Large animal models: baboons for trauma, shock, and sepsis studies

Physiology of Human Hemorrhage and Compensation

Hemorrhage is a leading cause of death following traumatic injuries in the United States. Much of the previous work in assessing the physiology and pathophysiology underlying blood loss has focused on descriptive measures of hemodynamic responses such as blood pressure, cardiac output, stroke volume, heart rate, and vascular resistance as indicators of changes in organ perfusion. More recent work has shifted the focus toward understanding mechanisms of compensation for reduced systemic delivery and cellular utilization of oxygen as a more comprehensive approach to understanding the complex physiologic changes that occur following and during blood loss. In this article, we begin with applying dimensional analysis for comparison of animal models, and progress to descriptions of various physiological consequences of hemorrhage. We then introduce the complementary side of compensation by detailing the complexity and integration of various compensatory mechanisms that are activated from the initiation of hemorrhage and serve to maintain adequate vital organ perfusion and hemodynamic stability in the scenario of reduced systemic delivery of oxygen until the onset of hemodynamic decompensation. New data are introduced that challenge legacy concepts related to mechanisms that underlie baroreflex functions and provide novel insights into the measurement of the integrated response of compensation to central hypovolemia known as the compensatory reserve. The impact of demographic and environmental factors on tolerance to hemorrhage is also reviewed. Finally, we describe how understanding the physiology of compensation can be translated to applications for early assessment of the clinical status and accurate triage of hypovolemic and hypotensive patients. © 2021 American Physiological Society. Compr Physiol 11:1531-1574, 2021.

Indices of muscle and liver dysfunction after surviving hemorrhage and prolonged hypotension

BACKGROUND

This study determined the long-term effects of prolonged hypotension (PH) on liver, muscle, and kidney dysfunction. The hypothesis was that longer duration of PH after hemorrhage will result in greater organ dysfunction.

METHODS

Baboons were sedated and hemorrhaged (30% blood volume). Systolic blood pressure greater than 80 mm Hg was maintained for 1 hour (1 hr-PH; n = 5), 2 hours (2 hr-PH; n = 5), or 3 hours (3 hr-PH; n = 5). After PH, hemorrhage volume was replaced. Animals were recovered and monitored for 21 days. Control animals were hemorrhaged and immediately resuscitated (0 hr-PH, n = 3). Data are Mean ± Standard Deviation, and analyzed by 2-way repeated measures ANOVA and Holm-Sidak test.

RESULTS

Hemorrhage resulted in mild hypotension. Minimal resuscitation was required during the hypotensive phase, and survival rate was 100%. Significant increases (p < 0.001) in alanine aminotransferase, aspartate aminotransferase, creatine phosphokinase, and lactate dehydrogenase occurred on Day 1 after PH, and were significantly greater (p < 0.001) in the 2 hr- and 3 hr-PH groups than the 0 hr-PH group. Maximum alanine aminotransferase levels (U/L) were 140 ± 56 (0 hr-PH), 170 ± 130 (1 hr-PH), 322 ± 241 (2 hr-PH), and 387 ± 167 (3 hr-PH). Maximum aspartate aminotransferase levels (U/L) were 218 ± 44 (0 hr-PH), 354 ± 219 (1 hr-PH), 515 ± 424 (2 hr-PH), and 711 ± 278 (3 hr-PH). Maximum creatine phosphokinase values (U/L) were 7834 ± 3681 (0 hr-PH), 24336 ± 22268 (1 hr-PH), 50494 ± 67653 (2 hr-PH), and 59857 ± 32408 (3 hr-PH). Maximum lactic acid dehydrogenase values (U/L) were 890 ± 396 (0 hr-PH), 2055 ± 1520 (1 hr-PH), 3992 ± 4895 (2 hr-PH), and 4771 ± 1884 (3 hr-PH). Plasma creatinine and blood urea nitrogen were unaffected by PH (p > 0.10).

CONCLUSION

These results indicate that PH up to 3 hours in duration results in transient liver and muscle dysfunction that was most severe after 2 hr-PH and 3 hr-PH. Prolonged hypotension produced minimal effects on the kidney.

LEVEL OF EVIDENCE

Basic science research, Level of evidence not required for basic science research.

