INTRODUCTION — A number of potential therapies have been investigated or are being evaluated in an attempt to improve clinical outcomes in sepsis. Therapies that are being investigated are the focus of this review, although none are recommended for use at this time because the ratio of potential benefits and harms has not been adequately studied. Therapies that have proven ineffective are also mentioned briefly. The definition, pathophysiology, and management of sepsis are discussed elsewhere. (See "Sepsis syndromes in adults: Epidemiology, definitions, clinical presentation, diagnosis, and prognosis" and "Evaluation and management of suspected sepsis and septic shock in adults".)
THERAPIES BEING INVESTIGATED — A number of potential therapies for sepsis appear promising in animal models, but have not yet been adequately studied in humans. Other potential therapies have been studied in humans, but have given conflicting results and require additional investigations to clarify their effects. These therapies are the focus of this section. Often there is no better therapy to offer a patient than enrollment onto a well-designed, scientifically valid, peer-reviewed clinical trial. Additional information and instructions for referring a patient to an appropriate research center can be obtained from the United States National Institutes of Health (www.clinicaltrials.gov).
Inhibition of innate immunity — Infecting microbes display highly conserved macromolecules (eg, lipopolysaccharides, peptidoglycans) on their surface. When these macromolecules are recognized by pattern-recognition receptors (called Toll-like receptors [TLRs]) on the surface of immune cells, the host’s immune response is initiated. This may contribute to the excess systemic inflammatory response that characterizes sepsis. Inhibition of several TLRs is being evaluated as a potential therapy for sepsis:
Inhibition of TLR-4 with the antagonist, E5564 (Eritoran) was tested in humans with severe sepsis in the ACCESS (A Controlled Comparison of Eritoran Tetrasodium and Placebo in Patients with Severe Sepsis) Trial, showing no effect on 28-day mortality [1]. An earlier clinical trial that evaluated the inhibition of TLR-4 with another antagonist, TAK 242 (Resatorvid), found a non-statistically significant reduction in 28-day mortality [2].
The triggering receptor expressed on myeloid cells 1 (TREM-1) is an activating receptor expressed on innate immune cells and activated endothelial cells. TREM-1 activation results in dysregulation of innate immune responses [3]. Cellular TREM-1 is upregulated in septic shock, where it also exists in a soluble form (sTREM-1) following cleavage from the membrane-bound form. Concentrations of sTREM-1 correlate with Sequential Organ Failure Assessment (SOFA) scores, suggesting a causal role for TREM-1 in septic organ failures. Nangibotide, a ligand-trapping molecule that inhibits TREM-1 activation, was well tolerated in a small trial of septic shock [4]. The ASTONISH Phase IIb trial of nangibotide in septic shock is currently enrolling with a target of 450 subjects [5].
Intravenous immune globulin — It has been hypothesized that polyclonal intravenous (IV) immune globulin (IVIG) may benefit patients with sepsis by binding endotoxin. However, we and others believe that IVIG should not be administered to patients with sepsis [6-13]:
●Suggesting that polyclonal IVIG has no benefit, a randomized trial of 653 patients with sepsis found that IVIG did not reduce 28-day mortality compared to placebo [6].
●Suggesting that polyclonal IVIG has a benefit, meta-analyses of randomized trials found that polyclonal IVIG decreased mortality compared to either placebo or no IVIG [7-10]. The mortality benefit was greatest when IgA- or IgM-enriched IVIG was used [9,10]. However, all of the meta-analyses were limited by heterogeneity and two of the meta-analyses found that the mortality benefit disappeared when only the well-designed trials were analyzed [8,10].
Well-designed randomized trials are warranted to evaluate the impact of IgA- and IgM-enriched IVIG [12]. (See "Overview of intravenous immune globulin (IVIG) therapy".)
Cytokine and endotoxin inactivation or removal — Strategies to improve sepsis by either inactivating or removing endotoxin and/or inflammatory cytokines have been investigated:
●Hemoperfusion through adsorptive materials – Perfusion of blood from patents with sepsis through adsorptive membranes or sorbent-containing cartridges can remove harmful circulating cytokines and endotoxin. While initial trials were hopeful [14-18], several trials have reported that hemoperfusion through a membranous polymyxin B fiber column (PBFC; high affinity for endotoxin) is not beneficial [19,20]. A post-hoc analysis suggested some possible benefit in those with high levels of endotoxin activity [21].
