Survival from sepsis. The significance of altered protein metabolism regulated by proteolysis inducing factor, the circulating cleavage product of interleukin-1

Effect of propranolol on nitrogen and energy metabolism in sepsis

Pharmacologic therapy designed to block adrenergic activity or alter hormonal milieu may modulate energy and protein metabolism in stress. The metabolic effects of propranolol (beta adrenergic receptor blocker) in sepsis was investigated in 22 well-nourished rats that underwent superior vena caval cannulation, cecal ligation, and puncture. Animals were randomly assigned to receive either a continuous infusion of 0.7 mg/day of propranolol combined with parenteral nutrition (n = 11) or parenteral nutrition alone (n = 11). Both groups received isocaloric, isonitrogenous, isovolemic, parenteral nutrition post-operatively for 24 hr. Nitrogen balance was better for the propranolol group than for the control group (+743 +/- 84 mg/kg/day versus +300 +/- 63 mg/kg/day, respectively, P less than 0.05). A significant difference between the pharmacologic therapy and control groups was noted for urinary 3-methylhistidine excretion versus control (0.99 +/- 0.08 micrograms/kg/day versus 7.5 +/- 0.37 micrograms/kg/day, respectively, P less than 0.01). Measured energy expenditure was similar for both pharmacologic therapy and control groups (149 +/- 20 kcal/kg/day versus 134 +/- 11 kcal/kg/day, respectively, P = N.S.). No statistically significant difference was demonstrated for 24-hr survival between propranolol and control groups (73 and 64%, respectively). Continuous, low-dose propranolol promotes nitrogen retention and decreases 3-methylhistidine excretion without altering energy expenditure in parenterally fed septic rats.

Protein synthesis in liver following infusion of the catabolic hormones corticosterone, epinephrine, and glucagon in rats

The mediator(s) and mechanism(s) of acute-phase protein synthesis in the liver following injury and sepsis are not fully known. Elevated plasma levels of the catabolic hormones cortisol, glucagon, and epinephrine have been reported in trauma and sepsis. In previous reports, when these hormones were infused simultaneously (triple hormone infusion), several, but not all, of the metabolic alterations characteristic of sepsis occurred. In the current investigation, the effect of triple hormone infusion on hepatic protein synthesis was studied. Rats were infused intravenously during 16 hours with a solution containing corticosterone (4.2 mg/kg/h), glucagon (2.5 micrograms/kg/h), and epinephrine (6 micrograms/kg/h). Control animals were infused with a corresponding volume of vehicle. Total hepatic protein synthesis in vivo was measured with a flooding dose technique using [14C]-leucine. The synthesis of total secretory proteins and of the individual proteins albumin, complement component C3, and alpha 1-acid glycoprotein was measured in isolated, perfused liver using [3H]-leucine and a recirculating technique. Urinary excretion of nitrogen and plasma concentration of glucose were higher and plasma total amino acid concentration was lower in hormone-infused than in control rats. Total hepatic protein synthesis in vivo, expressed as the proportion of the protein pool that was replaced each day, was increased from 39% +/- 2% per day to 48% +/- 3% per day (P less than .05) by hormone infusion, but synthesis of secretory proteins in perfused liver was not significantly altered. The results suggest that although total hepatic protein synthesis may be increased by catabolic hormones, other mediator(s) are probably responsible for the stimulation of acute-phase protein synthesis in sepsis.

Pentobarbital improves nitrogen retention in sepsis

Pentobarbital therapy has been associated with decreased urinary nitrogen excretion and resting energy expenditure in stressed patients. The metabolic effects of pentobarbital in sepsis were investigated in 29 well-nourished rats who underwent superior vena caval cannulation, cecal ligation, and puncture. Animals were randomly assigned to receive either a continuous infusion of 20 mg/kg/day of pentobarbital combined with parenteral nutrition (n = 13) or parenteral nutrition alone (n = 16). Both groups received isocaloric, isonitrogenous parenteral nutrition postoperatively for 24 hr. Mean nitrogen balance (+/- SEM) was better in the pentobarbital group (+169 +/- 76 mg/kg/day vs -190 +/- 66 mg/kg/day, p less than 0.01). No significant differences between the pentobarbital and control groups were noted for urinary 3-methylhistidine excretion (9 +/- 0.7 micrograms/kg/day vs 11 +/- 0.6 micrograms/kg/day, respectively) or 24 hr survival (77% vs 69%, respectively). Pentobarbital improves nitrogen retention without decreasing urinary 3-methylhistidine excretion in septic rats.

