The Story of Intracerebral Hemorrhage: From Recalcitrant to Treatable Disease

Abstract

This invited special report is based on an award presentation at the World Stroke Organization/European Stroke Organization Conference in November of 2020 outlining progress in the acute management of intracerebral hemorrhage (ICH) over the past 35 years. Intracerebral hemorrhage is the second most common and the deadliest type of stroke for which there is no scientifically proven medical or surgical treatment. Prospective studies from the 1990s onward have demonstrated that most growth of spontaneous ICH occurs within the first 2–3 hours and that growth of ICH and resulting volumes of ICH and IVH are modifiable factors that can improve outcome. Trials focusing on early treatment of elevated blood pressure (BP) have suggested a target systolic BP of 140 mm Hg but none of the trials were positive by their primary endpoint. Hemostatic agents to decrease bleeding in spontaneous ICH have included desmopressin, tranexamic acid, and recombinant Factor VIIa without clear benefit, and platelet infusions which were associated with harm. Hemostatic agents delivered within the first several hours have the greatest impact on growth of ICH and potentially on outcome. No large Phase III surgical ICH trial has been positive by primary endpoint but pooled analyses suggest that earlier ICH removal is more likely to be beneficial. Recent trials emphasize maximization of clot removal and minimizing brain injury from the surgical approach. The future of ICH therapy must focus on delivery of medical and surgical therapies as soon as possible if we are to improve outcomes.

Introduction

Intracerebral hemorrhage (ICH) accounts for more than 10% of the estimated 17 million strokes worldwide each year. ^1,2 ICH has a mortality of more than 40% and only 20% of survivors are functionally independent at 6 months. ^3–5 But while we have made tremendous strides in the treatment of acute ischemic stroke and aneurysmal subarachnoid hemorrhage over the past 35 years, ICH remains without a proven scientific medical or surgical treatment. The neurosurgical and interventional community has long embraced and “owned” the treatment of patients with ruptured aneurysms while the medical neurology and the neurointerventional communities have championed the acute treatment of ischemic stroke. Thus, when the first American Heart Association guidelines for management of acute stroke were first commissioned, discussed, and published in 1994, ischemic stroke and aneurysmal subarachnoid hemorrhage each had their own published guidelines with a great sense of hope for new therapeutic advances. ^6,7 But ICH, the second most common and the deadliest type of stroke, wasn’t even initially discussed as a topic for guidelines. This initial lack of engagement by clinicians and researchers in the early 1990s may have stemmed from the belief that ICH occurred over minutes which provided no realistic possibility of intervention to halt the bleeding. ⁸ In addition, patients with an ICH subsequently had very poor outcomes, with or without surgical removal that usually involved craniotomy. It wasn’t until 1999 that AHA published the first acute treatment guidelines for this neglected or “orphan” stroke subtype. ⁹

This invited special report is based on an award presentation at the 2020 European Stroke Organization and World Stroke Organization Meeting that told the story of acute treatment of ICH over the past 35 years. This paper focuses primarily on the acute treatment of ICH within the first 1–2 days and particularly within the early hours after onset. And as for ischemic stroke, understanding the pathophysiology of ICH and its timing is critical for successful therapies which still elude us. It will take a global village of engaged investigators and clinicians to finish the story, but the end of the first chapter is in sight.

