Trauma 2001; 3: 89100

The systemic inammatory response syndrome

Hanif Meeran and Mark Messent

The systemic inammatory response syndrome (SIRS) describes the clinical presentation of patients with systemic activation of the inammatory response from any underlying cause. SIRS is a common problem in acute medical and surgical practice and an important cause of morbidity and mortality. As a consequence of SIRS, patients may develop multiple organ dysfunction syndrome and acute respiratory distress syndrome (ARDS). Over the recent years our understanding of the inammatory response in SIRS has increased, but as yet specic immunomodulatory therapies have not proved useful. The mainstay of treatment for patients with SIRS and ARDS remains a general supportive care. It is in this area that more encouraging advances are being made, particularly in the management of invasive ventilation and nutrition. In this review we summarize the denitions, epidemiology and pathophysiology of SIRS, ARDS and related conditions. We then give a description of the clinical consequences and treatment of SIRS and ARDS with an emphasis on current aspects of supportive care. Key words: systemic inammatory response syndrome; sepsis; acute respiratory distress syndrome; multiple organ dysfunction syndrome; critical illness

Introduction
The inammatory response to injury or infection is usually a localized and protective physiological phenomenon. An overwhelming pro-inammatory stimulus or an exaggerated response can result in generalized inammation and dysfunction of organs distant from the initial site of injury. The term sepsis is often used to describe the clinical features of this process. This term is not ideal as sepsis implies an infective cause, which is not the case for some patients with systemic inammation. Following the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference in

1991 (Bone et al., 1992), the term systemic inammatory response syndrome (SIRS) was proposed to describe a clinical state arising from a non-specic cause, infective or otherwise. SIRS is dened as the presence of more than one of the following clinical manifestations: 1) a body temperature above 38C or less than 36C; 2) heart rate greater than 90 beats per minute; 3) respiratory rate greater than 20 breaths per minute or PaCO2 less than 4.3 kPa; 4) a white blood cell count of greater than 12 000/mm3 or less than 4000/mm3 or the presence of more than 10% immature neutrophils. These should represent acute changes not explained by other known causes. Sepsis is dened as SIRS, as a result of a conrmed infection. The Consensus Conference also proposed denitions for other conditions including severe sepsis, septic shock and multiple organ dysfunction syndrome (MODS) (Table 1).
14604086(01)TA096OA

Department of Intensive Care, London Chest Hospital, London, UK. Address for correspondence: M Messent, Department of Intensive Care, London Chest Hospital, Bonner Road, London E2 9JX, UK. E-mail: MarkMessent@btinternet.com Arnold 2001

90

H Meeran and M Messent

Table 1 Denitions associated with the SIRS (adapted from Bone et al., 1992) SIRS The systemic inammatory response to a variety of severe clinical insults. Manifested by two or more of the following: (1) (2) (3) (4) body temperature 38C or 36C, heart rate 90 beats per minute, tachypnoea: respiratory rate 20 breaths per minute or PaCO2 4.3 kPa (or 32 mmHg), white blood cell count 12 000/mm3 or 4000/mm3 or 10% immature bands.

These changes should represent an acute change from baseline and be unexplained by other causes. Sepsis SIRS, as dened above, as a result of a conrmed infection. Severe sepsis Sepsis associated with organ dysfunction, hypoperfusion or hypotension. Hypoperfusion and perfusion abnormalities may include but are not limited to lactic acidosis, oliguria or an acute alteration in mental status. Hypotension is dened as a systolic blood pressure 90 mmHg or a decrease in 40 mmHg from baseline in the absence of other causes. Septic shock Sepsis with hypotension despite adequate uid resuscitation, plus perfusion abnormalities as above. Patients who are on inotropic support may not be hypotensive when these perfusion abnormalities are measured. MODS The presence of altered organ function in an acutely ill patient such that homeostasis cannot be maintained without intervention.

Epidemiology
SIRS is a common condition and the multiple organ dysfunction, which results from SIRS, is a leading cause of death in intensive care units. In the United States the incidence of sepsis has been estimated as being 500 000 cases per year with a mortality rate of 35%. Sepsis has been listed as the 13th leading cause of death in the United States (Rangel-Frausto, 1995). The incidence appears to be increasing and despite an increase in the understanding of the pathophysiology of SIRS the mortality has not decreased. This may be explained by an increased awareness of the condition, resistant organisms and a greater number of patients who are immunocompromised or elderly with associated chronic diseases (Balk, 2000). In a prospective survey of 3708 patients admitted to a tertiary referral hospital, 68% met the criteria for SIRS with a similar distribution between medical and surgical patients. Twenty-one percent of patients with SIRS went on to develop sepsis, 18% developed severe sepsis and 4% developed septic shock. A continuum was demonstrated from SIRS to the development of sepsis, and progression to septic shock. There was also a trend of increasing multiple organ dysfunction and mortality from SIRS to sepsis, severe sepsis and septic shock. Patients classied as having septic shock had a 46% mortality. As more criteria for SIRS were met there was also a progressive increase in mortality and for
Trauma 2001; 3: 89100

patients meeting all four criteria mortality was 17% (Rangel-Frausto et al., 1995). Mortality has also been correlated with the number of failing organ systems. Less than three organ system failure has been associated with a 20% mortality, rising to 70% when more than three organ systems fail (Herbert et al., 1993).

Aetiology and pathophysiology of SIRS

Infection is an important cause of SIRS, usually bacterial in origin and when identied the condition can also be called sepsis. Bacteria identied can be Gram-negative or Gram-positive organisms, especially Staphylococcus aureus and enterococci (Balk, 2000). Endotoxin produced by Gram-negative organisms is an important trigger of inammation. Although infection may be thought to be the underlying cause for the majority of patients with SIRS, an infective organism is often never identied (Rangel-Frausto et al., 1995; Miller et al., 1999). There are also many recognized non-infective clinical conditions that can result in SIRS; examples include pancreatitis, multiple trauma, burns, aspiration, ischaemia and haemorrhagic shock. In any case SIRS due to non-infective causes and sepsis seem to be closely related entities in which there is an excessive and disordered inammatory response. The inammatory response is an incredibly complicated