Hemoperfusion through sorbent-containing cartridges (hemadsorption) provides a larger surface area than membranous columns to enhance the removal of cytokines and inflammatory molecules involved in the pathogenesis of sepsis. Among the existing sorbents, CytoSorb (porous polymer beads) appears to be the most promising in animal models with limited data in humans [22-25]. However, it does not remove the two more potent inflammatory molecules, endotoxin and interleukin-10.
●Plasma or whole blood exchange – Case series and observational studies with historical controls reported favorable outcomes when endotoxin was removed by plasma or whole blood exchange, including lower plasma endotoxin concentrations and, possibly, lower mortality [26-28].
●Coupled plasma filtration adsorption (CPFA) – CPFA combines the separation of plasma from the cellular components of blood with a highly permeable filter (plasma filter) followed by sorbent adsorption of the plasma component with a styrene resin. In one multicenter trial of 330 adult patients with septic shock, compared to conventional care, CPFA was not associated with a mortality benefit (45 versus 47 percent) nor did it prevent the development of new organ failures or decrease length of ICU stay [29].
●Interleukin-1 is a pro-inflammatory cytokine implicated in sepsis-related multiorgan dysfunction syndrome. Anakinra (a recombinant, non-glycosylated human interleukin-1 receptor antagonist) failed to improve survival in two clinical trials enrolling subjects with severe sepsis and septic shock [30,31]. Recent reports that anakinra is effective for patients with macrophage activation syndrome have rekindled interest in identifying subsets of severe sepsis in which the drug may be useful. When data from the original confirmatory phase III trial [30] were reanalyzed, grouping subjects according to the presence of hepatobiliary dysfunction and disseminated intravascular coagulation (as markers for macrophage activation), treatment with anakinra was associated with improved survival in those with macrophage activation, but not in those without [32]. The patients who met criteria for macrophage activation was a small subset of the overall group, so this finding (if confirmed) may pertain to few patients with septic shock.
●Xuebijing (XBJ) is an herbal preparation that has an antagonistic effect on endotoxin and is an inhibitor of inflammatory mediator release by endotoxin-stimulated monocytes/macrophages [33-35]. Preliminary data suggest possible benefit in patients with sepsis [36,37]. One randomized trial of 1817 patients with sepsis reported that intravenous XBJ reduced 28-day mortality compared with placebo (19 versus 26 percent) [38]. The most common adverse event associated with XBJ was elevated transaminases and white blood cell count, but no difference in adverse event rates was reported between the groups. However, several limitations in the trial design may have influenced the outcome.
While some of these therapies are promising, we believe that additional evidence from large, randomized trials are necessary before any of these strategies can be routinely used in patients with sepsis.
Interferon-gamma — A decrease in monocyte function has been observed late in the course of sepsis, which may predispose patients to life-threatening secondary infections. This observation prompted a study in which patients with sepsis were administered interferon-gamma [39]. The study found that interferon-gamma restored monocytic cell function. Controlled clinical trials measuring patient-important outcomes in patients with sepsis have not been reported, although a trial evaluating the impact of interferon-gamma in candidemia is underway [40].
Granulocyte-macrophage colony stimulating factor — Granulocyte-macrophage colony stimulating factor (GM-CSF, sargramostim, molgramostim) is a cytokine that promotes maturation of the progenitor cells of granulocytes, erythrocytes, megakaryocytes, and macrophages, as well as mature neutrophils, monocytes, macrophages, dendritic cells, T-lymphocytes, and plasma cells. Its use in sepsis has been studied in several controlled clinical trials.
