Surgical Anatomy, Hepatic Blood Flow, and the Biliary Tree

The adult liver can range from around 600 to over 1800 g making the liver one of the heaviest organs in the body. In healthy females the liver ranges in size from 603 to 1767 g, while in healthy males, the liver ranges in size from 968 to 1860 g. In newborns, infants, and children, the liver is also one of the largest organs and the contribution of its weight to total body weight decreases with age. Thus the liver in a term 3 to 3.5 kg newborn can weigh 150 to 170 g, which represents around 5% of the total body weight. In sharp contrast the adult liver represents 2% to 2.5% of the total body weight.

The liver receives approximately 25% of the resting cardiac output (CO). The blood supply is through both the arterial and venous systems in the form of the hepatic artery and the portal vein ( Fig. 16.1 ). The hepatic artery is responsible for 25% to 30% of the blood supply to the liver whereas the portal vein is responsible for 70% to 75%. The hepatic artery arises from the celiac trunk in 80% of the population. In the remainder, it arises from the superior mesenteric artery. After giving rise to the gastroduodenal artery, the common hepatic artery enters the hilum of the liver (porta hepatis) where it further branches into the right and left hepatic arteries, supplying the right and left sides of the liver, respectively. The right hepatic artery gives rise to the cystic artery that supplies the gallbladder. The arteries continue to branch throughout the liver ultimately running through the portal tracts and terminating in the hepatic sinusoids (capillaries). Although part of the venous system, the portal vein is the primary source of oxygenated blood to the liver. The portal vein carries blood from the gastrointestinal tract, pancreas, and spleen to the liver. It drains the superior mesenteric, splenic, and inferior mesenteric veins. It also drains the gastric, cystic, and pancreaticoduodenal veins. The portal vein enters the hilum and, like the hepatic artery, branches into the right and left portal veins, supplying the respective sides of the liver. The portal veins continue to branch throughout the liver in conjunction with the hepatic arteries. As with the arteries, they terminate in the hepatic sinusoids.

Schematic depiction of Couinaud segmental liver anatomy and the normal portal venous structures. Bracketed text shows hepatic segments resected during partial hepatectomies.

Modified from Venook AP, Curley SA. Management of potentially resectable colorectal cancer liver metastases. http://www.uptodate.com/contents/management–of–potentially–resectable–colorectal–cancer–liver–metastases.

Venous drainage of the liver is through the hepatic veins directly into the inferior vena cava (IVC). The right and middle hepatic veins serve the right half and middle portions of the liver, respectively, while the left hepatic vein drains the left half of the liver. The biliary system removes bile from the liver and delivers it to the duodenum through the ampulla of Vater. The intrahepatic bile ducts typically travel with the portal veins, draining into right and left collections systems that ultimately form the common bile duct (CBD; see Fig. 16.1 ).

From a historical perspective, the description of the gross anatomy of the liver has evolved from being rooted in the surface anatomy of the organ to its functional organization. Traditionally, the liver was divided into four lobes based on its surface features: right lobe, left lobe, quadrate, and caudate. The right and left lobes were divided by the falciform ligament, when viewed anteriorly. When viewed from below, the quadrate lobe was bounded by the porta hepatis posteriorly, the gallbladder fossa on the right, and the ligamentum teres on the left. The caudate lobe was bounded by the porta hepatis anteriorly, the IVC on the right, and the ligamentum venosum fissure on the left. In the late 1800s, Sir James Cantlie demonstrated that the right and left hemilivers were defined by independent portal circulations and thus the functional midline of the liver was at the bifurcation of the portal vein, along a line connecting the gallbladder bed and the IVC (“Cantlie’s line”), lateral to the falciform ligament. Cantlie recognized that the line defined a vascular watershed and described its implications for surgical resection of the liver. With advances in hepatic surgery, anatomic descriptions of the liver were developed that further divided the hemilivers into segments based on the vascular distribution and biliary drainage. Each segment has its own independent vascular inflow and outflow, and biliary drainage. As a result, surgical resection of a segment does not compromise adjacent segments. The most commonly used organizational system was developed by Couinaud (see Fig. 16.1 ). In the Couinaud model, the liver is divided into eight segments. The right and left hemilivers are divided at the bifurcation of the portal vein, along the middle hepatic vein. The right, middle, and left hepatic veins divide the liver vertically into four sectors: right posterior, right anterior, left medial, and left lateral sectors. The four sectors are divided in the horizontal plane by the branches of the portal vein, giving rise to the eight segments. In this system, the caudate lobe is referred to as segment 1 and the remainder of the segments are numbered in a clockwise fashion. Segments 2 and 3 are medial to the left hepatic vein, with segment 2 superior to segment 3. Segment 4 lies between the middle and left hepatic vein and is subdivided into 4a (superior) and 4b (inferior) subsegments. Segments 8 (superior) and 5 (inferior) are located between the middle and right hepatic veins. Segments 6 (inferior) and 7 (superior) are located between the right hepatic vein and the edge of the liver. In clinical practice, contrast-enhanced computed tomography (CT) scanning and intraoperative ultrasound are used to define the anatomy unique to each individual and plan for the appropriate resection. To standardize the nomenclature used to describe hepatic resections, the International Hepatopancreatobiliary Association published consensus terminology in 2000 based on the Couinaud segments, known as the Brisbane 2000 terminology. This system of terminology has gained traction but has not yet been uniformly adopted.

Cellular Anatomy

Liver Lobule and Acinus

The cellular architecture of the liver supports its functions of detoxifying the blood and metabolizing nutrients. Histologically, the liver parenchyma can be organized into anatomic units (liver lobules) or functional units (liver acinus). The liver lobule is the basic structural unit of the hepatic parenchyma ( Fig. 16.2 ). The lobule typically appears hexagonal in shape with a portal canal at each corner and a hepatic venule (central vein) located in the center. Through each portal canal run the lymphatics, nerve fibers, and a portal triad. Each portal triad consists of a bile ductule, hepatic arteriole, and portal venule. From a functional standpoint, the acinus is the smallest unit of the liver ( Fig. 16.3A ). It is comprised of a portal tract at the center with central vein at the periphery. Oxygenated and nutrient-rich blood flows from the portal triads to the central hepatic veins through the hepatic sinusoids (see Fig. 16.3B ). The walls of the sinusoids are composed of sinusoidal endothelial cells (SECs) that are separated by fenestrations of 50 to 150 nm in diameter. The fenestrations allow the passage of metabolites, plasma proteins, pharmaceutical molecules, lipoproteins, and other solutes into the space of Disse (SD) surrounding the sinusoids while retaining blood cells in the vessels. Larger macromolecules and potentially immunogenic peptides enter the SD by transcytosis through the SECs. Once in the SD, the molecules are taken up by hepatocytes.

Hepatic lobule. Liver arranged as a series of hexagonal lobules, each composed of a series of hepatocyte cords (plates) interspersed with sinusoids. Each lobule surrounds a central vein and is bounded by six peripheral portal triads (low magnification).

(A) Blood supply of a liver acinus. Schematic representation of the acinus and hepatocyte zones. Hepatocytes can be divided into three zones based on their location along the sinusoid from the portal triad (PT) to the central vein (CV) . Zone 1—Periportal zone—hepatocytes are closest to the portal triad; Zone 3—perivenous, pericentral, or perilobular zone—hepatocytes are closest to the central vein. Zone 2 (intermediary) hepatocytes are in between. (B) Diagram of the sinusoid. The sinusoid is lined with sinusoidal epithelial cells (EC ) separated by fenestrations that control the passage of solutes into the space of Disse. The space of Disse contains hepatic stellate cells (HSC) and is bound by the basolateral (sinusoidal) portion of the plasma membranes of the hepatocytes (HC) . The apical portion of the plasma membranes makes up the bile canaliculus (BC) . Oxygen and nutrient-rich blood flows from the terminal hepatic artery (THA) and portal vein (TPV) through the sinusoid to the central vein (CV) . There is an oxygen gradient along the length of the sinusoid. Kupffer cells (KC ) are resident hepatic macrophages and found in the sinusoids. (C) Hepatocyte zonation. The major metabolic pathways in the hepatocytes along the sinusoid differ based on their proximity to the portal triad or the central vein. Predominant metabolic pathways are listed for the periportal (pp) hepatocytes and the perivenous (pv) hepatocytes. AA, Amino acid; Cho, cholesterol synthesis; CYPs, cytochrome P450 enzymes; Ggn, glycogen; GPX, glutathione peroxidase; GS, glutamine synthesis; GST, glutathione transferase; Lac, lactate.

[B and C] From Kietzmann T. Metabolic zonation of the liver: the oxygen gradient revisited. Redox Biol . 2017;11:622–630.

