72 Infection, sepsis, and multiorgan dysfunction syndrome

Sepsis is the main cause of multiple organ failure and remains a concern because of the associated high morbidity and mortality. In recent years, important advances have been made in the understanding of the pathophysiology of sepsis. Sepsis and septic shock are the end result of complex interactions between infecting organisms and various elements of the host response. A key feature of the common sequence of organ failure is dysfunction of the cardiovascular system, including microcirculatory elements.

Outcome improvement in sepsis is based on recognizing the process early and instituting effective therapies. The time window for intervention is relatively short, and treatment must promptly control the source of infection, restore haemodynamic homoeostasis, and support failing organ systems. Thereafter, a de-escalation strategy should be used (withdrawal of catecholamines, negative fluid balance, etc.), timed according to individual patient characteristics and response to treatment.

Sepsis is the main cause of multi organ failure (MOF) and remains a concern, because of the associated high morbidity and mortality [1]. The management of sepsis is based on recognizing the process early and rapidly instituting effective therapies.

Sepsis is defined as a dysregulated host response to infection, responsible for some degree of organ dysfunction (see Figure 72.1) [2]. Septic shock refers to a state of acute circulatory failure, characterized by altered tissue perfusion, usually associated with arterial hypotension despite adequate fluid administration, so that vasopressor therapy is necessary to restore an acceptable arterial pressure. Hypotension is usually defined by a systolic arterial pressure of <90 mmHg, or a reduction of >40 mmHg from the baseline, and is associated with signs of altered tissue perfusion such as oliguria, altered mental status, or altered skin perfusion. The diagnosis is confirmed by increased blood lactate levels (typically above 2 mEq/L), reflecting abnormal O₂ metabolism. Serum lactate levels are also independently associated with mortality [3].

Sepsis and septic shock are major health care problems, affecting millions of individuals around the world each year, with mortality rates of 25–40% [4, 5]. Sepsis and septic shock are the tenth most common cause of death in the US, and the incidence of sepsis may even be increasing [5], with mortality rates decreasing [6], although this may be largely a reporting phenomenon, with less severe cases being included. Associated costs are also considerable, with one study estimating the total national hospital cost invoked by sepsis in the US at approximately $24.3 billion [6].

The pathophysiology of sepsis is complex involving multiple cellular interactions and mediators. Early responses to infectious microorganisms involve pathogen-associated molecular patterns (PAMPs) and pattern recognition receptors (PRRs), which trigger intracellular signalling systems with release of inflammatory mediators and early activation genes [5]. Effects are widespread and combined lead to altered coagulation, increased endothelial permeability, impaired cellular function, altered microcirculatory flow, among other effects; the end result is organ failure. Importantly, because of the global nature of the sepsis response and the effects of organ-organ crosstalk, most patients with sepsis will have dysfunction of multiple organs simultaneously; individual organ failure is rare.

Sepsis-induced vasodilation is quite variable in different vascular beds. The autoregulatory mechanisms that control the perfusion of the microcirculation, via multiple neuroendocrine, paracrine, and mechanosensory pathways, are altered in sepsis. Normally, these mechanisms adapt to the balance between the loco-regional tissue O₂ transport and metabolic needs to ensure that supply matches demand. Alteration of these processes in sepsis leads to blood flow heterogeneity within organs.

The coronary response is vasodilation with an increase in the coronary blood flow, even in septic shock. The mesenteric circulation is typically altered, as a result of increased sympathetic tone and renin–angiotensin activation inducing vasoconstriction of the mesenteric territory [7], whereas downregulation of endothelial nitric oxide synthase (eNOS) can decrease the endothelium-dependent vasodilation [8]. β-adrenergic stimulation increases mesenteric perfusion in animal models [9] and in septic shock patients [10]. Hepatic blood flow has been shown to be diminished or increased, depending on the model. In humans, the hepatic flow generally increases, in parallel with cardiac output [11], but hepatic venous O₂ saturation (ShO₂) can decrease due to an increase in oxygen demand.