Variability of mitochondrial respiration in relation to sepsis-induced multiple organ dysfunction

Ample experimental evidence suggests that sepsis could interfere with any mitochondrial function; however, the true role of mitochondrial dysfunction in the pathogenesis of sepsis-induced multiple organ dysfunction is still a matter of controversy. This review is primarily focused on mitochondrial oxygen consumption in various animal models of sepsis in relation to human disease and potential sources of variability in experimental results documenting decrease, increase or no change in mitochondrial respiration in various organs and species. To date, at least three possible explanations of sepsis-associated dysfunction of the mitochondrial respiratory system and consequently impaired energy production have been suggested: 1. Mitochondrial dysfunction is secondary to tissue hypoxia. 2. Mitochondria are challenged by various toxins or mediators of inflammation that impair oxygen utilization (cytopathic hypoxia). 3. Compromised mitochondrial respiration could be an active measure of survival strategy resembling stunning or hibernation. To reveal the true role of mitochondria in sepsis, sources of variability of experimental results based on animal species, models of sepsis, organs studied, or analytical approaches should be identified and minimized by the use of appropriate experimental models resembling human sepsis, wider use of larger animal species in preclinical studies, more detailed mapping of interspecies differences and organ-specific features of oxygen utilization in addition to use of complex and standardized protocols evaluating mitochondrial respiration.

Nonhuman primate species as models of human bacterial sepsis

Sepsis involves a disordered host response to systemic infection leading to high morbidity and mortality. Despite intense research, targeted sepsis therapies beyond antibiotics have remained elusive. The cornerstone of sepsis research is the development of animal models to mimic human bacterial infections and test novel pharmacologic targets. Nonhuman primates (NHPs) have served as an attractive, but expensive, animal to model human bacterial infections due to their nearly identical cardiopulmonary anatomy and physiology, as well as host response to infection. Several NHP species have provided substantial insight into sepsis-mediated inflammation, endothelial dysfunction, acute lung injury, and multi-organ failure. The use of NHPs has usually focused on translating therapies from early preclinical models to human clinical trials. However, despite successful sepsis interventions in NHP models, there are still no FDA-approved sepsis therapies. This review highlights major NHP models of bacterial sepsis and their relevance to clinical medicine.

Treatment for bacterial sepsis remains limited beyond the use of antibiotics. Lingye Chen, Karen Welty-Wolf, and Bryan Kraft review nonhuman primate models of sepsis and highlight their advantages and limitations compared to other preclinical models.

Inflammatory Profile in Response to Uncontrolled Hemorrhage in a Non-Human Primate (Rhesus Macaque) Model

BACKGROUND

Uncontrolled hemorrhage (UH), the leading cause of potentially survivable combat-related death, elicits a deleterious inflammatory response. Our group previously reported an increased secretion of pro-inflammatory cytokines in a novel non-human primate model of UH; however, to better understand the molecular profile of the inflammatory response to UH, we performed a comprehensive evaluation of inflammation at the proteomic and transcriptomic level.

METHODS

Anesthetized rhesus macaques (n = 8) underwent UH by 60% left lobe hepatectomy T = 0 min. At T = 5 min, animals received 11 mL of 5% albumin followed by normal saline infusion to a total resuscitation volume of 20 mL/kg by T = 120 min. Blood (T = 0, 5, 20, 120, 480 min) was collected for qPCR and multiplex cytokine quantification. Results from each non-human primate (NHP) per time-point are shown. Statistical analysis by one-way ANOVA with repeated measures, P <0.05 was considered significant.

RESULTS

Luminex analysis in serum revealed significant up-regulation compared with baseline of 8 cytokines/chemokines starting T = 120 min postinjury and significant down-regulation of 4 cytokines/chemokines as early as T = 20 min postinjury. Gene expression analysis in white blood cells uncovered 10 genes that were up-regulated greater than 3-fold compared with baseline and 29 genes that were down-regulated greater than 3-fold.

CONCLUSION

The present study confirms the presence of systemic inflammation after UH at the proteomic and transcriptomic level providing insight into the inflammatory mediators that are involved as well as their kinetics following UH. The data demonstrates that NHP hemorrhage models may be suitable for evaluating therapeutics to control inflammation following hemorrhage.

PET Imaging in Psychoneuroimmunology Research

Positron-emission tomography (PET) imaging is a valuable research tool that enables in vivo quantification of molecular targets in the brain or of a physiologic process. PET imaging can be combined with various experimental and clinical model systems that are commonly used in psychoneuroimmunology research. As PET imaging can be used in animals and humans, promising results can therefore often be translated from an animal model to human disease.

