Obstructive Jaundice

Patients with significant jaundice (see Chapters 2 and 7) often have anorexia and lose weight because of decreased oral intake. Approximately 45% to 70% of patients with obstructive jaundice present with malnutrition as evidenced by greater than 10% weight loss, albumin less than 3 g/dL, decreased triceps skin fold, and impaired delayed hypersensitivity reactivity (Foschi et al, 1986). The NRI is simple to use and can define a high-risk subgroup of patients with obstructive jaundice; an NRI less than 83.5 has been found to be significantly associated with an increased mortality risk and longer duration of hospital admission but not an increased complication rate in this population subgroup (Clugston et al, 2006). The primary nutritional deficit resulting from obstructive jaundice is malabsorption of fat and fat-soluble vitamins. In addition, there is loss of trace minerals, such as phosphate, calcium, magnesium, and zinc, owing to salt formation from unabsorbed dietary fat (Shronts, 1988).

Patients with obstructive jaundice may have ascites secondary to decreased serum albumin levels, but the metabolism of carbohydrates and proteins is rarely altered (Flannigan et al, 1985). Biliary sepsis in a patient with obstructive jaundice contributes to malnutrition by shifting protein synthesis from anabolic protein synthesis to acute-phase protein synthesis (O’Neill et al, 1997). This reprioritization of protein synthesis occurs as a result of endotoxin-stimulated Kupffer cell production of TNF, IL-6, eicosanoids, nitric oxide, and other inflammatory mediators that directly inhibit protein synthesis (Curran et al, 1990; Heinrich, 1990; see Chapters 9 and 10). Because of these derangements, some authors have advocated preoperative biliary drainage (PBD) in both liver and pancreas surgery. A Cochrane 2008 analysis demonstrated that PBD is not recommended in patients who need surgery for obstructive jaundice. However, another recent Cochrane review showed no evidence to support or refute routine endoscopic retrograde cholangiopancreaticography (ERCP) with stenting in patients with malignant pancreaticobiliary diseases awaiting surgery (Wang et al, 2008; Mumtaz et al, 2007). PBD in pancreatic adenocarcinoma should not be routine practice, as it is associated with a stent-related complication rate of 23% and has resulted in a twofold increase in postpancreatectomy infectious complications (Mezhir et al, 2009); however, trials have supported PBD in extended hepatectomy for hilar colangiocarcinoma, if the future liver remnant volume is anticipated to be less than or equal to 30%. To reverse the catabolic effects of chronic endotoxemia and restore hepatic protein synthesis, patients with biliary infection should be treated with biliary decompression for at least 4 weeks before major hepatobiliary surgery to allow hepatocytes to recover their protein synthetic capacity.

Cirrhosis and Liver Failure

Patients with cirrhosis (see Chapter 70A, Chapter 70B, Chapter 73, Chapter 74 ) provide the clinician with a major challenge, as they have multiple hormonal and metabolic alterations (see Chapter 2). Characteristics of cirrhotic patients include wasting symptoms, especially loss of fat and muscle mass; growth failure; glucose intolerance; hyperinsulinemia; insulin resistance; increased plasma glucagon and catecholamines; elevated serum free fatty acids; elevated glycerol; hypoproteinemia; hyperammonemia; hypophosphatemia; and alterations in plasma and cerebrospinal fluid amino acid profiles (Achord, 1987; Henriksen et al, 1985; Petrides & De Fronzo, 1989; Riggio et al, 1984). These hormonal and metabolic aberrations lead to altered metabolism of all three macronutrients: fat, protein, and carbohydrate. The hormonal and metabolic changes seen in cirrhosis also lead to an increased skeletal muscle proteolysis for energy provision, which leads to eventual muscle wasting. In addition, there is increased peripheral lipolysis with a decreased ability to use fat and carbohydrate, which leads to hyperglycemia and hyperlipidemia (Katz, 1986).