Administration of balanced or BCAA-enriched amino acid solution in septic rats. Effects on protein synthesis in the liver

Total hepatic protein synthesis was measured in vivo with a flooding-dose technique, and the production of total secreted proteins, albumin, complement component C3, and seromucoid fraction was measured in perfused livers of septic rats that received one of three different solutions infused intravenously; Group 1 received 16.4% dextrose; Group 2 received Aminosyn (25% BCAA) in 10.6% dextrose, and Group 3 received Freamine HBC (45% BCAA) in 10.6% dextrose. All solutions were isocaloric, and the amino acid solutions were isonitrogenous. The solutions were administered for 18 or 48 hours after the induction of sepsis. There were no significant differences in mortality rates in the three treatment groups. The negative nitrogen balance seen in the dextrose-infused animals was reversed to the same degree by the two different amino acid solutions. There were no significant differences in hepatic protein synthesis rates in vivo between the three groups of rats. Synthesis rates of secreted proteins in perfused liver were similar in the different treatment groups in the 18-hour experiments, whereas in the 48-hour experiments, synthesis rates of total secreted proteins, C3, and the serumucoid fraction were higher in Group 1 than in Groups 2 and 3. The results suggest that administration of an amino acid solution improves nitrogen balance in sepsis, but that this effect is not caused by stimulated hepatic protein synthesis. The nitrogen-sparing effect during sepsis of a branched chain amino acid (BCAA)-enriched solution does not seem to be superior to that of a balanced amino acid solution.

Methods for studying protein synthesis and degradation in liver and skeletal muscle

Different methods used for measuring protein turnover in liver and skeletal muscle are described, with special emphasis on technical and practical aspects and the advantages and limitations of different techniques. In the first part of the review, the concept of precursor specific radioactivity and its importance for accurate determination of protein synthesis rate is discussed. In the second part, different in vivo techniques for protein turnover measurements are reviewed, including continuous administration of tracer amino acid, flooding dose technique, indirect measurement of protein synthesis, and estimation of protein degradation in vivo. In the third part of the report, in vitro techniques are described, including measurement of protein turnover in incubated liver slices, perfused liver, isolated hepatocytes, incubated isolated muscles or muscle biopsies, and perfused rat hemicorpus. In vivo techniques are preferred when accurate absolute values of protein turnover rates are desired. In vitro techniques offer the advantage of standardized conditions, maintaining strict control of substrate and hormone concentrations, and eliminating complicating interactions with other tissues. For several in vitro techniques, a good correlation has been demonstrated between relative changes in protein turnover in vitro and in vivo in different conditions.

Synthesis of acute-phase proteins in perfused liver following administration of recombinant interleukin 1 alpha to normal or adrenalectomized rats

The purpose of this study was to investigate the effects of recombinant interleukin 1 alpha (rIL-1 alpha) on metabolic rate and synthesis of acute-phase proteins in intact and adrenalectomized rats. Animals were housed in metabolic cages with daily recording of food intake and body weight. Twice daily, for 3 days, the rats were injected intraperitoneally with 5000 LAF U of human rIL-1 alpha, purified from Escherichia coli. Control animals were pair-fed and received corresponding injections with saline. In the morning of the fourth day, resting energy expenditure (REE) was determined by indirect calorimetry, and synthesis of total secreted proteins, albumin, complement component C3, and seromucoid fraction was measured by radioimmunological method using rat-specific antisera and [3H]leucine in livers perfused for 2 hr. Food intake decreased by approximately 30% during rIL-1 alpha administration to intact rats. The decrease in food intake occurred later and was less pronounced in adrenalectomized rats receiving rIL-1 alpha. Growth rate was significantly reduced on the first day of rIL-1 alpha treatment in intact rats, while there was no effect on growth rate in adrenalectomized animals. After rIL-1 alpha administration, REE was increased by 26% in intact rats (P less than 0.001) and by 14% in adrenalectomized rats (N.S.). Increased synthesis rates of total secreted proteins, complement component C3, and seromucoid fraction were observed in livers of intact rats following rIL-1 alpha administration. In adrenalectomized rats, only production of C3 was significantly increased after treatment with rIL-1 alpha. Albumin synthesis rate was not changed in either group following rIL-1 alpha injections.(ABSTRACT TRUNCATED AT 250 WORDS)

