World J Surg DOI 10.1007/s00268-015-3110-9
ORIGINAL SCIENTIFIC REPORT
Surgical Resection for Hepatocellular Carcinoma with Concomitant Esophageal Varices Noboru Harada1 • Ken Shirabe2 • Takashi Maeda1 • Hiroto Kayashima1 Teruyoshi Ishida1 • Yoshihiko Maehara2
•
Ó Socie´te´ Internationale de Chirurgie 2015
Abstract Background The management of hepatocellular carcinoma (HCC) in patients with concomitant esophageal varices (EV) remains controversial. We assessed the surgical outcome of hepatectomy and aimed to clarify the indications and management of HCC in patients with concomitant EV. Methods We retrospectively enrolled 502 patients with HCC (100 with and 402 without EV), who underwent curative hepatectomy. We analyzed the prognostic outcomes and risk factors for EV bleeding after hepatectomy. Results Overall survival (OS) was significantly lower in HCC patients with EV than in those without EV (p = 0.003), although recurrence-free survival was similar in both groups. Multivariate analysis showed that indocyanine green retention test at 15 min (ICGR15)[17 % (p = 0.007) and a-fetoprotein [12.5 ng/ml (p = 0.003) was independent predictors of poorer OS. Among patients with EV who underwent hepatectomy, multivariate analysis identified ICGR15 [17 % (p = 0.03) as the only independent predictor of poorer OS. There was no significant difference in OS between HCC patients with EV and ICGR15 B17.0 % and HCC patients without EV. Ten patients experienced EV bleeding after hepatectomy. Multivariate analysis showed that preoperative endoscopic findings of blue color EV (p = 0.008) and red color sign (p = 0.0005) were independent predictors of EV bleeding in patients with HCC after hepatectomy. Conclusions These results suggest that HCC patients with EV and ICGR15 B17 % may be suitable for surgery, but patients with preoperative endoscopic blue color EV and red color sign need to be managed appropriately.
Introduction Many patients with hepatocellular carcinoma (HCC) have concomitant esophageal varices (EVs) which are considered a clinical manifestation of portal hypertension (PHT) [1].
& Noboru Harada [email protected] 1
Department of Surgery, Hiroshima Red Cross Hospital and Atomic Bomb Survivors Hospital, Hiroshima 730-8619, Japan
2
Department of Surgery and Medical Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
Increased portal venous pressure after hepatic resection may cause variceal bleeding, especially in relation to Pringle’s maneuver [2] or major hepatectomy. EV bleeding is a medical emergency associated with a mortality of around 20 % [3]. According to the Barcelona Clinic Liver Cancer Group [4] and American Association for the Study of Liver Disease [5] guidelines, HCC with clinically significant PHT is a contraindication for hepatectomy. Clinically significant PHT [hepatic venous pressure gradient (HVPG) C10 mmHg] is the most powerful predictor of postoperative liver failure or poor long-term survival in patients with Child-Pugh A liver function [6]. Preoperative HVPG measurement is invasive and not performed routinely in most liver centers, and indirect clinical parameters (EV
123
World J Surg
and/or splenomegaly associated with thrombocytopenia) are therefore considered clinical signs of PHT. However, several reports have suggested that PHT should not necessarily be a contraindication for hepatic resection in cirrhotic patients [1, 7, 8]. Increased HVPG was associated with postoperative liver dysfunction and mortality after liver resection in a small number of studies including a maximum of 46 patients with HCC and liver cirrhosis [6, 9–12], suggesting that some patients demonstrating indirect criteria of PHT may still be eligible for surgery in the presence of adequate selection criteria [1, 7, 8]. The surgical indications for HCC-associated EV thus remain controversial and the risk factors for EV bleeding after hepatectomy also remain unclear. This study aimed to assess the surgical outcomes of liver resection for HCC in patients with EV, and to clarify the indications and management of HCC patients with concomitant EV.
postoperative follow-up were not performed routinely in our hospital. However, five patients with apparent risky EV (F2 or F3, and RC positive) underwent endoscopic injection sclerotherapy (EIS) or endoscopic variceal ligation (EVL). Serious postoperative complications were defined as Clavien–Dindo grade III or higher [17]. Postoperative liver failure was defined according to the guidelines of the International Study Group of Liver Surgery [18]. Survival and recurrence
Methods
Patients underwent blood tests and computed tomography every 3 months after hepatic resection. Recurrence was diagnosed based on imaging findings. Patients with intrahepatic recurrence were managed with ablative therapy, such as radiofrequency ablation and percutaneous ethanol injection, transcatheter arterial chemoembolization, or surgery. In the event of death, survival time after surgery and cause of death were recorded. Postoperative survival time was also recorded for all patients.
Patients and surgical procedures
Statistical analysis
A total of 502 consecutive patients with HCC who underwent curative hepatectomy at the Department of Surgery, Hiroshima Red Cross Hospital and Atomic Bomb Survivors Hospital between September 2004 and October 2013, were included in the study. All patients had a confirmed pathological diagnosis of HCC and were diagnosed preoperatively with EV by upper-gastrointestinal endoscopy about 1 month before surgery. The type of hepatectomy was based on liver function and tumor extension [13]. Liver function was assessed by Child-Pugh classification [14] and indocyanine green retention test at 15 min (ICGR15). The hepatectomy procedures have been described elsewhere [13, 15]. Patients with HCC were divided into those with and without EV. Written informed consent was obtained from all patients for surgical treatment, according to institutional guidelines. The study protocol conformed to the updated ethical guidelines of the 2013 Declaration of Helsinki and was approved by our Institutional Review Board.
Continuous variables were analyzed parametrically using Student’s t tests. Wilcoxon’s rank-sum test was used for nonparametric analyses. Categorical variables were compared using v2 or Fisher’s exact tests. Overall survival (OS) and recurrence-free survival (RFS) rates were calculated by the Kaplan–Meier method and compared by univariate logrank test. Independent factors for OS were identified by multivariate analysis of factors significant in univariate analysis using a Cox proportional hazards model [19]. Cutoff values for continuous variables were calculated as the median values of the whole study population. Significant risk factors for EV bleeding were identified by multivariate logistic regression analysis of risk factors significant in univariate analysis. Statistical significance was set at p \ 0.05. Statistical analyses were performed using JMP 9.0 software (SAS Institute, Cary, NC, USA).
