Multiple organ failure

The syndrome of multiple organ failure (MOFS) has become recognized as a clinical entity only in the past ten to 20 years, primarily as a result of advances in technology, medical information, and scientific research that have allowed the successful treatment of single organ failures. Unfortunately, modern warfare has been the impetus for many of these medical advances. World War I led to the concepts of fracture immobilization and shock resuscitation. With World War II came blood banking and an emphasis on early evacuation and operative treatment of battlefield casualties. As early mortality rates from hemorrhage improved, survivors came to be at risk for late complications, and in the Korean War, acute renal failure became one of the leading causes of death. While early fluid resuscitation, with rapid evacuation and definitive operative treatment, reduced the incidence of acute renal failure, the Vietnam War introduced a new complication of shock, the adult respiratory distress syndrome (ARDS). Initially thought to be a result of overzealous fluid administration, it became clear that ARDS was actually a complication of circulatory shock, infection, and tissue injury in nonlung areas of the body. By the mid 1970s it became increasingly clear that the pathophysiology of ARDS was not confined to the lung but is part of a systemic injury-response pattern now known as the hypermetabolism-organ failure syndrome.

The organ failure syndrome is probably best viewed as the clinical endstage of the systemic hypermetabolic response to injury that is heralded by acute lung injury and followed by hepatic and renal failure and often by death. MOFS is the leading cause of death in critically ill surgical patients, accounting for 75 percent of deaths in the surgical ICU. The average length of ICU stay is 21 days at a cost of approximately $85,000. In survivors, rehabilitation may take months, which is spent in the recovery of skeletal-muscle function, at an estimated additional cost in the range of $300,000.

THE CLINICAL SYNDROME

The hypermetabolism-organ failure syndrome may develop after a variety of initial insults, including severe hemorrhage or sepsis, tissue injury or ischemia, and severe inflammation as in pancreatitis (Fig 1). Common to all of these initial events is a relative or absolute perfusion deficit that may or may not be associated with the usual clinical signs of circulatory insufficiency such as hypotension or oliguria. After initial resuscitation, there is a short period of relative hemodynamic stability.

After 48 to 72 h of relative stability, however, patients who may progress to MOFS enter a phase of persistent hypermetabolism, usually associated with some form of lung injury, which ranges from a mild capillary leak to fulminant ARDS. The phase of persistent hypermetabolism may last 14 to 21 days and is associated with progressive deterioration of renal and hepatic function. It is the development of clinical hepatic failure, however, that defines MOFS and its attendant mortality.

Patients who are going to develop the syndrome will develop low-grade fever, leukocytosis, tachycardia, and tachypnea. The development of diffuse lung infiltrates on chest x-ray film together with reduced pulmonary compliance, progressive arterial hypoxemia, and an increased minute ventilation that frequently necessitates mechanical ventilation characterize the respiratory failure component of the syndrome. In addition, the hypermetabolism phase is characterized by increased cardiac output, with the cardiac index often exceeding 4.5 L/min/[m.sup.2]; a decreased systemic vascular resistance (SVR), often falling below 600 dyne-cm, hyperglycemia; hyperlactatemia; an elevated oxygen consumption index, often exceeding 180 ml/min/[m.sup.2]; and an elevated urinary urea nitrogen excretion, often exceeding 15 g day. This array of clinical and hemodynamic findings is identical to that seen in infection with sepsis. Sepsis is distinguished from "sepsis syndrome" in that sepsis implies a culture-documented bacterial, fungal, or viral cause of the clinical findings.

The hypermetabolic phase of the illness is associated with a mortality range of 25 to 40 percent. Patients that are to recover generally do so by day 9 to 14 postinjury, after the source of the hypermetabolism is eliminated and tissue perfusion is restored. In patients who do not recover early, mortality rises progressively after seven to ten days of the hypermetabolism, with the greatest mortality at 14 to 20 days postinjury. The persistence of the hypermetabolism may be due to a failure to control the source or to a new cause such as infection, or it can occur without a new identifiable source. The serum creatinine rises and a prerenal azotemia occurs. There is a progressive need for volume to support blood pressure and cardiac output together with a need for inotropic support to maintain adequate oxygen delivery to the tissue. Repeated episodes of bacteremia and pneumonia, usually with gram-negative enteric organisms, and viral and fungal infections become a problem even in previously immunocompetent hosts. Nervous system function deteriorates as evidenced by encephalopathy and a peripheral motor and sensory neuropathy. Deteriorating GI function is characterized by stress ulceration and bleeding, ileus, and diarrhea. Biliary dilation and bile stasis are common and frequently occur in the absence of extrahepatic obstruction. Coagulopathy, impaired platelet function thrombocytopenia, and DIC may become manifest as the hypermetabolism/organ failure progresses.