Insights into the Early Pathophysiology of ICH

ICH occurs when old, damaged, or abnormal vascular structures in the brain break under pressure. ^10,11 Prior to the 1980s, completion of this bleeding was thought to occur over minutes but experiences with patients can change our understanding of a disease. The seminal patient that led to an entire new direction in research for ICH was a 58-year old woman who collapsed walking past the fire station in Cincinnati in the late 1980s and was brought to a hospital several blocks away within a short time after onset. ⁸ At that time, our stroke team had just started evaluating patients as part of the first pilot study of rt-PA for acute ischemic stroke, in which patients had to be treated within 90 minutes from onset. Because of her very early arrival, our research team saw her within 40 minutes of symptom onset and found that she was alert with a right hemiparesis and sensory loss. While she initially appeared like an excellent candidate for rt-PA, her computed tomography (CT) scan showed that she had an eight mL left thalamic hemorrhage (Figure 1). We were disappointed that she wasn’t a candidate for rt-PA, but shortly thereafter her nurse told us that she was getting worse and becoming poorly responsive. We repeated the CT of the head which showed major growth in her thalamic hemorrhage (Figure 1). As the pilot rt-PA trial continued, we noted a number of these ICH patients and repeated their initial CT imaging. A substantial proportion of these patients, imaged very soon after onset, had substantial growth of ICH with accompanying deterioration. ⁸

Figure 1:

Serial CT scans in 58-year old woman. Increase in volume of ICH of 8 cc at 50 minutes from onset of symptoms (upper three images) to 35 cc at 210 minutes from onset (lower three images). Reprinted with permission from Journal of Neurosurgery, 1990; 72: 195–199.

These very early clinical observations with repeated imaging led to a subsequent prospective NINDS-funded study of patients with ICH that asked the simple question: how often does substantial growth occur in patients with spontaneous ICH who had their CT of the head done within 3 hours from symptom onset. ¹² This study demonstrated that 1) 38% of these patients had ≥ 1/3 growth in the volume of their hemorrhage from baseline on repeated imaging, 2) the large majority of this growth occurred within the first several hours from onset, and 3) this growth was accompanied by neurologic deterioration.

Since publication of this study, there have been a number of prospective studies and clinical trials of ICH that have further expanded our understanding of the frequency of early substantial ICH growth. Al Shahi Salman and colleagues published a pooled analysis of these data (Figure 2) that demonstrated the strong exponential decay relationship between time from onset and predicted probability of growth of ICH of > 6cc. ¹³ Thus, if we are going to change the natural history of bleeding in ICH, we need to do this within the first hours after onset, and our systems of care have to incorporate these time demands as they have for ischemic stroke.

Figure 2:

Figure A demonstrates the strong exponential relationship between time from onset and predicted probability of growth of ICH of > 6cc. Figure B demonstrates relationship between baseline ICH volume and growth. Reprinted with permission from Elsevier (Lancet, Neurology 2018, 17: 885–894.)

The Natural History of ICH and Its Therapeutic Implications

The baseline factors associated with ICH mortality and functional outcome are volume of ICH, volume of intraventricular hemorrhage (IVH), growth of ICH during first hours of onset, age, Glasgow Coma Scale (GCS), and infratentorial location. ^5,14,15 Of these, only growth of ICH and the resulting volumes of ICH and IVH are biologically modifiable. Volume of ICH is the strongest determinant of outcome with over 90% mortality in patients with ≥ 60 cc hemorrhage and only 1 of 71 patients with ICH volume ≥30 cm³ functioned independently at 30 days in one study from the early 1990s. ⁵ Half of the 44% 30-day mortality in this study occurred within the first 2 days.

Given our improved understanding of the timing of ICH growth and the modifiable determinants of outcome, the potential therapeutic approaches include: 1) decreasing blood pressure pushing blood through the leak(s); 2) plugging the hole at the site of the leak with a clot, or ideally both; 3) removing blood that causes mechanical compression as well as inflammation and toxic effects on surrounding brain tissue; 4) limiting toxic effects of blood on surrounding brain tissue by various neuroprotective strategies.

Treating Elevated Blood Pressure to Limit Bleeding and Improve Outcome

Sustained elevation of blood pressure and the subsequent vasculopathy is the most important modifiable risk factor for ICH. ¹⁴ Lowering blood pressure in the first hours after onset of ICH is a logical approach to slow bleeding through the damaged vessel. A number of well-designed Phase II and Phase III trials have tested this approach.