Results Definitions Curative resection was defined as complete macroscopic and microscopic tumor removal. Liver cirrhosis was confirmed by histological examination of a resected specimen. Major hepatectomy was defined as resection of at least two subsegments, and minor resection, including partial resection, involved fewer than two subsegments. EV was classified preoperatively based on endoscopic findings [16], according to location, form, color, and red color signs (RC). Preoperative prophylactic EV treatment and
123
Clinicopathological characteristics of HCC patients with and without EV who underwent hepatectomy Preoperative background characteristics of HCC patients with and without EV who underwent hepatectomy are summarized in Table 1. Compared with patients without EV, patients with EV included more women (41.0 vs. 30.8 %, p = 0.037), more patients with liver cirrhosis (90.0 vs. 55.1 %, p \ 0.0001), and had a lower mean platelet count (104,000 vs. 146,000/ll, p \ 0.0001), lower mean
World J Surg Table 1 Preoperative background characteristics of HCC patients with and without EV who underwent hepatectomy Variables
Without EV (n = 402)
With EV (n = 100)
p
Age (years), mean (range)
69.0 (34–87)
69.1 (43–85)
0.740
Gender (male/female) (n)
281/121
59/41
0.037
Anti-HCVAb positive (n) HBV surface antigen positive (n)
262 63
73 16
0.155 0.985
Liver cirrhosis (n)
221 (55.1 %)
90 (90.0 %)
\0.0001
Platelet (9104/ll), mean (range)
14.6 (2.4–53)
10.4 (3.3–24)
\0.0001
Albumin (g/dl), mean (range)
4.0 (2.5–5.1)
3.9 (2.9–4.7)
0.002
Total bilirubin (mg/dl), mean (range)
0.7 (0.2–1.7)
0.9 (0.2–2.3)
0.014 \0.0001
Prothrombin activity ( %), mean (range)
92.6 (57.3–130)
84.4 (48.6–117)
Child-Pugh class A/B (n)
393/8
89/11
0.0002
ICGR15, mean (range)
17.5 (0.9–73.7)
25.1 (5.6–68.7)
\0.0001
AFP (ng/ml), mean (range)
808 (0.49–93721)
144 (2.6–2340)
0.032
HCC hepatocellular carcinoma, EV esophageal varices, HCVAb hepatitis C virus antibody, HBV hepatitis B virus, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
serum albumin level (3.9 vs. 4.0 g/dl, p = 0.002), greater mean serum total bilirubin level (0.9 vs. 0.7 mg/dl, p = 0.014), lower mean prothrombin activity (84.4 vs. 92.6 %, p \ 0.0001), higher frequency of Child-Pugh class B (11.0 vs. 2.0 %, p = 0.0002), greater mean ICGR15 (25.1 vs. 17.5 %, p \ 0.0001), and lower mean a-fetoprotein (AFP) level (144 vs. 808 ng/ml, p = 0.032). Tumor and surgical characteristics of HCC patients with and without EV who underwent hepatectomy are summarized in Table 2. Patients with EV had a lower frequency of major hepatectomy (12.0 vs. 22.9 %, p = 0.019) and shorter mean Pringle’s maneuver time (41 vs. 53 min, p = 0.002) than patients without EV. One in-hospital death occurred in a patient with EV secondary to sepsis. Patients with EV had a significantly higher rate of serious postoperative complications (Clavien–Dindo grade III or higher) than those without EV (18.0 vs. 9.2 %; p = 0.020). Eighteen serious postoperative complications (within 30 postoperative days) were recorded in patients with EV, including wound dehiscence (n = 4), bile leakage (n = 3), refractory ascites (n = 2), grade B liver failure (n = 2), and ileus, EV bleeding, pancreas fistula, pneumothorax, pneumonia, jejunal perforation, and sepsis (all n = 1). Thirty-seven serious postoperative complications were recorded in patients without EV, including bile leakage (n = 11), refractory pleural effusion (n = 8), wound dehiscence (n = 8), refractory ascites (n = 4), and grade B liver failure, ileus, postoperative bleeding, pneumonia, wound pain, and sepsis (all n = 1). OS and RFS Median postoperative follow-up after surgery was 39.9 months (range, 0.4–104.2 months). OS was significantly
lower in HCC patients with EV than those without EV (p = 0.003). The 1-, 3-, and 5-year OS rates in patients with EV were 93.7, 61.9, and 44.9 %, respectively, compared with 94.3, 80.4, and 67.2 % in patients without EV (Fig. 1a). The 1-, 3-, and 5-year RFS rates did not differ significantly between patients with (76.6, 41.5, and 29.6 %) and without EV (73.7, 43.1, and 30.3 %) (p = 0.906). Associations between clinicopathological factors and OS were determined (Table 3). Univariate analyses showed that age [70 years (p = 0.034), ICGR15 [17 % (p \ 0.0001), liver cirrhosis (p = 0.004), AFP [25.8 ng/ ml (p \ 0.0001), Child-Pugh class B (p = 0.002), EV (p = 0.003), maximum tumor diameter [22 mm (p = 0.028), multiple tumors (p = 0.021), microscopic vascular invasion (p = 0.016), intraoperative blood transfusion (p = 0.001), and serious postoperative complications (p = 0.039) were significantly associated with poorer OS, whereas sex, histological grading, extent of hepatic resection, and intraoperative blood loss [320 ml were not. Multivariate Cox proportional hazards analysis showed that ICGR15 [17 % [p = 0.007; hazard ratio (HR) 1.67; 95 % confidence interval (CI) 1.15–2.45] and AFP [12.5 ng/ml (p = 0.003; HR 1.78; 95 % CI 1.22–2.62) were independent predictors of poorer OS. Factors affecting OS in HCC patients with EV Univariate analyses showed that ICGR15 [17 % (p = 0.009) and serum albumin B4.0 g/dl (p = 0.035) were significantly associated with poorer OS in HCC patients with EV who underwent hepatectomy. Multivariate Cox proportional hazards analysis identified ICGR15 [17 % (p = 0.030; HR 2.60; 95 % CI 1.09–7.27) as the
123
World J Surg Table 2 Tumoral and surgical characteristics of HCC patients with and without EV who underwent hepatectomy Variables
Without EV (n = 402)
With EV (n = 100)
p
Mean (range)
2.7 (0.5–13)
2.3 (0.7–6)
0.056
Number of tumors, mean (range) Histological grading
1.6 (1–10)
1.4 (1–5)
0.343
57/339
11/88
0.514
244
67
0.417
92/310
12/88
0.019
Maximum tumor diameter (cm)
Well/moderately or poorly (n) Microscopic vascular invasion Extent of hepatic resection Major/minor hepatectomy (n) Pringle’s maneuver time (min) Mean (range) Surgical time (min), mean (range)
53 (0–195)
41 (0–120)
0.002
242 (60–633)
222 (54–483)
0.087
Intraoperative blood loss (ml) Mean (range) Intraoperative blood transfusion (n)
507 (10–4978)
594 (5–4220)
0.444
25
12
0.055
37
18
0.020
Postoperative serious complications (n) Clavien–Dindo CIII HCC hepatocellular carcinoma, EV esophageal varices
only independent predictor of poorer OS (Table 4). OS was similar in patients without EV and patients with EV with ICGR15 B17 % (p = 0.506) (Fig. 1b). The 1-, 3-, and 5-year OS rates in HCC patients with EV and ICGR15 B17 % were 96.4, 78.8, and 72.3 %, respectively, compared with 94.3, 80.4, and 67.2 % in patients without EV. There was no significant difference in the incidence of serious postoperative complications (Clavien–Dindo grade III or higher) between patients with EV and ICGR15 B17 % and patients without EV (13.3 vs. 9.2 %; p = 0.512). Incidence of and risk factors for EV bleeding in patients with HCC after hepatectomy Among 100 HCC patients with EV who underwent hepatectomy, 10 experienced postoperative EV bleeding, and hemostasis was achieved in all patients. The duration from the day of operation to bleeding was 2–867 days (median 172 days). There were three (30 %) and six (60 %) bleeding events in the first 3 and 6 months after liver resection, respectively. Although five patients [two F2 blue color (Cb) RC?, one F2 white color (Cw) RC?, and two F3CbRC?] underwent prophylactic preoperative EIS for EV, two (F2CbRC? and F3CbRC?) still had EV bleeding after hepatectomy. Nine patients underwent postoperative prophylactic EIS for EV and one underwent prophylactic EVL, none of whom had subsequent bleeding. By univariate analyses, EV with form CF2 (p = 0.0006), Cb
123
(p = 0.003), and RC? (p \ 0.0001) were significantly correlated with EV bleeding. Multivariate logistic regression analysis showed that Cb (p = 0.008; odds ratio 15.1; 95 % CI 1.90–402) and RC? (p = 0.0005; odds ratio 55.6; 95 % CI 4.81–1884) were independent predictors of EV bleeding in patients with HCC after hepatectomy (Table 5).
