Standardized measurement of the future

liver remnant prior to extended liver

resection: Methodology and clinical
associations
Jean-Nicolas Vauthey, MD, Amin Chaoui, MD, Kim-Anh Do, PhD, Malcolm M. Bilimoria, MD, Marc J.
Fenstermacher, MD, Chuslip Charnsangavej, MD, Marshall Hicks, MD, Guido Alsfasser, MD, Gregory
Lauwers, MD, Irvin F. Hawkins, MD, and James Caridi, MD, Houston, Texas, and Gainesville, Fla

Background. There is no agreement regarding the preoperative measurement of liver volumes and the
minimal safe size of the liver remnant after extended hepatectomy.
Methods. In 20 patients with hepatobiliary malignancy and no underlying chronic liver disease, volu-
metric measurements of the liver remnant (segments 2 and 3 ± 1) were obtained before extended right
lobectomy (right trisegmentectomy). The ratios of future liver remnant to total liver volume were calculat-
ed by using a formula based on body surface area. In 12 patients, response to preoperative right trisec-
toral portal vein embolization was evaluated. In 15 patients who underwent the planned resection, pre-
operative volumes were correlated with biochemical and clinical outcome parameters.
Results. The future liver remnants increased after portal vein embolization (26% versus 36%, P < .01).
Smaller size liver remnants were associated with an increase in postoperative liver function tests (P < .05)
and longer lengths of hospital stay (P < .02). Preliminary data indicates an increase in major complica-
tions for liver volumes ≤25% (P = .02).
Conclusions. A simple method of measurement provides an assessment of the liver remnant before
resection. It is useful in evaluating response to portal vein embolization and in predicating the outcome
before extended liver resections. (Surgery 2000;127:512-9.)

From the Departments of Surgical Oncology, Biostatistics and Biomathematics, and Radiology, University of
Texas M. D. Anderson Cancer Center, Houston, Texas, and the Departments of Radiology and Pathology,
University of Florida College of Medicine, Gainesville, Fla

WITH IMPROVEMENTS IN PERIOPERATIVE CARE, major
liver resections are increasingly performed for pri-
mary or metastatic liver cancer. These procedures
are associated with low mortality rates in patients
without cirrhosis although significant morbidity is
still reported.1-3 Currently, there is no consensus as
to the extent of liver that can be safely resected
with resection ranging from 60% to 65%,4,5 and up
to 80% to 90%.6-8 This disagreement regarding the
extent of resection results from a lack of standard-
ization in the volumetric assessment of the liver
before resection and from reports focusing on the
Accepted for publication December 28, 1999.
Reprint requests: Jean-Nicolas Vauthey, MD, Department of
Surgical Oncology, The University of Texas M. D. Anderson
Cancer Center, 1515 Holcombe Blvd, Box 106, Houston, TX
77030.
Copyright © 2000 by Mosby, Inc.
0039-6060/2000/$12.00 + 0 11/56/105294
doi:10.1067/msy.2000.105294
amount of liver resected rather than on the
amount of remaining liver.4,5
In patients who are candidates for resection of a
lobe or more, there is variability in the size of the
future liver remnant (FLR),4,5 and the safe mini-
mal size for FLR is not well defined. With the
advent of computed tomography (CT), measure-
ments of liver volumes can now be accurately esti-
mated5,9 and the volume of the FLR determined
before resection.10 Recently, a formula based on a
close association between liver volume and body
surface area was reported and its accuracy verified,
allowing for a precise determination of the total
liver volume in healthy adults.11
Twenty consecutive patients who were candidates
for an extended right lobectomy (right trisegmen-
tectomy) and who were expected to have a small
remaining liver volume (≤40%) underwent mea-
surement of the FLR with the use of CT. The ratio of
FLR to total liver volume, calculated on the basis of
Urata’s formula allowed for a comparison between
patients.11 With these standardized liver volumes,

512 SURGERY
Surgery Vauthey et al 513
Volume 127, Number 5

Fig 1. CT outline of the segments included in the measurement of FLR (patient 12).

