Long-term outcomes of synchronous splenectomy and hepatectomy or radiofrequency ablation for hepatocellular carcinoma and esophagogastric variceal bleeding


 Aim: This study aimed to compare the long-term outcomes of hepatectomy and radiofrequency ablation (RFA) combined with pericardial devascularization (PCDV) plus splenectomy for patients with cirrhosis having hepatocellular carcinoma and esophagogastric variceal bleeding.Materials and Methods: Between October 2008 and March 2018, 46 patients with cirrhosis having hepatocellular carcinoma and esophagogastric variceal bleeding for portal hypertension were included in this study. The overall survival curves, recurrence-free survival curves, and rebleeding-free survival curves were plotted using Kaplan–Meier analysis. The log-rank test was used to compare time-to-event curves between groups.Results: The median follow-up time was 38 months. Among 20 patients undergoing RFA, the 1-, 3-, and 5-year overall survival rates were 95.00%,60.00%, and 35.00%, respectively. The 1-, 3- and 5-year recurrence-free survival rates were 35.00%, 25.00%, and 10.00%, respectively. The 1,3- and 5-year rebleeding-free survival rates were 85.00%, 60.00%, and 40.00%, respectively. Among 26 patients undergoing hepatectomy, the 1-, 3-, and 5-year overall survival rates were 96.15%,50.00%, and 34.62%, respectively. The 1-, 3-, and 5-year recurrence-free survival rates were 65.38%, 19.23%, and 11.54%, respectively. The 1-, 3-, and 5-year rebleeding-free survival rates were 73.08%, 42.31%, and 26.92%, respectively. No significant differences were found in overall, recurrence-free, and rebleeding-free survival rates.Conclusions: Hepatectomy or RFA with PCDV plus splenectomy might be a safe and effective treatment for patients with cirrhosis having hepatocellular carcinoma and esophagogastric variceal bleeding. “Hepatectomy first” strategy may be considered due to its lower and later recurrence. More attention should be paid to background liver diseases after surgery.


Introduction
Hepatocellular carcinoma (HCC) is the sixth most commonly diagnosed cancer and the fourth cause of cancer-related death worldwide, accounting for 841,000 new cases and 782,000 deaths annually 1 .
Surgical resection is the rst choice of treatment for HCC, which has been recommended as the primary treatment for solitary hepatocellular cancer with Child's A stage and normal bilirubin level, and without signi cant portal hypertension 2,3 . The indication for hepatectomy has expanded with the improvement in surgical techniques and perioperative management 4 . A previous study suggested that portal hypertension should not be considered as the absolute contraindication 5 . Hepatectomy is one of the treatment choices for patients with Child's A and B early-stage tumor 6 . Radiofrequency ablation (RFA) is a minimally invasive therapy for HCC, with the advantages of ease of use, less invasiveness, good e cacy, and relatively low cost. It may have equivalent e cacy of resection for tumors with a diameter ≤3 cm 7 , 8 . RFA ablates the whole tumor while minimizing the damage to nontumor tissues. Covering the range of tumor in ltration and satellite lesions at the same time, it is also one of the curative modalities for HCC.
Cirrhosis is the most predisposing factor for HCC. Approximately 80% of HCCs develop in patients with cirrhosis 9 . Esophagogastric variceal bleeding (EVB) is a common complication in patients with HCC. The incidence of patients with HCC presenting with EVB ranges from 1% to 15% 10 . Tumor in ltration can aggravate portal hypertension, resulting in the development of collateral circulation and esophagogastric varices. HCC is one of several factors associated with increased mortality for an episode of variceal bleeding.
In China, for most cirrhotic patients with variceal bleeding, the available treatment is endoscopic therapy, surgery and interventional therapies. Interventional therapies such as TIPS, might best serve as a bridge to liver transplantation for patients with refractory variceal bleeding, but, timely liver transplantation is often not available and the cost of TIPS is signi cant. Pericardial devascularization (PCDV) plus splenectomy is widely used to treat patients with variceal bleeding. For patients with Child's A or B stage, PCDV is not inferior to TIPS in terms of treating portal hypertension and recurrent variceal bleeding and preventing rebleeding 11 , which is a safe and effective treatment 12 .
Meanwhile, studies indicated that splenectomy combined with hepatectomy could extend disease-free survival and overall survival 13,14 . Hepatectomy or ablation combined with splenectomy is a safe treatment in selected patients with HCC. It shows acceptable mortality and complications and favorable survival bene ts 15 .
The prognostic impacts of hepatectomy or RFA combined with PCDV and splenectomy have rarely been explored. The present study aimed to compare the clinical outcomes of RFA and hepatectomy plus PCDV and splenectomy. The ndings might help determine appropriate treatment choices for patients with cirrhosis having HCC and EVB.

