The incidence of sepsis (severe infection) has increased over the last four decades. Severe sepsis and septic shock are among the leading causes of death for patients admitted to critical care units with mortality ranging from 20-70% depending on totality of organ dysfunction. Outside of antibiotics and good bedside care, little has changed in the management of this life-threatening problem.

Therapeutic plasma exchange (TPE) involves the separation of plasma from whole blood. The removed plasma is 'exchanged or replaced' with either IV fluids, albumin, blood products or a combination thereof.

The primary objective of this study is to evaluate the safety of the TPE intervention protocol within 24 hours of study criteria being met. TPE is now a well-established program at the South Health Campus for neuro-muscular disorders. Since starting in May 2018, the investigators have performed over 150 runs making the SHC ICU one of the most experienced centers in Canada.

Background

The incidence of sepsis has increased over the last four decades (1). Sepsis is a life-threatening condition that arises when the body's response to an infection injures its own tissues and organs. Severe sepsis and septic shock are among the leading causes of death for patients admitted to critical care units with mortality ranging from 20-70% depending on totality of organ dysfunction (2, 3). The literature is replete with initial promising phase 2 therapies failing in definitive randomized trails (4-8). In fact, a recent systematic review concluded that no evidence exists for any pharmacologic intervention that has consistently reduced mortality in critically ill patients (9). This is both surprising and frustrating for the author. The most recent guidelines have tried to redefine sepsis as a 'syndrome' since neither validated criterion nor do standard diagnostic tests exist (10). The authors argue that sepsis should be viewed as organ dysfunction caused by a dysregulated or non-homeostatic host response. Most of the clinical manifestations of severe infections are caused by an intense, generalized inflammatory response in the host mediated by a multitude of interrelated cellular and humoral factors (3).

Plasmapheresis or therapeutic plasma exchange (TPE) involves the separation of plasma from whole blood. The removed plasma is 'exchanged or replaced' with crystalloids, albumin, fresh frozen plasma or a combination thereof. TPE use is well established in many neurological disorders including Guillain-Barre syndrome (11), Myasthenia Gravis (12, 13) and antibody mediated syndromes(14, 15). It is considered the standard of care for thrombotic thrombocytopenic purpura (TTP) (16, 17). The rationale for the use of TPE in sepsis, a non-selective intervention, is to remove multiple toxic mediators including endotoxins, activated complement, pro-inflammatory cytokines and pro-coagulant factors (18, 19). If fresh-frozen plasma is used as replacement fluid, consumed plasma factors are substituted, thereby possibly restoring the opsonic capacity and improving the coagulation abnormalities and microcirculation.

Plasma exchange has been reported since the late 1970s as a potential adjunctive or salvage therapy in severe sepsis in both pediatric and adult patients (20-24). These case reports, retrospective reviews and observational studies suggest a survival advantage when compared to historical controls. However, the obvious bias limits any meaningful interpretation. A literature review found only 4 studies with any attempt at randomization. One study enrolled only adults (25), two were exclusively pediatric patients (26, 27) and one study involved both adults and children (28). Excluding the pediatric studies, the adult protocols had few similarities:

1. Reeves and colleagues attempted a multi-center Australian study but terminated enrollment after 22 adult and 8 children (28). The mean APACHEII scores for adults were 25.2. They aggressively exchanged 5 plasma volumes continuously over 36 hours using a combination of fresh frozen plasma (FFP) and albumin (1/4 ratio). Mortality was reported at 14 days. No data on ICU or hospital length of stay was provided. This trial reported significant decrease in certain inflammatory markers.
2. Busund's larger trail involved 106 adults and reported mortality at 28 days (25). They performed a single 30-40mls/kg exchange that could be repeated once if no clinical improvement was observed. The replacement used was FFP and albumin in a 1:1 ratio. Six episodes of transient hypotension and 1 allergic reaction to FFP was reported (the only trial to report adverse events). There was an encouraging trend towards improved survival (33% vs 53%) versus historical controls. No data on ICU or hospital length of stay was provided.

A more recent German pilot study tried to evaluate the safety and feasibility of enrollment within 12 hours of shock for a proposed randomized controlled trial (RCT) (29). The inclusion criteria were simple: (a) Current sepsis-3 definitions; noradrenaline dose >0.4ug/kg/min despite >30mls/kg of crystalloid; (b) Less than 12 hours of vasopressor support. They performed a single 1.2x plasma exchange using entirely FFP. The major findings were significantly less vasopressor support, fluid balance and a decline in plasma concentrations of pro-inflammatory mediators. The "sustained -responders" mortality was 43% vs 77% of non-responders, but the small numbers prevented statistical significance.

The use of plasma exchange in severe sepsis is graded by the American Society for Apheresis as category III with grade 2C indications, indicating that there is a lack of reliable trials to support TPE use in the condition (30). The purpose of this phase 1 clinical trial is to assess the safety of using plasma exchange in critically ill adult patients with septic shock. TPE is now a well-established program at the South Health Campus (SHC) for neuro-muscular disorders. Since starting in May 2018, we have performed over 240 runs in 43 patients making the SHC ICU the most experienced TPE center in Canada in the past 2 years. One plasma volume is currently the standard dose used safely in patients with neuro-muscular disorders at SHC.

Justification of Research

1. The incidence of sepsis has increased over 4 decades but no single pharmacological invention reduces mortality/morbidity.
2. A potential treatment that increases the chance of survival and recovery in patients with septic shock would be both beneficial to patients and help decrease medical costs from long term stays in the ICU.
3. The maximum tolerable dose of TPE in septic patients is unknown.

Study Objectives

Primary Objective: To evaluate the safety of the TPE intervention protocol within 24 hours of study criteria being met.

Secondary Objectives:

1. To evaluate preliminary efficacy of the TPE protocol.
2. To determine maximum tolerable dose of TPE in septic patients.
3. To record adverse events from TPE intervention.

Hypothesis

It is hypothesized that TPE will be safe for use in patients with septic shock. We will determine the maximum tolerable dose via a dose escalation study.

Inclusion Criteria Adult patients (age ≥18) with a documented or strong clinical suspicion of infection that meets the definition of septic shock as per the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3).

Exhibiting two of the four clinical signs of inflammation:

1. Core temperature > 38oC or < 36oC
2. Heart rate > 90 beats per minute
3. Respiratory rate > 20 breaths per minute, or PaCO2 < 32 mmHg, or mechanical ventilation
4. White cell count > 12 x 109/L or < 4 x 109/L or > 10% immature neutrophils

We will further identify the subset with a hospital mortality in excess of 40%:

1. >30 mls/kg fluid resuscitation
2. Noradrenaline >0.1 ug/kg/min to maintain MAP> 65mmHg for at least 4 consecutive hours and present at initiation of TPE
3. Lactate >2 mmol/l.

Exclusion Criteria Patients will be excluded in cases where death is deemed inevitable or imminent during admission and either the attending physician, patient or surrogate legal decision maker is not committed to active treatment.

Once the results have been analyzed and published the de-identified data will be made available to other researchers

The data will be made available once published. It will be made available for 5 years as per University of Calgary data storage protocol.