The Phase II INTERACT 1 and Phase 3 INTERACT 2 trials were global trials that targeted a decrease in BP to 140 mm Hg as compared to standard care (< 180 mm Hg) using available intravenous medications in participating countries in patients randomized within 6 hours of onset. ^16–18 INTERACT 2 showed a non-significant trend toward less hemorrhage growth in the 140 mm Hg group with earlier treatment (3.1 cc more growth in standard group as compared to 140 mg Hg group when treated within 3 hours versus 1.2 cc in those treated at 3–6 hours). INTERACT 2 was not positive with regards to its primary endpoint, a modified Rankin (mRS) of 3–6, but was positive in its secondary endpoint of the distribution of the ordinal mRS. The results of this trial are the basis for the current Guidelines for Management of Acute ICH that target a systolic pressure of 140 mm Hg. ^14,19

The Phase II INTERACT 1 and Phase 3 INTERACT 2 trials were global trials that targeted a decrease in BP to 140 mm Hg as compared to standard care (range 160–180 mm Hg) using available intravenous medications in participating countries in patients randomized within 6 hours of onset. INTERACT 2 showed a non-significant trend toward less hemorrhage growth in the 140 mm Hg group with earlier treatment (3.1 cc more growth in standard group as compared to 140 mg Hg group when treated within 3 hours versus 1.2 cc in those treated at 3–6 hours). INTERACT 2 was not positive with regards to its primary endpoint, a modified Rankin (mRS) of 3–6, but was positive in its secondary endpoint of the distribution of the ordinal mRS. The results of this trial are the basis for the current Guidelines for Management of Acute ICH that target a systolic pressure of 140 mm Hg. In an analysis combining INTERACT I and II, there was evidence of a small decrease in the growth of ICH volume with the 140 target (ref).growth was significantly lower in the intensive BP reduction group compared with the standard treatment group (35 of 192 [18.2%] versus 46 of 162 [28.4%]; p = 0.02).

The Phase II pilot single-arm ATACH and Phase III randomized ATACH 2 trials focused on a more aggressive blood pressure target of a systolic BP of 110–139 mm Hg as compared to a standard control group of 140–179 mm Hg, using a single intravenous agent to lower blood pressure, nicardipine. ^20,21 ATACH 2 demonstrated a substantial and much quicker decrease in BP as compared to the INTERACT 2 trial but found no difference in outcome between the two groups. However, in a post-hoc analysis of those participants who received nicardipine within 2 hours of onset, 35 of the 192 (18.2%) of those treated with the 110–139 target had growth of ICH (> 33% growth from baseline to follow-up CT scans) as compared to 42 of 162 (28.4%) of those in the 140–179 target, p = 0.02. ²² This decrease in growth was accompanied by a greater proportion of participants with a modified Rankin Score (mRS) 0–2 at three months in the 110–139 target group as well (41.7% versus 27.8% in the 140–179 group). In another post-hoc analysis, those with a baseline systolic blood pressure ≥ 220 who were treated with the more aggressive target of a systolic blood pressure of 110–139 mm Hg had a greater rate of neurological deterioration within 24 hours as compared to the standard group (15.5% vs 6.8%; relative risk, 2.28 [95%CI,1.03–5.07]; P = .04). ²³ The rate of death and severe disability (39.0% vs 38.4%; relative risk, 1.02 [95%CI, 0.73–1.78]; P = .92) was not significantly different between the 2 groups. There was a significantly higher rate of kidney adverse events in participants randomized to intensive systolic blood pressure reduction (13.6% vs 4.2%; relative risk, 3.22 [95%CI, 1.21–8.56]; P = .01. Among the 48 participants with a pre-randomization SBP ≥ 220 mmHg, the rate of death or severe disability at 90 days was significantly higher for individuals randomized to intensive SBP reduction (66.7% vs 29.2%; RR, 2.29 [95%CI, 1.15–4.53]; P = .009). The rate of hematoma expansion (33.3% vs 26.1%; RR, 1.28 [95% CI, 0.52–3.11]; P = .59) was similar between the two treatment groups.

