro-
enterology 1992;39:237241
18. Okamoto E, Kyo A, Yamanaka N, et al. Prediction of the safe limits of
hepatectomy by combined volumetric and functional measurements in
patients with impaired hepatic function. Surgery 1984;95:586592
19. Izawa K, Sasaki M, Tomioka T, et al. Evaluation of ICG Bmax in hepa-
tectomized patients undergoing biliary decompression for obstructive
jaundice. Tando 1994;8:197203
20. Kumazawa K, Otani Y, Kubota K, et al. Study of bilirubin measurement
in bile drained from patients with obstructive jaundice. Nippon
Shoukaki Geka Gakkai Zasshi 1994;27:17711777
21. Nagano I, Kato S, Nimura Y, et al. Hepatic mitochondrial changes in
experimental obstructive jaundice complicated by biliary infection.
Hepatogastroenterology 1994;41:432437
22. Hatsuse K, Aoki H, Murayama M, et al. A clinical study of hyperbiliru-
binemia and pathophysiology after major hepatectomy: comparison
with liver with obstructive jaundice, liver with cirrhosis and liver with
normal function. Nippon Shoukaki Geka Gakkai Zasshi 1994;27:3744
23. Rimola A, Soto R, Bory F, et al. Reticuloendothelial systemphagocytic
activity in cirrhosis and its relation to bacterial infection and prognosis.
Hepatology 1984;4:5358
24. Cheng-Hsi S, Shyh-Haw T, Cheng-Chung W, et al. Factors influencing
postoperative morbidity, mortality, and survival after resection for hilar
cholangiocarcinoma. Ann. Surg. 1996;223:384394
25. Nagano Y, Nagahori K, Shimada H, et al. Improved functional reserve
of hypertrophied contralateral liver after portal vein ligation in rats. J.
Hepatol. 2002;37:7277
481 Takahashi et al.: Hepatectomy for Obstructive Jaundice