Myeloid Cells

The myeloid cells that can be found in the liver consist primarily of Kupffer cells (20%-30%) also known as resident tissue macrophages, in addition to dendritic cells and myeloid-derived suppressor cells. At first glance, it may seem that these cells have a less important role than hepatocytes and LSEC. However, while Kupffer cells constitute around 20% to 30% of nonparenchymal cells, they constitute 80% to 90% of all tissue macrophages. Kupffer cells reside in the portal and lobular liver sinusoids where they engulf both infectious and noninfectious particles by phagocytosis. Once phagocytosed, these particles are unable to induce proinflammatory responses in the liver. Thus by prevalence and location, these cells serve critical roles in innate and adaptive immunity by detoxification where they down-regulate potentially proinflammatory triggers that could disrupt hepatic homeostasis.

Dendritic cells and myeloid-derived suppressor cells are the least abundant of myeloid cells. Hepatic dendritic cells are present in the normal liver and reside in the portal area and are believed to promote tolerance to phagocytosed particles. Hepatic myeloid-derived supressor cells suppress immune response in the liver. In acute hepatitis they reduce inflammation and limit tissue injury. Their immune suppressive function has been associated with adverse effects in certain pathologic conditions. In chronic viral hepatitis, they may promote viral persistence. They have also been associated with suppression of immune response to hepatic tumors.

Lymphocytes

Cells of lymphatic origin that can be detected in the liver include natural killer (NK) cells, NK T cells (NKT), mucosal-associated invariant T cells, and γδ T cells in addition to major histocompatibility restricted CD4+ T cells, CD8+T cells, and B cells. These cells are distributed throughout the liver parenchyma and serve critical roles in the innate (NK, NKT, mucosal-associated invariant T cells, and γδ T cells) and adaptive (major histocompatibility restricted CD4+ T cells, CD8+T cells, and B cells) immune responses. These cells work primarily to maintain hepatic homeostasis by promoting tolerance to foreign substances. However, when necessary, these MHC-restricted cells can promote the clearance of foreign substances by expanding in response to them while recruiting additional cells from extrahepatic sources such as the lymph nodes and the spleen.

Drug Metabolism

The vast majority of drugs used in the conduct of anesthesia are metabolized in the liver. A variety of enzymes convert drug molecules into more water-soluble (hydrophilic) molecules or compounds to facilitate their excretion. These enzymes are designated as being either part of the Phase I pathway or Phase II pathway based on the types of reactions they mediate. Phase I enzymes consist of the cytochrome P450 family of enzymes that convert lipophilic drug molecules to hydrophilic molecules primarily through oxidation, reduction, or hydrolysis. Non-CYP450 enzymes include monoamine oxidases, alcohol dehydrogenases, and aldo-keto reductase. The phase II pathway consists of the conjugation of the products of the phase I pathway with hydrophilic endogenous moieties to make them more water-soluble. Polar molecules may undergo Phase II metabolism without having undergone Phase I metabolism. The most common Phase II reaction is glucuronidation, which is the conjugation of the drug compound to glucuronic acid. This reaction is carried out by a family of enzymes known as uridine 5′-diphospho-glucuronosyltransferases. Other Phase II enzymes include sulfotransferases (SULT), glutathione S-transferases (GST), and catechol O-methyltransferases. The phase III pathway involves the excretion of compounds into the sinusoids or canalicular bile by molecular transporters that are transmembrane proteins which facilitate the movement of large or ionized molecules across cell membranes. The majority of these transmembrane proteins are part of a superfamily of ATP-binding cassette (ABC) transporters that use ATP to actively transport molecules. Common ABC-transporters include multidrug resistance protein (MDR), cystic fibrosis transmembrane conductance regulator, and multidrug resistance–related protein (MRP).

Some orally administered medications undergo extensive metabolism in the gut or liver prior to entering the systemic circulation. This metabolism is termed the first-pass effect and is responsible for the lower oral bioavailability of these medications.

Drug metabolism is affected by a number of factors including genetic polymorphisms of metabolic enzymes, age, gender, pregnancy, liver disease, and concomitantly administered medications. The expression and function of Phase I and Phase II enzymes are reduced in neonates. The activities of some CYP450 enzymes are increased in women compared to men. Genetic polymorphisms in drug metabolizing enzymes and transporters can lead to wide variations in the pharmacokinetics of some drugs such as warfarin, with some patients having lower rates of metabolism based on the specific CYP450 polymorphism they carry. The concomitant administration of medications may also influence drug metabolism. A number of commonly encountered medications can serve as inducers or inhibitors of the enzymes involved in the different phases of drug metabolism. Table 16.1 lists some of the commonly used drugs, which are metabolized and excreted by each of the three phases along with drugs that may serve as inhibitors or inducers for each phase.

See the pharmacokinetics chapter for further discussion of hepatic extraction ratio.

Protein Metabolism

The liver is responsible for the synthesis and catabolism of proteins, amino acids, and peptides. It is the site of synthesis for 80% to 90% of the circulating proteins including hormones, coagulant factors, cytokines, and chemokines. As such it plays a significant role in the functioning of the body. Albumin is the predominant protein produced by the liver, accounting for over 50% of total plasma protein. It functions to transport lipids and hormones and maintain blood volume. The liver plays a central role in protein degradation. Amino acids are catabolized through one of two reactions: deamination or transamination. Both reactions lead to the production of ammonia, which the liver converts to urea through the urea cycle. Urea is then excreted by the kidneys in the urine.

Carbohydrate Metabolism

The liver is primarily responsible for storing and releasing glucose to meet the body’s needs. In the postprandial state, the liver stores glucose through glycogenesis. Once the glycogen stores are complete, the liver converts excess glucose into fat through lipogenesis. In the fasting state, the liver provides the body with glucose by breaking down glycogen (glycogenolysis) or by generating glucose from carbohydrate precursors (gluconeogenesis).

Lipid Metabolism

The liver plays an important role in lipid metabolism. Nonesterified fatty acids can arise from the lipase-mediated breakdown of complex lipids, or from thioesterase-mediated hydrolysis of fatty acid-CoA. These fatty acids can enter the liver following oral intake or they can enter the liver following the breakdown of adipose tissue. In the liver, fatty acid oxidation is regulated by two main factors: the supply of fatty acids to the liver (via lipolysis), and the amount of microsomal esterification that occurs. Lipid metabolism is also influenced by the carbohydrate metabolism, as the acetyl-CoA formed during carbohydrate metabolism can be utilized to synthesize fatty acids. Fatty acids can undergo biotransformation to supply energy for the needs of the body. Alternatively, the liver can convert amino acids and intermediate products of carbohydrates into fats and transport them to the adipose tissues.

Bile and Enterohepatic Circulation

The adult liver produces approximately 400 to 600 mL of bile each day. Bile facilitates the excretion of toxins as well as the absorption of dietary fats. It is the mechanism of excretion for compounds with molecular weights greater than 300 to 500 Daltons that are not readily excreted by the kidneys. It is used to excrete a host of endogenous and exogenous compounds, including bile acids, bilirubin, phospholipids, cholesterol, drugs, toxins, steroid hormones, and water-insoluble porphyrins. Box 16.1 lists drugs, chemicals, and their metabolites that are excreted in the bile. The other major function of bile is to assist in the digestion and absorption of dietary fats, cholesterol, and vitamins. Bile is 95% water by volume, with the remainder consisting of bile acids, phospholipids, cholesterol, bilirubin, as well as other exogenous and endogenous substances. The two primary bile acids are cholic acid and chenodeoxycholic acid. Bile acids are synthesized by hepatocytes from cholesterol. They are then conjugated to reduce hepatotoxicity and increase solubility and secreted into the canaliculi. The canaliculi drain into the biliary ductules, which connect to form hepatic ducts. The walls of the intrahepatic bile ducts are made up of cholangiocytes that modify the volume and composition of the bile. The ducts ultimately form the left and right hepatic ducts, which join into the common hepatic duct. Bile is stored and concentrated in the gallbladder, which connects to the biliary tree through the cystic duct. The common hepatic duct and cystic duct join to form the CBD, which connects to the duodenum through the sphincter of Oddi (hepatopancreatic sphincter). Following the ingestion of food, fatty acids in the duodenum stimulate the release of cholecystokinin (CCK) which causes the gallbladder to contract and the sphincter of Oddi to relax leading to the release of bile into the duodenum. The bile acids emulsify dietary fats and facilitate their absorption. The vast majority (95%) of the bile acids released into the duodenum are reabsorbed in the terminal ileum and returned to the liver to be reused. This pathway for recycling bile acids is known as the enterohepatic circulation (EHC). Enterohepatic cycling can impact the pharmacokinetics and pharmacodynamics of drugs that undergo biliary excretion by increasing their bioavailability, reducing their elimination, as well as altering their plasma concentration curves. The effect of EHC on the properties of a drug depends on the physiologic activity of the excreted form of the drug (i.e., prodrug or activated form), the ease with which the excreted form is reabsorbed through the intestines, and whether it is recycled through the liver into the bile or the systemic circulation. In the case of some drugs, EHC can lead to secondary and tertiary peaks in plasma concentration as the drug is recycled into the system.

