- Email: [email protected]
Efficacy and Safety of Portal Vein Embolization for Two-Stage Hepatectomy in Patients with Colorectal Liver Metastasis
CLINICAL STUDY
Efficacy and Safety of Portal Vein Embolization for Two-Stage Hepatectomy in Patients with Colorectal Liver Metastasis Steven Y. Huang, MD, Thomas A. Aloia, MD, Junichi Shindoh, MD, PhD, Joe Ensor, PhD, Colette M. Shaw, MD, Evelyne M. Loyer, MD, Jean-Nicolas Vauthey, MD, and Michael J. Wallace, MD
ABSTRACT Purpose: To examine the efficacy and safety of portal vein embolization (PVE) when used during two-stage hepatectomy for bilobar colorectal liver metastases (CLM). Materials and Methods: PVE was performed as an adjunct to two-stage hepatectomy in 56 patients with CLM. Absolute future liver remnant (FLR) volumes, standardized FLR ratios, degree of hypertrophy (DH), and complications were analyzed. Segment II and III volumes and DH were also measured separately. All volumetric measurements were compared with a cohort of 96 patients (n ¼ 37 right portal vein embolization [RPVE], n ¼ 59 right portal vein embolization extended to segment IV portal veins [RPVEþ4]) in whom PVE was performed before single-stage hepatectomy. Results: For patients who completed RPVE during two-stage hepatectomy (n ¼ 17 of 17), mean absolute FLR volume increased from 272.1 cm3 to 427.0 cm3 (P o .0001), mean standardized FLR ratio increased from 0.17 to 0.26 (P o .0001), and mean DH was 0.094. For patients who completed RPVEþ4 during two-stage hepatectomy (n ¼ 38 of 39), mean FLR volume increased from 288.7 cm3 to 424.8 cm3 (P o .0001), mean standardized FLR increased from 0.18 to 0.26 (P o .0001), and mean DH was 0.083. DH of the FLR was not significantly different between two-stage hepatectomy and single-stage hepatectomy. Complications after PVE occurred in five (8.9%) patients undergoing two-stage hepatectomy. Conclusions: PVE effectively and safely induced a significant DH in the FLR during two-stage hepatectomy in patients with CLM.
ABBREVIATIONS CLM = colorectal liver metastasis, DH = degree of hypertrophy, FLR = future liver remnant, PVE = portal vein embolization, RPVE = right portal vein embolization, RPVEþ4 = right portal vein embolization extended to segment IV portal veins, RV = resection volume, S2þ3 = segments II and III, S4 = segment IV, TLV = total liver volume
Surgical resection of metastatic disease is regarded as the most effective strategy for long-term survival in patients with colorectal liver metastasis (CLM), with reported From the Departments of Diagnostic Radiology (S.Y.H., E.M.L., M.J.W.), Surgical Oncology (T.A.A., J.S., J.-N.V.), and Biostatistics (J.E.), The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030; and Department of Radiology (C.M.S.), Jefferson University Hospitals, Philadelphia, Pennsylvania. Received May 16, 2013; final revision received October 8, 2013; accepted October 11, 2013. Address correspondence to S.Y.H.; E-mail: [email protected] From the SIR 2013 Annual Meeting. None of the authors have identified a conflict of interest. & SIR, 2014 J Vasc Interv Radiol 2014; 25:608–617 http://dx.doi.org/10.1016/j.jvir.2013.10.028
5-year survival rates of 58% (1–3). However, only 10%– 20% of patients with CLM are deemed surgical candidates (4). Two key reasons why CLM is considered unresectable are the risk of major morbidity and mortality related to an insufficient future liver remnant (FLR) (5) and the presence of bilobar hepatic metastases involving any portion of FLR. Preoperative portal vein embolization (PVE) was developed to induce hypertrophy in FLR to address the issue of postoperative hepatic insufficiency (6–8). Now that PVE has gained acceptance as a reliable tool to improve FLR adequacy, attention is focused on addressing disease within FLR by performing a two-stage hepatectomy with PVE performed between the two stages. Disease in FLR is resected in a limited single-stage hepatectomy, and then PVE is performed to induce hypertrophy of FLR to an
Volume 25
’
Number 4
’
April
’
2014
adequate volume. Finally, resection of the remaining metastatic disease is performed during the final stage (9). Two-stage hepatectomy is safe and effective in patients with bilobar CLM (9–12). Although hypertrophy of FLR does occur in response to PVE in the setting of two-stage hepatectomy (5,9,13,14), it is unclear whether the insult caused by first-stage liver resection impairs the magnitude of FLR hypertrophy induced by PVE. Prior studies have been limited by small numbers of patients undergoing two-stage hepatectomy and PVE (5,14), lack of an appropriate control cohort of patients undergoing PVE in the setting of single-stage hepatectomy (9,13), and differences in PVE technique limiting generalizability of results (5,14). A recent retrospective study from our institution analyzing the efficacy of PVE in patients with very low FLR volume found that adequate regeneration was observed in nearly all of these patients (96.5%) (13). Of 144 patients undergoing percutaneous right portal vein embolization extended to segment IV portal veins (RPVEþ4), 32 (22%) procedures were performed after a first-stage partial left hepatectomy as part of a two-stage hepatectomy approach. Our current work builds on this earlier analysis and is unique in many ways. First, our analysis includes an additional 23 patients who underwent PVE as an adjunct to two-stage hepatectomy (total n ¼ 55 patients; n ¼ 38 patients undergoing RPVEþ4, n ¼ 17 undergoing right portal vein embolization [RPVE]). Second, in our earlier work, all 144 patients (n ¼ 32 RPVEþ4 two-stage hepatectomy and n ¼ 112 RPVEþ4 single-stage hepatectomy) were grouped to calculate percentage of hypertrophy, making it difficult to determine if RPVEþ4 in the setting of two-stage hepatectomy is efficacious compared with a control cohort of patients who underwent RPVEþ4 in the setting of single-stage hepatectomy. Third, the focus of our previous work was to describe the operative morbidity and mortality after the second-stage operation of two-stage hepatectomy; specific morbidity data related to PVE was limited. The main purpose of our study was to examine the efficacy and safety of PVE on FLR hypertrophy when used in conjunction with two-stage hepatectomy in patients with CLM compared with a control cohort of patients in whom PVE was performed as an adjunct to single-stage hepatectomy. An additional issue that is controversial is the utility of performing RPVEþ4 (15). Prior studies of patients undergoing single-stage hepatectomy compared the changes in volume for segments II and III (S2þ3) after RPVE and RPVEþ4 (15–17), but results have differed with two groups finding an increase in S2þ3 volumes after RPVEþ4 (15,16) and one group finding no significant increase in S2þ3 volumes (17). The efficacy of RPVEþ4 in patients undergoing two-stage hepatectomy has yet to be described, and so a secondary purpose of our study was to evaluate the effect RPVEþ4 had on
609
S2þ3 hypertrophy for patients undergoing two-stage hepatectomy.
MATERIALS AND METHODS We retrospectively reviewed 168 consecutive patients with CLM who underwent PVE between May 1998 and January 2011 for this study, which was compliant with the Health Insurance Portability and Accountability Act and approved by our institutional review board. PVE was performed as a component of two-stage hepatectomy in 65 patients; the remaining 103 patients underwent PVE before single-stage hepatectomy and were used as a comparison cohort. Both the two-stage hepatectomy group and the single-stage hepatectomy group were divided into subgroups based on whether RPVEþ4 was performed. There were 16 patients excluded because of missing computed tomography (CT) volumetric data (n ¼ 9), staged PVE (ie, performing PVE in separate procedures; n ¼ 4), and percutaneous intraoperative radiofrequency ablation performed in lieu of surgical resection during first-stage hepatectomy (n ¼ 3). Our study population comprised the remaining 152 patients (Fig 1). Two-stage hepatectomy was performed in 56 patients (n ¼ 17 for RPVE and n ¼ 39 for RPVEþ4), and singlestage hepatectomy was performed in 96 patients (n ¼ 37 for RPVE and n ¼ 59 for RPVEþ4). No patients underwent transarterial chemoembolization or radioembolization, and no patients had evidence of cirrhosis. Data were retrospectively collected on age, sex, diabetes status, body mass index, and use or nonuse of neoadjuvant chemotherapy within 3 months of PVE. Clinical characteristics for the four subgroups of patients in this study (RPVE single-stage hepatectomy, RPVE twostage hepatectomy, RPVEþ4 single-stage hepatectomy, and RPVEþ4 two-stage hepatectomy) are shown in Table 1. During the study period, patients underwent PVE if the volume of FLR was r 20% of the standardized total liver volume (TLV) for normal liver (18), r 30% in patients with fibrosis or severe liver injury (19), and r 40% in patients with cirrhosis (20,21). Standardized TLV was calculated using a formula for body surface area in square meters: standardized TLV = 794.41 þ 1,267.28 body surface area (22). All patients generally underwent abdominal CT with volumetry 2–8 weeks after PVE. At our institution, PVE was performed via a transhepatic ipsilateral (ie, on the side of the liver being resected) approach, which has been described previously (23–25). We use a combination of trisacryl particles and coils to occlude the branches of the right portal vein with or without segment IV (S4) portal veins. Although we realize there are alternative occlusive agents, we prefer to use particles and coils because our technique results in
610
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
Huang et al
’
JVIR
Figure 1. Flow chart of patients included in analysis. RFA ¼ radiofrequency ablation. Table 1 . Patient Characteristics RPVE
Mean age (range) (y) Male sex (ratio) Presence of diabetes (ratio) Mean BMI (range) (kg/m2)
RPVEþ4
SSH (n ¼ 37)
TSH (n ¼ 17)
P Value
SSH (n ¼ 59)
TSH (n ¼ 39)
59.7 (37–81)
50.6 (35–64)
.0029
57.7 (37–78)
51.4 (33–68)
.0028
.2231 .9131
81.4% (48/59) 20.3% (12/59)
59.0% (23/39) 15.4% (6/39)
.0150 .5401
70.3% (26/37) 18.9% (7/37) 30.3 (29.4–41.2)
Use of neoadjuvant chemotherapy 91.9% (34/37) within 3 mo of PVE (ratio)
56.3 (9/17) 17.6% (3/17) 27.1 (20.2–35.9) 100.0% (17/17)
.0432 .2349
28.8 (19.8–44.5) 89.8% (53/59)
27.3 (17.8–42.8) 94.3% (37/39)
P Value
.1597 .3775
BMI ¼ body mass index, PVE ¼ portal vein embolization, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy.
very few patients in whom the FLR does not hypertrophy sufficiently to allow resection. In patients with very low FLR, only approximately 3.5% fail to develop hypertrophy sufficiently to allow definitive hepatectomy (13). Portal vein pressures were obtained via a multiside hole catheter before and after PVE. RPVEþ4 was performed when an extended right hepatectomy was planned on the basis of tumor location. Technical success of the procedure was defined as embolization of the intended segments with no residual flow to the liver after embolization detected by final portal venography.