Discussion The results of the current study demonstrated that OS rates were significantly lower in HCC patients with EV compared with those without EV. However, multivariate analysis suggested that EV had only a marginally significant effect on OS in HCC patients who underwent hepatectomy (p = 0.055), implying that similar survival outcomes could be achieved irrespective of the presence of EV. However, it is difficult to compare the outcomes of HCC patients with and without EV. Many hepatobiliary– pancreas surgeons are more cautious about the perioperative surgical management of HCC patients with EV than those without EV. HCC patients with EV are thus likely to have more advanced liver disease than patients without EV. However, we believe that surgery may be suitable in some HCC patients with EV, and new criteria are therefore needed to identify these patients. Kokudo recently proposed that liver resection could be an acceptable option for small HCCs in patients with PHT, provided ICG test results were normal or only slightly impaired, and that risky
World J Surg Fig. 1 Prognoses of 502 patients with and without EV who underwent liver resection for primary HCC. a Kaplan– Meier curves for OS after hepatectomy. OS in HCC patients with EV (bold black line) was significantly lower than in patients without EV (normal black line, p = 0.0032, log-rank test). b Kaplan–Meier curves for OS after hepatectomy among HCC patients without EV (normal black line), HCC patients with EV and ICGR15 B17 % (bold black line), and HCC patients with EV and ICGR15 [17 % (dotted black line). There was no significant difference in OS between HCC patients with EV and ICGR15 B17 % (bold black line) and HCC patients without EV (normal black line, p = 0.506, log-rank test). OS in HCC patients with EV and ICGR15 B17 % (bold black line) was significantly higher than in HCC patients with EV and ICGR15 [17 % (dotted black line, p \ 0.0001, log-rank test). OS in HCC patients without EV (normal black line) was significantly higher than in HCC patients with EV and ICGR15 [17 % (dotted black line, p = 0.009, log-rank test)
esophagogastric varices could be controlled in experienced tertiary centers [20]. The current study found no significant difference in OS rates between HCC patients without EV and those with EV and ICGR15 B17 %. Hepatectomy may therefore be indicated in HCC patients with concomitant EV and ICGR15 B17 %, because of the improved outcomes associated with surgical techniques and perioperative care [21]. We suggest that a cut-off value of ICGR15 B17 % provides a useful indication for surgery in HCC patients with EV. Given the serious shortage of available liver transplant donor organs
[22], surgical resection may represent a better option for the treatment of some HCC patients with concomitant EV. However, these patients should be selected carefully and their treatment should adhere to a strict surgical strategy, including EV management. Perioperative EV management is necessary to minimize the risk of bleeding. To the best of our knowledge, no previous studies have reported on the incidence of and risk factors for EV bleeding after hepatectomy in patients with resectable HCC. Miyoshi et al. reported 3-year bleeding rates of 50 % for the control group and 18 % for the
123
World J Surg Table 3 Univariate and multivariate analyses of overall survival in all patients Variables
Univariate analysis
Multivariate analysis
5-year survival
p value
HR (95 % CI)
p value
69.9/56.3
0.034
1.38 (0.97–1.97)
0.071
64.1/62.9
0.178
73.6/53.3
\0.0001
1.67 (1.15–2.45)
0.007
58.6/73.3
0.004
1.09 (0.71–1.71)
0.684
73.0/54.5
\0.0001
1.78 (1.22–2.62)
0.003
64.7/41.9
0.002
1.30 (0.63–2.45)
0.460
44.9/67.2
0.003
1.54 (0.99–2.34)
0.055
68.8/57.9
0.028
1.40 (0.99–1.98)
0.058
68.2/54.9
0.021
1.30 (0.90–1.87)
0.156
43.5/33.4
0.261
60.3/68.2
0.016
1.33 (0.93–1.93)
0.114
69.8/62.3
0.351
64.9/62.5
0.075
43.1/65.4
0.001
1.71 (0.98–2.84)
0.059
49.0/65.4
0.039
1.29 (0.75–2.10)
0.347
Age (years) B70/[70 Gender Male/female ICGR15 ( %) B17/[17 Liver cirrhosis Yes/no AFP (ng/ml) B12.5/[12.5 Child-Pugh Class A/B Esophageal varices Yes/no Maximum tumor diameter (mm) B22/[22 Multiple tumors Yes/no Histological grading Well/moderately or poorly Microscopic vascular invasion Yes/no Extent of hepatic resection Major/minor hepatectomy Intraoperative blood loss (ml) B320/[320 Intraoperative blood transfusion Transfused/nontransfused Postoperative serious complications Clavien–Dindo CIII (yes/no)
HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
prophylactic sclerotherapy group in patients with unresectable HCC and concomitant EV [23]. Akahoshi et al. reported a cumulative bleeding rate of 48.8 % at 6 months in patients with unresectable HCC and EVs that were likely to bleed, who were treated conservatively [24]. Despite some selection bias, the 3-year postoperative bleeding rate in the current study was 10 % in patients with resectable HCC and concomitant EV (n = 100). The median duration from day of operation to bleeding was \6 months. Preoperative endoscopy found that Cb and RC were strongly associated with EV bleeding in patients with HCC after hepatectomy. Cb and RC at preoperative endoscopy may thus be considered as indications for preoperative or postoperative EV treatment.
123
Most hepatobiliary surgeons aim to reduce the risk of EV bleeding by preoperative endoscopic treatment if upper-gastrointestinal endoscopy reveals moderately enlarged F2 varices with or without RC, or markedly enlarged nodular or tumor-shaped varices (F3) [7]. However, there have been no reports regarding the risk factors for EV bleeding after hepatectomy in patients with HCC. The current treatment strategy for EV is based on the finding that endoscopic sclerotherapy prevents EV bleeding in patients with PHT [24–26]. High-risk varices may bleed as a result of increased portal venous pressure after hepatic resection, associated with Pringle’s maneuver or major hepatectomy. In our study, EV-form F2/3 was not an independent risk factor for bleeding after hepatectomy in
World J Surg Table 4 Univariate and multivariate analyses of overall survival in HCC patients with EV who underwent hepatectomy
Table 5 Logistic regression analyses of risk factors for bleeding in HCC patients with EV who underwent hepatectomy
Variables
Variables
Univariate
Multivariate analysis
p value
HR (95 % CI)
p value
Univariate
Multivariate analysis
p value
Odds ratio (95 % CI)
p value
Age (years) B70/[70 Gender (male/female)
0.848
Age (years) [70
0.998
0.454
Gender (male/female)
0.184
Liver cirrhosis (yes/no)
0.164
ICGR15 ( %) [17
0.719
0.226
AFP (ng/ml) [12.6
0.735
Preoperative EV treatment (yes/no) Postoperative EV treatment (yes/no)
0.107
0.052
EV-location (Li/Ls, Lm)
0.085
EV-form (F2-3/F0-1)
0.0006
0.49 (0.02–5.91)
0.607
EV-color (blue/white)
0.003
15.1 (1.90–402)
0.008
EV-RC sign (presence/ absence)
\0.0001
55.6 (4.81–1884)
0.0005
ICGR15 (%) B17/[17
0.009
Liver cirrhosis (yes/no)
2.60 (1.09–7.27)
0.030
4
Platelet count (910 /ll) B12.6/[12.6 Total bilirubin (mg/dl) B0.7/[0.7
0.199
Serum albumin (g/dl) B4.0/[4.0
0.035
1.72 (0.79–4.20)
0.181
Prothronbin activity ( %) B91.4/[91.4
0.141
AFP (ng/ml) B12.5/[12.5
0.041
Child-Pugh class B/A
0.358 0.161 0.079
EV-location (Li/Ls, Lm)
0.390
Maximum tumor diameter [22 mm
EV-form (F2-3/F0-1)
0.238
Multiple tumors (yes/no)
EV-color (blue/white)
0.615
Histological grading
EV-RC sign (presence/ absence)
0.098
B22/[22
Microscopic vascular invasion (yes/no)
0.210
Extent of hepatic resection
0.847
Pringle’s maneuver time [45 min
0.836
0.321
Intraoperative blood loss [320 ml
0.103
Major/minor
0.147
Pringle’s maneuver time (min) B45/[45
0.678
Intraoperative blood loss (ml) B320/[320
0.312
0.400
Intraoperative blood transfusion
Extent of hepatic resection Major/minor
0.838
0.856
Histological grading Well/moderately or poorly differentiation
Well/moderately of poorly differentiation Microscopic vascular invasion (yes/no)
Maximum tumor diameter (mm) Multiple tumors (yes/no)
0.277
0.194
Intraoperative blood transfusion Transfused/nontransfused 0.349
Transfused/ nontransfused
0.103
Postoperative serious complications
0.838
Clavien–Dindo CIII
0.189
HCC hepatocellular carcinoma, EV esophageal varices, HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
Postoperative serious complications Clavien–Dindo CIII (yes/ no)
0.266
HCC hepatocellular carcinoma, EV esophageal varices, HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein, RC red color sign
patients with HCC, nor were surgical factors such as major hepatectomy, Pringle’s maneuver time, intraoperative blood loss, and blood transfusion associated with EV bleeding, according to univariate analysis. Further studies
