The effect of evolving trauma care on the development of multiple organ dysfunction syndrome

Abstract

Introduction

Multiple organ dysfunction syndrome (MODS) is still a major threat to polytrauma patients, since sepsis-related organ failure is the most common cause of late mortality in these patients. In this article, the development of trauma surgery and evolution of trauma care from early total care to damage control surgery is discussed. Increasing knowledge of the pathophysiology of trauma has enabled us to identify the inflammatory response induced by trauma. By understanding the pathophysiology, we may be able to fully comprehend the origin of multiple organ dysfunction related sepsis. Further, it is important to appreciate the influence of surgery on the inflammatory response induced by trauma, and subsequently on the development of inflammatory complications. It is crucial to offer the polytrauma patient the appropriate type of surgery at the right time to prevent further deterioration.

Conclusion

MODS is still highly lethal, and once it has developed it is difficult to treat, so it is vital to be able to predict its occurrence. If we knew how to predict MODS, we might be able to develop strategies to prevent this syndrome.

Introduction

Even though many advances in prehospital care and trauma resuscitation have enabled the early survival of severely injured patients, multiple organ dysfunction syndrome (MODS) is still a major threat to these patients, since sepsis-related organ failure is the most common cause of late mortality. The incidence of postinjury MODS has been reported to be between 7 and 66 %, with an associated mortality rate of between 31 and 80 % [1]. Over the years, the incidence, severity, and mortality of postinjury multiple organ failure has decreased despite an increased risk of MODS. Improvements in MODS outcomes can be attributed to several improvements in trauma and critical care, although organ failure remains a major cause of intensive care unit (ICU) resource use and mortality after injury [2, 3].

This article discusses the development of trauma surgery, the pathophysiology of trauma, and the influence of evolving trauma care on the development of inflammatory complications such as multiple organ failure. Further, future strategies for MODS prediction and prevention will be discussed.

Development of trauma surgery: from early total care to damage control surgery

In the 1970s and 1980s, studies reported that immediate stabilization of femur fractures considerably reduced problems with traumatic pulmonary failure and postoperative care when compared to traditional nonoperative fracture management. Thereafter, early total care for multi-trauma patients was said to represent the optimal treatment for the patient with multiple orthopedic injuries, and the benefits of this approach were demonstrated in many studies [4, 5]. However, the surgical burden of early total care increased the risk of postoperative complications such as adult respiratory distress syndrome (ARDS) and multiple organ dysfunction. Further, the physiologic derangements of massive hemorrhagic shock, such as coagulopathy, hypothermia, and acidosis, often led to the premature death of the patient. To prevent this derangement, the concept of “damage control surgery” was developed by Rotondo et al. [6] in 1993. The concept was defined as initial control of hemorrhage and contamination followed by intraperitoneal packing and rapid closure. This allowed resuscitation to normal physiology in the ICU and subsequent definitive re-exploration. Damage control surgery was initially used in exsanguinating penetrating abdominal injury, and it was noticed that survival was markedly improved in patients treated according to damage control surgery strategies. Since then, damage control has gained widespread use in all polytrauma patients [7–9].

Pape et al. [10] noticed that there was a higher incidence of posttraumatic ARDS and mortality in patients with severe chest trauma when early intramedullary femoral nailing was performed. He developed, in analogy to damage control surgery, the concept of initial temporary fixation and secondary conversion to a definitive procedure in orthopedic surgery [11]. This damage control orthopedics (DCO) approach has been shown to be an adequate alternative for patients at high risk of developing posttraumatic systemic complications such as ARDS and MODS, and has gained wide acceptance [12].