Compounding these metabolic alterations are issues that predispose patients with cirrhosis to malnutrition, including decreased dietary intake owing to nausea and vomiting, and the common practice of imposing protein restriction in an effort to prevent encephalopathy. This practice of protein restriction is questionable in an already malnourished patient. Protein restriction exacerbates the problems inherently associated with malnutrition and prohibits the goal of liver regeneration. An alteration of plasma and cerebrospinal fluid amino acid profiles caused by catabolism, impaired hepatocellular function, and portal shunting leads to decreased levels of branched-chain amino acids (BCAAs) valine, leucine, and isoleucine and preferred uptake into the brain of aromatic amino acids phenylalanine, tyrosine, and tryptophan. The increased uptake of aromatic amino acids is thought to alter the production of neurotransmitters, resulting in encephalopathy. This theory led to the use of BCAA as dietary treatment for patients with liver disease (Marchesini et al, 2005). Correcting the serum amino acid profile by BCAA administration aims to reverse mental status changes and promote anabolism (Fischer et al, 1976). Oral supplementation with a BCAA preparation that can be administered for a long period improves event-free survival, serum albumin concentration, and quality of life in patients with decompensated cirrhosis (Bianchi et al, 2005; Marchesini et al, 2003; Muto et al, 2005). Changes in Model for End-Stage Liver Disease (MELD) and Child-Turcotte-Pugh (CTP) scores were smaller in a BCAA-administered group than in a control group, and serum total bilirubin and serum albumin were better preserved. The incidence of major cirrhotic complications was also lower in the BCAA group than in the control group (Kawamura et al, 2009). ESPEN upgraded the recommendation of BCAA supplementation in decompensated liver cirrhosis in the latest revision of its guidelines in 2006. Recent work has demonstrated that BCAA may affect microinflammation in hepatitis C–positive patients with cirrhosis, reducing the production of oxidative stress and possibly leading to a decrease in the occurrence of hepatocellular carcinoma (HCC) (Ohno et al, 2008). More studies are needed to identify those who might benefit and what the benefit may be from BCAA supplementation.

Liver Resection

Metabolic alterations occur in the regenerating liver after liver resection (Diehl, 1991; see Chapters 5 and 64). Krebs cycle activity is depressed, as is the reduction–oxidation state of the hepatic mitochondria, with a switch from the use of glucose to fat as the preferred source of energy by way of β-oxidation (Nakatoni et al, 1981). Because hyperglycemia and hyperinsulinemia suppress the release of fatty acids from adipose tissue and decrease ketone body production by the liver (Riou et al, 1986), hypertonic glucose infusions and insulin administration should be avoided in the immediate (<6 hours) postoperative period (Ozawa, 1992).

These observations indicate that selective administration of fat or ketone bodies shortly after liver resection or transplantation may be beneficial. In rodents, the provision of intravenous fat (30% of total nonprotein calories) (Hamada, 1993; Nishiguchi et al, 1991) or the ketone body monoacetoacetate (Birkhahn et al, 1989) immediately after liver resection accelerates liver regeneration. The administration of medium-chain triglycerides after liver resection results in better hepatic energy charge and decreased lipid peroxidation compared with the effects of intravenous glucose or long-chain triglyceride infusions (Hamada, 1993).

Sarac and colleagues (1994) hypothesized that increasing fat oxidation preoperatively via fasting would improve liver function after extensive liver resection in rats. Rats subjected to 90% hepatectomy had improved survival to almost 100% when fasted 24 hours before operation and fed oral glucose immediately after operation. Greater use of ketone bodies was observed in the liver of fasted rats, suggesting that the enzyme machinery for using free fatty acids was induced by the previous fast. This and many animal models have demonstrated that glycogen is essential to maintaining hepatocellular integrity and function by supplying glucose for ATP generation. A recent trial that administered high concentrations of glucose to patients intravenously 24 hours before hepatic lesion resection with portal clamping found significantly improved liver function on the first and fifth days postoperatively by reducing liver I/R injury (Tang et al, 2007).

The regenerating liver has an increased demand for specific amino acids, and provision of these in the diet accelerates regeneration. Immediately after liver resection, an abrupt increase is seen in the synthesis of the system A amino acid transporter—which transports alanine, serine, and methionine—but not those of system N (glutamine, histidine, asparagine) or system ASC (e.g., cysteine) (Fowler et al, 1992). Increased system A activity depends on portal glucagon and insulin secretion and substrate amino acid supply (Dolais-Kitabgi et al, 1981); system N and system ASC are not similarly regulated (Fowler et al, 1992). This fact supports the argument by many liver surgeons that enteral administration of glucose is the preferred route of feeding because of the effect on insulin release, which is vital to the function of the regenerating liver (Ozawa, 1992; Ozawa et al, 1974).

