INCIDENCE

The concept that death from trauma has a trimodal distribution (with these deaths being caused by hemorrhage, head injury, and sepsis/organ failure) is well established. Because MODS is the most common cause of late trauma deaths, it has been the subject of intense investigation. It is now clear that certain clinical risk factors can be used to predict the likelihood of a patient developing MODS. These include age, injury severity score (ISS), number of blood transfusions, and lactate/base deficit levels.² However, it is only over the last decade that the incidence of MODS in high-risk trauma patients appears to have decreased. This decrease appears to be due to a better knowledge of the factors predisposing patients to its development, as well as to the immunoinflammatory response to shock and trauma. For example, a 12-year prospective study examining 1344 trauma patients noted that the actual incidence of MODS (25%) was lower than its predicted rate.³

The authors concluded that this decrease was likely due to the concomitant drop in the liberal use of blood transfusions, which have been shown to be an independent predictor of MODS, SIRS, and mortality.^4,5 Not only does the incidence of MODS appear to be decreasing but there is emerging data to suggest that the mortality rate of patients with MODS is also declining—as reflected in a retrospective study of MODS-related death after blunt multiple trauma during a 25-year period. This study revealed an approximately 50% reduction in MODS-related mortality, from 29% to 14% over this time period.⁶ As will be discussed later in this chapter, several therapeutic interventions have been developed that have been shown to reduce mortality or to attenuate organ dysfunction, which would help explain this decline in mortality. In spite of these improvements, once MODS has become established the risk of death is significant—with the patient’s prognosis being more closely related to the number of organs that have failed than to any other variable, including the underlying processes that initiated the MODS.⁷

MECHANISMS OF MODS

The clinical picture of MOF is indicative of a generalized systemic inflammatory response, which typically occurs as a result of infection or uncontrolled inflammation in the patient with severe trauma. Several distinct and often conflicting hypotheses have been proposed to explain the mechanisms underlying MODS.⁷ Nonetheless, MOF can be viewed as a systemic process involving the excessive stimulation of certain inflammatory responses mediated by circulating factors whose effects contribute to injury or dysfunction in organs not involved in the initial insult. To a large extent, the cascade of events culminating in MOF is likely to be mediated by the same factors irrespective of the exact nature of the triggering insult. In fact, it is the host’s inflammatory response to injury or infection that is probably more important in the genesis of SIRS, ARDS, and MODS than the microbial agent or the initiating insult. Thus, an appreciation of the role of the inflammatory response of the host in the pathogenesis of MOF is vital in order to develop new and effective modalities for the prevention and treatment of this syndrome.

The earliest reports of postinjury MODS identified occult intra-abdominal infection as the etiology in approximately half of the cases. However, the recognition that more than 50%–70% of patients with MOF do not have an identifiable focus of infection meant that uncontrolled infection could not be the universal cause of MODS. From this early work came the recognition that only a fraction of septic-appearing patients were infected and that the host’s own response to tissue injury or shock could result in a noninfectious septic state. In turn, this recognition that the host’s immunoinflammatory response to microbial infection, tissue injury, necrotic tissue, or shock was similar led to the hypotheses that immune cell products, such as cytokines, contributed to the development of MODS.

This hypothesis was based on the concept that an excessive immuno-inflammatory response due to activated macrophages and other immune cells led to cytokine-mediated tissue injury and thereby the development of SIRS and MODS. This hypothesis was supported by several experimental and clinical observations. For example, cytokine levels were increased in trauma patients and the administration of tumor necrosis factor alpha (TNF-α) to humans elicited a clinical response similar to SIRS, whereas preclinical animal studies documented that TNF-neutralization improved survival in animals receiving a lethal dose of endotoxin. However, things were not this simple—as soon became apparent from multiple failed clinical trials of anti-inflammatory agents and the results of more complex preclinical animal studies. In fact, it is now recognized that cytokines have many beneficial functions, such as the control of infection. It is also recognized that elevated cytokine levels appear to be more markers or predictors of the host response than inducers of MODS.

