Intracerebral Hemorrhage: Bundle Care, Reversal, and Surgical Evacuation
What current trial evidence supports for acute physiological control, anticoagulation reversal, and hematoma removal.
One-Minute Takeaway
- Acute intracerebral hemorrhage management rests on three complementary, evidence-based approaches — a physiological care bundle, anticoagulation reversal, and selective surgical evacuation — with the strongest [Verified] support from the INTERACT3 bundle trial.
- INTERACT3, a stepped-wedge cluster trial of 7,036 patients, found a goal-directed bundle lowered the odds of a poor functional outcome (common OR 0.86, 95% CI 0.76–0.97).
- For factor Xa inhibitor–related bleeding, ANNEXA-I improved hemostatic control (67.0% vs 53.1%) but raised thrombotic events and did not improve function.
- Minimally invasive evacuation (ENRICH) improved 180-day disability scores, yet the benefit was confined to lobar hemorrhages; open-craniotomy trials stayed neutral.
- Evidence is strongest for the bundle and warfarin reversal, moderate and mixed for factor Xa reversal, and preliminary for surgery in deep hemorrhage.
The Clinical Problem in Intracerebral Hemorrhage
Spontaneous intracerebral hemorrhage is the deadliest stroke subtype. It represents a minority of strokes yet accounts for a disproportionate share of stroke mortality. Early death is common, with roughly one in three patients dying within the first month.
Hematoma expansion in the opening hours drives much of that harm. Unlike the initial bleed, expansion is measurable and potentially modifiable. This makes the hyperacute window the central target for intracerebral hemorrhage treatment.
For years, trials of single interventions aimed at limiting expansion failed to change functional outcomes. Blood pressure lowering, hemostatic agents, and surgery each underwhelmed in isolation. The field has since shifted toward combined, time-critical, protocolized care.
Available Approaches to Intracerebral Hemorrhage
Current management converges on three approaches that layer onto one pathway. The first controls acute physiology through a bundle. The second reverses anticoagulation when present. The third removes the clot surgically in selected patients.
| Trial (Year) | Design | N | Population | Intervention vs Comparator | Key Verified Result |
|---|---|---|---|---|---|
| INTERACT3 (2023) | Stepped-wedge cluster RCT | 7,036 | Acute spontaneous ICH (mainly low/middle-income settings) | Goal-directed care bundle vs usual care | Lower odds of poor functional outcome: common OR 0.86 (95% CI 0.76–0.97); fewer serious adverse events (16.0% vs 20.1%) V |
| ANNEXA-I (2024) | RCT (stopped early) | 530 | FXa inhibitor–associated ICH; presenting ≤6 h from onset, last anticoagulant dose ≤15 h | Andexanet alfa vs usual care (mostly 4F-PCC) | Hemostatic efficacy 67.0% vs 53.1% (adjusted difference 13.4%, 95% CI 4.6–22.2); thrombotic events 10.3% vs 5.6% V |
| INCH (2016) | RCT, PROBE (stopped early) | 50 | Vitamin K antagonist–associated ICH, INR ≥2 | 4-factor PCC vs fresh frozen plasma | INR ≤1.2 within 3 h: 67% vs 9% (adjusted OR 30.6, 95% CI 4.7–197.9) V |
| ENRICH (2024) | Adaptive RCT | 300 | Lobar or anterior basal ganglia ICH, 30–80 mL, within 24 h | Minimally invasive evacuation + medical vs medical alone | Utility-weighted mRS at 180 d 0.458 vs 0.374 (difference 0.084, 95% BCI 0.005–0.163); benefit lobar-driven V |
| MISTIE III (2019) | RCT, blinded endpoint | 506 | Supratentorial ICH ≥30 mL | Catheter evacuation + alteplase vs medical care | mRS 0–3 at 365 d 45% vs 41% (not statistically significant) V |
| SWITCH (2024) | RCT, assessor-blinded (stopped early) | 201 | Severe deep supratentorial ICH | Decompressive craniectomy + medical vs medical alone | mRS 5–6 at 180 d, per-protocol 47% vs 60% (adjusted RR 0.76, 95% CI 0.58–1.00) V |
Physiological Control — The INTERACT3 Care Bundle
INTERACT3 tested a single goal-directed bundle rather than one drug. Its protocols targeted intensive systolic pressure lowering (below 140 mm Hg), glucose control, fever control (temperature at or below 37.5°C), and rapid warfarin reversal (INR below 1.5).
In this stepped-wedge cluster trial of 7,036 patients across 121 hospitals, the bundle reduced the odds of a poor functional outcome (common OR 0.86, 95% CI 0.76–0.97). Patients in the bundle group also had fewer serious adverse events.