The systemic inammatory response syndrome 91 cascade of events involving a myriad of cellular and molecular species, many of which have overlapping functions. There is normally close regulation of this system and in addition to pro-inammatory mediators there are also those with anti-inammatory functions. The bodys response can be up- or downregulated so that hyper-responsiveness or tolerance can occur depending on previous inammatory activation. To complicate matters further there are genetic variations in the response to pro-inammatory stimuli. Thus it may be that it is not just the nature of the initiating injury but also the hosts inammatory response that dictates the eventual outcome (Bone, 1996; Kumar et al., 1999). Usually inammation is localized and healing occurs, but in some circumstances activation of the inammatory cascade occurs on a systemic level. This gives rise to the clinical features of SIRS. In its milder forms a degree of systemic inammation might be benecial for an individual who is combating illness. A point may be reached, however, when the inammatory process becomes out of control and no longer benecial. A dysregulated inammatory process ensues in which there is inammation in organs distant from the original pro-inammatory insult leading to MODS. The anti-inammatory response is designed to downregulate inammation and limit its destructive effects. This too may become uncontrolled causing susceptibility to infection and may lead to the concept of the compensatory anti-inammatory response syndrome (CARS). A mixture of ongoing pro- and antiinammatory activation may coexist and this has been termed the mixed antagonists response syndrome (MARS). Only if the body is able to restore some balance to this immunological chaos, recovery is likely to occur (Bone et al., 1997).

The role of endotoxin

Endotoxin is a component of gram-negative bacterial cell walls and is also called lipopolysaccharide (LPS). LPS is an important trigger for SIRS, indeed normal volunteers develop features of SIRS after LPS administration (Suffredini et al., 1989). LPS consists of a core region, O antigen polysaccharide and a toxic lipid A region. LPS released from bacteria can associate with cell surface receptors including CD14 on monocyte/macrophages resulting in activation of these cells. This process is facilitated by a serum protein called lipopolysaccharide binding protein (LBP). In addition, LPS can activate neutrophils and endothelial cells to produce free radicals and NO. Further inammation and eventually disseminated intravascular coagulation can result from activation of complement and the coagulation cascade by LPS (Stephens and Hamilton-Davies, 2000).

Cytokines
Cytokines are produced by macrophage/monocytes, PMNs, endothelial and other cells, and can be considered to be messengers of inammation. They can act in an autocrine or paracrine or, during SIRS, in a hormonal fashion. Cytokines can be divided into pro-inammatory, of which TNF and IL-1 are the most important, and anti-inammatory, which include IL-4 and IL-10. TNF is produced by macrophage/monocytes and possibly mast cells and acts on a specic TNF receptor. Infusion of TNF into subjects produces symptoms and signs of SIRS, and serum levels of TNF correlate with the development of SIRS and MODS (Tracey et al., 1986; Chapman et al., 1987). TNF promotes activation of macrophages, endothelial cells and PMNs, as well as activating complement and the coagulation system. This results in the production of further cytokines and other inammatory mediators and the recruitment of more inammatory effector cells (Davies and Hagen, 1997; Kim and Deutschman, 2000). IL-1 is also released by macrophages alongside or possibly in response to TNF secretion and has most of the same effects. TNF, IL-1 and also IL-6 are responsible for the induction of the acute phase response by the liver. IL-6 is secreted by macrophages, PMNs, lymphocytes, endothelial cells, mast cells and other cells in response to a variety of stimuli including TNF, IL1, LPS, interferon- and platelet derived growth factor. Although often considered along with proinammatory cytokines IL-1 and TNF, IL-6 is important as a growth factor and some of the acute
Trauma 2001; 3: 89100

Biology of inammation
Inammation is the bodys non-specic response to tissue damage and can be produced by infectious, mechanical or chemical stimuli. Initial mediators of inammation include the cytokines, tumour necrosis factor alpha (TNF), interleukin 1 beta (IL1), macrophages, mast cells, complement and the coagulation cascade. Endothelial cells become activated and polymorphonucleocytes (PMNs) are recruited, leading to further release of cytokines and secondary inammatory mediators. These secondary mediators include arachidonic acid metabolites, reactive oxygen species (ROS), nitric oxide (NO), platelet activating factor (PAF) and many more (Davies and Hagen, 1997; Kim and Deutschman, 2000).

92

H Meeran and M Messent Monocytes and tissue macrophages are phagocytic cells involved in antigen presentation to lymphocytes and also produce toxic ROS. The role of the monocyte/macrophage in SIRS is to coordinate other inammatory cells, especially PMNs, by the production of cytokines and secondary mediators such as leucotrienes and prostaglandins. In addition to the role of the endothelium in leucocyte recruitment, the activated endothelial cell shifts from its normally anti-thrombotic activity to pro-thrombotic and antibrinolytic, alters vascular tone and produces additional inammatory mediators. Activated endothelial cells express tissue factor and produce TNF, both of which are activators of the extrinsic coagulation pathway. Von Willibrand factor, thromboxane (TXA2) and platelet adhesion molecules are produced promoting the production of further pro-inammatory and pro-coagulant substances. Anti-thrombotic factors such as protein C and anti-thrombin III are reduced. Fibrinolysis is inhibited by increased plasminogen activator inhibitors, PAI-1 and PAI-2. Coupled with endothelial cell injury these events lead to the development of microthrombi and impaired blood ow (Davies and Hagen, 1997; Gando et al., 1998). Vascular tone is decreased by elevated levels of NO, prostacyclin (PGI2), and bradykinin. Endothelial cells retract from each other when activated and permeability of the vasculature is further increased by PAF, TXA2, and leucotrienes. Many of these substances are produced by the activated endothelial cell as well as other inammatory cells (Davies and Hagen, 1997; Fry, 2000). The result of these changes is a familiar clinical picture of profound vasodilation, with or without hypotension and large uid requirements. Activated endothelial cells also increase production of vasoconstrictor substances called endothelins. The role of these in SIRS is unclear but endothelins might contribute to pulmonary hypertension and reduced blood ow to the gut and kidneys (Wort and Evans, 1999).

phase reactants that it promotes are anti-inammatory. IL-6 has also been used as a marker of severity and an indicator of prognosis in SIRS (Taniguchi et al., 1999; Deutschman, 1998). IL-8 belongs to the family of chemokines or chemotaxins and is produced by macrophages, PMNs and endothelial cells in response to pro-inammatory cytokines. It enhances chemotaxis of PMNs and lymphocytes. There are a large number of other chemokines including CC groups, which is specic for monocytes, and CXC for PMNs (nomenclature which refers to their amino acid sequences) (Fry, 2000). Anti-inammatory cytokines include IL-10, IL-4 and transforming growth factor-; these are synthesized by monocytes and lymphocytes and inhibit the release of pro-inammatory mediators. There are also endogenous antagonists to IL-1 and TNF, such as IL-1 receptor antagonist and soluble TNF receptors. These must be important regulators of the inammatory response, but can cause paradoxical immunosuppression in a critically ill patient and an inability to mount an appropriate response to further infection (Adrie and Pinsky, 2000).

