Hepatic resection for benign and malignant liver and biliary tract disease: Indications and outcomes

Preoperative evaluation

All patients for whom hepatic resection is planned will require evaluation of their baseline liver function. The cornerstone of assessment remains history and physical examination. History should focus on symptoms of liver disease, previous related diagnoses, and thorough social history to identify high-risk behavior. Physical examination should focus on stigmata of underlying liver disease, including the presence of jaundice, hepatomegaly, ascites, and varices. Liver function testing can augment the information obtained through history and physical examination but is often inadequate for discerning underlying liver dysfunction on its own.

Several chronic conditions can compromise liver function, and care should be taken to discern them before performing liver resection. Hepatitis C virus (HCV) and hepatitis B virus (HBV) can both lead to chronic liver inflammation through cycles of viral activation and ultimately progression to decompensated cirrhosis (see Chapter 68 ). Although merely being infected with viral hepatitis does not appear to increase the perioperative risk associated with hepatic resection, decompensated cirrhosis does (see Chapter 74 ). Other potential causes of chronic liver disease include chronic alcoholic liver disease, nonalcoholic fatty liver disease, hemochromatosis, primary biliary cirrhosis, primary sclerosing cholangitis, and autoimmune hepatitis. In some cases, the severity of the underlying liver disease can have a larger impact on long-term survival than the characteristics of the malignancy being resected. Importantly, the presence and degree of liver cirrhosis can dictate the extent of hepatectomy that would be considered safe and influence the minimum standardized functional liver remnant (sFLR) required to proceed with liver surgery.

Several scoring systems are used to quantify the extent of liver disease and indirectly select patients for hepatectomy (see Chapter 4 ). The Model for End Stage Liver Disease (MELD), originally developed to predict 3-month survival in patients with cirrhosis, can be used to predict posthepatectomy liver failure. It may be useful in the determination of hepatectomy versus liver transplantation in patients where both options represent reasonable approaches, and it has been adopted by the United Network for Organ Sharing (UNOS) to prioritize patients awaiting liver transplantation in the United States. However, MELD is not routinely used for calculating standardized future liver remnant (sFLR) thresholds before surgery. Instead, the Child-Pugh-Turcotte ( Table 101B.2 ) score has the advantage of considering clinical manifestations of liver disease when compared with MELD. For patients with Child-Pugh-Turcotte A cirrhosis, an sFLR of greater than 40% is typically required to proceed to hepatic resection. The same threshold applies for those with class B cirrhosis, although these patients are generally not suitable candidates for major hepatectomy. Patients with Child-Pugh-Turcotte C scores are considered poor surgical candidates. By comparison, in the absence of cirrhosis, sFLR greater than 20% is generally deemed adequate for patients with normal liver function. In addition to scoring systems, dynamic measures to calculate liver function are also used. Indocyanine green (ICG) clearance at 15 minutes was first introduced in Japan, and a meta-analysis found that values less than 7.1% are better predictors of posthepatectomy liver failure than either MELD or Child-Pugh-Turcotte score.

In addition to the severity of chronic liver disease, other factors will also need to be considered when determining the optimal sFLR and timing of surgery. Patients who receive 5-fluorouracil-based chemotherapy are at risk for hepatic steatosis; irinotecan-based chemotherapy can cause steatohepatitis, and oxaliplatin is associated with sinusoidal obstruction syndrome (SOS; see Chapter 69 ). Although steatosis and SOS minimally increase perioperative risks, patients found to have steatohepatitis on their resected liver specimen have postoperative 90-day mortality that is significantly increased from 1.6% to 14.7%, with most deaths caused by posthepatectomy liver failure. Additionally, patients with extensive exposure to chemotherapy (>12 cycles) have increased rates of posthepatectomy liver failure with sFLR less than 30%. For this reason, our group reserves major hepatectomy for patients treated with extensive chemotherapy to those whose sFLR is at least 30% (see Chapter 102 ).

For patients with inadequate sFLR to consider hepatic resection, preoperative volume expansion of the FLR can be attempted via several approaches. Preoperative portal vein embolization (PVE) by interventional radiology is the most commonly employed and best studied method for sFLR augmentation (see Chapter 102C ). This procedure is most frequently accomplished with percutaneous vascular access and embolization of the right portal vein, with or without embolization of segment IV branches. After PVE, patients undergo repeat volumetry; those who demonstrate more than 2.0% growth per week (referred to as kinetic growth rate) have a 0% rate of liver failure in the perioperative period after hepatectomy. This technique can be employed as part of a single stage or two-stage approach to hepatic lesions ( Fig. 101B.2 ).

Associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) is another strategy for patients who do not have adequate sFLR to undergo hepatectomy (see Chapter 102D ). In this two-stage operation, patients undergo parenchymal transection and portal ligation at a first-stage operation and then, 1 to 2 weeks later, undergo a second operation to remove the portion of the liver affected by disease. Advocates of this approach point to higher rates of liver hypertrophy and more patients undergoing definitive resection than PVE, whereas detractors note high perioperative mortality rates of 9% in the international ALPPS registry. For more information on PVE and ALPPS, please reference the discussions in Chapters 102C and 102D , respectively.

Although the factors previously discussed are common to all patients undergoing hepatectomy for almost any indication, disease-specific outcome determinants need to be considered as well. This chapter will explore both malignant and benign indications for liver surgery and their associated outcomes.

Hepatocellular carcinoma (see Chapter 89 )

HCC is the most common primary liver tumor and accounts for roughly 70% of primary liver tumor diagnoses. The incidence of HCC has increased over the last 30 years and is anticipated to continue increasing in the future. ^, HCC often occurs in the setting of chronic liver disease and cirrhosis. It is typically diagnosed late in its course, although screening initiatives have led to earlier diagnoses in the last 2 decades. Mortality from HCC remains high and is one of the fastest rising causes of cancer-related deaths. Underlying liver disease, which is the most significant risk factor for HCC, often complicates or precludes aggressive treatment via liver resection and other liver-directed modalities.

HCC has several established risk factors that can be broadly grouped into the following categories:

• 1.

  Chronic liver disease

• 2.

  Metabolic risk factors

• 3.

  Genetic risk factors

These categorizations can help with identification, risk mitigation, and, ultimately, treatment planning.

The most common causes of chronic liver disease in the world are the viral causes of hepatitis, and HBV and HCV are also the two most commonly identified risk factors preceding a HCC diagnosis worldwide (see Chapter 68 ). HBV is the incident risk factor identified most commonly globally, whereas HCV is the most common in the United States. Patients who have chronic infection with HBV are at risk for developing HCC even in the absence of cirrhosis. Progression to cirrhosis further increases the risk of developing HCC, with reported rates of 1% to 8% per year. ^, With the introduction of combination treatment with sofosbuvir and ledipasvir, HCV now represents a potentially modifiable risk factor. Similarly, HBV vaccination presents another opportunity to address a potentially modifiable target.

Nonalcoholic fatty liver disease (NAFLD) is the fastest rising cause of chronic liver disease in the world and the most frequent chronic liver disease in the United States (see Chapter 69 ). NAFLD can progress to nonalcoholic steatohepatitis (NASH). It represents a crossover between metabolic diseases and chronic liver disease because of the implication of insulin resistance. NAFLD/NASH do not appear to increase risk of HCC on their own but instead confer additional risk with progression to cirrhosis. Alcohol intake similarly does not appear to be related to carcinogenesis outside of the setting of cirrhosis. Diabetes is another modifiable metabolic risk factor for HCC development. In one meta-analysis, diabetes was associated with up to a 2.5 times greater risk of HCC. Additionally, there is some evidence that use of metformin can decrease HCC prevalence in both diabetics and nondiabetics with metabolic syndromes. ^,

Genetic disorders have been associated with increased prevalence of HCC. These include hemochromatosis, an autosomal recessive disorder of inborn iron metabolism, which is strongly associated with HCC, particularly once progression to cirrhosis has occurred. Additionally, alpha-1 antitrypsin deficiency, primary biliary cirrhosis, and Wilson disease have all been associated with a less profound risk of progression to cirrhosis and HCC development (see Chapters 74 and 89 ).