Infection in the compromised host

In spite of the development of successive generations of more powerful antibiotics, sepsis remains a common cause of death in the surgical patient. This fact is not surprising, since it is of little importance which organism is causing the infection if the patient's intrinsic antibacterial defenses cannot respond. Realization of the limitations of antibiotic therapy has prompted many investigators to study the immune and inflammatory systems with the ultimate goal of improving host defenses and increasing survival. In this article the local and systemic antibacterial defense systems have been reviewed. It is clear that the immunocompromised state is associated with multiple defects of the humoral and cellular components of both the nonspecific host defense system and the specific immune defense system. However, although consensus has not been reached on either the mediators responsible for the immunocompromised state or the prognostic and clinical significance of many of the described defects, work in this field is progressing rapidly. Nonetheless, knowledge that the host's antibacterial defense systems can be aided by good surgical technique, nutrition, appropriate use of antibiotics, and the sterile care of invasive catheters, lines, and tubes is critical in the prevention of fatal infections in the immunocompromised patient. When more information becomes available concerning the complex interrelations between the various cellular and humoral components of the host defense systems and their mediators and the symbiotic relations between man and his bacterial flora, it should be possible to develop specific strategies for the clinician to use to reduce the risk of infection in the high-risk surgical patient. Thus, in the future it may not be unusual for the clinician to alter the immunologic system of the host by the use of immunomodulators or vaccines to increase the host's resistance to infection. On the other hand, by manipulating the complex interrelations between the host and the indigenous bacterial flora, it may be possible to prevent the development of opportunistic infections originating from the gastrointestinal tract.

Use of scoring systems to assess patients with surgical sepsis

Several scoring systems for patients with severe surgical infections are described, compared, and contrasted. All show a strong relation between a high score and high mortality risk. Each has strengths and weaknesses. All of the systems need further refinement. Future investigations of patients with serious infections should incorporate one of the published severity scores as part of the patient description.

Increased synthesis of secreted hepatic proteins during abdominal sepsis

To study the effect of intraabdominal sepsis on hepatic protein synthesis, male Sprague-Dawley rats underwent celiotomy with either cecal ligation and puncture (CLP) or sham operation. Eight and sixteen hours later total hepatic protein synthesis was measured by flooding dose technique. Specific synthetic rates of structural or secreted hepatic proteins were further studied 16 hr after CLP in an isolated perfused liver model. Total hepatic protein synthesis was significantly elevated at 16 hr (59 +/- 6%/day vs 37 +/- 6%/day, P less than 0.05), but not 8 hr post-CLP. Structural hepatic protein synthesis was unchanged after CLP; however, the synthetic rates of the acute-phase secretory proteins alpha 1-acid glycoprotein, transferrin and complement component C3 were significantly increased 16 hr after CLP. However, the albumin synthetic rate was not increased during sepsis. We conclude that sepsis causes augmentation of hepatic protein synthesis primarily to increase acute-phase proteins for host defense.