Cells involved in the inammatory response

Macrophages, neutrophils and endothelial cells are important effectors of the inammatory response. Mast cells are also important early in the activation of inammation by releasing pre-formed proinammatory substances such as histamine. Endothelial cell activation results in the increased expression of adhesion molecules on their surface. Two of these are endothelial leucocyte adhesion molecule (ELAM) 1 which binds monocytes and PMNs, and intercellular adhesion molecule (ICAM) 1 which binds PMNs and lymphocytes. Simultaneously leucocytes produce complementary adhesion molecules on their surface, called integrins and include CD11 and CD18. Leucocytes then adhere to and roll along the endothelium, nally migrating through towards the site of inammation attracted by chemokines (Davies and Hagen, 1997; Wort and Evans, 1999). PMNs are of key importance in the production of tissue damage in SIRS. Margination of PMNs occurs along the endothelium not only at the initial site of injury, but also in other organs due to the inuence of high systemic levels of pro-inammatory cytokines. Once activated they produce potent proteases and ROS as well as activating platelets and the coagulation cascade. These damage endothelium and other tissues and disrupt the microcirculation, eventually producing organ dysfunction (Fry, 2000).
Trauma 2001; 3: 89100

Intracellular events
Once exposed to a pro-inammatory stimulus, the primary intracellular promoter of the inammatory response is nuclear factor-kappa B (NF-B). NF-B is activated by intracellular oxidative stress induced when the cell is exposed to cytokines such as TNF, IL-1 or other inammatory stimuli. Once activated NF-B promotes production of mRNA for inammatory cytokines in the cell nucleus. There is also an intracellular anti-inammatory system called the heat shock protein (HSP) system. This is a primitive and

The systemic inammatory response syndrome 93 important system for defending the cell against many stresses such as trauma and inammation. HSPs induced by the same stimuli that promote inammation, seem to protect against oxidative stress and may modulate NF-B (Adrie and Pinsky, 2000; Christman et al., 1998). The leucotrienes include LTB4, LTC4, LTD4 and LTE4; they are chemotactic agents that also increase vascular permeability (Heller et al., 1998; Wort and Evans, 1999).

Reactive oxygen species

ROS and an imbalance between oxidative stress and anti-oxidant activity play an important part in the pathology of SIRS. PMNs, monocyte/macrophages and endothelial cells produce ROS, such as superoxide, using the enzyme xanthine oxidase when activated. Other oxidative compounds include peroxynitrite formed from NO, and reactive iron species. These injure tissue by oxidising essential cellular components and altering the function of cellular enzyme systems. The inammatory response can also be amplied using ROS by upregulation of proinammatory cytokines, thus contributing to a cycle of tissue damage and further inammation in SIRS. Patients with ARDS and SIRS have raised indicators of oxidative stress such as lipid peroxidation and hypoxanthine levels. Many endogenous anti-oxidant agents exist to combat oxidative stress. These include vitamins E, C, A, the selenium containing glutathione peroxidase enzymes and some acute phase proteins. There is evidence that these anti-oxidant systems are overwhelmed in SIRS leading to the concept of redox imbalance (Zhang et al., 2000; Forceville et al., 1998; Gutteridge and Mitchell, 1999).

Secondary mediators of inammation Nitric oxide

Increased production of NO is probably due to triggering of the inducible form of NO synthetase (iNOS) during inammation. Synthesis is from L-arginine and occurs in endothelial cells, monocytes and PMNs. The high levels of NO identied in SIRS enhance vasodilation and vascular permeability. Smooth muscle tone is reduced via stimulation of guanylate cyclase to form cyclic GMP. NO decreases platelet aggregation and results in negative inotropy. NO is also a free radical and reacts with superoxide to produce the toxic molecule peroxynitrite which may contribute to tissue injury in SIRS (Wort and Evans, 1999; Gutteridge and Mitchell, 1999).

Platelet activating factor

PAF is released from macrophages, PMNs, endothelial cells and platelets, and is a potent inammatory cell activator. PAF promotes cell adhesion and activation, increases platelet activation, is negatively inotropic and enhances microvascular permeability. The production of oxygen free radicals and arachidonic acid metabolites can also be induced by PAF (Heller et al., 1998; Tetta et al., 1997).

Translocation theory
Although an important cause of SIRS is invasive infection at any site in the body, for example pneumonia or meningococcal disease, a second mechanism exists whereby bacteria and their products may cause SIRS. The gut is particularly vulnerable to hypoperfusion and hypoxia, due to the structure of the microcirculation in the mucosal villi and a high critical oxygen requirement (Marik, 1999). Contained within the lumen of the intestine are a large number of commensal organisms. Hypoperfusion of the gut is a common nding in the critically ill patient. Mucosal ischaemia is thought to increase permeability to bacteria and their toxins. Malnutrition can also impair gut barrier function and this is prevalent amongst critically ill patients (Reynolds et al., 1996). There may also be colonization with abnormal ora due to antibiotic therapy, and dysfunction of normal host defences which include lymphoid tissue and the liver. These factors combine to enable bacteria, endotoxin and cytokines from gut lymphoid tissue to enter the systemic circulaTrauma 2001; 3: 89100

Arachidonic acid metabolites

Arachidonic acid is metabolized via the cyclooxygenase (COX) pathway to produce prostaglandins or the lipoxygenase pathway to produce leucotrienes. Arachidonic acid metabolites are involved in vascular control, coagulation, oedema formation and chemotaxis. Prostacyclin (PGI2) is an important vasodilator and raised concentrations have been found in sepsis. PGI2 production in SIRS is due to induction of the enzyme COX-2, which is found in many inammatory cells, including endothelial cells. The enzyme COX-1 is present constitutively and assumed to contribute to the maintenance of homeostasis. The cytoprotective COX-1 may be downregulated in SIRS, thus compounding loss of homeostatic control. Thromboxane (TXA2) is also a prostaglandin released from endothelial cells, its actions are to increase platelet aggregation and also increase vascular permeability.