are needed to clarify the association between EV bleeding and surgical factors. Several clinical studies have concluded that prophylactic therapy of EV is ineffective, and it is generally not accepted in western countries [25, 27, 28]. Although five patients in our study (two F2CbRC?, one F2CwRC?, and two F3CbRC?) underwent prophylactic EIS for EV before hepatectomy, two (F2CbRC? and F3CbRC?) still
123
World J Surg
experienced postoperative EV bleeding episodes. These results support the idea that prophylactic EIS may have minimal benefits. There is currently no evidence regarding the benefit of postoperative prophylactic EIS or EVL in HCC patients with concomitant EV. We found no bleeding in patients who underwent postoperative prophylactic EIS or EVL after hepatectomy. Oncologically, the priority for surgical treatment of HCC may be higher than that for endoscopic treatment of EV. Considering the risk factors for EV bleeding, endoscopic treatment should be performed at least 6 months after hepatectomy to prevent bleeding of EV with Cb and RC. There were several limitations to our study. First, the sample size was relatively small, making it difficult to evaluate the effect of EV on OS in HCC patients after hepatectomy. Second, it is likely that there were differences in disease severity between HCC patients with and without EV after hepatectomy. However, the aim of this study was to clarify the surgical indication for HCC patients with EV, who have poorer liver function than those without EV. Furthermore, we did not compare OS between patients with EV treated with and without surgery. Further studies are therefore needed to show the effects of EV and surgical treatment on the prognosis of HCC patients. In conclusion, the results of this study suggest that hepatectomy may be indicated in HCC patients with EV and ICGR15 B17 %, though EV should be managed in patients with preoperative endoscopic findings of Cb and RC. Acknowledgments The authors thank Dr. Fujiwara and Ms. Kozono from the Hiroshima Red Cross Hospital and Atomic Bomb Survivor Hospital for diagnosing the pathological characteristics of HCC and for statistical advice, respectively. Conflict of interest The authors declare that they have no conflicts of interest.
References 1. Capussotti L, Ferrero A, Vigano` L, Muratore A, Polastri R, Bouzari H (2006) Portal hypertension: contraindication to liver surgery? World J Surg 30(6):992–999. doi:10.1007/s00268-0050524-9 2. Pringle JHV (1908) Notes on the arrest of hepatic hemorrhage due to trauma. Ann Surg. 48(4):541–549 3. D’Amico G, De Franchis R, Cooperative Study Group (2003) Upper digestive bleeding in cirrhosis. Post-therapeutic outcome and prognostic indicators. Hepatology 38(3):599–612 4. Forner A, Reig ME, Bruix J et al (2010) Current strategy for staging and treatment: the BCLC update and future prospects. Semin Liver Dis 30(1):61–74 5. Bruix J, Sherman M, American Association for the Study of Liver Diseases (2011) Management of hepatocellular carcinoma: an update. Hepatology 53(3):1020–1022
123
6. Bruix J, Castells A, Rode´s J et al (1996) Surgical resection of hepatocellular carcinoma in cirrhotic patients: prognostic value of preoperative portal pressure. Gastroenterology 111(4):1018–1022 7. Ishizawa T, Kokudo N, Makuuchi M et al (2008) Neither multiple tumors nor portal hypertension are surgical contraindications for hepatocellular carcinoma. Gastroenterology 134(7):1908–1916 8. Cucchetti A, Ercolani G, Pinna AD et al (2009) Is portal hypertension a contraindication to hepatic resection? Ann Surg 250(6):922–928 9. Llovet JM, Fuster J, Bruix J (1999) Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 30:1434–1440 10. Boleslawski E, Mathurin P, Pruvot FR et al (2012) Hepatic venous pressure gradient in the assessment of portal hypertension before liver resection in patients with cirrhosis. Br J Surg. 99(6):855–863 11. Llop E, Berzigotti A, Reig M et al (2012) Assessment of portal hypertension by transient elastography in patients with compensated cirrhosis and potentially resectable liver tumors. J Hepatol 56(1):103–108 12. Stremitzer S, Ferlitsch A, Gruenberger T (2011) Value of hepatic venous pressure gradient measurement before liver resection for hepatocellular carcinoma. Br J Surg. 98(12):1752–1758 13. Takeishi K, Shirabe K, Maehara Y et al (2011) Clinicopathological features and outcomes of young patients with hepatocellular carcinoma after hepatectomy. World J Surg 35(5):1063–1071. doi:10.1007/s00268-011-1017-7 14. Pugh RN, Murray-Lyon IM, Williams R et al (1973) Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 60(8):646–649 15. Yamashita YI, Shirabe K, Maehara Y et al (2014) Trends in surgical results of hepatic resection for hepatocellular carcinoma: 1,000 consecutive cases over 20 years in a single institution. Am J Surg 207(6):890–896 16. Tajiri T, Futagawa S, Idezuki Y et al (2010) General rules for recording endoscopic findings of esophagogastric varices (2nd edition). Dig Endosc 22(1):1–9 17. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240(2):205–213 18. Rahbari NN, Bu¨chler MW, Weitz J et al (2011) Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery 149(5):713–724 19. Royston P, Altman DG, Sauerbrei W (2006) Dichotomizing continuous predictors in multiple regression: a bad idea. Stat Med 25:127–141 20. Kokudo N (2015) Is it a time to modify the BCLC guidelines in terms of the role of surgery? World J Surg. 39(2):469–470. doi:10.1007/s00268-014-2915-2 21. Shirabe K, Kayashima H, Maehara Y et al (2011) Improvement of long-term outcomes in hepatitis C virus antibody-positive patients with hepatocellular carcinoma after hepatectomy in the modern era. World J Surg 35:1072–1084. doi:10.1007/s00268011-1013-y 22. Dutkowski P, Mu¨llhaupt B, Clavien PA et al (2015) Challenges to liver transplantation and strategies to improve outcomes. Gastroenterology 148(2):307–323 23. Miyoshi H, Hongou Y, Katsu K et al (1997) Efficacy of prophylactic sclerotherapy in patients with hepatocellular carcinoma and varices negative for the red color sign. Gastrointest Endosc 45(6):498–502 24. Akahoshi T, Hashizume M, Maehara Y et al (2014) Merits of prophylactic sclerotherapy for esophageal varices concomitant unresectable hepatocellular carcinoma: prospective randomized study. Dig Endosc 26(2):172–177
World J Surg 25. Lay CS, Wu KL, Lo KJ et al (1997) Endoscopic variceal ligation in prophylaxis of first variceal bleeding in cirrhotic patients with high-risk esophageal varices. Hepatology 25(6):1346–1350 26. D’Amico G, Pagliaro L, Bosch J (1995) The treatment of portal hypertension: a meta-analytic review. Hepatology 22(1):332–354 27. The Veterans Affairs Cooperative Variceal Sclerotherapy Group (1991) Prophylactic sclerotherapy for esophageal varices in men
with alcoholic liver disease. A randomized, single-blind, multicenter clinical trial. N Engl J Med 324(25):1779–1784 28. The PROVA Study Group (1991) Prophylaxis of first hemorrhage from esophageal varices by sclerotherapy, propranolol or both in cirrhotic patients: a randomized multicenter trial. Hepatology 14(6):1016–1024
123
ORIGINAL SCIENTIFIC REPORT
Surgical Resection for Hepatocellular Carcinoma with Concomitant Esophageal Varices Noboru Harada1 • Ken Shirabe2 • Takashi Maeda1 • Hiroto Kayashima1 Teruyoshi Ishida1 • Yoshihiko Maehara2
•
Ó Socie´te´ Internationale de Chirurgie 2015
Abstract Background The management of hepatocellular carcinoma (HCC) in patients with concomitant esophageal varices (EV) remains controversial. We assessed the surgical outcome of hepatectomy and aimed to clarify the indications and management of HCC in patients with concomitant EV. Methods We retrospectively enrolled 502 patients with HCC (100 with and 402 without EV), who underwent curative hepatectomy. We analyzed the prognostic outcomes and risk factors for EV bleeding after hepatectomy. Results Overall survival (OS) was significantly lower in HCC patients with EV than in those without EV (p = 0.003), although recurrence-free survival was similar in both groups. Multivariate analysis showed that indocyanine green retention test at 15 min (ICGR15)[17 % (p = 0.007) and a-fetoprotein [12.5 ng/ml (p = 0.003) was independent predictors of poorer OS. Among patients with EV who underwent hepatectomy, multivariate analysis identified ICGR15 [17 % (p = 0.03) as the only independent predictor of poorer OS. There was no significant difference in OS between HCC patients with EV and ICGR15 B17.0 % and HCC patients without EV. Ten patients experienced EV bleeding after hepatectomy. Multivariate analysis showed that preoperative endoscopic findings of blue color EV (p = 0.008) and red color sign (p = 0.0005) were independent predictors of EV bleeding in patients with HCC after hepatectomy. Conclusions These results suggest that HCC patients with EV and ICGR15 B17 % may be suitable for surgery, but patients with preoperative endoscopic blue color EV and red color sign need to be managed appropriately.