Eventually, hyperbilirubinemia that is disproportionate to elevations in levels of other hepatocellular enzymes and alkaline phosphatase ensues, which, together with reduced hepatic protein synthesis, increased ureagenesis, reduced hepatic amino acid clearance, and a falling hepatic redox potential, signifies the development of clinical hepatic failure and the transition to MSOF (Fig 2). Approximately 40 percent of the patients with hypermetabolism will undergo this transition, and in doing so incur a mortality risk exceeding 90 percent, particularly if a treatable cause such as an undrained abscess cannot be found. This rising bilirubin level in the absence of biliary obstruction or sequestered blood or hematoma is currently the most reliable indicator of the onset of hepatic failure. Under these conditions, a serum bilirubin of greater than 8 mg/dl is associated with a mortality exceeding 95 percent.

The commonly available schemes of severity indexing do not usually predict the occurrence of MOFS or its outcome. The reasons for this probably lie in the postresuscitation state of the patient at admission to the ICU and the time-sequence of the manifestations of the MOFS process as it unfolds days to weeks after injury.

PATHOPHYSIOLOGY

Global physiology is characterized by an increased metabolic rate that may exceed twice the basal metabolic rate. This is characterized by an increase in oxygen consumption, which must be met with increased oxygen delivery and an increase in [CO.sup.2] production that can require a minute ventilation as high as 15 to 20 L/min.

Pulmonary involvement can occur in several ways. Primary lung pathology such as pneumonia or a pulmonary contusion can be the inciting event for the development of the hypermetabolic response and ARDS. Secondary pulmonary pathology can result from a variety of nonlung events (sepsis, shock, etc) and presents in a spectrum ranging from mild acute lung injury to fulminant ARDS. The pulmonary pathology most characteristics is that of ARDS, which is characterized by endothelial cell injury and destruction, deposition of platelet and WBC aggregates in clots of fibrin and cellular debris, destruction of type I alveolar pneumocytes, and an acute inflammatory response that must proceed through all the phases of injury and repair, a process not unlike active wound healing in other parts of the body. These alterations become manifest clinically as decreased lung compliance, mismatching of ventilation and perfusion, arterial hypoxemia, and, frequently, pulmonary artery hypertension with its resultant effects on right heart function.

The cardiovascular response is characterized by high cardiac output and low systemic vascular resistance. The prime mover for this response appears to be an increased peripheral demand for oxygen. Failure to generate this hyperdynamic cardiovascular response is associated with an increased mortality risk and usually occurs as a result of inadequate preload, preexisting cardiac disease, or acquired cardiac dysfunction. Hemorrhage, inadequate volume resuscitation, and microcirculatory endothelial injury allowing interstitial sequestration of vascular volume all may contribute to inadequate preload. Preexisting cardiac disease such as cardiomyopathy or scar is one of the major causes of an inadequate hemodynamic response, particularly in the elderly. Acquired cardiovascular dysfunction occurs in a large percentage of patients and includes such factors as a primary decrease in contractility, reduced diastolic ventricular compliance, and a blunted cardiovascular responsiveness to endogenous or exogenous catecholamines.

Frequently, there is abnormal extraction of oxygen in the peripheral tissue that is characterized by a failure of desaturation of the venous hemoglobin. This phenomenon used to be interpreted as shunt. The metabolic response, however, does not support this interpretation, as will be explained in the metabolic section. This desaturation failure also accounts for the poor monitoring results of central venous oxygen saturation in this setting.

Renal function tends to be maintained early in the course, but a progressive fall in the GFR with oliguric renal failure occurs as the syndrome persists, and it is inevitable after the onset of clinical hepatic failure. In the preresuscitative phase, renal perfusion is compromised by splanchnic vasoconstriction and inadequate circulating vascular volume. Postresuscitation, a maldistribution of blood flow within the renal parenchyma also seems to occur. Catecholamines, angiotensin II, and circulating toxic mediators including prostaglandins probably contribute to the shunting of renal blood flow away from cortical glomeruli. Renal tubular injury from episodes of hypotension, diuretics, and nephrotoxic antibiotics also probably contribute to the renal dysfunction.