One trial reported that GM-CSF increased the peripheral neutrophil count, but did not affect the mortality, duration of ICU stay, or rate of hospital discharge [41]. Another trial found that GM-CSF reduced the rate of infectious complications, the length of hospital stay, and the duration of antibiotic therapy, without affecting mortality [42]. And, a third trial found that GM-CSF induced a trend toward a shorter duration of mechanical ventilation (148 versus 207 hours), a shorter hospital stay (59 versus 69 days), and a shorter ICU stay (41 versus 52 days) [43]. All of the trials were small with few events, which is the reason why large absolute differences were often not statistically significant. We believe that larger controlled clinical trials are necessary to confirm the clinical effects reported in these trials and that GM-CSF should not be routinely used to treat sepsis until then.
The effects of granulocyte colony stimulating factor (G-CSF) are mentioned below. (See 'Ineffective therapies' below.)
Augmentation of immunomodulation — Antibodies directed against macrophage migration inhibition factor (MIF) prevented death in a study that used a cecal puncture animal model of sepsis, even when administered up to eight hours after the onset of peritonitis [44]. The mechanism of action is uncertain, but MIF inhibition might restore or augment the immunomodulatory actions of endogenous glucocorticoids [45]. While it is known that elevated MIF levels correlate with poor outcomes in humans with sepsis [45,46], the impact of MIF inhibition has not been studied in humans.
Stem cell therapy — Multipotent mesenchymal stem cells modulate several aspects of the immune response, reprogram the immune system to ameliorate host damage, enhance tissue repair, display antimicrobial properties, and reduce the propensity to multiorgan failure in animal models of sepsis. A growing safety record in human trials provides grounds for optimism that stem cell-based therapies may find a place in clinical medicine [47].
Immunostimulation — Anti-inflammatory pathways activated in septic patients may raise the risk of secondary infections. Treatments that stimulate immunity, such as IL-7, IL-15, or anti-PDL1, offer a promising option [48].
Inhibition of proinflammatory gene expression — A synthetic peptide has been developed that inhibits bacterial superantigen-induced expression of certain proinflammatory genes (interleukin-2, gamma interferon, and tumor necrosis factor-beta) by limiting T-cell activation. In one study, the peptide protected mice against lethal challenges with staphylococcal and streptococcal superantigen, when administered up to three hours after the superantigen [49]. This peptide has not been tested in humans.
Hemofiltration — It has been proposed that hemofiltration, as a mode of dialysis, may remove proinflammatory molecules that induce hemodynamic collapse in septic shock, thereby improving outcomes. However, data from small randomized trials of hemofiltration (high volume or continuous) in patients with septic shock suggest that there is insufficient evidence for the routine use of this mode of renal replacement therapy over conventional hemodialysis.
Initial studies suggested that the high volume hemofiltration offered benefit over conventional hemodialysis for the treatment of acute renal failure in sepsis [50-53]. However, meta-analyses of small randomized studies and one multicenter prospective study (IVOIRE) suggested no benefit [54,55].
Similarly, a randomized trial that evaluated continuous venovenous hemofiltration (CVVH) in patients with sepsis did not demonstrate any improvement in the clearance of inflammatory mediators or clinical outcomes [56]. Another trial was discontinued after an interim analysis showed more frequent and more severe organ failure in the hemofiltration group [57].
Anticoagulants — Heparin has anti-thrombotic and immunomodulating effects, both of which have the potential to be beneficial in sepsis. In a retrospective cohort study of patients with septic shock, intravenous therapeutic heparin was associated with decreased mortality, discontinuation of vasoactive drug infusions, and liberation from mechanical ventilation [58]. The risk of major hemorrhage or the need for transfusion was not increased. This study was followed by the HETRASE trial, which randomly assigned 319 patients with sepsis to receive low dose intravenous heparin (500 units per hour) or placebo for seven days [59]. The trial found no improvement in any clinical outcome. It has been suggested that the conflicting results were due to the randomized trial using a low dose of heparin; as a result, a randomized trial comparing various doses of heparin in patients with septic shock is being conducted [60].
Several early studies and a subsequent meta-analysis suggested a potential benefit of recombinant thrombomodulin in sepsis-induced disseminated intravascular coagulation [61,62]. However, a large randomized trial found no benefit to thrombomodulin therapy in sepsis-associated coagulopathy [63].