94

H Meeran and M Messent

Antagonists: IL1RA, TNFR, IL4, IL10, Complement Cytokines: TNF IL1- Platelets Coagulation cascade, TXA2 PAI TF IL-1R

LBP CD14 Macrophage COX iNOS

PMN iNOS COX

Secondary mediators: PG, LTs NO ROS, PAF

CD11, CD14

TNFR ICAM, ELAM

Endothelial cell

iNOS

COX

Vasoactive substances: NO, ET, PG

Smooth muscle cell

Figure 1 Interactions between the cellular and humoral components of the inammatory response. This can be activated by a wide variety of stimuli including infection, endotoxin, ischaemia, trauma and haemorrhage. Abbreviations: COX, cyclo-oxygenase; iNOS, inducible nitric oxide synthetase; PG, prostaglandin; NO, nitric oxide; ROS, reactive oxygen species; PAF, platelet activating factor; TXA2, thromboxane; PAI, plasminogen activator inhibitor; LBP , lipopolysaccharide binding protein; PMN, polymorphonucleocyte; ICAM, intercellular adhesion molecule; ELAM, endothelial leucocyte adhesion molecule; TF , tissue factor; IL, interleukin; IL-1R, interleukin 1 receptor; IL1RA, interleukin 1 receptor antagonist; TNF, tumour necrosis factor; TNFR, TNF receptor; ET, endothelin.

tion and initiate or amplify systemic inammation (Rowlands et al., 1999; Deitch and Goodman, 1999).

Two-hit theory
SIRS and MODS may be initiated by a single episode of infection or massive injury and shock, a so-called one-hit model. Perhaps more commonly a patient might sustain an initial insult, for example pancreatitis or trauma, that initiates a mild inammatory response. This is followed by a second pro-inammatory challenge, for example infection or ischaemia that produces a severe uncontrolled SIRS. The inammatory process is primed somehow to produce an exaggerated response so that critically ill patients become more vulnerable to subsequent challenges (Bone, 1996; Deitch and Goodman, 1999).

and alteration in white blood cell count. The image of a patient who is non-specically described as septic or toxic with sweating, fever, chills and hypotension will be familiar to most clinicians. A frequent complication of SIRS is MODS. Dysfunction is dened as an inability of the organ to maintain homeostasis. Any organ system can become dysfunctional as a result of SIRS, and the number of organ systems involved is directly related to mortality (Bone et al., 1992; Herbert et al., 1993). The most important manifestations of MODS in SIRS include the spectrum of acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) and cardiovascular derangement.

Acute lung injury and acute respiratory distress syndrome

ALI and ARDS are processes of non-hydrostatic pulmonary oedema due to inammation and increased vascular permeability within the lung. This presents as hypoxia, bilateral pulmonary inltrates on the chest X-ray and other associated physiological abnormalities such as decreased compliance and

Clinical manifestations
The dening clinical features of SIRS are disturbance of body temperature, tachycardia, hyperventilation
Trauma 2001; 3: 89100

The systemic inammatory response syndrome 95 pulmonary vascular hypertension. ALI represents the less severe end of this pathological process and may or may not progress to ARDS. Uniform denitions for ARDS and ALI were proposed by the AmericanEuropean Consensus Conference of 1992: ALI should be of acute onset with a PaO2/FiO2 ratio of 40 kPa and a pulmonary artery wedge pressure of 18 mmHg with no evidence of left atrial hypertension; ARDS criteria are the same as for ALI but the PaO2/FiO2 ratio should be 27 kPa (Bernard et al., 1994) (Table 2). ARDS is associated with many aetiologies, including those that injure the lung directly such as aspiration, trauma or infection, and indirect causes such as SIRS, sepsis and pancreatitis. The likelihood of developing ARDS varies with the aetiology, as does the mortality. The annual incidence of ARDS has been estimated as 510 per 100 000 population. Mortality is lower due to trauma, high due to infection and is typically quoted as over 50%, although this may have reduced recently (Artigas et al., 1998; Wyncoll and Evans, 1999). pattern of lung disease with an increase in alveolar dead space. The intermediate phase represents a transition between early and late phases (Artigas et al., 1998; Gattinoni et al., 1994). Although the changes appear homogeneous on chest X-ray, when ARDS patients are examined by computed tomography (CT), considerable heterogeneity is revealed. There are areas of lung, particularly in less dependent areas, which appear normal and these may also function normally. Thus the patient with ARDS may be functioning with a greatly reduced volume of normal lung tissue or baby lung. CT scan may also reveal emphysematous bullae, pneumothoraces and abscesses not seen on chest X-ray (Gattinoni et al., 1994). The disease progression of ALI and ARDS is variable in its severity and time course, but patients frequently require mechanical ventilation for days or weeks. In survivors of late ARDS lung structure and function gradually returns to near normal (Ware and Matthay, 2000).

Cardiovascular dysfunction in SIRS Pathophysiology of ARDS

ALI and ARDS are a consequence of activation of the inammatory cascade within the lung. This may be secondary to SIRS, but ARDS from a primary pulmonary insult may also lead to systemic inammation and damage to other organs. ARDS is often divided into early, intermediate and late phases. Early ARDS occurs in the rst week and is characterized by intense lung inammation with increased epithelial and endothelial permeability, oedema, inammatory exudates and cells that ll air spaces, surfactant dysfunction and atelectasis. This tends to cause shunt and hypoxia resistant to oxygen therapy, decrease overall lung compliance and raise pulmonary vascular resistance. In late ARDS, which occurs after usually about two weeks, there is proliferation of broblasts, collagen deposition and lung brosis with emphysematous areas. Late ARDS patients have a restrictive Patients with SIRS tend to have a hyperdynamic circulation with tachycardia, raised cardiac output and systemic vasodilation. Hypovolaemia frequently occurs as a consequence of capillary leak, poor oral intake and uid loss due to sweating, tachypnoea, vomiting or bleeding. Hypotension may result from vasodilation, venous pooling, relative hypovolaemia and an inability to mount a sufcient increase in cardiac output. There is an increase in circulating catecholamines and also inammatory mediators such as NO which have vasodilator and cardiodepressant effects (Jindal et al., 2000). Septic shock is dened as sepsis with a systolic blood pressure 90 mmHg or a decrease of 40 mmHg and has a mortality of over 40%. A hypermetabolic state with increased energy consumption and oxygen requirement occurs early in SIRS (Plank and Hill, 2000). Failure to increase cardiac output and oxygen delivery to meet this need results in end organ hypoperfusion and hypoxia; this is evidenced by signs such as oliguria, confusion and lactic acidosis. Microcirculatory impairment and abnormal oxygen utilization also contribute to acidosis in SIRS that can be refractory to increased cardiac output and oxygen delivery (Wheeler and Bernard, 1999). Patients tend to become oedematous with SIRS and those with severe sepsis tend to retain more uid than those who are less ill. After some days, if recovery occurs, this extra uid is excreted often creating a marked diuresis (Plank and Hill, 2000).
Trauma 2001; 3: 89100

Table 2 Denitions of ARDS and ALI (adapted from Bernard, 1994) Acute onset. Bilateral inltrates on chest radiograph. Pulmonary artery wedge pressure 18 mmHg or the absence of clinical evidence of left atrial hypertension. ALI present if PaO2/FiO2 ratio is 40 kPa (or 300 mmHg). ARDS present if PaO2/FiO2 ratio is 27 kPa (or 200 mmHg).

