Lerner B. Hinshaw

Lethal staphylococcus aureus-induced shock in primates : prevention of death with anti-TNF antibody

A successful experimental treatment for gram-positive sepsis to our knowledge has not been achieved. The objectives of this study were to develop a nonhuman primate model of lethal gram-positive sepsis employing the micro-organism Staphylococcus aureus and to determine the efficacy of treatment using monoclonal antibody (MAb) to tumor necrosis factor alpha (TNF). The antibody was administered intravenously, 15 mg/kg, 30 minutes after the beginning of a 2-hour infusion of S. aureus, 4 x 10(10) colony forming units/kilogram. The baboons infused with S. aureus demonstrated the release of the cytokines TNF and interleukin-6 (IL-6), but endotoxin was not observed in the plasma at any time. Treatment with antibody to TNF abolished the rise in serum TNF levels and reduced the increased levels of IL-6. Treatment with MAb to TNF prevented multiple organ failure and achieved permanent (> 7 day) survival of all baboons.

Continuous arteriovenous hemofiltration therapy for Staphylococcus aureus-induced septicemia in immature swine.

Objectives.The goals of this study were: a) to evaluate the efficacy of controlled, continuous arteriovenous hemofiltration in improving morbidity and mortality rates in an immature swine model of Staphylococcus aureus-induced septicemia; b) to determine if ultrafiltrate from septic animals contained mediators that produce pathophysiologic changes observed in untreated S. aureus septic pigs. Design.Prospective, randomized, controlled study with age-matched controls. Setting.U.S. Department of Agriculture-licensed biomedical research facility. Subjects.Sixty-five weaned Poland-China swine (4 to 6 wks of age; 5 to 10 kg). Interventions: Part 1.Animals received a lethal dose of live S. aureus (8 ± 109 colony-forming units/kg) over 1 hr. The three treatment groups included: hemofiltration group 1 (eight filtered, eight nonfiltered animals), plasma filtration fraction = 5.5%; hemofiltration group 2 (six filtered, six nonfiltered animals), filtration fraction = 16.6%; and hemofiltration group 3 (six filtered, six nonfiltered animals), filtration fraction = 33.4%. A control, nonseptic group of animals (n = 4) was filtered to obtain “clean” ultra-filtrate (hemofiltration group 4). Part 2: Sterile ultrafiltrate concentrate batches obtained from each group of filtered, septic animals were concentrated and infused into healthy, nonseptic pigs (reinfusion groups 1 through 3). Measurements and Main Results.Physiologic, biochemical, and hematologic variables were measured in all animals every 1 to 3 hrs. Overall length of survival was also recorded. In hemofiltration groups 1 through 3, filtered animals survived significantly longer than matched, nonfiltered (sham-filtered) animals. Increments in survival time increased directly with filtration fraction. Ultrafiltrate concentrate from septic pigs produced death (LD41) and disease similar to those rates observed in untreated S. aureus-septic pigs. Infusion of clean ultrafiltrate concentrate produced no response. Conclusions.Continuous arteriovenous hemofiltration significantly improved survival rates in swine with S. aureus-induced sepsis. Resultant ultrafiltrate concentrate contained mediators responsible for some pathophysiologic responses observed in this animal model. (Crit Care Med 1993; 21:914–924)

Mechanisms of impaired cardiac function by vasopressin.

The mechanisms by which elevated levels of vasopressin (ADH) in man and animals cause serious myocardial dysfunction, evidenced by arrhythmias, reduction in cardiac output and coronary blood flow, are not settled. Experiments were conducted in 16 isolated working left ventricles to examine their metabolic and hemodynamic responses to the infusion of vasopressin and the combination of vasopressin and epinephrine. Contractile performance was evaluated by analysis of positive dP/dt, contractile element velocities, and ventricular work-curves using stroke work/end-diastolic pressure. Relaxation parameters, including negative dP/dt and the early diastolic relaxation time constant, were also studied. Coronary blood flow was reduced 22% or less by vasopressin while cardiac output was maintained at a constant level. Myocardial oxygen consumption, lactate and potassium balances were determined from arterial and coronary sinus concentrations. Vasopressin produced myocardial dysfunction indicated by decrements in contractile and relaxation indices, without evidence of global ischemia. Epinephrine restored the mechanical performance to normal without significant change in coronary blood flow, myocardial oxygen consumption, or lactate and potassium balance.

Effectiveness of Steroid/Antibiotic Treatment in Primates Administered LD100 Escherichia coli.

