Immunoparalysis in Septic Shock Patients Immunoparalysis in Septic Shock Patients

In the recent years, it has become clear that septic shock is characterized by the simulta- neous production of inflammatory and anti-inflammatory mediators; the primary role of the latter is to counterbalance the former, thus limiting the severity of their systemic effects. However, in a number of patients, the anti-inflammatory substances can cause a downregulation in both the innate and adaptive immune capabilities, leading a second phase characterized to secondary infections caused by opportunist germs and the reacti vation of latent viruses, muscle wasting; altogether, these abnormalities set the stage for a chronic critical condition. This condition, whose identification is relatively recent, is called immunoparalysis. Unfortunately, the current approach to septic shock is focused much more on the inflammatory phase than in the ensuing immunoparalysis, whose diagnosis can be challenging. In this chapter, the role played by both classes of mediators, the monitoring of the immune system, and the possible current and not yet available therapeutic strategies of immunoparalysis are reviewed and discussed.


Introduction
The classical clinical manifestations of septic shock (SS) include fever, tachycardia, arterial hypotension, and abnormalities of the white blood cell count (WBC) associated with a wide range of organ dysfunction carrying a substantial risk of death [1]; the current approach, issued under the auspices of the Surviving Sepsis Campaign on the basis of clinical trials fulfilling the evidence-based medicine (EBM) criteria, includes the rapid administration of widespectrum antibiotics, the maintenance of a proper perfusion pressure via the administration of fluids and/or to vasopressors, the drainage of septic foci, etc. [2]. Overall, it appears that both the description and the therapies apply to acutely ill patients suffering from an infectioninduced overwhelming reaction determined by a huge number of pro-inflammatory mediators produced and released by the innate immunity system. However, more than 20 years ago, Bone [3] hypothesized that this early hyperinflammatory phase could be accompanied by a compensatory anti-inflammatory response (CARS) aiming to limit the tissue damage. In the last decade, the concept of CARS has changed from a time-limited and somehow beneficial mechanism to a harmful reaction, potentially leading to a condition of marked reduction of the immune capabilities known as immunoparalysis [4][5][6]. Clinically, this condition is marked by recurrent and/or unresolving infections caused by germs with relatively low virulence; the reactivation of silent virus such as cytomegalovirus (CMV), Epstein-Barr virus (EBV), and herpesvirus (HV); a persisting low-grade inflammation; nutrition-resistant hypercatabolism; and muscle wasting [7,8] ( Table 1). The immunoparalysis characterizes also the clinical course of the chronic critically ill patients, namely, subjects who survived the initial insult (i.e., septic shock due to pneumonia, peritonitis, etc.) but fails to recover enough to be weaned from the mechanical ventilation and discharged from the intensive care unit (ICU) [9]. Moreover it should be noted that factors other than pathophysiological mechanisms can reduce the immune response, including the administration of steroids and norepinephrine [1,10]. The aims of this chapter are (1) to review the main mechanism determining a SS, (2) to describe the transition from an easily recognizable hyperinflammatory condition to a less straightforward diagnosable one featured by a downregulation of the immune capabilities, (3) to provide some monitoring tools of the immune function, and, finally, (4) to identify some possible therapeutic approaches.

Variable
Uncontrolled inflammatory response Immunoparalysis

The inflammatory response
Since the late 1970s, it has become clear that the clinical and biochemical manifestations of sepsis and its related complications are not caused directly by invading germ(s) but rather by the host's response to the infection. The innate immune response largely accounts for the above described signs and symptoms. The presence of microorganism-derived substances collectively known as pathogen-associated molecular pattern (PAMP) which include endotoxin, capsular antigens, elements derived from the cell wall, flagellins, and other substances derived from the bacterial lysis determines the rapid activation of genes encoding for an extremely elevated (and still partially unknown) mediators able to trigger a strong inflammatory reaction, including the tumor necrosis factor-α (TNF), a number of interleukins (IL), the platelet-activating factor (PAF), etc. ( Table 2). It is worthwhile to recall that (a) the list of mediators is incomplete because new elements are added on a weekly or at maximum monthly basis, (b) the rise of blood levels of inflammatory mediators is a matter of minutes since it represents the first line of defense to contrast the deleterious effects of PAMP and DAMPS, and (c) for this reason, the innate response is highly similar among all species of mammalians [6].
Independently from their biochemical structure, the term inflammasome lumps together all these heterogeneous mediators that are characterized by (a) the presence of many positive and negative feedback loops, determining an array that can be better conceived as a network  and not a cascade, thus making understandable the therapeutic failure demonstrated in many trials in which septic patients were treated with substances aimed to the block a single mediator via monoclonal or chimeric specific antibodies (Ab) such as anti-TNFαAb or with the administration of circulating antagonists (ra) (i.e., soluble IL1-ra and TNF TNFα-ra) directed to block the receptors present on the cell surface; (b) the pleiotropic and paracrine effects, accounting for the multiple effects exerted in different organs; (c) the interference with the mitochondria causing a disturbance of the O 2 uptake and consumption by the tissues; and (d) the interaction with other biological systems including the complement system and the coagulative cascade. Notably, the very same mediators are produced in noninfectious conditions, including trauma, low-flow states, surgery, burns, etc.; in these circumstances the trigger is represented by an intracellular substance derived from the injured tissues (DAMP, damage-associated molecular patterns). The endothelium is massively involved in this reaction causing a microvascular plugging and the abnormal production of nitric oxide (NO) which exert a profound vasodilation [11,12].
From an evolutionary perspective, it is likely that these mediators have been developed and maintained, aiming to contain the initial inoculum and to destroy the responsible organisms. This explains why in most cases an infection does not cause a SS: actually, the latter occurs only when the pro-inflammatory mediators exert their effects at a systemic level, thus determining the clinical phenotype of SS and the almost unavoidable presence of the simultaneous dysfunction of different organs and systems even not directly involved by the infection (MODS).