96

H Meeran and M Messent underlying cause, the mainstay of treatment remains a general supportive care. This is aimed at maintaining organ function and preventing further complications associated with critical illness. Specically this may include interventions such as mechanical ventilation, cardiovascular support, enteral feeding, and prevention and treatment of infection. Good nursing care is invaluable in all aspects, but especially in the prevention of skin breakdown. In mechanically ventilated patients prevention of stress ulceration and judicious use of sedation is important. Measures to prevent venous thromboembolism should not be forgotten.

Other organ system dysfunction

Haematological dysfunction as manifested by anaemia and mild coagulation abnormalities is a common nding in SIRS. In more severe cases disseminated intravascular coagulation and leucopoenia can occur. Coagulopathy is caused by activation of the extrinsic pathway and deciencies in some coagulation proteins such as protein C and anti-thrombin III (Gando et al., 1998; Wheeler and Bernard, 1999). A mild elevation of liver enzymes and bilirubin is common, but frank hepatic failure is unusual and is associated with a high mortality (Herbert et al., 1993). Ileus frequently occurs especially when there has been a period of hypoperfusion as in septic shock; this may be further compounded by inadequate uid resuscitation and injudicious use of vasoconstrictors. In addition sedatives and opioids are often given to patients on intensive care and these contribute to loss of gut motility. Much of the nutritional requirement of cells within the gut is derived from the gut lumen. Lack of enteral feeding adds to impairment of gut function, and normal mucosal structure deteriorates. Bacterial translocation is increased as is susceptibility to stress ulceration (Rowlands et al., 1999). Oliguria is a common finding in SIRS, but does not usually progress to acute renal failure provided adequate circulatory volume and blood pressure are maintained (Wheeler and Bernard, 1999). Cerebral dysfunction can be detected in SIRS, mild confusion is common but signicant deterioration in conscious level can occur and is associated with a poor prognosis. There is also a high incidence of neuromuscular dysfunction, this is partly attributable to drugs such as steroids and muscle relaxants but may also be due to disturbances in the microcirculation to peripheral nerves and muscle. The resultant loss of function may prolong recovery for weeks (Herbert et al., 1993; Bolton, 1996).

Mechanical ventilation
Less severely ill patients with SIRS may require no more than supplemental oxygen with a face mask, but many patients, including those with ALI and ARDS require mechanical ventilation in order to maintain adequate oxygenation. Although ARDS causes profound hypoxaemia, patients who die tend to die from SIRS and other organ dysfunction rather than respiratory failure. There has been an increasing understanding over recent years that the process of mechanical ventilation itself may lead to lung damage, release of inammatory mediators, and potentially SIRS leading to non-pulmonary organ failure (Ranieri et al., 1999). Traditional ventilatory strategies employ a tidal volume of 1015 ml/kg, a degree of positive end expiratory pressure (PEEP) and inspired oxygen concentration adjusted to achieve normal oxygenation, and minute volume adjusted to maintain normocapnia. Due to the small volume of aerated lung in ARDS patients (baby lung) this strategy tends to result in high end-inspiratory pressures. Lung damage can occur due to high alveolar pressure (barotrauma), this can include interstitial emphysema and pneumothorax. In addition to barotrauma, excessive stretch of ventilated alveoli due to large tidal volumes and high pressures is thought to cause volutrauma. Repeated opening and collapsing of alveoli and forces generated between alveoli of differing compliances during inspiratoryexpiratory cycling may cause injury via shear stresses. Finally, high inspired oxygen concentrations can produce lung injury in experimental models (Artigas et al., 1998; Roupie et al., 1995). An increased understanding of these problems has led most critical care clinicians to adopt more gentle forms of mechanical ventilation. These so-called lung protective ventilatory strategies do not have absolutely dened values as yet, but have several components in common. Lower tidal volumes are used to decrease airway pressure and limit alveolar over

Treatment of patients with SIRS

Supportive care
Despite advances in understanding the pathology of SIRS, interventions aimed at modifying specic aspects of the inammatory response have been largely disappointing. The most important aspect in the care of these patients is identication and treatment of the underlying cause of their illness. Without appropriate and timely medical and surgical intervention to remove the stimulus to systemic inammation, recovery is unlikely. Apart from treatment of the
Trauma 2001; 3: 89100