Carnitine excretion: a catabolic index of injury

In patients with trauma or sepsis, carnitine is known to be produced to a greater extent; deficient production could impair the energy management that is required in such patients. To clarify the requirements of carnitine after injury, we studied carnitine elimination (in 10 critically ill injured patients) both during fasting and early parenteral nutrition. Increased carnitine (mainly, free) output after injury (9.36 +/- 1.63 mumol/kg p less than 0.02 vs reference) was negatively related to nitrogen balance (p less than 0.05) and positively to 3-methyl-histidine output (p less than 0.01), acting as a market of body mass catabolism. The output of both total and free carnitine progressively decreased (p less than 0.01) throughout the course of total parenteral nutrition. In conclusion, our data definitively suggest that carnitine loss after injury reflects body cell mass wastage and does not necessarily mean an increased need.

Peripheral, visceral and body nitrogen balance of catabolic patients, without and with parenteral nutrition

The effect of major trauma and sepsis on skeletal muscle, central tissue and whole body nitrogen (N) metabolism was investigated in 5 patients before and during TPN (30 kcal, 0.30 g N kg-1 day-1). Fasting 3-methylhistidine (MEH) urinary excretion was elevated (407.9 +/- 67.6 mumol m-2 day-1), muscle and body N balances (NB) were markedly negative (-28.2 +/- 4.6 g m-2 day-1 and -15.7 +/- 3.1 g m-2 day-1), while central tissue NB was positive (13.0 +/- 2.4 g m-2 day-1). TPN effected a reduction in MEH excretion (261.8 +/- 27.5 mmol m-2 day-1 - p less than 0.05) and decreased the release of almost all amino acids from muscle tissue, some of them acting as catabolic markers. Muscle (-7.2 +/- 1.2 g m-2 day-1 - p less than 0.01) as well as body NB (-4.8 +/- 1.4 g m-2 day-1 - p less than 0.01) improved, whilst central tissue NB worsened, even though still positive (3.1 +/- 1.6 g m-2 day-1 - p less than 0.05). Gathering fasting and TPN data MEH excretion was significantly related to both body (r = 0.89) and muscle (r = 0.73) NB, that were highly related to each other (r = 0.93), being muscle always worse than body NB. In conclusion, the anticatabolic activity of TPN is confirmed, although our setting did not achieve muscle NB, it was consistently improved and seems to be the major determinant of body NB, in contrast central NB and central N utilization (46.4% +/- 5.4 vs 15.8% +/- 8.4 - p less than 0.05) worsened.

Interleukin 1 and tumor necrosis factor do not regulate protein balance in skeletal muscle

Recent studies have claimed that interleukin 1-containing preparations increase skeletal protein degradation similar to that seen during infection and inflammation. However, preparations employed have contained other products of activated macrophages, including tumor necrosis factor-alpha. In the present report, we investigated the capability of recombinant-derived murine and human interleukins 1-alpha and 1-beta and human tumor necrosis factor-alpha to affect skeletal protein synthesis and degradation both in vitro and in vivo. Partially purified products of Staphylococcus albus-stimulated human blood monocytes increased skeletal protein degradation both in vivo and in vitro. However, none of the recombinant interleukin 1 nor the human tumor necrosis factor-alpha preparations had any impact on skeletal protein balance. Both recombinant interleukin 1 and tumor necrosis factor-alpha stimulated the production of prostaglandin E2 (PGE2). Furthermore, a polyclonal antibody to human interleukin 1 eliminated the lymphoproliferative response to partially purified monocyte preparations (interleukin 1 activity), but failed to abrogate the increased skeletal protein degradation in vitro. This study demonstrates that although interleukin 1 and tumor necrosis factor-alpha induce a PGE2 response by skeletal muscle in vitro, some macrophage product distinct from either interleukin 1 or tumor necrosis factor-alpha is responsible for the accelerated skeletal protein degradation seen with partially purified human blood monocyte products. Elevated PGE2 levels do not appear to regulate skeletal protein balance in vitro.

