Search

In an attempt to meet better the needs of patients with hypermetabolic stress, amino acid formulas have been modified (MAA). As part of a total parenteral nutrition regimen, MAA have been associated with improvement in parameters of nutrition and survival. The safety and efficacy of these formulas in balanced enteral nutrition was assessed in malnourished patients with moderate to high levels of stress undergoing surgery. The nutritional targets were 5 gm/kg/day glucose, 0.8 gm/kg/day fat, and 0.27 gm/kg/day amino acid nitrogen. In an initial nine studies, the MAA nutrition was safe even in the presence of ileus. A double-blind, randomized, prospective trial was then undertaken in 18 patients in an isocaloric, isonitrogenous design with the nitrogen source as MAA (44% branched-chain amino acids) or standard amino acids (28% branched-chain amino acids). The patients receiving MAA had enhanced nitrogen retention, visceral protein mass, and indices of survival.

Although decreases in hepatic adenosine triphosphate (ATP) levels during ischemia are thought to reflect loss of hepatic energy reserves and decreased viability, such changes have not been correlated with a clinically relevant index of hepatic function or viability. Ability to clear amino acids from plasma has been shown to correlate with function of the allograft in hepatic transplantation and survival after portal decompression in patients with hepatic cirrhosis. The effects of 60 minutes of warm ischemia in two groups of mongrel dogs were studied to assess the relationship between loss of ATP and amino acid clearance. One group (shunted) had portal decompression during the ischemic period and the other (portal stasis) did not. There was a significant correlation between loss of ATP and amino acid clearance after ischemia. Although the effects of ischemia on the liver were similar in both groups, the portal stasis group demonstrated significantly elevated SGOT levels during reperfusion that were related to impaired net adenine monophosphate synthesis and suggestive of ongoing injury. These data support the contention that loss of ATP during ischemia is associated with reduced functional capacity. In addition, they suggest that portal stasis produces toxic products that can impede hepatic recovery from ischemia.

Sixteen canine myocardial pedicle preparations were divided into four groups. Group I received 60 minutes of reversible ischemic anoxia, Group II 90 minutes, Group III 120 minutes, and Group IV 180 minutes. Transmural ischemic infarction was present in Group I. With longer periods of anoxia, capillary disruption occurred, first in the subendocardium (90 minutes) and then transmurally (180 minutes). Reperfusion after ischemic, anoxic time periods of 90 minutes or greater produced hemorrhagic necrosis, the extent of which was directly proportional to the duration of the preceding ischemic, anoxic period. One hundred and twenty patients with aortic valve replacement surgery then were analyzed retrospectively. Of the 25 deaths, seven were due to immediate postinfarction, five from subendocardial hemorrhagic necrosis. The common factor in the patients with hemorrhagic infarction was a markedly reduced coronary flow (45 to 55 ml. per minute) for more than 70 minutes during the interval of cardiopulmonary bypass. Although other factors may be involved, it is hypothesized that the main cause is a degree of localized ischemia of sufficient duration to induce capillary disruption and subsequent postbypass hemorrhagic necrosis when normal coronary artery perfusion is resumed.

Twenty-two patients with cirrhosis and acute encephalopathy who were refractory to medical therapy were entered into a randomized, double-blind prospective trial. This trial consisted of either neomycin or a modified amino acid solution rich in branched chains and low in aromatic amino acids and methionine (F080) in the presence of isocaloric amounts of dextrose. The groups were indistinguishable from each other by clinical or laboratory criteria; they were primarily patients who had undergone operation and they would tolerate only 30 gm of oral protein or intravenous standard amino acids. The group receiving F080 had a faster and more complete improvement in encephalopathy. This improvement correlated with the plasma molar ratio and occurred with a lower mortality rate. In addition, the patients also tolerated twice the amino acid load without encephalopathy and were in positive nitrogen balance. Modified metabolic support is effective in the setting of acute liver failure in chronic cirrhosis, particularly in patients who have undergone operation.

The septic response in man appears to be a disease in which the infecting agent induces a state of disordered metabolic control in the host. The abnormal regulation of metabolic pathways causes a diversion of substrate utilization toward gluconeogenesis and ketone body formation and a reduction in oxidative energy-producing metabolism. This state of metabolic insufficiency is reflected in the pattern of cardiorespiratory, vascular, and physiological compensation. The precise magnitude of the resultant physiological compensation and its rate and direction of change can be quantified by the use of physiological state trajectories which also reflect the magnitude of underlying metabolic derrangements. The prognostic and therapeutic implications of these changes are discussed.

