2022 AHA/ASA Guideline for the Management of Spontaneous Intracerebral Hemorrhage

Spontaneous intracerebral hemorrhage (ICH) accounts for approximately 10–15% of all strokes but is responsible for a disproportionate share of stroke-related mortality and disability. With a 30-day case fatality rate approaching 40% and only 20% of survivors achieving functional independence at 6 months, ICH remains the deadliest and least treatable form of stroke. The 2022 AHA/ASA guideline (Greenberg et al., Stroke 2022;53:e282–e361) represents the first comprehensive update in nearly a decade, incorporating evidence from landmark trials including INTERACT3, ATACH-2, MISTIE III, CLEAR III, and TICH-2.

The pathophysiology of ICH involves primary mechanical injury from the hematoma itself and secondary injury from edema, inflammation, and blood product toxicity. Hematoma expansion (HE) — occurring in up to one-third of patients — remains the principal modifiable target in acute management. This guideline covers the full spectrum of ICH care: prehospital systems, acute medical management, surgical interventions, prognostication, rehabilitation, and secondary prevention.

1. Prehospital & Systems of Care

Prehospital recommendations for ICH mirror those for all stroke types, as clinical differentiation from ischemic stroke is unreliable without neuroimaging. Key principles include early recognition using validated stroke scales (FAST, LAPSS, CPSS), immediate EMS activation, and prenotification of receiving hospitals (all Class 1).

Regional systems of stroke care are recommended so all potentially beneficial therapies — including neurosurgical and neurocritical care capabilities — can be made available as rapidly as possible (Class 1, LOE C-LD). Mobile stroke units (MSUs) enable earlier diagnosis (Class 2a, LOE B-R), though their specific impact on ICH outcomes remains under investigation.

2. Diagnosis & Assessment

2.1 Neuroimaging

Rapid neuroimaging with CT or MRI is recommended to confirm ICH diagnosis (Class 1, LOE B-NR). Serial CT within the first 24 hours is useful to evaluate for HE, particularly in patients with low GCS or neurological deterioration (Class 2a). CT angiography (CTA) identifies the "spot sign" — active contrast extravasation associated with high HE risk — and is increasingly integrated into acute imaging protocols.

2.2 Etiologic Workup

An underlying macrovascular cause (AVM, aneurysm, dural AVF, cavernoma, CVT) is present in 14–25% of selected patients <70 years old without typical hypertension-related deep ICH. Vascular imaging is particularly important in the following populations:

• Age <70 with lobar ICH
• Age <45 with deep or posterior fossa ICH
• Age 45–70 with deep/posterior fossa ICH and no hypertension history or imaging signs of small vessel disease
• Isolated IVH (23% macrovascular cause on DSA)
• CT or MR evidence of macrovascular lesion

DSA remains the gold standard but CTA/MRA have >90% sensitivity and specificity for vascular malformations. MRI with blood-sensitive sequences (SWI, T2*) helps identify CAA markers (lobar microbleeds, cortical superficial siderosis) and exclude neoplasm.

3. Acute Blood Pressure Management

Blood pressure lowering in acute ICH aims to limit hematoma expansion, the most consistent predictor of poor outcome. Two landmark trials — INTERACT2 and ATACH-2 — defined the therapeutic landscape, with INTERACT3 validating a comprehensive care bundle approach.

INTERACT2 vs. ATACH-2: These trials used different approaches. INTERACT2 (n=2,839) randomized patients within 6 hours of onset (SBP 150–220 mmHg) to intensive (<140 mmHg) vs. guideline-recommended (<180 mmHg) targets. The primary composite endpoint was neutral, but ordinal mRS analysis suggested a functional benefit with intensive treatment (OR 0.87; p=0.04). ATACH-2 (n=1,000) tested more aggressive lowering (SBP 110–139 mmHg vs. 140–179 mmHg) within 4.5 hours, exclusively using IV nicardipine. The trial was stopped for futility — no benefit on mRS 4–6 at 3 months — with significantly more renal adverse events in the intensive arm.

