Abstract
Background: For eradicating portal venous tumor extension and intrahepatic metastasis in hepatocellular carcinoma (HCC), anatomical resection is, in theory, preferable. Patients and Methods: We carried-out a retrospective cohort study in 110 patients who underwent curative hepatic resection (anatomical resection; n=20, and limited resection; n=90) for solitary recurrent HCC from 1990-2010. Results: No significant difference was found in short-term surgical results such as mortality, morbidity, and duration of hospital stay between the two groups. Anatomical resection did not influence overall and disease-free survival for all patients with a solitary recurrent HCC. In patients with cancer spread, such as pathological vascular invasion and intrahepatic metastasis (n=61), or with des-γ-carboxy prothrombin (DCP)≥100 mAU/ml (n=73), the disease-free survival rate in the anatomical-resection group was significantly better than that in the limited-resection group (p=0.0452 and p=0.0345, respectively). Conclusion: Anatomical resection should be recommended only for HCC suspected of exhibiting cancer spread as reflected by DCP ≥100 mAU/ml in patients with solitary recurrent HCC.
- Recurrent hepatocellular carcinoma
- anatomical resection
- cancer spread
- overall survival
- disease-free survival
- prognostic factors
Recently, anatomical resection such as subsegmentectomy is the standard treatment for hepatocellular carcinoma (HCC) (1-3). HCCs often invade the portal vein and form intrahepatic metastasis (4). Anatomical resection, in theory, can eradicate intrahepatic metastasis confined to tumor-bearing portal tributaries, and may have prognostic impacts for patients with HCC.
HCC has a high incidence of intrahepatic recurrence in up to 68-98% of patients after initial hepatic resection (5). Repeat hepatectomy has been reported to be safe and to prolong survival after intrahepatic recurrence of HCC(6-8). But anatomical resection cannot always be selected for patients with recurrent HCC because of its technical difficulty and the high recurrent potential of the remnant liver. There is no report on the surgical outcomes of anatomical resection for recurrent HCC in the literature.
We herein presented a retrospective analysis of surgical results including the prognosis of anatomical resection for a solitary recurrent HCC.
Patients and Methods
Patients' characteristics. One hundred and ten patients with solitary recurrent HCC after initial curative hepatic resection who underwent repeat hepatic resections at the Department of Surgery, Hiroshima Red Cross and Atomic Bomb Survivors Hospital, from January 1990 to December 2011 were retrospectively analyzed.
Surgical procedures. Our indications for hepatic resection for recurrent HCC were previously described (9, 10). Essentially, patients with recurrent HCC with disease-free survival (DFS) of one year or less after initial hepatic resection were considered to be contraindicated for repeat hepatic resection (8). The type of operative procedure was principally determined according to the preoperative indocyanine green retention rate at 15 min (ICG 15R); ICGR15 <20% for anatomical resection, and ICGR15 ≥20% for limited resection. Anatomical resection (n=20) includes hemihepatectomy, segmentectomy, and subsegmentectomy or more based on Couinaud's classification (11). The operative procedures conducted in the anatomical resection group are summarized in Table I. In segmentectomy and hemihepatectomy, we divided the liver along the demarcation line appearing after the occlusion of extrahepatic Glisson's pedicle (12). In anatomical subsegmentectomy, we detected intrahepatic Glisson's branches charging the tumor-bearing area under intraoperative ultrasound guidance. We initially divided the liver toward the intrahepatic Glisson's branches, ligated and dissected the Glisson's branch near the pedicle, and finally divided the liver along the demarcation line without using Makuuchi's procedure involving the injection of dye in the portal vein (13). In limited resection (n=90), we divided the liver along a line so as to secure a surgical margin of at least 5 mm, if possible. When securing this margin was impossible, liver parenchymal transection was performed so as not to expose the tumor surface (14, 15).
There was no surgical mortality in any of 110 patients. Complications were evaluated by Clavien's classification of surgical complications, and those with a score of grade II or more were defined as positive (16).