In an attempt to determine the initial function of hepatic allografts, several metabolic indicators of hepatic function were studied intraoperatively in 12 cases of hepatic transplantation. The operation was divided into three sampling periods: baseline, anhepatic, and reperfusion. During the baseline period plasma lactate levels rose at 2.6 mmol/L/hr and continued to rise at a similar rate during the anhepatic period. Baseline period total free plasma amino acid levels (TFPAA) rose at a moderate rate of 0.4 mmol/L/hr. During the anhepatic period TFPAA levels rose at a fivefold greater rate than during baseline (p less than 0.01). The ability of the hepatic allograft to reduce abnormal levels of TFPAA and lactate during the reperfusion period was associated with reduced morbidity in the first 48 hours after transplantation. Intraoperative clearance of accumulated TFPAA is currently the best means of assessing initial allograft function. Elevated preoperative total serum bilirubin levels were also associated with increased early morbidity in hepatic transplant recipients.

A blinded, prospective trial of the effects of branched-chain amino acid (BcAA)-enriched total parenteral nutrition (TPN) versus standard TPN was undertaken in nonseptic noncirrhotic abdominal surgery patients and patients with multiple traumatic injuries. The study reflected data from the immediate 7-day postoperative period. With isocaloric and isonitrogenous input, the BcAA-TPN patients achieved positive nitrogen balance on day 3. Although the urinary nitrogen output was decreased by day 3 in BcAA-TPN, the plasma BcAA levels did not increase until day 6. With no change in 3-methylhistidine urinary excretion, the early nitrogen retention with BcAA-TPN probably reflects a stimulation of protein synthesis. The ability to favorably modulate the metabolic stress response with alternate fuels has become a clinical reality.

To evaluate the effects of pulsatile preservation, 34 heterotopic renal autografts with immediate contralateral nephrectomy were divided into four groups with either the pulsatile preservation or cold storage technique. Group I had pulsatile preservation with plasma-albumin perfusate at a mean pressure of 60 mm. Hg and a colloid osmotic pressure (COP) of 20 to 50 cm. H2O. Group II had pusatile preservation with the same plasma-albumin perfusates as in Group I, but at a mean pressure of 30 mm. Hg. Group III had pulsatile preservation with albumin-phosphate buffer solution at a mean pressure of 30 mm. Hg. Group IV used cold storage technique. Group I demonstrated severe endothelial destruction and denuding of basement membrane. The changes were inversely related to COP. None of these kidneys functioned and, following transplantation, had persistent severe tubular damage and plugging of glomerular capillary loops with platelets, fibrinogen, and white cells. In Groups II and III, there were minimal changes, consisting of widening of the endothelial pores and focal endothelial damage. All of these kidneys functioned following reimplantation with minimal ultrastructural alterations. Group IV had no glomerular changes following preservation or following transplantation. All preservation techniques produced proximal tubular damage. There is a definite mechanical lesion associated with pulsatile preservation, which can be minimized by utilizing low perfusion pressures and solutions of high colloid osmotic pressure and which has the potential of producing a picture resembling hyperacute rejection following transplantation.

Polyunsaturated fatty acid (PUFA) profiles are abnormal in a variety of clinical conditions that are commonly seen in the surgical intensive care unit. PUFA profiles in the serum phospholipids were studied by capillary gas chromatography in 22 critically ill, hypermetabolic surgical intensive care unit patients. All patients received continuous total parenteral nutrition (TPN) by central vein with trace elements and vitamins. Eleven patients received daily supplementation with 50 gm of safflower oil-based lipid emulsion (TPN + L group). No lipid was administered in the other 11 patients (TPN). Both groups showed deficiencies in 18:2 omega 6 linoleic acid from the time of onset of TPN. In TPN the deficiency was progressive; in the TPN + L group, lipid administration prevented this progression but did not restore 18:2 omega 6 levels to normal. In the TPN group levels of 18:1 omega 9 and its metabolite 20:3 omega 9 (the triene of deficiency) increased over time, consistent with a mild essential fatty acid deficiency. These changes in omega 9 acids were not seen with lipid supplementation. Despite low levels of 18:2 omega 6 in both groups, the levels of omega 6 metabolites were normal or increased. Levels of 20:4 omega 6 (arachidonate) remained normal or slightly decreased in the TPN group but were decreased in the TPN + L group. Levels of arachidonate metabolites, in particular 22:5 omega 6, were increased in the lipid-supplemented group. We concluded that stressed patients receiving TPN develop mild essential fatty acid deficiency that is only partially correctable by lipid supplementation and that administration of supplemental lipid to these patients stimulates arachidonic acid conversion to 22:5 omega 6.

