Laparoscopic liver resection for hepatocellular carcinoma: Indications and role
Potential conflict of interest: Nothing to report.
Watch a video presentation of this article
Answer questions and earn CME
body surface area
future liver remnant
hepatic venous pressure gradient
laparoscopic liver resection
Model for End-Stage Liver Disease
open liver resection
portal vein embolization
total liver volume
Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and is responsible for more than 500,000 deaths annually. Partial hepatectomy and liver transplantation are the only curative surgical options. With the employment of low central venous pressure anesthesia, more accurate axial imaging sequences for tumor location, and dedicated postoperative management teams, the mortality and morbidity of open liver resection for HCC has decreased dramatically over the last half-century. Five-year survival rates for hepatectomy in HCC approach 50% in most current series.[2, 3]
The use of laparoscopic techniques in the treatment of cancer has grown in acceptance over the last decade, as the initial concerns of comprising oncological principles and port-site metastases in laparoscopic resections have largely proven unfounded. However, the use of laparoscopic techniques in liver resection for HCC has not been fully embraced; the first laparoscopic liver resection for HCC was not reported until 1995. The purpose of this review is to detail the indications for hepatectomy for HCC and the role of laparoscopic liver resection (LLR).
Indications for Hepatectomy for HCC
The tenets for determining resectability of a HCC tumor are based on a combination of liver- and tumor-specific factors, due to tumors often arising in a background of cirrhosis. Liver-specific factors include not only the quantity but also the quality of the future liver remnant (FLR) following hepatectomy. The absolute FLR volume is not only dependent on FLR size but also the patient's body mass and the underlying degree of liver dysfunction. FLR is calculated by three-dimensional axial volumetric measure of liver volume left behind following resection divided by total liver volume (TLV) × 100. The most precise measure of TLV takes into account body surface area (BSA), TLV = −794 + 1267 × BSA. In patients with an otherwise normal liver with limited fibrosis, the risk of postoperative liver dysfunction and morbidity is reduced provided the FLR is >20%. However, in patients with cirrhosis or a high degree of fibrosis, mortality, major complications, and liver failure increase markedly if the FLR is ≤40%. Preoperative portal vein embolization (PVE) is a useful adjunct to not only increase the FLR of the nonembolized side, allowing for a safer resection, but also serves as a measure of potential postoperative hypertrophy. If the FLR fails to hypertrophy after PVE in a cirrhotic liver, the likelihood of postoperative complications from liver insufficiency increases and should be taken as a contraindication to hepatectomy. It is our policy to obtain tumor volumetry to determine the FLR and potentially plan for a PVE if we are resecting more than three segments of the liver, as occurring in a right hepatectomy, where the FLR will be typically ≤40%.
Determining the quality of the underlying liver in preparation for a planned hepatectomy can often be nuanced and varies from patient to patient. The Child-Pugh classification system is the most universally employed system for preoperative stratification. It is universally agreed upon that the postoperative complications from liver insufficiency in patients with Child-Pugh class B or C cirrhosis are prohibitively high enough to make only Child-Pugh class A patients acceptable for surgical resection. However, although the Child-Pugh score is useful in assessing global liver function, there is heterogeneity among Child-Pugh class A patients such that the Child-Pugh score alone is not enough to risk stratify patients undergoing hepatectomy. The use of the Model for End-Stage Liver Disease (MELD) scoring system has been demonstrated to help select the optimal candidates for hepatectomy. Specifically, patients with a preoperative MELD score >10 have 90-day mortality rates approaching 15% to 20%. Significant portal hypertension is generally considered a contraindication to surgical resection. Direct measurement of portal hypertension through an invasive measurement of hepatic venous pressure gradient (HVPG) has been correlated with hepatic insufficiency and postoperative mortality when HVPG is ≥10 mm Hg.[11, 12] However, measurement of HVPG is not commonly used in most large-volume centers due to the relative invasiveness of the procedure and lack of convincing data correlating HVPG and a survival benefit. Most centers rely on noninvasive, indirect measures of portal hypertension, including platelet count. A platelet count of ≤100 × 109/L has been demonstrated to be a surrogate for significant portal hypertension and is associated with an increase in major complications and mortality even when corrected for Child-Pugh/MELD scores and tumor extent.