Renal blood flow is generally increased in hyperkinetic models, but the repartition between the medullar and cortical flows is altered, probably largely because of an increased release of nitric oxide (NO) [12].

Cerebral blood flow is typically preserved, even in hypokinetic models. However, autopsy findings suggest ischaemic and apoptotic lesions in some patients with septic shock [13].

During systemic inflammation, endothelial damage results in the expression of tissue factor (TF) and platelet activation., with the subsequent release of thrombin, which converts fibrinogen into fibrin, induces release of pro-inflammatory cytokines and growth factors, and is involved in multiple other coagulation-related reactions [5]. There is also downregulation of physiological anticoagulant mechanisms. The end result is a hypercoagulable state with fibrin deposition and thrombus formation.Ultimately, clotting factors are depleted and disseminated intravascular coagulation (DIC) may develop; this indicates severe disease and is associated with worse outcomes [14].

Initial descriptions of the cardiovascular alterations in septic shock included a distinction between a hyperdynamic state, with full bounding pulses, flushing, fever, oliguria, and hypotension (so-called ‘warm’ shock), and a less common hypodynamic state with clammy, pale extremities and low-volume pulses (so-called ‘cold’ shock) [15]. The latter pattern is typically related to volume under-resuscitation and sometimes profound myocardial depression. With the ability to measure cardiac output and other haemodynamic variables at the bedside, the view of septic shock as a typically hyperdynamic state emerged.

Under conditions of adequate volume resuscitation, the profoundly reduced systemic vascular resistance (SVR), typically encountered in sepsis, is associated with a high cardiac index that obscures the myocardial dysfunction that also occurs.

Although a number of mediators and pathways has been shown to be associated with myocardial depression in sepsis, the pathophysiology is multifactorial and involves multiple pathways (see Figure 72.2) [16].

A provocative theory regarding the myocardial depression in sepsis suggests that it may play a protective role in the heart, similar to the phenomenon of hibernation in coronary ischaemia [17]. Myocardial depression may represent a protective adaptation by reducing the cellular energy expenditure in the heart during a situation of decreased energy production.

Early theories of myocardial dysfunction in sepsis included the hypothesis of global myocardial ischaemia. However, this theory was dismissed, based on studies in animals and humans showing that septic patients have a high coronary blood flow and a diminished coronary artery–coronary sinus O₂ difference [18]. Also there is no evidence supporting global ischaemia as an underlying cause of myocardial dysfunction in sepsis.

However, in septic patients with co-existent, and possibly undiagnosed, coronary heart disease (CAD), regional myocardial ischaemia or infarction secondary to CAD can occur. The manifestation of myocardial ischaemia due to CAD might even be facilitated by tachycardia, as well as by generalized microvascular dysfunction in sepsis [19]. Additional CAD-aggravating factors encountered in sepsis are inflammation and the activation of the coagulation system that may promote thrombosis of the coronary circulation.

The absence of significant myocardial cell death and the reversible nature of myocardial dysfunction in sepsis support a prominent role for functional, rather than anatomical, abnormalities in the underlying pathophysiology.

Many sepsis mediators can have myocardial depressant effects, including NO, tumour necrosis factor alpha (TNF-α), IL-1, platelet-activating factor (PAF), toll-like receptors, and oxygen free radicals.

NO is produced by all types of cardiac cells. NO exerts a plethora of biological effects in the cardiovascular system [21]. NO has been shown to modulate cardiac function under physiological and a multitude of pathophysiological conditions. The effects of NO relevant to sepsis-induced myocardial dysfunction include vasodilation, depression of mitochondrial respiration, and enhanced release of pro-inflammatory cytokines, which may exert their own cardiovascular effects [22]. NO is produced from the conversion of L-arginine to L-citrulline by NO synthase (NOS). NOS has two forms: constitutive (cNOS) and inducible nitric oxide synthase (iNOS). Early myocardial dysfunction in sepsis may occur through the overproduction of NO and the resultant cGMP through cNOS activation in cardiac cells. Through the expression of iNOS, the massive release of NO contributes to myocardial dysfunction, in part through the generation of cytotoxic peroxynitrite, a product of the reaction of NO with the superoxide anion [23].