Introduction Many patients with hepatocellular carcinoma (HCC) have concomitant esophageal varices (EVs) which are considered a clinical manifestation of portal hypertension (PHT) [1].
& Noboru Harada [email protected] 1
Department of Surgery, Hiroshima Red Cross Hospital and Atomic Bomb Survivors Hospital, Hiroshima 730-8619, Japan
2
Department of Surgery and Medical Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
Increased portal venous pressure after hepatic resection may cause variceal bleeding, especially in relation to Pringle’s maneuver [2] or major hepatectomy. EV bleeding is a medical emergency associated with a mortality of around 20 % [3]. According to the Barcelona Clinic Liver Cancer Group [4] and American Association for the Study of Liver Disease [5] guidelines, HCC with clinically significant PHT is a contraindication for hepatectomy. Clinically significant PHT [hepatic venous pressure gradient (HVPG) C10 mmHg] is the most powerful predictor of postoperative liver failure or poor long-term survival in patients with Child-Pugh A liver function [6]. Preoperative HVPG measurement is invasive and not performed routinely in most liver centers, and indirect clinical parameters (EV
123
World J Surg
and/or splenomegaly associated with thrombocytopenia) are therefore considered clinical signs of PHT. However, several reports have suggested that PHT should not necessarily be a contraindication for hepatic resection in cirrhotic patients [1, 7, 8]. Increased HVPG was associated with postoperative liver dysfunction and mortality after liver resection in a small number of studies including a maximum of 46 patients with HCC and liver cirrhosis [6, 9–12], suggesting that some patients demonstrating indirect criteria of PHT may still be eligible for surgery in the presence of adequate selection criteria [1, 7, 8]. The surgical indications for HCC-associated EV thus remain controversial and the risk factors for EV bleeding after hepatectomy also remain unclear. This study aimed to assess the surgical outcomes of liver resection for HCC in patients with EV, and to clarify the indications and management of HCC patients with concomitant EV.
postoperative follow-up were not performed routinely in our hospital. However, five patients with apparent risky EV (F2 or F3, and RC positive) underwent endoscopic injection sclerotherapy (EIS) or endoscopic variceal ligation (EVL). Serious postoperative complications were defined as Clavien–Dindo grade III or higher [17]. Postoperative liver failure was defined according to the guidelines of the International Study Group of Liver Surgery [18]. Survival and recurrence
Methods
Patients underwent blood tests and computed tomography every 3 months after hepatic resection. Recurrence was diagnosed based on imaging findings. Patients with intrahepatic recurrence were managed with ablative therapy, such as radiofrequency ablation and percutaneous ethanol injection, transcatheter arterial chemoembolization, or surgery. In the event of death, survival time after surgery and cause of death were recorded. Postoperative survival time was also recorded for all patients.
Patients and surgical procedures
Statistical analysis
A total of 502 consecutive patients with HCC who underwent curative hepatectomy at the Department of Surgery, Hiroshima Red Cross Hospital and Atomic Bomb Survivors Hospital between September 2004 and October 2013, were included in the study. All patients had a confirmed pathological diagnosis of HCC and were diagnosed preoperatively with EV by upper-gastrointestinal endoscopy about 1 month before surgery. The type of hepatectomy was based on liver function and tumor extension [13]. Liver function was assessed by Child-Pugh classification [14] and indocyanine green retention test at 15 min (ICGR15). The hepatectomy procedures have been described elsewhere [13, 15]. Patients with HCC were divided into those with and without EV. Written informed consent was obtained from all patients for surgical treatment, according to institutional guidelines. The study protocol conformed to the updated ethical guidelines of the 2013 Declaration of Helsinki and was approved by our Institutional Review Board.
Continuous variables were analyzed parametrically using Student’s t tests. Wilcoxon’s rank-sum test was used for nonparametric analyses. Categorical variables were compared using v2 or Fisher’s exact tests. Overall survival (OS) and recurrence-free survival (RFS) rates were calculated by the Kaplan–Meier method and compared by univariate logrank test. Independent factors for OS were identified by multivariate analysis of factors significant in univariate analysis using a Cox proportional hazards model [19]. Cutoff values for continuous variables were calculated as the median values of the whole study population. Significant risk factors for EV bleeding were identified by multivariate logistic regression analysis of risk factors significant in univariate analysis. Statistical significance was set at p \ 0.05. Statistical analyses were performed using JMP 9.0 software (SAS Institute, Cary, NC, USA).
