Yilei Mao and Shunda Du Abstract Postoperative liver failure still remains a major cause of mortality after partial hepatectomy, which results from an insuffi cient functional remnant liver. Therefore, the accurate evaluation of liver function is very important, particularly in cirrhotic patients who require hepatectomy. Traditional tests, such as serological indicators, Child-Pugh score, MELD score and ICG clearance test, are important in predicting and reducing the risks of hepatectomy. However, these tests only provide functional data on the entire liver, not on specifi c anatomic parts of the liver. Ideally, assessments of liver function should include both anatomical information and function of the whole and partial liver, providing reliable information for accurate evaluation of surgical risks. 99m Tc-galactosyl serum albumin scintigraphy, can assess the liver function quantitatively. It combined with single photon emission computed tomography, CT and three-dimensional reconstruction, may be a better measure of liver function, especially of remnant liver function. Keywords Hepatic reserve • Hepatectomy • Cirrhosis • Galactosyl serum albumin Introduction Liver resection is the accepted gold standard of treatment for liver tumors. Improvements in surgical methods and instruments have greatly reduced the perioperative mortality . However, the major cause of mortality after partial hepatectomy is liver failure , which results from an insuffi cient functional remnant liver mass [ 1 ]. Conversely, the erroneous results of liver function tests may mislead the surgeon to make a wrong decision such as precluding some patients with large liver tumor s from undergoing surgery , even if surgery is benefi cial. Therefore, the accurate Y. Mao (*) • S. Du Department of Liver Surgery , Peking Union Medical College (PUMC) Hospital , 1# Shuai-Fu-Yuan, Wang-Fu-Jing , Beijing 100730 , China e-mail: [email protected]; [email protected] 166 evaluation of liver function is very important, particularly in patients with damaged livers who require hepatectomy or liver transplant ation [ 2 ]. Liver function includes the uptake, metabolism, conjugation and excretion. Among the methods used to evaluate liver function in practice are serological tests which are the earliest and most commonly used in determining whole liver function. Clinical scoring systems, such as Child-Pugh and model for end-stage liver disease ( MELD ) scores can roughly evaluate the risks of hepatectomy . The indocyanine green (ICG) clearance test is a widely used quantitative test of liver function in patients who scheduled for major hepatectomy. Although these tests can assess whole liver function, they cannot assess remnant liver function and predict the risk of liver failure post-operation. Computed tomography ( CT ) volumetry can provide anatomic information on remnant liver volume (RLV), but anatomic volume is not equal to functional volume, especially in patients with cirrhosis . In recent years, 99m Tc- galactosyl serum albumin ( 99m Tc-GSA) scintigraphy combined with single photon emission computed tomography (SPECT) and CT with three-dimensional imaging, is relatively accurate in measuring the whole and regional liver function. 99m Tc-GSA scintigraphy may therefore be a promising method to plan surgical incisions and to predict operative risk. Based on a two-compartment kinetic model, a novel system was developed that provides 3D functional evaluation for any anatomical component of liver, and hepatectomy simulation with a freehand drawing tool. The result was showed by the parameter ‘UI’ which had high accuracy in predicting the risk of liver failure. In the future, many new methods will be established which can assess hepatic reserve accurately prior to liver resection in the cirrhotic patient. Search Strategy A literature search of English language publications since January, 2004 was used to identity published data on preoperative assessment of hepatic reserve in cirrhotic patients undergoing hepatectomy using the PICO outline (Table 14.1 ). Databases searched were PubMed, Embase, and Cochrane Evidence Based Medicine . Terms used in the search were “cirrhotic patients/ cirrhosis ”, “ liver resection / hepatectomy”, “liver function/hepatic reserve/ Child-Pugh Score /indocyanine green clearance test (ICG)/model for end-stage liver disease ( MELD ) score/ Monoethylglycinexylidide (MEGX) test/galactose elimination capacity Test/computed tomography volumetry/galactosyl serum albumin (GSA)/transient elastography (TE)”, “postoperative complications/postoperative hepatic failure/ ascites / hyperbilirubinemia /prolongedprothrombin time/length of stay/ mortality / morbidity / quality of life ”, and “preoperative/prior to liver resection ”. Eleven cohort studies, two systematic reviews and one meta-analysis, and four review articles were included in our analysis (Table 14.2 ). The other perspective cohort study [ 3 ] was enrolled about GSA which was accepted by the Annals of Surgical Oncology. The data was classifi ed using the GRADE system. Y. Mao and S. Du 167 Results Liver function includes the uptake, metabolism, conjugation and excretion. The serological tests are the earliest and most commonly used and still play important role. But any one serological indicator can show only one aspect not comprehensive function, and whole liver function not local. So different clinical scoring systems and metabolic quantitative liver function tests were developed to assess hepatic reserve . The Child-Pugh Scoring System The Child scoring system, fi rst proposed in 1964, was originally developed to predict the outcome of cirrhotic patients undergoing surgical therapy for portal hypertension. This system was modifi ed by Pugh et al. [ 4 ] in 1973, and called the Child-Pugh score. It includes total plasma bilirubin level, plasma albumin level, and prothrombin time together with the presence or absence of encephalopathy and ascites . Of all the tools for assessing liver function, the Child-Pugh system is simple but very useful [ 5 ]. It is widely used in hepatocellular carcinoma and cirrhosis patients, who will undergo resection or transplantation. Thus Child-Pugh is more relevant for liver resection s, compared with MELD score system. A classifi cation of grade A of the Child-Pugh grading system is a typical indication for liver resection. And liver transplant ation is selected if oncological indications meets the established criteria [ 6 ]. Schneider showed that, for patients classifi ed Child-Pugh A, the mortality is minimal at <5 % while for grade B cirrhotics the 1-year liver failure -related morality is almost 20 %, and for grade C cirrhosis is 55 %[ 7 ]. However, the Child-Pugh grading system only provides a rough evaluation for global liver function reserve, so more quantitative liver function tests may need for preoperative assessment . Table 14.1 PICO table for perioperative assessment of hepatic reserve in the cirrhotic