Fig 2. Three-dimensional reconstruction of FLR (segments 1, 2, and 3) before (left) and after (right) portal
vein embolization (patient 12).

the relative increase in liver size in 12 patients who
underwent portal vein embolization before liver
resection was also determined. The biochemical and
clinical outcome of patients who underwent the
anticipated resections were correlated with the stan-
dardized FLR volumes estimated preoperatively.
PATIENTS AND METHODS
Twenty consecutive patients with hepatobiliary
malignancies in whom an extended right lobecto-
my (right trisegmentectomy) was anticipated
underwent volume measurement of segments 1, 2,
and 3 on the basis of CT (Fig 1). All CT examina-
tions were performed with a helical scanner
(HiSpeed Advantage; General Electric Medical
Systems, Milwaukee, Wis). CT of the abdomen was
performed to include the whole liver in one
breath-hold, using a 10-mm collimation. All helical
CT liver volumes were obtained with an indepen-
dent workstation (Allegro; ISG Technologies,
Toronto, Canada). The helical CT data were trans-
ferred to the workstation for assessment. Liver vol-
umes were calculated by integrated software tech-
niques that use density threshold seeding. With this
technique, the levels of density desired for inclu-
sion in the data set are selected. Once the levels are
selected, a volumetric reconstruction is obtained
for the entire data set on the basis of the desired
514 Vauthey et al
Fig 3. Standardized FLR before and after preoperative
portal vein embolization (12 patients). PVE, Portal vein
embolization.
density. This 3-dimensional reconstruction will
include any voxels that are of the selected density.
Regions that are of the selected density but should
not be included in the data set—such as inferior
vena cava, gallbladder, abdominal and chest wall
muscles—are excluded. The total liver volume was
estimated on the basis of the body surface area of
the patients with the formula: liver volume (cm3) =
706 x body surface area (m2) + 2.4.11 The ratio of
FLR volume to absolute liver volume as estimated
by this formula was obtained for each patient. It
was defined as the standardized FLR and allowed
for a comparison between patients. The body sur-
face area was calculated as previously described.12
Portal vein embolization was performed in 12
patients as part of a protocol (University of Florida
IRB #036-94) evaluating the response and side
effects of the procedure before extended hepatec-
tomy in patients with small FLRs (≤40% of the total
estimated liver). The right portal vein and segment
4 branches were embolized by means of percuta-
neous transhepatic access of a peripheral branch of
the right or left portal vein. Fluoroscopic imaging
guidance using injection of iodinated contrast was
sufficient in the majority of patients for achieving
portal vein access. Carbon dioxide portography
provided a target for subsequent portal vein
branch catheterization in patients in whom initial
access to a portal vein branch was difficult. This was
performed by direct injection of 20 mL of carbon
dioxide into the hepatic parenchyma. Ultrasound
guidance was also used later in the experience and
provided the advantage of real time imaging for
access.5 After initial access into a portal vein branch
Surgery
May 2000
with a 21-gauge needle, the Seldinger technique
was used to place a 5F or 6F vascular sheath into the
main right or left portal vein. Initial portography was
performed with a 4F or 5F multisidehole angiograph-
ic catheter, and the embolization procedure was sub-
sequently performed with a selective 4F or 5F angio-
graphic catheter. Gelatin pledgets (1 to 2 mm2) or
powder (Gelfoam; Upjohn Co, Kalamazoo, Mich) was
used to embolize branch vessels, followed by place-
ment of Gianturco coils (Cook, Inc, Bloomington,
Ind) into larger branch vessels and into the main
right portal vein. Embolization was performed until
there was an absence of antegrade flow into the target
segments.
Resection was performed 4 to 6 weeks after portal
vein embolization. In all portal vein embolization
patients, volumetric measurement was repeated within
one week before surgical exploration for resection. An
example of 3-dimensional volume reconstruction of
FLR before and after portal vein embolization is shown
in Fig 2. One surgeon (J-N. V.) performed all the hepat-
ic resections, and the details of the surgical technique
were previously described.2,3 A biochemical assessment
of liver function was evaluated by serial measurements
of prothrombin time, alkaline phosphatase, and
bilirubin. Postoperative length of stay and complica-
tions were correlated with the preoperative standard-
ized volume measurements. In all patients, the
anatomical extent of the anticipated resection was
confirmed by postoperative CT performed one week
after resection to confirm segmental lines of resec-
tion. One pathologist (G. L.) performed the histolog-
ic review of the resected specimens.
Data management and analyses were carried out
with the SAS13 and S-PLUS 514 statistical software.
Statistical methods include paired and two-sample t-
tests, Fisher’s exact test for association and linear
regression modeling. Linear association is repre-
sented by regression coefficient Beta given with and
accompanying standard error and nominal P value.
RESULTS
Patients’ diagnoses, demographics, and absolute
and standardized volumes for the FLRs at presen-
tation and after embolization are presented in
Table I. In all patients, the extent of the anticipat-
ed resection was ≥60% of the total liver. FLR
accounted for ≤25% of the total liver in 8 of 20
patients (40%) at presentation. Patients’ absolute
and relative volumes were similar for patients who
did and did not undergo portal vein embolization:
absolute FLR (308 cm3 ± 94 versus 332 cm3 ± 67
respectively) and standardized FLR (25% ± 8% ver-
sus 27% ± 7% respectively).
There was a significant increase in absolute and
Surgery Vauthey et al 515
Volume 127, Number 5