Results Definitions Curative resection was defined as complete macroscopic and microscopic tumor removal. Liver cirrhosis was confirmed by histological examination of a resected specimen. Major hepatectomy was defined as resection of at least two subsegments, and minor resection, including partial resection, involved fewer than two subsegments. EV was classified preoperatively based on endoscopic findings [16], according to location, form, color, and red color signs (RC). Preoperative prophylactic EV treatment and
123
Clinicopathological characteristics of HCC patients with and without EV who underwent hepatectomy Preoperative background characteristics of HCC patients with and without EV who underwent hepatectomy are summarized in Table 1. Compared with patients without EV, patients with EV included more women (41.0 vs. 30.8 %, p = 0.037), more patients with liver cirrhosis (90.0 vs. 55.1 %, p \ 0.0001), and had a lower mean platelet count (104,000 vs. 146,000/ll, p \ 0.0001), lower mean
World J Surg Table 1 Preoperative background characteristics of HCC patients with and without EV who underwent hepatectomy Variables
Without EV (n = 402)
With EV (n = 100)
p
Age (years), mean (range)
69.0 (34–87)
69.1 (43–85)
0.740
Gender (male/female) (n)
281/121
59/41
0.037
Anti-HCVAb positive (n) HBV surface antigen positive (n)
262 63
73 16
0.155 0.985
Liver cirrhosis (n)
221 (55.1 %)
90 (90.0 %)
\0.0001
Platelet (9104/ll), mean (range)
14.6 (2.4–53)
10.4 (3.3–24)
\0.0001
Albumin (g/dl), mean (range)
4.0 (2.5–5.1)
3.9 (2.9–4.7)
0.002
Total bilirubin (mg/dl), mean (range)
0.7 (0.2–1.7)
0.9 (0.2–2.3)
0.014 \0.0001
Prothrombin activity ( %), mean (range)
92.6 (57.3–130)
84.4 (48.6–117)
Child-Pugh class A/B (n)
393/8
89/11
0.0002
ICGR15, mean (range)
17.5 (0.9–73.7)
25.1 (5.6–68.7)
\0.0001
AFP (ng/ml), mean (range)
808 (0.49–93721)
144 (2.6–2340)
0.032
HCC hepatocellular carcinoma, EV esophageal varices, HCVAb hepatitis C virus antibody, HBV hepatitis B virus, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
serum albumin level (3.9 vs. 4.0 g/dl, p = 0.002), greater mean serum total bilirubin level (0.9 vs. 0.7 mg/dl, p = 0.014), lower mean prothrombin activity (84.4 vs. 92.6 %, p \ 0.0001), higher frequency of Child-Pugh class B (11.0 vs. 2.0 %, p = 0.0002), greater mean ICGR15 (25.1 vs. 17.5 %, p \ 0.0001), and lower mean a-fetoprotein (AFP) level (144 vs. 808 ng/ml, p = 0.032). Tumor and surgical characteristics of HCC patients with and without EV who underwent hepatectomy are summarized in Table 2. Patients with EV had a lower frequency of major hepatectomy (12.0 vs. 22.9 %, p = 0.019) and shorter mean Pringle’s maneuver time (41 vs. 53 min, p = 0.002) than patients without EV. One in-hospital death occurred in a patient with EV secondary to sepsis. Patients with EV had a significantly higher rate of serious postoperative complications (Clavien–Dindo grade III or higher) than those without EV (18.0 vs. 9.2 %; p = 0.020). Eighteen serious postoperative complications (within 30 postoperative days) were recorded in patients with EV, including wound dehiscence (n = 4), bile leakage (n = 3), refractory ascites (n = 2), grade B liver failure (n = 2), and ileus, EV bleeding, pancreas fistula, pneumothorax, pneumonia, jejunal perforation, and sepsis (all n = 1). Thirty-seven serious postoperative complications were recorded in patients without EV, including bile leakage (n = 11), refractory pleural effusion (n = 8), wound dehiscence (n = 8), refractory ascites (n = 4), and grade B liver failure, ileus, postoperative bleeding, pneumonia, wound pain, and sepsis (all n = 1). OS and RFS Median postoperative follow-up after surgery was 39.9 months (range, 0.4–104.2 months). OS was significantly
lower in HCC patients with EV than those without EV (p = 0.003). The 1-, 3-, and 5-year OS rates in patients with EV were 93.7, 61.9, and 44.9 %, respectively, compared with 94.3, 80.4, and 67.2 % in patients without EV (Fig. 1a). The 1-, 3-, and 5-year RFS rates did not differ significantly between patients with (76.6, 41.5, and 29.6 %) and without EV (73.7, 43.1, and 30.3 %) (p = 0.906). Associations between clinicopathological factors and OS were determined (Table 3). Univariate analyses showed that age [70 years (p = 0.034), ICGR15 [17 % (p \ 0.0001), liver cirrhosis (p = 0.004), AFP [25.8 ng/ ml (p \ 0.0001), Child-Pugh class B (p = 0.002), EV (p = 0.003), maximum tumor diameter [22 mm (p = 0.028), multiple tumors (p = 0.021), microscopic vascular invasion (p = 0.016), intraoperative blood transfusion (p = 0.001), and serious postoperative complications (p = 0.039) were significantly associated with poorer OS, whereas sex, histological grading, extent of hepatic resection, and intraoperative blood loss [320 ml were not. Multivariate Cox proportional hazards analysis showed that ICGR15 [17 % [p = 0.007; hazard ratio (HR) 1.67; 95 % confidence interval (CI) 1.15–2.45] and AFP [12.5 ng/ml (p = 0.003; HR 1.78; 95 % CI 1.22–2.62) were independent predictors of poorer OS. Factors affecting OS in HCC patients with EV Univariate analyses showed that ICGR15 [17 % (p = 0.009) and serum albumin B4.0 g/dl (p = 0.035) were significantly associated with poorer OS in HCC patients with EV who underwent hepatectomy. Multivariate Cox proportional hazards analysis identified ICGR15 [17 % (p = 0.030; HR 2.60; 95 % CI 1.09–7.27) as the
123
World J Surg Table 2 Tumoral and surgical characteristics of HCC patients with and without EV who underwent hepatectomy Variables
Without EV (n = 402)
With EV (n = 100)
p
Mean (range)
2.7 (0.5–13)
2.3 (0.7–6)
0.056
Number of tumors, mean (range) Histological grading
1.6 (1–10)
1.4 (1–5)
0.343
57/339
11/88
0.514
244
67
0.417
92/310
12/88
0.019
Maximum tumor diameter (cm)
Well/moderately or poorly (n) Microscopic vascular invasion Extent of hepatic resection Major/minor hepatectomy (n) Pringle’s maneuver time (min) Mean (range) Surgical time (min), mean (range)
53 (0–195)
41 (0–120)
0.002
242 (60–633)
222 (54–483)
0.087
Intraoperative blood loss (ml) Mean (range) Intraoperative blood transfusion (n)
507 (10–4978)
594 (5–4220)
0.444
25
12
0.055
37
18
0.020
Postoperative serious complications (n) Clavien–Dindo CIII HCC hepatocellular carcinoma, EV esophageal varices
only independent predictor of poorer OS (Table 4). OS was similar in patients without EV and patients with EV with ICGR15 B17 % (p = 0.506) (Fig. 1b). The 1-, 3-, and 5-year OS rates in HCC patients with EV and ICGR15 B17 % were 96.4, 78.8, and 72.3 %, respectively, compared with 94.3, 80.4, and 67.2 % in patients without EV. There was no significant difference in the incidence of serious postoperative complications (Clavien–Dindo grade III or higher) between patients with EV and ICGR15 B17 % and patients without EV (13.3 vs. 9.2 %; p = 0.512). Incidence of and risk factors for EV bleeding in patients with HCC after hepatectomy Among 100 HCC patients with EV who underwent hepatectomy, 10 experienced postoperative EV bleeding, and hemostasis was achieved in all patients. The duration from the day of operation to bleeding was 2–867 days (median 172 days). There were three (30 %) and six (60 %) bleeding events in the first 3 and 6 months after liver resection, respectively. Although five patients [two F2 blue color (Cb) RC?, one F2 white color (Cw) RC?, and two F3CbRC?] underwent prophylactic preoperative EIS for EV, two (F2CbRC? and F3CbRC?) still had EV bleeding after hepatectomy. Nine patients underwent postoperative prophylactic EIS for EV and one underwent prophylactic EVL, none of whom had subsequent bleeding. By univariate analyses, EV with form CF2 (p = 0.0006), Cb
123
(p = 0.003), and RC? (p \ 0.0001) were significantly correlated with EV bleeding. Multivariate logistic regression analysis showed that Cb (p = 0.008; odds ratio 15.1; 95 % CI 1.90–402) and RC? (p = 0.0005; odds ratio 55.6; 95 % CI 4.81–1884) were independent predictors of EV bleeding in patients with HCC after hepatectomy (Table 5).