The important interpretive point is that the bundle — not any single component — is the intervention. Intensive blood pressure lowering on its own had not improved outcomes in earlier trials, which reframes the bundle as the unit that carries the signal.
Anticoagulation Reversal
Anticoagulant-associated bleeds expand more and carry a worse prognosis, and an estimated 12–20% of patients arrive on oral anticoagulants. Guidelines advise stopping the anticoagulant and reversing it immediately. The reversal agent should match the drug.
For warfarin, the INCH trial compared 4-factor prothrombin complex concentrate with fresh frozen plasma. PCC normalized the INR far faster, with 67% versus 9% of patients reaching an INR of 1.2 or lower within three hours (adjusted OR 30.6). The plasma arm had more deaths from hematoma expansion, and the trial was stopped early for safety.
For factor Xa inhibitors, ANNEXA-I compared andexanet alfa with usual care. Andexanet improved hemostatic efficacy (67.0% vs 53.1%) and rapidly reduced anti–factor Xa activity. However, thrombotic events were more frequent (10.3% vs 5.6%), driven by ischemic stroke, and death or disability at 30 days did not differ.
The hemostatic gain with andexanet therefore has not translated into a proven functional benefit, and it carries a thrombotic trade-off. Four-factor PCC remains a widely used off-label alternative for factor Xa reversal, though the two agents have not been compared head-to-head in a randomized trial.
Surgical Evacuation
Open craniotomy has the longest track record and the weakest routine case. The STICH and STICH II trials found no overall functional benefit from early open surgery, with at most a small possible survival edge in superficial lobar clots. Open evacuation is not standard for most patients.
Minimally invasive techniques were tested next. MISTIE III combined catheter evacuation with alteplase and was neutral for good functional outcome at one year (mRS 0–3 in 45% vs 41%). Greater clot reduction, however, correlated with better outcomes.
ENRICH then evaluated early minimally invasive parafascicular evacuation. It improved 180-day disability scores (utility-weighted mRS 0.458 vs 0.374), but the effect was concentrated in lobar hemorrhages. Deep and basal ganglia clots did not clearly benefit.
Decompressive craniectomy was assessed in SWITCH for severe deep hemorrhage. The trial provided only weak, non-significant evidence of benefit (mRS 5–6 in 47% vs 60% per-protocol; adjusted RR 0.76, 95% CI 0.58–1.00). It was underpowered and stopped early, leaving craniectomy a rescue option in selected patients.
Comparative Analysis
Across the three approaches, both effect sizes and certainty differ markedly. The bundle offers the broadest and most reproducible signal at minimal cost. Reversal is disease-modifying for the anticoagulated subgroup but agent-specific. Surgery helps a narrow, anatomically defined group.
A recurring theme unifies them: timing and patient selection. The bundle works by acting fast on several physiological targets at once. ENRICH’s benefit appeared only with early evacuation of lobar clots. Reversal matters most before expansion has occurred.
These approaches are complementary rather than competing. A single patient may receive bundled physiological control, targeted reversal, and — if the clot is lobar and appropriately sized — evacuation. The trials describe layers of one pathway, not a menu of alternatives.
Quality and Consistency of the Evidence
Strength of evidence varies by approach and by the outcome being claimed. The ratings below reflect design, size, consistency, and whether benefits are broad or subgroup-limited.
- Moderate–high — Care bundle (INTERACT3). Large randomized trial with a favorable shift consistent across sensitivity analyses; the main caveat is generalizability, as recruitment was concentrated in specific settings.
- Moderate — Warfarin reversal with PCC (INCH). Randomized and consistent with observational data, but small and stopped early for safety, which widens the uncertainty around the effect size.
- Moderate, mixed — Factor Xa reversal (ANNEXA-I). Randomized hemostatic benefit is offset by increased thrombotic events and no measured functional gain, so the net clinical value is uncertain.
- Moderate — Minimally invasive evacuation for lobar ICH (ENRICH). Randomized with a positive primary endpoint, but the benefit is subgroup-driven and tied to a specific device and technique.
- Low, preliminary — Decompressive craniectomy (SWITCH). Underpowered with a non-significant primary outcome; a per-protocol trend is hypothesis-generating rather than confirmatory.
What the Evidence Does Not Show
- No randomized trial shows that andexanet alfa improves functional outcome or survival; its hemostatic benefit has not translated into better 30-day disability, and it increases thrombotic events.
- No head-to-head randomized comparison exists between andexanet alfa and 4-factor PCC for factor Xa inhibitor reversal.
- Surgical evacuation has not been shown to help deep or basal ganglia hemorrhages; the ENRICH benefit was specific to lobar clots.
- The individual contribution of each bundle component is not quantified — INTERACT3 cannot separate the effect of pressure control from glucose, fever, or coagulation management.