standardized FLR in the 12 patients who underwent

preoperative portal vein embolization (308 cm3 ± 95
versus 432 cm3 ± 121 [P = .0001] and 26% ± 8% ver-
sus 36% ± 10%, [P = .0007], respectively) (Fig 3).
Likewise, the standardized FLR was higher for
patients with portal vein embolization (n = 12) when
compared with patients without portal vein emboliza-
tion (n = 8), though not statistically significant (36%
± 11% versus 27% ± 7%, respectively, P = .25).
Among 20 patients selected for resection, 15
underwent the planned extended right lobectomy
(right trisegmentectomy) with or without caudate
lobe resection (n = 2 and n = 13, respectively).
Additional resections included the vena cava (n = 1)
and the common bile duct with hepatoduodenal
lymphadenectomy (n = 1). In 5 patients who had
undergone portal vein embolization, no resection
was performed because of a lack of regeneration (n
= 1) or unanticipated extent of disease at the time of
exploration (n = 4). These included peritoneal car-
cinomatosis in 2 patients, portal lymph node metas-
tasis in 1 patient, and disease in segment 3 in the
other.
Histopathologic review confirmed the absence of
cirrhosis and chronic hepatitis in the 15 patients
resected. Only minimal changes secondary to the por-
tal vein embolization and tumor mass effect were
identified. Ten patients displayed mild pseudohepati-
tis with no evidence of piecemeal necrosis in the por-
tal triads immediately surrounding the tumors. Nine
of the 15 patients displayed mild fibrosis in the triads
immediately adjacent to the tumors with some bridg-
ing identified in 2. Changes related to reduced blood
flow were also noted in the 7 portal vein embolization
cases. It consisted of patchy centrolobular hepatocyt-
ic dropout in 6 patients (with zone 1 ischemic necro-
sis in 1), and of sinusoidal distention in 7. Minor his-
tologic alterations such as scattered macrovesicular
steatosis (5 patients) and lipogranulomas (2 patients)
were noted.
Perioperative characteristics did not differ
between the patients who did and did not have por-
tal vein embolization: duration of the operation
(365 minutes ± 131 versus 326 minutes ± 86, respec-
tively), estimated blood loss (930 mL ± 788 versus
981 mL ± 847, respectively), transfusions (1.5 units ±
2.7 versus 0.9 units ± 1.1, respectively), length of stay
(9.7 days ± 2.7 versus 12 days ± 5.8, respectively).
Overall, the length of stay was 6 to 24 days (mean,
10.8 days ± 4.5, median 9). There were no biliary or
septic complications, no perioperative mortality, or Fig 4. Linear regression of clinical outcome parameters
mortality within 3 months of the resections. Three on standardized FLR in 15 patients who underwent resec-
patients incurred major complications that resulted tion. Patients with (ο) and without (∆) portal vein
in extended lengths of stay (prolonged ileus and embolization are presented. Statistical data are detailed in
jaundice in 1 patient, ascites and wound separation Table II.
516 Vauthey et al Surgery
May 2000