Provision of adequate protein and calories is important to ensure adequate liver regeneration. Rats subjected to 50% of normal daily caloric intake for 1 week before and after liver resection had significantly impaired liver regeneration, even when administered exogenous IGF-I, compared with normally fed rats (Sato et al, 1994). Specific types of protein-supplemented diets, such as a nucleoside-nucleotide supplemented total parenteral nutrition (TPN) solution, may improve protein synthesis after hepatectomy (Ogoshi et al, 1989).

John and colleagues (1992) reported that postoperative TPN consisting of 45% fat calories significantly impaired hepatic regeneration and albumin synthesis, caused cholestasis, and increased mortality compared with the same diet administered by the enteral route after 70% hepatectomy in rats. Mortality was 68% in rats fed TPN, 9% in rats fed enterally, and 8% in rats fed chow. This increased mortality rate in TPN-fed rats could be due to increased bacterial translocation and lipopolysaccharide migration across the gut, which overwhelms the limited phagocyte capacity of the remnant Kupffer cell mass (van Leeuwen et al, 1991). The increased mortality rate may also be due to excessive administration of intravenous glucose calories, which adversely affects liver substrate metabolism (Ozawa, 1992; Ozawa et al, 1976). Investigators subsequently showed that the lethal consequences of parenteral nutrition after liver resection in rats could be ameliorated by decreasing the caloric and amino acid load (Delany et al, 1994). A more recent study has demonstrated that increased parenteral caloric intake, and not hyperglycemia, is an independent risk factor for bacteremia in patients receiving TPN (Dissanaike et al, 2007).

Transplantation

In the first 6 hours after liver transplantation (see Chapter 100, Chapter 97A, Chapter 97B, Chapter 97C, Chapter 97D, Chapter 97E ), glucose use by the transplanted liver is impaired until the redox state of the mitochondria improves (Ozaki et al, 1991). During this time, the liver preferentially uses fatty acid oxidation for ATP generation (Ozaki et al, 1991; Takada et al, 1993). After 6 hours, normally functioning liver allografts shift substrate use from fat to glucose, whereas failing livers continue to use fat. This shift in metabolism can be followed by measuring the plasma concentration of total ketone bodies and arterial ketone body ratio (Takada et al, 1993). Based on these observations, Takada and colleagues suggested that glucose should be administered in small quantities in the immediate postoperative period without insulin to avoid suppressing peripheral fat mobilization. Glucose infusion should be increased steadily, as mitochondria respiration recovers and the Krebs cycle becomes active. When the redox state of mitochondria is sufficiently improved (e.g., arterial ketone body ratio approximately 0.7), amino acids and increased amounts of glucose can be administered. It has been demonstrated that intraportal insulin administration augments liver regeneration during the first postoperative transplant week by improving hepatic function in live-donor liver transplant recipients (Xu et al, 2009).

The nutritional status of the donor liver may be an important factor affecting posttransplant allograft function. Initially, it was believed that liver allografts obtained from fasted subjects were more susceptible to anoxic liver injury than allografts obtained from subjects being administered glucose up to the time of liver donation because of an absence of glycogen as a source of glucose from glycolysis in the fasted group (Bradford et al, 1986). In addition, it was shown that livers from fasted animals had enhanced generation of oxygen radicals by Kupffer cells (Gasbarrini et al, 1993). Subsequent work in animal liver transplant models (Astarcioglu et al, 1991; Palombo et al, 1988) and human liver donors (Cywes et al, 1992) showed that glucose administration to the donor increases liver glycogen and ATP content before cold preservation and attenuates liver injury after transplantation.

Work by Sumimoto and colleagues (1993) shows, however, that the effects of fasting and feeding on liver allograft function after cold preservation and reperfusion are more complex. They reported no relationship between hepatic glycogen content and posttransplant function after cold preservation. The same investigators also showed that livers from fasted rats undergo two changes, one that sensitizes it to preservation and reperfusion injury (short-term fast [1-3 days]) and one that increases its tolerance to preservation injury (long-term fast [4 days]). Long-term fasting improves subsequent liver allograft function in rats (Sumimoto et al, 1993). The mechanisms for this protection are unclear, but it may be related to the accumulation of metabolites within the liver that are readily used as a source of energy or to the decreased apoptosis of sinusoidal endothelial cells. A study comparing rats that were orthotopically transplanted with donor rat livers, either fasted or fed, found that the number of apoptotic sinusoidal edothelial cells was significantly higher in the fed group than in the fasted group, suggesting that donor fasting decreases sinusoidal endothelial cell apoptosis after reperfusion, leading to the protection of the liver graft from reperfusion injury (Sun et al, 2001). The latter change may explain why livers from rats subjected to prolonged periods of fasting generate fewer oxygen radicals after reperfusion.

