Tumor recurrence is the major cause of death following resection of hepatocellular carcinoma (HCC) both in cirrhotic (1) and non-cirrhotic livers (2). Without any form of treatment most patients with recurrent HCC will die within one year (5). Hepatic resection is the only treatment for HCC that can be curative. Therefore an hepatectomy should be considered for any patient with tumor recurrence, provided a repeat resection is feasible and radical.
Frequency of recurrence
The frequency of tumor recurrence depends on the time of follow-up. In different series between 43% and 65% of the patients had recurrences within 2 years of removal of the first tumor (4, 6, 9, 13), and up to 85% within 5 years (14–16). The disease is confined to the liver in 30% to 91% (2, 4, 7, 9–11, 16). However, up to 50% of the intrahepatic recurrences are of multiple or diffuse pattern (1). Synchronous and metachronous multicentric carcinogenesis, are a common feature of HCC (particularly in cirrhosis), intrahepatic seeding and residual satellite tumors after surgery are different patterns which may explain intrahepatic recurrence (11–14). Extrahepatic recurrence predominantly in the lungs and bones (with or without intrahepatic recurrence) is found in up to 13% to 26% (7, 16) of recurrences.
Several factors have been involved in HCC recurrence according to the different series: presence of cirrhosis (1), alcohol abuse (6), size of the tumor (2), the number of nodules (2, 5), preoperative AFP (13), vascular involvement (2, 3, 6, 9, 17), intrahepatic metastases (1, 3, 17), absence of capsule formation (1, 2, 17), tumor at the resectional margin (4, 14, 16), and DNA aneuploid content (6).
Diagnosis of recurrence
One of the main factors that have enabled repeat hepatectomies to be carried out, as well as other forms of treatment thereby improving the outcome of patients with HCC, has been early detection during follow-up. Abdominal Ultrasound (US) is in our experience (2) particularly useful for diagnosing recurrence since it is easy to perform and noninvasive. US is a basic tool in our follow-up schemes together with serum AFP level. Our follow-up program includes clinical examination, US, liver function tests, serum AFP level and periodic abdominal and thoracic CT scan. Magnetic resonance imaging or angiography are only used for subsequent confirmation of the diagnosis of recurrence or as part of the preoperative evaluation. As most of the recurrences occur within the first 2 or 3 years (2, 7), close follow-up is required during this period, but evidence of late recurrence implies that follow-up should be continued for up to 10 years (2, 17). These principles are followed by most authors (6, 7, 11–15), and when the differential diagnosis from cirrhotic liver is considered, even needle biopsy (10) may be necessary.
Prevention of recurrence
Although no effective postoperative adjuvant treatment has been developed, in order to reduce the recurrence rate, some authors suggest the use of other forms of treatment postoperatively, particularly adjuvant chemotherapy. Different protocols have been described based on systemic doxorubicin and regional cisplatinum as an emulsion (15), intra-arterial 5-FU, adriamycin, and mitomycin C and intra-arterial or intra-venous epirubicin (17) resulting in improvement in survival rates (1).
Embolization or chemoembolization used in most cases for non operable patients (4, 13, 15, 17), may be used to confirm the diagnosis of recurrence, and in some cases these techniques may reduce the size of the tumor making it resectable (2). Other nodules not previously seen (either during the procedure or on a CT scan performed 4 weeks later to demonstrate persisting lipiodol fixation) would mean that resection should be ruled out in these cases.
Repeat liver resection
Assessment of liver resection should be done in terms of resectability, operative mortality, morbidity and survival.
Yet when dealing with these patients, one must remember that two different populations are involved, those in whom the tumor arises in a cirrhotic liver (even though for repeat hepatectomies a good preoperative liver function, classified as Child grade A, is necessary), and those who have non cirrhotic livers. The former have higher operative mortality (10%) (2), lower regeneration capacity (1), and a higher intrahepatic recurrence rate than the latter (probably due to the presence of more cases of multicentric tumors and more cases of positive surgical margins as a result of limited resections due to poorer residual liver function) (17).
In most series operative and hospital mortality was 0% (3, 7, 10, 11, 13), ranging from 0% to 8% (4, 5, 8) (the highest mortality was found in a series (8) where deterioration of the Child score, prolonged operating time and increased blood loss were blamed for the end results). This underlines the importance of strict patient selection (satisfactory liver function and limited intrahepatic tumor spread) (7, 10) and sophisticated peroperative management (8).