Gut malnutrition in patients with persistent hypermetabolism is hypothesized to be an important factor in postseptic multiple organ failure syndrome (MOFS). The hypothesis was made that enteral nutrition (EN) started at the onset of hypermetabolism could reduce the incidence of MOFS. Sixty-six patients with persistent hypermetabolism 4 to 6 days after onset of sepsis were prospectively randomized to receive either parenteral nutrition (PN) or enteral nutrition (EN) at 1.5 gm protein/kg/day and 30 nonprotein calories/kg/day; the EN and TPN were of the same composition. There was no reduction in either the incidence of MOFS or mortality attributable to the route of nutrition administration. The PN group tended to have better visceral protein support; the EN group had more gut complications. When analyzed, the type of formula given did have an effect on the nutritional outcome but not on the mortality rate. A formula with a nonprotein-calorie-to-nitrogen ratio of 100:1 was associated with more nitrogen retention, higher levels of visceral proteins, and better gut tolerance. The route of nutrition administration does not seem to affect the incidence of postseptic MOFS or mortality when hypermetabolism is already present and when commercially available nutritional formulas are used. The relationships among the route of nutrition, the type of enteral formula, and the disease process of hypermetabolism and MOFS appear to be complex and require much more investigation before the role of the gut and enteral nutrition can be defined.

Alterations in hepatic function are seen in sepsis or multiple-system organ failure. We have hypothesized that Kupffer cells (KCs) within the liver alter the function of contiguous hepatocytes after exposure to septic stimuli. Using an in vitro coculture system, we have found that lipopolysaccharide (LPS) induced a 40% to 60% decrease in cocultured hepatocyte protein synthesis but had no effect on hepatocytes alone. Coculture in the absence of LPS resulted in enhanced hepatocyte protein synthesis proportional to the number of KCs or macrophages (M0s). Conditioned medium (CM) from LPS-triggered coculture or KC alone decreased protein synthesis of hepatocytes whereas CM from hepatocytes alone had no effect. Gel filtration of active M0-CM showing maximal hepatocyte inhibitory activity was present in fractions between 15,000 and 30,000 daltons. This inhibitory activity was inactivated by exposure to 65 degrees C for 30 minutes. Although CM from untriggered M0 did not inhibit hepatocyte protein synthesis, lysates from both LPS-triggered and control M0 were inhibitory. These results show that M0s/KCs exposed to septic stimuli decrease hepatocyte protein synthesis via heat labile, soluble mediator(s) that may already be synthesized within untriggered cells. We hypothesize that soluble M0/KC mediators normally modulate hepatocyte function and that this normal homeostatic control is profoundly altered during sepsis.

To study the role of antibodies in promoting survival during gram-negative bacterial sepsis, we have developed several murine monoclonal antibodies (MAbs). One MAb (5B10) reacted in an enzyme-linked immunosorbent assay with only a single organism (Escherichia coli 0111:B4), while the other (8A1) reacted to all gram-negative whole-cell and lipopolysaccharide (LPS) antigens examined. Either 5B10 MAb, 8A1 MAb, or sterile saline solution was administered intravenously to outbred male Swiss-Webster mice immediately before one of three challenges: (1) viable bacteria intravenously, (2) viable bacteria with hemoglobin intraperitoneally, or (3) intravenous actinomycin D plus LPS. 5B10 MAb provided significant protection against either an E. coli 0111:B4 bacterial or LPS challenge but not against any other organism or type of LPS. 8A1 MAb provided protection against several challenge bacteria (intravenously or intraperitoneally) and against all types of LPS studied except Pseudomonas aeruginosa LPS. A higher dose (2 mg) of cross-reactive antibody (8A1 MAb) was required to produce protection when compared with the type-specific protection produced with 5B10 MAb (0.1 mg). Although ideal antibody therapy would consist of directing a specific MAb against a single microorganism, the acute nature of the disease process and time required to prepare reagents may preclude the use of type-specific MAbs. We believe that the cross-reactive and cross-protective capacity of 8A1 MAb or a similar MAb may be useful in averting the lethal effects of clinical gram-negative bacterial sepsis and warrants testing in the clinical setting.