INTERACT3 (2023) took a different approach by embedding BP lowering within a goal-directed care bundle targeting SBP <140 mmHg, glucose control, antipyrexia, and INR normalization. This stepped-wedge cluster RCT across 121 hospitals and 10 countries demonstrated a favorable mRS shift (OR 0.86; p=0.015) and reduced mortality (13.6% vs. 16.6%; OR 0.77; p=0.015; NNT = 35). This was the first positive phase 3 RCT in acute ICH treatment.

4. Hemostasis & Coagulopathy Management

4.1 Anticoagulant Reversal

Anticoagulant-associated ICH carries higher mortality due to larger initial hematoma volumes and increased risk of HE. Immediate reversal is mandated regardless of anticoagulant class (Class 1).

4.2 Antiplatelet-Related ICH

Management of ICH in patients on antiplatelet therapy remains challenging. Key guideline positions:

• Platelet transfusion for non-surgical patients: Harmful. The PATCH trial showed that platelet transfusion in ICH patients on antiplatelet therapy (who were not planned for surgery) shifted mRS toward worse outcomes and increased SAEs (Class 3: Harm).
• Platelet transfusion pre-surgery: May be reasonable before craniotomy/hematoma evacuation in patients on cyclooxygenase inhibitors (Class 2b). Data are limited to aspirin; not applicable to ticagrelor.
• Desmopressin (DDAVP): Uncertain benefit. No RCT data; retrospective studies show inconsistent results (Class 2b).

4.3 General Hemostatic Therapy

Despite HE occurring in up to one-third of ICH patients and being a consistent predictor of poor outcome, pharmacological hemostatic therapies have not shown functional benefit:

• Tranexamic acid (TXA): TICH-2 (n=2,325; within 8 hours) demonstrated modest HE reduction and fewer early deaths, but no significant improvement in 90-day mRS. TXA was safe, with no increase in VTE. STOP-AUST and ULTRA were also neutral on their primary endpoints, though STOP-AUST suggested a trend toward benefit with ultra-early treatment (≤2 hours). All classified as Class 2b, LOE B-R.
• Recombinant Factor VIIa: Phase III trial showed modest HE reduction but no functional benefit and increased arterial thrombotic events. Meta-analysis confirmed no benefit. Class 2b, LOE B-R.

5. General Inpatient & Neurointensive Care

5.1 Care Setting & Monitoring

Specialized inpatient care in a stroke unit with a multidisciplinary team is recommended (Class 1, LOE A). For moderate-to-severe ICH, IVH, hydrocephalus, or infratentorial location, a neuro-specific ICU is reasonable over a general ICU (Class 2a, LOE B-NR). The highest-risk period for neurological decline is within the first 12 hours, with deterioration becoming uncommon after 48 hours.

INTERACT3 demonstrated that standardized care bundles — targeting BP, glucose, temperature, and INR simultaneously — improve outcomes even in diverse healthcare settings. This bundle approach should be adopted as the standard of care for acute ICH.

5.2 Glucose, Temperature, and VTE Prophylaxis

• Glucose: Both hyperglycemia and hypoglycemia are associated with worse outcomes. Target glucose 7.8–10.0 mmol/L (140–180 mg/dL) in diabetics; 6.1–7.8 mmol/L (110–140 mg/dL) in non-diabetics. Avoid insulin-induced hypoglycemia (Class 2a).
• Temperature: Fever is associated with poor outcomes. Antipyrexia (target ≤37.5°C) is recommended as part of a standardized care bundle.
• VTE Prophylaxis: Intermittent pneumatic compression (IPC) should be applied from admission in non-ambulatory patients (Class 1). Pharmacologic prophylaxis (LMWH) may be considered after 24–48 hours once hemorrhage stability is confirmed (Class 2b).

6. Seizures & Antiseizure Drugs

Clinical seizures occur in ~8–16% of ICH patients, with electrographic-only seizures detected in an additional 18–28% on continuous EEG. Despite this high prevalence, prophylactic antiseizure medication has consistently failed to improve outcomes in multiple observational studies. The risk of seizures is highest with cortical involvement, lobar location, and larger hematoma volumes.