NO can also contribute to the alterations in myocardial mitochondria during sepsis [24, 25]. Increased levels of NO and superoxide can lead to the inhibition of oxidative phosphorylation and decreased production of ATP. This ‘cytopathic hypoxia’, the inability of cells to utilize O₂ and produce adenosine triphosphate (ATP), can contribute to the development of MOF in sepsis [26, 27]. The role of mitochondrial dysfunction in sepsis-induced myocardial depression is further supported by animal studies showing that inhibition of this process can improve cardiac function and reduce mortality [28].

Cellular apoptosis is also involved in sepsis-induced cardiovascular dysfunction [29]. Increased release of various substances implicated in the initiation of apoptosis, such as caspases and mitochondrial cytochrome c, has been shown in sepsis. Therapeutic strategies aimed at the inhibition of apoptosis have resulted in improved cardiovascular function in animal models of sepsis [29, 30]. However, the natural evolution of sepsis-induced myocardial dysfunction with recovery of cardiac function in survivors (after 7–10 days) suggests perhaps a less critical pathogenic role for apoptosis.

Echocardiographic evaluation at the bedside can be very useful in the evaluation of sepsis-related myocardial dysfunction, as it can reveal right ventricular (RV) or left ventricular (LV) systolic dysfunction, as well as LV diastolic dysfunction. Many septic shock patients are treated with mechanical ventilation, so that transoesophageal echocardiography (TOE) is the preferred approach, although technical improvements enable good images to be obtained using transthoracic views, even in mechanically ventilated patients. The incidence of cardiac dysfunction in septic shock varies between 35% and 60%, depending on the criteria used for definition [31, 32], and is associated with increased mortality rates [33]. The development of acute ventricular dilation is still debated [34]. However, it seems that acute LV dilation is observed only when septic myocardial injury results in acute LV systolic dysfunction, defined by reduced ejection fraction (EF), reduced stroke volume, or both [34]. Whether dilation occurs or not, LV dysfunction is transient and entirely reversible in surviving patients, lasting typically <1 week.

The majority of cardiac troponin T (cTnT) and cardiac troponin I (cTnI) is bound to myofilaments, and the remainder is free in the cytosol. When myocyte damage occurs, the cytosolic pool is released first, followed by a more protracted release from stores bound to deteriorating myofilaments. Direct effects of endotoxin, oxidative stress, and cytokines, or their effects on downstream pathways, can reduce myocyte integrity, explaining an increase in plasma troponin levels. For example, the executioners of apoptotic pathways, caspases, can induce sarcomere disarray and cleave α-actin, α-actinin, and troponin T [35]. The increase in blood troponin levels may result from myocardial ischaemia secondary to microcirculatory abnormalities, but increased myocyte permeability may also be involved. Cytokines, such as TNF-α or IL-6, may be good candidates to explain this phenomenon, because they can induce increased membrane permeability in vitro and their plasma levels have been found to be more elevated in troponin-positive than troponin-negative patients [36]. Elevation of troponin levels is correlated with a worse outcome [37].

Sepsis and septic shock are characterized by a decrease in vascular reactivity. Both vasodilation (endothelium-dependent and independent) and vasoconstriction (receptor-dependent and independent) mechanisms are altered. The principal mechanisms that explain the altered vascular tone in sepsis are the activation of iNOS, with massive release of NO [38], increased levels of prostaglandins [39], and hyporeactivity to catecholamines due to relative adrenal insufficiency in some patients and vasopressin deficiency [40]. Oxidative stress can also accelerate the degradation of catecholamines.