patient P (Patients) I (Intervention) C (Comparator group) O (Outcomes measured) Cirrhotic patient undergoing liver resection Novel preoperative liver function test, such as: indocyanine green(ICG) clearance test, model for end-stage liver disease ( MELD) score, transient elastography, 99m Tc- galactosyl serum albumin scintigraphy, etc Classical preoperative liver function test, such as: Child-Pugh score, model for end-stage liver disease ( MELD) score, indocyanine green(ICG)clearance test, computed tomography ( CT) volumetry, etc Postoperative complications, mortality 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… 168 Table 14.2 Incidence of different liver function assessments and clinical outcomes Author (year) N Age Child- Pugh score MELD score ICGR 15 TE (kPa) GSA LOS (d) Ascites Postoperative LF Mortality Study type ( quality of evidence) Matteo (2012) 90 64 A:90 % 8.3 NR 16.2 NR 9 20 % 28.9 % 2.2 % Prospective cohort (low) Other: 10 % Jeff (2013) 105 59 5 NR 4.2 % 9.4 NR NR NR NR 1.9 % Prospective cohort (low) Scheingraber (2008) 95 60 NR NR ICG-PDR was used NR NR NR NR 24.2 % 3.2 % Prospective cohort (low) Ohwada (2006) 75 63 A: 96 % NR ICG -k and ICGR 15 NR NR NR NR 11 % 1 % Prospective cohort (low) B: 4 % James (2013) 28 59 A: 96.4 % 7.2 11.8 % 10.2 NR 6 4/28 NR / Prospective cohort (low) B: 3.6 % Hirohisa (2014) 548 66 NR NR 12.5 % NR LHL15: 0.92 NR NR NR 0.89 % Prospective cohort (low) Cucchetti (2006) 200 64 5 8.8 NR NR NR NR 5 % 7.5 % NR Prospective cohort (low) Cescon (2009) 466 64 5.4 8.9 NR NR NR NR 4.9 % 4.9 % NR Prospective cohort (low) Kwon (2006) 178 62 A: 73 % NR GSA- Rmax and ICGR15 values correlate well NR GSA-RL NR 13.5 % NR 1.12 Retrospective cohort (low) B: 27 % Kaibori (2008) 191 66.5 A: 86.4 % NR 18.0 %: 9.7 % NR HA/ GSA- Rmax ratio NR NR 8.38 % 1.57 % Prospective cohort (low) B: 13.6 % Mao (2014) 142 53.3 A: 76.1 % NR Negative associate with UI NR UI NR UI can distinguish UI can predict NR Prospective cohort (low) B: 21.1 % C: 2.8 % MELD model for end-stage liver disease, TE transient elastography, GSA 99m Tc- galactosyl serum albumin scintigraphy, LOS length of stay, NR not reported, LHL the ratio of uptake by the liver to that by the liver and the heart at 15 min in GSA, LF liver failure Y. Mao and S. Du 169 The Model for End-Stage Liver Disease ( MELD ) Score The limitations of the Child-Pugh score led to the development of MELD . MELD score was originally developed to evaluate the survival rate of patients undergoing transjugular intrahepatic portosystemic shunt procedures, and was thereafter modifi ed to evaluate patients with liver disease undergoing surgery . MELD score is a constellation of serum bilirubin, creatinine concentration, INR and etiology of liver disease, and is calculated using the formula: 11.2 × Ln(INR) + 9.57 × Ln[creatinine( mg/dL)] + 3.78 × Ln [bilirubin(mg/dL)] + 6.43 × (etiology: 0 if cholestatic or alcoholic, 1 otherwise), with the score rounded to the nearest integer [ 8 ]. The MELD score is used to allocate organs for liver transplant ation [ 9 , 10 ]. The application of this system to determine organ allocation reduced 15 % of the mortality rate in liver transplant candidates [ 11 ]. Cholangitas et al. [ 9 ] stated that MELD score was shown to be useful for the prediction of long-term survival in patients with cirrhosis . Ascites, jaundice , prolonged prothrombin time, increase of serum creatinine and bilirubin levels, and decrease of albumin serum level are typical markers of impaired liver function. Cucchetti et al. has showed that MELD score can be used to predict the development of post-operative liver failure after hepatectomy for patients with cirrhosis undergoing resection of hepatocellular carcinoma, with a pre-operative score of 11 being associated with a poor outcome [ 12 ]. In subsequent clinical applications, outcomes were different in patients with the same score and different serum concentrations of sodium. So some modifi ed MELD formulas that have been proposed to predict the prognosis of liver disease, such as MELD-Na score, integrated MELD (iMELD), MELD to sodium (MESO), United Kingdom end-stage liver disease (UKELD), etc. However, they cannot accurately predict the actual survival time of patients undergoing hepatectomy . At present, they are mainly used to assess the severity and prognosis of chronic liver diseases, and to evaluate the patients awaiting liver transplant ation [ 13 ]. Computed Tomography ( CT ) Volumetry At present, CT volumetry is the most often used imaging method to determine whether hepatectomy can be performed safely. Pre-operative estimations have been shown to correlate well with actual volumes resected. But the safety limit for the remnant liver volume in patients with normal liver remains controversial. Kubota et al. found that resections of 60 % of non-tumorous liver was possible in patients with normal livers [ 14 ]. Shoup et al. stated that a liver resection can be safely performed if the functional remnant liver volume(RLV) is larger than 25–30 % when using computed tomography volumetry [ 15 , 16 ]. If the patients have underlying liver disease, then a margin of 40 % is taken into account [ 17 ]. Several studies found that in the presence of cirrhosis , a resection of >2 segments should only be performed of the estimated remnant functional liver was >40 %, while if it’s <40 %, a pre-resection portal vein embolization (PVE) should be advised [ 18 , 19 ]. 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… 170 However, if the patient has a compromised liver, then the liver volume does not truly refl ect liver function [ 20 ]. CT volumetry is used for preoperative calculations of the volume of resected livers, but does not demonstrate the effects of diseased liver parenchyma on liver function. The evaluation of liver function before liver surgery is dependent on the combination of the results of CT volumetry with those of other liver function tests. Transient Elastography Recently, noninvasive measurements to assess the degree of liver fi brosis and cirrhosis before operation, like transient elastography, acoustic radiation force impulse imaging and magnetic resonance elastography, have been developed. The clinical studies are ongoing to validate the strength and the power of these novel approaches [ 21 ]. Transient elastography (TE) measured by FibroScan is a rapid, non-invasive, and reproducible method for measuring liver stiffness that is increasingly explored to assess liver fi brosis. It measures the velocity of a low-frequency (50 Hz) elastic shear wave propagating through the liver. This velocity is directly related to tissue stiffness, called the elastic modulus. The stiffer the tissue, the faster the shear wave propagates. TE measures liver stiffness in a volume that approximates a cylinder that is 1-cm wide and 4-cm long, 25–65 mm below skin surface. The results are expressed in kilopascals (kPa) and range from 2.5 to 75 kPa; a normal