Early aggressive therapy with maintenance infusions of methylprednisolone sodium succinate and gentamicin sulfate significantly increases the probability for survival of baboons given LD100 Escherichia coli. The present study was designed to determine if baboons would recover when initiation of treatment was delayed until they had sustained E. coli-induced systemic hypotension for a period of approximately three hours. Sixteen adult baboons were each administered a two-hour infusion of LD100 E. coli. All eight untreated animals died within 42 hours. Five of the eight baboons treated after approximately three hours of hypotension with methylprednisolone sodium succinate and gentamicin sulfate survived. Treated animals had significantly higher blood glucose and insulin levels and lower blood urea nitrogen concentrations than baboons receiving E. coli alone. E. coli blood concentrations were lower in the treated than in the untreated baboon group by the sixth hour (less than 0.02). Heart rates increased in all animals but were not as high in the treated baboons. Both groups experienced similar decreases in mean systemic arterial pressure, PCO2, base excess, leukocyte, lymphocyte, and platelet concentrations, and increases in creatinine and lactate concentrations. Data from the present study indicate that the probability of recovery from shock is significantly increased even when initiation of steroid/antibiotic therapy is postponed until baboons have experienced sustained systemic hypotension.

Preclinical review of anti-tumor necrosis factor monoclonal antibodies

Objective: To review the preclinical evidence for the role of tumor necrosis factor (TNF) in the pathogenesis of septic shock and to assess the preclinical efficacy of anti-TNF therapies for this clinical problem. Data Sources: The international English language literature from 1986 to the present formed the basis for this review. MEDLINE was used to identify pertinent in vitro and animal studies pertaining to the pathobiology of TNF and the use of anti-TNF therapies, with special emphasis on antibody approaches. Study Selection: Those studies that focused on the mechanisms of action of TNF, its role in the inflammatory cascade, and the potential uses of anti-TNF therapies were emphasized. Investigations that described animal and human results served as the primary database. Data Extraction: Animal studies were selected based on the relevance of the model to the pathogenesis of the human clinical sepsis syndrome. Where they provided supportive evidence, patient studies were selected on the basis of study design. Data Synthesis: The administration of anti-TNF antibodies in baboons, monkeys, and other species that were administered lethal doses of bacteria or endotoxin suggest that this approach may limit organ damage and decrease the mortality rate caused by the septic shock syndrome. Therapy with anti-TNF monoclonal antibodies is reviewed. Conclusions: Bacterial challenge induces the release of TNF (among other mediators), which exerts both physiologic and toxic effects that may ultimately lead to organ dysfunction and death. New anti-TNF therapies such as anti-TNF antibodies appear to attenuate the injurious effects of TNF and promote survival in otherwise lethal septic shock animal models, suggesting a similar benefit might be obtained in the treatment of human septic shock. (Crit Care Med 1993; 21:S441-S446)

Plasma endotoxin concentration in healthy primates and during E. coli-induced shock.

The normal range for circulating plasma endotoxin concentration was determined in 62 healthy primates (vervet monkeys, Cerecopithecus aethiops) by the chromogenic substrate modification of the Limulus amoebocyte lysate test, and found to have a mean of 0.076 +/- 0.004 ng/ml (range 0.000 to 0.0127). Four anesthetized primates received an LD100 iv infusion of Escherichia coli over one hour. Plasma concentrations of endotoxin (lipopolysaccharide, LPS) and anti-LPS IgG, and viable E. coli colonies in circulating whole blood samples were determined at specified intervals. Plasma antiendotoxin IgG concentration was determined by an enzyme-linked immuno-absorbent assay, and viable bacterial counts were assayed by standard plate count techniques. LPS concentration increased during E. coli infusion to a mean of 1.13 +/- 0.068 ng/ml (p less than .001) with a concomitant decrease in the concentration of anti-LPS IgG to 59 +/- 5% of control values (p less than .005). Viable circulating E. coli colonies increased during the infusion to a maximum of 425 X 10(6) cfu/ml 10 min after the completion of the infusion, but fell precipitously 20 min later to 10.1 X 10(6) cfu/ml. When each animal succumbed, their respective plasma LPS concentrations were still raised, whereas no viable circulating E. coli colonies were present at a dilution of 10(2). Elevated plasma LPS could prove to be a significant circulating pathogen during Gram-negative bacterial shock and supports the possible association between plasma LPS and morbidity, and mortality in septic shock.