Assessment of PVE Effectiveness CT scans before and after PVE were performed to calculate absolute FLR volumes, change in FLR after
PVE, standardized FLR (ratio of FLR to standardized TLV), and degree of hypertrophy (DH) (difference between FLR before PVE and FLR after PVE) using density threshold seeding (26). Two patients (one patient from the RPVEþ4 two-stage hepatectomy group and one patient from the RPVEþ4 single-stage hepatectomy group) were excluded from the volume analysis because of portal vein thrombus that developed in the left lateral segmental portal vein during PVE. In both cases, the thrombus required overnight local thrombolytic administration, and PVE was not completed. Excluding these two patients, volume analysis for the RPVEþ4 twostage hepatectomy was performed on 38 patients, and RPVEþ4 single-stage hepatectomy group was performed on 58 patients. For patients undergoing twostage hepatectomy, CT was performed after first-stage
Volume 25
’
Number 4
’
April
’
2014
hepatectomy to assess the liver remnant, remeasure FLR, and detect any complications from the first-stage hepatectomy that would exclude patients from PVE. However, 16 of 56 patients who underwent two-stage hepatectomy did not undergo CT imaging after the firststage resection. The resected volumes from the first-stage hepatectomy were obtained from our surgical pathology database and manually subtracted from FLR volumes generated from the baseline CT scan obtained before hepatectomy. Correlation between DH and standardized resection volume (RV) (ratio of RV to standardized TLV) was assessed for the patients undergoing two-stage hepatectomy to determine if the amount of liver resected during the first-stage liver resection affected FLR hypertrophy after PVE.
Assessment of Effectiveness of RPVEþ4 Liver volumetry results were reviewed, and the absolute volume and DH of S2þ3 and S4 before and after PVE were compared between the RPVEþ4 and RPVE groups for both two-stage hepatectomy and single-stage hepatectomy. For the RPVE two-stage hepatectomy group, S2þ3 volumes were calculated based on 15 patients, whereas S4 volumes were calculated based on 17 patients. This discrepancy was due to two patients in the RPVE two-stage hepatectomy group who underwent complete resection of S2þ3 during the first stage of twostage hepatectomy.
Adverse Events Complications were evaluated by reviewing the electronic medical record from immediately after PVE through surgical resection for all patients (n ¼ 152). If the patient was unable to undergo surgery, the medical record was evaluated for periprocedural complications (up to 30 days). Complications were graded according to the Clavien-Dindo classification of surgical complications (27). Reasons for failure to progress to curative hepatectomy were also tabulated.
Statistical Analysis Volumetric results were reported as means and ranges. Continuous data were compared with a t test (pooled or Satterthwaite, as appropriate), and categorical data were compared with a 2-tailed Fisher exact test. Logistic regression models were used to evaluate the correlation between DH with resected liver volumes during the first stage of a two-stage hepatectomy and standardized FLR volumes before PVE. Pearson product-moment correlation was used to calculate the correlation between S2þ3 and S4 volume hypertrophy after RPVEþ4. P o .05 was considered statistically significant. Statistical analysis was performed with JMP software (SAS Institute, Cary, North Carolina).
611
RESULTS PVE was successfully completed in 55 of 56 patients undergoing two-stage hepatectomy and 95 of 96 patients undergoing single-stage hepatectomy. PVE was performed at a mean of 38.6 days (range, 4–288 d) after first-stage hepatectomy. The average interval between PVE and follow-up CT scan was 36.4 days (range, 15– 151 d). CT and portal venography images of a representative patient who underwent two-stage hepatectomy are shown in Figure 2a–d. When we compared singlestage hepatectomy with two-stage hepatectomy for RPVE and RPVEþ4, there was a significant difference detected in age; patients undergoing two-stage hepatectomy tended to be younger than patients undergoing single-stage hepatectomy. There were significantly more male patients in the RPVEþ4 single-stage hepatectomy group than the RPVEþ4 two-stage hepatectomy group. Body mass index was significantly lower in the RPVE two-stage hepatectomy group compared with the RPVE single-stage hepatectomy group. No difference was detected in the prevalence of diabetes and use or nonuse of neoadjuvant chemotherapy within 3 months of PVE between the single-stage hepatectomy and two-stage hepatectomy groups.
Assessment of PVE Effectiveness Mean absolute FLR volumes and standardized FLR ratios increased for all four groups after PVE (Table 2). DH was comparable between the RPVE two-stage hepatectomy and RPVE single-stage hepatectomy subgroups, measuring 0.094 and 0.097, respectively (P ¼ .8387). Similarly, no significant difference was noted in DH between the RPVEþ4 two-stage hepatectomy and RPVEþ4 single-stage hepatectomy subgroups, measuring 0.083 and 0.095, respectively (P ¼ .1816). FLR volumes before PVE were adjusted to account for the first-stage liver resection, as described in the Materials and Methods section. The amount of liver resected during the first-stage operation for patients undergoing twostage hepatectomy relative to the TLV ranged from 0.0001–0.421. Logistic regression did not detect a correlation between the standardized RV after first-stage liver resection and DH (ρ ¼ 0.12, P ¼ .38).
Assessment of Effectiveness of RPVEþ4 For single-stage hepatectomy, RPVEþ4 resulted in a significant increase to S2þ3 DH. The DH measured 0.089 for RPVEþ4 and 0.068 for RPVE (P ¼ .0095). For two-stage hepatectomy, RPVEþ4 yielded S2þ3 DH of 0.078 compared with 0.059 for RPVE (P ¼ .1602) (Table 3).
Adverse Events Complications occurred in 8 of 96 (8%) patients who underwent single-stage hepatectomy and 5 of 56 (9%)
612
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
Huang et al
’
JVIR
Figure 2. Sequence of events during two-stage hepatectomy. (a) Axial contrast-enhanced CT image at the level of the left portal vein shows a metastasis in the peripheral right lobe and a subtle metastasis in the lateral segment of the left hepatic lobe (arrow). (b) Axial contrast-enhanced CT image at the level of the left portal vein shows changes after surgery in the lateral segment of the left hepatic lobe (arrow). Metastasis in the peripheral right hepatic lobe is unchanged. During stage 1 hepatectomy, 50 cm3 of segment III was resected. FLR is depicted by the white dotted line. (c) Axial contrast-enhanced CT image through the level of the left portal vein performed 29 days after PVE shows interval growth of FLR (white dotted line), which includes segments I, II, and III. Standardized FLR increased from 0.155 to 0.274, yielding a DH of 0.119. (d) Axial contrast-enhanced CT image at the level of the left portal vein performed 203 days after second-stage hepatectomy shows hypertrophy of FLR.
Table 2 . Future Liver Remnant and Standardized Future Liver Remnant Hypertrophy after Right Portal Vein Embolization and Right Portal Vein Embolization Extended to Segment IV Portal Veins RPVE
RPVEþ4
Metric
SSH (n ¼ 37)
TSH (n ¼ 17)
P Value
SSH (n ¼ 58)
TSH (n ¼ 38)
P Value
FLRpre-PVE (range) (mL) FLRpost-PVE (range) (mL)
434.4 (122.2–1146.2) 612.4 (216.8–1351.9)
272.1 (55.1–524.4) 427.0 (148.1–853.4)
.0078 .0139
298.5 (145.2–564.3) 470.1 (277.2–682.4)
288.7 (78.6–526.3) 424.8 (85.0–888.5)
.6311 .1016
FLRchange (range) (mL)
177.9 (29.1–501.9)
154.9 (27.7–464.0)
.4661
171.5 (23.6–357.9)
136.1 (6.4–443.3)
.0372
0.24 (0.06–0.55) 0.34 (0.10–0.65)
0.17 (0.03–0.34) 0.26 (0.08–0.47)
.0256 .0448
0.17 (0.07–0.36) 0.26 (0.14–0.43)
0.18 (0.06–0.42) 0.26 (0.06–0.50)
.2765 .9182
0.097 (0.02–0.24)
0.094 (0.02–0.26)
.8387
0.095 (0.01–0.19)
0.083 (0.01–0.23)
.1816
sFLRpre-PVE sFLRpost-PVE DH
DH ¼ degree of hypertrophy (sFLRpostPVE sFLRprePVE), FLR ¼ future liver remnant, FLRchange ¼ change in future liver remnant, FLRpostPVE ¼ future liver remnant after portal vein embolization, FLRprePVE ¼ future liver remnant before portal vein embolization, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, sFLR ¼ standardized future liver remnant (FLR/standardized total liver volume), sFLRpostPVE ¼ standardized future liver remnant after portal vein embolization, sFLRprePVE ¼ standardized future liver remnant before portal vein embolization, SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy.
Volume 25
’
Number 4
’
April
’
2014
613
Table 3 . Changes in Segments II and III and Segment IV Volumes before and after Right Portal Vein Embolization and Right Portal Vein Embolization Extended to Segment IV Portal Veins SSH P Value
RPVE (n ¼ 15)
252.2 (83.2–555.6)
266.7 (65.3–506.6)
.4667
173.9 (31.1–292.5)
261.1 (111.9–478.7)
.0039
Volumepost-PVE (range) (mL)
372.7 (142.7–723.9)
426.8 (118.5–637.2)
.0325
273.3 (87.8–499.4)
385.4 (118.5–836.3)
.0101
Volumechange
120.5 (28.0–282.2)
160.1 (18.5–336.2)
.0082
99.4 (10.9–227.6)
127.8 (2.6–417.6)
.2408
(range) (mL) DH
0.068 (0.016–0.137)
0.089 (0.009–0.183)
.0095
0.059 (0.007–0.119)
0.078 (0.001–0.213)
.1602
S4 Volumepre-PVE
RPVE (n ¼ 37) 221.1 (97.9–527.8)
RPVEþ4 (n ¼ 58) 220.6 (34.2–553.0)
P Value .9808
RPVE (n ¼ 17) 194.0 (87.7–320.3)
RPVEþ4 (n ¼ 37) 225.5 (95.1–442.4)
P Value .1574
290.7 (154.2–544.2)
265.0 (70.1–453.4)
.2070
295.5 (162.0–496.7)
296.7 (57.5–494.4)
.9694
S2þ3 Volumepre-PVE
RPVE (n ¼ 37)
TSH
RPVEþ4 (n ¼ 58)
RPVEþ4 (n ¼ 38)
P Value
(range) (mL)
(range) (mL) Volumepost-PVE (range) (mL) Volumechange (range) (mL) DH
69.6 (8.3–254.6)
44.5 (202.1–175.2)
.0717
101.5 (22.0–261.5)
71.2 (46.4–216.7)
.1295
0.037 (0.006–0.133)
0.025 (0.079–0.103)
.0937
0.060 (0.016–0.145)
0.045 (0.029–0.140)
.2035
DH ¼ degree of hypertrophy, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ single-stage hepatectomy, S2þ3 ¼ segments II and III, S4 ¼ segment IV, TSH ¼ two-stage hepatectomy, volumechange ¼ volume change, volumepost-PVE ¼ volume after portal vein embolization, volumepre-PVE ¼ volume before portal vein embolization.