value is around 5 kPa [ 22 ]. Several advantages of TE have been reported, such as low invasiveness, a short procedure time (5 min), fast acquisition of results, and portability that enables testing at the bedside and in outpatient departments [ 50 ]. Although unreliable and unrepeatable measurements caused by host obesity, anatomical diffi culties such as a narrow intercostal space, and inadequate operator experience have also been reported, the overall diagnostic accuracy for advanced liver fi brosis and early cirrhosis is up to 90 % in various liver diseases including chronic viral hepatitis and nonalcoholic fatty disease [ 23 ]. To evaluate the effi cacy of preoperative assessment of liver fi brosis and cirrhosis using TE in predicting post- hepatectomy outcomes , several clinical studies has been carried out. In a prospective cohort [ 24 ], 90 patients undergoing hepatectomy for HCC were prospectively evaluated with FibroScan. Postoperative liver failure (PLF) occurred in 28.9 % of patients and receiver operating curves (ROC) analysis identifi ed patients with liver stiffness value higher than or equal to 15.7 kPa as being at higher risk of PLF, while patients with liver stiffness value lower than 14.8 kPa had no PLF. Multivariate analysis showed that along with low preoperative serum sodium levels (P = 0.012), histological cirrhosis (P = 0.024), elevated liver stiffness (P = 0.005) was an independent predictors of PLF. In a larger prospective cohort [ 25 ], 105 with a mean age of 59 years were included with both ICG retention rate at 15 min and TE were prospectively carried out. Using the calculated cutoff at 12.0 kPa, liver stiffness measurement was shown to have sensitivity of 85.7 % and speciY. Mao and S. Du 171 fi city of 71.8 % in the prediction of major postoperative complications. On ROC, only liver stiffness measurement but not ICG showed signifi cant correlation with major postoperative complications. The Indocyanine Green (ICG) Clearance Test ICG is a highly protein-bound, water-soluble, tricarbocyanine dye that bounds in plasma to albumin and β-lipoproteins and distributes uniformly in the blood within a few minutes after injection. It is selectively taken up by hepatocytes with a plasma extraction of 70–90 % and is excreted unchanged in the bile via a carrier-mediated mechanism. Therefore, it refl ects several liver functions, including the blood fl owdependent clearance and transporter functions [ 26 ]. The standard procedure involves a bolus injection of 0.5 mg/kg of ICG following an overnight fast, and blood samples are collected at 5-min intervals for 20 min. ICG concentrations are measured using a spectrophotometer. The ICG clearance test can also be automatically calculated under a dye densito-graph (DDG) analyzer using an optical sensor placed on the fi nger pulse [ 27 ]. The machine expands the application of ICG clearance test in current clinical situation. The results of ICG clearance test can be expressed in several ways, including the plasma disappearance rate (ICG-PDR), the ICG elimination rate constant (ICG- k ) and the ICGR 15 which describes the percentage of circulatory retention of indocyanine green during the fi rst 15 min after bolus injection [ 28 ]. In order to prospectively determine the effi cacy of ICG-PDR in the clinical course, 95 patients undergoing liver resection were included in a cohort [ 29 ], with ICG-PDR, bilirubin and prothrombin time selected and prospectively measured. After hepatectomy , 3 patients died due to liver failure and 21 patients developed signs of liver dysfunction. ROC analysis revealed that ICG-PDR did signifi cantly better indicate postoperative liver dysfunctions. Of date, pulse spectrophotometry was developed to noninvasively measure the ICG- k and a prospective clinical study was done [ 30 ]. Seventy fi ve patients who underwent anatomical liver resection for hepatocellular carcinoma were enrolled and ICG- k was measured instantaneously using pulse spectrophotometry before surgery , during infl ow occlusion and after hepatectomy. Eight patients suffered liver failure with one died in hospital. In a logistic regression model, the estimated remnant ICG- k was a signifi cant predictor of postoperative liver failure and real-time monitoring of ICG- k was shown to be helpful for evaluating the remnant liver functional reserve before, during and after hepatectomy. ICGR 15 , as the most commonly determined value, has been extensively investigated in various kinds of clinical setting and incorporated into a number of test combinations or score systems. A decision tree for deciding the safe limit of hepatectomy was developed [ 31 ] basing on three variables: whether ascites is present, the serum total bilirubin level, and the ICGR 15 . With strict application of this decision tree to 1,429 consecutive hepatectomy in 10 years, only one patient death was encountered. So ICGR 15 > 15 % is a high risk factor for serious post-hepatectomy 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… 172 complications [ 32 ], although a cutoff of 14 % has been suggested by Lau et al. [ 33 ]. ICGR 15 , along with TE, was performed preoperatively in 44 patients with hepatocellular carcinoma [ 34 ]. ICGR 15 was found to correlate well with preoperative factors and postoperative outcome (peak AST level). A classifi cation system for liver function using ICGR 15 and the ratio of uptake by the liver to that by the liver and heart at 15 min (LHL15) in 99m Tc-galactosyl human serum albumin scintigraphy for hepatic resection , was created [ 35 ]. A total of 548 consecutive patients who underwent hepatectomy were enrolled in a prospective study to validate the ranking system and the result confi rmed the usefulness of this system in predicting the safety of hepatic resection . 99m Tc-Galactosyl Serum Albumin Scintigraphy Molecular nuclear imaging techniques have developed these years. Some new agents, such as 99m Tc- galactosyl serum albumin scintigraphy (GSA) and 99m Tcmebrofenin hepatobiliary scintigraphy, can measure both total and future remnant liver function and potentially identify patients at risk for postresectional liver failure . GSA is an analogue of asialoglycoprotein, which binds to asialoglycoprotein receptors (ASGPR) on hepatocyte membranes, followed by receptor-mediated endocytosis. ASGPR density is closely related to hepatocyte function [ 36 , 37 ]. The level of expression of receptor is signifi cantly related to liver function and lower in diseased livers such as chronic hepatitis, cirrhosis and HCC [ 3 ]. Radio labeled ASGPR was developed originally by Vera et al. [ 38 ]. 