The GI tract function as both a target organ and as a potential effector of the hypermetabolism/MOFS. In the setting of severe or persistent perfusion deficits, GI ischemia is manifested as stress-induced gastritis, mesenteric ischemia, ischemic colitis, pancreatitis, acalculus cholecystitis, and ischemic hepatitis. Gastro-intestinal bleeding, ileus, and diarrhea can all occur. The GI tract may be the source of specific toxic mediators that may initiate or potentiate the process. Pancreatic injury or ischemia and pancreatitis can lead to the release of toxic oxygen radicals and activated lysosomal and zymogenic proteases, which in turn lead to activation of the complement, kallikrein-kinin, coagulation, and fibrinolytic cascades. Colonic bacteria colonize the proximal aerodigestive tract with gram-negative enteric organisms that appear to contribute to the repeated episodes of pneumonia and sinusitis. Further, the translocation of intestinal bacteria and toxins to extraintestinal sites may contribute to the ongoing generalized inflammatory response and be a source of recurrent or ongoing sepsis, thus contributing to the onset of organ failure.

METABOLISM

The increased oxygen consumption and carbon dioxide production is reflected in the elevated energy expenditure that can exceed twice resting (Table 1). Relative to starvation, the respiratory quotient (R/Q) is increased in the range of 0.80 to 0.85, reflecting the oxidation of a mixed carbon source for energy production. While there is an absolute increase in the utilization of all substrates, including glucose, there is a reduction in the fraction of calories derived from glucose and an increase in the fraction derived from amino acids. This reduction in the oxidation of pyruvate that is associated with a stoichiometric increase in the release of alanine and lactate, together with the oxidation of 2-carbon fragments from fat and amino acids in the Krebs cycle, represents a form of regulated energy production best characterized as aerobic glycolysis. The reduced activity of pyruvate dehydrogenase in animal models is consistent with this interpretation. This form of energy production is effective, although less efficient than normal aerobic metabolism, and an energy deficit does not appear to be a prominent feature of the metabolism until the late organ failure stage of the process, as long as oxygen transport is optimized. Energy production failure seems to be a preterminal event.

Carbohydrate metabolism is characterized by increased glycogenolysis and gluconeogensis that are not readily suppressible with exogenous glucose or insulin. Gluconeogenic substrates include lactate, alanine, glutamine, glycine, serine, and glycerol. Concomitant with the increased hepatic glucose production, there is increased glucose flow to and uptake in the peripheral tissues. When insulin levels and glucose utilization are actually increased, the hyperglycemia and relative excess in glucose production seem driven in part by an elevated glucagon/insulin ratio. In the periphery, there is increased lactate release with reconversion to glucose in the liver. This increased Cori cycle activity produces heat, but no net energy. Exogenous insulin does not seem to increase the oxidative use of glucose. Lactate also serves as a primary myocardial fuel. Excess calories (over 35 kcal/kg/day) and excess glucose (over 5 g/kg/day) can result in hyperosmolar complications, increased energy expenditure, fatty infiltration of the liver, and increased [CO.sub.2] production.

Fat metabolism is characterized by increased lipolysis and decreased lipogenesis. Plasma ketones are low, yet the hepatic production of ketones is increased relative to baseline, although in the presence of glucose administration, somewhat depressed relative to starvation. The turnover of medium and long chain fatty acids is increased. The plasma fatty acid profile changes in that oleic acid is increased but linoleic acid and arachidonic acid are decreased. The clearance rate of long chain fatty acid triglycerides becomes reduced, primarily through a reduction in peripheral lipoprotein lipase activity. If excess lipids are given (over 1 g/kg/day), several complications may occur, including hyperlipemia, a fall in [PaO.sub.2], bacteremia, and suppression of in vitro tests of PMN and lymphocyte function. [TABULAR DATA OMITTED]

As multiple organ failure develops, the endogenous respiratory quotient can exceed 1.0, indicating net lipogenesis, primarily in liver; and hypertriglyceridemia becomes manifest, perhaps reflecting reduced lipoprotein lipase activity in skeletal muscle and adipose tissue.