Nonhuman primate model of polytraumatic hemorrhagic shock recapitulates early platelet dysfunction observed following severe injury in humans

BACKGROUND

Platelet dysfunction has been described as an early component of trauma-induced coagulopathy. The platelet component of trauma-induced coagulopathy remains to be fully elucidated and translatable animal models are required to facilitate mechanistic investigations. We sought to determine if the early platelet dysfunction described in trauma patients could be recapitulated in a nonhuman primate model of polytraumatic hemorrhagic shock.

METHODS

Twenty-four male rhesus macaques weighting 7 to 14 kg were subjected to 60 minutes (min) of severe pressure-targeted controlled hemorrhagic shock (HS) with and without other injuries. After 60 min, resuscitation with 0.9% NaCl and whole blood was initiated. Platelet counts and platelet aggregation assays were performed at baseline (BSLN), end of shock (EOS; T = 60 min), end of resuscitation (EOR; T = 180 min), and T = 360 min on overall cohort. Results are reported as mean ± standard deviation (SD) or median (interquartile range). Statistical analysis was conducted using Spearmen correlation, one-way analysis of variance, two-way repeated-measures analysis of variance, paired t-test or Wilcoxon nonparametric test, with p < 0.05 considered significant.

RESULTS

Platelet count in all injury cohorts decreased over time, but no animals developed thrombocytopenia. Correlations were observed between platelet aggregation and platelet count for all agonists: adenosine diphosphate, thrombin recognition-activating peptide-6, collagen, and arachidonic acid. Overall, compared to BSLN, platelet aggregation decreased for all agonist at EOS, EOR, and T = 360 min. When normalized to platelet count, platelet aggregation in response to agonist thrombin recognition-activating peptide-6 demonstrated no change from BSLN at subsequent time points. Aggregation to adenosine diphosphate was significantly less at EOR but not EOS or T = 360 min compared to BSLN. Platelet aggregation to collagen and arachidonic acid was not significantly different at EOS compared to BSLN but was significantly less at EOR and T = 360 min.

CONCLUSION

Nonhuman primates manifest early platelet dysfunction in response to polytraumatic hemorrhagic shock, consistent with that reported in severely injured human patients. Nonhuman primate models potentially are translationally valuable for understanding the mechanisms and pathophysiology of trauma-induced platelet dysfunction.

OBESITY AND CRITICAL ILLNESS: INSIGHTS FROM ANIMAL MODELS

Critical illness is a major cause of morbidity and mortality around the world. While obesity is often detrimental in the context of trauma, it is paradoxically associated with improved outcomes in some septic patients. The reasons for these disparate outcomes are not well understood. A number of animal models have been used to study the obese response to various forms of critical illness. Just as there have been many animal models that have attempted to mimic clinical conditions, there are many clinical scenarios that can occur in the highly heterogeneous critically ill patient population that occupies hospitals and intensive care units. This poses a formidable challenge for clinicians and researchers attempting to understand the mechanisms of disease and develop appropriate therapies and treatment algorithms for specific subsets of patients, including the obese. The development of new, and the modification of existing animal models, is important in order to bring effective treatments to a wide range of patients. Not only do experimental variables need to be matched as closely as possible to clinical scenarios, but animal models with pre-existing comorbid conditions need to be studied. This review briefly summarizes animal models of hemorrhage, blunt trauma, traumatic brain injury, and sepsis. It also discusses what has been learned through the use of obese models to study the pathophysiology of critical illness in light of what has been demonstrated in the clinical literature.

The application of PET imaging in psychoneuroimmunology research

Positron emission tomography (PET) imaging is a research tool that allows in vivo measurements of brain metabolism and specific target molecules. PET imaging can be used to measure these brain variables in a variety of species, including human and non-human primates, and rodents. PET imaging can therefore be combined with various experimental and clinical model systems that are commonly used in psychoneuroimmunology research.