Alterations in hepatic function are seen in sepsis and/or multiple system organ failure. We hypothesized that Kupffer cells (KC) within the liver may mediate functional alterations in adjacent hepatocytes (HC) in response to bacterial products. We have previously described decreases in rat HC protein synthesis during in vitro cocultivation with peritoneal macrophages in the presence of gentamicin-killed Escherichia coli (GKEC) or endotoxin (LPS). The present studies demonstrate that purified (greater than 95%), syngeneic, or allogeneic KC exposed to GKEC or LPS impart a biphasic response in cultured HC. When HC were cultured alone there was no alteration in 3H-leucine incorporation into HC protein after the addition of GKEC or LPS. When HC were cocultured with KC there was increased protein synthesis compared with HC alone (p less than 0.001). After the addition of GKEC or LPS there was an immediate increase in coculture HC protein synthesis. However, a marked decrease in coculture protein synthesis was seen 16 degrees later (p less than 0.001). To ensure that KC alone were responsible, splenic lymphocytes were added to HC alone or HC/KC coculture, but they did not alter the results. HC viability and appearance were unchanged throughout the experiments. These results show that exposure of KC to microbial products can profoundly alter HC function and support the concept of local KC modulation of HC function during sepsis.

Diets high in n-3 fatty acids appear to have an anti-inflammatory effect, which is thought to be due to decreased macrophage prostaglandin (PG) and thromboxane (Tx) production after incorporation of these fatty acids into cell membrane phospholipids. The effect of n-3 fatty acids incorporation on macrophage monokine release in response to septic stimuli is not well established. Kupffer cells, the fixed macrophages of the liver, were obtained from rats fed diets with fat sources derived from corn oil (CO, control), fish oil (FO, high in n-3 fatty acids), or safflower oil (SO, high in n-6 fatty acids) for 2 or 6 weeks. After exposure to bacterial lipopolysaccharide, Kupffer cells from rats fed FO for 2 or 6 weeks produced less PG and Tx than Kupffer cells from rats fed CO or SO. After 2 weeks of defined diets, interleukin-1 (IL-1) and tumor necrosis factor release were not affected by dietary fat source. In contrast, after 6 weeks of feeding, Kupffer cells from both the FO and the SO groups released less IL-1 and tumor necrosis factor when triggered by lipopolysaccharide than Kupffer's cells from animals fed the control diet that contained CO. These data suggest that altered monokine release from macrophages may contribute to the anti-inflammatory effect of diets high in n-3 fatty acids. Also shown in our results is that prolonged changes in membrane phospholipid content induced by dietary fat source can influence not only PG and Tx production but monokine release as well.

The hypermetabolism organ failure complex remains the predominant reason for both prolonged stay and death in the surgical intensive care unit. What was perceived as isolated organ failure, such as adult respiratory distress syndrome, is now seen as part of the systemic response to injury and repair. Sepsis has become the systemic inflammatory response due to invading microorganisms. What was once perceived as diagnostic of sepsis has been recognized after severe perfusion deficits and in the presence of continuing sources of dead and injured tissue. The transition to organ failure is usually a distinct clinical event and probably represents the onset of clinical hepatic failure. Once present, the organ failure syndrome has a high mortality rate. From a treatment perspective, it is recognized that there is probably no "magic bullet"; that regimens will probably be time dependent and "multiple drug"; and that the best treatment is prevention. Malnutrition, as opposed to changes in body composition that occur as a result of disease process, has become a recognized cofactor in morbidity and death in patients with persistent hypermetabolism and organ failure. The metabolic processes of hypermetabolism have become increasingly categorized and understood. The result has been the development of metabolic support principles that are distinct from those of nutritional support and are designed to prevent the end-organ changes of malnutrition and the development of substrate-limited metabolism, to support organ structure and function, and to attempt to arrest the metabolic processes. The initial problem was to learn to do no harm, an outcome reasonably achieved. In addition, several beneficial results have been recognized including new techniques to better support total body protein synthesis, hepatic protein synthesis, and energy production. Techniques to better support organ structure and function are being tested. No techniques are currently available to control proteolysis and the redistribution of skeletal muscle nitrogen. A great deal of research is still necessary in this field, which is still in its infancy.