The systemic inammatory response syndrome 97 distension and a rise in pCO2 is tolerated. A degree of arterial hypoxaemia is also tolerated, typically a pO2 of 710 kPa is acceptable, as this helps limit inspired oxygen concentration and tidal volume (Tobin, 2000). PEEP is used to prevent repeated collapse of lung units at the end of each expiration, thus limiting shear forces and maintaining alveolar volume (Gattinoni et al., 1995). The level of PEEP can be set by determining a pressurevolume curve for each patient. The lower inection point of this compliance curve is theoretically the pressure at which less stable lung units collapse and re-open. It has been suggested that PEEP should be set above this pressure and this results in levels of PEEP somewhat higher than traditionally used (Keogh et al., 1999; Gattinoni et al., 1995). Several studies have examined the use of protective ventilation strategies in ARDS patients with conicting results (Amato et al., 1998; Brochard et al., 1998; Stewart et al., 1998), although the largest study so far does provide evidence of benet. This is by The ARDS Network and was stopped after the enrolment of 861 patients because mortality was 25% lower in the group ventilated with tidal volumes of 6 ml/kg than those receiving 10 ml/kg (The Acute Respiratory Distress Syndrome Network, 2000). Changing position from supine to prone has been used in severely hypoxic patients with ARDS. Over 50% of patients will show an improvement in oxygenation, although the magnitude of this is variable. The reasons for the improved oxygenation are unclear, but may be due to changes in chest wall compliance and more even distribution of alveolar ventilation with less atelectasis in dependent areas (Pelosi et al., 1998). Inhaled NO or prostacyclin have also been used in severely hypoxic patients with ARDS. These dilate pulmonary vessels in areas of ventilated lung, lower pulmonary vascular resistance and improve oxygenation. Unfortunately an improvement in mortality has not been demonstrated (Young, 1997; Wyncoll and Evans, 1999). Careful haemodynamic monitoring is essential as over zealous uid therapy may increase extravascular lung water and further impair oxygenation in ALI. Not one clinical sign or monitor alone can be used as an endpoint for resuscitation, although those that reect organ perfusion and function, such as urine output, may be the most useful. Base decit and serum lactate levels have been shown to be of value (Davis et al., 1988; Abramson et al., 1993) as metabolic acidosis frequently indicates poor tissue perfusion. In many patients invasive monitoring of arterial, central venous and pulmonary artery occlusion pressures is performed although these do not always correlate well with adequacy of uid therapy. Patients who survive SIRS tend to be those who are able to increase their cardiac output (CO) and oxygen delivery (DO2) to greater than normal. Unfortunately, once circulating volume is restored, attempts to increase CO and DO2 to supranormal levels with anything other than minimal doses of inotropic drugs has not been shown to improve mortality in critically ill patients (Hayes et al., 1994; Gattinoni et al., 1995). Inotropes and vasopressors may, however, be essential to maintain adequate cardiovascular function in severe SIRS, for example in septic shock. Noradrenaline is a logical choice for many patients to overcome the profound vasodilation that is frequently present. -agonists can be used if reduced myocardial contractility is a feature. Dopexamine is also of interest as it appears to improve splanchnic blood ow and splanchnic ischaemia seems to inuence outcome in the critically ill patients (Maynard et al., 1995). Gastric tonometry has been advocated to monitor gut perfusion and cardiovascular therapy, but its role is not yet clear (Kolkman et al., 2000). Critically ill patients with SIRS frequently become anaemic necessitating red cell transfusions to maintain oxygen carrying capacity. Concern has grown recently over the complications of blood transfusion and it seems that the traditional transfusion trigger of 10 g/dl may be too liberal for most patients. Maintaining a haemoglobin of 7.09.0 g/dl has been suggested to be safe for patients without ischaemic heart disease (Herbert et al., 1999).

Cardiovascular support and uid therapy

In addition to protection of the airway and ensuring adequate oxygenation it is essential to restore an optimal circulatory volume in SIRS. Hypovolaemia can be treated with either crystalloid or colloid solutions. Although crystalloid is safe and cheap it must be administered in volumes 46 times the actual intravascular decit due to its short intravascular half life. Patients with SIRS are already prone to tissue oedema and for this reason many clinicians favour colloids (Boldt, 2000).

Nutritional support
The benets of providing enteral nutrition to critically ill patients have been established over the last decade. These include improved gut structure and function, improved antibacterial defences and decreased infection rate (Pastores et al., 1996). When oral or nasogastric feeding is not possible, for example due to ileus,
Trauma 2001; 3: 89100

98

H Meeran and M Messent context of septic shock and adrenocortical insufciency. Modest doses of hydrocortisone may allow reduction in vasopressor therapy and mortality may even be improved (Bollaert et al., 1998; Briegel et al., 1999). Steroids may also be useful in the treatment of late, non-resolving ARDS. Improved lung function has been demonstrated, possibly by attenuation of the inammatory process thus decreasing the advance of broproliferation (Meduri et al., 1998). In other situations the use of steroids in SIRS may be associated with an increase in infections, no improvement or a decrease in survival (Marras et al., 1999; Ware and Matthay, 2000).

parenteral nutrition can be used, but its usefulness in critically ill patients has been questioned (Kennedy and Hall, 2000). Certain compounds such as arginine, glutamine, omega-3-fatty acids and RNA enhance immune and metabolic function. These have been added to enteral feeds in the hope of further improvements in morbidity and mortality. Current evidence suggests that there may be an improved infectious complication rate and reduced hospital stay but no difference in mortality when compared with standard enteral feed (Bower et al., 1995; Kennedy and Hall, 2000). Ranitidine or sucralphate are often administered to prevent stress ulcers and gastrointestinal bleeding in patients receiving mechanical ventilation. A recent meta-analysis suggested that both agents are no better than placebo at preventing gastrointestinal bleeding, and that ranitidine might increase the risk of pneumonia (Messori et al., 2000).

Anti-endotoxin and anticytokine therapies

Endotoxin is implicated in activating SIRS and MODS and is therefore an obvious target for therapy. Anti-endotoxin treatments include monoclonal antibodies, vaccines and substances that bind endotoxin. As yet none of these are clinically useful (Stephens and Hamilton-Davies, 2000). A large number of agents have been produced to antagonize inammatory mediators such as TNF, IL-1, PAF and NO. These have been unsuccessful and sometimes harmful. This may be due to a disproportionate emphasis on targeting pro-inammatory mediators and failure to recognize their benecial effects, or because the inammatory cascade is too complex for a single agent to make any impact (Bone, 1996; Fry, 2000). Perhaps combination treatments or those that non-specically lower plasma concentrations of cytokines such as haemoltration may be more useful (Rodby, 1998).

Control and treatment of infection

Good hygiene practices by all hospital staff are essential to limit nosocomial infection. Many patients with SIRS have an infectious origin for their illness and those who develop SIRS for other reasons, for example multiple trauma, frequently acquire an infection whilst in hospital. Unnecessary use of antibiotics may disrupt normal ora and enhance selection of resistant organisms, but untreated infection may lead to death. Every effort must be made to search for the site of infection and identify the offending organism prior to commencing treatment. As it can be very difcult to distinguish whether patients are infected, antibiotics are usually started empirically after testing culture samples (Wheeler and Bernard, 1999). To help discriminate between patients with SIRS who have sepsis and those who are not infected, some have suggested the use of serum markers such as C-reactive protein and procalcitonin, both of which are elevated in sepsis (Miller et al., 1999; Brunkhorst et al., 2000). Prophylactic administration of antibiotics to remove potentially pathogenic organisms from the gut (selective digestive decontamination) can reduce the risk of nosocomial infections and perhaps improve mortality. This strategy might increase antibiotic resistance, may not be cost effective and popular (DAmico et al., 1998).