Hepatic Cancer

Hepatic cancer (see Chapter 80, Chapter 81A, Chapter 81B, Chapter 81C ) is commonly accompanied by progressive weight loss, anorexia, and malnutrition. Patients with hepatocellular carcinoma (HCC) usually have advanced cirrhosis, and for that reason, their malnutrition is often more pronounced (see Chapter 2, Chapter 70A, Chapter 70B, Chapter 72, Chapter 73 ). Inefficient energy use resulting from increased Cori cycle activity (Holroyde & Reichard, 1981) and other common occurrences with liver damage may be seen, such as increased protein turnover rates with altered amino acid profiles, increased gluconeogenesis, and increased lipid oxidation (Holroyde & Reichard, 1981; Waterhouse et al, 1979; Young, 1977). If jaundice is present, fat malabsorption often occurs in these patients. Profoundly malnourished patients with cirrhosis or chronic active hepatitis and HCC usually are seen initially in an advanced clinical stage, such as Okuda class III. These patients generally should not undergo operation because of a prohibitive operative mortality (Okuda et al, 1985).

Pancreatic Cancer

Pancreatic cancer (see Chapter 58A, Chapter 58B ) is associated with a severe metabolic derangement referred to as the cancer anorexia-cachexia syndrome (Ryan et al, 1998; Inui, 2002). This syndrome is associated with anorexia tissue wasting, malnutrition, weight loss, and a loss of compensatory increase in feeding. The pathogenesis is dependent on disorders of carbohydrate, protein, lipid, and energy metabolism mediated by cytokine elaboration (IL-1, IL-6, TNF, interferon [IFN] γ, leukocyte-inhibiting factor [LIF], TGF-β) and an overall increase in leptin levels (Bruera et al, 2000; Walker, 2001). Pancreatic cancer has the highest incidence of cachexia among patients with cancer, and up to 80% of such patients have cachexia at the time of their initial diagnosis. For this reason, patients with pancreatic cancer are perhaps the most likely of all patients undergoing hepatopancreatobiliary surgery to benefit from nutritional support strategies.

Obstructive Jaundice (See Chapter 7)

If patients are profoundly malnourished, and preoperative nutritional repletion is indicated, an argument can be made for reestablishing bile flow by placing a biliary stent for decompression of the biliary tree before proceeding with a major operation. Before undertaking this step, however, patients should be assessed for their ability to tolerate fat, and dietary recommendations should be made based on this assessment. Generally, restoration of bile flow improves absorption of fat and fat-soluble vitamins. Foschi and colleagues (1986) reported a prospective randomized trial in which patients with obstructive jaundice were randomized to biliary decompression followed by operation or biliary decompression combined with 2 weeks of alimentation followed by operation. Malnutrition was present in 70% of the patients. Of the patients receiving alimentation, 85% received at least partial enteral feeding. Patients receiving alimentation for 2 weeks had lower morbidity (17% vs. 46%), mortality (3.5% vs. 12.5%), and postoperative infection (14% vs. 28%) rates compared with patients receiving only biliary decompression.

When bile flow is not reestablished in patients with biliary obstruction who require nutritional support, ability to tolerate fat in the diet is impaired. These patients should ingest a low-fat diet supplemented with water-soluble forms (e.g., Aquasol E, vitamins A, D, E, and K) of the fat-soluble vitamins. They also should receive supplements of calcium, phosphorus, and magnesium (Shronts, 1988). Rather than struggle with the problems of inadequate oral intake in these anorectic patients, they are more easily and effectively repleted nutritionally by tube feedings because of the ability to administer more physiologically appropriate diets, containing predigested protein (dipeptides and tripeptides) and a blend of medium-chain triglycerides and long-chain triglycerides. These specialized diets help avoid diarrhea from fat maldigestion and malabsorption, because the absorption of medium-chain triglycerides depends on neither bile salts nor pancreatic lipase, as does absorption of long-chain triglycerides (McCullough et al, 1989; Snopko et al, 1991).

Patients with obstructive jaundice require approximately 25 to 35 kcal/kg dry weight to maintain nitrogen balance, and it is recommended that 25% to 40% of total calories be provided by fat as tolerated, but this is unlikely without biliary decompression (McCullough et al, 1989). Patients with externally draining biliary catheters represent a particularly difficult problem, because they not only have problems related to biliary obstruction but also suffer under the added insult of continuous fluid, electrolyte, and protein loss. This situation has become increasingly less common as interventional radiologists and gastroenterologists have gained greater proficiency with biliary drainage procedures.