The gut origin septic states in blunt multiple trauma (ISS = 40) in the ICU

The association between support elements (ventilator days = Vd, enteral protein = EnP, number of antibiotics per day = AB/d) and the magnitude of the septic state (SSS) and its bacteriologic manifestations (bacti. log) in 66 patients with blunt multiple trauma (mean HTI-ISS = 40) over 1649 days have been studied retrospectively. SSS is measured by summing the standard deviation units of change in the septic direction for the 16 measurements taken every day in the intensive care unit. Increasing Vd is tightly associated with an increasing SSS (r = +0.52), after day 10 an increasing bacti. log (r = +0.21 to +0.32), and an increasing AB/d (r = +0.26) (all p less than 0.001, N = 1615 - 1626). The independent variables that best predicted Vd were delayed operations (DORS), day of rising EnP, and total positive blood cultures (TPC) (adj. R sq. = 0.84, F = 104, dF = 3/59). An increasing AB/d was associated with an increasing SSS (r = +0.38), increasing Vd (r = +0.26), and an increased bacti. log (r = +0.14 to +0.18) (all p less than 0.001, N = 1615). Only an increased EnP was consistently associated with a reduced SSS (r = -0.38) and a reduction in bacti. log (r = -0.10 to -0.21) (all p less than 0.001, N = 1626-1636). The independent variables Vd, EnP, AB/d, and TPC best predicted SSS for all surviving patients (adj. R sq. = 0.42, F = 268, dF = 4/1496). The patients who died of sepsis were not different in terms of bacti. log from those with equal Vd but were distinguished by zero EnP, high AB/d, and persistent ventilatory support. In conclusion, DORS is tightly associated with increased Vd, SSS, AB/d, and zero EnP. If Vd exceeds 10, there is an increasing bacti. log and evidence of infection probably from the gut. This responds only to increased EnP and not to AB/d. Death due to sepsis is not associated with increased bacti. log but with zero EnP and high AB/d and their consequences.

Transport kinetics of amino acids across the resting human leg

Flux rates of amino acids were measured across the leg after an overnight fast in resting human volunteers. A balanced amino acid solution was, after a primed infusion, continuously infused for 2 h at each of three step-wise and increasing rates corresponding to 8.3, 16.7, 33.2 mg N/kg per h that were equivalent to 0.2, 0.4, 0.8 g N/kg per d. Flux of amino acids across the leg was compared with the flux of glucose, glycerol, lactate, free fatty acids, and oxygen. The size of the muscular tissue pool of amino acids was measured. Whole body amino acid oxidation was estimated by means of the continuous infusion of a 14C-labeled mixture of amino acids. Arterial steady state levels were obtained for most amino acids within 30 to 45 min after the primed constant infusion. Leg flux of amino acids switched from a net efflux after an overnight fast to a balanced flux between infusion rates corresponding to 0.2-0.4 g N/kg per d. At 0.8 g N/kg per d essentially all amino acids showed uptake. The infusion of amino acids stimulated leg uptake of glucose and lactate production and decreased FFA release. Oxygen uptake and leg blood flow increased significantly with increased infusion of amino acids. There was significant variability in transport rate among individual amino acids. Branched chain amino acids showed rapid transport and methionine slow transport rate. Only small changes in the muscle tissue concentration of certain amino acids were registered after 6 h of amino acid infusion despite uptake for several hours. When amino acids were infused at a rate corresponding to 0.8 g N/kg per d, the leg uptake of amino acids was 6% and the simultaneous whole body oxidation of infused amino acids was approximately 10%. Net uptake of leucine across the leg per hour was 62% of the muscle pool of free leucine when amino acids were infused at a rate corresponding to 0.4 g N/kg per d. Multiple regression analysis showed that the arterial concentration of an amino acid was the most important factor for uptake, more so than insulin concentration and blood flow. It is concluded that leg exchange of amino acids is large enough to rapidly change the pool size of the amino acids in skeletal muscle, if not counter-regulated by changes in rates of protein synthesis and degradation. Estimates of the capacity for protein synthesis and transfer RNA acceptor sites in muscles agree in order of magnitude with the net uptake of amino acids at high infusion rates of amino acids. Therefore, measurements of the balance of tyrosine, phenylalanine, and particularly methionine at steady state may reflect net balance of proteins across skeletal muscles even in short-time experiments.