Protein metabolism is characterized by a marked increase in protein catabolism. Protein synthesis is increased relative to starvation, but is significantly decreased relative to the rate of protein catabolism. Thus, the rate of net protein catabolism is increased and the lean body mass becomes rapidly depleted (autocannibalism). Amino acids are mobilized from skeletal muscle, connective tissue and unstimulated gut and used to support wound healing, the cellular inflammatory response and the hepatic synthesis of acute phase reactant proteins. This amino acid mobilization far exceeds that which occurs from bedrest alone and can exceed 20 g/day loss in the urine as urea. In addition to this redistribution of mobilized amino acids, the oxidative use of amino acids, and in particular the branched-chain amino acids, is also increased, occurring primarily in skeletal muscle.

The excessive protein catabolism is not readily suppressible by exogenous amino acids or by exogenous fuel sources such as fat or glucose, as occurs in starvation. The synthetic rate, however, is responsive to exogenously administered amino acids and if adequate amino acids are given, the synthetic rate can be increased to approach the catabolic rate and nitrogen equilibrium can be achieved. Doses of 1.5 to 2.0 g/kg/day are necessary to achieve this end point. There does appear to be a survival benefit in the attainment of nitrogen balance, although the loss of lean body mass can be expected to continue through the course of the illness.

As multiple organ failure ensues, total body and hepatic synthesis rates fall, and the absolute and relative catabolic rates increase. Ureagenesis increases and plasma aromatic amino acids rise. As a preterminal event, plasma levels of all amino acids begin to rise, signaling the near-total failure of hepatic protein synthesis.

ETIOLOGY

Mechanistically, the hypermetabolism/MOFS is best thought of as a systemic extension of the local inflammatory response. The initial events appear to occur in the microcirculation. The initiating event is usually associated with a reduction in microcirculatory perfusion that induces endothelial cell injury, platelet aggregation, infiltration of neutrophils, and activation of the complement kallikrein-kinin, coagulation and fibrinolysis cascades, as well as cytokine release. Increased capillary permeability leads to further losses of intravascular volume into the extravascular space, aggravating the existing perfusion deficit. With reperfusion, cell debris and a variety of toxic inflammatory mediators are released into the systemic circulation, contributing to continued injury locally and at distant sites such as lung and liver. A great many mediators that may ultimately contribute to the pathogenesis of the MOFS have been described. Although a detailed discussion of individual mediators is beyond the scope of this review, the major groups of mediators include: platelet factors such as platelet-activating factors; cytokines and prostanoids from macrophages and endothelial cells; metabolic byproducts such as octopamine; numerous peptides such as those promoting proteolysis and reduced muscle uptake of amino acids; oxidants from PMNs; and a number of lysosomal proteases.

The importance of bacterial endotoxin should not be underestimated, although it most certainly requires other mediators to elicit its toxic effects. The precise interaction of these mediators is the subject of intense research.

At this point, it seems reasonable to hypothesize that the shock and reperfusion phases have the microcirculation and the endothelial cell as their targets. The neuroendocrine axis, platelets, and the PMN appear to be the primary effectors in these early phases. After resuscitation, the phase of stable hypermetabolism is entered. This later metabolic response appears to be driven primarily by the mononuclear cell mass, with the macrophage playing a central role.

The macrophage appears to act at several organizational levels during the hypermetabolism and organ failure phases of the syndrome. Inflammatory macrophages produce a variety of substances that regulate the local inflammatory and repair processes. Some of these products, such as the interleukins, prostanoids, and tumor necrosis factor (TNF), also produce a variety of systemic effects including changes in microcirculatory tone and permeability. Others affect different cell systems and interfere with their function, such as [PGE.sub.2] inhibition of antigen-induced lymphocyte proliferative responses.

Stationary macrophages also appear to play a central role in the modulation of metabolic activity of anatomically associated target cells. Supernatants from endotoxin-stimulated Kupffer cells appear to alter protein synthesis in hepatocyte cell cultures, causing decreased production of albumin and increased production of acute-phase reactant proteins. Interestingly, this interaction between Kupffer cells and hepatocytes appears to be bidirectional. In co-culture systems, endotoxin stimulation leads to production of much greater quantities of interleukin-1 and TNF than is observed from Kupffer cells alone. Excessive or repetitive stimulation in this model has the potential of causing hepatotoxicity.

