PATCH
PlAtelet Transfusion versus standard care after acute stroke due to spontaneous Cerebral Haemorrhage associated with antiplatelet therapy
Year of Publication: 2016
Authors: Baharoglu MI, Cordonnier C, Al-Shahi Salman R, ..., Roos YB; on behalf of The PATCH Investigators
Journal: The Lancet
Citation: Lancet 2016;387(10038):2605-2613
Link: https://doi.org/10.1016/S0140-6736(16)30392-0
PDF: https://www.pure.ed.ac.uk/ws/files/19361...lean_002_2_.pdf
PDF: Clinical Question
Does platelet transfusion reduce death or dependence compared to standard care in patients with acute spontaneous intracerebral hemorrhage who were taking antiplatelet therapy?
Study Overview
Objective
To test whether platelet transfusion reduces death or dependence compared to standard care in patients with acute spontaneous intracerebral hemorrhage associated with antiplatelet therapy use
Study Summary
- Platelet transfusion WORSENED outcomes compared to standard care (adjusted OR 2.05 for worse mRS, p=0.01)
- Higher rates of death or dependence (mRS 4-6) with platelet transfusion: 72% vs 56%
- No reduction in hematoma growth with platelet transfusion
Intervention
Platelet transfusion (1 concentrate for COX inhibitor users, 2 concentrates for ADP inhibitor users) within 6 hours of symptom onset and 90 minutes of imaging
Patients per Arm
97 vs 93
Bottom Line
Platelet transfusion is HARMFUL in acute ICH patients on antiplatelet therapy. It increased the odds of death or dependence by 2-fold compared to standard care and cannot be recommended for this indication.
Major Points
- First randomized trial of platelet transfusion for acute ICH in patients on antiplatelet therapy
- Platelet transfusion significantly worsened functional outcomes (adjusted common OR 2.05, 95% CI 1.18-3.56, p=0.0114)
- 72% of platelet transfusion group had mRS 4-6 vs 56% in standard care group
- No significant reduction in hematoma growth with platelet transfusion
- Findings consistent across subgroups (single vs dual antiplatelet, country, ICH volume)
- Possible mechanisms of harm: pro-thrombotic effects, inflammation, impaired collateral perfusion
- Trial stopped at pre-specified sample size, not for futility or harm
Design
Study Type: Randomized, open-label, masked endpoint (PROBE), phase 3 trial
Randomization: 1
Blinding: Open-label for participants and treating clinicians; outcome assessors and data analysts masked to treatment allocation
Enrollment Period: February 2009 to October 2015
Follow-up Duration: 3 months
Centers: 60
Countries: Netherlands, United Kingdom, France
Sample Size: 190
Analysis: Intention-to-treat for efficacy; as-treated for safety. Ordinal logistic regression of mRS shift adjusted for stratification variables (antiplatelet type) and ICH Score
Inclusion Criteria
- Age ≥18 years
- Non-traumatic supratentorial ICH confirmed by brain imaging
- Glasgow Coma Scale score 8-15
- Platelet transfusion could be initiated within 6 hours of symptom onset (or last seen well) and within 90 minutes of diagnostic brain imaging
- Use of antiplatelet therapy (COX inhibitor [aspirin or carbasalate calcium], ADP receptor inhibitor [clopidogrel], or adenosine reuptake inhibitor [dipyridamole]) for ≥7 days pre-ICH
- Pre-ICH modified Rankin Scale score 0-1
Exclusion Criteria
- Epidural or subdural hematoma, or underlying aneurysm or AVM on imaging
- Planned surgical evacuation of ICH within 24 hours
- Intraventricular hemorrhage more than sedimentation in posterior horns of lateral ventricles
- Previous adverse reaction to platelet transfusion
- Known use of vitamin K antagonist (unless INR ≤1.3) or history of coagulopathy
- Known thrombocytopenia <100×10⁹/L
- Lacking mental capacity before ICH
- Death appeared imminent
Baseline Characteristics
| Characteristic | Platelet Transfusion | Standard Care |
|---|---|---|
| N | 97 | 93 |
| Mean Age (years) | 74.2 (range 49-94) | 73.5 (range 40-92) |
| Female Sex | 43.3% | 38.7% |
| Prior Ischemic Stroke or TIA | 40.4% | 43.0% |
| Prior ICH | 4.1% | 5.4% |
| Hypertension | 72.3% | 72.8% |
| Diabetes Mellitus | 15.5% | 18.9% |
| Hypercholesterolemia | 48.9% | 47.6% |
| Ischemic Heart Disease | 23.6% | 24.4% |
| Peripheral Arterial Disease | 16.5% | 4.4% |
| Antiplatelet - COX inhibitor alone | 73.2% | 83.9% |
| Antiplatelet - COX + dipyridamole | 18.6% | 14.0% |
| Antiplatelet - ADP inhibitor alone | 4.1% | 1.1% |
| Antiplatelet - COX + ADP inhibitor | 3.1% | 1.1% |
| Antiplatelet - None | 1.0% (1/97) | 0% (0/93) |
| Statin therapy pre-ICH | 56.3% | 52.1% |