99m Tc-GSA is very stable and only distributes in the blood and liver after intravenous injection [ 36 ]. The liver is the only uptake site for 99m Tc-GSA, making 99m Tc-GSA an ideal agent for predicting hepatocyte mass and function by monitoring the functional status and distribution of ASGPR [ 39 , 40 ]. After liver uptake, 99m Tc-GSA remains trapped in the liver for at least 30 min, and there is practically no biliary excretion. Thus, SPECT can assess both liver function and functional volume at the same time [ 41 ]. The 99m Tc-GSA liver uptake ratio (LHL15) and blood clearance ratio (HH15) are quantitative indices frequently used in planar dynamic 99m Tc-GSA scintigraphy. LHL15 defi ned as 15 min after bullet injection of 99m Tc-GSA and calculated by dividing the radioactivity in regions of interest (ROIs) of the liver by the radioactivity in the liver and heart, it represents the number of hepatocytes. HH15 is calculated by dividing the radioactivity in ROIs of the heart 15 min by the radioactivity 3 min after injection of 99m Tc-GSA, it represents the rate of blood clearance [ 42 ]. Harada and his colleagues recently developed a simple software program to automatically calculate the pixel counts of the area between the hepatic curve and heart curve from 3 to 15 min [ 43 ]. Both LHL15 and HH15 refl ect the liver function and the severity of liver disease [ 44 ]. For LHL15 and HH 15 measures preoperative total liver function, not the function of the remnant liver, postoperative liver failure has been observed in patients with normal LHL15 values [ 45 ]. Y. Mao and S. Du 173 LHL15 and HH15 are readily calculated from the radioactivity in the heart and liver ROIs, it may not refl ect the actual liver function. So some complex and perfect compartmental models of 99m Tc-GSA kinetics are developed for the assessment of liver function ([ 46 ] #174, [ 47 ] #30, [ 48 ] #149). Many different parameters can be calculated from different kinetic models for the quantitative evaluation of liver function. The liver blood fl ow and maximal asialoglycoprotein receptor binding rate assessed by 99m Tc-GSA are signifi cantly correlated with other quantitative measures of liver function [ 48 ]. Total ASGPR amount are proportional to the number of viable hepatocytes and the correlation of total ASGPR amount with hepatocyte number was signifi cantly higher than the correlation of ICG-k with total hepatocyte number [ 53 ]. Kwon etc. reported previously that the maximal removal rate of GSA(GSARmax) values correlated well with the results from the transferrin, prealbumin, retinol binding protein, fi brinogen, prothrombin time, hepaplastin test, antithrombin III, and ICG tests [ 49 ]. In another retrospective study [ 50 ], this team reviewed 178 patients for elective hepatectomy . Preoperative estimation of the GSA-Rmax in the predicted remnant liver (GSA-RL) is used a parameter. In this study, seven patients postoperative hyperbilirubinemia were recorded with GSA-RL <0.15 mg/min. Two patients died of postoperative liver failure 1–2 months after surgery , the GSA-RL values were 0.078 and 0.090, respectively. They considered a margin of safety (0.05) and determined 0.15 as the cutoff value. Preoperative percutaneous transhepatic portal embolization should be performed for cases with a GSA-RL less than 0.15 to avoid postoperative hyperbilirubinemia or hepatic failure. In another study, this team [ 51 ] followed 191 patients more than 1 year after hepatectomy with 16 patients suffered from liver failure and 3 of them died. Total 35 clinicopathologic factors were performed to identify independent predictors of postoperative liver failure after resection of HCC by univariate and multivariate analyses. In univariate analyse, elder, a lower serum albumin level, lower cholinesterase level, longer prothrombin time, lower platelet count, and lower GSA-Rmax, higher values of ICGR15, total bilirubin, AST, type IV collagen 7S, hyaluronate (HA), AFP, type IV collagen 7S/GSA-Rmax ratio, and HA/GSA-Rmax ratio, are the factors easy to the postoperative liver failure. Patients in the liver failure group had signifi cantly more intraoperative blood loss and a longer postoperative hospital stay. Multivariate logistic regression analysis showed that HA/GSA-Rmax ratio 500 mg min/dl (OR 23.60; 95 % confi dence interval (CI) 1.91–62.09; P = 0.0138) was the only independent predictor of postoperative liver failure. An increase of the HA/ GSA-Rmax ratio was associated with more severe liver dysfunction. The HA/GSARmax ratio was also positively correlated with various conventional liver function tests, such as the ICGR 15 , AST, total bilirubin, platelet count, albumin, cholinesterase, prothrombin time, type IV collagen 7S, HA and GSA-Rmax, etc. They conclude that the HA/GSA-Rmax ratio can predict postoperative liver failure, and a ratio 500 mg min/dl is a relative contraindication to liver resection with a sensitivity of 88 % and a specifi city of 92 %, and its negative predictive rate was 99 %. Recently, Mao and Du [ 52 , 53 ] set up a computerized image system based on a two-compartment model, which could provide liver images, a freehand drawing 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… 174 tool for hepatectomy simulation, assess liver function and predict postoperative remnant liver function, using uptake index (UI) as a parameter. That study [ 54 ] recruited 71 pre-hepatectomy patients and 71 healthy volunteers. They found that median UI = 2.81 was the normal reference, lower UI values were associated with the more impaired liver functions. ROC analysis indicated that lower UI values could be used to predict the presence of ascites with high accuracy (AUC = 0.88, P < 0.0001). Preoperative UI values were also able to distinguish patients with and without elevated bilirubin (AUC = 0.86, P < 0.0001). Preoperative UI was also negatively associated with ICGR 15 values, i.e., the lower UI value was, the larger ICGR 15 value would be(r = −0.92, P < 0.0001). In this system, for each simulated liver resection plan, the corresponding anatomic and functional remnant liver volume, and the risk of postoperative liver failure were presented. There 33 patients had both preoperative and postoperative measures of UI values for the remnant liver via the system. Regression analysis using predicted UI as an explanatory variable showed a linear equation as: Post Surgery UI = −0.09 + 1.04(Predicted UI). It supported the accuracy of the preoperative prediction. To further evaluate the reliability of predicted UI values for the future remnant liver (FRL), predicted UIs were further compared with the parameters of the actual post operative liver functions tests. The results demonstrated that predicted UI negatively correlated with PT and total bilirubin level (Pearson’s correlation coeffi cient r = −0.67 and −0.68 respectively, P < 0.0001). The AUC for predicted UI to distinguish patients with and without postoperative ascites was at 0.85, P < 0.0001. While Child score of 9 or larger was defi ned as high risk of liver failure, the ROC analysis results indicated that UI values had a high accuracy in predicting the risk of liver failure (AUC = 0.95, P < 0.0001). The threshold for very high risk was defi ned as P = 0.05 which corresponds to UI of 0.73 (FLVI = 