Therapeutic studies of anticoagulation in coronavirus disease 2019 (COVID-19) are discussed separately. (See "COVID-19: Hypercoagulability".)
Prostacyclin analogs — Iloprost is a prostacyclin analog that targets the microcirculatory perturbations of septic shock through vasodilatory and anti-thrombotic mechanisms. An initial case series of three patients receiving iloprost described improvements of hypoperfusion and arterial lactate concentrations without effects upon blood pressure [64]. A phase 2 trial enrolled 24 subjects randomized to placebo versus iloprost combined with the glycoprotein IIb/IIIa inhibitor eptifibatide, administered within 24 hours of onset of septic shock [65]. Compared to placebo, iloprost/eptifibatide was associated with a reduction in biomarkers of endothelial injury and fibrinolysis. Iloprost/eptifibatide was also associated with a reduction in SOFA scores. Phase 3 trials of iloprost are currently recruiting subjects [66,67].
Naloxone — A meta-analysis of three randomized trials (61 patients) comparing naloxone to either placebo or no naloxone found that naloxone therapy led to hemodynamic improvement, but did not improve the case-fatality rate [68]. Adverse effects that have been associated with naloxone therapy include pulmonary edema, hypertension, and seizures [69]. Given the limitations of the meta-analysis that found a physiological benefit, the absence of patient-important benefits, and the potential for adverse effects, we believe that naloxone therapy is not warranted for patients with septic shock. However, large controlled trials are justified, since the studies included in the meta-analysis had too few events to definitively exclude a beneficial effect.
Pentoxifylline — Sepsis results in decreased red cell deformability and increased erythrocyte aggregation, effects that may be mitigated by pentoxifylline [70]. Pentoxifylline also inhibits neutrophil adhesion and activation, and modulates endotoxin-induced expression of proinflammatory cytokines [71]. In a trial of 51 surgical patients with severe sepsis who were randomly assigned to receive pentoxifylline or placebo, pentoxifylline improved the multiple organ dysfunction score and the arterial oxygen tension to fraction of inspired oxygen (PaO2/FiO2), but there was no improvement in the 28-day mortality [72]. Additional trials are needed to identify a true beneficial clinically meaningful effect.
Statins — HMG CoA reductase inhibitors (statins) are in wide clinical use and have an established role in the management of hyperlipidemia and cardiovascular disease. Statins also appear to have beneficial anti-inflammatory properties, such as suppression of endotoxin-induced up-regulation of TLR-4 and TLR-2. Although early studies of statins in patients with sepsis suggested benefit, most studies since then, including meta-analyses of randomized trials, show no benefit from statin use in this population [73-82].
Heart rate control — Beta-blocker therapy may potentially attenuate the deleterious effects of the sympathetic adrenergic response that occurs during septic shock. One open-label, single-center trial randomized 154 patients with septic shock to an esmolol infusion or standard therapy [83]. All patients on esmolol achieved the preset target heart rate (HR) of 80 to 94 beats/min without a significant drop in mean arterial pressure (median dose 100 mg/hour). Compared to patients on standard therapy, patients receiving esmolol had improvements in the following variables at 96 hours:
●Greater decline from baseline HR (by 28 beats/min versus 6 beats/min)
●Higher left ventricular stroke work index (43 versus 31 mL/m2)
●Reduced need for vasopressors (-0.11 mcg/kg/min versus -0.01 mcg/kg/min)
●Reduced need for fluid replacement therapy (4L/day versus 5.4 L/day)
Improvements were also observed in secondary outcomes that included metabolic variables (eg, lactate and pH), markers of end organ function (eg, glomerular filtration rate), and 28-day mortality (50 versus 81 percent). Another trial of landiolol in patients with sepsis-related tachycardia resulted in a reduction in the rate of new-onset tachyarrhythmia [84]. A meta-analysis of seven randomized trials of esmolol or landiolol reported a possible increase in survival associated with short-acting beta blockade [85]. Further validation of these findings is warranted before esmolol can be routinely recommended as a therapy in patients with septic shock.