Finally, the Phase III Rapid Intervention with Glyceryl Trinitrate in Hypertensive Stroke Trial 2 tested the prehospital initiation of antihypertensive medication (transdermal glyceryl trinitrate - GTN) within 4 hours of onset in patients with suspected stroke that was continued for four days in addition to standard care of blood pressure. ^24,25 The trial found no difference in outcomes between the GTN and sham groups, but a predefined subgroup analysis found that patients with ICH in the GTN group had larger hematomas (1·95 [1·07–3·58]; p=0·030) on baseline imaging and a trend toward worse outcome. The authors speculated that early initiation of GTN might have prevented the initial vasoconstrictive response in the setting of an arterial rupture and so led to very early hematoma expansion. Secondly, venodilators, such as sodium nitroprusside and GTN, have been shown experimentally and clinically to raise intracerebral pressure and reduce cerebral blood flow, particularly if intracerebral pressure is already elevated. Reduced blood flow might then induce peri-hematoma ischemia. The findings of diffusion positive changes with MR imaging of ICH patients in other studies, particularly in those with highest and the greatest decrease in BP, is supportive of this speculation. ^26,27

Thus, while there is some evidence of benefit with lowering BP to a target systolic BP of 140 mm Hg in the INTERACT 2 trial and that very rapid initiation of blood pressure treatment may be beneficial, there is also evidence that very aggressive lowering of the highest levels of elevated BP may have adverse outcomes. The ideal approach to lowering of BP in patients with ICH, including the targeted decrease, the speed of decrease, the effectiveness of treatment by time from onset to treatment, and types of agents used, requires more research.

Treatment to Plug Leaky Blood Vessel with Clot

ICH Related to Anticoagulants

Rapid hemostasis at the site of an intracranial arteriolar rupture is another logical way to decrease growth of ICH and improve outcome. The randomized, open-label, blinded-endpoint International Normalized Ratio (INR) Normalization in Coumadin Associated Intracerebral Hemorrhage (INCH) Trial, a clear demonstration of the effectiveness of this approach, tested prothrombin complex concentrate (PCC) versus fresh frozen plasma (FFP) in 54 patients with ICH related to the use Vitamin K antagonists within 12 hours of onset. ²⁸ The trial was terminated on Feb 6, 2015, after inclusion of 50 patients, after a safety analysis demonstrated safety concerns. Two (9%) of 23 patients in the FFP group versus 18 (67%) of 27 in the PCC group reached the primary endpoint of INR ≤ 1.2 within 3 hours of treatment initiation (adjusted odds ratio 30.6, 95% CI 4·7–197·9; p=0·0003). At 24 hours, the adjusted difference in hematoma expansion was 16.4 mL less in the PCC group as compared to the FFP group (95% CI 2.9–29.9; p =0.018). 13 patients died: eight (35%) of 23 in the FFP group (five from hematoma expansion, all occurring within 48 h after symptom onset) and five (19%) of 27 in the PCC group (none from hematoma expansion).

Several single-arm trials of reversal agents for ICH related to novel oral anticoagulants, idarucizumab for dabigatran and andexanet alpha for Factor X inhibitors, have been published, providing additional therapeutic options to halt bleeding in these patients. ^29,30

Spontaneous ICH: Platelets and Platelet Stimulation in Setting of Antiplatelet Use

The Platelet Transfusion in Cerebral Hemorrhage (PATCH) Trial tested the use of platelet infusions in 190 patients with ICH who had been taking antiplatelet agents prior to the ICH. ³¹ The trial was stopped early for safety since the odds of death or dependence at 3 months were higher in the platelet transfusion group than in the standard care group (adjusted common odds ratio 2·05, 95% CI 1·18–3·56; p=0·01) and this was associated with a nonsignificant increase in thrombo-embolic events and increased serious adverse events related to intracerebral hemorrhage in the transfusion group. While the modest-sized trial has multiple limitations and the differences in mortality likely has multiple explanations, the authors speculate that infusion of platelets may have stimulated prothrombotic and inflammatory responses and wasn’t sufficient to reverse the effects of anti-platelet use on growth of hemorrhage.