Table 4 . Surgical Complications according to the Clavien-Dindo Classification System RPVE SSH (n ¼ 37)
RPVEþ4 SSH (n ¼ 59)
RPVE TSH (n ¼ 17)
I II
1a 0
3b,c,d 3e,f,g
0 0
3i,j,k 1l
IIIa
0
1h
0
1m
IIIb IVa
0 0
0 0
0 0
0 0
IVb
0
0
0
0
V
0
0
0
0
Surgical Complication Grade
RPVEþ4 TSH (n ¼ 39)
RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ singlestage hepatectomy, TSH ¼ two-stage hepatectomy. a Fever requiring antibiotics. b Right upper quadrant pain requiring narcotics. c Fever requiring antipyretic. d Nontarget embolization (coil misplaced in segment III portal vein branch). e Main portal vein thrombus requiring anticoagulation. f Main portal vein and superior mesenteric vein thrombus requiring anticoagulation. g Subcapsular hematoma requiring blood transfusion. h Left portal vein thrombosis during RPVEþ4 requiring catheter-directed thrombolysis. i Right upper quadrant pain requiring narcotics. j Right upper quadrant pain requiring narcotics. k Nontarget embolization (coil misplaced at segment III portal vein origin). l Nonocclusive left portal vein thrombus requiring anticoagulation for 3 months. m Left portal vein thrombosis during RPVEþ4 requiring catheter-directed thrombolysis.
patients who underwent two-stage hepatectomy (P ¼ .90) (Table 4). For the two-stage hepatectomy group, there were five surgical complications; only one was greater than grade II in severity. This complication arose from acute thrombosis of the left lateral segmental portal vein during RPVEþ4, which required catheter-directed thrombolysis for 24 hours. Follow-up portal venography
showed persistent thrombus, and the patient was started on anticoagulation therapy. A CT scan performed 1 month later showed disease progression, and the patient was no longer a surgical candidate. For the single-stage hepatectomy group, only one of eight complications measured greater than grade II in severity. For this patient, RPVEþ4 was complicated by
614
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
partial thrombosis in the left lateral segmental portal vein, requiring catheter-directed thrombolysis for 24 hours and 3 months of anticoagulation with subsequent resolution of the thrombus. Despite the complication, the patient underwent definitive hepatectomy, although the patient’s original extended right hepatectomy was changed to a partial right hepatectomy extended to S4 with intraoperative radiofrequency ablation of residual metastatic lesions along the resection margin. Complications related to nontarget thrombus formation requiring thrombolysis or anticoagulation or both were unique to RPVEþ4. The incidence was 5 of 98 cases (5%), whereas there were no incidents of nontarget thrombus Table 5 . Factors Relating to Patient Ineligibility for Curative Resection SSH (n ¼ 96) TSH (n ¼ 56) P Value Disease progression Inadequate hypertrophy
13 3
10 5
.4894 .1451
Thrombocytopenia High-risk surgery*
1 4
1 0
1.0000 .2971
Development of new or
0
4
.0172
Chemotherapy-related liver toxicity
2
0
.2532
Steatohepatitis
1
0
1.0000
worsening imaging manifestations of portal hypertension
SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy. *Elevated body mass index or abdominal adhesions from prior surgery.
Huang et al
’
JVIR
formation when PVE was limited to the right portal vein (0 of 54 cases) (P ¼ .1613). Curative hepatectomy was performed in 72 of 96 (75%) patients who underwent single-stage hepatectomy and 36 of 56 (64%) patients who underwent two-stage hepatectomy (P ¼ .1621). Reasons patients were unable to undergo definitive resection are listed in Table 5. Of 56 patients undergoing two-stage hepatectomy, 4 (7%) were unable to undergo definitive hepatectomy because of new or worsening imaging findings of portal hypertension; none of the 96 patients undergoing single-stage hepatectomy developed this finding (P ¼ .0077). New or worsening imaging manifestations of portal hypertension included new gastroesophageal varices or new or worsening splenomegaly on follow-up CT scan performed approximately 1 month after PVE (Fig 3a, b). The portal venous pressures before and after PVE were available for three of these four patients. Portal venous pressures before PVE measured 5 mm Hg, 5 mm Hg, and 18 mm Hg. Portal venous pressures after PVE measured 13 mm Hg, 13 mm Hg, and 22 mm Hg. In comparison, for patients undergoing two-stage hepatectomy who underwent definitive hepatectomy, portal venous pressures before PVE were available for 29 of 36 (80.6%) patients (mean, 10.1 mm Hg; range, 1–18 mm Hg), and portal venous pressures after PVE were available for 28 of 36 (77.8%) patients (mean, 15.5 mm Hg; range, 3–29 mm Hg). For the four patients with new or worsening imaging manifestations of portal hypertension, standardized RV after first-stage liver surgery was 0.103, 0.117, 0.218, and 0.211. Standardized RV for 36 patients undergoing two-stage hepatectomy who underwent definitive hepatectomy without developing
Figure 3. New gastroesophageal varices after first-stage hepatectomy and RPVEþ4. (a) Axial contrast-enhanced CT image at the gastroesophageal junction 21 days before RPVE shows no definite varices (arrow). (b) Axial contrast-enhanced CT image at the gastroesophageal junction 28 days after RPVE shows varices (arrow).
Volume 25
’
Number 4
’
April
’
2014
new or worsening imaging manifestations of portal hypertension was 0.048 (range, 0.005–0.420). After definitive hepatectomy, the postoperative 90-day mortality rate was 4% for the single-stage hepatectomy group (3 of 72 patients) and 6% for the two-stage hepatectomy group (2 of 36 patients). All three of the deaths in the single-stage hepatectomy group and one of the deaths in the two-stage hepatectomy group were related to progressive postoperative liver failure and occurred 10 days, 28 days, and 40 days after singlestage hepatectomy and 18 days after two-stage hepatectomy. The second death in the two-stage hepatectomy group was due to cardiac arrest 5 days after two-stage hepatectomy.
DISCUSSION As an adjunct to two-stage hepatectomy, PVE is an effective and safe technique to induce FLR hypertrophy. The increases in absolute FLR volume, standardized FLR, and DH after PVE for two-stage hepatectomy were similar to the increases witnessed with PVE before single-stage hepatectomy. Complication rates with PVE in the setting of two-stage hepatectomy and single-stage hepatectomy were also similar. PVE is an effective means to induce hypertrophy of FLR in the setting of two-stage hepatectomy and does not appear to be negatively affected by recent liver resection. Previous studies have documented the increase in absolute FLR volume after PVE during two-stage hepatectomy (5,9,14). In our series, FLR volume increased from 272.1 mL to 427.0 mL for RPVE (n = 17) and 288.7 mL to 424.8 mL for RPVEþ4 (n = 38); this is comparable to FLR hypertrophy described by Jaeck et al (429 mL to 564 mL, n = 33) (9) and Tanaka et al (243.8 mL to 350.5 mL, n ¼ 16) (14). These results reinforce the utility of PVE in the setting of two-stage hepatectomy. Our study was notable for two reasons. First, our PVE technique involved a transhepatic ipsilateral approach with particles and coils, as opposed to the technique used by Jaeck et al (9), which involved a contralateral approach without describing the occlusive agent, and Tanaka et al (14), which involved cannulation of the ileocolic vein during surgery followed by administration of gelatin pellets and oleic acid. Second, we evaluated liver volume changes after RPVE and RPVEþ4 separately, and we were better able to determine the effect of two-stage hepatectomy on each group using the single-stage hepatectomy cohorts for comparison. As previously described, RPVEþ4 yielded a significant increase in S2þ3 volumes for patients undergoing single-stage hepatectomy (15). Our results confirmed this finding for the single-stage hepatectomy cohort. For the two-stage hepatectomy cohort, we did not detect a significant increase in S2þ3 volumes when RPVEþ4 was
615
performed (P ¼ .1602). The limited number of patients in our RPVE two-stage hepatectomy subgroup (n ¼ 15) may have hindered our ability to detect a significant increase in S2þ3 volumes after RPVEþ4. The effect of extended chemotherapy may also play a role because patients undergoing two-stage hepatectomy are generally treated with extensive systemic chemotherapy before surgery, which has been shown to impair hypertrophy of FLR after PVE (28,29). In our analysis, we evaluated whether or not chemotherapy had been administered during the 3 months before PVE. The use of perioperative chemotherapy seems to be ubiquitous in our patient population. Future analysis may consider comparing the duration of chemotherapy for patients undergoing two-stage hepatectomy versus single-stage hepatectomy to determine if length of treatment can affect FLR hypertrophy. Part of the technically challenging nature of embolization of S4 portal veins is obtaining complete embolization of S4 portal branches without causing reflux and showering embolic particles into S2þ3. It may be possible that during S4 embolization, the operators either stopped short of complete stasis or did not perform embolization of all branches, blunting the increased S2þ3 hypertrophy seen with RPVEþ4 in the single-stage hepatectomy cohort. After all, despite embolization extended to S4 portal veins, the DH of S4 still generally increased on follow-up CT scans for single-stage hepatectomy (mean, 0.025) and twostage hepatectomy (mean, 0.045) (Table 3). Finally, and most importantly, the decision to perform RPVEþ4 is made based on tumor location. Several studies have described the possibility of accelerated tumor growth in portions of the liver that did not receive embolization (30–32). At our institution, if metastases are situated in the right liver and S4, RPVEþ4 is generally performed to prevent tumor growth, regardless of baseline S2þ3 volumes. Complication rates after PVE in the setting of twostage hepatectomy and single-stage hepatectomy were similar. Complications were generally grade I and II according to the Clavien-Dindo classification system and did not alter the surgical plan. One particular complication, nontarget portal vein thrombosis, occurred only in the setting of RPVEþ4. Although this association was not statistically significant (P ¼ .1613), analysis was limited by the relative infrequency of this complication (n ¼ 5). Nontarget thrombosis of portal veins may potentially be attributed to the challenging nature of PVE performed in S4, which requires longer procedure times with indwelling catheters that alter portal vein hemodynamics. The percentage of patients who ultimately underwent definitive surgical resection after PVE was similar for the two-stage hepatectomy (64%) and single-stage hepatectomy (75%) groups. A small subset of patients in the two-stage hepatectomy group (n ¼ 4) did not undergo definitive hepatectomy because of new or worsening
616
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
radiographic signs of portal hypertension. Previous reports have described increases in portal pressure after partial hepatectomy involving 4 30% of the liver (33,34). Although RVs during the first-stage hepatectomy of our study were usually o 30%, it may stand to reason that larger volumes of liver resected during the first stage of two-stage hepatectomy, in combination with PVE, may predispose patients to new or worsening portal hypertension. In our study, standardized RV (ratio of RV to TLV) for the patients who developed new or worsening portal hypertension was 0.103, 0.117, 0.218, and 0.211, as opposed to a mean of 0.048 for the 36 patients who underwent two-stage hepatectomy, who completed second-stage hepatectomy without developing new or worsening imaging manifestations of portal hypertension. In addition, the use of preoperative chemotherapy in our two-stage hepatectomy population (n ¼ 54 of 56; 96%) may play a role. Specifically, oxaliplatin-based chemotherapy is known to cause sinusoidal injury and contribute to the development of noncirrhotic portal hypertension (35,36). This study has several limitations. The study design was retrospective in nature, which has inherent limitations. In addition, the number of patients in this study was small, limiting statistical analysis. Our reliance on medical chart review may not have captured all of the medical decision making or complications. Portal pressures before and after PVE were available for 80.6% and 77.8%, respectively, of embolization procedures performed as an adjunct to two-stage hepatectomy, limiting full evaluation of the influence of immediate changes in portal pressure on the development of portal hypertension. In conclusion, two-stage hepatectomy is an important treatment modality for patients with multiple bilobar CLMs. In these patients, PVE is an effective, safe, and reliable technique to induce hypertrophy of standardized FLR with results similar to patients undergoing PVE before single-stage hepatectomy. RPVEþ4 increased S2þ3 volumes before single-stage hepatectomy. A significant increase was not seen in the two-stage hepatectomy cohort; however, this should be interpreted with caution until larger studies with more patients can be conducted.