| Median GCS (IQR) | 14 (13-15) | 15 (13-15) |
| Median NIHSS (IQR) | 12 (7-19) | 13 (7-17) |
| Mean Platelet Count (×10⁹/L) | 229 (range 120-622) | 241 (range 91-461) |
| ICH Location - Deep | 64.6% | 76.1% |
| ICH Location - Lobar | 33.3% | 23.9% |
| ICH Location - Infratentorial | 2.1% | 0% |
| Median ICH Volume mL (IQR) | 13.1 (5.4-42.4) | 8.0 (4.4-25.8) |
| Intraventricular Extension | 12.6% | 21.7% |
| Median ICH Score (IQR) | 1 (0-2) | 1 (0-1) |
Arms
| Field | Platelet Transfusion | Control |
|---|---|---|
| Intervention | Standard care plus platelet transfusion initiated within 6 hours of symptom onset (or last seen well) and within 90 minutes of diagnostic brain imaging. Patients on COX inhibitor ± dipyridamole received 1 platelet concentrate; patients on ADP receptor inhibitor ± other antiplatelet received 2 platelet concentrates. | Standard care according to contemporary European and national guidelines, without platelet transfusion |
| Duration | Single transfusion | NA |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| Functional outcome at 3 months rated using modified Rankin Scale (mRS 0-6), analyzed by ordinal logistic regression of the shift of all mRS categories; effect measure is the adjusted common (proportional-odds) OR | Primary | mRS distribution (n=93): 0=3 (3.2%), 1=6 (6.5%), 2=8 (8.6%), 3=24 (25.8%), 4=26 (28.0%), 5=5 (5.4%), 6=21 (22.6%) | mRS distribution (n=97): 0=1 (1.0%), 1=5 (5.2%), 2=5 (5.2%), 3=16 (16.5%), 4=30 (30.9%), 5=9 (9.2%), 6=31 (32.0%) | 2.05 | 0.0114 |
| Survival at 3 months | Secondary | 72 (77.4%) | 66 (68.0%) | 0.62 | 0.15 |
| mRS 4-6 (death or dependence) at 3 months | Secondary | 52 (55.9%) | 70 (72.2%) | 2.04 | 0.0190 |
| mRS 3-6 (poor outcome) at 3 months | Secondary | 76 (81.7%) | 86 (88.7%) | 1.75 | 0.18 |
| Median ICH growth at 24 hours, mL (IQR) [n=80 platelet transfusion, n=73 standard care] | Secondary | 1.16 (IQR 0.03-4.42) | 2.01 (IQR 0.32-9.34) | 0.81 | |
| Any SAE (ITT) | Adverse | 27/93 (29.0%) | 41/97 (42.3%) | 1.79 | NS |
| Any Fatal SAE (ITT) | Adverse | 15/93 (16.1%) | 24/97 (24.7%) | 1.71 | NS |
| SAE due to ICH complications (ITT) | Adverse | 13/93 (14.0%) | 24/97 (24.7%) | 2.02 | NS |
| SAE due to thromboembolism (ITT) | Adverse | 1/93 (1.1%) | 4/97 (4.1%) | 3.96 | NS |
| Transfusion reaction (ITT) | Adverse | 0 (0%) | 1 (1.0%) - minor non-hemolytic | ||
| Any SAE (as-treated) | Adverse | 28/95 (29.5%) | 40/95 (42.1%) | 1.74 | NS |
| SAE due to ICH complications (as-treated) | Adverse | 13/95 (13.7%) | 24/95 (25.3%) | 2.13 | Significant (only significant safety difference between groups) |
| SAE due to thromboembolism (as-treated) | Adverse | 1/95 (1.1%) | 4/95 (4.2%) | 4.13 | NS |
Subgroup Analysis
No significant interaction with treatment effect in pre-specified subgroups: single vs dual antiplatelet therapy, country of randomization (Netherlands vs France vs UK), and trichotomized baseline ICH volume (≤7mL vs >7-30mL vs >30mL). Post hoc sensitivity analysis excluding 36 participants who met at least one exclusion criterion showed consistent results (adjusted OR 2.22, 95% CI 1.20-4.09, p=0.0108).
Criticisms
- Small sample size (n=190) relative to other acute stroke trials, resulting in some baseline imbalances
- Baseline imbalances: platelet transfusion group had larger median ICH volume (13.1 vs 8.0 mL) but fewer with IVH extension (12.6% vs 21.7%)
- Majority of participants used aspirin; relatively few on ADP inhibitors (clopidogrel) - limits generalizability to clopidogrel users
- No screening logs maintained - potential selection bias
- Antiplatelet adherence was self-reported; no platelet function testing performed
- 19% of participants (36/190) met at least one exclusion criterion (protocol deviations)
- Open-label design (though outcome assessment was masked)
- No clear mechanism identified to explain harmful effect of platelet transfusion
- Unknown generalizability to low-middle income countries
Funding
Netherlands Organization for Health Research and Development (ZonMW, 170881002), Sanquin Blood Supply, Chest Heart and Stroke Scotland (CHSS project grant Ref. Res10), MRC clinician scientist (Ref. G108/613) and senior clinical fellowships (Ref. G1002605), Le Programme Hospitalier de Recherche Clinique (PHRC)
Based on: PATCH (The Lancet, 2016)
Authors: Baharoglu MI, Cordonnier C, Al-Shahi Salman R, ..., Roos YB; on behalf of The PATCH Investigators
Citation: Lancet 2016;387(10038):2605-2613
Content summarized and formatted by NeuroTrials.ai.