Table I. Patient diagnoses, demographics, absolute and standardized FLR

Total estimated
ID Diagnosis Age/sex BSA liver (cm3)
1 Met colorectal 71/F 1.7 1252
2 Met colorectal 52/F 1.6 1146
3 Hilar ca 75/F 1.6 1189
4 Gallbladder ca 68/F 1.6 1153
5 HCC 55/M 2.05 1450
6 HCC 69/M 1.7 1238
7 Met colorectal 76/M 1.6 1160
8 Met colorectal 46/F 1.6 1168
9 Met colorectal 37/M 1.8 1330
10 Met colorectal 68/F 1.4 1061
11 HCC 63/M 1.8 1295
12 Hilar ca 62/F 1.7 1217
13 Met colorectal 72/M 2.1 1499
14 Met colorectal 33/F 1.8 1273
15 Met colorectal 79/F 1.5 1062
16 Met colorectal 71/F 1.6 1189
17 Met colorectal 70/F 1.7 1203
18 Met colorectal 55/M 1.9 1379
19 Met colorectal 60/F 1.6 1146
20 Met colorectal 50/F 1.52 1076
Mean — 61.6 1.69 1224.3
Median — 65.5 1.7 1196
SD — 12.9 0.17 119.7
Met colorectal, Metastatic colorectal carcinoma; Hilar ca, hilar carcinoma; Gallbladder ca, gallbladder carcinoma; HCC, hepatocellular carcinoma; BSA,
body surface area; FLR, future liver remnant; PVE, portal vein embolization; postembo, postembolization; Y/N, Yes/No; SD, standard deviation.

in 1 patient, and partial superior mesenteric vein
thrombosis and pancreatitis in 1 patient). These
complications occurred in 3 of 5 patients with stan-
dardized FLR volumes of ≤25% compared with no
major complications in the remainder (n = 10) of
the resected group with an FLR of >25% (P = .02).
There was a significant association between the
standardized FLR size and the postoperative peaks
for alkaline phosphatase (P = .002), prothrombin
time (P = .029), bilirubin (P = .048), and lengths of
stay (P = .015) (Fig 4). The results of the regression
analyses associating FLR with the outcome variables
for all resected patients and the subsets with or with-
out preoperative portal vein embolization are pre-
sented in Table II.
DISCUSSION
A prospective series of 20 patients without under-
lying chronic liver disease underwent volumetric
measurements of the portion of the liver that was
anticipated to remain after resection. These mea-
surements provide an opportunity to correlate post-
operative outcome with FLR size. The method also
allows for a comparison between patients before
and after portal vein embolization. Extended right
lobectomy (right trisegmentectomy) only was select-
ed for this study to allow for a close anatomical cor-
relation between the surgical resection lines (umbil-
ical fissure and falciform ligament) and the left lat-
eral bisegmentectomy or trisegmentectomy
volumes (segments 2 and 3 ± 1) as defined preop-
eratively by CT. The patient outcome assessment
was made on the basis of the measurement of the
FLR, which is free of disease, as opposed to the
resected liver, which may be compromised by the
presence of a tumor, cholestasis, or the effects of the
portal vein embolization. The total liver volume is
estimated with a formula that closely correlates the
volume measured by CT with the body surface area
in adult patients. This formula, which accounts for
the total liver volume that is functionally available,
has been validated in healthy adults regardless of
gender.11
Previous methods of measurement of the extent of
resection have been made on the basis of ratios that
include the measurements of the total liver, the
resected volume, and the total tumor volume (resect-
ed volume–tumor volume/total liver volume–tumor
volume).4,15,16 In patients with multiple tumors, mea-
surement of individual volumes cumulates the error
variability associated with each measurement, result-
ing in an inaccurate estimate of the total tumor vol-
Surgery Vauthey et al 517
Volume 127, Number 5

