POSTOPERATIVE JAUNDICE

Postoperative jaundice occurs as a result of overproduction and underexcretion of bilirubin, direct hepatocellular injury, or extrahepatic obstruction. ^[130] Most etiologies of postoperative jaundice become manifested within 3 weeks of surgery and can be classified as either mild (<4 mg/dL) or severe (>4 mg/dL). Table 55-2 lists the most common etiologies of postoperative jaundice based on the nature of the pathophysiology.^[130] If jaundice is secondary to hemolysis, the liver's capacity to conjugate bilirubin (normal rate of production, approximately 250 to 300 mg/day) has been exceeded. Hemolysis-induced anemia is therefore associated with unconjugated (indirect) hyperbilirubinemia and, in the perioperative period, can usually be attributed to either drug or mechanically induced erythrocyte destruction. The primary mechanisms by which certain drugs cause hemolysis and subsequent jaundice are (1) the drug adsorption type, whereby an antibody (IgG) reacts with a drug bound to the red blood cell membrane; (2) the neoantigen type (so-called

TABLE 55-2 -- Etiology of postoperative jaundice by time course

Immediate postoperative jaundice (<3 wk)

Hemolysis

Anesthesia

Hypotension/hypovolemia

Drugs

Infection/sepsis

Bleeding/resorption of hematoma

Bile duct ligation/stricture/surgical injury

Hepatic artery ligation

Retained common duct stone

Postoperative pancreatitis/cholecystitis

Acute viral hepatitis

Gilbert's syndrome/Dubin-Johnson syndrome

Inflammatory bowel syndrome

Heart failure

Pulmonary postoperative jaundice

Blood transfusion

Delayed postoperative jaundice (>3 wk)

Drugs

Blood transfusion

Post-intestinal bypass status

Total parenteral nutrition

innocent bystander), whereby the drug combines with the erythrocyte membrane and an antibody reacts with the newly formed antigenic site to activate the complement cascade; and (3) the autoimmune type caused by an autoantibody (IgG) to erythrocytes. Medications commonly used perioperatively and associated with one or more forms of hemolytic anemia are listed in Table 55-3 .^[131]

The degree of associated jaundice is related to the rate of hemolysis and excess production of bilirubin, with the neoantigen mechanism most likely to cause acute anemia, hemoglobinuria, and renal failure. The diagnosis is made by the laboratory constellation of anemia, indirect hyperbilirubinemia, positive direct antiglobulin test, low serum haptoglobin, and a peripheral blood smear note-worthy for fragmented erythrocytes and reticulocytosis. Mechanical destruction of erythrocytes can also occur from surgically implanted prosthetic heart valves or from intrinsically diseased valves.^[132]

Multiple blood transfusions can increase levels of unconjugated bilirubin because approximately 10% of stored

TABLE 55-3 -- Medications associated with hemolysis

Acetaminophen

Cephalosporins

Hydralazine

Ibuprofen and other nonsteroidal anti-inflammatory drugs (e.g., diclofenac, tolmetin)

Insulin

Intravenous contrast media

Penicillin and all derivatives (e.g., ampicillin, methicillin)

Procainamide

Ranitidine

Sodium thiopental

2220

Hepatocellular injury can cause postoperative jaundice (as previously discussed) by drug, ischemia, or virally mediated mechanisms. In addition to potential anesthetic-induced hepatotoxicity, a number of commonly prescribed drugs can cause hepatocellular injury that mimics either hepatitis or cholestasis based on liver test abnormalities ( Table 55-4 and Table 55-5 ). With the exception of acetaminophen, which can directly cause hepatocyte necrosis, drug-induced hepatotoxicity is primarily the result of either idiosyncratic reactions or alterations in bile flow resulting in cholestasis.

Postoperative jaundice may also occur as a result of hepatic hypoperfusion. Cardiogenic and noncardiogenic shock can decrease HABF and PBF and cause hepatocellular necrosis. Marked elevations in aminotransferase levels are common, with hyperbilirubinemia usually a late finding. In addition to hemodynamic alterations in hepatic blood flow, sepsis or the systemic inflammatory response syndrome is associated with cholestasis and jaundice, presumably because of the effects of circulating endotoxins or other inflammatory mediators on bile formation and flow.^[182] Finally, it is important to exclude viral hepatitis as the cause of postoperative jaundice because viral exposure may have occurred before surgery.

Obstruction of the common bile duct from a gallstone, stricture, or tumor can result in failure to excrete conjugated bilirubin (obstructive jaundice) and has been associated with significant perioperative morbidity and mortality. Factors associated with a poor outcome after surgery in patients with obstructive jaundice include extrahepatic obstruction as a result of malignancy, malnutrition, hypoalbuminemia, low hematocrit (<30%), azotemia, and the level and duration of hyperbilirubinemia.^{[183] [184]} Dixon and coauthors^[185] cited an overall mortality rate of 9.1% in a retrospective analysis of 373 patients requiring surgery for obstructive jaundice, but a 60% mortality rate for patients with anemia, malignancy, and marked hyperbilirubinemia (>11 mg/dL). Of paramount significance is the association of acute renal failure with obstructive jaundice.^{[186] [187]} The incidence of postoperative acute renal failure is approximately 8% to 10% and correlates directly with the degree of hyperbilirubinemia, with mortality rates as high as 70% to 80% reported. Neither bilirubin nor bile acids themselves are directly nephrotoxic, but experimental evidence suggests a negative chronotropic, vasodilatory, and diuretic effect of bile salts resulting in systemic and thus renal hypoperfusion. The changes appear to be related to enhanced absorption of endotoxin from the gastrointestinal tract because of low intestinal levels and high serum levels of bile acids. This stimulates a release of inflammatory mediators, in particular endothelin, a potent renal vasoconstrictor that leads to a further decline in renal perfusion.^{[188] [189]} Other experimental data have implicated endotoxemia-induced nitric oxide release as a further mediator of renal hypoperfusion.^[190] These patients clearly pose a challenge to the anesthesiologist, who is often forced to rely on the limitations of urine output as a method of monitoring renal perfusion. It is more reliable to monitor effective blood volume and cardiac function with either a central venous or pulmonary artery catheter or perform transesophageal echocardiography, with the goal of maintaining renal perfusion by augmenting cardiac output.^[90] No data have documented the clinical benefit of mannitol, furosemide, or dopamine in perioperative renal protection,^{[90] [191]} and the potential benefit of endothelin receptor blockers has yet to be thoroughly assessed in clinical studies.^[188]