The diagnosis of HCC is largely dependent on imaging and may be established on the basis of imaging characteristics alone in cirrhotic patients. Under screening protocols for at-risk individuals, initial findings of a mass on liver ultrasound should lead to follow-up cross-sectional imaging. The combination of arterial hyperenhancement and early contrast washout in the venous and/or delayed phases is highly suggestive of HCC, and in high-risk patients, there is no need for routine biopsy in the presence of these computed tomography (CT)/magnetic resonance imaging (MRI) findings. In fact, these imaging findings make up part of the Liver Reporting & Data System (LI-RADS), a standardized reporting system developed by the American College of Radiology to apply consistent terminology and facilitate communication between clinicians when dealing with patients at risk of developing HCC ( Fig. 101B.3 ; see Chapter 14 ).

Once a diagnosis of HCC is established, a wide array of treatment options is available that must consider both tumor and underlying liver characteristics. There is no single consensus ideal staging algorithm to dictate treatment of HCC, although the Barcelona Clinic Liver Cancer and American Joint Committee on Cancer (AJCC) TNM systems are the most widely used. Two prior consensus statements from the Americas Hepato-Pancreato-Biliary Association (AHPBA) have counseled against routine use of either system for all patients. ^,

Often the etiology and severity of the underlying chronic liver disease will direct a patient toward and away from individual treatment modalities. In many patients with localized disease, the severity of their underlying liver disease may preclude the safe performance of liver resection, which may otherwise have been a good option. For patients who are not eligible for resection, transplantation is the treatment of choice should criteria be met because it would address both the cancer and the underlying diseased liver condition (see Chapter 105 ). Ablation represents another potentially curative option but is best reserved for small tumors (see Chapter 96 ). When surgery is considered, preoperative selection of patients should focus on the likelihood of disease being confined to the liver, anatomic constraints of resectability, and limits of underlying liver dysfunction.

Resection should be considered in patients who have disease confined to the liver and with normal underlying liver function or compensated liver cirrhosis. Evidence of portal hypertension, including presence of ascites, splenomegaly with associated thrombocytopenia, or varices and/or a recanalized umbilical vein ( Fig. 101B.4 ), or other signs of decompensated cirrhosis will render most patients ineligible for resection-based therapy (see Chapter 74 ). All patients with known liver disease should undergo evaluation with a validated system such as the Child-Turcotte-Pugh or MELD systems. If the status of the patient’s liver disease is unclear from these systems, or when concern persists for portal hypertension despite reassuring liver function tests, direct hepatic vein-portal vein gradient can be measured to assess the presence and degree of portal hypertension.

Patients who have signs of decompensated cirrhosis with refractory ascites, hepatorenal syndrome, hepatopulmonary syndrome, or spontaneous bacterial peritonitis should not be considered for resection. Similarly, only patients with Child-Pugh-Turcotte class A cirrhosis should be considered for major resection. Based on these clinical criteria, in the absence of portal hypertension and with adequate sFLR, approximately 5% to 10% of patients with cirrhosis and HCC will be eligible for resection. In these patients, resection is relatively safe, with perioperative mortality of 2% and transfusion rates of 10%. After successful resection, cirrhotic patients with HCC can expect 5-year survival rates up to 50% to 70%.

For patients who are not suitable candidates for hepatic resections, other available treatment modalities include systemic therapy (see Chapter 99 ), radiofrequency ablation (RFA; see Chapter 96B ), microwave ablation (see Chapter 96C ), percutaneous alcohol injection (see Chapter 96D ), bland embolization (see Chapter 94A ), trans-arterial chemoembolization (TACE; see Chapter 94A ), transarterial radioembolization (TARE) with yttrium-90 (Y-90; see Chapter 94B ), stereotactic body radiation therapy (SBRT; see Chapter 95 ), proton beam irradiation (see Chapter 95 ), and liver transplantation (see Chapters 105 and 108A ). These therapies are often used in conjunction with one another and sometimes as strategies for local control to bridge to transplant or resection. They are covered in greater detail in the chapters dedicated to HCC treatment.

Several randomized controlled trials (RCTs) have compared RFA and liver resection for small, potentially resectable HCC. ^, These trials focused on different populations, and all but one showed inferiority of RFA to surgery. Subsequent meta-analyses of these trials found that RFA is inferior to surgery for patients who can undergo liver resection. One meta-analysis included patients who had HCC less than 2 cm and were Child-Pugh Class A with or without cirrhosis. It showed similar 1-year overall survival (OS) but demonstrated that resection was associated with improved 3- and 5-year OS (see Chapter 96B ).

For tumors that are larger than 2 cm but less than 5 cm, RFA is generally discouraged because of incomplete coverage of the ablation zone. The question shifts instead to the type and extent of surgical resection, with several studies investigating the question of anatomic versus nonanatomic resection. In this regard, meta-analyses on the topic have reached different conclusions. One early meta-analysis showed similar OS for both approaches but with decreased rates of local intrahepatic recurrence and early postoperative recurrence in the anatomic resection group. Morbidity and mortality were similar between the two approaches. A more recent meta-analysis contradicted these findings and showed that anatomic resection was associated with disease-free survival (DFS) benefits at 1, 3, and 5 years. Additionally, the meta-analysis showed improved OS at 5 years. Our group recommends anatomic resection for HCC when feasible.

Historically, surgeons have been reticent to approach large (>5 cm) and giant (>10 cm) HCCs surgically because of concerns for both poor short-term operative outcomes and long-term survival outcomes. More recent reports, however, have indicated that surgery can be safe in well-selected patients with perioperative mortality less than 3%. Long-term outcomes in recent case series have also been encouraging, with 5-year survival ranging from 29% to 53% for tumors greater than 5 cm and 27% to 35% for tumors greater than 10 cm. TACE has previously been recommended for these patients, but a recent comparison indicated that surgical resection is associated with superior survival. With improving surgical technique and perioperative pathways, we do not recognize a tumor size limit for surgical resection and offer this option to patients whenever possible. It should be noted that TACE or TARE may represent good initial treatments before surgical resection by offering a window into the tumor biology.

Macrovascular invasion is one of the strongest predictors of poor survival in HCC and is associated with intrahepatic and extrahepatic metastases (see Chapter 89 ). Additionally, portal vein and hepatic vein thromboses, when present, can significantly increase the operative morbidity and mortality of hepatectomy, which approaches 8% in these patients. In this setting, survival tends to be poor, regardless of the treatment approach. Median survival for patients with macrovascular invasion managed with best supportive care is 5 months and stretches to 6 months with sorafenib. One multicenter surgical series was able to demonstrate 10% 5-year survival for patients who underwent surgical resection, with an 11-month median survival.

For patients who are not eligible for resection but still fulfill the Milan criteria, liver transplantation is an option. The original report from which the Milan criteria were developed showed 75% OS at 5 years and recurrence-free survival of 83% for patients who underwent transplantation for HCC. Since that time, 70% has become the benchmark OS at 5 years after liver transplantation for HCC. After concerns that the Milan criteria may be too restrictive, various centers adopted less restrictive guidelines, including the University of California San Francisco (UCSF) criteria that expanded transplant criteria for HCC ( Table 101B.3 ). When compared, neither set of criteria was predictive of improved survival (see Chapter 108A ).

In patients who undergo successful resection of their primary HCC, about half will recur within 2 years. Median survival after recurrence is less than 2 years. At the time of recurrence, only 5% to 30% of patients will be eligible for salvage resection. Many patients will undergo other strategies for local control, including RFA, TACE, TARE, and systemic chemotherapy. Patients who have liver-only recurrences are potentially eligible for salvage liver transplant. Those who do undergo salvage liver transplant have comparable 1-, 3-, and 5-year survival to primary liver transplant for HCC (100%, 95%, 85%, respectively) but are more likely to develop tumor recurrence (28% vs. 15.6%). For further information on HCC, please refer to Chapter 89 .

Intrahepatic cholangiocarcinoma (see Chapters 50 and 118 )

IHC is a relatively rare cancer. It is the second most common primary liver malignancy and accounts for 10% to 20% of primary liver cancer diagnoses. Among cholangiocarcinomas, IHC is the least common and is thought to represent 5% to 10% of all cholangiocarcinoma diagnoses, although this relative percentage might be rising with the growing incidence of this condition. For the purposes of this chapter, IHC is treated as a distinct entity from perihilar cholangiocarcinoma. Despite their similar histology, these malignancies have different biologies, treatment options, and outcomes.

In the United States, the overall incidence of IHC is low at 0.95 cases per 100,000; it can be up to 100 times more frequent in East Asian countries. The incidence in America has risen over the past several decades, mostly with new diagnoses at early stage disease. Epidemiologically, the rise in the incidence of IHC has nearly mirrored a corresponding decline in the rate of cancers of unknown primary (CUPs), suggesting that IHC might have previously been misidentified as CUP. At the same time, there appears to be a true increase in the incidence of IHC that is possibly related to the epidemic of metabolic syndrome. Risk factors for IHC include viral hepatitis, cirrhosis, primary sclerosing cholangitis (PSC), NAFLD, parasitic infection ( Clonorchis sinensis and Opisthorchis viverrini ), and diabetes. Like other biliary cancers, incidence also increases with age.

Patients with cholangiocarcinoma at extrahepatic sites are likely to develop jaundice as a presenting symptom. For IHC, jaundice is less common, whereas right upper quadrant pain, weight loss, or incidental discovery are more likely. The majority of cases are diagnosed via cross-sectional imaging for the mass-forming morphology. Typical features on CT imaging are a hypodense lesion with rim enhancement during both arterial and portal venous phases. When no mass is apparent on imaging, other highly suggestive findings include unexplained biliary dilation and liver atrophy. On MRI, IHCs appear hypointense on T1-weighted imaging and hyperintense on T2-weighted imaging. Cross-sectional imaging is also useful for staging and may demonstrate suspicious lymph nodes, intrahepatic spread, or carcinomatosis (see Chapter 14 ).

In addition to imaging abnormalities, patients may have laboratory abnormalities including elevated levels of alkaline phosphatase. Bilirubin is usually normal in patients with IHC. Tumor markers including carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), and alpha-fetoprotein (AFP) should be monitored and can point to other potential etiologies, including colorectal liver metastases with CEA elevation and HCC with elevated AFP. Markedly elevated CA19-9 can be indicative of more advanced disease, and levels over 100 U/mL have been associated with worse DFS after surgical resection. Tumor markers alone are insufficient for establishing a diagnosis of IHC however.

Accurately staging patients with IHC is important when considering curative-intent surgery. Patients who are found to have extra-regional distant metastases or invasion of critical structures not amenable to resection are not candidates for surgery. In the absence of these features, staging laparoscopy may still be indicated in the presence of high-risk features such as suspicious lymph nodes, multifocal nodules, and/or high serum CA19-9. Up to a third of these patients will be found to have occult peritoneal disease or intrahepatic metastases at the time of exploration.

In patients where there is no contraindication to surgery and an adequate sFLR can be maintained after R0 resection, surgery is the only potentially curative therapy. In patients undergoing surgery for IHC, R0 resections are obtained only 75% of the time. Large primary tumor size, locally advanced disease, or invasion of critical structures are risk factors for positive margins. Surgeons may be tempted to undertake more extensive and extended resection to achieve R0 resections, but margin status must be balanced against the likelihood of surgical complications. Indeed, postoperative complications have been found to be an independent predictor of worse long-term outcomes, which correlate with the severity of the complications.

Lymphadenectomy during surgical resection for IHC remains controversial. Although there is no definitive evidence that lymphadenectomy is associated with improved survival, it does allow for more accurate disease staging. The information gained has been shown to direct decision making regarding adjuvant therapy and refine prognostication. Lymphadenectomy identifies positive nodes up to 30% to 40% of the time, which is associated with decreased median survival. For these reasons, consensus statements, including the National Comprehensive Cancer Network (NCCN) guidelines, recommend routine lymphadenectomy at the time of hepatectomy for IHC. ^, Despite these recommendations, some controversy remains, with a recent meta-analysis showing no difference in OS or recurrence but increased postoperative morbidity associated with lymphadenectomy (see Chapter 50 ). At our institution, we routinely perform lymphadenectomy for the purpose of accurate disease staging.

In patients who undergo curative-intent hepatectomy, disease stage is an important prognostic consideration. Five-year survival ranges from 90% for resected stage IA IHC to 16.2% for stage IIIB using the AJCC 8th edition staging system ( Table 101B.4 ). In addition to stage, microvascular invasion, perineural invasion, and surgical margins have been associated with worse prognosis. Several nomograms have been developed that consider other factors including tumor size and number, vascular invasion, and tumor markers to further refine prognostication after hepatectomy.

Local recurrence is the most frequent pattern of disease recurrence, even with R0 resection after hepatectomy. Up to two-thirds of patients with IHC will experience recurrence within 2 years of surgery. Although recurrent tumors are associated with worse outcomes, patients who can be salvaged with repeat hepatectomy see increased median survival from 11.1 months to 26.7 months. Among patients who cannot undergo salvage hepatectomy, median survival is shortest for patients who receive best supportive care at 8 months and increases to 16.8 months with systemic therapy and 18 months with liver-directed therapy ( Fig. 101B.5 ).

Data for adjuvant therapy in resected IHC remain sparse. To date, only one RCT has been completed on this subject that has demonstrated a survival benefit with adjuvant therapy. In the BILCAP trial, which included all bile duct and gallbladder cancers, patients randomized to the capecitabine arm after resection had significantly longer OS compared with those in the observation arm (53 months vs. 36 months) on a predetermined per-protocol analysis ( Fig. 101B.6 ), as well as longer DFS (24.4 months vs. 17.5 months). Large retrospective series have also suggested a benefit to adjuvant therapy, especially for patients with lymph node positive disease and advanced T-stage with gemcitabine-based regimens. On the other hand, the PRODIGE 12 trial failed to demonstrate a benefit to adjuvant gemcitabine and oxaliplatin for resected biliary tract cancers (including 88 patients with IHC) compared with observation (see Chapter 50 ).

For patients with unresectable disease, there is a well-defined role for both systemic therapy and liver-directed treatment strategies besides hepatectomy. The ABC-02 trial showed improvement in OS (11.7 months vs. 8.1 months) and progression-free survival (PFS; 8 months vs. 5 months) for gemcitabine and cisplatin versus gemcitabine alone. Liver-directed therapies including hepatic artery infusion (HAI), TARE with Y-90, TACE, and drug-eluting bead TACE (DEB-TACE) have all been identified as potential local control strategies. In one meta-analysis, HAI was associated with the best survival outcomes and response rates compared with Y-90 and TACE. For further information on IHC, please refer to Chapter 50 .

Gallbladder cancers (see Chapters 49 and 119A )

Gallbladder cancer is the most common biliary tract cancer worldwide, although its incidence in the Unites States is low. It is endemic in certain geographic areas of the world and among some ethnic groups in countries in Asia and South America. Gallbladder cancer is unusual in that it is the only primary liver or biliary cancer with a higher incidence in women than in men. It is an aggressive malignancy with an exceedingly poor prognosis and often presents at advanced stages. Diagnosis at an early stage often occurs in the setting of an incidental finding on cholecystectomy performed for what is thought to be a benign indication. For this reason, surgical therapy for gallbladder cancer often occurs in a staged fashion.

Before curative-intent surgery for gallbladder cancer, accurate staging is of utmost importance. Two paradigms begin to emerge in the treatment of gallbladder cancer: patients who are diagnosed incidentally at the time of cholecystectomy for presumed cholecystitis or symptomatic cholelithiasis and patients who are diagnosed with gallbladder cancer before surgical resection. In both scenarios, characteristics of the primary tumor such as T-stage and molecular analysis can provide insights about future therapies. Lymphatic spread is common and even patients with T2 disease can have lymph node involvement more than 40% of the time.

Incidental diagnosis

Patients who are found to have incidental T1a tumors resected with negative margins on cholecystectomy are felt to be cured, and extended resection does not increase long-term survival. These tumors are associated with low rates of lymph node metastases of less than 5%. For individuals with T1b tumors, which penetrate the muscle layer, most guidelines and retrospective analyses favor more extensive resection. The need for completion radical or extended cholecystectomy, which includes partial hepatectomy of the gallbladder bed and regional lymphadenectomy, is supported by the 15% rate of nodal involvement with T1b disease. Decision analysis shows a potential survival benefit to extended resection despite equivalence in small retrospective series.

For patients with T2 disease, aggressive surgery can have marked effects on long-term survival. Series have shown increased long-term survival with extended resection versus simple cholecystectomy, a finding that is also demonstrated in large population-based registries, including the Surveillance, Epidemiology, and End Results (SEER) database. T2 tumors have higher rates of positive margins and local recurrence when simple cholecystectomy is performed and these patients should undergo completion radical cholecystectomy to realize the survival benefit of patients who are identified preoperatively. Those with T3 and T4 tumors are unlikely to be discovered incidentally because the tumor extends outside the confines of the gallbladder. When that is the case, however, patients with T3 tumors will have residual disease in more than 77.3% of cases in one retrospective study.

Even in this re-operative setting, use of minimally invasive techniques has been shown to be safe and feasible to achieve a complete regional lymphadenectomy and hepatectomy and may be aided by the use of fluorescence angiography and cholangiography for surgical navigation ( Fig. 101B.7 ).

Preoperative diagnosis

Patients who are diagnosed with gallbladder cancer before surgery should undergo imaging to determine the possibility of spread and ultimately staging laparoscopy if the cancer is thought to be resectable. Liver metastases, peritoneal metastases, malignant ascites, distant lymphatic spread, extensive involvement of the hepatoduodenal ligament, and/or encasement of the hepatic artery or portal vein are contraindications to proceeding with definitive surgical resection. Staging laparoscopy is an important tool for determining resectability and avoiding nontherapeutic laparotomy because up to a quarter of patients who are thought to be resectable based on imaging alone will have findings precluding resection on staging laparoscopy (see Chapter 24 ).

T3 tumors perforate the gallbladder serosa and/or invade the liver or another adjacent organ. For these patients, extended resection is warranted to achieve a margin negative resection and may require en bloc resection of the involved organ. T4 disease may be apparent on preoperative imaging or at the time of exploration; in well-selected patients, surgical resection may be possible, although major vascular invasion is a terrible prognostic sign and precludes resection. For further discussion on the extent of lymphadenectomy and extent of resection, please refer to Chapter 119A .

Perioperative outcomes depend on the type of resection performed. Simple cholecystectomy is well tolerated with little in the way of mortality, whereas radical resection has historically been associated with mortality rates as high as 5%. Morbidity and mortality rates are correlated with resection extent and are higher when bile duct excision and major hepatectomy are required to clear disease (see Chapters 102 and 119A ). Given their greater rates of complications, these aggressive maneuvers should only be used as needed to achieve negative margins because their routine application has not been shown to confer any survival benefit. Similarly, anatomic resection of segments IVB and V has been shown to have increased morbidity without survival benefit when compared with nonatomic (2-cm rim) resection.

If radical resection is attempted and negative margins are achieved, the most frequent site of recurrence is distant disease as opposed to locoregional recurrence. This reflects the aggressive biology of the tumor and potentially points to a role for adjuvant therapy in advanced cases. There is a dearth of information regarding optimal regimens and timing. Patients with gallbladder cancer were included in the BILCAP trial, where adjuvant capecitabine was shown to improve survival in the aggregate per protocol analysis. Although capecitabine did not appear to be associated with a clear survival benefit for gallbladder cancer on subset analysis, the study was not intended nor powered to answer this specific question. Combination gemcitabine and cisplatin is frequently used in the setting of unresectable disease and is another potential option, although level I data to support this regimen are still lacking. The ongoing ACTICCA-1 RCT is investigating this combination, with the control arm having recently been changed from observation to capecitabine in response to the BILCAP trial.

OS in gallbladder cancer is poor because of a combination of aggressive biology and advanced stage at diagnosis in most patients. T-stage is a surprisingly good discriminator of OS absent other criteria in gallbladder cancer because more advanced T-stage is directly correlated with higher rates of advanced disease. Additionally, in patients with T2 tumors, the location of the tumor seems to impact OS. Tumors located on the peritoneal side of the gallbladder, designated T2a in AJCC 8th edition, have higher 3- and 5-year survival versus those on the hepatic parenchymal side (T2b; 73.7% vs. 52.1% and 64.7% vs. 42.6%, respectively; Fig. 101B.8 ). Beyond staging, for patients who undergo curative-intent resection, positive margins, lymph node metastasis, poor pathologic differentiation, and the presence of ascites have all been associated with worse OS. Common bile duct involvement of the primary tumor is also associated with worse OS in some analyses. Several nomograms have been developed to help predict survival, with some specifics regarding geographic areas reflecting a potential difference in disease biology. One robust model developed from the SEER database identified T-stage, lymph node metastases, CA19-9, surgical margin, tumor grade, and the presence of high-risk features including lymphovascular and perineural invasion as factors that can be used to determine prognosis ( Fig. 101B.9 ). For further information regarding gallbladder cancer, please see Chapter 49 .