The microcirculation, a network of vessels of <150 microns in diameter, comprising arterioles, capillaries, and venules, is critical for supplying tissues with substrates and removing metabolites. The microcirculation is the principal site of O₂ exchange between blood and underlying tissues, and there is profound disruption of microcirculatory homeostasis in sepsis [41, 42].

Microcirculatory alterations are a hallmark of the maldistributive defect characterizing sepsis. After aggressive resuscitation of the septic patient, a normal or high cardiac output is typically achieved, yet tissue perfusion can remain markedly impaired. Clinically, this may manifest with persistent acidosis, mottled skin, or progressive MOF. Limiting goal-directed resuscitation solely according to macrocirculatory perfusion indices (e.g. cardiac filling pressure, mean arterial pressure (MAP), cardiac output, or even mixed/central venous O₂ saturation) may therefore not be sufficient to optimize blood flow to the tissues in many patients.

It has been proposed that the failure of the microvasculature is the key event that leads to MOF [43]. The microcirculatory unit is the landscape where most of the pivotal events of sepsis pathogenesis take place, including the loss of vasomotor reactivity, endothelial cell injury leading to capillary leakage and interstitial oedema, which increase the diffusion distance of O₂ to cells, activation of coagulation, and disordered leucocyte trafficking. Stiff leucocytes and red blood cells, platelet/fibrin clots, and endothelial cell swelling are thought to be involved in microcirculatory occlusion.

The activation of circulating mature leucocytes or the release of immature leucocytes from the bone marrow may result in a population of less deformable cells that can be entrapped in the capillaries [41]. The decrease in red blood cell deformability may also contribute to the loss of perfused capillaries [44]. DIC can be involved in the decrease in perfused capillaries. Finally, capillary endothelial swelling and the formation of pseudopod extensions are thought to reduce the capillary lumen, thus contributing to the trapping of blood cells [45].

An obvious consequence of the microvascular abnormalities in sepsis is that both diffusive and convective O₂ transport is impaired. Diffusive O₂ transport may be compromised in the lung, because of pulmonary oedema, or in the peripheral microcirculation. Convective transport may be impaired because of myocardial dysfunction, blood flow maldistribution, and inadequate capillary perfusion.

An objective and reliable method of monitoring microcirculatory organ perfusion is still not available. Minimally invasive video microscopy techniques (e.g. side-stream (SDF) or incident (IDF) dark field imaging) enable direct visualization of the microcirculatory network beneath thin mucosal surfaces, using a handheld instrument [46, 47]. The sublingual site has emerged as the preferred site for microcirculatory assessment in human subjects with overt or impending shock. Monitoring sublingual blood flow can yield important information for use in clinical studies of circulatory shock, because: (1) the sublingual mucosa shares the same embryological (and therefore anatomical) origin as the splanchnic mucosa; (2) alterations in sublingual perfusion can reflect alterations in splanchnic blood flow [48]; and (3) the sublingual space is easily accessible.

Near infrared spectroscopy (NIRS) is another technique that has been used to monitor tissue oxygenation (StO₂) [46]. Analysing changes in StO₂ during a short episode of forearm ischaemia enables the degree of microvascular dysfunction to be quantified repeatedly over time.

Despite considerable progress in this field in recent years, there are still many limitations that prevent the routine use of these techniques at the bedside, so, at the moment, their use remains investigational.

