Extracorporeal Membrane Oxygenation Support Following Stem Cell Transplant—When Is All That We Have Still Not Enough?*

    loading  Checking for direct PDF access through Ovid


Even as the use of extracorporeal membrane oxygenation (ECMO) support becomes more commonplace, many questions remain regarding appropriate indications for its use. In this issue of Critical Care Medicine, Wohlfarth et al (1) review the outcomes of ECMO support of acute respiratory failure (ARF) following adult allogeneic stem cell transplantation (ASCT), herein presenting the largest single series collected to date.
This study has important implications for critical care and bone marrow transplant practitioners considering offering ECMO support to ASCT recipients who develop severe ARF. Historical data suggest that this population carries high mortality despite conventional treatment; the question of whether or not referring for ECMO can contribute to salvaging more of these patients represents a major clinical decision point.
Historically, active immunosuppression has been associated with worse outcomes in ECMO (2, 3). A history of hematologic transplantation is considered a deterrent or outright contraindication to initiating ECMO by many centers. Given the reported improved outcomes and rapid increase in adult respiratory ECMO support over the past 5 years (4), it is very relevant to reevaluate the use of extracorporeal support in this context.
This report covers a multinational, retrospective, observational study of veno-venous-ECMO following ASCT conducted in 12 European centers between 2010 and 2015. The authors identified 37 adult ASCT recipients who received ECMO after developing severe ARF. The criteria for initiating ECMO were not explicitly controlled between centers. The patients in this series were young (median age, 37 yr; range, 26–49 yr) and critically ill, with low P/F ratios, high rates of vasopressor use, renal replacement therapy, and nonpulmonary, nonhematologic organ dysfunction.
The group was heterogeneous with respect to underlying disease process (hematologic malignancy, acute leukemia, and other malignant and nonmalignant diseases), concurrent receipt of immunosuppressive therapies, status of remission, and presence of graft versus host disease (GvHD). Given the small number of patients overall, the authors are appropriately cautious in discussing interpretation of statistical findings.
Importantly, the authors analyzed survival by time from ASCT and identified that only one of 24 patients (4%) receiving ECMO in the early posttransplant period, within 240 days of transplant, survived. A much higher survival rate of seven of 13 patients (46%) was seen in patients in whom ECMO was initiated greater than 240 days following transplant. Although the number of patients evaluated was small, this is a potentially very meaningful difference, particularly if survival in similar patients without ECMO is only 10-20%.
The authors were also able to identify other high-mortality subgroups in this series. Acute leukemia was the most common underlying diagnosis, with an associated mortality of 92% (22/24). There were no survivors in the lymphoma group, the acute GvHD group, nor in the patients receiving more than one ECMO run. There was only one survivor in seven malignancy patients who were not in remission before or after engraftment. Conversely, survivors were predominantly seen in the group with ARF secondary to pneumonia, which was the most represented diagnosis. Only one or two patients were seen in other groups that included diffuse alveolar hemorrhage, vasculitis, bronchiolitis obliterans, or abdominal sepsis.
ECMO was usually initiated early in these patients, at a median interval of 2 days following intubation. The median time on extracorporeal support was 2 weeks, with a median ICU length of stay of 1 month. As would be expected, significant thrombocytopenia was often present, and the bleeding complication rate was substantial. Anticoagulation management was altered or held entirely in almost 40% of patient; despite this, 38% of patients experienced bleeding events, including intracranial events in 13% of patients, and uncontrolled bleeding ensued in three others. Six cases of hemorrhage directly resulted in fatal outcomes.

Related Topics

    loading  Loading Related Articles