REFERENCES 1. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–825; discussion 825–827. 2. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005; 241:715–722; discussion 722–724. 3. Fernandez FG, Drebin JA, Linehan DC, Dehdashti F, Siegel BA, Strasberg SM. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 2004; 240: 438–447; discussion 447–450.
Huang et al
’
JVIR
4. Abulkhir A, Limongelli P, Healey AJ, et al. Preoperative portal vein embolization for major liver resection: a meta-analysis. Ann Surg 2008; 247:49–57. 5. Jaeck D, Bachellier P, Nakano H, et al. One or two-stage hepatectomy combined with portal vein embolization for initially nonresectable colorectal liver metastases. Am J Surg 2003; 185:221–229. 6. Azoulay D, Castaing D, Smail A, et al. Resection of nonresectable liver metastases from colorectal cancer after percutaneous portal vein embolization. Ann Surg 2000; 231:480–486. 7. Kawasaki S, Makuuchi M, Kakazu T, et al. Resection for multiple metastatic liver tumors after portal embolization. Surgery 1994; 115: 674–677. 8. de Baere T, Roche A, Elias D, Lasser P, Lagrange C, Bousson V. Preoperative portal vein embolization for extension of hepatectomy indications. Hepatology 1996; 24:1386–1391. 9. Jaeck D, Oussoultzoglou E, Rosso E, Greget M, Weber JC, Bachellier P. A two-stage hepatectomy procedure combined with portal vein embolization to achieve curative resection for initially unresectable multiple and bilobar colorectal liver metastases. Ann Surg 2004; 240:1037–1049; discussion 1049–1051. 10. Adam R, Laurent A, Azoulay D, Castaing D, Bismuth H. Two-stage hepatectomy: a planned strategy to treat irresectable liver tumors. Ann Surg 2000; 232:777–785. 11. Kianmanesh R, Farges O, Abdalla EK, Sauvanet A, Ruszniewski P, Belghiti J. Right portal vein ligation: a new planned two-step all-surgical approach for complete resection of primary gastrointestinal tumors with multiple bilateral liver metastases. J Am Coll Surg 2003; 197: 164–170. 12. Brouquet A, Abdalla EK, Kopetz S, et al. High survival rate after twostage resection of advanced colorectal liver metastases: response-based selection and complete resection define outcome. J Clin Oncol 2011; 29: 1083–1090. 13. Shindoh J, Vauthey JN, Zimmitti G, et al. Analysis of the efficacy of portal vein embolization for patients with extensive liver malignancy and very low future liver remnant volume, including a comparison with the associating liver partition with portal vein ligation for staged hepatectomy approach. J Am Coll Surg 2013; 217:126–133; discussion 133–134. 14. Tanaka K, Shimada H, Matsuo K, Ueda M, Endo I, Togo S. Remnant liver regeneration after two-stage hepatectomy for multiple bilobar colorectal metastases. Eur J Surg Oncol 2007; 33:329–335. 15. Kishi Y, Madoff DC, Abdalla EK, et al. Is embolization of segment 4 portal veins before extended right hepatectomy justified? Surgery 2008; 144:744–751. 16. Nagino M, Kamiya J, Kanai M, et al. Right trisegment portal vein embolization for biliary tract carcinoma: technique and clinical utility. Surgery 2000; 127:155–160. 17. Capussotti L, Muratore A, Ferrero A, Anselmetti GC, Corgnier A, Regge D. Extension of right portal vein embolization to segment IV portal branches. Arch Surg 2005; 140:1100–1103. 18. Abdalla EK, Barnett CC, Doherty D, Curley SA, Vauthey JN. Extended hepatectomy in patients with hepatobiliary malignancies with and without preoperative portal vein embolization. Arch Surg 2002; 137: 675–680; discussion 680–681. 19. Azoulay D, Castaing D, Krissat J, et al. Percutaneous portal vein embolization increases the feasibility and safety of major liver resection for hepatocellular carcinoma in injured liver. Ann Surg 2000; 232: 665–672. 20. Kokudo N, Makuuchi M. Current role of portal vein embolization/hepatic artery chemoembolization. Surg Clin North Am 2004; 84:643–657. 21. Shirabe K, Shimada M, Gion T, et al. Postoperative liver failure after major hepatic resection for hepatocellular carcinoma in the modern era with special reference to remnant liver volume. J Am Coll Surg 1999; 188:304–309. 22. Vauthey JN, Abdalla EK, Doherty DA, et al. Body surface area and body weight predict total liver volume in Western adults. Liver Transpl 2002; 8: 233–240. 23. Madoff DC, Hicks ME, Vauthey JN, et al. Transhepatic portal vein embolization: anatomy, indications, and technical considerations. Radiographics 2002; 22:1063–1076. 24. Madoff DC, Hicks ME, Abdalla EK, Morris JS, Vauthey JN. Portal vein embolization with polyvinyl alcohol particles and coils in preparation for major liver resection for hepatobiliary malignancy: safety and effectiveness—study in 26 patients. Radiology 2003; 227:251–260. 25. Madoff DC, Abdalla EK, Gupta S, et al. Transhepatic ipsilateral right portal vein embolization extended to segment IV: improving hypertrophy
Volume 25
26.
27.
28.
29.
30.
’
Number 4
’
April
’
2014
and resection outcomes with spherical particles and coils. J Vasc Interv Radiol 2005; 16:215–225. Vauthey JN, Chaoui A, Do KA, et al. Standardized measurement of the future liver remnant prior to extended liver resection: methodology and clinical associations. Surgery 2000; 127:512–519. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009; 250:187–196. Aussilhou B, Dokmak S, Faivre S, Paradis V, Vilgrain V, Belghiti J. Preoperative liver hypertrophy induced by portal flow occlusion before major hepatic resection for colorectal metastases can be impaired by bevacizumab. Ann Surg Oncol 2009; 16:1553–1559. de Baere T, Teriitehau C, Deschamps F, et al. Predictive factors for hypertrophy of the future remnant liver after selective portal vein embolization. Ann Surg Oncol 2010; 17:2081–2089. Hoekstra LT, van Lienden KP, Doets A, Busch OR, Gouma DJ, van Gulik TM. Tumor progression after preoperative portal vein embolization. Ann Surg 2012; 256:812–817; discussion 817–818.
617
31. Kokudo N, Tada K, Seki M, et al. Proliferative activity of intrahepatic colorectal metastases after preoperative hemihepatic portal vein embolization. Hepatology 2001; 34:267–272. 32. Pamecha V, Levene A, Grillo F, Woodward N, Dhillon A, Davidson BR. Effect of portal vein embolisation on the growth rate of colorectal liver metastases. Br J Cancer 2009; 100:617–622. 33. Kanematsu T, Takenaka K, Furuta T, Ezaki T, Sugimachi K, Inokuchi K. Acute portal hypertension associated with liver resection: analysis of early postoperative death. Arch Surg 1985; 120:1303–1305. 34. Nagasue N, Yukaya H, Ogawa Y, Higashi T. Portal pressure following partial to extensive hepatic resection in patients with and without cirrhosis of the liver. Ann Chir Gynaecol 1983; 72:18–22. 35. Slade JH, Alattar ML, Fogelman DR, et al. Portal hypertension associated with oxaliplatin administration: clinical manifestations of hepatic sinusoidal injury. Clin Colorectal Cancer 2009; 8:225–230. 36. Rubbia-Brandt L, Audard V, Sartoretti P, et al. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann Oncol 2004; 15:460–466.