The redox potential of mitochondria is decreased in sepsis [49]. This phenomenon is associated with ATP depletion and is correlated with a poor outcome in septic shock patients [50]. Various mechanisms can contribute to the altered mitochondrial function. NO and nitrosative stress derivatives (peroxynitrite, ONOO–), the production of which are increased during sepsis, inhibit cytochrome oxidase (complex IV) [51]. The inhibition by NO is reversible, but peroxynitrite irreversibly inhibits F0F1 ATPase complexes I and II. The nuclear enzyme poly(ADP-ribose) polymerase (PARP-1) is activated by DNA breaks caused by oxidative stress. This activation decreases the cellular NAD⁺/NADH content. In endotoxic shock models, inactivation of PARP-1 can prevent vascular contractile dysfunction [52]. Both the alterations in oxidative phosphorylation and the decrease in antioxidant defences (notably catalase activity [53]) can explain the increase in oxidative stress. Stress hormones (cortisol, catecholamines, vasopressin, growth hormone) are released in greater amounts in sepsis and can impact cellular respiration [54].

Cytosolic and mitochondrial apoptotic pathways are both activated during sepsis. The phenomenon known as mitochondrial permeability transition leads to the transition of cytochrome c from the mitochondria to the cytosol and the activation of caspases. A positive impact of the inhibition of the permeability pore transition has been shown in several models [28].

Mitochondrial biogenesis is altered during sepsis. Experimental evidence suggests that restoration of mitochondrial bioenergetics is associated with improved haemodynamic parameters, organ function, and overall survival [55], and early activation of mitochondrial biogenesis has been associated with survival in patients with sepsis [56].

The management of sepsis (see Figure 72.3) is essentially based on rapid and adequate source control with antibiotics, and source removal where appropriate, adequate cardiovascular resuscitation with fluids and vasopressor agents, and supportive therapy, including renal replacement therapy (RRT) and/or ventilator support, where necessary, glucose control, and adequate nutrition. Although considered separately for the purposes of this chapter, in practice, all these aspects will be considered and applied (where appropriate) almost simultaneously.

Because the type of infecting organism(s) is usually unknown at the time of antibiotic initiation, empirical antibiotic therapy is based on the clinical presentation and epidemiological factors, including the local flora, resistance patterns, and previous antibiotic exposure. The initial empirical anti-infective therapy should include one or more drugs that have activity against all likely pathogens and penetrate in adequate concentrations into the presumed source of sepsis [57]. Combination therapy is recommended for patients in septic shock or with difficult-to-treat, multidrug-resistant bacterial pathogens, such as Acinetobacter and Pseudomonas species [57].

IV antibiotic therapy should be started as early as possible in patients with sepsis [57]. Timing of the initial administration of effective antimicrobial therapy is the most important predictor of survival [58]. Appropriate cultures should be obtained before initiating antibiotic therapy, whenever possible, but should not prevent the prompt administration of antimicrobial therapy. De-escalation to the most appropriate single therapy should be performed whenever possible as soon as the susceptibility profile is known. An antimicrobial treatment duration of 7 to 10 days is adequate for most serious infections, but longer courses may be needed in some patients [57]. Procalcitonin concentrations may be used to support a decision to stop antibiotics [57] but should not be used in isolation.

Source control is a critical issue in the optimal management of infection associated with severe sepsis, yet too often precious time is wasted before measures to control an underlying source of sepsis are taken. All patients presenting with severe sepsis should be carefully evaluated for the presence of a focus of infection amenable to source control measures, specifically the drainage of an abscess or a local focus of infection, the debridement of infected necrotic tissue, the removal of a potentially infected device, or the definitive control of a source.

Timing is of crucial importance in the haemodynamic management of patients with sepsis. Four phases of haemodynamic strategy can be distinguished, defined as ‘SOSD’ [59]:

Efficient restoration of the circulating blood volume is the primary goal of resuscitation in septic patients. Infusing fluids is a cornerstone of supportive care during sepsis, but optimal modalities and volumes are difficult to determine, and choices should be driven by objectives in the individual patient. There has been considerable controversy over the years regarding the preferential use of crystalloids or colloids, but crystalloid solutions are usually used as the initial resuscitation fluid, with addition of albumin if large amounts of crystalloids are needed [57]. Hydroxyethyl starch solutions are no longer recommended in patients with sepsis. An initial fluid challenge [60], using 20–30 mL/kg of crystalloids, is recommended [57], and repeated fluid challenges performed to assess the ongoing fluid needs. Routine maintenance of a superior vena cava oxygenation saturation (ScvO₂) or mixed venous O₂ saturation (SvO₂) of 70% or 65%, respectively, is not recommended [57]. The negativity of recent trials of early goal-directed therapy [PRISM Investigators, 2017] does not mean that some patients may not benefit from this strategy, but ScvO₂ must be considered as just one variable among others.