Another mechanism by which hemorrhagic shock and trauma could predispose to the developments of MODS is through an ischemia-reperfusion injury and/or damage to the microcirculation. Because shock is essentially a total-body ischemia-reperfusion insult and the microcirculation of various tissues and organs are highly susceptible to ischemia-reperfusion–mediated insults, this process has been termed the microcirculatory hypothesis of MODS.⁷ Physiologically, circulatory shock could contribute to MOF through inadequate global oxygen delivery, the ischemia-reperfusion phenomenon, and/or the promotion of deleterious endothelial-leukocyte interactions.

Although prolonged tissue hypoxia leads to inadequate ATP generation and potentially irreversible cell damage, under most clinical conditions the shock period is not long enough for this process to occur. Thus, in clinical situations it appears that most of the tissue damage occurs after ischemia is relieved by reperfusion and that this damage is due to the production of reperfusion-induced oxygen radicals and proinflammatory factors (such as oxidants, nitric oxide, chemokines, and cytokines). In fact, recent studies show that the combination of reperfusion-induced increased levels of nitric oxide and superoxide anion synergistically increase cell injury via the production of peroxynitrite, which is a long-lasting and potent oxidant that causes direct cell injury through lipid peroxidation. This notion that increased nitric oxide production is important in the pathogenesis of MODS is supported by clinical studies showing that serum nitrate levels (an index for the systemic production of nitric oxide) correlated well with MOF scores in critically ill patients.⁸

Endothelial-leukocyte interactions leading to tissue injury also seem to be a key step in the pathogenesis of SIRS, ARDS, and MODS. Many factors related to shock and tissue injury, including cytokines, necrotic tissue, endotoxins, and oxidants, can convert endothelial cells from a quiescent state to a proinflammatory procoagulant one and can activate neutrophils. The combination of these changes in endothelial cell phenotype and neutrophil activation has been documented to lead to increased neutrophil adherence to the microcirculatory endothelium, thereby promoting neutrophil-mediated microvascular injury.⁷ Experimentally, inhibition of neutrophil-endothelial interactions has been shown to limit shock- and sepsis-induced injury to a number of organs, including the lung. Furthermore, neutrophil activation in trauma patients has been identified as a predictor of the development of SIRS, ARDS, and MODS. Therefore, endothelial cell–neutrophil interactions, whether induced by shock, sepsis, or an augmented inflammatory response, appear to be an important effector mechanism in the development of ARDS and MODS.

The gut hypothesis of MOF has been used to explain why no identifiable focus of infection can be found in as many as 30% of bacteremic patients who die from MOF.⁹ An extensive body of experimental as well as clinical studies supports this hypothesis. For example, clinical studies indicate that intestinal permeability is increased in patients with sepsis after major thermal injury or trauma and that loss of intestinal barrier function correlates with the development of systemic infection, ARDS, and MODS.¹⁰ Likewise, studies in intensive care unit (ICU) and trauma patients indicated that gut ischemia, as measured by gastric tonometry, is a better predictor of the development of ARDS and MODS than global indices of oxygen delivery.¹¹ Although both clinical and experimental studies implicated intestinal injury and bacterial translocation in the development of SIRS and MODS, a study by Moore et al.¹² began to cast doubt on the clinical relevance of bacterial translocation.

These investigators failed to find bacteria or endotoxin in the portal blood of severely injured patients, including a subgroup of patients developing MODS. One potential explanation for this failure to find endotoxin or bacteria in the portal blood was that the gut-derived factors contributing to SIRS, ARDS, and MODS were exiting the gut via the lymphatics. Studies testing this possibility have documented that nonbacterial factors exiting the ischemic gut contribute to acute ARDS, MODS, neutrophil activation, and endothelial cell injury/activation in both rodent and primate models of trauma-hemorrhagic shock and have led to the gut-lymph hypothesis of MODS.¹⁰ This gut lymph hypothesis of MODS proposes that nonbacterial noncytokine factors released from the stressed gut via the lymphatic system activate neutrophils and endothelial cells, thereby leading to organ dysfunction. Thus, over the last several years the gut hypothesis has expanded beyond bacterial translocation and now also implicates gut-derived nonbacterial proinflammatory and tissue-injurious factors in the pathogenesis of SIRS, ARDS, and MODS.