Finally, the macrophage appears to be involved in the translocation of intestinal bacteria. Luminal organisms appear to be engulfed by the macrophage and transported to extrainstestinal sites. Due to a failure of intracellular killing, these organisms become a potential source of infection and sepsis.

While the inciting events have been identified, many of the mediators characterized and some of the cellular mechanisms described, the importance of infection in the MOFS cannot be overemphasized. Even when sepsis is not the initiating event, it is virtually always a problem as the syndrome progresses, whether as the result or the cause of the organ failure process. Episodes of bacteremia, urinary tract infection, line infections and especially repeated episodes of pneumonia characterize the process. Virtually all in vitro tests of immune function show suppression, T4/T8 ratios are reduced, and skin test anergy is almost universal.

The gut is hypothesized to play a major role in the origin of these recurrent infections with enteric organisms. Progressive colonization of the proximal aerodigestive tract with enteric flora in the presence of ileus, obstruction, nonuse, or manipulation of the gastric pH appears to be a primary source of organisms causing pulmonary or sinus infections. Additionally, translocation of intestinal bacteria through grossly intact intestinal mucosa, causing infection in mesenteric lymph nodes, liver, and spleen, has been demonstrated in experimental animals and appears to be potentiated by shock, burns, trauma, and malnutrition. While gut bacterial translocation does occur in humans, its clinical significance in hypermetabolism/organ failure has not been established. Circumstantial evidence of a role for translocated intestinal bacteria exists in that in over 30 percent of bacteremic trauma patients dying of sepsis and multiple organ failure, a source of infection is never identified, either clinically or at autopsy.

CURRENT THERAPY

The therapeutic approach to the hypermetabolism/organ failure syndrome is primarily one of prevention and support and includes source control, resuscitation of the microcirculation, and metabolic support. The key to the successful management of the problem lies in prevention, so that the vicious cycle of tissue injury and perfusion deficit, followed by more tissue injury, is broken by early and aggressive therapeutic interventions, and organ support is provided to "buy time" for the patient to heal on his own.

Source control is directed at eliminating the inciting event or risk factors for hypermetabolism and organ failure. Such treatment would include control of hemorrhage, drainage of infection, fracture stabilization, and early excision and grafting of burn wounds. Necrotic tissue should be debrided and appropriate antibiotic therapy instituted.

Together with source control measures, resuscitation of the microcirculation is the most important therapeutic intervention. Unfortunately, hemodynamic restoration to the clinical end points of blood pressure and urine output are often inadequate; and persistent, subclinical circulatory shock remains one of the primary causes of hypermetabolism and organ failure. Ideally, the end point of adequate microcirculatory resuscitation is the attainment of flow-independent oxygen consumption and lactate production, meaning that oxygen delivery ([DO.sub.2]) is increased until oxygen consumption ([VO.sub.2]) no longer rises and lactate production no longer falls. Inherent in this treatment mode are restoration of adequate oxygen-carrying capacity and, often, the attainment of supranormal blood volume and cardiac output.

Finally, current treatment includes the provision of adequate metabolic support. Control of malnutrition as a covariable of morbidity and mortality does appear to be of survival value. The correlate of this outcome appears to be the attainment of nitrogen balance, although, as noted previously, the excess catabolic rate is not responsive to exogenously administered amino acids. Amino acid formulas supplemented with branched chain amino acids appear to be more efficient at promoting nitrogen retention, supporting hepatic protein synthesis and minimizing urea production. In addition to providing adequate nitrogen, current metabolic support is aimed at minimizing the complications of excess carbohydrate and fat administration. Current recommendations for metabolic support include 30 to 35 nonprotein cal/kg/day supplied as 4 to 5 g/kg/day of glucose and 0.5 to 1 g/kg/day of fat. Protein requirements are in the range of 1.5 to 2 g/kg/day, depending on the degree of stress. The ideal goal of amino acid therapy is the attainment of positive nitrogen balance. In addition, surviving patients generally tend to respond to adequate nutritional support with rising visceral protein levels.

In general, malnutrition appears to be a covariable in morbidity and mortality, but once present, it can become a primary effector. Thus, nutritional support must be started early, although the precise timing remains controversial. In patients at risk for hypermetabolism and organ failure, the institution of nutritional support as soon as resuscitation and hemodynamic stability have been achieved seems to be the most rational approach.