FLR Estimated total Postembo Estimated total liver

(cm3) liver FLR (%) PVE FLR (cm3) postembo FLR (%) Resection
252 0.2 + 452 0.36 N
374 0.32 + 536 0.47 N
346 0.3 + 436 0.36 N
144 0.12 + 193 0.16 N
329 0.22 + 453 0.36 N
401 0.32 + 569 0.46 Y
162 0.14 + 269 0.23 Y
236 0.2 + 296 0.25 Y
357 0.27 + 539 0.4 Y
428 0.4 + 537 0.53 Y
401 0.31 + 506 0.4 Y
266 0.27 + 405 0.33 Y
389 0.26 - - - Y
311 0.24 - - - Y
312 0.29 - - - Y
419 0.35 - - - Y
307 0.25 - - - Y
199 0.14 - - - Y
347 0.3 - - - Y
378 0.35 - - - Y
317.9 0.26 - 432.5 0.36
337.5 0.27 - 452.5 0.36
83.9 0.07 - 121.0 0.10

ume. In addition, lesions beyond the resolution of
imaging or areas of nonfunctional liver resulting from
tumor growth are underestimated. Specifically, in
patients with bile duct tumors, the measurement of
liver volume that is compromised by cholestasis,
cholangitis, or vascular obstruction may not be rele-
vant with the formula noted above.4,15,16
Furthermore, in patients undergoing portal vein
embolization, volume measurement of the embolized
liver is probably not appropriate given the impair-
ment in portal vein flow and histologic changes
described in association with the procedure.17
We did not compare our method with the actual
volumetric measurement of the total liver volume.
This measurement would have compared het-
erogenous groups of patients, some with normal
liver parenchyma and others with compromised
biliary flow from bile duct obstruction or impaired
portal vein flow from portal vein embolization.
Such a comparison would have been quite inaccu-
rate. Conversely, the standardized measurement
provides a uniform comparison between patients
that is based on a single formula estimating normal
functional total liver volume for all patients. This
new method of measurement is validated by the
correlation with the clinical outcome parameters.
Standardized FLR volume is useful in patients
with small liver remnants who require extended
resection because of multiple small tumors or cen-
trally located tumors. In these patients, there is often
a lack of hypertrophy of the contralateral liver and
the critical threshold for a safe resection needs to be
determined. This study points to the variability in the
size of FLR with segments 2 and 3 ± 1 accounting for
≤25% of the total liver in 8 of 20 patients (40%) at
presentation, emphasizing the utility of preoperative
volume measurements. Experimentally, the chain of
pathologic events ensuing extensive liver resection
that leaves a small liver remnant has been partially
elucidated.18-20 These events include an increase in
portal pressure and mesenteric flow per unit of liver
tissue and possibly endothelial and Kupffer cell
injury leading to a release of inflammatory
cytokines.21 In this study, all 3 major complications
are felt to be part of the spectrum of complications
associated with a small liver remnant. Ascites and par-
tial superior mesenteric vein thrombosis in 2 patients
likely resulted from increased portal pressure and
low portal flow.19,20 Transient jaundice in another
patient pointed to cholestatic dysfunction as a result
of parenchymal injury secondary to increased sinu-
soidal flow.21
518 Vauthey et al
Table II. Results of linear regression of outcome variables on FLR
Postoperative PVE (n = 7)
parameters Beta (SE) R2
Prothrombin time –12 (6) 0.431
Bilirubin –1 (3.5) 0.016
Alkaline phosphatase –536 (193) 0.607
Length of stay –12 (10) 0.227
PVE, Portal vein embolization; R2, total variation in outcome variable that can be explained by variation in FLR; SE, standard error; Beta, slope of linear
regression line. A negative Beta indicates decreasing FLR as the postoperative parameter increases.
The patients included in this series had no signif-