26 %). In fact, there are some weak points in this study. Without Child C patients enrolled in the study might lead to conservative decision making rule. The small sample size also might affect the accuracy of the threshold to defi ne the high risk region. Further improving the accuracy and validating the system in phase III clinical trial is needed before bring it to clinical practice. Recommendations Based on the Data The clinical methods to evaluate liver function including serological tests, various evaluation scoring systems, ICG clearance, 3D- CT volumetric calculation are all useful in clinical practice. They all have advantages and disadvantages, and cannot be replaced, currently. The preoperative liver function evaluation must be a comprehensive process. In order to make a safe and thorough evaluation, multiple indices, as well as general condition of the patient, type of planned surgery, and profi ciency of surgeons should be considered and combined. The maturation and application of new GSA based three-dimension imaging system may bring a new promising tool for the preoperative liver function evaluation. Y. Mao and S. Du 175 References 1. Fujii Y, et al. Risk factors of posthepatectomy liver failure after portal vein embolization. J Hepatobiliary Pancreatol Surg. 2003;10(3):226–32. 2. Clavien PA, et al. Strategies for safer liver surgery and partial liver transplantation. N Engl J Med. 2007;356(15):1545–59. 3. Mao YL, et al. Application of technetium galactosyl human serum albumin diethylenetriamine pentaacetic acid injection on liver imaging in mouse models with different hepatic injuries. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2008;30(4):404–8. 4. Pugh RN, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg. 1973;60(8):646–9. 5. Poon RT, Fan ST. Assessment of hepatic reserve for indication of hepatic resection: how I do it. J Hepatobiliary Pancreatol Surg. 2005;12(1):31–7. 6. Mizuguchi T, et al. Preoperative liver function assessments to estimate the prognosis and safety of liver resections. Surg Today. 2014;44(1):1–10. 7. Schneider PD. Preoperative assessment of liver function. Surg Clin N Am. 2004;84(2):355–73. 8. Malinchoc M, et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology. 2000;31(4):864–71. 9. Cholongitas E, et al. A systematic review of the performance of the model for end-stage liver disease (MELD) in the setting of liver transplantation. Liver Transpl. 2006;12(7):1049–61. 10. Huo TI, Lee SD, Lin HC. Selecting an optimal prognostic system for liver cirrhosis: the model for end-stage liver disease and beyond. Liver Int. 2008;28(5):606–13. 11. Dutkowski P, et al. The model for end-stage liver disease allocation system for liver transplantation saves lives, but increases morbidity and cost: a prospective outcome analysis. Liver Transpl. 2011;17(6):674–84. 12. Cucchetti A, et al. Impact of model for end-stage liver disease (MELD) score on prognosis after hepatectomy for hepatocellular carcinoma on cirrhosis. Liver Transpl. 2006;12(6):966–71. 13. Abradelo M, Jimenez C. Splitting liver grafts for two adults: suboptimal grafts or suboptimal matching? Hepatobiliary Surg Nutr. 2013;2(5):242–3. 14. Kubota K, et al. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatology. 1997;26(5):1176–81. 15. Shoup M, et al. Volumetric analysis predicts hepatic dysfunction in patients undergoing major liver resection. J Gastrointest Surg. 2003;7(3):325–30. 16. Vauthey JN, et al. Standardized measurement of the future liver remnant prior to extended liver resection: methodology and clinical associations. Surgery. 2000;127(5):512–9. 17. Clavien PA, et al. Protection of the liver during hepatic surgery. J Gastrointest Surg. 2004;8(3):313–27. 18. Azoulay D, et al. Percutaneous portal vein embolization increases the feasibility and safety of major liver resection for hepatocellular carcinoma in injured liver. Ann Surg. 2000;232(5):665–72. 19. Hemming AW, et al. Preoperative portal vein embolization for extended hepatectomy. Ann Surg. 2003;237(5):686–91; discussion 691–3. 20. de Graaf W, et al. Assessment of future remnant liver function using hepatobiliary scintigraphy in patients undergoing major liver resection. J Gastrointest Surg. 2010;14(2):369–78. 21. Castera L, et al. Biopsy and non-invasive methods for the diagnosis of liver fi brosis: does it take two to tango? Gut. 2010;59(7):861–6. 22. Sandrin L, et al. Transient elastography: a new noninvasive method for assessment of hepatic fi brosis. Ultrasound Med Biol. 2003;29(12):1705–13. 23. Tsochatzis EA, et al. Elastography for the diagnosis of severity of fi brosis in chronic liver disease: a meta-analysis of diagnostic accuracy. J Hepatol. 2011;54(4):650–9. 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… 176 24. Cescon M, et al. Value of transient elastography measured with FibroScan in predicting the outcome of hepatic resection for hepatocellular carcinoma. Ann Surg. 2012;256(5):706–12; discussion 712–3. 25. Wong JS, et al. Liver stiffness measurement by transient elastography as a predictor on posthepatectomy outcomes. Ann Surg. 2013;257(5):922–8. 26. Morris-Stiff G, et al. Quantitative assessment of hepatic function and its relevance to the liver surgeon. J Gastrointest Surg. 2009;13(2):374–85. 27. Akita H, et al. Real-time intraoperative assessment of residual liver functional reserve using pulse dye densitometry. World J Surg. 2008;32(12):2668–74. 28. Sakka SG. Assessing liver function. Curr Opin Crit Care. 2007;13(2):207–14. 29. Scheingraber S, et al. Indocyanine green disappearance rate is the most useful marker for liver resection. Hepatogastroenterology. 2008;55(85):1394–9. 30. Ohwada S, et al. Perioperative real-time monitoring of indocyanine green clearance by pulse spectrophotometry predicts remnant liver functional reserve in resection of hepatocellular carcinoma. Br J Surg. 2006;93(3):339–46. 31. Imamura H, et al. Assessment of hepatic reserve for indication of hepatic resection: decision tree incorporating indocyanine green test. J Hepatobiliary Pancreatol Surg. 2005;12(1):16–22. 32. Das BC, Isaji S, Kawarada Y. Analysis of 100 consecutive hepatectomies: risk factors in patients with liver cirrhosis or obstructive jaundice. World J Surg. 2001;25(3):266–72; discussion 272–3. 33. Lau H, et al. Evaluation of preoperative hepatic function in patients with hepatocellular carcinoma undergoing hepatectomy. Br J Surg. 1997;84(9):1255–9. 34. Fung J, et al. Use of liver stiffness measurement for liver resection surgery: correlation with indocyanine green clearance testing and post-operative outcome. PLoS ONE. 2013;8(8), e72306. 35. Okabe H, et al. Rank classifi cation based on the combination of indocyanine green retention rate at 15 min and (99 m)Tc-DTPA-galactosyl human serum albumin scintigraphy predicts the safety of hepatic resection. Nucl Med Commun. 2014;35(5):478–83. 36. Kudo M, et al. Functional hepatic imaging with receptor-binding radiopharmaceutical: clinical potential as a measure of functioning hepatocyte mass. Gastroenterol Jpn. 1991;26(6):734–41. 