Desmopressin is another medication that has been tested in small pilot studies in patients with ICH who have been taking anti-platelet agents. ³² It stimulates release of Von Willebrand Factor (VWF) and factor VIII from endothelial Weibel-Palade bodies. VWF is responsible for platelet adhesion to collagen and may also bind platelets through their glycoprotein IIb/IIIa receptors, so increased levels of VWF have the potential to compensate for the platelet function defect associated with antiplatelet drugs. A pilot trial is ongoing (Desmopressin for Reversal of Antiplatelet Drugs in Stroke Due to Haemorrhage – DASH, Clinical Trials.gov).

Spontaneous ICH: Antifibrinolytic Therapy

Tranexamic acid (TXA) is an antifibrinolytic agent that has been extensively studied in patients with bleeding of various causes, including ICH. The largest randomized trial for spontaneous ICH was the Tranexamic acid for hyperacute primary IntraCerebral Haemorrhage (TICH-2) Trial which enrolled 2325 participants with ICH within 8 hours of onset. ³³ Two grams of TXA (one gram over 10 minutes followed by one gram over 8 hours) was compared to matching placebo. TXA decreased growth of ICH significantly by 1.4 ml. However, there was no difference between the TXA and placebo groups who were dead or dependent (mRS > 3), a OR 0.82 (95% CI 0.65 to 1.03), p = 0.08.

The Spot Sign and Tranexamic Acid on Preventing ICH Growth - Australasia Trial (STOP-AUST) randomized 100 ICH patients with a spot sign on CT angiography within 4 ½ hours either 2g TXA (1g over 10 min followed by 1g over 8h) or matching placebo. ³⁴ The primary outcome was intracerebral hemorrhage growth (>33% relative or >6 mL absolute) at 24 h. The median time from onset to treatment was 150 minutes. There was no significant difference in the primary outcome between the two groups although there was a trend towards reduced presence of hematoma growth in the prespecified subgroup treated ≤3h from onset. Additional trials of TXA focused on even earlier treatment time from onset are ongoing.

Spontaneous ICH: Recombinant Factor VIIa (rFVIIa)

Recombinant factor VIIa (rFVIIa) is a manufactured recombinant protein identical to factor VIIa in humans that is approved for use in patients with congenital hemophilia, acquired hemophilia, and Glanzmann Thrombasthenia. ^35,36 rFVIIa interacts with tissue factor at the site vessel injury which leads to thrombin generation. In addition, rFVIIa activates factor X on the surface of activated platelets, leading to an enhanced thrombin burst at the site of injury. Thrombin converts fibrinogen to fibrin, producing a stable clot. The risk of rFVIIa is related to its generation of thrombin at the site of injured vessels or activated platelets that can lead to myocardial infarction and ischemic stroke. rFVIIa is easily and rapidly administered intravenously with rapid onset of action and ½ life of several hours.

rFVIIa is the only medication to substantially decrease bleeding in spontaneous ICH in large randomized trials of ICH but its effectiveness is time dependent (Table 1). ^37–39 After several small pilot trials, it was first tested in a Phase IIb dose finding trial of 399 participants in which rVIIa significantly decreased growth of ICH, decreased mortality (combined 3 doses compared to placebo), and improved outcome on the modified Rankin score. In addition, the effect of rFVIIa on decreasing ICH group was greatest in the subgroup (269 patients) treated within 3 hours of onset. The absolute increase in volume of ICH was 10.7 mL for the placebo group, as compared with 4.4 mL for the rFVIIa-treated patients (p = 0.009). Among those treated more than 3 hours after onset (115 patients), the mean increase in ICH volume was 14% for the placebo group, as compared with 16% for the rFVIIa groups (p = 0.86), and the absolute increase was 3.1 mL, as compared with 3.8 mL (p = 0.76). Serious thromboembolic AEs, mainly myocardial or cerebral infarction, occurred in 7% of rFVIIa-treated patients, as compared with 2% of those given placebo (p = 0.12).