7. ICP & Edema Management

• EVD for hydrocephalus: Recommended for clinical hydrocephalus with impaired consciousness (Class 1, LOE B-NR).
• ICP monitoring: May be considered in patients with GCS 3–8, based primarily on TBI literature (Class 2b). Maintain ICP <22 mmHg and CPP 50–70 mmHg.
• Hyperosmolar therapy: Mannitol or hypertonic saline for acute ICP elevation (Class 2b). No strong evidence favoring one agent over the other.
• Corticosteroids: NOT beneficial for ICP management in ICH and associated with complications including hyperglycemia and infections (Class 3: No Benefit).

8. Surgical Interventions

8.1 Minimally Invasive Surgery (MIS) for Supratentorial ICH

Minimally invasive approaches represent the most actively evolving area of ICH surgery. Three major trials define the current evidence base:

MISTIE III (n=499) randomized patients with supratentorial ICH ≥30 mL to stereotactic catheter aspiration with alteplase instillation vs. medical care. The primary outcome (mRS 0–3 at 365 days) was neutral (45% vs. 41%; p=0.33), but mortality was significantly reduced (15% vs. 23%; HR 0.67; p=0.037). A key finding was that achieving a residual clot <15 mL was associated with a 10.5% improvement in favorable outcomes.

ENRICH (2024; n=300) evaluated minimally invasive parafascicular surgery (MIPS) within 24 hours for lobar or anterior basal ganglia ICH (30–80 mL). MIPS improved utility-weighted mRS at 180 days (0.47 vs. 0.37; p=0.04), with the effect driven primarily by lobar ICH (Uw-mRS difference 0.127; p=0.007). MIPS dramatically reduced the need for decompressive hemicraniectomy (3.3% vs. 20%) and shortened both ICU stay (6.9 vs. 9.7 days) and hospital stay (14.9 vs. 18.1 days).

MIND (2025; n=236) tested MIS within 72 hours for supratentorial ICH (20–80 mL). Stopped early due to publication of the positive ENRICH trial, MIND showed no difference in 180-day death and disability (OR 1.03; p=0.45), though 30-day mortality trended lower with surgery (7.2% vs. 9.8%). Exploratory analyses suggested early benefit at 30 days.

8.2 IVH Management: EVD ± Thrombolytic

Intraventricular hemorrhage (IVH) complicates ~45% of ICH cases and independently predicts worse outcomes through obstructive hydrocephalus and direct neurotoxicity of intraventricular blood products.

CLEAR III (n=500) randomized patients with IVH requiring EVD to intraventricular alteplase (1 mg q8h, up to 12 doses) vs. saline. The primary outcome (mRS 0–3 at 180 days) was not significantly different (48% vs. 45%; p=0.45), but mortality was reduced in the alteplase group (18% vs. 29%; HR 0.50; p<0.001). Achieving an 80% clot reduction was associated with better functional outcomes.

8.3 Craniotomy for Supratentorial ICH

Conventional craniotomy has been evaluated in multiple RCTs, most notably the STICH trials. STICH II (n=601) randomized patients with lobar ICH (10–100 mL, within 48 hours, GCS motor 5–6) to early craniotomy vs. initial conservative treatment. The primary outcome was neutral (favorable GOSE: 41% vs. 38%; OR 0.86; p=0.37), with a non-significant trend toward reduced mortality (18% vs. 24%). A subgroup with poor prognosis showed significant surgical benefit (OR 0.49; p=0.02).

The guideline classifies craniotomy as uncertain benefit for most patients (Class 2b, LOE A) but acknowledges it as a potentially lifesaving measure in deteriorating patients (Class 2b, LOE C-LD).

8.4 Posterior Fossa ICH

Cerebellar ICH represents a surgical emergency due to the confined posterior fossa space and risk of rapid brainstem compression. Immediate surgical removal ± EVD is recommended (Class 1, LOE B-NR) in patients who are:

• Deteriorating neurologically
• Showing brainstem compression
• Developing hydrocephalus from ventricular obstruction
• Cerebellar ICH volume ≥15 mL

8.5 Decompressive Craniectomy

SWITCH (2024; n=197) randomized patients with severe deep supratentorial ICH to decompressive craniectomy (≥12 cm, without hematoma evacuation) plus best medical treatment vs. medical treatment alone. Stopped early due to funding, the trial showed a non-significant reduction in mRS 5–6 at 180 days (44% vs. 58%; aRR 0.77; p=0.057). mRS shift analysis was significant (common OR 0.57; p=0.04), and mortality trended lower (17% vs. 27%). A post-hoc analysis showed the benefit was preserved regardless of ICH location within deep structures.

