Introduction: Thrombocytopenia-associated multiple organ failure (TAMOF) is a clinical phenotype characterized by endothelial dysfunction, ADAMTS13 deficiency, and consumptive coagulopathy. Though often caused by sepsis, TAMOF can also arise from noninfectious triggers like cardiopulmonary bypass. Therapeutic plasma exchange (TPE) may mitigate the effects of TAMOF and can be integrated with extracorporeal membrane oxygenation (ECMO). However, its role in pediatric patients, especially across sepsis phenotypes and in cardiac disease, remains unclear. We aimed to assess survival and organ dysfunction outcomes in children with sepsis or TAMOF who received TPE versus those who did not while on ECMO.
Methods: We conducted a single-center retrospective cohort study of patients aged 0–21 years who received ECMO and met sepsis or TAMOF criteria in the pediatric or cardiac ICU from 2011–2024. Patients who received TPE for other indications (e.g., antibody-mediated rejection, myocarditis) were excluded. The primary outcome was survival to hospital discharge.
Results: Of 120 patients included, 27% (n=32) underwent TPE while on ECMO. Median age was 6 months [IQR 1–71]. 52% were female. Venoarterial ECMO (vs. venovenous) was associated with worse survival (p=0.018). Median ECMO duration was 4 days [3–7]; TPE use correlated with longer ECMO runs (p=0.006). Overall survival to discharge was 49%, with no significant difference between TPE and no-TPE groups (p=0.332). Among patients with Gram-positive infections who underwent TPE, survival correlated with a greater median number of sessions (4 vs. 2; p=0.001). Among TPE recipients, confirmed infection was linked to earlier TPE initiation (within the first day of ECMO; p< 0.001). Cardiac disease was present in 64% and associated with lower TPE use (OR 0.32 [0.14-0.74]; p=0.007); in this subgroup, TPE was linked to lower survival (p=0.018).
Conclusions: TPE was not associated with overall survival benefit in pediatric patients with sepsis or TAMOF on ECMO. In patients with Gram-positive infections, more TPE sessions correlated with better survival. The timing of TPE initiation and infection burden may influence outcomes and warrant further investigation. In cardiac patients, TPE was associated with lower survival, though this likely reflects use in more critically ill cases.