Table 1:

Decrease in ICH growth between baseline and follow-up imaging in published trials

Treatment	Time Interval to Start of Treatment	Decrease in ICH growth (cc)
Lowering BP (Interact 2)	3–4 ½ hours	1.3
Lowering BP (Interact 2)	0–3 hours	3.1
Lowering BP (ATACH 2)	0–2 hours	2.0
Tranexamic Acid (TICH 2)	0–8 hours	1.4
Tranexamic Acid (STOP-AUST)	0–4 ½ hours (SPOT sign)	1.8
Tranexamic Acid (STOP-AUST)	0–3 hours (SPOT sign)	4.9
rFVIIa 20 μg/kg (FAST Trial)	0–4 hours	2.8
rFVIIa 80 μg/kg (FAST Trial)	0–4 hours	3.8
rFVIIa 80 μg/kg (FAST Trial)	0–2 hours	5.6
rFVIIa 160 μg/kg (Phase IIb Trial)	0–4 hours	6

Substantial growth (> 1/3 increase of initial volume of ICH) occurred in 35–40% of ICH subjects in the rFVIIa Phase IIb trial, although nearly ¾ of patients had some growth in bleeding between the baseline and 24-hour scan. In this trial, percentage growth (cumulative OR 0.84 [95% CI: 0.75, 0.92; p < 0.0001]), initial ICH volume (cumulative OR 0.94 [95% CI: 0.91, 0.97; p < 0.0001]), Glasgow Coma Scale (cumulative OR 1.46 [95% CI: 1.21, 1.82; p < 0.0001]), and age (cumulative OR 0.95 [95% CI: 0.92, 0.98; p = 0.0009]) predicted outcome on the mRS. ⁴⁰ For each mL increase in ICH from the baseline ICH volume and for each 10% increase in ICH growth, patients were 6% and 16% more likely to increase 1 full point on the modified Rankin Scale (mRS). ⁴⁰

The subsequent Phase III Recombinant Factor VIIa in Acute Intracerebral Hemorrhage Trial (FAST) of 841 participants also demonstrated decreased growth of ICH as compared to placebo, which was greatest when administered within 2 hours of onset, but no difference was seen in functional outcome between the two groups at 90 days. ³⁹ The overall frequency of thromboembolic SAEs was similar in the 3 groups; however, arterial events were more frequent in the group receiving 80 μg/kg of rFVIIa than in the placebo group (9% vs. 4%, p = 0.04).

Some of the differences in outcome between the Phase IIb and Phase III trials relate to imbalances between the placebo group and rFVIIa groups in the proportion of patients with IVH, a known factor associated with mortality and outcome (higher in placebo patients in the Phase IIb trial and higher in rFVIIa groups in the Phase III Trial). However, a subsequent post-hoc analysis focusing on ICH patients with hemorrhages < 60 cc and smaller IVHs, demonstrated the strong effect of time from onset to treatment, and to a lesser extent age, upon the potential effectiveness of rFVIIa. ³⁸ The importance of time to treatment is also demonstrated by the “Spot Sign” Selection of Intracerebral Hemorrhage to Guide Hemostatic Therapy (SPOTLIGHT) trial. ⁴¹ This trial tested rFVIIa versus placebo in participants with a spot sign on CT angiogram, a marker of ongoing bleeding. Unfortunately, the time from onset to treatment with study medication was longer due to the additional imaging and interpretation. As a result, almost all of the growth in ICH in both the rFVIIa and placebo groups occurred between the baseline CT and follow-up CT done at the start of study medication (Table 2- written communication, Andrew Demchuk, 2019). In other words, rFVIIa was given at a time point when most bleeding had stopped. Additional analyses of the rFVIIa Phase 3 and Phase 2b trials, using the rFVIIa for Acute hemorrhagic Stroke Administered at Earliest Time (FASTEST) inclusion criteria, demonstrate the importance of time to treatment in future trials of rFVIIa (Table 3).