Role of the Liver in Coagulation

The liver plays a significant role in the coagulation system. It synthesizes all coagulation factors except factors III (thromboplastin), IV (calcium), and VIII (von Willebrand factor [vWF]). It also synthesizes proteins that regulate coagulation and fibrinolysis such as protein S, protein C, plasminogen activator inhibitor, and antithrombin III. Furthermore, it removes activated clotting and fibrinolysis products through the hepatic reticuloendothelial system. A number of factors require vitamin K to become active. Coagulation factors II, VII, IX, X, as well as protein C and protein S undergo posttranslational modification with vitamin K to become active. Briefly, glutamic acid in the amino terminus of these proteins is converted to gamma-carboxyglutamic acid. These gamma-carboxylated procoagulants can then bind calcium ions and form bridges to phospholipid surfaces that are essential for the formation of activation complexes. Warfarin acts by inhibition gamma-carboxylation. In addition to these vitamin K–dependent factors, hepatocytes also synthesize factor V, XIII, fibrinogen, antithrombin, α ₂ plasmin inhibitor, and plasminogen. Thrombomodulin, tissue plasminogen activator, tissue factor plasma inhibitor, vWF, and urokinase are not synthesized in the liver. Instead these proteins are synthesized in endothelial cells, whereas urokinase is expressed by endothelial cells, macrophages, and renal epithelial cells. Tissue plasminogen activator is primarily removed from the bloodstream through the hepatic reticuloendothelial system.

Heme Metabolism, Bilirubin, and Porphyrias

The liver is involved in both heme synthesis and metabolism. Eighty to 90% of heme synthesis takes place in the bone marrow with the resultant heme used to produce hemoglobin. Most of the remainder of the heme is produced in the liver and used primarily to synthesize cytochrome P450 enzymes. Whereas the rate of heme synthesis in the bone marrow is a function of the availability of iron, the rate of synthesis in the liver is a function of the available free heme pool in the body. Heme is synthesized through an eight-step enzymatic cascade known as the Shemin pathway. Synthesis begins with glycine and succinyl CoA and proceeds through porphyrinogen intermediaries. A deficiency in any of the enzymes involved in heme synthesis leads to the development of porphyria. The specific type of porphyria and its clinical manifestations depend on the specific enzyme that is deficient and the substrate that accumulates as a result. The most common porphyria is acute intermittent porphyria with an estimated incidence of 5 to 10 per 100,000. It is caused by a deficiency in porphobilinogen deaminase, which catalyzes the conversion of porphobilinogen to hydroxymethylbilane. Patients typically have adequate levels of the enzyme for heme homeostasis; however, in response to endogenous or exogenous triggers that induce the Shemin pathway, the capacity of the system is exceeded and they accumulate precursors leading to symptoms. Common triggers include erythromycin, trimethoprim, rifampicin, phenytoin, and barbiturates. Clinical symptoms of an attack include severe, poorly localized abdominal pain (in >90% of cases), nausea, vomiting, agitation, and confusion. Hyponatremia occurs in 40% of attacks. Change in urine color to dark red (especially on exposure to light) is a common finding. Treatment consists of discontinuing the triggering agent, administering pain medication, carbohydrates, and hematin.

Bilirubin is a product of heme catabolism. The primary source is senescent erythrocytes that are phagocytosed by macrophages in the spleen, liver, and bone marrow. The released heme is metabolized by heme oxygenase into bilirubin, yielding carbon monoxide and iron in the process. Unconjugated bilirubin is water insoluble and thus tightly bound to albumin in the circulation. Hepatocytes convert bilirubin into a water-soluble form by conjugating it to glucuronic acid via the enzyme glucuronyl transferase. Conjugated bilirubin is then transported across bile canaliculi and excreted in the bile. In the colon, bilirubin is deconjugated, metabolized by bacteria, and converted into urobilinogen. Urobilinogens are either reabsorbed through the EHC or excreted in the urine and stool, giving urine and stool their characteristic colors.

Hepatic Regulation of Hormones

The liver can participate in endocrine functions through hormone synthesis or hormone degradation. Hepatocytes synthesize hormones or prohormones such as hepcidin, insulin-like growth factor, and angiotensinogen, respectively. In addition to these hormones, thrombopoietin is also synthesized by hepatocytes and LSECs. These hormones and prohormones have specialized roles in the human body. Thus hepcidin is responsible for iron homeostasis and regulates intestinal iron absorption, plasma iron concentrations, and tissue iron distribution by inducing degradation of the hepcidin receptor, ferroportin. Insulin-like growth factor promotes systemic growth, especially bone growth in children. Angiotensinogen, the precursor of all angiotensin proteins, regulates the systemic blood pressure as well as the water and sodium composition of the body. Thrombopoietin regulates platelet production by stimulating production and differentiation of megakaryocytes. In addition to hormone synthesis, the liver participates in endocrine function by inactivating many hormones, including thyroxine, aldosterone, antidiuretic hormone, estrogens, androgens, and insulin.

Clinical Assessment

There are often no signs or symptoms of liver disease until it is quite advanced. Even then, the only clues may be mild or nonspecific symptoms such as loss of appetite, fatigability, malaise, disrupted sleep patterns, or subtle cognitive changes. Major risk factors for liver disease include: alcohol use; illicit drug use; sexual promiscuity; blood transfusions; exposure to hepatotoxins; prior bouts of jaundice; and a family history of genetic diseases such as hemochromatosis, α ₁ -antitrypsin deficiency, and Wilson disease. Patients with advanced liver disease may have these nonspecific symptoms, as well as pruritus, easy bruising, and changes in urine or stool color. In advanced liver disease physical examination findings include jaundice, scleral icterus, ascites, spider angiomas, xanthelasma, asterixis, and palmar erythema.

Standard Laboratory Tests

Standard panels used to evaluate the hepatobiliary system are often called “liver function tests” ( Table 16.2 ). In fact, these tests do not measure specific liver functions. Instead, they help define broad categories of hepatobiliary pathology: hepatitis, hepatobiliary dysfunction, or insufficient protein synthesis.

Detection of Hepatocellular Injury

Detection of Cholestatic Disorders

Serum Bilirubin

Serum bilirubin is the most commonly used test to monitor for excretory dysfunction of the liver. Bilirubin is measured in serum assays as either direct or indirect bilirubin. Direct bilirubin measures the water-soluble form, which interacts directly with the assay’s reagents. While direct bilirubin correlates with conjugated bilirubin levels, and indirect with unconjugated levels, the two terms are not synonymous. The level of unconjugated bilirubin is typically underestimated by the indirect bilirubin test. Distinguishing conjugated from unconjugated bilirubin is key to the differential diagnosis. The presence of bilirubin in the urine can also help differentiate the clinical cause. Bilirubinuria usually reflects conjugated hyperbilirubinemia, since only the water-soluble conjugated form is excreted by the kidney ( Fig.16.4 ).

Differential diagnosis of jaundice depending on whether bilirubin is or is not present in the urine.

Elevation in the concentration of unconjugated bilirubin is either due to excessive heme breakdown, or the inability of the liver to conjugate bilirubin. One major cause is hemolysis, in which increased erythrocyte breakdown creates more unconjugated bilirubin than the liver is able to conjugate. Massive hemolysis will cause elevations in both forms of bilirubin, but with unconjugated predominance. Gilbert syndrome is a benign condition with genetically low levels of the hepatic enzyme glucuronyl transferase, associated with mild or intermittent elevation in unconjugated bilirubin. Various drugs may also produce unconjugated hyperbilirubinemia. High levels of unconjugated bilirubin are neurotoxic, particularly in infants.

Conjugated hyperbilirubinemia occurs due to either problems with the secretion of conjugated bilirubin from hepatocytes into canalicular bile or to the blockage of bile flow within the hepatobiliary tree. This can be due to a genetic defect in the excretion of bilirubin, or cholestasis (either intrahepatic or extrahepatic). Intrahepatic cholestasis is caused by an inflammatory or infiltrative process compressing small intrahepatic bile ducts, while extrahepatic cholestasis is due to biliary obstruction such as from stones or a pancreatic mass. Primary sclerosing cholangitis (PSC) can involve both intra- and extrahepatic bile ducts.

Assessment of Hepatic Protein Synthesis

Testing for Diagnosis of Specific Diseases

In addition to the above standard laboratory tests, targeted testing can aid in the diagnosis of specific hepatobiliary diseases. Such testing includes serologic testing for viral and autoimmune diseases, genetic testing, and tumor marker assays. Viral markers, including antibodies, antigens, and genetic material, are used to diagnose hepatitis from hepatotropic viruses (A, B, C, D, and E) and herpesviruses such as cytomegalovirus and Epstein-Barr virus. Patients with acute or chronic hepatitis B infection and those with immunity due to resolved infection or vaccination can be distinguished based on measurements of hepatitis B surface antigens and surface and core antibodies ( Table 16.3 ). HBV DNA testing has also emerged to monitor treatment effect and the development of chronic hepatitis B after resolution of acute hepatitis.