Discussion The results of the current study demonstrated that OS rates were significantly lower in HCC patients with EV compared with those without EV. However, multivariate analysis suggested that EV had only a marginally significant effect on OS in HCC patients who underwent hepatectomy (p = 0.055), implying that similar survival outcomes could be achieved irrespective of the presence of EV. However, it is difficult to compare the outcomes of HCC patients with and without EV. Many hepatobiliary– pancreas surgeons are more cautious about the perioperative surgical management of HCC patients with EV than those without EV. HCC patients with EV are thus likely to have more advanced liver disease than patients without EV. However, we believe that surgery may be suitable in some HCC patients with EV, and new criteria are therefore needed to identify these patients. Kokudo recently proposed that liver resection could be an acceptable option for small HCCs in patients with PHT, provided ICG test results were normal or only slightly impaired, and that risky
World J Surg Fig. 1 Prognoses of 502 patients with and without EV who underwent liver resection for primary HCC. a Kaplan– Meier curves for OS after hepatectomy. OS in HCC patients with EV (bold black line) was significantly lower than in patients without EV (normal black line, p = 0.0032, log-rank test). b Kaplan–Meier curves for OS after hepatectomy among HCC patients without EV (normal black line), HCC patients with EV and ICGR15 B17 % (bold black line), and HCC patients with EV and ICGR15 [17 % (dotted black line). There was no significant difference in OS between HCC patients with EV and ICGR15 B17 % (bold black line) and HCC patients without EV (normal black line, p = 0.506, log-rank test). OS in HCC patients with EV and ICGR15 B17 % (bold black line) was significantly higher than in HCC patients with EV and ICGR15 [17 % (dotted black line, p \ 0.0001, log-rank test). OS in HCC patients without EV (normal black line) was significantly higher than in HCC patients with EV and ICGR15 [17 % (dotted black line, p = 0.009, log-rank test)
esophagogastric varices could be controlled in experienced tertiary centers [20]. The current study found no significant difference in OS rates between HCC patients without EV and those with EV and ICGR15 B17 %. Hepatectomy may therefore be indicated in HCC patients with concomitant EV and ICGR15 B17 %, because of the improved outcomes associated with surgical techniques and perioperative care [21]. We suggest that a cut-off value of ICGR15 B17 % provides a useful indication for surgery in HCC patients with EV. Given the serious shortage of available liver transplant donor organs
[22], surgical resection may represent a better option for the treatment of some HCC patients with concomitant EV. However, these patients should be selected carefully and their treatment should adhere to a strict surgical strategy, including EV management. Perioperative EV management is necessary to minimize the risk of bleeding. To the best of our knowledge, no previous studies have reported on the incidence of and risk factors for EV bleeding after hepatectomy in patients with resectable HCC. Miyoshi et al. reported 3-year bleeding rates of 50 % for the control group and 18 % for the
123
World J Surg Table 3 Univariate and multivariate analyses of overall survival in all patients Variables
Univariate analysis
Multivariate analysis
5-year survival
p value
HR (95 % CI)
p value
69.9/56.3
0.034
1.38 (0.97–1.97)
0.071
64.1/62.9
0.178
73.6/53.3
\0.0001
1.67 (1.15–2.45)
0.007
58.6/73.3
0.004
1.09 (0.71–1.71)
0.684
73.0/54.5
\0.0001
1.78 (1.22–2.62)
0.003
64.7/41.9
0.002
1.30 (0.63–2.45)
0.460
44.9/67.2
0.003
1.54 (0.99–2.34)
0.055
68.8/57.9
0.028
1.40 (0.99–1.98)
0.058
68.2/54.9
0.021
1.30 (0.90–1.87)
0.156
43.5/33.4
0.261
60.3/68.2
0.016
1.33 (0.93–1.93)
0.114
69.8/62.3
0.351
64.9/62.5
0.075
43.1/65.4
0.001
1.71 (0.98–2.84)
0.059
49.0/65.4
0.039
1.29 (0.75–2.10)
0.347
Age (years) B70/[70 Gender Male/female ICGR15 ( %) B17/[17 Liver cirrhosis Yes/no AFP (ng/ml) B12.5/[12.5 Child-Pugh Class A/B Esophageal varices Yes/no Maximum tumor diameter (mm) B22/[22 Multiple tumors Yes/no Histological grading Well/moderately or poorly Microscopic vascular invasion Yes/no Extent of hepatic resection Major/minor hepatectomy Intraoperative blood loss (ml) B320/[320 Intraoperative blood transfusion Transfused/nontransfused Postoperative serious complications Clavien–Dindo CIII (yes/no)
HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
prophylactic sclerotherapy group in patients with unresectable HCC and concomitant EV [23]. Akahoshi et al. reported a cumulative bleeding rate of 48.8 % at 6 months in patients with unresectable HCC and EVs that were likely to bleed, who were treated conservatively [24]. Despite some selection bias, the 3-year postoperative bleeding rate in the current study was 10 % in patients with resectable HCC and concomitant EV (n = 100). The median duration from day of operation to bleeding was \6 months. Preoperative endoscopy found that Cb and RC were strongly associated with EV bleeding in patients with HCC after hepatectomy. Cb and RC at preoperative endoscopy may thus be considered as indications for preoperative or postoperative EV treatment.
123
Most hepatobiliary surgeons aim to reduce the risk of EV bleeding by preoperative endoscopic treatment if upper-gastrointestinal endoscopy reveals moderately enlarged F2 varices with or without RC, or markedly enlarged nodular or tumor-shaped varices (F3) [7]. However, there have been no reports regarding the risk factors for EV bleeding after hepatectomy in patients with HCC. The current treatment strategy for EV is based on the finding that endoscopic sclerotherapy prevents EV bleeding in patients with PHT [24–26]. High-risk varices may bleed as a result of increased portal venous pressure after hepatic resection, associated with Pringle’s maneuver or major hepatectomy. In our study, EV-form F2/3 was not an independent risk factor for bleeding after hepatectomy in
World J Surg Table 4 Univariate and multivariate analyses of overall survival in HCC patients with EV who underwent hepatectomy
Table 5 Logistic regression analyses of risk factors for bleeding in HCC patients with EV who underwent hepatectomy
Variables
Variables
Univariate
Multivariate analysis
p value
HR (95 % CI)
p value
Univariate
Multivariate analysis
p value
Odds ratio (95 % CI)
p value
Age (years) B70/[70 Gender (male/female)
0.848
Age (years) [70
0.998
0.454
Gender (male/female)
0.184
Liver cirrhosis (yes/no)
0.164
ICGR15 ( %) [17
0.719
0.226
AFP (ng/ml) [12.6
0.735
Preoperative EV treatment (yes/no) Postoperative EV treatment (yes/no)
0.107
0.052
EV-location (Li/Ls, Lm)
0.085
EV-form (F2-3/F0-1)
0.0006
0.49 (0.02–5.91)
0.607
EV-color (blue/white)
0.003
15.1 (1.90–402)
0.008
EV-RC sign (presence/ absence)
\0.0001
55.6 (4.81–1884)
0.0005
ICGR15 (%) B17/[17
0.009
Liver cirrhosis (yes/no)
2.60 (1.09–7.27)
0.030
4
Platelet count (910 /ll) B12.6/[12.6 Total bilirubin (mg/dl) B0.7/[0.7
0.199
Serum albumin (g/dl) B4.0/[4.0
0.035
1.72 (0.79–4.20)
0.181
Prothronbin activity ( %) B91.4/[91.4
0.141
AFP (ng/ml) B12.5/[12.5
0.041
Child-Pugh class B/A
0.358 0.161 0.079
EV-location (Li/Ls, Lm)
0.390
Maximum tumor diameter [22 mm
EV-form (F2-3/F0-1)
0.238
Multiple tumors (yes/no)
EV-color (blue/white)
0.615
Histological grading
EV-RC sign (presence/ absence)
0.098
B22/[22
Microscopic vascular invasion (yes/no)
0.210
Extent of hepatic resection
0.847
Pringle’s maneuver time [45 min
0.836
0.321
Intraoperative blood loss [320 ml
0.103
Major/minor
0.147
Pringle’s maneuver time (min) B45/[45
0.678
Intraoperative blood loss (ml) B320/[320
0.312
0.400
Intraoperative blood transfusion
Extent of hepatic resection Major/minor
0.838
0.856
Histological grading Well/moderately or poorly differentiation
Well/moderately of poorly differentiation Microscopic vascular invasion (yes/no)