Cirrhosis and Hepatic Failure (See Chapter 70A, Chapter 70B, Chapter 72, Chapter 73 )

The following discussion summarizes the essentials of nutritional management of cirrhotic patients. For more detailed discussions, the interested reader is referred to other authors on this topic (Cabre & Gassull, 2001; McCullough & Tavill, 1991; Nompleggi & Bonkovsky, 1994; Shronts, 1988). Hospitalized patients with cirrhosis requiring operation have substantial morbidity and mortality (Rice et al, 1997; Alberino et al, 2001). Because malnutrition is a major problem in these patients, aggressive nutritional support should be undertaken when surgery is being contemplated or carried out. Prospective, randomized clinical trials have shown that hospitalized patients with cirrhosis do not meet their nutritional needs with aggressive daily dietary counseling and have reduced hospital mortality and morbidity when fed via nasogastric or duodenal tubes (Cabre et al, 1990). It is recommended that all hospitalized cirrhotic patients undergoing emergency operation be fed via the enteral route unless otherwise contraindicated; use of parenteral nutrition in these patients is not indicated if the GI tract is functional.

Because the metabolic derangement of advanced liver disease affects all three macronutrients, any dietary intervention needs to supplement all three. The caloric needs of patients with liver disease are increased (35 to 40 kcal/kg body weight/day), particularly if they have ascites (McCullough & Tavill, 1991; ESPEN guidelines, 2006). Protein needs depend on the patient’s current nutritional status, ability to tolerate protein, and history or presence of encephalopathy. In a nonencephalopathic patient, 1 to 1.5 g protein/kg dry weight is recommended. If encephalopathy is present, a thorough diet history should be taken to determine the tolerable amount of protein with recommendations to supply at least a minimum 0.5 to 0.7 g/kg dry weight per day.

Fluid and sodium restriction is often necessary, depending on the severity of ascites and edema. A level of at least 2 g of sodium per day is recommended, because lower levels are unpalatable and usually decrease dietary intake further, exacerbating the patient’s malnutrition. In cases of intractable ascites, some nutritional support experts (Shronts & Fish, 1993) have recommended a 500 to 1000 mg sodium diet and 1 to 1.5 L of fluid. The authors believe, however, that treatment of massive ascites should rely more on aggressive diuretics rather than drastic limits on fluid intake. Patients with cirrhosis and those with hepatic failure commonly have micronutrient deficiencies of phosphate, magnesium, and zinc; these should be measured, monitored, and aggressively repleted as necessary by the intravenous route (Shronts, 1988). As stated earlier, the use of BCAA-enriched formula should be used in patients with encephalopathy and decompensated cirrhosis, leading to improved long-term event-free survival and decreased major complications.

Preliminary data suggest that use of recombinant growth hormone may improve nutritional recovery and liver regeneration in cirrhotic patients (Luo et al, 2004). A smaller study demonstrated increased albumin levels and energy metabolism in patients with cirrhosis after a 4-month administration of IGF-I (Conchillo et al, 2005). More recently, patients treated with parenteral nutrition and recombinant growth hormone versus parenteral nutrition alone were found to have increased levels of serum prealbumin, growth hormone, IGF-1, IGF binding protein (IGF-BP)-3, hepatic IGF-1 mRNA, IGF-BP-3 mRNA, and liver Ki67 labeling index (Cao et al, 2007). Although these studies resulted in no actual clinical improvements, this data would encourage further testing of growth hormone in this subgroup of patients.

Liver Resection

Most patients in Western countries undergoing liver resection (see Chapter 90A, Chapter 90B, Chapter 90C, Chapter 90D, Chapter 90E, Chapter 90F, Chapter 91A, Chapter 91B, Chapter 92 ) have no associated cirrhosis and no need for specialized nutritional support preoperatively or postoperatively. Most can begin an oral diet safely on the first or second postoperative day, and most are able to tolerate a full regular diet within 5 days. Malnourished patients being considered for elective liver resection should receive nutritional repletion preoperatively and postoperatively by the oral or enteral route. In addition, patients undergoing extensive liver resection, particularly patients with compromised preoperative hepatocellular function (e.g., steatosis, chronic viral hepatitis, or cirrhosis), may benefit from specialized nutritional support.