Antioxidants
Oxidative stress plays an important role both in the activation of the inammatory response and in tissue damage in SIRS. Several anti-oxidant substances have been proposed as treatments in SIRS and ARDS including N-acetylcysteine, selenium and vitamin E, but so far none of these are widely recommended (Artigas et al., 1998; Angstwurm et al., 1999; Walsh and Lee, 1999).

Summary
The term SIRS was invented in an attempt to standardize denitions surrounding sepsis. SIRS is a useful concept as it recognizes that patients who show clinical manifestations of systemic inammation are not all infected. The denition of SIRS is a very

Specic treatments Corticosteroids

The anti-inammatory effects of steroids have led to their use as treatment for sepsis and ARDS for many years. The use of steroids in SIRS is indicated in the
Trauma 2001; 3: 89100

The systemic inammatory response syndrome 99 sensitive denition, thus it encompasses a wide range of aetiologies and severities of illness. Although some patients who meet SIRS criteria may not be severely ill, all may have the potential to become so if their disease is allowed to progress. It is therefore important for all clinicians in acute medical and surgical practice to have a good knowledge of SIRS and its implications. Patient care should be directed towards treatment of the underlying cause for SIRS, resuscitation and supportive care to prevent further complications so that recovery may take place. Despite advances in our understanding of the inammatory response and improvements in supportive care mortality in conditions such as sepsis and ARDS remains high. Future improvements in supportive care are likely, for example with mechanical ventilation and feeding, but useful immunomodulatory therapies seem some way off. Nevertheless as appreciation of the complexities of the inammatory and anti-inammatory responses grows, there is hope that some successful way of manipulating the immune system in SIRS may emerge.
Bone RC. 1996. Immunological dissonance: a continuing evolution in our understanding of the systemic inammatory response syndrome (SIRS) and the multiple organ dysfunction syndrome (MODS). Ann Intern Med 125: 680 87. Bone RC, Balk RA, Cerra FB et al. 1992. American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: denitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101: 1644 55. Bone RC, Grodzin CJ, Balk RA. 1997. Sepsis: a new hypothesis for pathogenesis of the disease process. Chest 112: 23543. Bower RH, Cerra FB, Bershadsky B et al. 1995. Early enteral administration of a formula (Impact) supplemented with arginine, nucleotides, and sh oil in intensive care unit patients: result of a multicenter, prospective, randomized, clinical trial. Crit Care Med 23: 43649. Briegel J, Forst H, Haller M et al. 1999. Stress doses of hydrocortisone reverse hyperdynamic septic shock: a prospective, randomised, double-blind, single-center study. Crit Care Med 27: 723 32. Brochard L, Roudot-Thorval F, Roupie E et al. 1998. Tidal volume reduction for prevention of ventilator-induced lung injury in acute respiratory distress syndrome. Am J Respir Crit Care Med 158: 183138. Brunkhorst FM, Wegscheider K, Forycki ZF et al. 2000. Procalcitonin for early diagnosis and differentiation of SIRS, sepsis, severe sepsis, and septic shock. Intensive Care Med 26: S14852. Chapman PB, Lester TJ, Casper ES et al. 1987. Clinical pharmacology of recombinant human tumour necrosis factor in patients with advanced cancer. J Clin Oncol 5: 194251. Christman JW, Lancaster LH, Blackwell TS. 1998. Nuclear factor B: a pivotal role in the systemic inammatory response syndrome and new target for therapy. Intensive Care Med 24: 113138. DAmico R, Pifferi S, Leonetti C et al. 1998. Effectiveness of antibiotic prophylaxis in critically ill adult patients: systematic review of randomised controlled trials. Br Med J 316: 127585. Davies MG, Hagen PO. 1997. Systemic inammatory response syndrome. Br J Surg 84: 92035. Davis JW, Shackford SR, Mackersie RC et al. 1988. Base decit as a guide to volume resuscitation. J Trauma 28: 146467. Deitch EA, Goodman ER. 1999. Prevention of multiple organ failure. Surg Clin North Am 79: 147178. Deutschman CS. 1998. Acute-phase responses and SIRS/ MODS: the good, the bad, and the nebulous. Crit Care Med 26: 163031. Forceville X, Vitoux D, Gauzit R et al. 1998. Selenium, systemic immune response syndrome, sepsis, and outcome in critically ill patients. Crit Care Med 26: 153644. Fry DE. 2000. Sepsis syndrome. Am Surg 66: 12632. Gando S, Nanzaki S, Sasaki S et al. 1998. Activation of the extrinsic coagulation pathway in patients with severe sepsis and septic shock. Crit Care Med 26: 20052009. Gattinoni L, Bombino M, Pelosi P et al. 1994. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA 271: 177279. Gattinoni L, Brazzi L, Pelosi P et al. 1995. A trial of goal-oriented haemodynamic therapy in critically ill patients. N Engl J Med 333: 1025 32. Trauma 2001; 3: 89100

References
Abramson D, Scalea TM, Hitchcock R et al. 1993. Lactate clearance and survival following injury. J Trauma 35: 58488. Adrie C, Pinsky MR. 2000. The inammatory balance in human sepsis. Intensive Care Med 26: 364 75. Amato MBP, Barbas CSV, Medeiros DM et al. 1998. Effect of a protective ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 338: 34754. Angstwurm MWA, Schottdorf J, Schopohl J et al. 1999. Selenium replacement in patients with severe systemic inammatory response syndrome improves clinical outcome. Crit Care Med 27: 180713. Artigas A, Bernard GR, Carlet J et al. 1998. The American European Consensus Conference on ARDS, Part 2. Am J Respir Crit Care Med 157: 1332 47. Balk RA. 2000. Severe sepsis and septic shock denitions, epidemiology, and clinical manifestations. Crit Care Clin 16: 17991. Bernard GR, Artigas A, Brigham KL et al. 1994. Report of the AmericanEuropean Consensus Conference on ARDS: denitions, mechanisms, relevant outcomes and clinical trial coordination. Intensive Care Med 20: 22532. Boldt J. 2000. Volume replacement in the surgical patient does the type of solution make a difference? Br J Anaes 84: 78393. Bollaert PE, Charpentier C, Levy B et al. 1998. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 26: 64550. Bolton CF. 1996. Sepsis and the systemic inammatory response syndrome: neuromuscular manifestations. Crit Care Med 24: 140816.