FUTURE THERAPY

While the three-part approach to the management of MOFS--including source control, resuscitation of the microcirculation, and metabolic support--has been effective in reducing the occurrence of, and thus the mortality risk from, the problem, it seems that we have progressed as far as possible using current treatment modalities. In addition to continued attempts to prevent the problem, future therapy will likely be aimed at the specific cellular mechanisms of hypermetabolism, mediators of injury, and stimulants of organ healing and repair. Several potential areas of intervention are discussed briefly below.

MODULATION OF THE ENDOGENOUS MICROFLORA

Because the endogenous GI flora is thought to play a significant role in the development of nosocomial infection in critically ill patients, manipulation of the flora is potentially beneficial. Elimination of the protective acidic gastric content with antacids and [H.sub.2]-receptor antagonists potentially allows colonization of the proximal aero-digestive tract with coliform organisms, thus promoting infection in the lungs and sinuses. Whether cytoprotective or mucous barrier therapies will be successful in controlling stress ulceration without alkalinization while preventing this colonization, remains to be elucidated in clinical studies.

In addition to proximal migration of coliform organisms, the translocation of intestinal organisms through the gut wall in critically ill patients is considered a potential cause of persistent sepsis and hypermetabolism. Selective alteration of the gut flora with non-absorbable enteral antibiotics directed at aerobic enteric organisms and yeast, while sparing the anaerobic bacteria that apparently have little propensity to translocate, appears to be of benefit in reducing the colonization of the oropharynx and nasopharynx and genitourinary tract, and to reduce the incidence of serious nosocomial infections with enteric organisms. To date, however, there has been little effect on either the incidence of MOFS or the mortality rate in this group of patients. Perhaps, while sepsis is sufficient for entrance into the process, the recurrent bacteremias and infections are signs of both gut and immune failure.

NUTRITIONAL MANIPULATION

There is potential to alter cell-cell interaction, the synthesis of specific inflammatory mediators and to support some specific metabolic pathways by directed dietary manipulation. Fish oil, which is high in the [omega]-3 polyunsaturated fatty acids eicosapentanoic acid and docosahexanoic acid, is known to reduce inflammation in animal models of arthritis, systemic lupus erythematosus, and amyloidosis. [omega]-3 Polyunsaturated fatty acids are known to be incorporated into cell membranes and then alter the synthesis of inflammatory mediators. Eicosapentanoic and docosahexanoic acids are competitive inhibitors of both the cyclo-oxygenase and lipo-oxygenase pathways of arachidonic acid metabolism and, therefore, lead to reduced synthesis of the dienoic prostaglandins and the leukotrienes. Additionally, when these [omega]-3 fatty acids are metabolized via the lipo-oxygenase pathway, they are converted to leukotrienes that are much less metabolically active than those synthesized from arachidonic acid. These effects are reproducible in cell cultures and animal models and are currently in clinical testing.

In addition, the [omega]-3 fatty acids can alter the production of tumor necrosis factor and interleukin 1 by Kupffer cells in response to endotoxin stimulation (Fig 3). The precise mechanism of these effects remains to be clarified.

Glutamine is a primary nitrogen transporter from the periphery to the visceral organs; is one of the major amino acids released from skeletal muscle in the catabolic state; and is a primary gut fuel and trophic factor. Glutamine given to protein-depleted rats leads to increased small bowel weight, increased villus height, and increased DNA formation. Glutamine may also protect the gastric mucosa from aspirin-induced ulceration, and adequate glutamine may be required for maintenance of gut mucosal integrity. A beneficial role in hypermetabolism/organ failure, however, remains to be proved.

Metabolizable fiber with the production of medium chain fats in the colon has been shown in animal models to promote colonic anastomotic healing and mucosal integrity. Clinical trials to demonstrate efficacy are being planned.

The enteral route for nutrition, in theory. should be superior to the IV route. A beneficial effect of early enteral feeding three to four days postinjury, however, has not been elucidated. Whether starting within hours of injury will be beneficial, remains the subject of current research protocols.

BLOCKERS OF MEDIATOR SYNTHESIS OR EFFECT

A variety of compounds have been used to block the synthesis or effect of many of the presumed mediators of shock, hypermetabolism, and multiple organ failure in animal models. The list includes inhibitors of prostaglandin and leukotriene synthesis and/or effect; antagonists of PAF; antibodies against TNF; and various types of antioxidant therapies. Although a great deal has been learned about the mechanisms of injury and regulation of metabolism using these agents, clinical trials to date have not demonstrated outcome efficacy.