icant underlying liver disease as indicated by the
pathologic review of the specimens. Preliminary
data indicate an increase in major complications in
liver volumes ≤25%. Further studies reporting stan-
dardized liver remnant size will be needed to con-
firm these results and determine the safe limit of
extended resection in liver compromised by steato-
sis, fibrosis, or hepatitis and in resections combined
with other major procedures (such as pancreatecto-
my) or in association with complete vascular exclu-
sion. At this time, FLRs of ≤25% liver volumes,
although associated with an increased risk of com-
plications, are not a contraindication to resection
because the minimal FLR precluding safe resection
has not been determined. For comparison, it is
interesting to note that the minimal size that has
recently been reported for living-related liver trans-
plantation is 32% of the total liver volume.22
With the goal of increasing the volume of the
FLR, Makuuchi23 first described portal vein
embolization before major hepatectomy for hilar
bile duct carcinoma. Portal vein embolization is now
a well-described technique that has been used by
clinicians to increase the FLR, and in the laboratory
portal branch ligation is performed to study the
effects of liver regeneration.24,25 The technique has
allowed a reduction in the liver volume requiring
resection while inducing hypertrophy in the FLR. In
this study, portal vein embolization was performed
as part of a protocol evaluating response and com-
plications associated with portal vein embolization.
Although all patients had an FLR of 40% or less at
presentation, no preset criteria were used to deter-
mine whether the patient would undergo preopera-
tive portal vein embolization.
The effect of the embolization led to an increase
in the absolute FLR volumes of 25% to 80% in the
12 patients who underwent portal vein emboliza-
tion. No complications specific to portal vein
embolization occurred. Likewise, the standardized
FLR measured preoperatively was higher in patients
who had undergone portal vein embolization.
Bilirubin and prothrombin time were not associated
Surgery
May 2000
No PVE (n = 8) All patients (N = 15)
P value Beta (SE) R2 P value Beta (SE) R2 P value
.109 –41 (12) 0.647 .016 –16 (6) 0.316 .029
.786 –48 (15) 0.633 .018 –16 (7) 0.269 .048
.039 –1643 (434) 0.706 .009 –821 (208) 0.546 .002
.280 –72 (18) 0.716 .008 –28 (10) 0.375 .015
with FLR size in this group (Table II), suggesting
minimal functional impairment in excretory or syn-
thetic function, likely as a result of larger-size FLRs.
However, alkaline phosphatase, an index of postop-
erative regeneration,26-28 remained significantly
associated with FLR in this subset of patients, there-
by indicating the value of FLR measurement in case
of portal vein embolization and resections of lesser
extent. This study presents a valuable method to
compare response to portal vein embolization.
However, the small number of patients precludes a
definite conclusion regarding the effect of portal
vein embolization on postoperative mortality or
morbidity.
Recently a small remnant liver volume measured
postoperatively was associated with liver failure after
resection.29 Unfortunately, this evaluation does not
help the surgeon determine which patients can safe-
ly undergo resection because regeneration normally
occurs at a very rapid rate during the first week after
resection.30 Conversely, the preoperative volumetric
assessment as described in this study provides a pre-
liminary method to determine which patients can
undergo liver resection without portal vein emboliza-
tion, which patients may require portal vein
embolization, and which patients are at increased
