Lafayette, Richard A. MD
Up to 750,000 cases of sepsis are seen each year in the United States alone, involving 25% of all intensive care unit (ICU) patients. About one-quarter of those with sepsis die, and the mortality rate is higher in septic shock—nearly 50%.1
Richard A. Lafayette...Image Tools
Nephrologists are commonly brought into the care of these individuals when acute kidney injury (AKI) or fluid and electrolyte disturbances develop. The treatment of patients with AKI has left nephrologists well experienced with both intermittent and continuous extracorporeal therapies.
Results from recent studies may drive increasing support for extracorporeal therapies in patients with sepsis—not for organ support alone, but to directly alter the course of the condition—which may also mean more demand for these types of services.
Nephrologists must stay abreast of progress in the field and actively participate in trials testing the efficacy of these interventions.
Rationale for Extracorporeal Therapy
Sepsis is caused by infection, but it is the inflammatory response to infection that seems to underlie the likelihood of multiorgan failure and death. Thus, the cellular and cytokine response to sepsis has been closely evaluated.
An optimal inflammatory response is inhospitable to organisms and clears them from the body, but a deleterious response can lead to systemic injury, including multiple organ failure, through cytotoxicity.
Prolonged sepsis perversely can lead to impaired immunity and increased susceptibility to secondary infections, either through depletion of factors/cells or through the activation of anti-inflammatory cytokines and cells.2
Therefore, it is highly desirable to find a way to optimally control both the infection and the inflammatory response in septic patients.
Standard treatment focuses on antimicrobial therapy and hemodynamic support with fluids, as well as organ support when indicated. Extracorporeal therapy has been advocated as a potential way to limit the inflammatory response or even to assist in antimicrobial treatment.
Initial enthusiasm for the approach was heightened by findings that high-volume (35 and 45 mL/kg/h versus 20 mL/kg/h) continuous ultrafiltration therapy for AKI seemed to confer hemodynamic and mortality benefits, particularly in septic patients.3
These results were attributed to reductions in peak concentrations of cytokines and other mediators of septic injury. Unfortunately, extensive randomized and observational study results have not confirmed the observed advantages.4,5
However, the desire to affect the inflammatory nature of sepsis through extracorporeal therapy remains strong, and many novel approaches continue to show promise.
In fact, authors use the term “extracorporeal purification” to describe these efforts to treat the systemic inflammatory response.6
These systems focus on ways to remove harmful mediators of sepsis, such as endotoxin or cytokines, or shift the balance toward more anti-inflammatory mediators than inflammatory mediators. Other techniques aim to support the immune system in sepsis by aiding in immune reconstitution.
Hemofiltration: Volume and Solute Control
Hemofiltration can remove or bind large amounts of circulating cytokines and allows for volume and solute control.
As discussed above, the use of moderately high-dose ultrafiltration versus standard-dose ultrafiltration has not proven helpful in septic AKI patients or in those with sepsis alone. Thus, even higher doses have been utilized to attempt to control inflammation in sepsis.
High-volume hemofiltration aims to provide an average of over 45 mL/kg/h, either continuously or in pulsed doses. Reported benefits from small trials include reduction in vasopressor requirements, improvement in blood pressure, and even a decrease in mortality rates compared with calculated expected mortality rates.7
There is no obvious effect of the technique on circulating cytokine levels, which is perhaps surprising given the design of these studies. The mechanism of any benefit is uncertain, although reductions in apoptotic and anaphylactic mediators have been suggested.8
Downsides of the approach include costs, risks of medication and nutrient losses, and risks associated with the circuit itself, including blood loss, inflammation, and infection.
A relatively large trial in subjects with sepsis and AKI is slowly progressing, and perhaps more information will emerge in the next couple of years.
Higher Cutoff Membranes
Due to the intrinsic size of cytokines, efforts to increase their removal have included the design and use of higher molecular weight cutoff membranes that incorporate larger pore diameters of up to 10 nm.
Small pilot studies have demonstrated superior removal of some, but not all, cytokines using these membranes and have even suggested certain hemodynamic benefits such as reduced pressor needs.9
However, given concerns about cost and higher loss of nutrients, medications, albumin, and other proteins, stronger findings of clinical benefits will be needed before moving toward clinical application. One approach to using this technique focuses on greater use of dialytic clearance to limit protein loss.
Hemoperfusion with Polymyxin B
Polymyxin B was effective as an antibiotic given its ability to bind endotoxin, especially in gram-negative infections. Systemic use of polymyxin B is limited due to its substantial toxicity, especially neurotoxicity and nephrotoxicity.
As an alternate approach, hemoperfusion with polymyxin B-immobilized fibers has been tested in patients with sepsis.10 Polymyxin B can absorb endotoxin and, theoretically, attenuate the inflammatory process of sepsis.
The greatest efficacy is expected for gram-negative sepsis, but a more general utility may be seen, as these columns also are able to absorb activated leukocytes and thereby may limit inflammation by this avenue as well.