37. Shuke N, et al. Estimation of fractional liver uptake and blood retention of 99mTc-DTPAgalactosyl human serum albumin: an application of a simple graphical method to dynamic SPECT. Nucl Med Commun. 2003;24(5):503–11. 38. Vera DR, Stadalnik RC, Krohn KA. Technetium-99 m galactosyl-neoglycoalbumin: preparation and preclinical studies. J Nucl Med. 1985;26(10):1157–67. 39. Kokudo N, et al. Predictors of successful hepatic resection: prognostic usefulness of hepatic asialoglycoprotein receptor analysis. World J Surg. 2002;26(11):1342–7. 40. Kaibori M, et al. Usefulness of Tc-99 m-GSA scintigraphy for liver surgery. Ann Nucl Med. 2011;25(9):593–602. 41. Kudo M, et al. Synthesis and radiolabeling of galactosyl human serum albumin. Methods Enzymol. 1994;247:383–94. 42. Kwon AH, et al. Use of technetium 99 m diethylenetriamine-pentaacetic acid-galactosylhuman serum albumin liver scintigraphy in the evaluation of preoperative and postoperative hepatic functional reserve for hepatectomy. Surgery. 1995;117(4):429–34. 43. Harada K, et al. Area between the hepatic and heart curves of (99 m)Tc-galactosyl-human serum albumin scintigraphy represents liver function and disease progression for preoperative evaluation in hepatocellular carcinoma patients. J Hepatobiliary Pancreatol Sci. 2012;19(6):667–73. 44. Wu J, et al. The functional hepatic volume assessed by 99mTc-GSA hepatic scintigraphy. Ann Nucl Med. 1995;9(4):229–35. Y. Mao and S. Du 177 45. Nanashima A, et al. Relationship between indocyanine green test and technetium-99 m galactosyl serum albumin scintigraphy in patients scheduled for hepatectomy: clinical evaluation and patient outcome. Hepatol Res. 2004;28(4):184–90. 46. Miki K, et al. Asialoglycoprotein receptor and hepatic blood fl ow using technetium- 99mDTPA-galactosylhuman serum albumin. J Nucl Med. 1997;38(11):1798–807. 47. Miki K, et al. Receptor measurements via Tc-GSA kinetic modeling are proportional to functional hepatocellular mass. J Nucl Med. 2001;42(5):733–7. 48. Ha-Kawa SK, et al. Compartmental analysis of asialoglycoprotein receptor scintigraphy for quantitative measurement of liver function: a multicentre study. Eur J Nucl Med. 1997;24(2):130–7. 49. Kwon AH, et al. Preoperative determination of the surgical procedure for hepatectomy using technetium-99mgalactosyl human serum albumin (99mTc-GSA) liver scintigraphy. Hepatology. 1997;25(2):426–9. 50. Kwon AH, et al. Preoperative regional maximal removal rate of technetium-99 m-galactosyl human serum albumin (GSA-Rmax) is useful for judging the safety of hepatic resection. Surgery. 2006;140(3):379–86. 51. Kaibori M, et al. HA/GSA-Rmax ratio as a predictor of postoperative liver failure. World J Surg. 2008;32(11):2410–8. 52. Du S, et al. A novel liver function evaluation system using radiopharmacokinetic modeling of technetium-99 m-DTPA-galactosyl human serum albumin. Nucl Med Commun. 2013;34(9):893–9. 53. Ge PL, Du SD, Mao YL. Advances in preoperative assessment of liver function. Hepatobiliary Pancreat Dis Int. 2014;13(4):361–70. 54. Mao Y, et al. Using dynamic 99mT c-GSA SPECT/CT fusion images for hepatectomy planning and postoperative liver failure prediction. Ann Surg Oncol. 2015;22(4):1301–7. 14 What Is the Best Way to Assess Hepatic Reserve Prior to Liver Resection… © Springer International Publishing Switzerland 2016 179 J.M. Millis, J.B. Matthews (eds.), Diffi cult Decisions in Hepatobiliary and Pancreatic Surgery, Diffi cult Decisions in Surgery: An Evidence-Based Approach, DOI 10.1007/978-3-319-27365-5_15 Chapter 15 Treatment Protocols for Small Hepatocellular Carcinoma (3 cm): RFA or Resection? Yudong Qiu and Yilei Mao Abstract Treatment selection for small hepatocellular carcinoma remains controversial. Although there are various studies showed different prognostic results in patients with small HCC by resection compared with RFA or LT, some other important factors, not only the tumor size, which may correlate with prognosis are still lack especially for gross classifi cation. The identifi cation of gross classifi cation is crucial for the discrimination of small HCC and may play a great role for the fi nal decision. Our results showed that not all the patients with small HCC are applicable for RFA treatment, so as to say, resection may be more benefi cial for patients with the nonboundary type of small HCC. Keywords Small hepatocellular carcinoma • RFA • Hepatic resection • Gross classifi cation Introduction Hepatocellular carcinoma ( HCC ) is a major health problem worldwide and a prevalent tumor type in mainland China [ 1 ]. Progresses in diagnostic imaging have allowed detection of HCC at an early stage which can be curable by multiple treatment protocols. According to BCLC staging system, patients with very early or early-stage HCC should be considered for resection , ablation or transplantation [ 2 , 3 ]. Also, the use of Milan Criteria to select patients for liver transplant ation (LT) leads to good results for a solitary HCC up to 5 cm or for multiple HCC up to 3 in number and up to 3 cm for each tumor [ 4 , 5 ]. In 2014, a new staging system called Y. Qiu Department of Hepatobiliary and Pancreas Surgery , Drum Tower Hospital, Medical School of Nanjing University , Nanjing , China Y. Mao (*) Department of Liver Surgery , Peking Union Medical College Hospital, Peking Union Medical College & Chinese Academy of Medical Sciences , 1# Shuai-Fu-Yuan, Wang-Fu-Jing , Beijing 100730 , China e-mail: [email protected]; [email protected] 180 HKLC system was erected which may be more applicable to Asian patients (Fig. 15.1 ). For the HCC patients with stage I and IIa, resection, transplantation and ablation are all recommended [ 6 ]. Hence, three various therapies including resection, transplantation and ablation have been adopted in patients with small HCC although these three approaches have respective distinct indications which are not mentioned in the current staging systems. Strategy Discussion Hepatitis B is endemic in China and this results in a heavy burden of hepatocellular carcinoma ( HCC ) because hepatitis B virus is a major risk factor in the development of the disease [ 7 , 8 ]. Most HCC patients with chronic infection with HCV have remarkable cirrhosis with impaired liver function, whereas patients with HBVrelated HCC in general have better preserved liver function. Individuals would be considered for liver transplant ation (LT) if they were with poor liver function reserve and especially small HCC within Milan Criteria (solitary tumour 5 cm and up to three nodules 3 cm) [ 9 ]. Nevertheless, this treatment which gives the potential to both resect the entire potentially tumor-bearing liver and eliminate the cirrhosis can be offered only to a minority of patients because of the shortage of donors and high Fig. 15.1 The HKLC prognostic classifi cation scheme. EVM extrahepatic vascular invasion/ metastasis. Early tumor: 5 cm, 3 tumor nodules and no intrahepatic venous invasion; Intermediate tumor: (1) 5 cm, either >3 tumor nodules or with intrahepatic venous invasion, or (2) >5 cm, 3 tumor nodules and no intrahepatic venous invasion; and Locally-advanced tumor: (1) 5 cm, >3 tumor nodules and with intrahepatic venous invasion, or (2) >5 cm, >3 tumor nodules or/and with intrahepatic venous invasion, or (3) diffuse tumor Y. Qiu and Y. Mao 181 cost [ 10 ]. Specifi cally, resection was more likely to be recommended over transplantation for patients with small solitary tumors, while patients with small multifocal lesion were much more likely to be referred for transplantation [ 11 ]. Pomfret EA et al. [ 12 ] also mentioned that the application of liver transplantation at very early stages of HCC development may be futile when treated in patients with wellcompensated cirrhosis and very early HCC (single tumour of <2 cm in size; T1 stage [ 13 , 14 ]). Due to the current issues about transplantation in China, this kind of therapy may be excluded from our fi rst treatment of choice for small HCC. Hepatectomy and ablation are another two treatment options available for small HCC that will potentially have a positive impact on survival . Not surprisingly, resection and ablation have achieved excellent survival outcomes in this setting, in the range of 60–70 % at 5 years [ 15 , 16 ]. Resection has generally been accepted as the fi rst treatment of choice for HCC in many centers. It is recommended by surgeons and allows for better local control, with an overall mortality rate less than 5 % in cirrhotic patients and long-term survival up to >50 % after adequate anatomical resections [ 17 , 18 ]. Anatomical hepatectomy is defi ned to preliminarily make blood occlusion of hepatic segments and sectors where tumor located and then undergo liver resection according to anatomical range. This approach resected the whole tumor and the hepatic segments and sectors which its portal venous branches allocated. It may ensure the negative incisal margin and decrease the intrahepatic spread of the tumor. After reforming the operation skill, our new approach is probably able to precisely dissect the hepatic pedicle which the required resected segments are affi liated and not need to excessively dissect more fi brous connective tissues of hilar plate and gallbladder bed (Figs. 15.2 and 15.3 ), consequently, more normal liver are remained by the skill. Previous study also revealed that precise hemihepatectomy guided by middle hepatic vein resulted in fewer incidences of postoperative complications and had the potential to achieve more adequate tumor- free resection margin, which may result in higher tumor-free survival rate [ 19 ]. For the nodules in the left lateral lobe, laparoscopic hepatectomy may be another kind of surgical choice. In addition, when patients’ performance status permitted, resection is still the fi rst-line treatment for small nodules in right posterior and middle lobe. Therefore, surgical resection was regarded as a fantastic treatment option for small HCC. However, less than 30 % of patients with small HCC are eligible for surgery , mainly because of the multiplicity and heterogeneity of the lesions that often occurs in a background of chronic liver disease, bad liver function, and deteriorating general condition [ 20 , 21 ]. So, many nonsurgical ablative methods have been developed, such as cryoablation, percutaneous ethanol injection (PEI), acetic acid injection, radiofrequency ablation ( RFA ), microwave coagulation, and transcatheter arterial chemoembolization ( TACE ) [ 22 ]. Among these modalities, Radiofrequency (RFA) is now the fi rst-line technique for ablation [ 23 ]. Treatment strategy for HCC is mainly decided according to the tumor size, tumor number, liver function and performance status. For small HCC, liver function, which is evaluated by three parts such as Child-Pugh grade, MELD score and the retention rate of ICG in 15 min (R15), may play a pivotal role in guiding decision making. If a patient is diagnosed as small HCC with bad liver function (MELD score >9 [ 24 ] or ICG-R15 >20 % 15 Treatment Protocols for Small Hepatocellular Carcinoma (3 cm): RFA or Resection? 182 [ 25 ]), RFA will be considered as fi rst line probably. Nevertheless, there is still controversy regarding the treatment choices for small HCC [ 26 ] although recent advances in RFA technology have enabled clinicians to use RFA for larger tumors [ 27 ]. While a robust trial appropriately comparing resection and ablation is still not available [ 28 ], large case-control series and modelling studies support RFA as a non-inferior [ 29 ] and more cost -effective [ 30 ] treatment for very early HCCs. Wakai T et al. [ 31 ] proved that hepatectomy provides both similar local control and better long-term survival for patients with HCC 4 cm in comparison with percutaneous ablation. A nonrandomized prospective study suggested that resection is superior to RFA in long-term survival [ 29 ]. Moreover, a recently reported randomized trial confi rmed that in patients with small HCC, percutaneous RFA showed similar local control and long-term survival compared with hepatectomy but are accompanied with a lower complication rate and shorter hospital stay day [ 32 ]. Results As far as we know, there have been rare randomized trials to compare the effi cacy of RFA with that of surgical resection for an operable early-stage HCC in terms of survival for HCCs 3 cm [ 33 , 34 ]. In our opinion, a new risk factor like gross Fig. 15.2 Anatomical liver S4, 5, 8 resection . Boundary type of single nodular small HCC Y. Qiu and Y. Mao 183 classifi cation should be added in to further help make decision in treatment choice for small HCC. The concept of gross classifi cation for HCC was fi rst put forward in 1984 by Okuda K et al. [ 35 ]. According to those studies made by the Japanese scholars, HCC nodules were divided into four groups based on the classifi cation in The general rules for the clinical and pathological study of primary liver cancer , 4th ed ., established by the Liver Cancer Study Group of Japan [ 36 ]: single nodular type (Fig. 15.4 ), single nodular type with extranodular growth, confl uent multinodular type and invasive type (Fig. 15.5 ). In total, 88 patients with small HCC treated by RFA were divided into two groups on the basis of gross classifi cation distinguished through preoperative imaging data. Our incipient results revealed that the single nodular type group (SN) had signifi cantly better overall survival (OS) and recurrence- free survival (RFS) than the non-single nodular type group (non-SN) ( P < 0.05) (Figs. 15.6 and 15.7 ). This signifi cance indicated that not all the patients with small HCC are applicable for RFA treatment, so as to say, resection may be more benefi cial for patients with the nonboundary type of small HCC. Fig. 15.3 Anatomical liver S5 resection . Infi ltrating small HCC lesion 15 Treatment Protocols for Small Hepatocellular Carcinoma (3 cm): RFA or Resection? 