PERITONITIS IN THE BABOON: A PRIMATE MODEL WHICH SIMULATES HUMAN SEPSIS

The physiological, hemostatic, and immunological responses of 12 chronically instrumented conscious baboons with sepsis due to Escherichia coli peritonitis were compared with that of similarly instrumented controls. Chronic indwelling cannulae were placed in the aorta and pulmonary artery to monitor pressure, cardiac output, and obtain blood samples. At t = 0 a sterile or E. coli-laden fibrin clot containing 1.9-6.7 x 10(11) CFU/kg was introduced into the peritoneal cavity. The control animals were group 1 (n = 3). The animals with peritonitis were divided into three groups depending on their clinical response. Group 2 animals (n = 3) were clinically well at the time of sacrifice (day 14), group 3 (n = 4) survived but were obviously sick on day 14, and group 4 (n = 5) died of sepsis. Implantation of a sterile fibrin clot was well tolerated with little hemodynamic change and a transient minimal inflammatory response in group 1. Implantation of an E. coli-containing clot elicited a hyperdynamic cardiovascular response and evoked a marked inflammatory reaction and a disseminated intravascular coagulopathy. Five of 12 (42%) E. coli animals died from sepsis. In general, the physiological, hemostatic, and immunological disturbances tended to be greatest in these animals. Autopsy revealed residual peritoneal inflammation and varying degrees of inflammation in the lungs, adrenal, spleen, liver, and kidneys in all the animals that received E. coli with the inflammatory infiltrate increasing in severity from group 2 through group 4. Tissue necrosis was observed only in the latter group. We conclude that the cardiovascular, hemostatic, and immunological responses of baboons with sepsis due to E. coli peritonitis exhibit a variable course that resembles the clinical manifestations of gram-negative sepsis in humans.

Development of a strain of spontaneously hypertensive rabbits.

Summary We have bred and raised a colony of spontaneously hypertensive rabbits. It is concluded that there is a hereditary factor in hypertension and that the characteristic changes of blood pressure in spontaneously hypertensive animals are similar to the normal rabbit population. These findings would indicate that spontaneously hypertensive rabbits have characteristics in common with man in his development of essential hypertension.

Further Observations on Development of a Colony of Spontaneously Hypertensive Rabbits.

Summary (1) Two new generations of spontaneously hypertensive rabbits showed about the same incidence of elevated systolic pressures as their progenitors at 6 and 8 months of age, 58 and 71% respectively. At 4 months, mean systolic pressure in the 2 new generations was significantly higher than in their progenitors at same age. (2) Male rabbits of the new generations in all 3 age groups had higher mean systolic values than females and than their male or female progenitors. (3) Direct arterial pressure measurements in a group of spontaneously hypertensive rabbits showed average diastolic pressure to be 10 mm Hg higher and systolic pressure 30 mm Hg higher than in normo tensive stock rabbits. (4) Direct arterial recordings from unanesthetized animals demonstrated spontaneously hypertensive rabbits to have larger Traube-Hering waves than normotensives. (5) The presence of interstitial nephritis in normotensive as well as spontaneously hypertensive animals precludes it as an etiological factor in the development of spontaneous hypertension.

Comparison of the capacity of rhTNF-alpha and Escherichia coli to induce procoagulant activity by baboon mononuclear cells in vivo and in vitro.

The procoagulant activity of mononuclear cells (MNCs) may play an important role in the disseminated intravascular coagulation seen in septic shock. This study compares the capacity of Escherichia coli (E. coli) and recombinant human TNF-α (rhTNF-α) to induce procoagulant activity by baboon MNCs. In vivo studies showed that MNC procoagulant activity was significantly increased at T + 120 min after LD100 E. coli infusion into baboons. Most of this procoagulant activity was attributable to tissue factor. In contrast, a bolus infusion of rhTNF-α (150 μg/kg) and a monoclonal antibody to activated protein C (2 mg/kg) did not induce any increase of MNC procoagulant activity at T + 120 min even though the plasma TNF-a level was 10 times higher than that seen following infusion of E. coli. In vitro studies showed that E. coli at concentrations comparable to that observed in the in vivo study and LPS at a concentration of 2.5 ng/mL induced more intense tissue factor expression by both human and baboon monocytes than rhTNF-α in the concentrations ranging from 10 to 1,000 ng/mL. These results suggest that TNF-α alone is not sufficient to induce noticeable MNC procoagulant activity, at least, in the early stage of this septic shock model.