Departments of Internal Medicine and Community Health Sciences, University of Manitoba; and Department of Medical Oncology and Haematology, CancerCare Manitoba, Winnipeg, Canada Section of Critical Care Medicine, University of Manitoba, Winnipeg, Canada Section of Critical Care Medicine, University of Manitoba, Winnipeg, Canada; and Department of Medical Oncology and Haematology, CancerCare Manitoba, Winnipeg, Canada
Sepsis and septic shock are leading causes of intensive care unit (ICU) mortality. They are characterized by excessive inflammation, upregulation of procoagulant proteins and depletion of natural anticoagulants. Plasma exchange has the potential to improve survival in sepsis by removing inflammatory cytokines and restoring deficient plasma proteins. The objective of this study is to evaluate the efficacy and safety of plasma exchange in patients with sepsis.We searched MEDLINE, EMBASE, CENTRAL, Scopus, reference lists of relevant articles, and grey literature for relevant citations. We included randomized controlled trials comparing plasma exchange or plasma filtration with usual care in critically ill patients with sepsis or septic shock. Two reviewers independently identified trials, extracted trial-level data and performed risk of bias assessments using the Cochrane Risk of Bias tool. The primary outcome was all-cause mortality reported at longest follow-up. Meta-analysis was performed using a random-effects model.Of 1,957 records identified, we included four unique trials enrolling a total of 194 patients (one enrolling adults only, two enrolling children only, one enrolling adults and children). The mean age of adult patients ranged from 38 to 53 years (n = 128) and the mean age of children ranged from 0.9 to 18 years (n = 66). All trials were at unclear to high risk of bias. The use of plasma exchange was not associated with a significant reduction in all-cause mortality (risk ratio (RR) 0.83, 95% confidence interval (CI) 0.45 to 1.52, I(2) 60%). In adults, plasma exchange was associated with reduced mortality (RR 0.63, 95% CI 0.42 to 0.96; I(2) 0%), but was not in children (RR 0.96, 95% CI 0.28 to 3.38; I(2) 60%). None of the trials reported ICU or hospital lengths of stay. Only one trial reported adverse events associated with plasma exchange including six episodes of hypotension and one allergic reaction to fresh frozen plasma.Insufficient evidence exists to recommend plasma exchange as an adjunctive therapy for patients with sepsis or septic shock. Rigorous randomized controlled trials evaluating clinically relevant patient-centered outcomes are required to evaluate the impact of plasma exchange in this condition.
Objectives: To characterize the prevalence, incidence, and temporal evolution of thrombocytopenia (platelets < 100 × 10 9 /L) in septic shock and to investigate the independent association of thrombocytopenia on clinical outcomes. Design: Retrospective, propensity-matched, cohort study. Setting: Two academic ICUs in Winnipeg, Canada. Patients: Nine-hundred eighty adult patients diagnosed with septic shock between 2007 and 2012. Interventions: Propensity-matched cohort analysis and Cox proportional hazard model evaluating thrombocytopenia over time. Measurements and Main Results: Of 980 adults, 165 patients (16.8%) had thrombocytopenia at ICU admission (prevalent), whereas 271 (27.7%) developed thrombocytopenia during ICU admission (incident). Among patients with incident thrombocytopenia, the median time from ICU admission to thrombocytopenia was 2 days (interquartile range, 1–3 d). Among survivors, the median time from incident thrombocytopenia to platelet recovery was 6 days (interquartile range, 4–8 d). The median time from liberation of vasopressors to recovery of platelets concentration (≥ 100 × 10 9 /L) was 2 days (interquartile range, 0–4 d). In a propensity-matched analysis, thrombocytopenia was associated with increased durations of ICU length of stay (9 vs 6 d; p < 0.01), mechanical ventilation (7 vs 4 d; p < 0.01), and vasopressor use (4 vs 3 d; p < 0.01), as well as increased major bleeding events (41% vs 18%; p < 0.01). In an adjusted Cox proportional hazards model, thrombocytopenia was significantly associated with both increased ICU mortality (hazard ratio, 1.99; 95% CI, 1.51–2.63) and hospital mortality (hazard ratio, 1.93; 95% CI, 1.48–2.51). Conclusions: Both the prevalence and incidence of thrombocytopenia are high in septic shock. Incident thrombocytopenia occurs early in septic shock, and platelet recovery lags behind clinical recovery. In septic shock, thrombocytopenia is associated with increased length of stay, longer duration of organ support, major bleeding events, and mortality.