Complications occurred in 9% to 32% (4, 5, 8, 9, 11, 13, 15) and were mainly due to upper gastrointestinal bleeding (5), bile leak (5, 13), hepatic failure, bacterial peritonitis (11), pleural effusion (5, 13), wound infection (13). Reoperation for these complications is as rare as after the first hepatectomy. The length of hospital stay postoperatively after the second hepatectomy has been reported shorter than after the first hepatectomy (9), probably due to the fact that these patients are very strictly selected.
Survival rates after repeat resection range from 37% to 83% after 3 years and from 37% to 64% after 5 years (3–5, 7–11, 13). A longer interval between hepatectomies predicts improved total survival (4, 7, 13). The survival rate of patients treated by repeat hepatectomy has been reported to be better than the rate of those treated by other palliative methods (5, 7, 11).
A new approach (that our group favors) in dealing with intrahepatic recurrence of the liver involving both lobes, of multiple or diffuse patterns not amenable to resection, particularly in patients with less than three nodules, with tumors less than 3 cm in diameter and without vascular involvement, is total liver resection followed by orthotopic liver transplantation. Complete removal of the liver avoids the risk of synchronous nodules or metachronous nodules. On the other hand when performed for tumors larger than 3 cm and three or more nodules, the tumor may have already spread outside the liver and this may explain the bad results of transplantation in this group (18). The larger series from Asia hardly include any liver transplantation because of the great number of socioeconomic obstacles (15).
All series report their results on repeat partial liver resection. The increase in the number of liver transplantation makes it possible to include this therapy either as the initial or as the repeat hepatectomy.
Experience at the Paul Brousse Hospital
From June 1975 to November 1998, 402 patients underwent hepatic resection for primary liver cancer, twenty two patients of which had partial liver resection as repeat hepatectomy for tumor recurrence. Seven re-resections were carried out on cirrhotic livers (all of them in men) and 15 on non-cirrhotic livers (11 men and 4 women). The tumor was hepatocellular carcinoma in 17 cases, fibrolamellar carcinoma in 2, hepatoblastoma in one and in 2 cases necrotic tissue after chemoembolization was found but the diagnosis was compatible with that of HCC.
The mean time of recurrence after the first hepatectomy was 21 months (4–63) in cirrhotic patients and 31 months (5–99) in non cirrhotic patients.
Before rehepatectomy, residual liver function was assessed by means of the Child Paul Brousse classification (19) - a modification of the Child-Pugh score - and indocyanine green clearance. There were 6 patients (86%) with Child A grade, one patient (14%) with Child B and none with Child C. Two major resections (10%) - that is more than 3 Couinaud's segments - were carried out at rehepatectomy. Segmental liver resection was preferred along with the resection of all the potentially contaminated portal territory.
After a systematic examination of the abdominal cavity to exclude the presence of peritoneal metastases - we start by carrying out complete mobilization and exposure of the liver. Then we make a thorough examination by both palpation and intraoperative US to determine and locate the number and size of any lesions present, thus establishing their relation to the intrahepatic vascular structures. Parenchymal dissection is done using an ultrasonic dissector, with ligation of bile ducts and small vessels.
Repeat hepatectomies were technically more difficult than first hepatectomies. There were multiple adhesions of the raw surface of the previous hepatectomy to adjacent organs, specially to the diaphragm following right liver resections. Regeneration of the liver, in the absence of cirrhosis, led to changes in its shape and in the position of the vascular structures. Previous dissection of the hepatic hilum and vena cava made vascular control more difficult. In cirrhotic patients, portal hypertension and impaired coagulation increased the risk of bleeding during surgery. Mean blood transfusion was 2.0 blood units (5–8) in cirrhotic patients and 1.7 blood units (0–10) in the whole series.
There was no intraoperative or postoperative mortality during the first 2 months.
Postoperative morbidity included minor complications in 6 of the seven cirrhotic patients (85%) and in 5 of the non cirrhotic patients (33%). In five (23%) cases more than one complication was noted. Minor complications included: 2 patients with ascites (8%), 2 patients presenting fever (8%), 2 sub-hepatic collection (8%), pleural effusion was noted in two patients (8%), and general complications (3 cases of urinary infections (13%), one case of gout (4%), one case of lymphangitis (4%), and one infection of intravenous catheter (4%)). The patient with a major complication following surgery suffered burns which required plastic surgery following a preoperative leak in his water-filled thermal blanket. These results compare favorably with those reported in the literature (table I).