Red blood cell transfusion should be considered in the presence of anaemia, especially when the haemoglobin concentration decreases to <7 g/dL [57]. A recent study has shown that overtransfusion is not beneficial [64]. For critically ill patients with acute ischaemic heart disease, a haemoglobin concentration >9 g/dL may be more appropriate [65]. In all cases, the need for a red blood cell transfusion should be individualized, based on a patient’s clinical circumstances, rather than an arbitrary haemoglobin concentration.

The choice of vasopressor is important. Depending on its receptor modulation, the vasopressor agent may have an impact on myocardial function, different organ perfusion (renal, hepatosplanchnic), and microcirculatory perfusion. Noradrenaline is the vasopressor of choice in patients with sepsis [66]. Adrenaline should not be used as a first-line agent but may be used with, or replace, noradrenaline if an additional agent is needed to maintain adequate blood pressure [57]. A recent study has shown that angiotensin II administered as an adjunct to patients with vasodilatory shock not responding to norepinephrine (0.2 μg/kg/min), was associated with a greater increase in mean arterial pressure compared to placebo [Khanna A, 2017].

Vasopressin administration has been suggested on the basis of inadequate blood levels in septic shock. The VASST study showed a similar survival rate with vasopressin, compared to noradrenaline, in septic shock, with perhaps a decreased mortality in patients with less severe forms of shock [67]. Vasopressin administration may be of interest in the context of impending renal failure, because it acts on V1R that are located on the efferent arterioles, and not on afferent arterioles as for α-1 receptor agonists (dopamine, noradrenaline, adrenaline). A post hoc analysis of the VASST study [68] indicated that vasopressin was associated with a trend towards a lower rate of progression to renal failure and a lower rate of use of RRT, compared to noradrenaline. However, a more recent trial reported no differences in kidney failure-free days in patients treated with vasopressin or noradrenaline [69]. The reduction in cardiac output and the decrease in hepatosplanchnic and coronary blood flows induced by vasopressin are of some concern, so that, when used, doses should be limited. The latest guidelines suggest that vasopressin (0.03 U/min) can be added to noradrenaline to raise the MAP to target or to decrease the noradrenaline dose, but it should not be used as the initial vasopressor [57].

Despite a lack of randomized data demonstrating its efficacy, dobutamine is the first-choice inotropic agent for patients with measured or suspected low cardiac output in the presence of adequate filling pressures [70]. The challenge in interpreting myocardial dysfunction in sepsis is that the most important physiological variable is cardiac output. Patients with pre-existing cardiac dysfunction, who may have a decreased cardiac output, are candidates for inotropic therapy to improve O₂ delivery (DO₂). Even patients without an underlying cardiomyopathy may have an inadequate cardiac output to meet the O₂ demands of the cells. The presence of a low central venous oxygen saturation (SvO₂)/mixed venous oxygen saturation (ScvO₂) or persistent hyperlactataemia may suggest an inadequate cardiac output (see Figure 72.4) and an indication for dobutamine. A strategy of routinely increasing the cardiac index or DO₂ to predefined ‘supranormal’ levels has not been shown to improve outcome and may be deleterious.

Phosphodiesterase inhibitors, like milrinone, can be considered as an adjunct to adrenergic agents, although their vasodilating effects and prolonged half-life complicate their routine use.