The compensatory reaction
The secretion of inflammasome is accompanied by the production of other substances aimed to limit their action at a local level and, at the same time, to prevent their systemic spread ( Table 3). As stated above for the inflammatory mediators, their list is incomplete for the very same reasons. Actually, it was hypothesized that during the initial phase (almost), only proinflammatory mediators were produced and that these conditions subsided due to the action of the CARS-associated mediators. Despite its popularity, it became clear that this scheme represents an oversimplification as (a) both classes of substances are produced since the initial phase of sepsis albeit in different rates; (b) the action of anti-inflammatory mediators is responsible for the late-onset immunoparalysis; and finally (c) a low-level production of pro-inflammatory substances can be maintained even during the advanced stages of sepsis leading to malnutrition, protein waste, and reduced adaptive immunity. Overall, the sepsisassociated immunoparalysis resembles the normal aging process of the immune system (immunosenescence) that is characterized by the overall downregulation of both the innate and adaptive immunity functions. This appears particularly relevant as the ever-increasing age of septic patients exposes them to both conditions.
Put shortly, it appears that the mediators implicated in the CARS can represent a doubleedged sword, as they both can exert (a) a beneficial role when they determine the restoration of the immune condition existing prior to the sepsis (immune restoration) and (b) can trigger a life-threatening condition when their excess production and/or duration of action causes the shutdown of the immune response [13,14].

Sepsis
In conclusion, (at least) three clinical trajectories can be hypothesized (Figure 1): the first includes patients with an intense hyperinflammatory reaction that subsides once the CARS is well established and the immune function is restored; in the second the initial phase is shorter and weaker, and the CARS determines a short-lived immunoparalysis preceding the return   toward the baseline immune function; and in the third one, the CARS prevails and causes the loss of the immune capabilities.

The determinants of immunoparalysis
Only recently it became clear that the CARS does not represent only a physiologic counterbalance to the inflammatory response to PAMP and DAMP but that it can determine a critical condition in and by itself [13,15].
Actually, different experimental and clinical studies indicate that the advanced stage of sepsis and SS is characterized by a reduction of both the innate and adaptive immune responses ( Table 4). Extensive evidence supports this model, even if large inter-patient differences exist. First, monocytes present a reduced expression of membrane HLA-DR in association to either a decreased secretion of inflammatory mediators when stimulated or a diminished antigen presentation. Second, different membrane-bound receptors able to potentiate the immune response, including IL-2α, IL-7R α, CD86, etc., are reduced. Third, the production of immunosuppressant substances, such and programmed death 1 (PD1) and its ligand (PD-L1), is increased in antigenpresenting cells, thus inhibiting the activation of T lymphocytes. Fourth, there is an increased appearance of immunosuppressive T-cell subpopulations, such as myeloid-derived suppressor cell and CD4 + and CD25 + T-regulatory cells (T reg ), which suppress adaptive immunity. These appear to be particularly relevant, as Treg (a) actively produce anti-inflammatory cytokines including TGF-β and IL-10, (b) downregulate the secretion of pro-inflammatory mediators, (c)

Factors involved Marker
Monocyte deactivation ↓ mHLA-DR expression  Finally, the phagocytosis of apoptotic cells by fixed and circulating macrophages leads to a switch of the latter to the M2 phenotype, whose feature is an increased production of the antiinflammatory substances IL-10 and IL-1ra. Put briefly, all these mechanisms exert their action via relatively few common pathways, which include the increased apoptosis determining the reduction of immune cells, the loss of antigen presentation, the blunted response to PAMP, and the reduction of energy production caused by the impairment of the glucose metabolism ( Table 5) [16,17]. All these reactions are driven by epigenetic changes causing in different time frames the activation or deactivation of genes involved in the immune response, and the resulting phenotype is an intense inflammatory response or, conversely, an immunoparalysis.