Pathophysiology of SIRS, ARDS, and MODS

In the 1980s, Goris et al. [13] hypothesized that ARDS and MODS were not a result of infectious diseases, but were instead initiated by a systemic inflammatory response induced by the immune system as a reaction to tissue damage. This theory was later confirmed by others; tissue damage results in activation of the innate immune system, which can lead to a systemic inflammatory response syndrome (SIRS) with an increased risk of dangerous inflammatory complications such as acute respiratory distress syndrome (ARDS) and multiple organ dysfunction syndrome (MODS) [14, 15]. Polymorphonuclear granulocytes (PMNs) are among the most important cells of the innate immune system. They play an essential role in the early inflammatory phase after injury, recognizing and responding to pathogens and providing immediate defense against infection. The function of PMNs is to neutralize bacterial pathogens using mechanisms employing reactive oxygen species (ROS) and cytotoxic proteins. In response to local inflammation, activated endothelial cells and innate immune cells such as macrophages release pro-inflammatory mediators such as interleukin (IL) 1β, IL-6, tumor necrosis factor (TNF)-α, and platelet activation factor (PAF). Interaction with these mediators and bacterial components (such as lipopolysaccharide, LPS) results in pre-activation of neutrophils (priming). Priming is necessary before a stimulus can prompt full activation [16–18], and is mediated by danger-associated molecular patterns (DAMPs) and pathogen-associated molecular patterns (PAMPs). Both DAMPs and PAMPs stimulate PMNs to release proteases and free oxygen radicals into the environment [19]. DAMPs are nuclear or cytosolic proteins that are released outside the cell or exposed on the surface of the cell following tissue injury. They act as endogenous danger signals to promote and exacerbate the inflammatory response. A similar systemic inflammatory response is seen in sepsis [20]. In sepsis, however, the immune system is activated by PAMPs liberated from invading microbes. Both DAMPs and PAMPs are recognized by pattern recognition receptors (PRRs), among which Toll-like receptors (TLRs) are the most well known. TLRs, expressed by neutrophils for example, can detect the threat from microbes and tissue damage, and start the required responses [21, 22]. Upon activation, PMNs marginate and adhere to the endothelial wall. Selectins expressed on the endothelial cells (E- and P-selectin) and leukocytes (L-selectin) are responsible for the “rolling behavior” of PMNs. Firm adhesion to the endothelial wall is dependent on upregulation of integrins such as CD11b/CD18 (macrophage antigen-1 or Mac-1) [23]. Integrins adhere to an endothelial- and leukocyte-associated transmembrane protein (such as intercellular adhesion molecule-1, ICAM-1), facilitating leukocyte endothelial transmigration. After adhesion to the endothelial wall, PMNs migrate through the endothelial layer, infiltrating and accumulating in different and uninvolved tissues—a process called “homing” (Fig. 1) [23, 24]. This is thought to be important for the development of secondary organ and tissue damage [16, 25].

Fig. 1

The inflammatory response after injury. In response to local inflammation, activated endothelial cells and innate immune cells release pro-inflammatory mediators such as interleukin (IL)-1 beta, IL-6, and tumor necrosis factor (TNF)-alpha. Interaction with these mediators and bacterial components results in the priming of neutrophils (PMNs). Priming is necessary before a stimulus can prompt full activation, and it is mediated by danger-associated molecular patterns (DAMPs) and pathogen-associated molecular patterns (PAMPs). Both DAMPs and PAMPs stimulate PMNs to release proteases and free oxygen radicals into the environment, and are recognized by Toll-like receptors (TLRs). TLRs, expressed by neutrophils, can detect the threat from microbes and tissue damage and start the required responses (inset). Upon activation, PMNs marginate and adhere to the endothelial wall. Selectins expressed on the endothelial cells (E- and P-selectin) and leukocytes (L-selectin) are responsible for the “rolling behavior” of PMNs. Integrins adhere to an endothelial- and leukocyte-associated transmembrane protein (such as intercellular adhesion molecule-1, ICAM-1), facilitating leukocyte endothelial transmigration. After adhesion to the endothelial wall, PMNs migrate through the endothelial layer, infiltrating and accumulating in different and uninvolved tissues