risk for postoperative complications. Of note is that
this study was conducted in a patient population that
was without significant underlying liver disease. The
majority of patients with hepatocellular carcinoma
will have underlying cirrhosis with impaired liver
regeneration. Therefore, volumetric measurements
of total liver volume and FLR may be more applica-
ble to patients undergoing resection for metastases
or cholangiocarcinoma.31,32 For patients with
advanced chronic liver disease requiring major liver
resection, volumetric measurements may not be
relevant and other predictors of outcome such as
Child-Pugh’s classification or grading of fibrosis
and hepatocellular injury should be evaluated.33
In summary, this study describes a new method
to evaluate the FLR before extended resection with
or without portal vein embolization. It is based on
the FLR volume adjusted for the estimated total
Surgery
Volume 127, Number 5
liver volume regardless of the volume resected. The
measurements appear to correlate with the bio-
chemical postoperative changes. The advantage of
preoperative volumetric liver measurements as
applied in this study lies in their ability to predict
which patients can safely undergo major hepatic
resection and which patients may require portal
vein embolization in hopes of increasing the FLR.
This method should contribute to the preoperative
evaluation of patients in whom major resections
are anticipated.
The authors thank A. G. Encarnacion for data man-
agement and Prof Ph. Gertsch for helpful suggestions in
the preparation of the manuscript.
REFERENCES
1. Tsao JI, Loftus JP, Nagorney DM, Adson MA, Ilstrup DM.
Trends in morbidity and mortality of hepatic resection for
malignancy. A matched comparative analysis. Ann Surg
1994;220:199-205.
2. Vauthey JN, Baer HU, Guastella T, Blumgart LH.
Comparison of outcome between extended and non-
extended liver resections. Surgery 1993;114:968-75.
3. Cunningham JD, Fong Y, Shriver C, Melendez J, Marx WL,
Blumgart LH. One hundred consecutive hepatic resections.
Blood loss, transfusion, and operative technique. Arch Surg
1994;129:1050-6.
4. Kubota K, Makuuchi M, Kusaka K, Kobayashi T, Miki K,
Hasegawa K, et al. Measurement of liver volume and hepat-
ic functional reserve as a guide to decision-making in resec-
tional surgery for hepatic tumors. Hepatology 1997;
26:1176-81.
5. Soyer P, Roche A, Elias D, Levesque M. Hepatic metastases
from colorectal cancer: influence of hepatic volumetric analy-
sis on surgical decision making. Radiology 1992;184:695-7.
6. Nagasue N, Yukaya H, Ogawa Y, Kohno H, Nakamura T.
Human liver regeneration after major hepatic resection.
Ann Surg 1987;206:30-9.
7. Monaco AP, Hallgrimsson J, McDermott WV. Multiple ade-
noma (hamartoma) of the liver treated by subtotal (90%)
resection: morphological and functional studies of regener-
ation. Ann Surg 1964;159:513-9.
8. Starzl T, Putnam C, Groth C, Corman J, Taubman J.
Alopecia, ascites, and incomplete regeneration after 85 to
90 percent liver resection. Am J Surg 1975;129:587-90.
9. Heymsfield S, Fulenwider T, Nordlinger B, Barlow R, Sones
P, Kutner M. Accurate measurement of liver, kidney, and
spleen volume and mass by computerized axial tomogra-
phy. Ann Intern Med 1979;90:185-7.
10. de Baere T, Roche A, Elias D, Lasser P, Lagrange C, Bousson
V. Preoperative portal vein embolization for extension of
hepatectomy indications. Hepatology 1996;24:1386-91.
11. Urata K, Kawasaki S, Matsunami H, Hashikura Y, Ikegami T,
Ishizone S, et al. Calculation of child and adult standard liver
volume for liver transplantation. Hepatology 1995;21:1317-21.
12. Mosteller RD. Simplified calculation of body surface area