ANTI-ENDOTOXIN THERAPY

Whether primarily or through a variety of secondary mediators, bacterial endotoxin is thought to be one of the major effectors of sepsis, hypermetabolism, and organ failure. In addition to standard antimicrobial therapy, other options for treating or preventing the effects of endotoxin are being investigated. Polyclonal and monoclonal antibodies have been developed against lipid A, a common core lipopolysaccharide of bacterial endotoxin in a variety of bacterial species. A murine monoclonal antibody has shown cross-species reactivity in vitro (ELISA) and reduced [LD.sub.50] in a murine model of endotoxemia using a variety of bacterial endotoxins. However, the protective effect was restricted to a more limited number of bacterial species in a bacteremic animal model. Clinical testing with these antibodies is currently being completed and the results are anxiously awaited.

The problem of bacterial endotoxin and its effects is also being approached from a nonimmunologic standpoint. Lipid X, a monosaccharide precursor of lipid A, and 3-aza-lipid X, the diamine analog of lipid X, have been shown to inhibit the neutrophil-priming effect of bacterial endotoxin and to attenuate the effect of LPS to markedly enhance the "respiratory burst" of neutrophils when stimulated with a variety of activators. This competitive binding of lipopolysaccharide receptor sites by lipid X is distinct from the effect of Polymixin B, which inactivates lipopolysaccharide by binding to it.

HORMONAL MANIPULATION

To date, hormonal manipulation with insulin and somatostatin has not been useful in modulating the hypermetabolism-organ failure syndrome. Recombinant human growth hormone has shown promise for its anabolic effects in studies in healthy human volunteers and in uncomplicated post surgical patients. A randomized multicenter trial to evaluate the effects of growth hormone on recovery and survival in patients with multiple organ failure is ongoing.

Epidermal growth factor (EGF) is released from salivary glands and is thought to promote mucosal development in the proximal GI tract. EGF is also an anabolic hormone that promotes nitrogen retention in experimental animals. EGF promotes increased DNA and protein synthesis in small intestinal mucosa and also promotes elevated mucosal sucrase and maltase activity. These effects are distinct from but additive with the effects of glutamine. Whether it will be clinically useful awaits the results of clinical trials.

MISCELLANEOUS EXPERIMENTAL THERAPY

Dichloroacetate is a compound known to reverse the inhibition of the pyruvate dehydrogenase enzyme complex that exists in the setting of sepsis and hypermetabolism. By reversing pyruvate dehydrogenase activity inhibition, dichloroacetate could permit increased glucose metabolism and decreases in sepsis induced elevation in pyruvate and lactate. Perhaps as a result of this restoration of glucose metabolism, dichloroacetate may allow decrease protein catabolism and reduce the branched chain amino acid dependence of the septic state. Whether this compound will be of clinical usefulness is controversial.

Patients with ARDS from trauma and sepsis demonstrate low levels of plasma fibronectin (opsonic [[alpha].sub.2]-surface-binding glycoprotein), and infusion of cryo-precipitate or purified fibronectin may improve pulmonary function in these patients. Fibronectin may act as a circulating opsonin to precipitate removal of bacteria and bacterial products, but may also function as a tissue intercellular "glue" to prevent alveolar capillary membrane permeability changes associated with sepsis. Whether fibronectin infusion will be useful clinically awaits prospective, randomized testing.

REFERENCES

[1] Cerra FB. Hypermetabolism, organ failure and metabolic support. Surgery 1987; 101:1-14

[2] Shoemaker WC, Appel PL, Kram HB. Role of oxygen transport patterns in the pathophysiology, prediction of outcome, and therapy of shock. In: Bryan-Brown CW, Ayres SM, eds. New horizons: oxygen transport. Fullerton, CA: Society of Critical Care Medicine, 1987:65-92

[3] Cerra F, Negro F, Eyer S, Abrams J, Perry J. Multiple organ failure syndrome: patterns and effect of current therapy. Arch Surg (in press)

[4] J Parenteral Enteral Nutrition. 1988; 12(suppl 6)

[5] Bihari DJ, Cerra FB, eds. New horizons: multiple organ failure. Fullerton, CA: Society of Critical Care Medicine, 1989

COPYRIGHT 1989 American College of Chest Physicians
COPYRIGHT 2004 Gale Group