100

H Meeran and M Messent

patients with severe sepsis or major blunt trauma. World J Surg 24: 630 38. Rangel-Frausto MS, Pittet M, Costigan M et al. 1995. The natural history of the systemic inammatory response syndrome (SIRS). JAMA 273: 11723. Ranieri VM, Suter PM, Tortorella C et al. 1999. Effect of mechanical ventilation on inammatory mediators in patients with acute respiratory distress syndrome a randomized controlled trial. JAMA 282: 54 61. Reynolds JV, OFarrelly C, Feighery C et al. 1996. Impaired gut barrier function in malnourished patients. Br J Surg 83: 128891. Rodby RA. 1998. Haemoltration for SIRS: bloodletting, twentieth century style? Crit Care Med 26: 1940 41. Roupie E, Dambrosio M, Servillo G et al. 1995. Titration of tidal volume and induced hypercapnia in acute respiratory distress syndrome. Am J Respir Crit Care Med 152: 12128. Rowlands BJ, Song CV, Gardiner KR. 1999. The gastrointestinal tract as a barrier in sepsis. Br Med Bull 55: 196211. Stephens R, Hamilton-Davies C. 2000. Update on antiendotoxin therapies. Hospital Medicine 61: 254 58. Stewart TE, Meade MO, Cook DJ et al. 1998. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. N Engl J Med 338: 355 61. Suffredini AF, Fromm RE, Parker MM et al. 1989. The cardiovascular response of normal humans to the administration of endotoxin. N Engl J Med 321: 280 87. Taniguchi T, Koido Y, Aiboshi J et al. 1999. Change in the ratio of interleukin-6 to interleukin-10 predicts a poor outcome in patients with systemic inammatory response syndrome. Crit Care Med 27: 1262 64. Tetta C, Mariano F, Buades J et al. 1997. Relevance of platelet activating factor in inammation and sepsis: mechanisms and kinetics of removal in extracorporeal treatments. Am J Kidney Dis 30(Suppl. 4): 57 65. The Acute Respiratory Distress Syndrome Network. 2000. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 342: 1301308. Tobin M. 2000. Culmination of an era in research on the acute respiratory distress syndrome. N Engl J Med 342: 136061. Tracey KJ, Beutler B, Lowry SF et al. 1986. Shock and tissue injury induced by recombinant human cachectin. Science 234: 47074. Walsh TS, Lee A. 1999. N-acetylcysteine administration in the critically ill. Intensive Care Med 25: 43234. Ware LB, Matthay MA. 2000. The acute respiratory distress syndrome. N Engl J Med 342: 133449. Wheeler AP, Bernard GR. 1999. Treating patient with severe sepsis. N Engl J Med 340: 20714. Wort SJ, Evans TW. 1999. The role of the endothelium in modulating vascular control in sepsis and related conditions. Br Med Bull 55: 3048. Wyncoll DLA, Evans TW. 1999. Acute respiratory distress syndrome. Lancet 354: 497501. Young JD. 1997. Inhaled nitric oxide in acute respiratory failure. Br J Anaes 79: 69596. Zhang H, Slutsky AS, Vincent JL. 2000. Oxygen free radicals in ARDS, septic shock and organ dysfunction. Intensive Care Med 26: 474 76.

Gattinoni L, Pelosi P, Crotti S et al. 1995. Effects of positive endexpiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med 151: 180714. Gutteridge JMC, Mitchell J. 1999. Redox imbalance in the critically ill. Br Med Bull 55: 4975. Hayes MA, Timmins AC, Yau EHS et al. 1994. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med 330: 171722. Heller A, Koch T, Schmeck J et al. 1998. Lipid mediators in inammatory disorders. Drugs 55: 48796. Herbert PC, Drummond AJ, Singer J et al. 1993. A simple multiple system organ failure scoring system predicts mortality of patients who have sepsis syndrome. Chest 104: 23035. Herbert PC, Wells G, Blajchman MA et al. 1999. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med 340: 40917. Jindal N, Hollenberg SM, Dellinger RP. 2000. Pharmacological issues in the management of septic shock. Crit Care Clin 16: 23348. Kennedy BC, Hall GM. 2000. Metabolic support of critically ill patients: parenteral nutrition to immunonutrition. Br J Anaes 85: 18588. Keogh BF, Bateman CJ. 1999. Lung protective ventilatory strategies will these prevail in the next millennium? Br J Anaes 83: 82932. Kim PK, Deutschman CS. 2000. Inammatory responses and mediators. Surg Clin North Am 80: 88594. Kolkman JJ, Otte JA, Groeneveld ABJ. 2000. Gastrointestinal luminal pCO2 tonometry: an update on physiology, methodology and clinical applications. Br J Anaes 84: 7486. Kumar A, Short J, Parillo JE. 1999. Genetic factors in septic shock. JAMA 282: 57981. Marik PE. 1999. Total splanchnic resuscitation, SIRS, and MODS. Crit Care Med 27: 25759. Marras T, Herridge M, Mehta S. 1999. Corticosteroid therapy in acute respiratory distress syndrome. Intensive Care Med 25: 119193. Maynard ND, Bihari DJ, Dalton RN et al. 1995. Increasing splanchnic blood ow in the critically ill. Chest 108: 164854. Meduri GU, Headley AS, Golden E et al. 1998. Effect of prolonged methylprednisolone therapy in unresolving acute respiratory distress syndrome. JAMA 280: 15965. Messori A, Trippoli S, Vaiani M et al. 2000. Bleeding and pneumonia in intensive care patients given ranitidine and sucralfate for prevention of stress ulcer: meta-analysis of randomised controlled trials. Br Med J 321: 1103 106. Miller PR, Munn DD, Wayne Meredith J et al. 1999. Systemic inammatory response syndrome in the trauma intensive care unit: who is infected? J Trauma, Injury, Infection Crit Care 47: 1004 1008. Pastores MS, Katz DP, Kvetan V. 1996. Splanchnic ischaemia and gut mucosal injury in sepsis and the multiple organ dysfunction syndrome. Am J Gastroenterology 91: 1697709. Pelosi P, Tubiolo D, Mascheroni D et al. 1998. Effects of the prone position on respiratory mechanics and gas exchange during acute lung injury. Am J Respir Crit Care Med 157: 38793. Plank LD, Hill GL. 2000. Sequential metabolic changes following induction of systemic inammatory response in

Trauma 2001; 3: 89100