Patients
Between October 2008 and March 2018, 426 patients with cirrhosis underwent PCDV plus splenectomy in our medical center. Further, 46 eligible patients diagnosed with both HCC and EVB for portal hypertension were included in this retrospective study. The exclusion criteria were as follows: (i) patients with other serious concurrent illnesses expected to decrease expectancy; (ii) patients who received both the treatments (RFA and hepatectomy) for HCC; and (iii) patients who were found to have extrahepatic metastasis and/or macroscopic vascular invasion on preoperative examinations (ultrasonography, enhanced computed tomography, and magnetic resonance imaging). HCC diagnosis was based on AASLD or Chinese criteria8, 16 or histology. Further, 26 patients underwent hepatectomy combined with PCDV plus splenectomy, and 20 patients underwent RFA combined with PCDV plus splenectomy.
The study was approved by the ethics committee of Beijing Ditan Hospital, Capital Medical University. All patients signed informed consent before treatments. All the data of this study come from the research which registered in the Chinese Clinical Trial Registry (2/11/2018, ChiCTR1800019265, all the data were available on http://www.chictr.org.cn/showproj.aspx?proj=32565). Details of the trail is provided in the research protocol.

Preoperative assessments
The data on baseline characteristics were collected prior to surgery, including clinical characteristics and laboratory values, such as white blood cell count, red blood cell count, platelet count, lymphocyte count, neutrophilic granulocyte percentage, hemoglobin content, alanine aminotransferase level, aspartate transferase level, gamma-glutamyl transpeptidase level, cholinesterase level, total bilirubin level, albumin level, creatinine level, prothrombin time, alpha-fetoprotein (AFP) level, and so on. Trans-abdominal ultrasonography values, such as portal vein diameter and spleen thickness, were also determined. Child-Pugh score and BCLC stage were recorded. All patients underwent an endoscopic examination before surgery.

Surgical and RFA treatments
Conventional open surgery was performed in all patients, including curative wedge hepatectomy and PCDV plus splenectomy. PCDV and splenectomy were performed as previously described. Brie y, the gastric and esophageal branches of the gastric coronary vein were disconnected after the routine splenectomy. The esophageal branch was disconnected and ligated. Then, the gastric posterior vein and the left subphrenic vein were ligated. Finally, arteries accompanied by the veins (including the left gastric artery, left gastroepiploic artery, gastric posterior artery, and left subphrenic artery) were disconnected, as previously reported 17,18 . Daily injection of low-molecular-weight heparin over the rst week after surgery and oral administration of aspirin for 4 weeks were performed for prophylactic anticoagulation to prevent portal and splenic vein thrombosis.
RFA and hepatectomy were performed under the Chinese criteria: a single HCC smaller than 5 cm or no more than three tumor nodules, each smaller than 3 cm in diameter. Curative hepatectomy was de ned as the complete removal of all gross lesions of tumor-free margins. The diagnosis of HCC was con rmed after the review of a senior pathologist. RFA was performed by the open or percutaneous approach, immediately before or after PCDV plus splenectomy. Dynamic enhanced CT and/or MRI were used to assess the response about 1 month after the therapy. Repeated ablation therapy was used for patients with a residual tumor. Patients were followed up after a complete response was achieved. Patients were free to choose the treatment option they preferred.
All methods were performed in accordance with the relevant guidelines and regulations.

Follow-up evaluations
The follow-up was initiated after discharge from the hospital. Routine blood count, liver and kidney functions, AFP, and dynamic contrast-enhanced CT or MRI were performed on all of the patients in the rst month after the surgery, and then every 3 or 6 months based on the CT/MRI and AFP. An endoscopic examination was performed every 6 months. The follow-up ended at patients' death or in May 2020. Perioperative mortality or complication was de ned as death or complication occurring within 1 month after the surgery. The primary end-point was death. The second end-point was 3-and 5-year survival.