Ivabradine also lowers heart rate through inhibition of ion channels in the sinoatrial node. In an animal model of sepsis, ivabradine mitigated septic tachycardia similar to atenolol [86]. Unlike atenolol, ivabradine did not decrease cardiac output or left ventricular ejection fraction. In a randomized, controlled study of clinical septic shock, enteral ivabradine led to decreased heart rates, lower SOFA scores, and increased stroke volume and ejection fraction compared to placebo. However, ivabradine had no effect on cardiac output or mortality. Accordingly, ivabradine cannot be recommended therapy for septic shock at this time.
Vitamin C — Vitamin C has gained popularity as a therapy for sepsis, either alone or in combination with thiamine and hydrocortisone. Hydrocortisone and vitamin C act synergistically on the inflammatory cascade. Vitamin C is also an antioxidant. Thiamine (vitamin B1) prevents renal oxalate crystallization when vitamin C is administered in high doses.
While one early retrospective study suggested a possible reduction in mortality, most studies since then have not been favorable for vitamin C-containing regimens [87-94], supporting the Society of Critical Care Medicine/European Society of ICU Medicine recommendation against its use [11].
Studies favoring vitamin C regimens:
●One retrospective single institution study of 47 patients with sepsis or septic shock and a procalcitonin level >2 ng/mL, examined the impact of high-dose intravenous (IV) vitamin C (1.5 g every six hours for four days or until ICU discharge), IV thiamine (200 mg every 12 hours for four days or until ICU discharge), and IV hydrocortisone (50 mg every six hours for seven days or until ICU discharge) given early during the course of sepsis (ie, within the first 24 hours of intensive care [ICU] admission) [87]. This regimen was associated with a reduction in in-hospital mortality (8.5 versus 40.4 percent), more rapid weaning of vasopressors (18 versus 55 hours), and the prevention of progressive organ dysfunction compared with 47 historical controls who were treated in an era prior to the institution of the vitamin C protocol.
●A preliminary randomized trial of 167 patients with sepsis and acute respiratory distress syndrome reported no effect of a 96-hour infusion of vitamin C on organ failure (as assessed by the SOFA score), levels of inflammatory markers (C-reactive protein), or vascular injury (thrombomodulin) [90]. While vitamin C reduced the crude 28-day mortality (30 versus 46.3 percent) and ICU- and hospital-free days, the analyses were not adjusted for confounding variables and are therefore considered exploratory.
Studies not favoring vitamin C:
●Three randomized trials of patients with sepsis or septic shock compared a regimen of IV vitamin C (1.5 g every six hours), IV hydrocortisone (50 mg every six hours), and IV thiamine (200 mg every 12 hours) with either IV hydrocortisone alone [91] or placebo [93,94]. None of the trials showed improvement in 28-day mortality, ventilator- or vasopressor-free days, or rate of acute kidney injury. Adverse events included hyperglycemia (6 versus 3 percent), hypernatremia (5 versus 3 percent), and new hospital-acquired infection (6 percent each). However, the trials had several limitations, including open-label design and lack of blinded outcome assessment for one [91] and another was underpowered to detect a significant difference, which may affect the interpretation of these data.
●Another study of 872 patients with septic shock who were receiving a vasopressor, IV vitamin C alone had no effect on 28-day mortality (53 versus 32 percent) or persistent organ dysfunction (defined by the use of vasopressors, invasive mechanical ventilation, or new renal-replacement therapy; 9 versus 7 percent) [95]. One patient had severe hypoglycemia and another had anaphylaxis in response to vitamin C.
●Several meta-analyses suggest similar lack of benefit [96-98], but several studies remain in progress.
Targeting bioenergetic failure — Mitochondrial dysfunction is present in sepsis and could contribute to immune failure, cell death, and multiorgan impairment. For example, activating mitochondrial biogenesis using inhaled carbon monoxide reduced liver injury in a model of sepsis [99]. Treatments that protect mitochondria or enhance mitochondrial biogenesis might prove effective.
INEFFECTIVE THERAPIES — A number of potential therapies for sepsis have been investigated, but either caused harm or failed to improve clinical outcomes.