Follow-up and treatment strategy for recurrent HCC. After discharge, all patients were examined for recurrence by ultrasonography and tumor markers such asα-fetoprotein (AFP) and des-γ-carboxy prothrombin (DCP) every month and by dynamic computed tomography every three months (17). The median follow-up period after repeat hepatic resection was 4.2 years (range=0.7-10.0 years). When second or third recurrence was suspected, we treated by repeat hepatic resection according to the same indications for first recurrent HCC (9,10), ablation therapy, and liodolization (18).
Statistics. The survival curves of the anatomical and limited resection groups were generated by the Kaplan Meier method and compared by the log-rank test. To evaluate the good prognostic factors after repeat hepatic resection for recurrent HCC, we performed multivariate analysis with the Cox proportional hazard model, using a variable-selection method by the backward-elimination procedure. A value of p<0.05 was set as the cutoff for the elimination using the following 18 clinical, surgical, and tumor-related variables according to the previous reports (2,8,13,19-22): age (< vs. ≥65 years); diabetes mellitus (present vs. absent);, preoperative serum total bilirubin level (< vs. ≥1 mg/dl); ICG R15 (< vs. ≥20%); albumin level (> vs. ≤3.5 g/dl); Child-Pugh class (A vs. B); Liver damage (A vs. B) (15); background liver status as assessed histologically (cirrhosis vs. non-cirrhosis); preoperative AFP level (< vs. ≥100 ng/ml); preoperative DCP level (< vs. ≥100 mAU/ml); tumor size (< vs. ≥3.0 cm); cancer spread including pathological vascular invasion and intrahepatic metastasis (present vs. absent); tumor cell differentiation (well or moderately vs. poorly); surgical time (< vs. ≥300 min); surgical blood loss (< vs. ≥1000 g); operative procedure (anatomical vs. limited); intra-operative transfusion (present or absent); surgical margin (> vs. ≤5 mm).
Continuous variables were expressed as the mean±S.D. and compared using Student's t-test. Categorical variables were compared using the χ2 test. All analyses were performed with Statview 5.0 software (Abacus Concepts, Berkeley, CA, USA). A p-values less than 0.05 were considered to indicate statistical significance.
Results
The background characteristics, surgical outcomes, and tumor-related factors in the anatomical and limited-resection groups are summarized in Table II. In the anatomical resection group, the serum level of albumin was significantly higher than that in the limited-resection group (3.9±0.4 g/dl vs. 4.2±0.4 g/dl; p=0.0162). Although the difference was statistically not significant, there were more patients with Liver damage A (75.6% vs. 95.0%; p=0.0531) and less histological cirrhosis (67.8% vs. 45.0%; p=0.0514) in the anatomical-resection group than in the limited-resection group. As for the surgical outcomes, surgical time (203±85 min vs. 281±120 min; p=0.0009) and resected volume (36±39 g vs. 130±81 g; p<0.0001) were significantly higher in the anatomical-resection group than in the limited-resection group. But the intraoperative transfusion rate, complication rate, and duration of hospital stay were similar between the two groups. As for tumor-related factors, the tumor size in the anatomical-resection group was significantly larger than in the limited-resection group (1.9±0.8 cm vs. 2.3±1.0 cm; p=0.0362). In addition, there were significantly more patients with cancer spread in the anatomical-resection group than in the limited-resection group (52.2% vs. 80.0%; p=0.0213). But pathological tumor differentiation, the values of AFP and DCP were similar between the two groups. Briefly summarized, the anatomical-resection group included more patients with good liver function, more surgical stress, and advanced tumor stage.
The overall survival (OS) and disease-free (DFS) curves of the anatomical and limited resection groups are illustrated in Figure 1. The OS was similar in the two groups. The 2-year DFS rates were 48% in the limited resection group and 63% in the anatomical-resection group, but this difference was not statistically significant (p=0.2846). Multivariate analysis identified two factors reflecting good prognosis (Liver damage A, and ICGR15 <20%) influencing OS and three good prognostic factors (Liver damage A, Blood loss<1000 g, and without cancer spread) influencing DFS (Table III). Therefore, in our study, anatomical resection had no prognostic impact beyond that of limited resection for a solitary recurrent HCC.
The OS and DFS curves of the anatomical (n=16)-and limited-resection (n=45) groups are illustrated in Figure 2 in patients with cancer spread. DFS was significantly better in the anatomical resection group than in the limited resection group (p=0.0452). The two-year DFS rates were 35% in the limited-resection group and 59% in the anatomical-resection group, respectively.