Recurrence after repeat partial hepatectomy was 71% after a mean follow up of 56 months. Recurrence was confined to the liver in 8 cases (53%), and extrahepatic disease was found in 6 cases (40%) (lung in 3 cases, bone in 2 patients and peritoneum in one).
Two patients were lost to follow-up. Overall survival in this group is 59% at 5 years and 26% at 10 years. Non cirrhotic patients have 61% survival at 5 years and 25% at 10 years. Cirrhotic patients have a 3 years survival of 62% after re-resection with a limited number of patients and follow-up.
Repeat resection for primary liver tumor recurrence is feasible in patients with recurrence limited to the liver and good liver function. Although technically more difficult, low mortality rates together with an acceptable morbidity are obtained. A survival at 5 years of 59% similar to that after the first hepatectomy, suggests a benefit from a hepatectomy for recurrent HCC. However, resectability rates in these cases are low due mainly to limited residual liver function in cirrhotic patients, and to the pattern of recurrence. Non-resectable forms of intrahepatic recurrence may, in the absence of extrahepatic spread, benefit from liver transplantation. This attitude may broaden the curative possibilities.
Recent developments of in situ destruction techniques (cryotherapy, radiofrequency …) may allow in the near future comparative studies between these methods and the resective approach when facing recurrent primary liver tumors, even with curative intention.
- Bismuth H, Chiche L, Castaing D. Surgical treatment of hepatocellular carcinomas in noncirrhotic liver: Experience with 68 liver resections. World J Surg. (1995);19:35–41. [PubMed: 7740808]
- Kawasaki S, Makuuchi M, Miyagawa S. et al. Results of hepatic resection for hepatocellular carcinoma. World J Surg. (1995);19:31–34. [PubMed: 7740807]
- Neeleman N, Andersson R. Repeated liver resection for recurrent liver cancer. Br J Surg 1996. (1996);83:893–901. [PubMed: 8813771]
- Hu R H, Lee P H, Yu S C. et al. Surgical resection for recurrent hepatocellular carcinoma: Prognosis and analysis of risk factors. Surgery. (1996);120:23–29. [PubMed: 8693418]
- Okada S, Shimada K, Yamamoto J. et al. Predictive factors for postoperative recurrence of hepatocellular carcinoma. Gastroenterology. (1994);106:1618–1624. [PubMed: 8194710]
- Farges O, Regimbeau J M, Belghiti J. Aggressive management of recurrence following surgical resection of hepatocellular carcinoma. Hepatogastroenterology. (1998);45:1275– 1280. [PubMed: 9730388]
- Nagasue N, Kohno H, Hayashi T. et al. Repeat hepatectomy for recurrent hepatocellular carcinoma. Br J Surg. (1996);83:127–131. [PubMed: 8653337]
- Shimada M, Matsumata T, Taketomi A. et al. Repeat hepatectomy for recurrent hepatocellular carcinoma. Surgery. (1994);115:703– 706. [PubMed: 8197562]
- Suenaga M, Sugiura H, Kokuba Y. et al. Repeated hepatectomy for recurrent hepatocellular carcinoma in eighteen cases. Surgery. (1994);115:452–457. [PubMed: 7513088]
- Lange J, Leese T, Castaing D. et al. Repeat hepatectomy for recurrent malignant tumors of the liver. Surg Gyn Obst. (1989);169:119–126. [PubMed: 2547252]
- Chen M F, Hwang T L, Jeng L B. et al. Postoperative recurrence of hepatocellular. Arch Surg. (1994);129:738–742. [PubMed: 7517662]
- Sugioka A, Tsuzuki T, Kanai T. Postresection prognosis of patients with hepatocellular carcinoma. Surgery. (1993);113(6):612–618. [PubMed: 8389492]
- Nagasue N, Uchida M, Makino Y. et al. Incidence and factors associated with intrahepatic recurrence following resection of hepatocellular carcinoma. Gastroenterology. (1993);105:488–494. [PubMed: 8392955]
- Bismuth H, Adam R, Mathur S. et al. Options for elective treatment of portal hypertension in the transplantation era. Am J Surg. (1990);160:105–110. [PubMed: 2368870]
R. Adam, G. Marin-Hargreaves, D. Azoulay, D. Castaing, F. Meriggi, and H. Bismuth.
Adam R, Marin-Hargreaves G, Azoulay D, et al. Repeat hepatectomy for primary liver cancer. In: Holzheimer RG, Mannick JA, editors. Surgical Treatment: Evidence-Based and Problem-Oriented. Munich: Zuckschwerdt; 2001.