Testing for hepatitis C is recommended for those with various risk behaviors and exposures (such as history of intravenous drug use, long term hemodialysis, or HIV infection) and as part of an evaluation for unexplained liver disease. American guidelines also recommend one-time screening for hepatitis C in all persons in the United States born between 1945 and 1965, as this birth cohort has the highest prevalence of hepatitis C. Screening entails testing for antibodies to HCV; a positive screen indicates exposure to the virus. Active infection can then be confirmed with nucleic acid testing for HCV RNA. Sustained viral response to treatment is defined as the absence of HCV RNA detected at least 12 weeks after the completion of therapy. Currently, patients also receive HCV genotype testing prior to the initiation of treatment. New direct antiviral agents are highly effective in achieving sustained viral response for most genotypes and most clinical conditions. Therefore genotype testing may soon no longer be routinely required.

Rapid HCV antibody testing and point of care RNA tests are useful in community-based settings serving high-risk populations and remote settings without access to centralized laboratories. The measurement of HCV core antigen (HCVcAg) can diagnose active hepatitis C infection with high sensitivity and specificity as a single test, although it is currently not the standard of care. HCVcAg measurement is particularly useful as a rapid, single test for diagnosis in populations who are at high risk of loss to follow-up. However, it is not yet available as a point of care test.

Markers of hepatic malignancy include alpha fetoprotein (AFP) and protein induced by vitamin K absence or antagonist-II (PIVKA-II). AFP is a glycoprotein synthesized in the liver, fetal yolk sac, and GI tract. The normal value of AFP is less than 20 ng/mL. A significantly elevated concentration of AFP, often greater than 1000 ng/mL, is common in patients with hepatocellular carcinoma (HCC). AFP is also elevated in testicular germ cell tumors, and less frequently in prostate and other gastrointestinal cancers. The concentration of AFP is used to monitor HCC disease progression and response to treatment. It is also widely used to screen for HCC in high-risk patients, along with ultrasound. However, AFP’s role in HCC surveillance has recently been debated, as studies suggest that AFP has insufficient sensitivity as a screening tool for HCC, particularly for small and solitary tumors. AFP may be normal in 40% of HCC patients. PIVKA-II, also known as des-γ-carboxylated prothrombin, is a biomarker for HCC with high specificity. With a shorter half-life than AFP, PIVKA-II is useful in monitoring for treatment response and recurrence. PIVKA-II is also associated with worse survival. AFP and PIVKA-II combined can diagnose HCC with further increased sensitivity and specificity.

Testing in Management and Prognosis of Liver Disease

Laboratory tests can help define the type of liver dysfunction, and even identify specific causes. In addition, some of these markers are used to assess disease severity, monitor disease progression and treatment, and predict mortality. Taken alone, thrombocytopenia is the earliest sign of cirrhosis among the routine serum tests, indicating both decreased hepatic function and splenic sequestration from portal hypertension. Some markers are combined to create scoring systems, thereby improving the sensitivity and specificity of single tests. For example, the Model for End-Stage Liver Disease (MELD) score (a calculation that combines INR, bilirubin, and creatinine) was created to assess risk of mortality in cirrhotic patients undergoing transjugular intrahepatic portosystemic shunt (TIPS) procedures. The MELD score now is used for liver transplant listing, as it correlates with wait-list mortality. Standard tests such as INR, lactate and platelet count, as well as novel quantitative tests have been used to predict short-term outcome after liver transplantation. The HCC-MELD score, which combines AFP, MELD, and tumor size, can predict survival after liver transplantation in HCC patients. Elevated bilirubin predicts poor prognosis in acute or chronic liver disease, and AP and bilirubin together predict poor outcome in primary biliary cirrhosis.

Measurement of Hepatic Blood Flow

Blood flow to the liver can be measured using clearance techniques, indicator dilution techniques, and direct measurements.

Radiologic Methods

Radiologic techniques play important roles in diagnosing liver disease. Splenoportography evaluates the splenic and portal veins; it is useful in identifying varices and thromboses. Portal venography uses three-dimensional CT to create vascular maps of portosystemic collateral vessels. These noninvasive imaging techniques are particularly useful in surgical planning, such as before liver transplantation.

Inexpensive and routinely available, ultrasound can help diagnose cirrhosis with high sensitivity and specificity. Ultrasonic signs of cirrhosis include surface nodularity, hepatomegaly, and hypertrophy of the caudate lobe, as well as signs of portal hypertension, such as ascites, splenomegaly, and increased portal vein diameter. Doppler ultrasonography can show diminished portal flow and portal venous flow reversal. Fibrosis can also be estimated using elastography, a noninvasive measure of liver stiffness (LS). A shear wave is delivered across the liver via an ultrasound probe, and the wave propagation is then measured. Transient elastography is reproducible and easily performed in the office. Magnetic resonance elastography is especially useful in obese patients and those with ascites, where transient elastography is less accurate. Stiffness is a continuous measure which correlates with the histological stage of fibrosis, with high sensitivity and specificity. The combination of serum tests along with elastography offers much improved accuracy in diagnosing fibrosis. These tests are now widely used in practice, although they have not yet supplanted the role of liver biopsy.

Patients with chronic liver disease and chronic hepatitis B virus carriers are at increased risk for developing HCC and require regular screening for the cancer. Ultrasound is the most widely used screening tool for HCC worldwide. Multiphasic, contrast-enhanced CT and MRI are used to evaluate focal liver lesions suspicious for HCC with high sensitivity and specificity. While normal liver parenchyma receives only 25% blood flow from the hepatic artery (the rest via the portal vein), HCC cells receive their blood supply primarily from the hepatic artery. Contrast-enhanced imaging reflects this fact with HCCs showing hyperintensity in the arterial phase and washout during the portal venous phase. The Liver Imaging Reporting and Data System (LI-RADS) is a standardized tool for the radiologic classification of liver lesions in patients with chronic liver disease. Use of the LI-RADS has improved radiologic diagnostic accuracy and reduced the need for biopsy in these patients. In fact, confirmatory biopsy is rarely used, even prior to liver transplantation for HCC.

Cholestatic Disease

Cholestasis is the impairment of bile production or flow. Cholestatic disorders are characterized primarily by elevations of serum AP and gamma-glutamyl transferase (GGT), with or without elevations in bilirubin. Cholestasis can present as an acute or chronic condition and is thought to affect up to 20% of the population. The impairment of flow leads to an increase in the concentration of bile salts in the serum and hepatocytes. As the bile salts accumulate in the liver, they dissolve hepatocyte cell membranes, leading to the release of AP and GGT, resulting in an increase in the serum concentrations of these enzymes. Serum concentrations of AST and ALT, which are found in the cytoplasm of hepatocytes, are not increased to as great an extent. As the disease progresses, the serum concentration of conjugated bilirubin increases, giving rise to jaundice. The majority of cases of cholestasis are benign; however, severe and prolonged cholestasis can lead to cirrhosis.

Signs and symptoms of cholestasis include fatigue, pruritus, dark urine, and pale stools. An elevation of AP on screening laboratory studies may be the first indication in asymptomatic patients. The evaluation consists of a history and physical exam followed by abdominal ultrasonography. Abdominal ultrasonography is used to distinguish between extrahepatic and intrahepatic cholestasis. Extrahepatic cholestasis results from mechanical obstruction of the extrahepatic bile ducts, typically by stones, strictures, or masses leading to ductal dilation. Definitive diagnosis and treatment is by endoscopic retrograde pancreatography (ERCP). If the etiology is unclear, further imaging can be conducted using endoscopic ultrasound or magnetic resonance cholangiopancreatography to identify the biliary lesion leading to obstruction. Normal imaging results indicate intrahepatic cholestasis. Intrahepatic cholestasis can have several etiologies including immune-mediated, infectious, drug-induced, paraneoplastic, and ischemic. The first step in the evaluation is the measurement of antimitochondrial antibody (AMA) titers to rule out primary biliary cholangitis (PBC). Further evaluation may require ERCP and liver biopsy.