Table 2:

SPOTLIGHT Trial: Volume of ICH+IVH at 3 time Points in Both Treatment Groups

Timing of Scan	Stroke onset to CT (hours)	rFVIIa (N=19) ICH+IVH vol (ml) Median (IQR)	Placebo (N=25) ICH+IVH vol (ml) Median (IQR)
Baseline CT	1.4 (1.2,2.6)	24.1 (16.0,41.4)	23.1 (11.5,53.0)
Immediate Post-dose CT	3.0 (2.5,4.3)	35.9 (20.8,63.0)	30.4 (21.4,63.1)
24-hour CT	26.6 (26.1,27.8)	31.3 (17.4,64.8)	33.3 (18.5,59.6)

Table 3:

Subgroup analyses from FAST and Phase IIb rFVIIa Trials which include FASTEST imaging criteria: baseline ICH volume of ≥ 2 and < 60 cc, and no or a small volume of IVH, i.e, IVH score ≤ 7.

FAST Trial: Minutes from onset to treatment in patients age ≤ 80	mRS 0–2 FVIIa	mRS 0–2 Placebo	Absolute % in mRS 0–2 in favor of rFVIIa at 90 days
≤ 150	42%	42%	0%
≤ 140	46%	41%	5%
≤ 130	49%	41%	8%
≤ 120*	52%	38%	14%
Phase IIB Trial: Minutes from onset to treatment in patients age ≤ 80	mRS 0–2 FVIIa	mRS 0–2 Placebo	Absolute % in mRS 0–2 in favor of rFVIIa at 90 days
≤150	42%	32%	10%
≤ 140	47%	30%	17%
≤ 130	50%	25%	25%
≤ 120	50%	20%	30%

^*N=25 in FVIIA and 32 in placebo group

Surgical removal of ICH

No large, well-designed Phase III randomized controlled surgical trial of ICH, whether using craniotomy alone, minimally invasive stereotactic removal of ICH, or both, has been positive per its primary endpoint. This lack of benefit is reflected in the guidelines for acute management of ICH. ^14,19 Pooled and meta-analysis of present trial to date have suggested that patients with earlier removal, mid-range volumes of ICH, minimally-invasive removal, and younger age may be more likely to benefit with surgical removal. ^42,43 In addition, the Minimally Invasive Surgery plus Alteplase in Intracerebral Hemorrhage Evacuation (MISTIE) 3 trial of minimally invasive removal of ICH using rt-PA after stabilization of ICH volume by repeated imaging, indicated that more complete removal of ICH, and standardization of surgical techniques were associated with better outcomes. ^44,45 Trials of minimally invasive mechanical removal of ICH without rt-PA are ongoing. ⁴⁶