Based on the physiology discussed previously, it is strongly recommended that patients not receive hypertonic glucose infusions in the immediate (<6 hours) postoperative period, because this can interfere with the liver’s preferential use of fatty acids for energy and hepatic regeneration. If nutritional support is initiated within 24 or 48 hours after liver resection, it should be administered via the enteral route through a nasoduodenal or jejunal tube. If the patient is profoundly malnourished preoperatively, and the surgeon intends to administer postoperative nutrition, placement of a feeding jejunostomy tube at the time of surgery should be considered.

Nutritional therapy has the potential to attenuate damage to the residual liver in patients scheduled to undergo elective liver resection by preventing further reductions in endogenous antioxidants and by repleting deficient antioxidant defenses. Patients at risk for depleted hepatic antioxidant defenses are alcoholics, smokers, and patients with a long history of obstructive jaundice, chronic biliary infection, or chronic active viral hepatitis (Bulger & Helton, 1998; Helton, 1994). It is imperative that all patients stop smoking and cease ethanol ingestion approximately 2 weeks preoperatively to replete their intrinsic antioxidant defenses and avoid further oxidation stress and lipid peroxidation after operation. It can take 2 weeks of oral vitamin E supplementation to increase vitamin E content significantly in some tissues (Yau et al, 1994). Patients at risk for vitamin E deficits should receive supplements for 2 weeks preoperatively when possible.

A prospective randomized trial evaluated the effect of an antioxidant vitamin infusion (Omnibionta) that included 10 mg of α-tocopherol acetate, 2 mg of dl-α-tocopherol, and 1 g of ascorbate in patients undergoing liver surgery (Cerwenka et al, 1998). An antioxidant protective effect was shown by significantly less postischemic increases in transaminases and prothrombin time (PT) compared with patients who did not receive the vitamin infusion. A 2005 placebo-controlled double-blind study of 47 patients who underwent partial liver resection again demonstrated that the group receiving infusions containing vitamin E had a significantly shorter length of stay in the intensive care unit (ICU) and showed improvement in aspartate aminotransferase, alanine aminotransferase, and glutamic dehydrogenase (Bartels et al, 2004). When considering the fact that vitamins are cheap, safe, and readily available and that patients undergoing hepatobiliary operations are at risk for oxygen radical–mediated injury, it is reasonable to consider administration of vitamin E and vitamin C for several days preoperatively in patients at risk for postoperative complications.

To explore the hypothesis that the remnant liver after liver resection has limited capacity to metabolize glucose, Nishizaki and colleagues (1996) randomized patients after liver resection to receive peripheral glucose infusion (10 kcal/kg body weight/day) or hypertonic glucose via a central catheter for 7 days; both groups were allowed ad libitum food intake when tolerated. The average caloric intake of the two groups was 20 kcal/kg body weight/day (peripheral dextrose) and 30 kcal/kg body weight/day (hypertonic glucose) with no untoward effects from the hypertonic glucose infusion (Nishizaki et al, 1996); patients in that group had improved prealbumin, decreased urinary 3-methylhistidine excretion, and improved nitrogen balance compared with those who received peripheral glucose. The results of this study showed that the remnant liver is not harmed by increased loads of glucose.

A 2006 systematic review of nutritional support after open liver resection concluded that early enteral nutrition is safe, resulting in a significantly lower rate of wound infection and catheter-related complications when compared to parenteral nutrition; however, no differences in mortality were found with either enteral or parenteral nutrition, but patients who received enteral nutrition did show an improved postoperative immune competence (Richter et al, 2006).

Liver Transplantation (See Chapter 100, Chapter 97A, Chapter 97B, Chapter 97C, Chapter 97D, Chapter 97E )

Clinical and experimental observations show that nutrition plays an extremely important role in just about all aspects of liver transplantation. The status of the donor, the function of the transplanted liver, the overall health of the recipient, the rejection process, the absorption and metabolism of immunosuppressive drugs, and a variety of metabolic pathways are influenced by nutrients and how they are administered. Clinical observations show that malnourished patients undergoing liver transplantation have a significantly poorer outcome than well-nourished patients (Porayko et al, 1991; Shaw et al, 1985). Pretransplant nutritional status, hemoglobin levels, and disease severity were independently associated with the number of infectious episodes during hospitalization (Meril et al, 2009). Degree of malnutrition correlates significantly with days spent in the ICU, days spent mechanically ventilated, and length of hospital stay, and it also correlates significantly with mortality after transplantation (Pikul et al, 1994; Meril et al, 2009). Patients who were seen initially at less than 90% of ideal body weight had a longer hospital stay, needed longer ventilatory support, and had a higher rate of infectious complications; and conversely, patients who were obese also required a longer period of support (Hade et al, 2003).