Excessive activation of PMNs leads to enhanced mobilization and migration to the tissues, and can contribute to tissue damage that is not necessarily caused by the injury. This immune response is referred to as systemic inflammatory response syndrome (SIRS). The overwhelming accumulation of neutrophils in the organ parenchyma may lead to organ failure and conditions such as ARDS and MODS [14, 15]. Excessive inflammatory response is not only associated with organ dysfunction early after injury; it is also associated with an increased risk for septic complications and organ dysfunction in a later phase, also known as compensatory anti-inflammatory response syndrome (CARS) [15, 24, 26]. CARS is associated with inactivation and/or exhaustion of PMNs and leads to a decreased immune defense, which allows microorganisms to breach the immune system and invade the body during the period of reduced surveillance by the immune system. During this period, patients are prone to develop sepsis, possibly followed by septic shock [15]. Lately, pro- and anti-inflammatory responses have been described as occurring simultaneously [27]. Hietbrink et al. [28] have demonstrated this phenomenon in severely injured patients: PMNs in the peripheral blood had an impaired functionality, whereas pulmonary PMNs showed a fully primed phenotype, suggesting that primed cells are likely to leave the circulation for the tissues, leaving partially refractory cells behind in the circulation. This migration could make the patient more susceptible to infectious complications later.

Influence of evolving surgical strategies on the development of MODS

Effect of surgery on the inflammatory response

Both pro- and anti-inflammatory cytokines are induced by surgical trauma [29]. The imbalance between pro- and anti-inflammatory cytokines following surgery after trauma induces immunological dysfunction, resulting in postoperative complications. Several researchers have attempted to determine the impact of surgery on the patient’s condition [4, 30, 31]. Damage control laparotomy (DCL) with abdominal packing, generally considered a life-saving procedure in polytrauma patients, was shown to induce an inflammatory response itself, and may contribute to septic and organ failure complications [32]. Pape et al. [31] compared the effects of surgery in both isolated fracture patients and polytrauma patients by measuring pro-inflammatory cytokines as a measure of systemic inflammatory response. In their study, polytrauma patients showed higher cytokine levels than isolated fracture patients after initial trauma. Further, both polytrauma patients and patients with isolated fractures showed an extra increase in cytokines (on top of the initial response) proportional to the magnitude of surgery. They concluded that pro-inflammatory cytokine measurements can be of value when assessing a patient’s perioperative inflammatory status. They also randomized severely injured patients with femur fractures, who were graded as patients with borderline physiology, either by primary intramedullary nailing or by external fixator and secondary intramedullary nailing. Serum IL-6 levels after the primary and secondary interventions were analyzed. While intramedullary nailing directly after trauma caused a large increase in circulating IL-6 in addition to already preoperatively elevated levels due to the accidental trauma, the external fixator did not cause any additional IL-6 production. In contrast, secondary intramedullary nailing 5 days after trauma caused only a minor increase in serum IL-6 [33]. Damage control orthopedic surgery reduced the impact of the second hit, and Pape et al. [34] suggested that DCO should be used in polytrauma patients who are at risk of developing systemic complications such as MODS.

Who will benefit from damage control (orthopedic) surgery?

One of the most frequently debated topics is deciding whether to use DCO or early total care (ETC). A prospective, randomized interventional trial carried out in 10 level-1 trauma centers investigated whether temporary fracture fixation would be beneficial for certain patient groups. The study illustrated the clear advantage of ETC in stable polytrauma patients in terms of a shorter ICU stay and less time on the ventilator. In patients with borderline physiology, ETC was associated with a higher incidence of acute lung injury, and the authors concluded that the damage control approach should be followed in borderline patients [35]. Lichte et al. [36] demonstrated similar findings in their investigation of the efficacy of damage control orthopedics: DCO was beneficial for borderline and unstable patients in comparison to early total care.

In contrast, Nicholas et al. [37] stated that DCO is overused in many centers, because indications for DCO established in the early 1990s are not appropriate anymore due to improvements in trauma and critical care. They compared their polytrauma patients with borderline physiology with the results of the prospective study conducted by Pape et al. [35]. They showed that early total care of femur fractures is safe in polytrauma patients with both stable and borderline physiologies. These findings are in line with a retrospective study which showed low rates of ARDS and mortality after reamed intramedullary nailing of femoral fractures in polytrauma patients. The authors performed reamed nailing only after adequate resuscitation had been shown by normalizing lactate plus optimized ventilatory and hemodynamic parameters. Damage control orthopedics with primary external fixation was reserved for patients who did not respond to resuscitation. They concluded that ETC was safe even in severely injured patients, and that DCO was rarely needed [38].