- The optimal systolic pressure target remains unsettled; lowering below 130 mm Hg appears potentially harmful, and earlier intensive-lowering trials did not establish a functional benefit.
Practical Implications
The evidence supports treating intracerebral hemorrhage as a time-critical emergency and starting a goal-directed bundle early. Reasonable physiological targets include a systolic pressure near 140 mm Hg while avoiding drops below 130 mm Hg, alongside glucose, temperature, and coagulation management.
Every patient should be screened for anticoagulation and reversed immediately when it is present. Agent choice follows the drug: PCC for warfarin, andexanet alfa or PCC for factor Xa inhibitors, and idarucizumab for dabigatran.
Minimally invasive evacuation is a reasonable consideration for appropriately sized lobar hemorrhages removed early. The lobar-specific signal should not be extrapolated to deep clots. Decompressive craniectomy may be discussed for selected severe deep hemorrhages, with candid acknowledgment that the supporting evidence is weak.
Evidence Grade and Bottom Line
Overall evidence grade
- The strongest and most generalizable advance is the INTERACT3 bundle: simple, low-cost, and beneficial across care settings.
- Reversal is essential and agent-specific; faster normalization limits expansion, but the factor Xa antidote carries a thrombotic trade-off without a proven functional gain.
- Surgery benefits a defined minority — early lobar evacuation — and remains investigational for deep hemorrhage.
Taken together, the current evidence favors fast, protocolized physiological control for everyone, reversal tailored to the anticoagulant, and surgery reserved for anatomically suitable clots. The largest gains come from doing simple things quickly and consistently.
Article Information and References
- Ma L, Hu X, Song L, et al. The third Intensive Care Bundle with Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial (INTERACT3): an international, stepped wedge cluster randomised controlled trial. Lancet. 2023;402(10395):27–40. DOI: https://doi.org/10.1016/S0140-6736(23)00806-1
- Connolly SJ, Sharma M, Cohen AT, et al. Andexanet for factor Xa inhibitor–associated acute intracerebral hemorrhage. N Engl J Med. 2024;390(19):1745–1755. DOI: https://doi.org/10.1056/NEJMoa2313040
- Steiner T, Poli S, Griebe M, et al. Fresh frozen plasma versus prothrombin complex concentrate in patients with intracranial haemorrhage related to vitamin K antagonists (INCH): a randomised trial. Lancet Neurol. 2016;15(6):566–573. DOI: https://doi.org/10.1016/S1474-4422(16)00110-1
- Pradilla G, Ratcliff JJ, Hall AJ, et al. Trial of early minimally invasive removal of intracerebral hemorrhage (ENRICH). N Engl J Med. 2024;390(14):1277–1289. DOI: https://doi.org/10.1056/NEJMoa2308440
- Hanley DF, Thompson RE, Rosenblum M, et al. Efficacy and safety of minimally invasive surgery with thrombolysis in intracerebral haemorrhage evacuation (MISTIE III): a randomised, controlled, open-label, blinded endpoint phase 3 trial. Lancet. 2019;393(10175):1021–1032. DOI: https://doi.org/10.1016/S0140-6736(19)30195-3
- Beck J, Fung C, Strbian D, et al. Decompressive craniectomy plus best medical treatment versus best medical treatment alone for spontaneous severe deep supratentorial intracerebral haemorrhage (SWITCH): a randomised controlled clinical trial. Lancet. 2024;403(10442):2395–2404. DOI: https://doi.org/10.1016/S0140-6736(24)00702-5
- Anderson CS, Heeley E, Huang Y, et al. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage (INTERACT2). N Engl J Med. 2013;368(25):2355–2365. DOI: https://doi.org/10.1056/NEJMoa1214609
- Qureshi AI, Palesch YY, Barsan WG, et al. Intensive blood-pressure lowering in patients with acute cerebral hemorrhage (ATACH-2). N Engl J Med. 2016;375(11):1033–1043. DOI: https://doi.org/10.1056/NEJMoa1603460
- Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas (STICH): a randomised trial. Lancet. 2005;365(9457):387–397. DOI: https://doi.org/10.1016/S0140-6736(05)17826-X
- Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial lobar intracerebral haematomas (STICH II): a randomised trial. Lancet. 2013;382(9890):397–408. DOI: https://doi.org/10.1016/S0140-6736(13)60986-1
- Greenberg SM, Ziai WC, Cordonnier C, et al. 2022 Guideline for the management of patients with spontaneous intracerebral hemorrhage: a guideline from the American Heart Association/American Stroke Association. Stroke. 2022;53(7):e282–e361. DOI: https://doi.org/10.1161/STR.0000000000000407