Primary Biliary Cholangitis

Formerly known as primary biliary cirrhosis, PBC is an autoimmune disease that is characterized by the destruction of intrahepatic cholangiocytes and the presence of AMAs. It is one of the most common cholestatic liver disorders. Reported prevalence rates vary from 20 to 490 cases per million inhabitants with the highest rates reported in the United States. Over 90% of patients are female and the average age at diagnosis is typically in the sixth decade. The diagnosis is made based on an elevation of AP over 1.5 times the upper limit of normal for a duration greater than 24 weeks along with the presence of an AMA titer greater than 1:40. A liver biopsy in these patients will reveal interlobular bile duct lesions. Up to 60% of patients may be asymptomatic at diagnosis. If present, the most common symptoms at presentation are fatigue and pruritus. A large number of patients may present with other autoimmune disorders such as keratoconjunctivitis sicca and Raynaud phenomenon. PBC may be associated with osteopenia as well as hyperlipidemia. The disease develops as a result of the loss of tolerance to mitochondrial antigens, leading to autoimmune-mediated destruction of biliary epithelial cells (cholangiocytes). PBC can lead to cirrhosis and liver failure, ultimately requiring transplantation.

Treatment consists of ursodeoxycholic acid (UDCA) administration. UDCA, a bile acid, is thought to act by modifying the bile acid pool and reducing inflammation. Approximately 60% to 70% of patients with PBC respond to UDCA with improvement in serum AP and bilirubin levels and a delay in the progression to cirrhosis and liver transplantation. Patients who have a suboptimal response to UDCA have an increased risk of developing complications and liver failure. In these patients, the administration of obeticholic acid, a farnesoid-X receptor agonist, has been shown to reduce the serum levels of AP. Fibrates and corticosteroids have also shown benefit when combined with UDCA in patients who are unresponsive to UDCA alone.

Primary Sclerosing Cholangitis

PSC is a rare cholestatic, immune-mediated disease that occurs most frequently in men aged 30 to 40 years old. It is characterized by inflammation and fibrosis of the bile ducts. The majority of patients with PSC have ulcerative colitis as well. Approximately half of the patients are asymptomatic at diagnosis. The most common presenting symptoms are fatigue and pruritus. Blood tests demonstrate a cholestatic pattern. Patients may also have hypergammaglobulinemia and elevations in serum autoantibodies. Diagnosis is made based on the presence of multifocal strictures and segmental biliary dilations on cholangiography, and the exclusion of a number of other possible etiologies including bacterial cholangitis, choledocholithiasis, and surgical biliary trauma. Patients have an increased risk of developing hepatobiliary carcinoma. The disease is progressive, with biliary fibrosis leading to cirrhosis and ultimately end-stage liver disease (ESLD). Liver transplantation is the only effective therapy.

Cirrhosis

Cirrhosis is the common pathologic endpoint of a variety of mechanisms of hepatic injury. In the United States, it is the 14th most common cause of death and has a prevalence of 0.3%. The most common causes of cirrhosis in western countries are alcoholic liver disease, hepatitis C, and nonalcoholic steatohepatitis. In the Asia-Pacific region the most common cause is hepatitis B. The morbidity and mortality associated with cirrhosis result mainly from portal hypertension. Chronic liver disease promotes apoptosis and regeneration of hepatocytes. The resultant inflammation leads to parenchymal injury and fibrosis, causing distortion and obstruction of the hepatic vascular architecture. These structural changes increase hepatic resistance to portal blood flow (PBF) and give rise to portal hypertension. Hepatic vascular resistance is further increased due to hepatic endothelial dysfunction, which results in a reduction of vasodilator (primarily nitric oxide [NO]) production and an increase in vasoconstrictor (i.e., thromboxane) production. Portal hypertension and the release of angiogenic factors lead to the development of gastroesophageal varices as well as portosystemic shunts. As portal hypertension worsens, the locally released NO and prostaglandins cause the splanchnic circulation to vasodilate. This further increases portal inflow and exacerbates portal hypertension. Splanchnic vasodilation also leads to systemic hypotension that, in turn, increases sympathetic tone, activates the renin-angiotensin-aldosterone system, and increases antidiuretic hormone release, leading to the development of ascites and the hepatorenal syndrome (HRS). Other significant complications of cirrhosis and portal hypertension are spontaneous bacterial peritonitis, encephalopathy, HRS, portopulmonary hypertension (PoPH), cirrhotic cardiomyopathy, and HCC.

Laboratory abnormalities in cirrhotic patients include elevations in ALT, AST, AP, GGT, as well as serum bilirubin. Hypoalbuminemia and elevations in the PT are indications of hepatic synthetic dysfunction. Although liver biopsy is the gold standard for diagnosing cirrhosis, it is often not necessary in the presence of an appropriate patient history, and laboratory and radiologic data. In patients who are suspected of having cirrhosis, abdominal ultrasonography is used to aid in the diagnosis and evaluation for HCC and portal hypertension. Serologic studies and ultrasound-based techniques for evaluating hepatic fibrosis are also available.

The risk stratification of patients with cirrhosis is based on the presence and degree of portal hypertension, gastroesophageal varices, and synthetic dysfunction. There are two stages of cirrhosis: compensated and decompensated. Decompensated cirrhosis is characterized by the presence of ascites, variceal hemorrhage, and/or hepatic encephalopathy (HE). Compensated cirrhosis is characterized by the absence of these clinical conditions. While the median survival in patients with compensated cirrhosis is greater than 12 years, in those with decompensated cirrhosis it is approximately 2 years. Patients with compensated cirrhosis can be further stratified based on their degree of portal hypertension. Portal hypertension is defined as a hepatic venous pressure gradient (HVPG) greater than 5 mm Hg. Patients with an HVPG greater than 10 mm Hg are deemed to have clinically significant portal hypertension. They are at increased risk of developing varices, overt hepatic decompensation, postsurgical decompensation, and HCC compared to those with mild portal hypertension (HVPG between 5 mm Hg and 10 mm Hg). Although HVPG is the gold standard for monitoring portal hypertension, its invasive nature precludes its routine use. A noninvasive measure of clinically significant portal hypertension is LS using transient elastography, an ultrasound-based approach. An LS value greater than 20 kPa has been shown to effectively discriminate mild from clinically significant portal hypertension. Liver Doppler ultrasound may also aid in the diagnosis. The presence of portosystemic collaterals or reversal of portal venous flow on ultrasound imaging is also sufficient to diagnose clinically significant portal hypertension. In patients with clinically significant portal hypertension, the presence of gastroesophageal varices portends a worse prognosis.

The management of patients with compensated cirrhosis who don’t have gastroesophageal varices focuses on preventing the advancement of the disease by treating the underlying etiology. Patients with varices are treated with nonspecific beta-blockers (i.e., propranolol, timolol), carvedilol or endoscopic variceal ligation to prevent variceal hemorrhage. Nonspecific beta-blocker therapy reduces HVPG.

Acute variceal hemorrhage is a medical emergency. Management consists of airway protection, volume resuscitation, and endoscopy with variceal ligation. The concomitant initiation of antibiotic prophylaxis (ceftriaxone 1 g/24 h) and vasopressor infusion (somatostatin, octreotide, or terlipressin) has been shown to improve outcome and is the standard of care. Once stabilized, preemptive TIPS may be performed in patients who are at high risk of rebleeding.

Ascites

The presence of ascites is a marker of decompensated cirrhosis. The 1-year mortality rate of patients who develop ascites is 20%. Patients who develop ascites are at increased risk of developing further complications including spontaneous bacterial peritonitis and acute kidney injury (AKI). Management consists of dietary sodium restriction, nonspecific beta-blocker therapy to reduce portal pressure, and diuretic administration (spironolactone, furosemide). Midodrine can be added for refractory ascites. TIPS can improve transplant-free survival in certain patients with refractory ascites.

Hepatic Encephalopathy

HE is brain dysfunction secondary to hepatic insufficiency or portosystemic shunting. It is a sign of decompensated cirrhosis. It occurs in 30% to 40% of patients with cirrhosis and 50% of patients with portosystemic shunts. The clinical presentation is variable and can range from mild changes in personality to disorientation, somnolence, and coma. Asterixis is usually present in the early to middle stages. The severity of HE is typically graded from I to IV using the West Haven criteria with Grade I indicating mild symptoms and Grade IV, coma ( Table 16.4 ). Common precipitating factors of HE include infections, variceal hemorrhage, and diuretic overdose. Diagnosis is based on the exclusion of other causes of brain dysfunction and the application of clinical criteria. Treatment consists of amelioration of precipitating factors and the administration of lactulose, a nonabsorbable disaccharide. Rifaximin can be added on to prevent recurrence.