Even though damage control laparotomy (DCL) is generally considered a life-saving procedure in polytrauma patients, there is little evidence that supports the efficacy of damage control surgery with respect to traditional laparotomy in patients with major abdominal trauma. A Cochrane database systematic review conducted in 2010 and repeated in 2013 revealed no randomized controlled trials comparing damage control surgery with immediate and definitive repair in patients with major abdominal trauma [39, 40].

Timing of secondary surgery after damage control (orthopedic) surgery

Although using the DCO concept seems to decrease the systemic inflammatory response, the concept does warrant secondary surgical procedures. The timing of these secondary operations in severely injured patients is difficult and has not yet been specified. Retrospective analysis revealed that the timing of secondary surgery after severe injury can influence the incidence of multiple organ dysfunction after trauma. A retrospective analysis of more than 4000 severely injured patients revealed a higher incidence of multiple organ failure in patients subjected to surgery during days 2 and 4 than in those undergoing surgery later than the sixth day after trauma [41]. Multiply injured patients who were operated on during a period of increased inflammatory response developed postoperative organ dysfunction more often than those with normal homeostasis. A prospective trial showed that early secondary surgery (days 2–4 after trauma) in severely injured patients was associated with a higher incidence of organ failure than operations on day 6 after trauma (late secondary surgery). The surgery-induced increase in serum IL-6 was significantly higher in the early surgery group. In addition, high IL-6 levels at the time of admission were only predictive for subsequent organ failure in the early secondary surgery group [42]. These data point to the idea of an “immunological vulnerable phase” for secondary operations in multiply injured patients. Intervention during this vulnerable phase of dysregulated immune response may lead to an overshooting inflammatory response or an increase in compensatory anti-inflammatory reactions with subsequent infectious complications. Ideally, an operation should be performed after normalization of the immune response. However, the immune system of a trauma patient remains vulnerable for a long period after injury, as shown in a study by Flohé et al. [43]. Analysis of cellular immune functions after secondary surgery in patients primarily treated with the DCO concept showed that, even after 17 days, major secondary surgery such as intramedullary nailing or stabilization of pelvis fractures caused a suppression of immune functions. This long period of susceptibility to immunological problems raises issues regarding the local fracture treatment. Swelling in the area of surgical approach or problems with fracture reduction in secondary treatment and the risk of pin-track infection after primary external fixation have to be taken into consideration.

Comparable problems are seen after damage control laparotomy. Although damage control with abdominal packing is often life-saving, the resulting open abdomen is associated with serious complications, and the management of an open abdomen itself is challenging [44–49]. Further, there is little data on how long abdominal packs can be left in situ. A retrospective study by Abikhaled [50] suggested that the packs should be removed or repacked within 72 h to minimize septic complications. The most effective way to prevent or reduce complications is to close the abdominal wall as soon as possible, ideally within 5–7 days [51]. This is, however, not always feasible.

Prediction of MODS

Multiple organ failure is still associated with high mortality rates, and once organ failure has developed it is difficult to treat, and its treatment is mainly supportive. In order to prevent MODS in the future, it is important to be able to predict its occurrence. Several researchers have established clinical predictors for MODS development. In 1998, Sauaia et al. [52] created a prediction model that identified shock parameters, injury severity score (ISS), age, and platelet count as predictors of MODS. Consequently, an effort has been made to minimize the influence of shock by focusing on early hemostatic resuscitation. Since then, the independent predictors of MODS have fundamentally changed, with ISS and shock parameters no longer predicting MODS, and the incidence of the syndrome has further decreased. Recently, Dewar et al. [3] have shown that MODS can now be predicted at admission based on age and platelet count on admission, with the prediction power improving at 24 h with the addition of creatinine and bilirubin [3].

General markers of inflammation, including cytokines, have been investigated as prognostic markers for posttraumatic complications. Unfortunately, they have been found to be of little use due to low sensitivity and specificity [53].

A promising method of predicting MODS development may be the determination of immune cell functionality. Hietbrink et al. [54] showed that the extent of the sustained injury and the subsequent cellular innate immune response is reflected in changes in a functional PMN phenotype of N-formyl-methionyl-leucyl-phenylalanine (fMLP)-induced active FcγRII (CXC-chemokine receptor 1) in the peripheral blood. Further, they demonstrated that phenotyping blood PMNs enables the kinetics and magnitude of the initial systemic inflammatory response after injury to be identified. The decreased functionality of PMNs reaches its minimum before sepsis development, and could be an important contributing factor in the early identification of patients at risk [28].