Although each of the various MODS hypotheses was presented individually, in patients many of these pathways overlap and the induction of one pathway can lead to activation of others. For example, severe bacterial infection activates the immuno-inflammatory response, which in turn leads to microcirculatory dysfunction and gut ischemia. Likewise, nonbacterial gut-derived factors have been shown to activate neutrophils, lead to an augmented inflammatory response, and promote microcirculatory dysfunction. Furthermore, shock states are associated with microcirculatory failure, gut injury, induction of an inflammatory response, and augmented neutrophil-endothelial cell interactions. In addition, in many patients who develop MOF no one major insult seems to have occurred. Instead, it appears that the development of MODS is related to the summation of several minor insults rather than one major event.

This clinical observation has led to the “two-hit” hypothesis of MODS, where potentially clinically modest events prime the host so that the host’s response to subsequent secondary events becomes exaggerated, culminating in SIRS, ARDS, and MOF. Although this two-hit theory needs to be further understood, it is a feasible explanation of how trauma or burn injury can convert a nonlethal infectious or hypoxic challenge into a lethal insult. In fact, as illustrated in Figure 1 it is clear that the difficulty in finding an effective therapy to prevent or treat MODS relates to the overlapping nature of the multiple systems activated by shock and trauma as well as the ability of one system to prime other systems for an exaggerated physiologic response to secondary insults. Nonetheless, the knowledge gained from these basic studies of the physiology of inflammation and MODS have provided important therapeutic insights. For example, they highlight the importance of prompt and adequate volume resuscitation and microcirculatory blood flow to prevent organ ischemia, the need for early excision of nonviable tissue to limit systemic inflammation, and the need for therapies to better preserve gut barrier dysfunction and limit uncontrolled inflammation.

Figure 1 Overlapping nature of multiple hypotheses believed to be responsible for causing MODS.

DIAGNOSIS

A key step in the treatment of a disease process is the establishment of an accurate diagnosis. To that end, a number of consensus conferences have been held in an attempt to provide classification schemes that allow SIRS, ARDS, and MODS to be accurately diagnosed. Based on these conferences, SIRS is defined as the response to a variety of severe clinical insults, which is manifested by two or more of the four conditions listed in Table 1.¹³ Furthermore, SIRS should be viewed as an evolved dynamic process that has adaptive survival value for the host under most circumstances because it signals the body to respond to injury or to an external threat such as a bacterial infection. However, if this protective inflammatory response becomes uncontrolled or excessive it has maladaptive consequences due to its potential to injure the host’s own tissues.

Table 1 Definition of Systemic Inflammatory Response Syndrome

The term “MODS” was introduced by a consensus conference of the American College of Chest Physicians (ACCP) and the Society of Critical Care Medicine (SCCM) in 1991.¹⁴ Prior to that time, this syndrome had many different names, including sequential organ dysfunction syndrome and multiple-organ failure syndrome. Several MODS scoring systems have been established that grade the severity of MODS and emphasize the concept that there exists a continuous spectrum from mild to full-blown dysfunction that correlates, on a patient population level, with mortality and morbidity (Table 2).^15–18

These systems, much like the sequential organ failure assessment (SOFA) score developed by Vincent et al.,¹⁸ score organ failure by assigning a numerical scale in which more points are given to the higher degree of organ dysfunction in several organ systems (Table 3).^16,19 Although not developed to predict mortality in individual patients, there are several areas where the use of such scoring systems can be beneficial in critically ill patients. This includes their use in the daily clinical evaluation of a patient’s response, in research involving epidemiologic studies, and in the assessment of new therapies in clinical trials. Although these scoring systems use slightly different parameters to grade organ failure, most studies have found that the clinical utility of these scoring systems is comparable.^20,21

MANAGEMENT