Pulmonary Complications

Hepatopulmonary syndrome (HPS), PoPH, and hepatic hydrothorax (HH) are pulmonary conditions that can develop in patients with cirrhosis and advanced liver disease. HPS has been reported in up to 30% of patients presenting for liver transplantation evaluation and is associated with increased mortality independent of the severity of liver disease. It is defined as abnormal alveolar to arterial oxygen gradient (≥15 mm Hg) while breathing room air in sitting position. It is caused by intrapulmonary vascular dilation in the setting of cirrhosis. Its severity is determined by the PaO ₂ : greater than 80 mm Hg is mild, 60 to 80 mm Hg is moderate, 50 to 60 mm Hg is severe, and less than 50 mm Hg is very severe. Patients may present with dyspnea at rest or on exertion. Approximately 25% of patients have platypnea (dyspnea when moving from supine to standing) or orthodeoxia (decrease in PaO ₂ more than 5% or 4 mm Hg when moving from supine to standing). In advanced disease, patients may exhibit digital clubbing and cyanosis. Patients can be screened with pulse oximetry (SpO ₂ < 96% on room air) and diagnosed based on PaO ₂ values. Contrast-enhanced transthoracic echocardiography (TTE) may demonstrate intrapulmonary vascular dilation. No medical therapy has been shown to change outcome in patients with HPS. Supplemental oxygen is provided to maintain SpO ₂ over 88%. Liver transplantation is the only effective treatment. As such, MELD exception points are granted for patients with severe HPS (PaO ₂ 50-60 mm Hg). Patients with very severe HPS (PaO ₂ < 50 mm Hg) have an increased risk for complications and mortality following liver transplantation although this risk may be lower in experienced centers.

PoPH is pulmonary arterial hypertension in the setting of portal hypertension. It is found in 5% of patients presenting for liver transplant evaluation and, in the absence of treatment, is associated with a 1-year survival of 46%. The diagnosis is made by right heart catheterization showing a mean pulmonary artery pressure (mPAP) greater than 25 mm Hg, pulmonary vascular resistance (PVR) greater than 240 dynes/s/cm , and a pulmonary artery wedge pressure less than 15 mm Hg. PoPH is classified as mild (mPAP: 25-35 mm Hg), moderate (mPAP: 35-45 mm Hg), and severe (mPAP more than 45 mm Hg). The pathophysiology of the disease remains to be elucidated. Patients are typically screened using TTE to measure right ventricular systolic pressure (RVSP). Those with RVSP greater than 50 mm Hg undergo right heart catheterization to confirm the diagnosis and determine the degree of PoPH. PoPH is treated with phosphodiesterase-5 inhibitors, prostacyclin analogs, and endothelin receptor antagonists to reduce PVR. mPAP greater than 45 mm Hg is an absolute contraindication to liver transplantation. MELD exception is considered for treatment-responsive PoPH.

HH has a prevalence of 5% to 10% in cirrhotic patients. It develops due to the passage of ascitic fluid from the peritoneal cavity into the pleural space through defects in the diaphragm, most often on the right side. Patients may present with dyspnea, cough, chest discomfort, and hypoxia. Chest radiography and thoracentesis are used for diagnosis. Medical therapy consists of sodium restriction and diuretic administration. Refractory HH is treated with TIPS placement. For patients who have failed TIPS, VATS with pleurodesis is an option.

Hepatocellular Carcinoma

HCC is the most common primary liver malignancy and the third most common cause of cancer death globally. It occurs with greater frequency in men than woman and with a higher incidence in Asia and Africa, although the incidence has been increasing in the United States. Chronic liver disease is the most important risk factor for the development of HCC. Hepatitis B, hepatitis C, hemochromatosis, alcohol use, nonalcoholic fatty liver disease, diabetes mellitus, and obesity all increase the risk of developing HCC. Screening is recommended for all patients with cirrhosis at 6-month intervals using ultrasound with or without serum AFP measurement. Masses over 1 cm should be imaged with dynamic MRI or four-phase multidetector CT to make the diagnosis noninvasively. A percutaneous biopsy is obtained for nodules that have nontypical appearance on imaging. Staging is based on the size of the tumor, degree of extension into adjacent structures, presence of metastases, liver function, and the patient’s physical status. There is no universally accepted system for staging. The tumor, node, metastasis and the Barcelona staging system (BCLC) are two of the more commonly used systems.

Surgical resection of HCC is the definitive treatment for the disease. However, the diagnosis of HCC is typically made late in the course of the disease. In patients with preserved liver function and a solitary tumor confined to the liver, treatment with surgical resection carries a 5-year survival rate from 40% to 75%. Transplantation may be an option for patients with hepatic dysfunction and tumors confined to the liver that meet certain size criteria. Nonsurgical options for patients who are ineligible for resection or transplantation include radiofrequency ablation, transarterial chemoembolization, and systemic therapy.

Hepatic Effects of Anesthetic Agents

Nonhepatic Surgery in Patients With Liver Disease

The preoperative evaluation of patients with liver disease should determine the severity of the disease and the presence of comorbid conditions associated with hepatic dysfunction. Given the mortality risk, elective surgery is contraindicated in patients with acute hepatitis or acute liver failure. In the absence of cirrhosis and significant hepatic dysfunction, patients with chronic liver disease generally have no increased risk for undergoing elective surgery. Patients with cirrhosis who undergo nonhepatic surgery have a higher incidence of postoperative morbidity and mortality when compared to patients without cirrhosis. A wide range of mortality rates has been reported in the literature (1%-50%), mostly in small, uncontrolled, institutional studies spanning 30 years. Mortality in these patients is a function of the severity of liver disease, its attendant comorbidities, and the type of surgical procedure. The largest study of in-hospital mortality in patients with cirrhosis used data from the Nationwide Inpatient Sample. The outcomes of 22,569 cirrhotic patients undergoing one of four index operations (cholecystectomy, colectomy, coronary artery bypass grafting [CABG], and abdominal aortic aneurysm repair) from 1998 to 2005 were compared to those of 2.8 million control patients. It was found that 4214 of the cirrhotic patients had portal hypertension. Patients with cirrhosis had a three- to eight-fold increase in the risk of in-hospital mortality, depending on the operation, compared to control patients. Mortality risk worsened in those with more severe liver disease, ranging from 12- to 23-fold, depending on the specific operation. The greatest risk was seen in patients undergoing CABG. A Taiwanese study comparing 24,282 cirrhotic patients undergoing major nonhepatic surgery with 97,128 matched control patients found an over two-fold increased risk of 30-day mortality in patients with cirrhosis (1.2% vs. 0.7%). Mortality was further increased in those cirrhotic patients with viral hepatitis, jaundice, ascites, gastrointestinal hemorrhage, and hepatic coma.

In cirrhotic patients undergoing nonhepatic surgery, the Child-Turcotte-Pugh (CTP) and MELD scores have been shown to identify those who are at higher risk of morbidity and mortality. The CTP score is calculated based on the values of five variables: degree of HE, degree of ascites, INR, serum albumin, and total bilirubin ( Table 16.5 ). The score is then used to classify the patient into one of three groups: Class A (score of 5-6), Class B (score of 7-9), and Class C (score > 10). A higher CTP class is associated with a greater risk of mortality. The mortality risk associated with each CTP class has decreased in more recent studies. In early retrospective series of cirrhotic patients undergoing abdominal surgery, the mortality risks associated with CTP Class A, B, and C were 10%, 30%, and 73% to 82%, respectively. In a more recent series of 194 patients undergoing 212 general surgical procedures, 30-day mortality was reported to be 6%, 13%, and 53% for CTP Classes A, B, and C, respectively. The risk has been found to be even lower in patients undergoing laparoscopic procedures (Class A: 2%; Class B: 12%; Class C: 12%), suggesting that improvements in perioperative care and surgical techniques have made surgery safer for patients with cirrhosis. A disadvantage of the CTP score is that the grading of encephalopathy and ascites is subjective. Furthermore, the score doesn’t take into account other predictors of operative risk such as the type of surgery or the etiology of the cirrhosis.

Table 16.5

Child-Turcotte-Pugh (CTP) Score

From Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg . 1973;60(8):646–649. https://doi-org.easyaccess2.lib.cuhk.edu.hk/ .

	Points assigned
	1	2	3
Ascites	Absent	Slight	Moderate
Bilirubin	<2 mg/dL	2-3 mg/dL	>3 mg/dL
Albumin	>3.5 g/dL	2.8-3.5 g/dL	<2.8 g/dL
prothrombin Time
Seconds over control	<4	4-6	>6
International normalized ratio	<1.7	1.7-2.3	>2.3
Encephalopathy	None	Grades 1-2	Grades 3-4
Point total	CTP class	Operative mortality
5-6	A	2-10%
7-9	B	12-31%
10-15	C	12-82%

 

The 30-day postoperative mortality predicted by the Child-Turcotte-Pugh (CTP) score is based on retrospective studies cited in text. Lower mortality rates have been reported in more recent studies.