Endotoxin-induced systemic inflammation activates microglia: [¹¹C]PBR28 positron emission tomography in nonhuman primates

Microglia play an essential role in many brain diseases. Microglia are activated by local tissue damage or inflammation, but systemic inflammation can also activate microglia. An important clinical question is whether the effects of systemic inflammation on microglia mediate the deleterious effects of systemic inflammation in diseases such as Alzheimer's dementia, multiple sclerosis, and stroke. Positron Emission Tomography (PET) imaging with ligands that bind to Translocator Protein (TSPO) can be used to detect activated microglia. The aim of this study was to evaluate whether the effect of systemic inflammation on microglia could be measured with PET imaging in nonhuman primates, using the TSPO ligand [(11)C]PBR28.

METHODS

Six female baboons (Papio anubis) were scanned before and at 1h and/or 4h and/or 22 h after intravenous administration of E. coli lipopolysaccharide (LPS; 0.1mg/kg), which induces systemic inflammation. Regional time-activity data from regions of interest (ROIs) were fitted to the two-tissue compartmental model, using the metabolite-corrected arterial plasma curve as input function. Total volume of distribution (V(T)) of [(11)C]PBR28 was used as a measure of total ligand binding. The primary outcome was change in V(T) from baseline. Serum levels of tumor necrosis factor alpha (TNFα), interleukin-1 beta (IL-1β), interleukin-6 (IL-6), and interleukin-8 (IL-8) were used to assess correlations between systemic inflammation and microglial activation. In one baboon, immunohistochemistry was used to identify cells expressing TSPO.

RESULTS

LPS administration increased [(11)C]PBR28 binding (F(3,6)=5.1, p=.043) with a 29 ± 16% increase at 1h (n=4) and a 62 ± 34% increase at 4h (n=3) post-LPS. There was a positive correlation between serum IL-1β and IL-6 levels and the increase in [(11)C]PBR28 binding. TSPO immunoreactivity occurred almost exclusively in microglia and rarely in astrocytes.

CONCLUSION

In the nonhuman-primate brain, LPS-induced systemic inflammation produces a robust increase in the level of TSPO that is readily detected with [(11)C]PBR28 PET. The effect of LPS on [(11)C]PBR28 binding is likely mediated by inflammatory cytokines. Activation of microglia may be a mechanism through which systemic inflammatory processes influence the course of diseases such as Alzheimer's, multiple sclerosis, and possibly depression.

Similarities in thromboelastometric (ROTEM®) findings between humans and baboons

INTRODUCTION

Interest in visco-elastic testing in different clinical scenarios has increased but few data are available on thromboelastometric findings in primates.

MATERIALS AND METHODS

Blood cell count (hemoglobin, hematocrit, platelet count), coagulation parameters (prothrombin time, International Normalized Ratio, fibrinogen), and ROTEM® (Tem International GmbH, Munich, Germany) variables were analyzed using blood from 25 anesthetized male baboons and 21 non-anesthetized healthy volunteers. The platelet component of the clot was calculated as the difference in maximum clot elasticity (MCE) between the whole blood clot (EXTEM test) and the fibrin-based clot (FIBTEM test). In subgroups of each species, 10 μg abciximab was added to the regular FIBTEM reagent (cytochalasin D) for additional platelet inhibition.

RESULTS

Blood cell count was comparable between humans and primates. Both fibrinogen concentration (p<0.0001) and maximum clot firmness (MCF) in FIBTEM assays were significantly lower in baboons (p>0.0001, and p=0.006, respectively). PT, INR, and clotting time in NATEM assays were significantly prolonged in humans compared with baboons. MCF in NATEM, EXTEM and INTEM assays was not different between baboons and humans. Clot lysis in NATEM, EXTEM and INTEM assays was significantly higher in humans (p<0.0001). In contrast FIBTEM clot lysis was significantly higher in baboons (p=0.01). Addition of abciximab into the FIBTEM assay resulted in a significant reduction in MCF and MCE (p<0.001) and, consequently, the platelet component increased similar in both humans and baboons (p<0.001).

CONCLUSION

Activated ROTEM® tests revealed broad similarities between humans and baboons. ROTEM® assays developed for use in humans can also be used in baboons.

