PATCH Trial: Platelet Transfusion in Antiplatelet-Associated Spontaneous ICH
In adults with spontaneous supratentorial ICH on antiplatelet therapy for at least 7 days, randomized within 6 hours of onset with GCS ≥8 (no planned surgery), does platelet transfusion within 90 minutes of brain imaging reduce death or dependence at 3 months compared with standard care? Ordinal-shift primary on the full mRS distribution.
Baharoglu MI, et al. for the PATCH Investigators (Lancet 2016;387:2605-2613) · doi:10.1016/S0140-6736(16)30392-0 · 190 patients
Population
Included
- Adults with spontaneous supratentorial intracerebral hemorrhage confirmed on CT
- On antiplatelet therapy (aspirin, clopidogrel, dipyridamole, or combinations) for at least 7 days before ICH
- Randomization within 6 hours of symptom onset
- Platelet transfusion possible within 90 minutes of diagnostic brain imaging
- GCS at least 8 at randomization
Excluded
- Infratentorial (cerebellar or brainstem) hemorrhage
- Underlying structural cause (aneurysm, AVM, tumor, hemorrhagic transformation of infarct)
- Planned neurosurgical evacuation within 24 hours of randomization
- Coagulopathy unrelated to antiplatelet therapy (INR >1.5, platelet count <100 x 10^9/L)
- Death imminent within 24 hours
Primary Outcome — ORDINAL SHIFT
Shift toward death or dependence on the full mRS distribution at 3 months (ordinal logistic regression). Direction: HARM. NNT is not appropriate for ordinal-shift outcomes.
mRS 3–6 at 3 months (proxy for ordinal shift; primary was full mRS distribution)
HARM trial. The adjusted common odds ratio of 2.05 (95% CI 1.18–3.56, P=0.0114) means platelet transfusion approximately DOUBLED the odds of a worse mRS at 3 months. Mortality 24% vs 17%; serious adverse events 42% vs 30% — same direction as the primary. Hematoma expansion did not differ; mechanism is not simple rebleeding. Trial excluded patients with planned neurosurgical evacuation — does NOT apply to bridging platelet transfusion before craniotomy.
Safety
3-month all-cause mortality
24%
17%
Death by 3 months: 23 of 97 (24%) platelet transfusion vs 16 of 93 (17%) standard care. Numerically higher with platelet transfusion; consistent direction with the primary mRS shift toward worse outcome.
Serious adverse events during hospitalization
42%
30%
Serious adverse events: 40 of 97 (42%) platelet transfusion vs 28 of 93 (30%) standard care. The platelet-transfusion arm carried higher serious-event burden in addition to worse mRS at 3 months.
Trial Design
Type
- Open-label, multicenter, masked-endpoint phase 3 RCT (PROBE design)
- 1:1 randomization stratified by hospital and antiplatelet type
- Outcome adjudication blinded
- Ordinal logistic regression on full mRS distribution
Timeline
Enrolled February 2009 to October 2015; 3-month mRS follow-up
N
190
Enrollment
190 patients (97 platelet transfusion / 93 standard care) at 60 hospitals in the Netherlands, UK, and France. Enrolled February 2009 to October 2015. Open-label, multicenter, masked-endpoint phase 3 RCT (PROBE design). 1:1 randomization stratified by hospital and antiplatelet type. Ordinal logistic regression on full mRS distribution at 3 months as the primary analysis; outcome adjudication blinded. Published Lancet 2016.
Bedside Pearl
Do NOT transfuse platelets routinely for spontaneous antiplatelet-associated ICH. PATCH 2016 showed platelet transfusion DOUBLES the odds of death or dependence at 3 months (adjusted common OR 2.05, P=0.0114). AHA/ASA 2022 Class III: Harm. Exception: bridging to emergent craniotomy, off-trial case-by-case.