As described above, the MELD score, calculated using the patient’s INR, serum creatinine, serum bilirubin, and etiology of liver disease ( Fig. 16.5 ) is a validated measure of mortality risk in patients with ESLD and is the basis of the liver allocation system for transplantation. A number of studies have validated the MELD score as a predictor of risk in cirrhotic patients undergoing abdominal, orthopedic, and cardiac surgery. In a large, retrospective study of cirrhotic patients undergoing major abdominal, orthopedic, or cardiac surgery, MELD score was a predictor for mortality at 30 days, 90 days, 1 year, and 5 years. The 30-day mortality ranged from 5.8% in patients with a MELD under 8 to over 50% in patients with a MELD score greater than 20. The incorporation of serum sodium into the MELD score (MELD-Na) as well as a combination of serum sodium and patient age (integrated MELD or iMELD) have been shown to improve the accuracy of predictions of survival over MELD. The use of MELD-Na and iMELD may allow more accurate prediction of mortality following nonhepatic surgery; however more studies are needed.

Estimated 3-month survival as a function of the Model for End-Stage Liver Disease (MELD) score.

Studies comparing the predictive power of CTP to MELD have yielded conflicting results likely due to small sample sizes and differences in primary outcome measures and surgical procedures. Small studies in cardiac and general surgical patients have found the abilities of CTP and MELD to predict postoperative mortality to be comparable. A study in cirrhotic patients undergoing extrahepatic abdominal surgery found MELD to be a better predictor than CTP for mortality or hepatic decompensation within 90 days of surgery. A more recent study found neither to be a good predictor for death and hepatic decompensation in cirrhotic patients undergoing elective surgery requiring general anesthesia. It has been suggested that both scores should be used in conjunction with other available patient data when attempting to risk-stratify cirrhotic patients for nonhepatic surgery.

Apart from the severity of cirrhosis and the specific surgical procedure, risk factors associated with increased postoperative morbidity and mortality in cirrhotic patients include emergency surgery, American Society of Anesthesiologists (ASA) physical status classification, the presence of renal insufficiency, male gender, and advanced age.

In addition to determining the severity of cirrhosis, the preoperative evaluation should determine the presence and severity of comorbidities associated with cirrhosis, including HE, pulmonary disease, cardiac disease, and renal dysfunction (see previous discussion). Preoperative laboratory evaluation should include a complete blood count to evaluate for anemia, thrombocytopenia, and leukocytosis. Additionally, INR, fibrinogen, serum electrolytes, creatinine, glucose, transaminases, bilirubin, and albumin should be measured.

Anesthetic Considerations for Procedures Involving the Liver

Transjugular Intrahepatic Portosystemic Shunt

TIPS placement is a catheter-based, endovascular procedure during which a shunt is created between the portal and hepatic venous systems through the parenchyma of the liver ( Fig. 16.6 ). The goal of the procedure is to reduce portal hypertension. The main indications for TIPS are the secondary prevention of variceal hemorrhage and the management of refractory ascites. During acute variceal hemorrhage in high-risk patients (HVPG > 20), TIPS placement following endoscopic intervention has been shown to be reduce rebleeding. Contraindications to the procedure include congestive heart failure, tricuspid regurgitation, and moderate to severe pulmonary hypertension. During the procedure, a catheter is typically inserted through the right internal jugular vein into a hepatic venous branch. A needle is then passed through the hepatic vein into the portal vein. The tract is dilated and a stent is placed across it. HVPG should be less than 12 mm Hg following the procedure. The incidence of major complications such as intra abdominal hemorrhage is on the order of 1% to 2%.

Transjugular intrahepatic portosystemic shunt procedure. A stent (or stents) is passed through the internal jugular vein over a wire into the hepatic vein (A); dilated esophageal varices (EV) are apparent. The wire and stent or stents are then advanced into the portal vein (B), after which blood can pass through the portal vein into the hepatic vein and bypass and decompress dilated esophageal veins (C). (Reproduced with permission from University of Michigan Health System: www.med.umich.edu/1libr/topics/liver09.htm/ (not able to access/TG 9–21–18)

TIPS may be carried out electively or emergently. The preoperative assessment should determine the extent of liver dysfunction and associated morbidity. Patients with advanced liver disease may be in a hyperdynamic state. They may have significant ascites with reductions in FRC as well as HH or HPS. Pulmonary function may be further compromised when the patient is in a supine position for the procedure. Patients may also have renal dysfunction, anemia, coagulopathy, and thrombocytopenia. Laboratory studies should screen for anemia, thrombocytopenia, and coagulopathy, as well as hyponatremia and elevations of creatinine and potassium. Patients should be transfused for a hemoglobin concentration less than 7 to 9 mg/dL, INR greater than 2, and a platelet count less than 50,000/μL. The procedure can be conducted under local anesthesia, sedation, or general anesthesia. Consideration should be given to the acuity of the patient, their ability to tolerate supine positioning, and the anticipated length of the case. In patients with significant ascites or recent variceal hemorrhage, general anesthesia with rapid sequence induction for airway protection is preferred. Intraoperative pain may be experienced during the establishment of the intrahepatic shunt and the dilation of the stent. In a series of 150 patients undergoing TIPS with general anesthesia using TIVA, the authors reported a postoperative ICU admission rate of 6.6% primarily for intraoperative hemodynamic instability. They advocated for the use of general anesthesia for all TIPS procedures, given the concerns for airway protection, ability to maintain supine position, and pain associated with the procedure. Postprocedure complications include heart failure due to increased venous return, HE, contrast-induced nephropathy, hemolytic anemia, and sepsis.

Hepatic Resection

The most common indications for hepatic resection are for the treatment of secondary metastases (i.e., metastatic colon cancer), primary hepatic malignancies, biliary tract malignancies, and benign hepatic masses. Hepatic resection is carried out along the functional segments of the liver. The liver is divided into eight functional segments based on the distribution of blood supply and biliary drainage. The left liver is comprised of segments II, III, and IV, while the right liver is comprised of segments V, VI, VII, and VIII. Segment I is the caudate lobe (see Fig.16.1 ). Right hepatectomy involves resection of segments V-VIII. Left hepatectomy involves resection of segments II-IV. Right lobectomy, also described as an extended right hepatectomy or right trisegmentectomy, involves resection of segments IV-VIII with or without segment I. An extended left hepatectomy (left trisegmentectomy) is the resection of segments II-V and VIII. Mortality and morbidity rates are impacted by the extent of the resection. A review of the 4881 hepatic resections over a 5-year period in the American College of Surgeons – National Surgical Quality Improvement Program database reported 30-day mortality and morbidity rates of 1.9% and 13.1%, respectively, for patients undergoing partial hepatectomy. Those mortality and morbidity rates increased significantly for patients undergoing extended hepatectomies (lobectomies or trisegmentectomies) to 5.8% and 22.5%, respectively. Other factors which have been shown to impact outcomes are the case volume of the performing center, patient age, history of cardiac, pulmonary or renal disease, blood loss, the presence of ascites, and reduced hepatic function.

Laparoscopic hepatic resection is considered most appropriate for small (<5 cm), solitary lesions located in peripheral segments (II-VI), when performed in experienced institutions. More extensive resections have been reported in experienced centers. A metaanalysis of 83 comparative case series (2900 patients) found a significantly lower rate of complications, transfusions, blood loss, and hospital stay in case-matched cohort of patients undergoing laparoscopic liver resection compared to open liver resection. To date, the only randomized, controlled trial comparing laparoscopic to open liver resection found a significant reduction in postoperative complications in patients undergoing laparoscopic hepatic resection (19% vs. 31%; 95% confidence interval [CI], 1.67-21.8; P = .021) for colorectal cancer liver metastases requiring resection of fewer than three consecutive segments. Although the hospital length of stay was significantly shorter, there were no differences in blood loss or mortality. Conversion from a laparoscopic to an open procedure is usually due to intraoperative hemorrhage and is reported to occur in 6% to 14% of cases. Risk factors for conversion include elevated BMI, increased age, diabetes mellitus, hypertension, and large tumor size. A number of recent observational, retrospective studies have described the performance of robot-assisted hepatectomy for both minor and major resections. A metaanalysis of seven studies published between 2010 and 2014, comparing laparoscopic to robotic hepatectomy, found that the laparoscopic approach is associated with significant reductions in blood loss and operative time without any difference in the conversion rate, postoperative morbidity, and the hospital length of stay.

The preoperative evaluation of the patient should focus on determining the severity of liver disease and the presence of other comorbidities. Both cirrhosis and steatosis have been associated with increased mortality in patients undergoing liver resection. The decision to proceed is made based on the anticipated future liver remnant (FLR) that will remain following resection. In patients with an anticipated FLR size that is considered unsafe, selective portal vein embolization (PVE) may be conducted in the weeks prior to the hepatic resection. Disruption of portal flow to the diseased portion of the liver leads to atrophy of the portion to be resected and hypertrophy of the FLR, allowing the patient to have adequate liver mass following resection. PVE is typically indicated for patients with an anticipated FLR less than 20% in the absence of cirrhosis, and less than 40% in the presence of cirrhosis. Further preoperative evaluation and risk assessment should take place as outlined above. Laboratory investigations prior to hepatic resection should include a complete blood count, serum electrolytes, liver chemistries, albumin, coagulation studies, and a type and screen.