Advances in assay of complement function and activation

The main function of the complement system is pattern recognition of danger. Typical exogenous danger signals are pathogen associated molecular patterns inducing a protective inflammatory response. Other examples are exposure to foreign surfaces of biomedical materials including nanoparticles, which principally induce the same inflammatory response. If a surface is "foreign" to the host, it induces complement activation. Development of monoclonal antibodies to neoepitopes on complement activation products introduced an entirely new set of methods for assay of complement activation. Activation of complement by a surface occurs by impairment of the fine balance of the control system, e.g. by preferred binding of factor B at the expense of factor H. Sensitive methods to detect complement activation on surfaces and in the fluid phase are a prerequisite for investigation of the biocompatibility of artificial materials. This information can be used to develop new materials with enhanced biocompatibility. Here we review available methods to study human and animal complement function and activation in vitro and in vivo.

Experimental trauma models: an update

Treatment of polytrauma patients remains a medical as well as socioeconomic challenge. Although diagnostics and therapy improved during the last decades, multiple injuries are still the major cause of fatalities in patients below 45 years of age. Organ dysfunction and organ failure are major complications in patients with major injuries and contribute to mortality during the clinical course. Profound understanding of the systemic pathophysiological response is crucial for innovative therapeutic approaches. Therefore, experimental studies in various animal models are necessary. This review is aimed at providing detailed information of common trauma models in small as well as in large animals.

Experimentally approaching the ICU: monitoring outcome-based responses in the two-hit mouse model of posttraumatic sepsis

To simulate and monitor the evolution of posttraumatic sepsis in mice, we combined a two-hit model of trauma/hemorrhage (TH) followed by polymicrobial sepsis with repetitive blood sampling. Anesthetized mice underwent femur fracture/sublethal hemorrhage and cecal ligation and puncture (CLP) 48 h later. To monitor outcome-dependent changes in circulating cells/biomarkers, mice were sampled daily (facial vein) for 7 days and retrospectively divided into either dead (DIE) or surviving (SUR) by post-CLP day 7. Prior to CLP, AST was 3-fold higher in DIE, while all other post-TH changes were similar between groups. There was a significant post-CLP intergroup separation. In SUR, RBC and Hb were lower, platelets and neutrophils higher, and lymphocytes mixed compared to DIE. In DIE, all organ function markers except glucose (decrease) were few folds higher compared to SUR. In summary, the combination of daily monitoring with an adequate two-hit model simulates the ICU setting, allows insight into outcome-based responses, and can identify biomarkers indicative of death in the acute posttraumatic sepsis in mice.

Intramedullary nailing as a 'second hit' phenomenon in experimental research: lessons learned and future directions

BACKGROUND

The 'second hit' phenomenon is based on the fundamental concept that sequential insults, which are individually innocuous, can lead to overwhelming physiologic reactions. This response can be expressed in several organic systems and can be examined by measurement of several parameters.

PURPOSES

The purpose of this study was to evaluate the incidence of systemic effects of intramedullary nailing and the role of concurrent head and thoracic injuries as they have been recorded in vivo. We also wanted to determine what would be the optimal animal model for future research and what variables should be investigated.

METHODS

We reviewed the available literature of animal studies that used surgery, and particularly nailing, as a second hit. The reviewed studies were retrieved through an electronic search of the MEDLINE database. We analyzed the methods of creating the first and second hits (nailing), the characteristics of the animal models, the variables examined, and the pathophysiologic responses, which appeared after the second hit.

RESULTS

Second hit reamed intramedullary nailing was found to provoke consumption of coagulation factors, whereas the effect of unreamed nailing on coagulation factors appears inconsistent. Hemodynamic factors were affected only transiently by the second hit, whereas the pulmonary function was affected only when the first hit included lung injury.

CONCLUSIONS

Thoracic and head injuries predispose to an aggravated second hit. Primate animal models are considered to be closer to clinical reality and should be preferred for future studies. Future studies should include measurements of proinflammatory and antiinflammatory markers.

Assisted Reproductive Technologies (ART) with baboons generate live offspring: a nonhuman primate model for ART and reproductive sciences

Human reproduction has benefited significantly by investigating nonhuman primate (NHP) models, especially rhesus macaques. To expand the Old World monkey species available for human reproductive studies, we present protocols in baboons, our closest Old World primate relatives, for assisted reproductive technologies (ART) leading to live born offspring. Baboons complement rhesus by confirming or modifying observations generated in humans often obtained by the study of clinically discarded specimens donated by anonymous infertility patient couples. Here, baboon ART protocols, including oocyte collection, in vitro fertilization (IVF), intracytoplasmic sperm injection (ICSI), preimplantation development to blastocyst stage, and embryo transfer techniques are described. With baboon ART methodologies in place, motility during baboon fertilization was investigated by time-lapse video microscopy (TLVM). The first ART baboons produced by ICSI, a pair of male twins, were delivered naturally at 165 days postgestation. Genetic testing of these twins confirmed their ART parental origins and demonstrated that they are unrelated fraternal twins not identicals. These results have implications for ART outcomes, embryonic stem cell (ESC) derivation, and reproductive sciences.