Furthermore, it is well established that circulating endotoxin levels go up in many patients with sepsis, especially severe sepsis, regardless of the initiating organism.
Small studies showing positive effects of these polymyxin-B cartridges on endotoxin levels, hemodynamics, and especially mortality have led to their ongoing clinical use in Japan for two decades and their recent regulatory approval in Europe.
Patients are treated early in the course of sepsis with hemoperfusion at low blood flow rates for two to three hours, with repeat treatment after 24 hours.
In a meta-analysis of 28 randomized and observational studies that included more than 1,400 patients, hemoperfusion with polymyxin B had favorable effects on blood pressure, dopamine use, oxygen status, and mortality.11
The Early Use of Polymyxin B Hemoperfusion in Abdominal Sepsis (EUPHAS) study was a multicenter, randomized trial of 64 patients with abdominal sepsis that compared standard therapy with or without two sessions of polymyxin B hemoperfusion.12 The trial was stopped early due to lower mortality in the treated arm, which also enjoyed better hemodynamics and a lower Sequential Organ Failure Assessment (SOFA) score at Day 3.
However, the study was not designed as a mortality study, and it has been criticized for stopping early.
A larger study, EUPHRATES (Evaluating the Use of Polymyxin-B Hemoperfusion in a Randomized, Controlled Trial of Adults Treated for Endotoxemia and Septic Shock), is under way in the United States to evaluate the potential of this therapy. The trial organizers hope to enroll 360 patients, and the study should be adequately powered to assess clinical benefits in a broader population of patients with sepsis. It seems prudent to wait for these results before implementing the clinical use of this device in the ICU.
Efforts have also been made to assess plasma filtration as a way of lowering the inflammatory potential of circulating blood.
Plasma therapy has been accomplished through plasmapheresis or plasma exchange with limited data thus far but some suggestions of improvements, particularly in gram-negative sepsis.
In another approach, plasma filtration has been coupled to devices, allowing adsorption of specific molecules. In coupled plasma filtration, the plasma is returned to the patient; there is no need for fluid replacement. Using plasma rather than whole blood limits effects on circulating cells, preventing thrombocytopenia, which can be a significant side effect of adsorptive therapy.
Preliminary studies have suggested that coupled plasma filtration and adsorption can reduce proinflammatory cytokines, with a tendency toward improved hemodynamics and less organ failure, but the evidence is quite limited, and further studies are planned.13
Use of cartridges that absorb cytokines is also under active investigation. One such system is the CytoSorb, which utilizes a synthetic polymer-based cytokine-absorbent system. It is effective at removing cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-6, and IL-1.
Animal experiments support the hypothesis that the treatment confers benefits in sepsis. In humans, these cartridges have shown some ability to significantly reduce targeted cytokines while providing some moderate clinical benefits.14 Further development of the cartridges and techniques should yield interesting results.
Efforts have also been made to expose circulating blood to cells outside the body in order to add antimicrobial or inflammatory-modulating properties.
In one method, patients' plasma is separated and then exposed to a cartridge of donor granulocytes in an effort to clear mediators of sepsis and perhaps add immune responses that may have been depleted systemically. Animal studies have suggested significant benefits.
A pilot study of 10 patients with septic shock demonstrated reduced need for pressor support and improvements in some biomarkers of sepsis, including endotoxin.15
Likewise, an effort was made to use a “renal-assist device,” which exposed ultra-filtered blood to a cartridge of renal tubule cells. In animals, improvements in hemodynamics and cytokine profiles were seen. Despite promising initial human studies in acute kidney injury and sepsis, clinical benefits were not validated in subsequent research. Further studies are awaited.
Lead the Way
The past decades have seen tremendous innovation in the development of techniques to alter the blood and plasma composition of patients with sepsis through extracorporeal circuits. The desire has been to create a more favorable environment, allowing individuals to overcome sepsis and avoid, or more readily recover from, the complications of multiple organ failure.
These approaches also aim to allow the immune system to stay stronger, preventing secondary infections—a major cause of mortality in prolonged sepsis. There is great hope that some of the techniques will lead to improvements in the rather dismal prognosis of patients with sepsis, especially those we commonly see in the ICU.
Additional optimism comes from the idea that these approaches could be used in combination, perhaps further improving their efficacy—for example, combining high cutoff membranes with highly efficient dialysis.
Functionally, these approaches could also be used in combination with extracorporeal therapies already aiding subjects with organ failure, such as dialysis, artificial liver therapy, or even extracorporeal membrane oxygenation.
Unfortunately in our field, we have seen many early positive studies not fulfill their promise when subjected to well-designed randomized clinical trials.
All of these approaches to sepsis are yet to be proven in a convincing manner. None of them have been adequately studied to recommend and, in fact, are not part of the international best practice campaign of “Surviving Sepsis.”16
Nonetheless, many large trials are under way. In the effort to improve the outcome of subjects with sepsis through extracorporeal therapy, nephrologists should lend a hand and lead the way.