Efficacy and Safety of Portal Vein Embolization for Two-Stage Hepatectomy in Patients with Colorectal Liver Metastasis Steven Y. Huang, MD, Thomas A. Aloia, MD, Junichi Shindoh, MD, PhD, Joe Ensor, PhD, Colette M. Shaw, MD, Evelyne M. Loyer, MD, Jean-Nicolas Vauthey, MD, and Michael J. Wallace, MD
ABSTRACT Purpose: To examine the efficacy and safety of portal vein embolization (PVE) when used during two-stage hepatectomy for bilobar colorectal liver metastases (CLM). Materials and Methods: PVE was performed as an adjunct to two-stage hepatectomy in 56 patients with CLM. Absolute future liver remnant (FLR) volumes, standardized FLR ratios, degree of hypertrophy (DH), and complications were analyzed. Segment II and III volumes and DH were also measured separately. All volumetric measurements were compared with a cohort of 96 patients (n ¼ 37 right portal vein embolization [RPVE], n ¼ 59 right portal vein embolization extended to segment IV portal veins [RPVEþ4]) in whom PVE was performed before single-stage hepatectomy. Results: For patients who completed RPVE during two-stage hepatectomy (n ¼ 17 of 17), mean absolute FLR volume increased from 272.1 cm3 to 427.0 cm3 (P o .0001), mean standardized FLR ratio increased from 0.17 to 0.26 (P o .0001), and mean DH was 0.094. For patients who completed RPVEþ4 during two-stage hepatectomy (n ¼ 38 of 39), mean FLR volume increased from 288.7 cm3 to 424.8 cm3 (P o .0001), mean standardized FLR increased from 0.18 to 0.26 (P o .0001), and mean DH was 0.083. DH of the FLR was not significantly different between two-stage hepatectomy and single-stage hepatectomy. Complications after PVE occurred in five (8.9%) patients undergoing two-stage hepatectomy. Conclusions: PVE effectively and safely induced a significant DH in the FLR during two-stage hepatectomy in patients with CLM.
ABBREVIATIONS CLM = colorectal liver metastasis, DH = degree of hypertrophy, FLR = future liver remnant, PVE = portal vein embolization, RPVE = right portal vein embolization, RPVEþ4 = right portal vein embolization extended to segment IV portal veins, RV = resection volume, S2þ3 = segments II and III, S4 = segment IV, TLV = total liver volume
Surgical resection of metastatic disease is regarded as the most effective strategy for long-term survival in patients with colorectal liver metastasis (CLM), with reported From the Departments of Diagnostic Radiology (S.Y.H., E.M.L., M.J.W.), Surgical Oncology (T.A.A., J.S., J.-N.V.), and Biostatistics (J.E.), The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030; and Department of Radiology (C.M.S.), Jefferson University Hospitals, Philadelphia, Pennsylvania. Received May 16, 2013; final revision received October 8, 2013; accepted October 11, 2013. Address correspondence to S.Y.H.; E-mail: [email protected] From the SIR 2013 Annual Meeting. None of the authors have identified a conflict of interest. & SIR, 2014 J Vasc Interv Radiol 2014; 25:608–617 http://dx.doi.org/10.1016/j.jvir.2013.10.028
5-year survival rates of 58% (1–3). However, only 10%– 20% of patients with CLM are deemed surgical candidates (4). Two key reasons why CLM is considered unresectable are the risk of major morbidity and mortality related to an insufficient future liver remnant (FLR) (5) and the presence of bilobar hepatic metastases involving any portion of FLR. Preoperative portal vein embolization (PVE) was developed to induce hypertrophy in FLR to address the issue of postoperative hepatic insufficiency (6–8). Now that PVE has gained acceptance as a reliable tool to improve FLR adequacy, attention is focused on addressing disease within FLR by performing a two-stage hepatectomy with PVE performed between the two stages. Disease in FLR is resected in a limited single-stage hepatectomy, and then PVE is performed to induce hypertrophy of FLR to an
Volume 25
’
Number 4
’
April
’
2014
adequate volume. Finally, resection of the remaining metastatic disease is performed during the final stage (9). Two-stage hepatectomy is safe and effective in patients with bilobar CLM (9–12). Although hypertrophy of FLR does occur in response to PVE in the setting of two-stage hepatectomy (5,9,13,14), it is unclear whether the insult caused by first-stage liver resection impairs the magnitude of FLR hypertrophy induced by PVE. Prior studies have been limited by small numbers of patients undergoing two-stage hepatectomy and PVE (5,14), lack of an appropriate control cohort of patients undergoing PVE in the setting of single-stage hepatectomy (9,13), and differences in PVE technique limiting generalizability of results (5,14). A recent retrospective study from our institution analyzing the efficacy of PVE in patients with very low FLR volume found that adequate regeneration was observed in nearly all of these patients (96.5%) (13). Of 144 patients undergoing percutaneous right portal vein embolization extended to segment IV portal veins (RPVEþ4), 32 (22%) procedures were performed after a first-stage partial left hepatectomy as part of a two-stage hepatectomy approach. Our current work builds on this earlier analysis and is unique in many ways. First, our analysis includes an additional 23 patients who underwent PVE as an adjunct to two-stage hepatectomy (total n ¼ 55 patients; n ¼ 38 patients undergoing RPVEþ4, n ¼ 17 undergoing right portal vein embolization [RPVE]). Second, in our earlier work, all 144 patients (n ¼ 32 RPVEþ4 two-stage hepatectomy and n ¼ 112 RPVEþ4 single-stage hepatectomy) were grouped to calculate percentage of hypertrophy, making it difficult to determine if RPVEþ4 in the setting of two-stage hepatectomy is efficacious compared with a control cohort of patients who underwent RPVEþ4 in the setting of single-stage hepatectomy. Third, the focus of our previous work was to describe the operative morbidity and mortality after the second-stage operation of two-stage hepatectomy; specific morbidity data related to PVE was limited. The main purpose of our study was to examine the efficacy and safety of PVE on FLR hypertrophy when used in conjunction with two-stage hepatectomy in patients with CLM compared with a control cohort of patients in whom PVE was performed as an adjunct to single-stage hepatectomy. An additional issue that is controversial is the utility of performing RPVEþ4 (15). Prior studies of patients undergoing single-stage hepatectomy compared the changes in volume for segments II and III (S2þ3) after RPVE and RPVEþ4 (15–17), but results have differed with two groups finding an increase in S2þ3 volumes after RPVEþ4 (15,16) and one group finding no significant increase in S2þ3 volumes (17). The efficacy of RPVEþ4 in patients undergoing two-stage hepatectomy has yet to be described, and so a secondary purpose of our study was to evaluate the effect RPVEþ4 had on
609
S2þ3 hypertrophy for patients undergoing two-stage hepatectomy.
MATERIALS AND METHODS We retrospectively reviewed 168 consecutive patients with CLM who underwent PVE between May 1998 and January 2011 for this study, which was compliant with the Health Insurance Portability and Accountability Act and approved by our institutional review board. PVE was performed as a component of two-stage hepatectomy in 65 patients; the remaining 103 patients underwent PVE before single-stage hepatectomy and were used as a comparison cohort. Both the two-stage hepatectomy group and the single-stage hepatectomy group were divided into subgroups based on whether RPVEþ4 was performed. There were 16 patients excluded because of missing computed tomography (CT) volumetric data (n ¼ 9), staged PVE (ie, performing PVE in separate procedures; n ¼ 4), and percutaneous intraoperative radiofrequency ablation performed in lieu of surgical resection during first-stage hepatectomy (n ¼ 3). Our study population comprised the remaining 152 patients (Fig 1). Two-stage hepatectomy was performed in 56 patients (n ¼ 17 for RPVE and n ¼ 39 for RPVEþ4), and singlestage hepatectomy was performed in 96 patients (n ¼ 37 for RPVE and n ¼ 59 for RPVEþ4). No patients underwent transarterial chemoembolization or radioembolization, and no patients had evidence of cirrhosis. Data were retrospectively collected on age, sex, diabetes status, body mass index, and use or nonuse of neoadjuvant chemotherapy within 3 months of PVE. Clinical characteristics for the four subgroups of patients in this study (RPVE single-stage hepatectomy, RPVE twostage hepatectomy, RPVEþ4 single-stage hepatectomy, and RPVEþ4 two-stage hepatectomy) are shown in Table 1. During the study period, patients underwent PVE if the volume of FLR was r 20% of the standardized total liver volume (TLV) for normal liver (18), r 30% in patients with fibrosis or severe liver injury (19), and r 40% in patients with cirrhosis (20,21). Standardized TLV was calculated using a formula for body surface area in square meters: standardized TLV = 794.41 þ 1,267.28 body surface area (22). All patients generally underwent abdominal CT with volumetry 2–8 weeks after PVE. At our institution, PVE was performed via a transhepatic ipsilateral (ie, on the side of the liver being resected) approach, which has been described previously (23–25). We use a combination of trisacryl particles and coils to occlude the branches of the right portal vein with or without segment IV (S4) portal veins. Although we realize there are alternative occlusive agents, we prefer to use particles and coils because our technique results in
610
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
Huang et al
’
JVIR
Figure 1. Flow chart of patients included in analysis. RFA ¼ radiofrequency ablation. Table 1 . Patient Characteristics RPVE
Mean age (range) (y) Male sex (ratio) Presence of diabetes (ratio) Mean BMI (range) (kg/m2)
RPVEþ4
SSH (n ¼ 37)
TSH (n ¼ 17)
P Value
SSH (n ¼ 59)
TSH (n ¼ 39)
59.7 (37–81)
50.6 (35–64)
.0029
57.7 (37–78)
51.4 (33–68)
.0028
.2231 .9131
81.4% (48/59) 20.3% (12/59)
59.0% (23/39) 15.4% (6/39)
.0150 .5401
70.3% (26/37) 18.9% (7/37) 30.3 (29.4–41.2)
Use of neoadjuvant chemotherapy 91.9% (34/37) within 3 mo of PVE (ratio)
56.3 (9/17) 17.6% (3/17) 27.1 (20.2–35.9) 100.0% (17/17)
.0432 .2349
28.8 (19.8–44.5) 89.8% (53/59)
27.3 (17.8–42.8) 94.3% (37/39)
P Value
.1597 .3775
BMI ¼ body mass index, PVE ¼ portal vein embolization, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy.
very few patients in whom the FLR does not hypertrophy sufficiently to allow resection. In patients with very low FLR, only approximately 3.5% fail to develop hypertrophy sufficiently to allow definitive hepatectomy (13). Portal vein pressures were obtained via a multiside hole catheter before and after PVE. RPVEþ4 was performed when an extended right hepatectomy was planned on the basis of tumor location. Technical success of the procedure was defined as embolization of the intended segments with no residual flow to the liver after embolization detected by final portal venography.