Low Central Venous Pressure

While the Pringle maneuver occludes inflow to the liver, retrograde flow through the hepatic veins can still lead to substantial venous hemorrhage. The pressure in the hepatic veins is directly related to vena caval pressure, hence the strategy of maintaining a low CVP (<5 mm Hg) to minimize intraoperative blood loss. Retrospective observational studies have reported reduced blood loss in patients managed with CVP less than 5 mm Hg without any significant adverse events. Two randomized controlled trials comparing patients with CVP under 5 mm Hg with liberal volume administration have shown reductions in both intraoperative blood loss and the need for transfusion. These studies have all been at risk for significant bias. Furthermore, in some of the trials the difference in estimated blood loss, though statistically significant, was clinically inconsequential. Although standard practice in hepatic resections is to maintain low CVP, some studies have found no association between CVP and intraoperative blood loss. In retrospective studies of patients undergoing donor hepatectomies, no correlation was found between the CVP and intraoperative blood loss. This finding may be due to the fact that living liver donors are healthy patients with normal livers. The largest of these studies consisted of over 900 patients. Although it found no correlation between CVP and estimated blood loss, nearly all of the patients were managed with CVPs less than 10 mm Hg, and the majority had CVPs in the 4 to 6 mm Hg range.

A number of techniques have been described for reducing CVP during hepatic resection. The most commonly reported approach is the use of intraoperative fluid restriction with rates of 1 mL/kg/h. Vasopressors are used as needed to maintain SBP over 90 mm Hg and urine output more than 25 mL/h. Though seldom required, vasodilation using nitroglycerin or morphine, or forced diuresis with furosemide, can be instituted if fluid restriction is insufficient. The use of milrinone, reverse Trendelenburg position, and epidural anesthesia have also been described, though there is insufficient data to recommend one particular approach.

Clamping of the infrahepatic IVC (IIVC) has been described as a means of minimizing hemorrhage in the setting of an elevated CVP despite the interventions described earlier (see Fig. 16.7E ). A number of randomized controlled trials have demonstrated a significant reduction in blood loss and transfusion with the use of IIVC clamping in patients who tolerate the clamp or have a CVP greater than 5 mm Hg. A metaanalysis of these trials, comprising of 714 patients, showed a significant reduction in intraoperative blood loss (mean difference −353 mL) with no difference in the rates of postoperative complications. Of note, in pooled analysis, there was no significant difference in reported CVP values between the two groups.

Other Approaches to Reduce Blood Loss

Two multicenter, randomized, placebo-controlled trials have examined the efficacy of prophylactic recombinant Factor VIIa (rFVIIa) administration to reduce transfusion patients undergoing hepatic resection. The studies found no benefit of prophylactic rFVIIa with respect to mortality, red blood cell transfusions, or adverse events. A single randomized, controlled trial in 214 patients undergoing primarily minor hepatic resections (82% of cases) found that the administration of tranexamic acid (TXA) resulted in fewer transfusions and less blood loss compared to placebo. In the study group, 500 mg of TXA was administered intravenously prior to incision followed by 250 mg every 6 hours for 3 days.

A recent Cochrane database analysis of 67 randomized clinical trials of methods to decrease blood loss during liver resection found a high risk of bias in all trials. Based on the outcomes that were available in more than one trial, there is low-quality evidence that the “clamp and crush” parenchymal resection technique may be associated with fewer adverse events than radiofrequency resection techniques. They reported very low quality evidence that blood loss, operating time, and total hospital stay were reduced with intraoperative low CVP management.

Anesthetic Management

The anesthetic management of these cases should be tailored to the comorbidities of the patients, the surgical approach (open, laparoscopic), the anticipated extent of resection, and the need for vascular occlusive maneuvers. Adequate venous access should be obtained for the anticipated degree of hemorrhage. In cases where significant blood loss is anticipated, red blood cell salvage can be used and consideration should be given to the use of acute normovolemic hemodilution. In addition to standard ASA monitors, intraarterial blood pressure monitoring should be used in cases during which significant hemorrhage or the need for vascular occlusion maneuvers is anticipated. Central venous catheters may be inserted for both vascular access and CVP monitoring, although some high-volume centers forgo the routine use of CVP monitoring. Peripheral venous pressure monitoring may be considered as an alternative. Values are obtained through transduction of an antecubital vein and have been shown to correlate with the CVP in patients undergoing hepatic resection. Stroke volume variation (SVV) has been investigated as an alternative to CVP for volume management in patients undergoing liver resection. A study of patients undergoing donor hepatectomy showed an SVV value ≤6% to be predictive of blood loss over 700 mL. In two studies, targeting SVV values greater than 12% to 15% in patients undergoing open or laparoscopic hepatic resection yielded blood loss comparable to, or better than, maintaining low CVP. In a study using milrinone to reduce CVP, SVV of 9% yielded a favorable surgical field. No correlation with CVP was noted in this study, possibly due to the lusitropic effects of milrinone. The use of transesophageal echocardiography may be considered in patients with cardiac risk factors, although care must be taken in the presence of significant esophageal varices.

Hepatic resection may be associated with hemorrhage and, in extensive cases, the development of coagulopathy. There is insufficient data to recommend the routine use of viscoelastic hemostatic assays (rotational thromboelastography (TEG) or rotational thromboelastometry). A single randomized, controlled trial of TEG-guided transfusion in cirrhotic patients with coagulopathy (INR > 1.8 and/or platelet count <50 × 10 ⁹ /L) undergoing invasive procedures found a significant reduction in blood product administration with TEG compared to the standard lab value–based group. However, only half of the patients enrolled in the trial were undergoing procedures associated with greater than 3% risk of hemorrhage.

With respect to pain management, thoracic epidural analgesia is a highly effective means of pain control for hepatic resection. Although its use for this purpose has been described in a number of case series, it remains controversial due to the development of postoperative coagulopathy and thrombocytopenia following hepatic resection and the associated risk of epidural hematoma. Postoperative coagulopathy and thrombocytopenia occur in 21% to 100% of patients undergoing liver resection, depending on the volume of tissue resected. The INR peaks around postoperative day (POD) one or two while the platelet count is lowest on POD three or four. The values typically return to baseline by POD four or five; however in some patients they may remain elevated for up to a week, leading to delays in epidural catheter removal or the decision to transfuse FFP to facilitate catheter removal. An additional concern is the accidental dislodgment of the epidural catheter while the patient is coagulopathic which has been reported to occur in up to 7% of patients. The degree of coagulopathy is influenced by the extent of the resection. In a review of 759 patients undergoing hepatic resection, the incidence of coagulopathy was 39.1% in patients who had undergone major hepatectomies (>2 segments) versus 21.3% in patients who had undergone minor hepatectomies (≤2 segments). Other factors that are independently associated with postoperative coagulopathy included preexisting cirrhosis, preoperative INR ≥ 1.3, preoperative platelet count <150,000/μL, estimated blood loss ≥ 1000 mL, and the duration of surgery. Although no case series has reported the complication of an epidural hematoma, epidural hematoma is quite rare and the numbers of published cases may have been inadequate to detect its occurrence.

Rapid sequence induction should be considered in patients with significant ascites. Maintenance of general anesthesia has been described with both volatile anesthetics and intravenous anesthetics. There is no compelling clinical evidence for the benefit of one specific agent over another. Regardless of the approach, anesthetic dosing should be titrated to the patient’s requirements, taking into account the pharmacokinetics and pharmacodynamics of common anesthetic agents in patients with liver disease. The anesthesiologist should understand the vascular occlusion strategies that may be employed during the operation along with their hemodynamic effects. The Pringle maneuver, both during open and laparoscopic hepatic resections, leads to an increase in MAP and systemic vascular resistance and a decrease in CO. During open hepatic resection, portal triad clamping leads to a decrease in left ventricular (LV) end-diastolic area (i.e., LV preload). During laparoscopic hepatic resection, in the presence of pneumoperitoneum, PTC also increases LV end-systolic wall stress (i.e., LV afterload).

Enhanced Recovery After Surgery and Liver Resection

A number of randomized controlled trials have been conducted comparing the use of enhanced recovery after surgery (ERAS) protocols with traditional management in patients undergoing both open and laparoscopic liver resections. Although ERAS Society consensus guidelines exist, these trials differed with respect to the number and nature of ERAS elements. Most incorporated preoperative education, early enteral nutrition, and early mobilization. The implementation of an ERAS protocol for liver resection surgery has been shown to reduce minor adverse events, hospital length of stay, hospital cost, and to improve patient quality of life. A metaanalysis published by the Cochrane group found that the use of ERAS protocols resulted in a reduction in mild adverse events, hospital length of stay, and cost. However, all studies were at high risk of bias and the quality of available evidence was determined to be low.