The interaction of thrombocytopenia, hemorrhage, and platelet transfusion in venoarterial extracorporeal membrane oxygenation: a multicenter observational study

Abstract

Background: Thrombocytopenia, hemorrhage and platelet transfusion are common in patients supported with venoarterial extracorporeal membrane oxygenation (VA ECMO). However, current literature is limited to small single-center experiences with high degrees of heterogeneity. Therefore, we aimed to ascertain in a multicenter study the course and occurrence rate of thrombocytopenia, and to assess the association between thrombocytopenia, hemorrhage and platelet transfusion during VA ECMO.

Methods: This was a sub-study of a multicenter (N = 16) study on transfusion practices in patients on VA ECMO, in which a retrospective cohort (Jan-2018-Jul-2019) focusing on platelets was selected. The primary outcome was thrombocytopenia during VA ECMO, defined as mild (100-150·10⁹/L), moderate (50-100·10⁹/L) and severe (< 50·10⁹/L). Secondary outcomes included the occurrence rate of platelet transfusion, and the association between thrombocytopenia, hemorrhage and platelet transfusion, assessed through mixed-effect models.

Results: Of the 419 patients included, median platelet count at admission was 179·10⁹/L. During VA ECMO, almost all (N = 398, 95%) patients developed a thrombocytopenia, of which a significant part severe (N = 179, 45%). One or more platelet transfusions were administered in 226 patients (54%), whereas 207 patients (49%) suffered a hemorrhagic event during VA ECMO. In non-bleeding patients, still one in three patients received a platelet transfusion. The strongest association to receive a platelet transfusion was found in the presence of severe thrombocytopenia (adjusted OR 31.8, 95% CI 17.9-56.5). After including an interaction term of hemorrhage and thrombocytopenia, this even increased up to an OR of 110 (95% CI 34-360).

Conclusions: Thrombocytopenia has a higher occurrence than is currently recognized. Severe thrombocytopenia is strongly associated with platelet transfusion. Future studies should focus on the etiology of severe thrombocytopenia during ECMO, as well as identifying indications and platelet thresholds for transfusion in the absence of bleeding.

Trial registration: This study was registered at the Netherlands Trial Registry at February 26th, 2020 with number NL8413 and can currently be found at https://trialsearch.who.int/Trial2.aspx?TrialID=NL8413.

Keywords: Hemorrhage; Platelet transfusion; Thrombocytopenia; Venoarterial extracorporeal membrane oxygenation.

Fig. 1

Progression of thrombocytopenia over time. This figure describes the proportion of patients suffering from a thrombocytopenia during VA ECMO. In Panel (A), the y-axis shows the absolute number of patients. In Panel (B), a stacked bar-plot shows the proportion of patients per day with a certain degree of thrombocytopenia. No., number, VA ECMO, venoarterial extracorporeal membrane oxygenation

Fig. 2

Course of platelet count over time, stratified by severity of thrombocytopenia at admission. This figure describes the course of platelet count over time as median with 1st–3rd quartile, as stratified by the degree of severity of thrombocytopenia as measured at the day of admission (before VA ECMO). VA ECMO, venoarterial extracorporeal membrane oxygenation

Fig. 3

Odds to receive a platelet transfusion during VA ECMO. Odds ratio + 95% confidence interval. In case of adjusted: adjusted for confounding factors. Confounding factors include: sex, age, history of cardiovascular disease, SOFA score at day of ECMO initiation, cannulation site (reference: peripheral), daily aPTT, a thrombotic complication during ECMO and anticoagulation type (reference: unfractionated heparin)