Statistical analysis
Statistically analyses were performed using SPSS 19.0 (IBM Corp., NY, United States). Quantitative data with normal distribution were analyzed using the Student t-test. Data without normal distribution were analyzed using the Mann-Whitney U test. Qualitative data were analyzed using the χ2-test. Univariate survival analysis was performed using Kaplan-Meier statistical analysis. Time-to-event curves between different groups were compared using the log-rank test. The signi cance levels were set to Two-sided P values <0.05.
Recurrence-free survival was calculated from the day of surgery to the day of any recurrence or death due to any cause. If recurrence did not occur, patients were censored on the date of death or the last followup.

Baseline characteristics
The characteristics of 46 patients 38 male and 8 female, with an average age of 50.48 ± 8.47 years (range: 34-68 years)] were investigated; 20 underwent RFA and simultaneous PCDV plus splenectomy, while the remaining 26 underwent hepatectomy and PCDV plus splenectomy. The demographic and clinical characteristics of the patients are shown in Table 1. The amount of blood loss and intraoperative blood transfusion was higher in the hepatectomy group. However, the tumor size, differentiation status, surgery time, Child's score, and so on did not differ between the RFA and hepatectomy groups.
The RFA and hepatectomy groups comprised 7 and 15 recurrence-rebleeding-free patients. The survival rate in recurrence-rebleeding-positive and recurrence-rebleeding-free groups was similar (P = 0.225, Fig.   D). Four patients died in the recurrence-rebleeding-free group; all of them were HBsAg-positive and/or HBV DNA-positive men aged more than 40 years and died of liver failure.