Tumor-specific factors in determining the suitability of hepatectomy for HCC include tumor size, tumor number, and presence of vascular invasion. Multiple studies have demonstrated that tumor size alone does not correlate with overall survival after hepatectomy for HCC.[14, 15] Provided there is an adequate FLR and no evidence of major vascular invasion, resection of tumors ≥10 cm have shown survival rates similar to those of tumors ≤10 cm. The use of resection in multinodular HCC is more controversial, as published reports have traditionally grouped patients with both multicentric tumors and intrahepatic metastases as one group, with seemingly equivalent survival and recurrence rates. However, in more recent reports, multicentric or tumors related to the “field defect” of underlying liver cirrhosis offer a better prognosis after resection than intrahepatic metastases, typically seen as large tumors with smaller satellite lesions surround it.[16, 17] Major vascular invasion as evidenced by portal or hepatic venous involvement is generally regarded as a contraindication to hepatectomy, with multiple published reports demonstrating early recurrence and low survival rates.[18, 19]
Role of Laparoscopic Liver Resection in HCC
LLR in HCC has lagged behind laparoscopic resections in other cancers, as LLR requires not only the expertise of hepatobiliary surgery but also advanced laparoscopic skills. There has never been a published randomized controlled trial comparing LLR with open liver resection (OLR) in HCC, and only seven published reports, with at least 20 patient in the LLR group, directly compare outcomes after LLR and OLR in the same institution (Table 1).[20-26] Currently, the most common indication for LLR in HCC appears to be small tumors more peripherally located within the liver, allowing for wedge or segmental resections.
Table 1. Published Reports Comparing OLR and LLR for Patients With HCC With at Least 20 Patients in the LLR Group
|Sarpel et al.|| || || || || || || || |
|OLR||56||4.3 ± 2.1||NA||NA||165 ± 53||NA||NA||78%|
|LLR||20||4.3 ± 2.2||NA||NA||161 ± 37||NA||NA||92%|
|Belli et al.|| || || || || || || || |
|OLR||125||6.0 ± 2.3||31.2%||26.0%||185 ± 65||9.2 ± 3.1||4.0%||65% (3 yr.)|
|LLR||54||3.8 ± 1.3||6%||11.0%||167 ± 36||8.4 ± 2.5||2%||67%|
|Lai et al.|| || || || || || || || |
|Tranchart et al.|| || || || || || || || |
|OLR||42||3.7 ± 2.1||11.9%||16.7%||222 ± 46||9.6 ± 3.4||2.4%||48%|
|LLR||42||3.6 ± 1.7||11.9%||9.5%||233 ± 93||6.7 ± 5.9||2.4%||60%|
|Lee et al.|| || || || || || || || |
|Truant et al.|| || || || || || || || |
|OLR||53||3.1 ± 1.2||0%||3.8%||215 ± 89||9.5 ± 4.8||7.5%||46%|
|LLR||36||2.9 ± 1.2||0%||2.8%||193 ± 104||6.5 ± 2.7||0%||71%|
|Hu et al.|| || || || || || || || |
|OLR||30||8.7 ± 2.3||0%||NA||170 ± 32||20 ± 3.2||0%||53%|
|LLR||30||6.7 ± 2.1||0%||NA||180 ± 45||13 ± 2.1||0%||50%|
The paucity of published reports comparing LLR and OLR in HCC tumors is most likely a result of the progression of laparoscopic liver surgery from resection of a benign lesion, to malignant lesions with normal liver parenchyma, to LLR in a background of cirrhosis as seen in HCC. Since the majority of the published reports originated in the last 3 years, it is fully expected that as hepatobiliary surgeons become increasingly more facile with advanced laparoscopic techniques, additional larger series will be reported. With careful selection of patients, making sure both liver- and tumor-specific factors are optimized, a significant difference in terms of survival measures between OLR and LLR will be very low, making the use of a prospective, randomized controlled trial cost-ineffective due to the large number of patients that would be required to show a significant survival difference.
Despite apparent equivalence in long-term survival with OLR, LLR has been demonstrated to offer significant short-term clinical benefits such as less postoperative pain, shorter length of stay in the hospital, and earlier return of functional status. As technological advances in minimally invasive surgery have been made and long-term clinical equipoise demonstrated, the remaining unsolved issue is whether the short-term clinical benefits of LLR translate into a fiscal advantage of LLR over OLR. Bhojani et al. demonstrated in a dedicated LLR program at the University of Toronto that LLR offered not only the short-term clinical advantage of reduced operating room time and length of stay but that this translated into decreasing overall costs compared with OLR. Obviously, the limitation with this study, as with all retrospective studies comparing LLR with OLR, is that no matter how the patients are matched or what factors are controlled for, a significant selection bias is present.
Determining the resectability of HCC tumors is dependent on both liver- and tumor-specific factors. If these factors are optimized and local expertise in both hepatobiliary and advanced minimally invasive surgery is present, LLR is equivalent to OLR in long-term survival in selected patients. LLR may also have a short-term clinical benefit of shorter length of stay and postoperative return to function that may translate into a financial benefit of LLR to OLR. Randomized controlled trials comparing LLR with OLR are most likely not feasible due to the large costs of accruing such a large sample size to demonstrate a benefit and the relative acceptance of LLR for small, peripherally located tumors.