Maximum tumor diameter (mm) Multiple tumors (yes/no)
0.277
0.194
Intraoperative blood transfusion Transfused/nontransfused 0.349
Transfused/ nontransfused
0.103
Postoperative serious complications
0.838
Clavien–Dindo CIII
0.189
HCC hepatocellular carcinoma, EV esophageal varices, HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein
Postoperative serious complications Clavien–Dindo CIII (yes/ no)
0.266
HCC hepatocellular carcinoma, EV esophageal varices, HR hazard ratio, CI confidence interval, ICGR15 indocyanine green retention test at 15 min, AFP a-fetoprotein, RC red color sign
patients with HCC, nor were surgical factors such as major hepatectomy, Pringle’s maneuver time, intraoperative blood loss, and blood transfusion associated with EV bleeding, according to univariate analysis. Further studies
are needed to clarify the association between EV bleeding and surgical factors. Several clinical studies have concluded that prophylactic therapy of EV is ineffective, and it is generally not accepted in western countries [25, 27, 28]. Although five patients in our study (two F2CbRC?, one F2CwRC?, and two F3CbRC?) underwent prophylactic EIS for EV before hepatectomy, two (F2CbRC? and F3CbRC?) still
123
World J Surg
experienced postoperative EV bleeding episodes. These results support the idea that prophylactic EIS may have minimal benefits. There is currently no evidence regarding the benefit of postoperative prophylactic EIS or EVL in HCC patients with concomitant EV. We found no bleeding in patients who underwent postoperative prophylactic EIS or EVL after hepatectomy. Oncologically, the priority for surgical treatment of HCC may be higher than that for endoscopic treatment of EV. Considering the risk factors for EV bleeding, endoscopic treatment should be performed at least 6 months after hepatectomy to prevent bleeding of EV with Cb and RC. There were several limitations to our study. First, the sample size was relatively small, making it difficult to evaluate the effect of EV on OS in HCC patients after hepatectomy. Second, it is likely that there were differences in disease severity between HCC patients with and without EV after hepatectomy. However, the aim of this study was to clarify the surgical indication for HCC patients with EV, who have poorer liver function than those without EV. Furthermore, we did not compare OS between patients with EV treated with and without surgery. Further studies are therefore needed to show the effects of EV and surgical treatment on the prognosis of HCC patients. In conclusion, the results of this study suggest that hepatectomy may be indicated in HCC patients with EV and ICGR15 B17 %, though EV should be managed in patients with preoperative endoscopic findings of Cb and RC. Acknowledgments The authors thank Dr. Fujiwara and Ms. Kozono from the Hiroshima Red Cross Hospital and Atomic Bomb Survivor Hospital for diagnosing the pathological characteristics of HCC and for statistical advice, respectively. Conflict of interest The authors declare that they have no conflicts of interest.
References 1. Capussotti L, Ferrero A, Vigano` L, Muratore A, Polastri R, Bouzari H (2006) Portal hypertension: contraindication to liver surgery? World J Surg 30(6):992–999. doi:10.1007/s00268-0050524-9 2. Pringle JHV (1908) Notes on the arrest of hepatic hemorrhage due to trauma. Ann Surg. 48(4):541–549 3. D’Amico G, De Franchis R, Cooperative Study Group (2003) Upper digestive bleeding in cirrhosis. Post-therapeutic outcome and prognostic indicators. Hepatology 38(3):599–612 4. Forner A, Reig ME, Bruix J et al (2010) Current strategy for staging and treatment: the BCLC update and future prospects. Semin Liver Dis 30(1):61–74 5. Bruix J, Sherman M, American Association for the Study of Liver Diseases (2011) Management of hepatocellular carcinoma: an update. Hepatology 53(3):1020–1022
123
6. Bruix J, Castells A, Rode´s J et al (1996) Surgical resection of hepatocellular carcinoma in cirrhotic patients: prognostic value of preoperative portal pressure. Gastroenterology 111(4):1018–1022 7. Ishizawa T, Kokudo N, Makuuchi M et al (2008) Neither multiple tumors nor portal hypertension are surgical contraindications for hepatocellular carcinoma. Gastroenterology 134(7):1908–1916 8. Cucchetti A, Ercolani G, Pinna AD et al (2009) Is portal hypertension a contraindication to hepatic resection? Ann Surg 250(6):922–928 9. Llovet JM, Fuster J, Bruix J (1999) Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation. Hepatology 30:1434–1440 10. Boleslawski E, Mathurin P, Pruvot FR et al (2012) Hepatic venous pressure gradient in the assessment of portal hypertension before liver resection in patients with cirrhosis. Br J Surg. 99(6):855–863 11. Llop E, Berzigotti A, Reig M et al (2012) Assessment of portal hypertension by transient elastography in patients with compensated cirrhosis and potentially resectable liver tumors. J Hepatol 56(1):103–108 12. Stremitzer S, Ferlitsch A, Gruenberger T (2011) Value of hepatic venous pressure gradient measurement before liver resection for hepatocellular carcinoma. Br J Surg. 98(12):1752–1758 13. Takeishi K, Shirabe K, Maehara Y et al (2011) Clinicopathological features and outcomes of young patients with hepatocellular carcinoma after hepatectomy. World J Surg 35(5):1063–1071. doi:10.1007/s00268-011-1017-7 14. Pugh RN, Murray-Lyon IM, Williams R et al (1973) Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 60(8):646–649 15. Yamashita YI, Shirabe K, Maehara Y et al (2014) Trends in surgical results of hepatic resection for hepatocellular carcinoma: 1,000 consecutive cases over 20 years in a single institution. Am J Surg 207(6):890–896 16. Tajiri T, Futagawa S, Idezuki Y et al (2010) General rules for recording endoscopic findings of esophagogastric varices (2nd edition). Dig Endosc 22(1):1–9 17. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240(2):205–213 18. Rahbari NN, Bu¨chler MW, Weitz J et al (2011) Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery 149(5):713–724 19. Royston P, Altman DG, Sauerbrei W (2006) Dichotomizing continuous predictors in multiple regression: a bad idea. Stat Med 25:127–141 20. Kokudo N (2015) Is it a time to modify the BCLC guidelines in terms of the role of surgery? World J Surg. 39(2):469–470. doi:10.1007/s00268-014-2915-2 21. Shirabe K, Kayashima H, Maehara Y et al (2011) Improvement of long-term outcomes in hepatitis C virus antibody-positive patients with hepatocellular carcinoma after hepatectomy in the modern era. World J Surg 35:1072–1084. doi:10.1007/s00268011-1013-y 22. Dutkowski P, Mu¨llhaupt B, Clavien PA et al (2015) Challenges to liver transplantation and strategies to improve outcomes. Gastroenterology 148(2):307–323 23. Miyoshi H, Hongou Y, Katsu K et al (1997) Efficacy of prophylactic sclerotherapy in patients with hepatocellular carcinoma and varices negative for the red color sign. Gastrointest Endosc 45(6):498–502 24. Akahoshi T, Hashizume M, Maehara Y et al (2014) Merits of prophylactic sclerotherapy for esophageal varices concomitant unresectable hepatocellular carcinoma: prospective randomized study. Dig Endosc 26(2):172–177
World J Surg 25. Lay CS, Wu KL, Lo KJ et al (1997) Endoscopic variceal ligation in prophylaxis of first variceal bleeding in cirrhotic patients with high-risk esophageal varices. Hepatology 25(6):1346–1350 26. D’Amico G, Pagliaro L, Bosch J (1995) The treatment of portal hypertension: a meta-analytic review. Hepatology 22(1):332–354 27. The Veterans Affairs Cooperative Variceal Sclerotherapy Group (1991) Prophylactic sclerotherapy for esophageal varices in men
with alcoholic liver disease. A randomized, single-blind, multicenter clinical trial. N Engl J Med 324(25):1779–1784 28. The PROVA Study Group (1991) Prophylaxis of first hemorrhage from esophageal varices by sclerotherapy, propranolol or both in cirrhotic patients: a randomized multicenter trial. Hepatology 14(6):1016–1024
123