Studies on the possible role of thyroid hormone in altered muscle protein turnover during sepsis

Five days after thyroidectomy (Tx) or sham-Tx in young male Sprague-Dawley rats, sepsis was induced by cecal ligation and puncture (CLP). Control animals underwent laparotomy and manipulation of the cecum without ligation or puncture. Sixteen hours after CLP or laparotomy, protein synthesis and degradation were measured in incubated extensor digitorum longus (EDL) and soleus (SOL) muscles by determining rate of 14C-phenylalanine incorporation into protein and tyrosine release into incubation medium, respectively. Triiodothyronine (T3) was measured in serum and muscle tissue. Protein synthesis was reduced by 39% and 22% in EDL and SOL, respectively, 16 hours after CLP in sham-Tx rats. The response to sepsis of protein synthesis was abolished in Tx rats. Protein breakdown was increased by 113% and 68% in EDL and SOL, respectively, 16 hours after CLP in sham-Tx animals. The increase in muscle proteolysis during sepsis was blunted in hypothyroid animals and was 42% and 49% in EDL and SOL, respectively. T3 in serum was reduced by sepsis, both in Tx and sham-Tx rats. T3 in muscle, however, was maintained or increased during sepsis. Abolished or blunted response of muscle protein turnover after CLP in hypothyroid animals may reflect a role of thyroid hormones in altered muscle protein metabolism during sepsis. Reduced serum levels of T3, but maintained or increased muscle concentrations of the hormone, suggests that increased T3 uptake by muscle may be one mechanism of low T3 syndrome in sepsis, further supporting the concept of a role for thyroid hormone in metabolic alterations in muscle during sepsis.

Survival from hepatic transplantation. Relationship of protein synthesis to histological abnormalities in patient selection and postoperative management

Forty-one patients, all in end stage hepatic failure, underwent 46 liver transplantations with a long-term survival rate of 63%. Six patients died of uncontrollable bleeding due to primary graft malfunction at or immediately after operation. Nine died early or late with overwhelming infection. In addition to clinical assessment, needle liver biopsy, central plasma clearance rate of amino acids (CPCR-AA), and routine "liver function tests" were employed to aid in selection of patients for transplantation and for guidance in postoperative management. Although liver biopsies usually afforded an exact diagnosis, neither they nor the routine liver function tests quantitated the extent to which hepatocyte function was impaired. CPCR-AA, which measures the rate of amino acid uptake by the liver and other central tissues for oxidation, gluconeogenesis, and protein synthesis was 91 +/- 9 ml/M2/min in the preoperative transplant group. This compares with a value of 97 +/- 16 in a previously studied series of cirrhotics who died following other forms of surgery and a CPCR-AA of 220 +/- 26 ml/m2/min in those who survived. In addition, the preoperative CPCR-AA was found to correlate with the in vitro hepatic protein synthetic rate of slices from the resected recipient liver (r = 0.72, p less than 0.02). After operation, serial hepatic needle biopsies were classified by histology into four grades of injury, ranging from normal liver transplant (Grade I) to mild hypoxic or rejection injury (Grade II), viral hepatitis (Grade III), and severe hypoxic or rejection injury (Grade IV). Significant relationships of the histological grades to ultimate mortality, CPCR-AA, and prothrombin times were found. CPCR-AA and prothrombin time correlate inversely (r = 0.57, p less than 0.001), further demonstrating the relationship of CPCR-AA to protein synthesis of clotting factors. These patterns of posttransplant response were delineated by serial CPCR-AA values. "Early" responders had values over 290 ml/M2/min and all survived. Twelve patients with delayed response were characterized by values of 150 +/- 12, rising to over 200 ml/M2/min after 2 weeks. Two who failed to increase CPCR-AA died. In six "poor" responders, CPCR-AA with Grade IV injury remained below 110 ml/M2/min. All died except for one whose CPCR-AA subsequently rose following retransplantation. It is concluded that percutaneous hepatic needle biopsies and CPCR-AA measurements in combination are of proven value, not only in understanding the nature of injury and functional impairment of the liver, but are also important as guides to selection of patients and for their posttransplant management.