[letter]. N Engl J Med 1987;317:1098.
13. SAS/STAT User’s Guide [computer program]. Version 6,
4th ed, Volume I, Cory, NC: SAS Institute Inc, 1990.
14. S-PLUS 5 for Unix Guide to Statistics [computer program].
Seattle: Data Analysis Products Division, Mathsoft, 1998.
15. Scheele J, Stangl R. Segment oriented anatomical liver
Vauthey et al 519
resections. In: Blumgart LH, editor. Surgery of the liver and
biliary tract. 2nd ed. New York: Churchill Livingstone; 1994.
p. 1557-78.
16. Ogasawara K, Une Y, Nakajima Y, Uchino J. The significance
of measuring liver volume using computed tomographic
images before and after hepatectomy. Jpn J Surg
1995;25:43-8.
17. Nagino M, Nimura Y, Kamiya J, Kondo S, Uesaka K, Kin Y,
et al. Changes in hepatic lobe volume in biliary tract cancer
patients after right portal vein embolization. Hepatology
1995;21:434-9.
18. Kawasaki T, Moriyasu F, Kimura T, Someda H, Fukuda Y,
Ozawa K. Changes in portal blood flow consequent to par-
tial hepatectomy: Doppler estimation. Radiology
1991;180:373-7.
19. Hanna SS, Pagliarello G, Ing A. Liver blood flow after major
hepatic resection. Can J Surg 1988;31:363-7.
20. Kahn D, van Hoorn-Hickman R, Terblanche J. Liver blood
flow after partial hepatectomy in the pig. J Surg Res
1984;37:290-4.
21. Panis Y, McMullan DM, Emond JC. Progressive necrosis
after hepatectomy and the pathophysiology of liver failure
after massive resection. Surgery 1997;121:142-9.
22. Kawasaki S, Makuuchi M, Matsunami H, Hashikura Y,
Ikegami T, Nakazawa Y, et al. Living related liver transplan-
tation in adults. Ann Surg 1998;227:269-74.
23. Makuuchi M, Thai BL, Takayasu K, Takayama T, Kosuge T,
Gunven P, et al. Preoperative portal embolization to
increase safety of major hepatectomy for hilar bile duct car-
cinoma: a preliminary report. Surgery 1990;107:521-7.
24. Katoh T, Tanaka M, Nimura Y, Kanai M, Nagino M, Ozawa
T. Enhancement of rat liver mitochondrial function by por-
tal branch ligation secures subsequent extended hepatecto-
my. Biochem Intl 1991;24:107-16.
25. Morsiani E, Fogli L, Rozga J, Ricci D, Azzena G, Demetriou
AA. Growth of intraportally transplanted islets under liver
regeneration stimulus and restoration of normoglycemia in
streptozocin-diabetic rats. Surgery 1998;123:398-406.
26. Suc B, Panis Y, Belghiti J, Fékété F. ‘Natural history’ of hepa-
tectomy. Br J Surg 1992;79:39-42.
27. Aronsen KF, Ericsson B. Metabolic changes following major
hepatic resection. Ann Surg 1969;169:102-20.
28. Burke J. Serum alkaline phosphatase in liver disease. A con-
cept of significance. Gastroenterology 1950;16:660-2.
29. Shirabe K, Shimada M, Gion T, Hasegawa H, Takenaka K,
Utsunomiya T, et al. Postoperative liver failure after major
hepatic resection for hepatocellular carcinoma in the mod-
ern era with special reference to remnant liver volume. J
Am Coll Surg 1999;188:304-7.
30. Chaoui AS, Mergo PJ, Vauthey JN, Taylor HM, Ros PR,
Caridi JG. 3D volumetric helical CT quantification of liver
volumes following major hepatic resection: comparison in
patients with and without pre-surgical portal vein emboliza-
tion. Radiology 1998;209:215-6.
31. Chen MF, Hwang TL, Hung CF. Human liver regeneration
after major hepatectomy. A study of liver volume by com-
puted tomography. Ann Surg 1991, 213:227-9.
32. Yamanaka N, Okamoto E, Kawamura E, Kato T, Oriyama T,
Fujimoto J, et al. Dynamics of normal and injured human
liver regeneration after hepatectomy as assessed on the
basis of computed tomography and liver function.
Hepatology 1993;18:79-85.
33. Farges O, Malassagne B, Flejou JF, Balzan S, Sauvanet A,
Belghiti J. Risk of major liver resection in patients with
underlying chronic liver disease. A reappraisal. Ann Surg
1999;229:210-5.