Can children receiving extracorporeal membrane oxygenation (ECMO) be enrolled early and reliably managed using different prophylactic platelet transfusion thresholds? The ECSTATIC feasibility trial addressed this practical question in 50 children across ten centers.
The trial compared transfusion triggers of <50 × 109/L and <90 × 109/L in children with no or minimal bleeding. It achieved clear separation in pre-transfusion platelet counts and excellent protocol adherence. These findings support a larger definitive trial; they do not establish that the lower threshold is noninferior, safer, or appropriate for every child on ECMO.
Study Design and Population
Nellis and colleagues conducted an open-label, randomized feasibility trial at nine centers in the United States and one in Israel. Children younger than 18 years were eligible within 24 hours of cannulation if bleeding was absent or minimal. Randomization was stratified by age: 28 days or younger versus older children.
Of 159 patients screened, 123 were eligible. Caregivers of 80 eligible patients were approached, and 50 consented. Twenty-five children were assigned to each threshold. Median age was 0.2 years, and 44 of 50 patients received venoarterial ECMO; six received venovenous ECMO. Median time from cannulation to randomization was 18.1 hours.
What the Protocol Tested
The lower-threshold group received prophylactic platelets when the count fell below 50 × 109/L (50,000/µL); the higher-threshold group used 90 × 109/L (90,000/µL). Counts were monitored at least twice daily, and transfusion was to occur within 12 hours of reaching the assigned trigger. These details describe the trial protocol, not a new universal clinical recommendation.
Temporary suspension was permitted for circumstances such as surgery, chest-tube insertion, circuit change, or preparation for decannulation. The assigned strategy continued for up to 21 days, with prespecified stopping conditions including progression to severe bleeding or clotting, ECMO cessation, prolonged suspension, or withdrawal.
Anticoagulation and other hemostatic care followed each center’s local guidance, which could not be changed because of treatment assignment. The primary feasibility endpoint was separation in pre-transfusion platelet count outside temporary suspensions.
Key Results
| Measure | Lower threshold (n=25) | Higher threshold (n=25) |
|---|---|---|
| Platelet transfusion trigger | <50 × 109/L | <90 × 109/L |
| Pre-transfusion count, median (IQR), × 109/L | 43 (39–47) | 79 (70–86) |
| Primary safety composite | 6/25 (24%) | 5/25 (20%) |
| Progression to severe bleeding | 4/25 (16%) | 3/25 (12%) |
| Progression to severe clotting | 1/25 (4%) | 1/25 (4%) |
| 28-day mortality | 3/25 (12%) | 4/25 (16%) |
Selected results from the author manuscript. The safety composite combined progression to severe bleeding or clotting outside temporary suspensions with mortality during the intervention or within 24 hours after it ended. The separate 28-day mortality outcome uses a different observation period.
Meaningful separation and high adherence
After accounting for repeated measurements, the model-adjusted mean difference in pre-transfusion platelet count was 32 × 109/L (P<0.001). Compliance with assigned transfusion decisions was 99.2%. Compliance included both giving platelets when indicated and withholding them when the assigned trigger had not been reached.
Overall, 28 of 50 children received at least one platelet transfusion during the intervention period outside temporary suspensions. Thus, successful protocol adherence should not be confused with every participant receiving a transfusion.
Seven of ten feasibility criteria were met
The study met its prespecified overall feasibility standard, achieving seven of ten criteria when at least five were required. Successful measures included platelet-count separation, eligibility and consent rates, randomization of consented patients, transfusion compliance, adherence to non-platelet protocols, and adjudication of severe bleeding or clotting events.
Three operational targets were missed: approaching eligible families within the early enrollment window, limiting temporary suspensions unrelated to decannulation, and limiting withdrawal or loss to follow-up. These findings identify practical improvements needed for a larger trial.
How to Interpret the Safety Findings
Eleven children experienced the primary safety composite. Severe bleeding occurred in seven and severe clotting in two. A committee blinded to treatment assignment adjudicated the reported bleeding and clotting events.
The small event counts cannot rule out clinically important differences between thresholds. This study was sized to demonstrate separation in platelet counts, not to establish definitive clinical safety, noninferiority, or survival benefit. A nonsignificant comparison in 25 patients per group is not proof of equivalence.
Relevance to Perfusion Practice
ECSTATIC shows that an early, protocolized comparison of prophylactic platelet thresholds is achievable in pediatric ECMO. For perfusion and intensive-care teams, the operational lessons include consistent bleeding assessment, timely platelet monitoring, documentation of transfusion triggers, and explicit handling of procedures or circuit changes that require temporary protocol suspension.
The findings concern children with no or minimal bleeding. They should not be generalized directly to active major bleeding, urgent procedures, or other populations outside the trial’s eligibility and management framework. The study supports further investigation rather than an automatic change to institutional platelet targets.
Limitations and Evidence Assessment
Randomization, multicenter participation, age stratification, and blinded event adjudication strengthen the feasibility evidence. However, the trial was open-label, included only 50 patients, and predominantly enrolled infants and children receiving VA-ECMO. Only two countries were represented, and local anticoagulation practices varied.
Enrollment, protocol suspensions, and follow-up also require refinement. The most defensible conclusion is that a definitive trial is feasible, with important implementation lessons; the optimal platelet transfusion threshold remains unresolved by this pilot.
Related IPA Coverage
For a separate question about platelet-product characteristics and post-transfusion count response, see our summary of the ECSTATIC secondary analysis. That report complements this feasibility trial and should not be treated as a separate randomized comparison of thresholds.
Source and Full Text
This original IPA educational summary is based on the supplied HHS Public Access author manuscript. The manuscript corresponds to the final article in Critical Care Medicine; formatting may differ from the publisher’s version. The cover is conceptual artwork, not a literal depiction of blood-cell scale or circuit assembly.
Original Article
Nellis ME, Barney BJ, Coles G, et al. Platelet Transfusion Thresholds for Children Supported by Extracorporeal Membrane Oxygenation: The ECSTATIC Feasibility Clinical Trial. Critical Care Medicine. 2026;54(3):410–421. doi:10.1097/CCM.0000000000006995. Trial registration: NCT05796557.