For patients with DCP ≥100 mAU/ml (n=73), the OS and DFS curves of the anatomical (n=15) and limited resection (n=58) groups are illustrated in Figure 3. The DFS was significantly better in the anatomical-resection group than in the limited-resection group (p=0.0345). The two-year DFS rates were 42% in the limited-resection group and 61% in the anatomical-resection group.
Discussion
Repeat hepatic resection for recurrent HCC was first reported to be effective more than two decades ago, and it is still the main option for recurrent HCC treatment (6-8). However, there was no evidence that anatomical resection should be recommended for recurrent HCC in the literature. As far as we are aware of, this is the first report on the surgical outcomes of anatomical resection for recurrent HCC. In our retrospective analysis of 110 patients with a solitary recurrent HCC, anatomical resection was safely performed but the prognostic impacts of anatomical resection were not confirmed.
In patients with recurrent HCC, anatomical resection would be a difficult procedure because of the intra-abdominal adhesions, especially around the hepatic hilum, caused by previous hepatic resection. This was true in our series, in which the surgical time was significantly prolonged in the anatomical-resection group (203±85 min vs. 281±120 min; p=0.0009). However, there were no significant differences in short-term surgical results such as complication rate (13% vs. 17%; p=0.3910) and the duration of hospital stay (19±22 days vs. 23±28 days; p=0.4772) between the two groups. According to the patient backgrounds shown in Table II, patients in the anatomical-resection group showed a better preservation of liver function. We previously reported that liver dysfunction was a strong predictive factor linked to postoperative mortality and morbidity (23). The maintenance of short-term surgical results in the anatomical-resection group is attributable to the adequate selection of candidates for anatomical resection for recurrent HCC.
In the treatment of HCC, the eradication of intrahepatic metastasis via vascular invasion is one of the most important considerations (4). Anatomical resection such as subsegmentectomy taking into consideration both the eradication of intrahepatic metastasis and preservation of liver functional parenchyma would be a theoretically reasonable procedure (1-3). However, in our retrospective analysis, anatomical resection had no prognostic impacts for recurrent HCC, as shown in Figure 1. As shown in Table III, multivariate analysis identified lack of cancer spread as a good prognostic factor influencing DFS. There were significantly more patients with cancer spread in the anatomical resection group than in the limited resection group (52.2% vs. 80.0%; p=0.0213). Therefore, there is a possibility that anatomical resection for recurrent HCC leads to favorable DFS in patients with highly recurrent HCC after repeat hepatic resection.
For patients with cancer spread (n=61), DFS was significantly better in the anatomical-resection group than in the limited-resection group (p=0.0452). This result may indicate that anatomical resection also had greater potential for the eradication of intrahepatic metastasis via vascular invasion than limited resection also in patients with recurrent HCC. However, anatomical resection was associated with significantly higher surgical stress, including greater surgical time (203±85 min vs. 281±120 min; p=0.0009) and resected volume (36±39 g vs. 130±81 g; p<0.0001), and the high surgical stress and the decrease of the remnant hepatic reserve resulting from anatomical resection had adverse effects on patient prognosis (15, 24). Anatomical resection, of course, has no beneficial effect against multicentric recurrence. Patients with recurrent HCC after repeat hepatic resection would have high recurrent potential in the remnant liver compared to those with primary HCC (8, 25). Therefore, the better local control of anatomical resection did not lead to better DFS in patients with recurrent HCC.
We have previously reported that a high DCP level could predict cancer spread of HCC (26-28). Koike et al. performed a prospective study to clarify the significance of DCP and concluded positivity of DCP to be the strongest predictor for portal vein invasion (29). In patients with DCP ≥100 mAU/ml in our series, DFS was significantly better in the anatomical-resection group than in the limited-resection group (p=0.0452). DCP≥100 mAU/ml should be one of the criteria indicating anatomical resection for recurrent HCC.
In conclusion, anatomical resection should be recommended only for HCC cases suspected of having cancer spread, as reflected by DCP ≥100 mAU/ml in patients with recurrent HCC.
- Received May 9, 2014.
- Revision received June 16, 2014.
- Accepted June 17, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved