Liver resection for benign and malignant conditions has evolved significantly over the past 2 decades. Moreover, considerable interest in the field of liver surgery has led to an increase in the number of surgeons subspecializing in hepatobiliary surgery. This interest has brought significant innovation and evolution to the field including new technologies in minimally invasive approaches, expanded indications for patients with liver metastases, and the ability to plan and perform more complex resections. At the same time, improvements in patient selection, liver function assessment, and perioperative care have significantly improved the safety of liver resection at experienced tertiary centers. The concepts and principles of liver surgery are expertly reviewed in Chapters 56 (Hepatic Abscess and Cystic Disease of the Liver) and 57 (Benign Liver Neoplasms). Herein, we provide a perspective on recent advances in the field of liver resection, their impact on patient outcomes, and where future developments are anticipated.

Indications for Resection

Although the majority of liver resections today are performed for malignant indications, hepatobiliary surgeons should be familiar with the management of benign conditions as well (Table 60-1). Benign simple cysts can typically be characterized with cross-sectional imaging and do not require resection unless they are symptomatic (unusual unless >10 cm) or contain features worrisome for biliary cystadenoma or cystadenocarcinoma. In the former (simple cyst), laparoscopic fenestration is the preferred treatment, whereas in the latter, formal resection or enucleation should be performed. One notable and frequent cause of liver cystic disease globally is echinococcal or hydatid cyst. Surgical resection of the cyst-bearing area of the liver or operative drainage of the cyst (to ensure prevention of spillage and potential anaphylactic shock) is occasionally required. Despite improvements in antimicrobial therapy and percutaneous aspiration and drainage techniques, the need for occasional surgical management of nonechinococcal hepatic abscesses remains. Finally, hepatic adenomas are benign tumors of the liver that have a low rate of malignant transformation and risk of rupture that increases with size, pregnancy, and location of the tumor. Most authors therefore recommend resection of lesions >5 cm or in women of childbearing age with tumors at risk for rupture. Hemangiomas and focal nodular hyperplasias are benign conditions that do not require resection (or surveillance) unless they are symptomatic, which, in general, is rare.

However, the vast majority of liver resections performed by experienced hepatobiliary surgeons are for malignant indications. Although hepatocellular carcinoma (HCC) is one of the most common solid tumors worldwide, only a minority of patients are candidates for resection because of either poor underlying liver function or advanced disease (eg, multifocality, extrahepatic disease, macrovascular invasion). Unlike HCC, cholangiocarcinoma more commonly develops in patients with normal liver function. Resection of cholangiocarcinomas often requires extended resections with or without biliary reconstruction but is nevertheless recommended in patients with adequate liver function and absence of metastatic disease. Similarly, most liver metastases occur in patients with normal liver function, and colorectal liver metastases (CRLM) are now the most common indication for liver resection among Western hepatobiliary surgeons. Largely based on nonrandomized prospective and retrospective data, liver resection for CRLM when feasible has become largely accepted as standard of care given its association with improved survival rates. Other liver metastases from various histologies, for example, neuroendocrine liver metastases, are also regularly considered for surgical resection.

Patient Selection

Although advances in perioperative anesthesia, minimally invasive surgery, parenchymal transection techniques, and enhanced recovery protocols have been instrumental, patient selection is paramount to optimizing postoperative outcomes following liver resection. In general, patient selection for liver surgery should be conducted along 3 domains: physiologic, oncologic, and technical. Physiologic resectability refers to the patient’s capacity to safely tolerate major abdominal surgery. Multiple risk calculators, frailty indices, and other measures have recently been developed to identify patients at highest risk for major complications.^1,2 Next, oncologic resectability refers to the indications for resection based on the underlying tumor biology. Consideration should be given to the presence of extrahepatic disease, the histopathologic features, disease-free interval (for liver metastases), response to previous therapies (if applicable), and, increasingly, the molecular features of the tumor. Finally, technical resectability should be evaluated based on high-quality cross-sectional imaging. In general, resectability requires retention of 2 contiguous liver segments with adequate vascular inflow, outflow, and biliary drainage as well as sufficient future liver remnant (FLR) volume and function to prevent postoperative hepatic insufficiency (PHI).

Accurate preoperative assessment of the FLR has been one of the most important advances over the past 2 decades, leading to improved risk stratification as well as prevention of PHI and postoperative mortality. Since FLR volume correlates with function and the risk of PHI, a systematic analysis of liver volumetry is imperative in patients undergoing extended hepatectomies or those with compromised liver function. Previous studies have identified FLR size thresholds at which the risk of PHI is prohibitively high: <20% in chemotherapy-naïve patients, <30% in chemotherapy-treated patients, and <40% to 50% in patients with cirrhosis.³ In addition to volumetry, several modalities now aim to directly estimate liver function. For example, since indocyanine green (ICG) is exclusively cleared by the liver, the ICG clearance test is useful because the retention rate at 15 minutes has been correlated with postoperative mortality.⁴ In addition, ^99mTc-galactosyl human serum albumin (GSA) scintigraphy, which uses an analogue ligand of asialoglycoprotein that binds to asialoglycoprotein receptors on the hepatocyte cell membrane, has recently been introduced as a more sensitive indicator of liver function and, when combined with single-photon emission computed tomography (SPECT)/computed tomography (CT), can directly estimate function of the FLR.⁵

FLR Augmentation

Limiting liver resection to patients with adequate FLR volume and function is essential to minimizing postoperative morbidity and mortality. Nevertheless, for patients with inadequate FLR by volumetric analysis, advances in FLR augmentation strategies have expanded the proportion of patients eligible for complex liver resection. Portal vein embolization (PVE) diverts portal blood flow and its inherent growth factors preferentially to the FLR, resulting in 30% to 40% hypertrophic response in most patients.⁶ PVE is most frequently used among patients undergoing extended right hepatectomy or as part of a 2-stage hepatectomy (TSH) strategy.⁷ While various techniques have been described, PVE should be performed in a transhepatic, ipsilateral fashion using microsphere particles and can be extended to segment IV when extended hemihepatectomy is planned.⁸ Among patients who do not achieve an adequate degree of hypertrophy (DH), hepatic vein embolization has been described as a method of promoting additional hypertrophy.⁹

TSH allows for the resection of bilobar CRLM that would otherwise be unresectable using a single-stage technique. In the first stage, the FLR (typically the left lateral section) is cleared of metastatic disease. One to 4 weeks after surgery, a right PVE is performed, which leads to hypertrophy of the FLR. In patients who demonstrate adequate hypertrophy on CT volumetry 3 to 4 weeks after PVE, as well as absence of disease progression, a second-stage operation (typically right or extended right hepatectomy) is performed. TSH with PVE has consistently demonstrated high completion rates, low perioperative morbidity and mortality rates, and excellent overall survival outcomes in experienced centers.¹⁰

The associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) procedure has been introduced as an alternative to traditional TSH.¹¹ In the first stage of ALPPS, right portal vein ligation is combined with parenchymal transection and clearance of the FLR of metastatic disease. The second stage is performed during the same hospital admission 1 to 2 weeks after the first stage and involves completion hemihepatectomy. Although ALPPS results in rapid FLR hypertrophy and higher completion rates compared to traditional TSH, concerns remain over the relatively high postoperative morbidity and mortality as well as uncertain long-term oncologic outcomes.¹²

It is also important to recognize that the liver’s response to PVE provides an estimation of the FLR’s regenerative capacity. For example, patients who experience an absolute DH of <5% have significantly elevated risks of PHI and postoperative mortality.¹³ In addition, a kinetic growth rate (KGR), measured as the DH divided by the number of weeks since PVE, less than 2% is one of the strongest predictors of PHI.¹⁴ Using a traditional TSH approach, the response of the liver to PVE can be measured and second-stage surgery (and its inherent morbidity) avoided in patients with inadequate FLR and/or demonstrated regenerative capacity. Given the high rates of PHI and postoperative mortality following second-stage ALPPS, mostly among patients with an FLR >30%, better predictors of PHI are urgently needed. Recent evidence suggests that KGR¹⁵ and hepatobiliary scintigraphy with SPECT/CT¹⁶ can accurately identify patients at higher risk for PHI following second-stage ALPPS.

Repeat Hepatectomy

As systemic therapies improve and survival durations increase, the proportion of patients who develop hepatic recurrence will only increase. This is especially true among patients with CRLM in whom the recurrence rate following liver resection approaches 50%. However, multiple retrospective series have demonstrated the feasibility, safety, and oncologic outcomes of patients undergoing repeat hepatectomy.¹⁷ In general, similar outcomes can be expected in these situations as long as a similar approach to patient selection, optimization, and resection techniques is employed. A trend in recent years has been an emphasis on parenchymal sparing approaches—that is, performing nonanatomic resections or minor anatomic resections (eg, segmentectomy, bisegmentectomy) rather than hemihepatectomies. Indeed, parenchymal sparing hepatectomy allows for the opportunity for repeat hepatectomy in the case of future liver recurrence without compromising margin-negative rates or oncologic outcomes compared to traditional approaches.¹⁸

Definitions and Terminology

Advances in hepatobiliary techniques have permitted complex liver resections to be performed at experienced centers while minimizing the risk of perioperative complications. Much of this advancement can be attributed to an improved understanding of segmental liver anatomy. Couinaud¹⁹ originally described the 8 segments of the liver, each with its own segmental portal pedicle, separated into 4 sectors based on 3 portal scissura marked along the path of the hepatic veins (Fig. 60-1). Since the left hepatic vein divides segments II and III, this partitions the left liver into a left paramedian sector (IV, III) and left lateral sector (II). The Brisbane terminology reclassified the liver into more surgically relevant sections,²⁰ which is important given that en bloc resection of segments IV and III cannot be technically performed while leaving segment II in situ. In this description, the left medial section (IVa and IVb) is distinct from the left lateral section (segment II and III), whereas the right liver sectors and sections are concordant. This standardized classification informs the terminology used today to describe common partial liver resections (Table 60-2).