Discussion
Patients with cirrhosis having HCC and EVB are di cult to treat with monotherapy. The tumor, hepatic function, and secondary prevention of variceal bleeding should all be considered in the therapeutic schedule. The present study showed that RFA/hepatectomy combined with PCDV plus splenectomy had similar effects on tumor recurrence control, rebleeding control, and overall survival for patients with cirrhosis having HCC and EVB. No patient died during the perioperative period in both groups. The perioperative complications were similar, too. PVT was the major complication.
Liver resection plus Hassab's operation could be safely performed in selected patients with HCC and EVB. Concomitant Hassab's operation improves the long-term prognosis 19 . PCDV plus splenectomy is an effective and simple method for hemostasis and widely used in China. EVB, thrombocytopenia, and/or leukopenia can be resolved at the same time 20 , without the disadvantage of hepatic encephalopathy 21 .
The combination of splenectomy and PCDV decreased portal pressure 22 and led to a consequent decrease in the severity of varices 23,24 .
Surgical treatment is the most important and best treatment choice for patients with HCC. Hepatectomy is a curative approach, especially in stage Ia, Ib, and IIa patients. However, patients suitable for hepatectomy are limited. Hepatic failure is often seen in patients after the resection of excessive liver volume. Patients even showed good hepatic reserve function before the surgery, and the 5-year recurrence and metastasis rate after hepatectomy was up to 40%-70% 8 . RFA provided a curative outcome in earlystage patients and had a comparable survival rate as hepatectomy. Studies demonstrated that the recurrence rate was similar after the two treatment modalities 25 . The 1-, 2-, and 3-year recurrence rates were 10.7%, 18.4%, and 24.6%, respectively, in 65 patients of the resection group, and 8.5%, 19.1%, and 23.4%, respectively, in 47 patients in the RFA group 18 . However, other studies showed that the disease-free survival and recurrence-free survival rates were better in the hepatectomy group than in the RFA group.
The clinical outcomes of RFA were not comparable to those of surgery 26 , 27 , 28 . In the present study, the 1-, 2-, and 3-year recurrence rates were 0%, 19.23%, and 23.08%, respectively, in resection group and 15.00%, 30.00%, and 45.00%, respectively, in the RFA group. The same trend of lower recurrence rate was observed after hepatectomy compared with RFA. Early recurrence seemed more often in the RFA group, and late recurrence occurred more in the resection group. Early recurrence was generally associated with tumorrelated factors. It was hard to distinguish minimal tumor residual from recurrence among patients with early recurrence in the RFA group. However, late recurrence might be due to de novo chronic liver disease, considering viral loads, baseline HBsAg level, in ammation activity, and brosis degree 29,30 . All laterecurrence patients (recurrence-free period ≥3 years) in the present study had the baseline HBsAg level of ≥250 IU/mL. Some studies showed that the overall survival in the RFA and surgical resection groups was the same 18,31,32 . The 1-year overall survival was more than 90%, the 2-year overall survival was about 80%-90%, but the 3-year overall survival was 60%-80%. The low survival was likely due to the hepatic insu ciency and tumor recurrence. Studies enrolling patients with better liver function (Child's A) and smaller and single tumors showed better long-term survival rate. In the present study, the 3-year overall survival was lower (50%-60%) than previously reported, especially in the hepatectomy group. However, only one patient died of variceal bleeding, two of tumor recurrences, and the other four of liver failure.
Severe portal hypertension was not the main reason of high mortality in the present study; deterioration of liver function might be the reason.
A previous study included 203 patients undergoing open splenectomy and esophagogastric devascularization. Three patients had variceal rebleeding during the perioperative period, and seven patients rebled during the follow-up period. The total rebleeding rate was 4.93% 33 . Another retrospective study retrieved 283 patients with Child's A or B class cirrhosis with portal hypertension and recurrent variceal bleeding. The perioperative rebleeding rate was 3.2%, and the total rebleeding rate during the follow-up period (29 months) was 4.6% 34 . In the present study, no one rebled during the perioperative period, but 6 of 46 patients had variceal rebleeding during the follow-up period (2 in the RFA group and 4 in the hepatectomy group). The rebleeding rate was 13.04%, much higher than that reported in previous studies. The complication PVT should not be the main cause of rebleeding because only one of six patients who rebled had PVT and a much longer follow-up time (12-129 months vs 4-75 months), hepatectomy might cause secondary portal hypertension 35 , and tumor in ltration might aggravate portal hypertension 36 .
Seven patients died during the follow-up period; two died of HCC, one died of EVB, and the other four recurrence-rebleeding-free patients died of liver failure (HBsAg-positive and/or HBV DNA-positive men aged more than 40 years). HBsAg-positive and/or HBV DNA-positive adults were at high risk of HCC 37 .
Older age and ongoing HBV replication with persistent or intermittent detection of HBV DNA have been found to be related to cirrhosis progression 38 . An EUROGEP cohort study found a linear and slow increase in the incidence rate of HCC, hepatic decompensation, and liver-related death in HBsAg-positive patients; the most common cause of death or liver transplantation in these patients was liver failure (53%), compared with HCC (35%) and non-hepatic causes (12%) 39 . Active HBV replication was considered to be a negative prognostic factor 40,41 . The 5-year survival rate of HBV-DNA-positive patients was much lower than that of HBeAg-negative/HBV-negative patients with compensated cirrhosis (79% vs 86% vs 97%, P = 0.01) 33 . The risk of decompensation and mortality was 4.0-fold and 5.9-fold in DNA-negative patients 33 . The variceal bleeding and HCC might be cured by synchronous splenectomy and hepatectomy or RFA, but decompensation persisted. Hepatitis B virus could be reactivated in patients with HCC undergoing hepatectomy or RFA 42 . Hepatic blood in ow occlusion during hepatectomy might in uence postoperative liver regeneration 43 . The reactivation of virus, reduction of blood supply to the liver, and the loss of functioning liver parenchyma all contributed to liver failure.
This study had some limitations. First, the data were obtained from a single center, and the study had small sample size. A multicenter study with more patients should be performed to draw a more conclusive statistical statement. Second, the retrospective design of the study suffered from different degrees of bias; differences in patients' baseline status might in uence the clinical outcomes. Third, very few patients had a moderate-to-severe impaired liver function; the safety and effectiveness in these patients should be validated.
In conclusion, hepatectomy or RFA with simultaneous PCDV plus splenectomy might be a safe and effective treatment choice for patients with cirrhosis having hepatocellular carcinoma and EVB. A trend of lower recurrence rate was observed after hepatectomy compared with RFA combined with PCDV. "Hepatectomy rst" strategy may be considered for enlisting liver transplantation or other appropriate salvage treatments for patients at high risk for recurrence. We should pay more attention to the background liver diseases, besides variceal rebleeding and tumor recurrence.  Note: Data accorded with normal distribution was expressed in mean±standard deviation. Data not accorded with normal distribution was expressed in median with interquartile range.