Assessment of PVE Effectiveness CT scans before and after PVE were performed to calculate absolute FLR volumes, change in FLR after
PVE, standardized FLR (ratio of FLR to standardized TLV), and degree of hypertrophy (DH) (difference between FLR before PVE and FLR after PVE) using density threshold seeding (26). Two patients (one patient from the RPVEþ4 two-stage hepatectomy group and one patient from the RPVEþ4 single-stage hepatectomy group) were excluded from the volume analysis because of portal vein thrombus that developed in the left lateral segmental portal vein during PVE. In both cases, the thrombus required overnight local thrombolytic administration, and PVE was not completed. Excluding these two patients, volume analysis for the RPVEþ4 twostage hepatectomy was performed on 38 patients, and RPVEþ4 single-stage hepatectomy group was performed on 58 patients. For patients undergoing twostage hepatectomy, CT was performed after first-stage
Volume 25
’
Number 4
’
April
’
2014
hepatectomy to assess the liver remnant, remeasure FLR, and detect any complications from the first-stage hepatectomy that would exclude patients from PVE. However, 16 of 56 patients who underwent two-stage hepatectomy did not undergo CT imaging after the firststage resection. The resected volumes from the first-stage hepatectomy were obtained from our surgical pathology database and manually subtracted from FLR volumes generated from the baseline CT scan obtained before hepatectomy. Correlation between DH and standardized resection volume (RV) (ratio of RV to standardized TLV) was assessed for the patients undergoing two-stage hepatectomy to determine if the amount of liver resected during the first-stage liver resection affected FLR hypertrophy after PVE.
Assessment of Effectiveness of RPVEþ4 Liver volumetry results were reviewed, and the absolute volume and DH of S2þ3 and S4 before and after PVE were compared between the RPVEþ4 and RPVE groups for both two-stage hepatectomy and single-stage hepatectomy. For the RPVE two-stage hepatectomy group, S2þ3 volumes were calculated based on 15 patients, whereas S4 volumes were calculated based on 17 patients. This discrepancy was due to two patients in the RPVE two-stage hepatectomy group who underwent complete resection of S2þ3 during the first stage of twostage hepatectomy.
Adverse Events Complications were evaluated by reviewing the electronic medical record from immediately after PVE through surgical resection for all patients (n ¼ 152). If the patient was unable to undergo surgery, the medical record was evaluated for periprocedural complications (up to 30 days). Complications were graded according to the Clavien-Dindo classification of surgical complications (27). Reasons for failure to progress to curative hepatectomy were also tabulated.
Statistical Analysis Volumetric results were reported as means and ranges. Continuous data were compared with a t test (pooled or Satterthwaite, as appropriate), and categorical data were compared with a 2-tailed Fisher exact test. Logistic regression models were used to evaluate the correlation between DH with resected liver volumes during the first stage of a two-stage hepatectomy and standardized FLR volumes before PVE. Pearson product-moment correlation was used to calculate the correlation between S2þ3 and S4 volume hypertrophy after RPVEþ4. P o .05 was considered statistically significant. Statistical analysis was performed with JMP software (SAS Institute, Cary, North Carolina).
611
RESULTS PVE was successfully completed in 55 of 56 patients undergoing two-stage hepatectomy and 95 of 96 patients undergoing single-stage hepatectomy. PVE was performed at a mean of 38.6 days (range, 4–288 d) after first-stage hepatectomy. The average interval between PVE and follow-up CT scan was 36.4 days (range, 15– 151 d). CT and portal venography images of a representative patient who underwent two-stage hepatectomy are shown in Figure 2a–d. When we compared singlestage hepatectomy with two-stage hepatectomy for RPVE and RPVEþ4, there was a significant difference detected in age; patients undergoing two-stage hepatectomy tended to be younger than patients undergoing single-stage hepatectomy. There were significantly more male patients in the RPVEþ4 single-stage hepatectomy group than the RPVEþ4 two-stage hepatectomy group. Body mass index was significantly lower in the RPVE two-stage hepatectomy group compared with the RPVE single-stage hepatectomy group. No difference was detected in the prevalence of diabetes and use or nonuse of neoadjuvant chemotherapy within 3 months of PVE between the single-stage hepatectomy and two-stage hepatectomy groups.
Assessment of PVE Effectiveness Mean absolute FLR volumes and standardized FLR ratios increased for all four groups after PVE (Table 2). DH was comparable between the RPVE two-stage hepatectomy and RPVE single-stage hepatectomy subgroups, measuring 0.094 and 0.097, respectively (P ¼ .8387). Similarly, no significant difference was noted in DH between the RPVEþ4 two-stage hepatectomy and RPVEþ4 single-stage hepatectomy subgroups, measuring 0.083 and 0.095, respectively (P ¼ .1816). FLR volumes before PVE were adjusted to account for the first-stage liver resection, as described in the Materials and Methods section. The amount of liver resected during the first-stage operation for patients undergoing twostage hepatectomy relative to the TLV ranged from 0.0001–0.421. Logistic regression did not detect a correlation between the standardized RV after first-stage liver resection and DH (ρ ¼ 0.12, P ¼ .38).
Assessment of Effectiveness of RPVEþ4 For single-stage hepatectomy, RPVEþ4 resulted in a significant increase to S2þ3 DH. The DH measured 0.089 for RPVEþ4 and 0.068 for RPVE (P ¼ .0095). For two-stage hepatectomy, RPVEþ4 yielded S2þ3 DH of 0.078 compared with 0.059 for RPVE (P ¼ .1602) (Table 3).
Adverse Events Complications occurred in 8 of 96 (8%) patients who underwent single-stage hepatectomy and 5 of 56 (9%)
612
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
Huang et al
’
JVIR
Figure 2. Sequence of events during two-stage hepatectomy. (a) Axial contrast-enhanced CT image at the level of the left portal vein shows a metastasis in the peripheral right lobe and a subtle metastasis in the lateral segment of the left hepatic lobe (arrow). (b) Axial contrast-enhanced CT image at the level of the left portal vein shows changes after surgery in the lateral segment of the left hepatic lobe (arrow). Metastasis in the peripheral right hepatic lobe is unchanged. During stage 1 hepatectomy, 50 cm3 of segment III was resected. FLR is depicted by the white dotted line. (c) Axial contrast-enhanced CT image through the level of the left portal vein performed 29 days after PVE shows interval growth of FLR (white dotted line), which includes segments I, II, and III. Standardized FLR increased from 0.155 to 0.274, yielding a DH of 0.119. (d) Axial contrast-enhanced CT image at the level of the left portal vein performed 203 days after second-stage hepatectomy shows hypertrophy of FLR.
Table 2 . Future Liver Remnant and Standardized Future Liver Remnant Hypertrophy after Right Portal Vein Embolization and Right Portal Vein Embolization Extended to Segment IV Portal Veins RPVE
RPVEþ4
Metric
SSH (n ¼ 37)
TSH (n ¼ 17)
P Value
SSH (n ¼ 58)
TSH (n ¼ 38)
P Value
FLRpre-PVE (range) (mL) FLRpost-PVE (range) (mL)
434.4 (122.2–1146.2) 612.4 (216.8–1351.9)
272.1 (55.1–524.4) 427.0 (148.1–853.4)
.0078 .0139
298.5 (145.2–564.3) 470.1 (277.2–682.4)
288.7 (78.6–526.3) 424.8 (85.0–888.5)
.6311 .1016
FLRchange (range) (mL)
177.9 (29.1–501.9)
154.9 (27.7–464.0)
.4661
171.5 (23.6–357.9)
136.1 (6.4–443.3)
.0372
0.24 (0.06–0.55) 0.34 (0.10–0.65)
0.17 (0.03–0.34) 0.26 (0.08–0.47)
.0256 .0448
0.17 (0.07–0.36) 0.26 (0.14–0.43)
0.18 (0.06–0.42) 0.26 (0.06–0.50)
.2765 .9182
0.097 (0.02–0.24)
0.094 (0.02–0.26)
.8387
0.095 (0.01–0.19)
0.083 (0.01–0.23)
.1816
sFLRpre-PVE sFLRpost-PVE DH
DH ¼ degree of hypertrophy (sFLRpostPVE sFLRprePVE), FLR ¼ future liver remnant, FLRchange ¼ change in future liver remnant, FLRpostPVE ¼ future liver remnant after portal vein embolization, FLRprePVE ¼ future liver remnant before portal vein embolization, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, sFLR ¼ standardized future liver remnant (FLR/standardized total liver volume), sFLRpostPVE ¼ standardized future liver remnant after portal vein embolization, sFLRprePVE ¼ standardized future liver remnant before portal vein embolization, SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy.
Volume 25
’
Number 4
’
April
’
2014
613
Table 3 . Changes in Segments II and III and Segment IV Volumes before and after Right Portal Vein Embolization and Right Portal Vein Embolization Extended to Segment IV Portal Veins SSH P Value
RPVE (n ¼ 15)
252.2 (83.2–555.6)
266.7 (65.3–506.6)
.4667
173.9 (31.1–292.5)
261.1 (111.9–478.7)
.0039
Volumepost-PVE (range) (mL)
372.7 (142.7–723.9)
426.8 (118.5–637.2)
.0325
273.3 (87.8–499.4)
385.4 (118.5–836.3)
.0101
Volumechange
120.5 (28.0–282.2)
160.1 (18.5–336.2)
.0082
99.4 (10.9–227.6)
127.8 (2.6–417.6)
.2408
(range) (mL) DH
0.068 (0.016–0.137)
0.089 (0.009–0.183)
.0095
0.059 (0.007–0.119)
0.078 (0.001–0.213)
.1602
S4 Volumepre-PVE
RPVE (n ¼ 37) 221.1 (97.9–527.8)
RPVEþ4 (n ¼ 58) 220.6 (34.2–553.0)
P Value .9808
RPVE (n ¼ 17) 194.0 (87.7–320.3)
RPVEþ4 (n ¼ 37) 225.5 (95.1–442.4)
P Value .1574
290.7 (154.2–544.2)
265.0 (70.1–453.4)
.2070
295.5 (162.0–496.7)
296.7 (57.5–494.4)
.9694
S2þ3 Volumepre-PVE
RPVE (n ¼ 37)
TSH
RPVEþ4 (n ¼ 58)
RPVEþ4 (n ¼ 38)
P Value
(range) (mL)
(range) (mL) Volumepost-PVE (range) (mL) Volumechange (range) (mL) DH
69.6 (8.3–254.6)
44.5 (202.1–175.2)
.0717
101.5 (22.0–261.5)
71.2 (46.4–216.7)
.1295
0.037 (0.006–0.133)
0.025 (0.079–0.103)
.0937
0.060 (0.016–0.145)
0.045 (0.029–0.140)
.2035
DH ¼ degree of hypertrophy, RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ single-stage hepatectomy, S2þ3 ¼ segments II and III, S4 ¼ segment IV, TSH ¼ two-stage hepatectomy, volumechange ¼ volume change, volumepost-PVE ¼ volume after portal vein embolization, volumepre-PVE ¼ volume before portal vein embolization.