Levosimendan exerts beneficial effects on the LV and RV, which are independent of β-adrenergic signalling or changes in the intracellular Ca²⁺ concentration, by increasing the contractile myofilament sensitivity to Ca²⁺. As the administration of adrenergic agents may be associated with tachyarrhythmia or an increase in myocardial O₂ demand, levosimendan may represent an alternative in the setting of sepsis-induced myocardial dysfunction [71]. However, a recent randomized placebo-controlled study reported no beneficial effect on organ dysfunction or mortality of the routine addition of levosimendan to standard treatment in adults with sepsis [73], although patients were not selected before randomization for the conditions in which this drug is most likely to be of benefit: myocardial depression and microcirculatory abnormalities.

Oxygen should be provided to prevent hypoxaemia, with mechanical ventilation used early if mask/nasal prong ventilation is not possible or effective. Ventilator settings should be set to maintain so-called ‘lung-protective ventilation’; one can start with a tidal volume of around 6 mL/kg of (predicted) body weight and an upper limit of plateau pressure of 30 cmH₂O (considering the chest wall compliance) [57]. Positive end-expiratory pressure (PEEP) should be set to avoid extensive lung collapse at end-expiration. There is some evidence that higher PEEP levels may be associated with improved survival rates in patients with more severe disease [75]. Neuromuscular-blocking agents may be beneficial during the first 24 hours of severe ARDS [76] and prone positioning should be considered when possible in patients with severe ARDS [77].

Although there is no consistent evidence that supports the superiority of continuous renal replacement therapy (CRRT) over intermittent haemodialysis (IHD), the use of continuous therapies can facilitate the management of fluid balance in haemodynamically unstable septic patients [57]. Because there is no evidence to support an early RRT strategy systematically, decisions have to be made on an individual basis for each patient. However, since acute renal failure and its associated metabolic alterations appear to increase the risk of severe extrarenal complications, the initiation of RRT should not be delayed in patients with severe, rapidly developing, and oliguric forms of acute renal failure. The ‘optimal dose’ of renal support is still a matter of debate, but the most recent randomized trials showed no benefit of high compared to conventional, doses of RRT [78, 79].

The use of steroids in septic shock is still controversial. The administration of IV hydrocortisone (200 mg/day as a continuous infusion) is currently suggested in severe cases of septic shock, not responsive to fluid resuscitation and vasopressor therapy [57]. The results of a recent French study (APROCCHSS) in more than 1200 patients showed a reduction in mortality in patients with septic shock who received hydrocortisone (200 mg/day) plus fludrocortisone (50 µg/day) (Annane D, presented at 2017 ISICEM, Brussels, Belgium), and publication of the full results is eagerly awaited. The patient should be weaned from the steroids when vasopressor support is no longer required.

In addition to haemodynamic management and respiratory and renal support, general aspects of management relevant to all critically ill patients must not be forgotten. Checklists can help to enhance the efficiency, safety, and efficacy of care. One suggested approach is to ‘give your patient a FASTHUG at least once a day’. FASTHUG is an acronym for simple interventions in the critically ill patient that can contribute to the quality of care and may improve outcomes [74].

In recent years, important advances have been made in the understanding of the pathophysiology of sepsis. Multiple cells and cellular pathways have been identified as being involved in the response to sepsis, including inflammation, thrombosis, and apoptosis. Macrocirculatory and microcirculatory dysfunctions are important contributors to the development of progressive organ failure. Importantly, general haemodynamic parameters cannot always predict the impact of treatments on the microcirculation and cells.

Outcome improvement in sepsis is based on recognizing the process early and instituting effective therapies. The time window for intervention is short, and treatment must promptly control the source of infection and restore haemodynamic homoeostasis. After initial aggressive treatment, minimizing complications, rapid weaning from vasopressors, and achieving a negative fluid balance must be priorities. The improved ability to diagnose sepsis early and to measure the response to sepsis and to treatment will facilitate a more appropriate targeting of therapies to individual patients.