The diagnosis of immunoparalysis
The recognition of sepsis-induced immunoparalysis is not straightforward because the clinical manifestations associated with the switch from the hyperinflammatory state to CARS and the full-blown depression of the immune capabilities are not so protean as the symptoms of SS [18]. Moreover, the SSC guidelines focus almost exclusively on the former and pay much less attention, if any, to the latter. From a practical and clinical point view, some issues appear particularly relevant.

Timing of onset
The transition from the hyperinflammatory phase to immunoparalysis can be challenging to identify and to monitor at the bedside and represents a kind of no man's land in the clinical course of patients which survived from the initial phase of SS.
The onset is highly variable. Actually, although the secretion of immunomodulatory substances can occur relatively early, their clinical consequences present wide variations. Some authors [19] observed a substantial difference of mHLA-DR starting from 3 to 7 days in a small group of surgical septic patients, and other authors demonstrated that significant decrease of

Immune cell number anergy
Energy failure Immune cell anergy Apoptosis Epigenetic regulation ↓ Pro-inflammatory mediators the CD14/HLA-DR and of heat-shock proteins (HSP) 70 and 90 was present already within 24 hours from the onset of sepsis [5]; in both studies, these alterations were more marked in patients who developed SS later on. More recently, Morris et al. [20] in association with raised percentage of regulatory T cells (T reg ) were predictive for infections occurring between 3 and 9 days after ICU admission, and a similar timing has been demonstrated also in another study in which the mortally rate of secondary infection was ~14% [17]. On the basis of these findings, it is reasonable to hypothesize that (a) a combination of cellular and soluble factors able to blunt the immune response is present since the very initial phase of sepsis; (b) their effects on the clinical course, namely, the appearance of secondary infections and/or viral reactivation, can occur within the initial 10 days from the admission; and (c) these are associated with a substantial mortality of patients surviving the initial insult.

Monitoring of the immune function
In ICU patients, every organ system is monitored to allow a change in the treatment tailored on the variation observed. An ideal monitoring system should be accurate, cheap, and not labor-intensive, and the information gathered should be readily if not continuously available. Since it has become clear that the immune system in sepsis undergoes modifications not reflected by the commonly measured biological variables such as the arterial pressure, the heart rate, the urinary output, etc., different investigations aimed to identify one or more markers of changes of its functions whose follow-up could be valuable to modify the therapy according to its changes: as an example, the occurrence of immunoparalysis contraindicates the administration of steroids whose use is recommended by the SSC guidelines.
Several monitoring systems exploring both legs of the immune response have been developed so far, based on the repeated assessments of the cells involved, their response to different   [14,15,21,22]. It could be useful to describe separately those currently available and those which will be used likely in the next future. Most of the former ( Table 6) can be obtained cheaply and on a daily basis; among all, the neutrophil-to-lymphocyte ratio has been indicated as the less costly and more rapidly available monitoring tool [23,24].
Other advanced, expensive, and not yet widely available monitoring tools take advantage of more sophisticated lab techniques ( Table 7) requiring lab expertise and financial resources putting them at risk of not being used outside the research center. Another dynamic approach, which shares the very same limitations of the previously described advanced techniques, consists in challenging the immune cells with substances able to trigger their activation, including LPS, other PAMP, and phytohemoagglutinin; actually, a number of investigators demonstrated that a blunted response to the stimulation is associated with an increased rate of severe infectious complications in different patient populations [25][26][27].   it should be limited to the subjects at risk; as an example, it is worthwhile to monitor the immune function in patients undergoing multiple abdominal surgical procedures for suture dehiscence but not in another one safely recovering after peritonitis.

The identification of patients at risk of immunoparalysis
Even with the exclusion of clinical conditions and/or treatments known to cause an immunoparalysis (i.e., solid and hematologic cancers, autoimmune disorders), etc., this circumstance can occur in virtually all ICU patients; however, different studies identified some predisposing factors that should be considered particularly relevant, including septic shock, advanced age, health care-associated infections, elevated Charlson's score indicating a substantial underlying fragility, comorbidities, prolonged hospital and ICU length of stay, and multiple surgical procedures [17,28,29]. The latter, which are associated with the repeated activation of the inflammatory and anti-inflammatory responses, according to the multiple hits model, ultimately lead to the exhaustion of the immune response [30] ( Figure 2).