Table 4 . Surgical Complications according to the Clavien-Dindo Classification System RPVE SSH (n ¼ 37)
RPVEþ4 SSH (n ¼ 59)
RPVE TSH (n ¼ 17)
I II
1a 0
3b,c,d 3e,f,g
0 0
3i,j,k 1l
IIIa
0
1h
0
1m
IIIb IVa
0 0
0 0
0 0
0 0
IVb
0
0
0
0
V
0
0
0
0
Surgical Complication Grade
RPVEþ4 TSH (n ¼ 39)
RPVE ¼ right portal vein embolization, RPVEþ4 ¼ right portal vein embolization extended to segment IV portal veins, SSH ¼ singlestage hepatectomy, TSH ¼ two-stage hepatectomy. a Fever requiring antibiotics. b Right upper quadrant pain requiring narcotics. c Fever requiring antipyretic. d Nontarget embolization (coil misplaced in segment III portal vein branch). e Main portal vein thrombus requiring anticoagulation. f Main portal vein and superior mesenteric vein thrombus requiring anticoagulation. g Subcapsular hematoma requiring blood transfusion. h Left portal vein thrombosis during RPVEþ4 requiring catheter-directed thrombolysis. i Right upper quadrant pain requiring narcotics. j Right upper quadrant pain requiring narcotics. k Nontarget embolization (coil misplaced at segment III portal vein origin). l Nonocclusive left portal vein thrombus requiring anticoagulation for 3 months. m Left portal vein thrombosis during RPVEþ4 requiring catheter-directed thrombolysis.
patients who underwent two-stage hepatectomy (P ¼ .90) (Table 4). For the two-stage hepatectomy group, there were five surgical complications; only one was greater than grade II in severity. This complication arose from acute thrombosis of the left lateral segmental portal vein during RPVEþ4, which required catheter-directed thrombolysis for 24 hours. Follow-up portal venography
showed persistent thrombus, and the patient was started on anticoagulation therapy. A CT scan performed 1 month later showed disease progression, and the patient was no longer a surgical candidate. For the single-stage hepatectomy group, only one of eight complications measured greater than grade II in severity. For this patient, RPVEþ4 was complicated by
614
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
partial thrombosis in the left lateral segmental portal vein, requiring catheter-directed thrombolysis for 24 hours and 3 months of anticoagulation with subsequent resolution of the thrombus. Despite the complication, the patient underwent definitive hepatectomy, although the patient’s original extended right hepatectomy was changed to a partial right hepatectomy extended to S4 with intraoperative radiofrequency ablation of residual metastatic lesions along the resection margin. Complications related to nontarget thrombus formation requiring thrombolysis or anticoagulation or both were unique to RPVEþ4. The incidence was 5 of 98 cases (5%), whereas there were no incidents of nontarget thrombus Table 5 . Factors Relating to Patient Ineligibility for Curative Resection SSH (n ¼ 96) TSH (n ¼ 56) P Value Disease progression Inadequate hypertrophy
13 3
10 5
.4894 .1451
Thrombocytopenia High-risk surgery*
1 4
1 0
1.0000 .2971
Development of new or
0
4
.0172
Chemotherapy-related liver toxicity
2
0
.2532
Steatohepatitis
1
0
1.0000
worsening imaging manifestations of portal hypertension
SSH ¼ single-stage hepatectomy, TSH ¼ two-stage hepatectomy. *Elevated body mass index or abdominal adhesions from prior surgery.
Huang et al
’
JVIR
formation when PVE was limited to the right portal vein (0 of 54 cases) (P ¼ .1613). Curative hepatectomy was performed in 72 of 96 (75%) patients who underwent single-stage hepatectomy and 36 of 56 (64%) patients who underwent two-stage hepatectomy (P ¼ .1621). Reasons patients were unable to undergo definitive resection are listed in Table 5. Of 56 patients undergoing two-stage hepatectomy, 4 (7%) were unable to undergo definitive hepatectomy because of new or worsening imaging findings of portal hypertension; none of the 96 patients undergoing single-stage hepatectomy developed this finding (P ¼ .0077). New or worsening imaging manifestations of portal hypertension included new gastroesophageal varices or new or worsening splenomegaly on follow-up CT scan performed approximately 1 month after PVE (Fig 3a, b). The portal venous pressures before and after PVE were available for three of these four patients. Portal venous pressures before PVE measured 5 mm Hg, 5 mm Hg, and 18 mm Hg. Portal venous pressures after PVE measured 13 mm Hg, 13 mm Hg, and 22 mm Hg. In comparison, for patients undergoing two-stage hepatectomy who underwent definitive hepatectomy, portal venous pressures before PVE were available for 29 of 36 (80.6%) patients (mean, 10.1 mm Hg; range, 1–18 mm Hg), and portal venous pressures after PVE were available for 28 of 36 (77.8%) patients (mean, 15.5 mm Hg; range, 3–29 mm Hg). For the four patients with new or worsening imaging manifestations of portal hypertension, standardized RV after first-stage liver surgery was 0.103, 0.117, 0.218, and 0.211. Standardized RV for 36 patients undergoing two-stage hepatectomy who underwent definitive hepatectomy without developing
Figure 3. New gastroesophageal varices after first-stage hepatectomy and RPVEþ4. (a) Axial contrast-enhanced CT image at the gastroesophageal junction 21 days before RPVE shows no definite varices (arrow). (b) Axial contrast-enhanced CT image at the gastroesophageal junction 28 days after RPVE shows varices (arrow).
Volume 25
’
Number 4
’
April
’
2014
new or worsening imaging manifestations of portal hypertension was 0.048 (range, 0.005–0.420). After definitive hepatectomy, the postoperative 90-day mortality rate was 4% for the single-stage hepatectomy group (3 of 72 patients) and 6% for the two-stage hepatectomy group (2 of 36 patients). All three of the deaths in the single-stage hepatectomy group and one of the deaths in the two-stage hepatectomy group were related to progressive postoperative liver failure and occurred 10 days, 28 days, and 40 days after singlestage hepatectomy and 18 days after two-stage hepatectomy. The second death in the two-stage hepatectomy group was due to cardiac arrest 5 days after two-stage hepatectomy.
DISCUSSION As an adjunct to two-stage hepatectomy, PVE is an effective and safe technique to induce FLR hypertrophy. The increases in absolute FLR volume, standardized FLR, and DH after PVE for two-stage hepatectomy were similar to the increases witnessed with PVE before single-stage hepatectomy. Complication rates with PVE in the setting of two-stage hepatectomy and single-stage hepatectomy were also similar. PVE is an effective means to induce hypertrophy of FLR in the setting of two-stage hepatectomy and does not appear to be negatively affected by recent liver resection. Previous studies have documented the increase in absolute FLR volume after PVE during two-stage hepatectomy (5,9,14). In our series, FLR volume increased from 272.1 mL to 427.0 mL for RPVE (n = 17) and 288.7 mL to 424.8 mL for RPVEþ4 (n = 38); this is comparable to FLR hypertrophy described by Jaeck et al (429 mL to 564 mL, n = 33) (9) and Tanaka et al (243.8 mL to 350.5 mL, n ¼ 16) (14). These results reinforce the utility of PVE in the setting of two-stage hepatectomy. Our study was notable for two reasons. First, our PVE technique involved a transhepatic ipsilateral approach with particles and coils, as opposed to the technique used by Jaeck et al (9), which involved a contralateral approach without describing the occlusive agent, and Tanaka et al (14), which involved cannulation of the ileocolic vein during surgery followed by administration of gelatin pellets and oleic acid. Second, we evaluated liver volume changes after RPVE and RPVEþ4 separately, and we were better able to determine the effect of two-stage hepatectomy on each group using the single-stage hepatectomy cohorts for comparison. As previously described, RPVEþ4 yielded a significant increase in S2þ3 volumes for patients undergoing single-stage hepatectomy (15). Our results confirmed this finding for the single-stage hepatectomy cohort. For the two-stage hepatectomy cohort, we did not detect a significant increase in S2þ3 volumes when RPVEþ4 was
615
performed (P ¼ .1602). The limited number of patients in our RPVE two-stage hepatectomy subgroup (n ¼ 15) may have hindered our ability to detect a significant increase in S2þ3 volumes after RPVEþ4. The effect of extended chemotherapy may also play a role because patients undergoing two-stage hepatectomy are generally treated with extensive systemic chemotherapy before surgery, which has been shown to impair hypertrophy of FLR after PVE (28,29). In our analysis, we evaluated whether or not chemotherapy had been administered during the 3 months before PVE. The use of perioperative chemotherapy seems to be ubiquitous in our patient population. Future analysis may consider comparing the duration of chemotherapy for patients undergoing two-stage hepatectomy versus single-stage hepatectomy to determine if length of treatment can affect FLR hypertrophy. Part of the technically challenging nature of embolization of S4 portal veins is obtaining complete embolization of S4 portal branches without causing reflux and showering embolic particles into S2þ3. It may be possible that during S4 embolization, the operators either stopped short of complete stasis or did not perform embolization of all branches, blunting the increased S2þ3 hypertrophy seen with RPVEþ4 in the single-stage hepatectomy cohort. After all, despite embolization extended to S4 portal veins, the DH of S4 still generally increased on follow-up CT scans for single-stage hepatectomy (mean, 0.025) and twostage hepatectomy (mean, 0.045) (Table 3). Finally, and most importantly, the decision to perform RPVEþ4 is made based on tumor location. Several studies have described the possibility of accelerated tumor growth in portions of the liver that did not receive embolization (30–32). At our institution, if metastases are situated in the right liver and S4, RPVEþ4 is generally performed to prevent tumor growth, regardless of baseline S2þ3 volumes. Complication rates after PVE in the setting of twostage hepatectomy and single-stage hepatectomy were similar. Complications were generally grade I and II according to the Clavien-Dindo classification system and did not alter the surgical plan. One particular complication, nontarget portal vein thrombosis, occurred only in the setting of RPVEþ4. Although this association was not statistically significant (P ¼ .1613), analysis was limited by the relative infrequency of this complication (n ¼ 5). Nontarget thrombosis of portal veins may potentially be attributed to the challenging nature of PVE performed in S4, which requires longer procedure times with indwelling catheters that alter portal vein hemodynamics. The percentage of patients who ultimately underwent definitive surgical resection after PVE was similar for the two-stage hepatectomy (64%) and single-stage hepatectomy (75%) groups. A small subset of patients in the two-stage hepatectomy group (n ¼ 4) did not undergo definitive hepatectomy because of new or worsening
616
’
Efficacy and Safety of PVE for Two-Stage Hepatectomy
radiographic signs of portal hypertension. Previous reports have described increases in portal pressure after partial hepatectomy involving 4 30% of the liver (33,34). Although RVs during the first-stage hepatectomy of our study were usually o 30%, it may stand to reason that larger volumes of liver resected during the first stage of two-stage hepatectomy, in combination with PVE, may predispose patients to new or worsening portal hypertension. In our study, standardized RV (ratio of RV to TLV) for the patients who developed new or worsening portal hypertension was 0.103, 0.117, 0.218, and 0.211, as opposed to a mean of 0.048 for the 36 patients who underwent two-stage hepatectomy, who completed second-stage hepatectomy without developing new or worsening imaging manifestations of portal hypertension. In addition, the use of preoperative chemotherapy in our two-stage hepatectomy population (n ¼ 54 of 56; 96%) may play a role. Specifically, oxaliplatin-based chemotherapy is known to cause sinusoidal injury and contribute to the development of noncirrhotic portal hypertension (35,36). This study has several limitations. The study design was retrospective in nature, which has inherent limitations. In addition, the number of patients in this study was small, limiting statistical analysis. Our reliance on medical chart review may not have captured all of the medical decision making or complications. Portal pressures before and after PVE were available for 80.6% and 77.8%, respectively, of embolization procedures performed as an adjunct to two-stage hepatectomy, limiting full evaluation of the influence of immediate changes in portal pressure on the development of portal hypertension. In conclusion, two-stage hepatectomy is an important treatment modality for patients with multiple bilobar CLMs. In these patients, PVE is an effective, safe, and reliable technique to induce hypertrophy of standardized FLR with results similar to patients undergoing PVE before single-stage hepatectomy. RPVEþ4 increased S2þ3 volumes before single-stage hepatectomy. A significant increase was not seen in the two-stage hepatectomy cohort; however, this should be interpreted with caution until larger studies with more patients can be conducted.