Another method of predicting MODS has been demonstrated by Morrison et al. They showed that in trauma patients with hemorrhagic shock, peripheral blood PMN apoptosis in the early resuscitative period is associated with a decreased incidence of subsequent infections [55]. Patients with MODS have been shown to have delayed PMN apoptosis, enhanced oxidative burst activity, and increased end-organ sequestration [56]. The risk of tissue injury and susceptibility to infection could possibly be reduced if the PMNs could be removed from the circulation before end-organ infiltration [55].

Treatment strategies to prevent MODS

Therapeutic strategies such as damage control surgery have been implemented to prevent secondary damage and further deterioration in severely injured patients. Even though the incidence of MODS has decreased over the years, it is still a deadly syndrome. To be able to prevent MODS in the future, several methods have been developed to modulate the inflammatory response. Spruijt et al. [57] conducted a systematic review exploring the effect of immunomodulative interventions on infection, organ failure, and mortality in trauma patients. They showed that although several interventions led to a decrease in inflammatory response, only the administration of immunoglobulin, IFN-γ, or glucan was efficacious at reducing infection and/or mortality.

A large retrospective analysis with patients with major injuries showed that there was less organ failure when cryoprecipitate instead of fresh frozen plasma (FFP) was administered [58]. A possible explanation for this might be that cryoprecipitate contains large amounts of fibrinogen compared to FFP. An in vitro study showed that soluble fibrinogen might limit collateral inflammatory damage by protecting endothelium from activated leukocytes, and therefore contribute to the resolution of inflammation. Pillay et al. [59] hypothesized that low physiologic concentrations of soluble fibrinogen are important for the initiation of inflammation (through the deposition of fibrin), and that the late increase in soluble fibrinogen as an acute-phase protein contributes to the resolution of inflammation. This hypothesis was confirmed in a murine model of intravenous endotoxin, in which the inflammatory responses were delayed in mice lacking fibrinogen, but peak values of cytokines and cell influxes into tissues were higher than in wild-type mice [60]. Fibrinogen might be of value in preventing further deterioration of the patient by attenuation of the inflammatory response.

A promising intervention is treatment with C1-esterase inhibitor (C1-INH). C1-INH is an acute-phase protein produced by the liver that is important in regulating complement and contact activation, which play a role in opsonization and the regulation of coagulation. Currently, C1-INH is applied in patients with hereditary angioedema [61]. Several animal studies have demonstrated that high levels of C1-INH preserve endothelial function, prevent adhesion of leucocytes to the endothelium and capillary leak, and improve survival [62–64]. Dorresteijn et al. [65] demonstrated in a human LPS study that C1-INH shows anti-inflammatory effects in the absence of classic complement activation. Currently, the effect of C1-esterase inhibitor on systemic inflammation in trauma patients with a femur or pelvic fracture is being tested in a randomized controlled trial [66].

Discussion

In conclusion, a better understanding of the pathophysiology of trauma has contributed to the improvements made in trauma care over the last few decades [6–11]. Damage control and hemostatic resuscitation strategies have played a role in the decreases in the incidence and mortality of multiple organ dysfunction. However, damage control (orthopedic) surgery has its downsides as well, and timing of surgery is essential. Care must be taken to select the right patient who will benefit from the damage control concept [32, 35, 37–40, 44–50].

Further, early prediction of MODS will add to the decrease in the incidence of MODS. Clinical predicting parameters have been developed and improvements in trauma care have changed these parameters [3, 52]. Recently, promising methods of determining the functionality of immune cells have been developed as a means to predict MODS development [54, 55]. These methods may be implemented in daily trauma care in the future.

Treatment strategies to prevent MODS need to be developed. Many studies have been conducted to investigate immunomodulative interventions. Even though several interventions have led to a decrease in the inflammatory response, only a few studies have been shown to be efficacious at reducing infection and/or mortality [56]. To develop successful methods that are able to alter the inflammatory response, it is likely that these processes must influence more than one step in the cascade of the complex pathophysiological mechanism of systemic inflammation.