The treatment of immunoparalysis
In the last decade, a number of drugs have been developed to restore a normal immune function in patients with solid or hematologic tumors on the basis of many investigations Sepsis demonstrating the tumor cells are able to suppress in many different ways the host's immune response against themselves. Independently from the substance use and the molecular target, these innovative treatments have been demonstrated to be effective but somehow difficult to handle, as they are associated with a number of side effects ranging from mild to life-threatening [31]. As several similarities exist between tumor-and sepsis-induced blunting of the immune response [32], it is likely that in the next future the immune-boosting treatments will be developed to treat the latter, aiming to develop a precision medicine also in ICU patients [33] ( Table 8).
Presently, according to the SSC guidelines [2], the immune-targeted approaches are limited to the administration of steroids in not fluid and catecholamine-responding SS, whereas the use of intravenous immunoglobulins (IvIg) is discouraged. Actually, this latter position is questionable as a number of trials performed in several thousands of patients demonstrated that (a) the administration of IvIg is associated with the reduction of mortality in different subsets of SS patients; (b) among the different preparations available, the only ones containing supranormal concentrations of IgM and IgA appears more effective, and (c) the improvement of survival is time-dependent, as a ~6% increase of mortality has been observed for every day of delay in the administration [34].
Besides steroids and IvIg, other treatments aimed to modulate the immune response include blood purification (BPT) techniques and a number of substances able to boost it.

Cells/factors involved Alterations Possible therapies
Myeloid cells ↑ Immature neutrophils

Blood purification techniques
Since the 1980s, a number of extracorporeal techniques have been developed aiming to remove the "toxic" mediators responsible for the clinical manifestations of SS.
Independently from their principle of functioning (see later), the BPT consists in an extracorporeal circuit where the patient's blood flows till enters in the depurative device; once the latter is passed, the blood returns to the patient. According to the principle used, the BPT can be subdivided into (a) blood processing or (b) plasma processing techniques. In the former, the whole blood is depurated via a number techniques, which differ in terms of type and surface of the membranes used, their permeability to the high molecular weight of the septic mediators, etc., whereas in the latter the plasma is separated from the blood, processed in a cartridge, and reinfused downstream. The mediators can be eliminated through the membranes or adsorbed over it. In both cases, the neutralizing capabilities are time-limited. A detailed description of the BPT is beyond the aim of this chapter, but some considerations are necessary. First, there are no studies clearly demonstrating the superiority of one of them, even if some meta-analysis indicates that the those using the adsorption are more effective; (b) they can remove also antibiotics, nutrients, vitamins, hormones, etc.; (c) they require anticoagulation; and, most importantly; and (d) they are not selective and thus remove pro-as well as anti-inflammatory mediators [35].

Immune-boosting agents
Different substances have been used or likely will be used in the next future ( Table 9) to enhance the depressed immune function in septic and non-septic critically ill patients, including [36,37]: • Interferon-γ (IFN-γ) is a cytokine produced by helper T cell and an activator of monocytes. Different case series and case report performed in a limited number of patients demonstrated that its administration was associated with an increased HLA-DR expression; however, presently there are no RCT fulfilling the EBM criteria demonstrating a beneficial effect on the outcome of patients with SS.
• Granulocyte-macrophage colony-stimulating factor (GMC-SF) stimulates the production of neutrophils from the bone marrow. Even if prophylactic use in neutropenic patients did not demonstrate any beneficial effect, a number of investigations demonstrated that its administration was associated with an improved outcome especially in patients with a decreased HLA-DR expression.
• Interleukin-7 (IL-7) is a cytokine released by bone marrow and thymus cells that prompts the growth and the differentiation of T cells. This substance is considered an immuneboosting agent in patients with cancer and multifocal leukoencephalopathy and in septic patients suffering from immunoparalysis.
• Programmed death inhibitors (PD1i) are proteins whose effect is to block the programmed death of immune cells, which appears to be a critical factor for the progression of cancer.
This approach is new as it is aims to increase the immune response to the cancer cells without interfering with their metabolism. Due to their mechanism of action, their administration could determine a potentially life-threatening inflammatory reaction caused by the sudden release of mediators determining a "cytokine storm"; although their use is not codified yet in critically ill septic patients, in a recent RCT, the restoration of the immune response in the absence of a hyperinflammatory reaction was demonstrated in some SS patients given a novel PD1i at different doses [38].

Conclusions
Independently from its source, septic shock can be considered a double-step process: the initial phase is characterized by an intense inflammatory response that is counterbalanced by the production of several anti-inflammatory substances aiming to restore the immunity pre-sepsis steady state. However, in many cases this compensatory mechanism prevails and not only extinguishes the initial response but determines a condition of immunoparalysis that dominates the clinical course and influences the outcome. Unfortunately, the current approach is mainly directed against the initial inflammatory phase although some techniques of monitoring of the immune function are currently developed and others are being studied. The same concepts apply to treatments directed to potentiate the immune capabilities, but in this case the goal appears to be still far.