REFERENCES 1. Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239: 818–825; discussion 825–827. 2. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005; 241:715–722; discussion 722–724. 3. Fernandez FG, Drebin JA, Linehan DC, Dehdashti F, Siegel BA, Strasberg SM. Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 2004; 240: 438–447; discussion 447–450.
Huang et al
’
JVIR
4. Abulkhir A, Limongelli P, Healey AJ, et al. Preoperative portal vein embolization for major liver resection: a meta-analysis. Ann Surg 2008; 247:49–57. 5. Jaeck D, Bachellier P, Nakano H, et al. One or two-stage hepatectomy combined with portal vein embolization for initially nonresectable colorectal liver metastases. Am J Surg 2003; 185:221–229. 6. Azoulay D, Castaing D, Smail A, et al. Resection of nonresectable liver metastases from colorectal cancer after percutaneous portal vein embolization. Ann Surg 2000; 231:480–486. 7. Kawasaki S, Makuuchi M, Kakazu T, et al. Resection for multiple metastatic liver tumors after portal embolization. Surgery 1994; 115: 674–677. 8. de Baere T, Roche A, Elias D, Lasser P, Lagrange C, Bousson V. Preoperative portal vein embolization for extension of hepatectomy indications. Hepatology 1996; 24:1386–1391. 9. Jaeck D, Oussoultzoglou E, Rosso E, Greget M, Weber JC, Bachellier P. A two-stage hepatectomy procedure combined with portal vein embolization to achieve curative resection for initially unresectable multiple and bilobar colorectal liver metastases. Ann Surg 2004; 240:1037–1049; discussion 1049–1051. 10. Adam R, Laurent A, Azoulay D, Castaing D, Bismuth H. Two-stage hepatectomy: a planned strategy to treat irresectable liver tumors. Ann Surg 2000; 232:777–785. 11. Kianmanesh R, Farges O, Abdalla EK, Sauvanet A, Ruszniewski P, Belghiti J. Right portal vein ligation: a new planned two-step all-surgical approach for complete resection of primary gastrointestinal tumors with multiple bilateral liver metastases. J Am Coll Surg 2003; 197: 164–170. 12. Brouquet A, Abdalla EK, Kopetz S, et al. High survival rate after twostage resection of advanced colorectal liver metastases: response-based selection and complete resection define outcome. J Clin Oncol 2011; 29: 1083–1090. 13. Shindoh J, Vauthey JN, Zimmitti G, et al. Analysis of the efficacy of portal vein embolization for patients with extensive liver malignancy and very low future liver remnant volume, including a comparison with the associating liver partition with portal vein ligation for staged hepatectomy approach. J Am Coll Surg 2013; 217:126–133; discussion 133–134. 14. Tanaka K, Shimada H, Matsuo K, Ueda M, Endo I, Togo S. Remnant liver regeneration after two-stage hepatectomy for multiple bilobar colorectal metastases. Eur J Surg Oncol 2007; 33:329–335. 15. Kishi Y, Madoff DC, Abdalla EK, et al. Is embolization of segment 4 portal veins before extended right hepatectomy justified? Surgery 2008; 144:744–751. 16. Nagino M, Kamiya J, Kanai M, et al. Right trisegment portal vein embolization for biliary tract carcinoma: technique and clinical utility. Surgery 2000; 127:155–160. 17. Capussotti L, Muratore A, Ferrero A, Anselmetti GC, Corgnier A, Regge D. Extension of right portal vein embolization to segment IV portal branches. Arch Surg 2005; 140:1100–1103. 18. Abdalla EK, Barnett CC, Doherty D, Curley SA, Vauthey JN. Extended hepatectomy in patients with hepatobiliary malignancies with and without preoperative portal vein embolization. Arch Surg 2002; 137: 675–680; discussion 680–681. 19. Azoulay D, Castaing D, Krissat J, et al. Percutaneous portal vein embolization increases the feasibility and safety of major liver resection for hepatocellular carcinoma in injured liver. Ann Surg 2000; 232: 665–672. 20. Kokudo N, Makuuchi M. Current role of portal vein embolization/hepatic artery chemoembolization. Surg Clin North Am 2004; 84:643–657. 21. Shirabe K, Shimada M, Gion T, et al. Postoperative liver failure after major hepatic resection for hepatocellular carcinoma in the modern era with special reference to remnant liver volume. J Am Coll Surg 1999; 188:304–309. 22. Vauthey JN, Abdalla EK, Doherty DA, et al. Body surface area and body weight predict total liver volume in Western adults. Liver Transpl 2002; 8: 233–240. 23. Madoff DC, Hicks ME, Vauthey JN, et al. Transhepatic portal vein embolization: anatomy, indications, and technical considerations. Radiographics 2002; 22:1063–1076. 24. Madoff DC, Hicks ME, Abdalla EK, Morris JS, Vauthey JN. Portal vein embolization with polyvinyl alcohol particles and coils in preparation for major liver resection for hepatobiliary malignancy: safety and effectiveness—study in 26 patients. Radiology 2003; 227:251–260. 25. Madoff DC, Abdalla EK, Gupta S, et al. Transhepatic ipsilateral right portal vein embolization extended to segment IV: improving hypertrophy
Volume 25
26.
27.
28.
29.
30.
’
Number 4
’
April
’
2014
and resection outcomes with spherical particles and coils. J Vasc Interv Radiol 2005; 16:215–225. Vauthey JN, Chaoui A, Do KA, et al. Standardized measurement of the future liver remnant prior to extended liver resection: methodology and clinical associations. Surgery 2000; 127:512–519. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009; 250:187–196. Aussilhou B, Dokmak S, Faivre S, Paradis V, Vilgrain V, Belghiti J. Preoperative liver hypertrophy induced by portal flow occlusion before major hepatic resection for colorectal metastases can be impaired by bevacizumab. Ann Surg Oncol 2009; 16:1553–1559. de Baere T, Teriitehau C, Deschamps F, et al. Predictive factors for hypertrophy of the future remnant liver after selective portal vein embolization. Ann Surg Oncol 2010; 17:2081–2089. Hoekstra LT, van Lienden KP, Doets A, Busch OR, Gouma DJ, van Gulik TM. Tumor progression after preoperative portal vein embolization. Ann Surg 2012; 256:812–817; discussion 817–818.
617
31. Kokudo N, Tada K, Seki M, et al. Proliferative activity of intrahepatic colorectal metastases after preoperative hemihepatic portal vein embolization. Hepatology 2001; 34:267–272. 32. Pamecha V, Levene A, Grillo F, Woodward N, Dhillon A, Davidson BR. Effect of portal vein embolisation on the growth rate of colorectal liver metastases. Br J Cancer 2009; 100:617–622. 33. Kanematsu T, Takenaka K, Furuta T, Ezaki T, Sugimachi K, Inokuchi K. Acute portal hypertension associated with liver resection: analysis of early postoperative death. Arch Surg 1985; 120:1303–1305. 34. Nagasue N, Yukaya H, Ogawa Y, Higashi T. Portal pressure following partial to extensive hepatic resection in patients with and without cirrhosis of the liver. Ann Chir Gynaecol 1983; 72:18–22. 35. Slade JH, Alattar ML, Fogelman DR, et al. Portal hypertension associated with oxaliplatin administration: clinical manifestations of hepatic sinusoidal injury. Clin Colorectal Cancer 2009; 8:225–230. 36. Rubbia-Brandt L, Audard V, Sartoretti P, et al. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann Oncol 2004; 15:460–466.