The guideline classifies decompressive craniectomy as potentially reducing mortality (Class 2b, LOE C-LD) but with uncertain functional benefit (Class 2b, LOE C-LD).

9. Outcome Prediction & Goals of Care

The ICH Score and other severity measures provide a useful framework for communicating clinical severity but have critical limitations when used to guide withdrawal of life-sustaining treatment. Evidence strongly suggests a self-fulfilling prophecy exists: early withdrawal of care in patients predicted to do poorly ensures the predicted outcome.

10. Recovery & Rehabilitation

10.1 Neurobehavioral Complications

Depression affects 30–50% of ICH survivors and is associated with increased disability, cognitive decline, and mortality. Screening for depression should be performed routinely, and treatment with SSRIs (with monitoring for bleeding risk) is reasonable. Cognitive screening and therapy for impaired attention, memory, and executive function should be integrated into rehabilitation programs. Fatigue and apathy are common and underrecognized post-ICH complications.

11. Secondary Prevention

11.1 Long-Term BP Management

Hypertension accounts for 73.6% of the global population-attributable risk for ICH. BP control after ICH is the single most important modifiable factor for preventing recurrence (Class 1, LOE B-R). The guideline recommends a target of SBP ≤130 mmHg and DBP ≤80 mmHg for long-term management (Class 2a, LOE B-NR). A significant proportion of ICH survivors continue to have poorly controlled BP, highlighting the need for aggressive outpatient management.

11.2 Prognostication of Recurrence Risk

ICH recurrence risk ranges from 1.2–3%/year across undifferentiated patients but varies substantially based on underlying pathogenesis. Risk factors to incorporate into decision-making (Class 2a, LOE B-NR):

• Lobar location of initial ICH (HR ~4.8 for CAA-related)
• Number and lobar distribution of MRI microbleeds (HR 1.88 for 1, 2.93 for 2–4, 4.12 for >4 microbleeds)
• Disseminated cortical superficial siderosis (HR 4.69)
• APOE ε2 or ε4 alleles (HR 2.5–3.3)
• Poorly controlled outpatient BP
• Older age (HR 2.8 for age >65)
• Asian or Black race

11.3 Antithrombotic Resumption

Resuming antithrombotic therapy after ICH requires careful balancing of recurrent hemorrhage risk against thromboembolic risk. The guideline provides a nuanced framework:

Antiplatelet resumption: RESTART (n=537; median randomization 76 days post-ICH) showed that resuming antiplatelet therapy did not increase recurrent ICH (4% vs. 9%, favoring restart) and may reduce major vascular events. The guideline considers antiplatelet resumption "may be reasonable" (Class 2b, LOE B-R).

Anticoagulation resumption for AF: This remains the most complex decision. PRESTIGE-AF (n=319) found that DOACs effectively prevented ischemic stroke in ICH survivors with AF (HR 0.05; p<0.0001) but did not meet non-inferiority for recurrent ICH (HR 10.89; ICH rate 5.00 vs. 0.82 per 100 patient-years). The guideline suggests resumption "may be considered" (Class 2b, LOE B-NR), with initiation approximately 7–8 weeks post-ICH (Class 2b, LOE C-LD).

Mechanical valves/LVAD: Early resumption of anticoagulation is reasonable given high thrombotic risk (Class 2a, LOE C-LD).

Left atrial appendage closure: May be considered in AF patients deemed ineligible for anticoagulation (Class 2b, LOE C-LD).

11.4 Lifestyle Modifications

Lifestyle modification is reasonable to reduce BP and recurrence risk (Class 2a, LOE C-LD). Avoiding heavy alcohol consumption is specifically recommended (Class 2a). Supervised training and counseling may improve functional recovery (Class 2b). Psychosocial education for patients and caregivers is reasonable to improve outcomes and quality of life.

12. Major ICH Trial Comparison