Gram-negative bacterial sepsis and shock remain a cause of substantial morbidity and mortality in hospitalized patients despite appropriate antimicrobial therapy, fluid resuscitation, and monitoring. We sought to test the ability of equine antibody directed against core endotoxin, a portion of bacterial outer membrane lipopolysaccharide common to many gram-negative microorganisms, to bind to various gram-negative bacteria in vitro, to promote bacterial phagocytosis by leukocytes, and to protect against lethal gram-negative bacteremia in mice. The importance of the IgG Fc leukocyte attachment site was examined by comparing the ability of intact IgG and IgG F(ab')2 fragments to protect against lethality during murine sepsis. A single horse was immunized with Escherichia coli J5, an organism that expresses a portion of core endotoxin extensively on the cell surface. Preimmunization IgG and F(ab')2 possessed no titer as determined by enzyme-linked immunosorbent assay, did not promote in vitro phagocytosis, and did not protect in vivo. Postimmunization IgG and F(ab')2 possessed a significant titer to E. coli J5 whole cell and lipopolysaccharide antigens and provided significant (p less than 0.05) protection in vivo during lethal intravenous sepsis caused by either E. coli J5, E. coli 0111:B4, Klebsiella pneumoniae, or Pseudomonas aeruginosa. Only postimmunization IgG, but not F(ab')2, promoted in vitro phagocytosis of these same organisms. We therefore hypothesized that protection occurred as a result of antitoxin activity rather than opsonization and phagocytosis, as F(ab')2 fragments were as active as the intact molecule. Further studies must be done to determine the role of anticore endotoxin antibody in conjunction with antibiotics so that appropriate clinical studies may be undertaken.

A major function of the liver is to maintain normal levels of plasma amino acids. This function may depend in part on tissue levels of high-energy phosphate. Experiments were performed in adult mongrel dogs to assess the relationship between plasma amino acid clearance (PAAC) and tissue high-energy phosphate after 90 minutes of warm hepatic ischemia. In addition, when PAAC was assessed in the anhepatic dog, PAAC fell to low levels after hepatectomy. After 90 minutes of warm ischemia, both tissue adenosine triphosphate (ATP) and PAAC fell to low levels, with PAAC similar to those levels observed in anhepatic dogs. Recovery of ATP and PAAC progressed over a 48-hour period but did not reach control values. Mortality rate was 33% in a group of 12 animals at 48 hours after ischemia. At 24 hours after ischemia, total free plasma amino acid levels were significantly higher in those animals that were dying as compared with those that were surviving (4352 mumol/L versus 2850 mumol/L; p less than 0.05). There was a strong correlation between PAAC and ATP (r = 0.81). PAAC appears to be an indicator of hepatic functional recovery and tissue ATP levels after ischemia.

Peripheral postcapillary venous pressure (PCVP) and mixed venous oxygen saturation (SvO2 or PASO2) have been shown to be sensitive indicators of volume status and appear to reflect the adequacy of peripheral perfusion during controlled bleeding. This study demonstrates that in an open-chest dog model with controlled venous return, PCVP is closely and linearly (r2 = 0.6) correlated with cardiac output (CO). Furthermore, oxygen saturation as measured in the central venous system (CVSO2) and peripheral vein PVSO2) were found to be closely and linearly related to PASO2 (r = 0.72 to 0.99 and 0.91 to 0.98, respectively). Thus PCVP, CVSO2, and PVSO2 represent easily and safely obtainable parameters that closely reflect major physiologic variables. During resuscitation after controlled hemorrhage, the PCVP and PVSO2 accurately reflected the restoration of blood volume and were as good as CO and central saturations. Central venous and pulmonary wedge pressures both poorly reflected the return to full volume repletion (P less than 0.01). Thus, PCVP and PVSO2 seem to be reliable indices of volume status and perfusion and do not require invasive, central monitoring.