Despite promising initial data, recombinant human activated protein C (rhAPC) has not been confirmed to improve survival in patients with severe sepsis or septic shock, prompting withdrawal of this drug from the market [100]. Recombinant human activated protein C (also called drotrecogin alfa) promotes fibrinolysis and inhibits thrombosis. It was hypothesized that rhAPC may benefit patients with sepsis because it modulates the procoagulant response that is believed to contribute to multisystem organ dysfunction. This hypothesis was initially tested in the PROWESS trial, which reported that rhAPC improved 28-day mortality in patients with severe sepsis or septic shock, with its greatest benefit among patients with a high risk of death (ie, APACHE II score ≥25) [101]. However, conflicting data from subsequent studies eventually led to a new trial, the PROWESS-SHOCK trial. In this trial, 1696 patients with vasopressor-dependent septic shock were randomly assigned to receive rhAPC or placebo [102]. Preliminary analyses done by the maker of the drug indicated that rhAPC did not improve 28-day mortality (26.4 versus 24.2 percent for placebo, RR 1.09, 95% CI 0.92-1.28).
Additional therapies that have proven ineffective include the following:
●The Toll-like receptor (TLR)-4 antagonist, TAK 242 (Resatorvid) [2]
●The human anti-endotoxin monoclonal antibody, HA-1A [103,104]
●The human anti-Enterobacteriaceae common antigen (ECA) monoclonal antibody [105]
●Alkaline phosphatase [106]
●Granulocyte colony-stimulating factor (filgrastim, G-CSF) [107]
●Anti-tumor necrosis factor monoclonal antibody [108-110]
●Tumor necrosis factor receptor antagonist [109,111]
●Interleukin-1 receptor antagonist [30]
●Antithrombin (formerly known as antithrombin III) [11,112-119]
●Recombinant human tissue factor pathway inhibitor (tifacogin) [120,121]
●N-acetylcysteine [123,124]
●Nitric oxide inhibitors [125-131]
●The bradykinin antagonist, deltibant [132]
●Growth hormone [133]
●Intravenous selenium supplementation [11,134]
●Talactoferrin – restores the barrier properties of the gastrointestinal mucosa [135-139]
●Calcitriol [140]
●Levosimendan [141]
●Hypothermia [142]
●Hyperoxia [143]
●Hemoperfusion through a membranous polymyxin B fiber column (PBFC) [19,20]
●Recombinant human soluble thrombomodulin (for sepsis-associated coagulopathy) [63]
●High-dose vitamin D (in vitamin D deficient patients) [144]
SUMMARY AND RECOMMENDATIONS
●A number of potential therapies for sepsis appear promising in animal models but have not yet been adequately studied in humans. These include toll-like receptor antagonists and neutralizing antibodies, iloprost, interferon gamma, macrophage migration inhibition factor neutralizing antibody, and a synthetic peptide that inhibits bacterial superantigen-induced expression of certain proinflammatory genes. (See 'Inhibition of innate immunity' above and 'Interferon-gamma' above and 'Augmentation of immunomodulation' above and 'Inhibition of proinflammatory gene expression' above.)
●Other potential therapies for sepsis have been studied in humans, but have provided conflicting results and require additional investigations to clarify their effects. These include polyclonal intravenous immune globulin (IVIG), interleukin-1 receptor antagonists, hemoperfusion through adsorptive materials or membranes, plasma exchange, whole blood exchange, coupled plasma filtration adsorption, granulocyte-macrophage colony stimulating factor (GM-CSF), hemofiltration, anticoagulants, naloxone, pentoxifylline, and statins. (See 'Intravenous immune globulin' above and 'Cytokine and endotoxin inactivation or removal' above and 'Granulocyte-macrophage colony stimulating factor' above and 'Hemofiltration' above and 'Anticoagulants' above and 'Naloxone' above and 'Pentoxifylline' above and 'Statins' above.)
●Potential therapies that require further validation of benefit in patients with septic shock include therapy with short-acting beta blockers esmolol or landiolol. (See 'Heart rate control' above.)
●Potential therapies for sepsis have also been investigated that were found to either cause harm or not improve clinical outcomes. (See 'Ineffective therapies' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Peter F Clardy, MD, who contributed to earlier versions of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