Liver transplant recipients are predisposed to postoperative infectious complications not only because of severe preoperative protein calorie malnutrition but also because of immunosuppressive drug therapy (O’Keefe et al, 1980, 1981). Profoundly malnourished recipients have an incidence of infectious morbidity and 3-month mortality rates twice that of well-nourished patients undergoing liver transplant (Harrison et al, 1997).

Several studies have investigated the effects of nutritional support in liver transplant patients (Charlton et al, 1992; Chin et al, 1990; Harrison et al, 1997; Hasse et al, 1995; Martin et al, 1993; Mehta et al, 1995; Reilly et al, 1990; Wicks et al, 1994). Based on these studies, patients awaiting transplantation who are malnourished should receive aggressive nutritional counseling and support (Cabre & Gassull, 2001; Harrison et al, 1993, 1997; Plauth et al, 1997). Nutritional supplements, including vitamins, should be taken in an effort to attenuate further nutritional deficits and to replenish losses. Patients who ingest inadequate amounts of calories should be fed via a soft nasogastric feeding tube. In contrast to what most people and physicians believe, these tubes are well tolerated and can be used for months without much difficulty or patient discomfort.

Reilly and colleagues (1990) prospectively studied the effects of three intravenous diets in hypoalbuminemic patients after liver transplantation. The three diets were a BCAA-based TPN solution, a standard TPN diet, and intravenous glucose (Reilly et al, 1990). Both TPN diets provided 1.5 g protein/kg body weight/day and 35 kcal/kg body weight/day. Glucose was limited to 5 mg/kg body weight/min, and the balance of energy was provided as Intralipid 10%. Patients in both TPN groups were extubated and discharged from the ICU a mean of 2.3 days earlier compared with patients receiving only glucose. In addition, overall hospital charges were estimated to be $21,000 less in the patients who received either standard TPN or BCAA-based TPN. The investigators concluded that malnourished patients undergoing transplantation can and should receive postoperative TPN.

Numerous transplant centers subsequently showed that liver transplant patients can be fed successfully and safely by a nasoduodenal or nasojejunal tube within hours after transplantation (Hasse et al, 1995; Martin et al, 1993; Mehta et al, 1995; Sekido et al, 2003; Wicks et al, 1994). Enteral nutrition as early as 12 hours after surgery seemed to improve nutritional parameters and measures of immune function such as total lymphocyte count (Bengmark, 2002; Sekido et al, 2003). In three studies, early initiation of feeding after operation (<24 hours) was associated with a postoperative infection rate that was lower than expected (Hasse et al, 1995; Martin et al, 1993; Mehta et al, 1995; Wicks et al, 1994). In the prospective, randomized study reported by Hasse and colleagues (1995), the enterally fed patients had no viral infections in the first 2 postoperative weeks compared with 17% of patients fed intravenously (P < .05). Bacterial and overall infection rates also were decreased in the enterally fed group (14% vs. 29% and 21% vs. 41%, respectively), although the differences did not reach statistical significance. Mehta and colleagues (1995) reported that postoperative liver transplant patients fed immediately via the enteral route had decreased incidence of postoperative ileus (5% vs. 28%; P < .01) and started on oral feedings much sooner (8 days vs. 12 days; P = .07) than patients fed intravenously. Based on these studies, postoperative nutritional support should be administered via a nasoduodenal or jejunal tube if indicated for nutritional repletion or maintenance.

Based on the hypothesized benefits of increasing glycogen levels of the liver before cold preservation, the University of Toronto Liver Transplant Group carried out a prospective, randomized trial investigating treatment of the donor liver before procurement with a short-term intraportal glucose and insulin infusion (Cywes et al, 1992). This study showed that increasing liver glycogen content decreased preservation and reperfusion injury compared with untreated livers. The protective effect of glycogenation was not clinically apparent and manifested only as differences in serum transaminase levels. The protective effects were predominantly evident in patients with prolonged warm ischemia times, suggesting that the extra glycogen provided substrate for anaerobic glycolysis as the liver rewarmed before reperfusion. Another area under investigation is the use of both prebiotics and probiotics to enhance the GI immune response. A prospective, randomized, double-blind trial treated all postoperative liver transplant recipients with enteral nutrition and a combination of lactic acid bacteria and fibers, thereby reducing bacterial infection rates from 48% to 3% following liver transplantation (Rayes et al, 2005).