184 Risk of Recurrence One of the greatest problems plaguing potential curative treatment for HCC is the high risk of recurrence (i.e., ablation and surgical resection ). Whichever modality we choose to treat small HCC, recurrence and follow-up work should not be ignored. Early recurrence due to dissemination is likely to have poorer prognosis than late recurrence as it happens after resection. Tumor seeding due to tumor puncture for diagnosis or ablation is the most important, as it is associated with a poor prognosis among the patterns of recurrence [ 37 , 38 ]. In current study, local recurrence was found to be more frequent after RFA than after HR. Local recurrences after RFA may be attributable to insuffi cient ablation of the primary tumor and/or the presence of tumor venous invasion in the adjacent liver [ 22 ]. Solving these problems, Fig. 15.4 Based on imaging examination, the macroscopic type was single nodular small HCC . ( a ) plain CT . ( b ) arterial phase. ( c ) portal phase CT computed tomography Fig. 15.5 Based on imaging examination, the macroscopic type was invasive small HCC . ( a ) plain CT . ( b ) arterial phase. ( c ) portal phase CT computed tomography Y. Qiu and Y. Mao 185 prolongation of the follow-up time is needed and might be benefi cial for the comparison of the disease-free and overall survival rates between RFA and hepatectomy . Resection or TACE when indicated would be the great treatment of choice against intrahepatic recurrences. Conclusion As previously stated, treatment decision in patients with small HCC should be individualized according to the parameters at fi rst diagnosis . Combined with current research, three curative therapies (surgical resection , transplantation, RFA ) are effi – cacious for small HCC. How to select an appropriate treatment seems to be a bit vague in order to achieve a better prognosis. Shown in our results, gross classifi cation may play a pivotal role in therapy decision making for small HCC. In summary, comprehensive diagnosis and treatment is essential for future survival in patients with HCC. For these results to take place, all factors should be considered in .00 SN non-SN non-SN-censored SN-censored 0.0 0.2 0.4 0.6 0.8 1.0 20.00 40.00 60.00 Months after Surgery Cum Recurrence-free Survival Fig. 15.6 Recurrence-free survival curves for patients with small HCC treated by RFA between SN and non-SN group ( P = 0.21) 15 Treatment Protocols for Small Hepatocellular Carcinoma (3 cm): RFA or Resection? 186 treating small HCC. Only by combination of the past clinical experience, the current recommendations from guidelines and the latest research results will allow those patients to achieve benefi ts. Recommendations • For Asian patients with HBV-related small HCC , we recommend the use of RFA and resection to treat small HCC. • For patients with Child-Pugh A and MELD score <9, resection should be considered. However, patients with MELD score >10, cirrhosis and portal hypertension are tending to be treated with RFA . • There is evidence for RFA to treat those patients with the boundary type of small HCC in accordance with gross classifi cation . • Specifi cally, infi ltrating hepatocellular carcinoma (iHCC) described as invasive type in gross classifi cation was suggested to be treated with anatomical resection due to the high recurrence rate if managed with RFA . .00 non-SN SN non-SN-censored SN-censored Months after Surgery Cum Overall Surgery 0.0 0.2 0.4 0.6 0.8 1.0 20.00 40.00 60.00 80.00 100.00 Fig. 15.7 Overall survival curves for patients with small HCC treated by RFA between SN and non-SN group ( P = 0.19) Y. Qiu and Y. Mao 187 References 1. Bruix J, Gores GJ, Mazzaferro V. Hepatocellular carcinoma: clinical frontiers and perspectives. Gut. 2014;63:844–55. 2. Llovet JM, Brú C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classifi – cation. Semin Liver Dis. 1999;19:329–38. 3. Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology. 2011;53:1020–2. 4. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996;334:693–9. 5. Lau WY, Lai EC. Salvage surgery following downstaging of unresectable hepatocellular carcinoma – a strategy to increase resectability. Ann Surg Oncol. 2007;14:3301–9. 6. Yau T, Tang VY, Yao TJ, et al. Development of Hong Kong Liver Cancer staging system with treatment stratifi cation for patients with hepatocellular carcinoma. Gastroenterology. 2014;146:1691–700. 7. Yin L, Li H, Li AJ, et al. Partial hepatectomy vs. transcatheter arterial chemoembolization for resectable multiple hepatocellular carcinoma beyond Milan Criteria: a RCT. J Hepatol. 2014;61:82–8. 8. Llovet JM, Bruix J. Novel advancements in the management of hepatocellular carcinoma in 2008. J Hepatol. 2008;48 Suppl 1:S20–37. 9. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42:1208–36. 10. Befeler AS, Hayashi PH, Di Bisceglie AM. Liver transplantation for hepatocellular carcinoma. Gastroenterology. 2005;128:1752–64. 11. Nathan H, Herlong HF, Gurakar A, et al. Clinical decision-making by gastroenterologists and hepatologists for patients with early hepatocellular carcinoma. Ann Surg Oncol. 2014;21:1844–51. 12. Pomfret EA, Washburn K, Wald C, et al. Report of a national conference on liver allocation in patients with hepatocellular carcinoma in the United States. Liver Transpl. 2010;16:262–78. 13. Vauthey JN, Lauwers GY, Esnaola NF, et al. Simplifi ed staging for hepatocellular carcinoma. J Clin Oncol. 2002;20:1527–36. 14. OPTN U, editor. United network for organ sharing. Liver Transplant Candidates with hepatocellular carcinoma (HCC) Policy 3.6.4.4. http://optn.transplant.hrsa.gov/PoliciesandBylaws2/ policies/pdfs/policy_8.pdf. Accessed 23 Jun 2013. 15. Livraghi T, Meloni F, Di Stasi M, et al. Sustained complete response and complications rates after radiofrequency ablation of very early hepatocellular carcinoma in cirrhosis: is resection still the treatment of choice? Hepatology. 2008;47:82–9. 16. Roayaie S, Obeidat K, Sposito C, et al. Resection of hepatocellular cancer or=5.0-cm) HCCs: multipolar RF ablation with three internally cooled bipolar electrodes—initial experience in 26 patients. Radiology. 2008;248:288–96. 28. Majno PE, Mentha G, Mazzaferro V. Partial hepatectomy versus radiofrequency ablation for hepatocellular carcinoma: confi rming the trial that will never be, and some comments on the indications for liver resection. Hepatology. 2010;51:1116–8. 29. Cho YK, Kim JK, Kim WT, et al. Hepatic resection versus radiofrequency ablation for very early stage hepatocellular carcinoma: a Markov model analysis. Hepatology. 2010;51:1284–90. 30. Roberts JP, Venook A, Kerlan R, et al. Hepatocellular carcinoma: ablate and wait versus rapid transplantation. Liver Transpl. 2010;16:925–9. 31. Wakai T, Shirai Y, Suda T, et al. Long-term outcomes of hepatectomy vs percutaneous ablation for treatment of hepatocellular carcinoma