Animal models of sepsis

In this review, we start with a general discussion of relevant factors that can determine the validity of a sepsis animal model. We briefly review some of the currently used animal models of sepsis (small animal models and large animal models). We discuss the clinical relevance of animal models in sepsis research today and address potential reasons for the apparent underperformance of animal models in predicting therapeutic success of novel drugs in clinical trials.

Mathematical modeling of posthemorrhage inflammation in mice: studies using a novel, computer-controlled, closed-loop hemorrhage apparatus

Hemorrhagic shock (HS) elicits a global acute inflammatory response, organ dysfunction, and death. We have used mathematical modeling of inflammation and tissue damage/dysfunction to gain insight into this complex response in mice. We sought to increase the fidelity of our mathematical model and to establish a platform for testing predictions of this model. Accordingly, we constructed a computerized, closed-loop system for mouse HS. The intensity, duration, and time to achieve target MAP could all be controlled using a software. Fifty-four male C57/black mice either were untreated or underwent surgical cannulation. The cannulated mice were divided into 8 groups: (a) 1, 2, 3, or 4 h of surgical cannulation alone and b) 1, 2, 3, or 4 h of cannulation + HS (25 mmHg). MAP was sustained by the computer-controlled reinfusion and withdrawal of shed blood within +/-2 mmHg. Plasma was assayed for the cytokines TNF, IL-6, and IL-10 as well as the NO reaction products NO2-/NO3-. The cytokine and NO2-/NO3- data were compared with predictions from a mathematical model of post-hemorrhage inflammation, which was calibrated on different data. To varying degrees, the levels of TNF, IL-6, IL-10, and NO2/NO3 predicted by the mathematical model matched these data closely. In conclusion, we have established a hardware/software platform that allows for highly accurate, reproducible, and mathematically predictable HS in mice.

Animal models for trauma research: what are the options?

Even if trauma patients initially avoid death after trauma (due to massive blood volume loss, primary severe brain injury), they are still at risk for multiple organ failure. Thus, it is crucial to elucidate the underlying pathophysiological mechanisms of trauma/hemorrhagic shock and the immune response involved. As of now, many hemorrhagic shock/trauma studies have used various types of animal models. Despite a large number of results from these efforts, some authors have argued that animal model results are difficult to translate directly into the clinical scenario. This review summarizes the advantages and the disadvantages of using animal models in trauma/hemorrhagic shock studies and discusses the relevance of various animal studies to the clinical scenario.

Translational research in the development of novel sepsis therapeutics: logical deductive reasoning or mission impossible?

The successful translation of promising research findings from basic research laboratories into useful clinical products for the management of septic patients has proven to be a daunting challenge. The complexity and variability of the clinical entity referred to as sepsis makes it intrinsically difficult to model preclinical systems and predict efficacy of potentially useful, experimental, therapeutic agents. Technological innovations in microarrays, microfluidics, and nanotechnology make it feasible to study the evolution of sepsis in small animal models in considerable detail. The recognized limitations of standard preclinical platforms used to study sepsis have lead to innovative approaches to study sepsis in silico, and in more complex and clinically more valid ex vivo tissue perfusion models and animal systems. It is abundantly clear that sepsis researchers need to do a better job informing clinicians about the possible benefits and potential risks of new treatment interventions as they traverse the gap between the bench and the bedside.

The disconnect between animal models of sepsis and human sepsis

Frequently used experimental models of sepsis include cecal ligation and puncture, ascending colon stent peritonitis, and the i.p. or i.v. injection of bacteria or bacterial products (such as LPS). Many of these models mimic the pathophysiology of human sepsis. However, identification of mediators in animals, the blockade of which has been protective, has not translated into clinical efficacy in septic humans. We describe the shortcomings of the animal models and reasons why effective therapy for human sepsis cannot be derived readily from promising findings in animal sepsis.