Hepatic Cancer

Many liver cancer patients experience major metabolic alterations coupled with some degree of wasting. Each individual patient should be evaluated for current nutritional status, and an individual nutritional plan should be devised to maintain or replete the patient’s nutritional status. This includes increased calorie and protein provision when necessary (35 kcal/kg dry weight/day, 2 g protein/kg dry weight/day, and micronutrient supplementation as needed; Fan et al, 1994). Dietary restrictions of any kind should be avoided if at all possible in these patients, including sodium and fluid restrictions.

Fan and colleagues (1994) showed the utility of preoperative and postoperative TPN in patients undergoing elective liver resection for HCC. In this study, the authors randomized 120 patients to one of two groups: standard oral diet preoperatively and intravenous dextrose postoperatively or TPN in addition to oral diet preoperatively and postoperatively. Perioperative nutritional therapy consisted of a solution enriched with 35% BCAA, dextrose, and lipid emulsion (50% medium-chain triglycerides), given intravenously for 14 days (7 days preoperatively for 12 h/day and 7 days postoperatively continuously). The TPN diet provided 30 kcal/kg body weight/day and 1.5 g amino acid/kg body weight/day. The TPN-supplemented group of patients had a reduction in overall postoperative morbidity compared with the control group (34% vs. 55%), predominantly because of fewer septic complications (17% vs. 37%). In addition, less deterioration of liver function was seen, as measured by the change in the rate of clearance of indocyanine green, and less ascites. Postoperative mortality in the perioperative TPN group also was 45% less than in the control group.

Two important aspects of this study are that 90% of the patients had either cirrhosis or chronic active hepatitis, and the average nutritional status of patients in both groups was adequate. The complication rates and deaths were not stratified according to the preoperative severity of malnutrition, and it is unknown whether the beneficial treatment effect of perioperative TPN was beneficial to all patients or only for patients with malnutrition. Nevertheless, the general conclusion of this well-conducted study was that patients with cirrhosis or chronic active hepatitis with HCC who undergo liver resection seem to benefit from perioperative parenteral nutrition. Postoperative parenteral nutrition with solutions enriched with BCAAs in major liver resections for colorectal metatasis has a positive impact on liver function tests during the postoperative period (Vyhnánek et al, 2008).

Pancreatic Cancer

Although the pathogenesis of the cancer anorexia–cachexia syndrome is better understood, few clinical studies have addressed prevention or treatment of this syndrome in patients with pancreatic cancer. A 2005 Cochrane Database meta-analysis demonstrated a benefit of megastrol as compared with placebo for increased weight gain and appetite in anorexia-cachexia related to cancer, AIDS, and other pathologies (Berenstein et al, 2005). Omega-3 fatty acids have been shown to modulate both proinflammatory cytokines and acute-phase reactants. In pancreatic cancer patients, n-3 fatty acids have decreased the acute-phase protein response and increased serum albumin and fibrinogen (Barber et al, 2004). Eighteen patients with advanced pancreatic cancer received oral supplementation with fish oil and after 3 months had an increase in weight, a significant reduction of acute-phase proteins, and stabilization of resting energy expenditure (Wigmore et al, 1996). Similarly, a report of 20 patients receiving eicosapentaenoic acid, an n-3 fatty acid, also showed a benefit in terms of weight gain and an improvement in performance status and appetite (Barber et al, 1999). Barber and colleagues also found a decrease in IL-6, a rise in insulin, and a fall in the cortisol–insulin ratio in patients with pancreatic cancer who received a nutritional supplement (600 kcal/day and 2 g/day of EPA; Barber et al, 2001).

A prospective, randomized, double-blinded clinical trial showed, in the post ad hoc analysis, that although a diet enriched with n-3 fatty acids did not provide a clear-cut therapeutic advantage, a correlation was found between supplement intake and weight gain, lean body mass gain, and quality of life as compared with intake of an isocaloric control diet not rich in omega-3 fatty acids (Fearon et al, 2003). Further investigations need to be made into the above supplements in pancreatic cancer patients to clarify a possible role in improving survival benefit and quality of life.