Much of the rationale for removal of ICH has been based upon preclinical work highlighting the toxicity of blood products on surrounding brain which likely do have a longer time course. ^47–49 Randomized trials of medical therapies of neuroprotection to decrease the toxicity of ICH on surrounding brain tissue have been neutral thus far but remain an area of future study. ^50–52 However, much of the damage associated with ICH is mechanical whose time course is likely much shorter based upon the limited preclinical animal data exploring this issue. ^53–55 If mechanical damage is time-dependent, the future success of surgical removal of ICH will depend upon very early removal, stabilization of ongoing bleeding during surgery, removal of sufficient blood to limit mechanical damage and toxicity, and minimization of brain injury during access to the ICH. One advantage of surgical removal of ICH over medical therapies, if done very quickly, is that it could be done in patients with larger volumes of ICH in whom outcomes are uniformly very poor with medical therapies designed to slow or halt bleeding. Hyperacute removal is likely to involve minimally invasive techniques that monitor and treat ongoing bleeding. ^56,57 Pilot trials of ultra-early surgery are ongoing or are in planning stages. Effective hemostatic treatments that limit ongoing bleeding could also lead to improved hemostasis during ultra-early removal of ICH.

Where We Are Now and the Path Forward

The large majority of bleeding in ICH occurs within the first 2–3 hours after onset and if we are going to change the natural history, as we have for ischemic stroke, we will have to replicate and surpass the time restrictions of the initial t-PA trials. rFVIIa has the best data thus far for decreasing ICH growth although data from tranexamic acid and blood pressure trials also indicate the importance of rapid treatment. All medical approaches must strive to limit adverse events that can limit the impact of slowing the growth of ICH. We need to exclude patients in the medical trials who have an extremely poor prognosis even if we can slow the growth of ICH (for example, patients with an ICH volume ≥ 60 cc or those with large volumes of intraventricular hemorrhage). Surgical trials have to address the early pathophysiology of mechanical damage due to ICH while limiting brain injury from clot removal. New trials focusing on very early treatment are in the planning stages.

The NINDS has funded a trial of medical therapy for ICH that attempts to address the previously discussed issues. The objective of the FASTEST Trial is to establish the first treatment for acute ICH within a time window and subgroup of patients that is most likely to benefit. ⁵⁸ FASTEST is a randomized, double-blind controlled efficacy trial of rFVIIa plus best standard therapy vs. placebo plus best standard therapy (including INTERACT 2 goal of target BP of 140 mm Hg). The trial Includes subjects with a baseline volume of ICH < 60 cc, no or a small volume of IVH (IVH score ≤ 7 ⁵⁹), age ≤ 80 years, and most importantly, treatment begun within 2 hours. To enroll subjects within 2 hours, and as many as possible within one hour, we are using mobile stroke units, ^60–62 exception from informed consent or EFIC, and improved acute stroke treatment processes as for ischemic stroke including automatic calculation of volume of ICH, decreased door to needle, etc.

We will randomly assign patients in a 1:1 ratio to intravenous rFVIIa or placebo at a dose of 80 μg/kg (maximum 10,000 μg or 10 mg) and administered intravenously over 2 minutes. All investigators and patients will be blinded throughout the course of the study. To limit time to completion and interpretation of imaging, we require only baseline non-contrast CT but will collect CT angiography if it is done as standard of care. The trial plans for a maximum of 860 participants at 100+ hospitals and 15+ mobile stroke units in the United States, Canada, Japan, Germany, Spain, and the United Kingdom. The primary outcome measure is the distribution of the ordinal mRS at 180 days (specifically, mRS 0–2, 3, 4–6 per discussions with the FDA).

Improving outcome after acute ICH is likely to require a multi-pronged approach of decreasing bleeding in the first hours, rapid removal of blood with minimization of brain injury, therapies that limit toxicity of blood within the brain, and innovative approaches to neuro-recovery which for ICH is still in its infancy. Our successes in changing the outcomes of ischemic stroke makes us hopeful that the next chapters in the story of ICH treatment will also lead to improved outcomes for our patients.

Non-standard Abbreviations and Acronyms

ATACH

Antihypertensive Treatment of Acute Cerebral Hemorrhage

CT

Computed tomography

INTERACT

Intensive blood pressure reduction in acute cerebral hemorrhage trial

FAST

Recombinant Factor VIIa in Acute Intracerebral Hemorrhage

FASTEST

rFVIIa for Acute hemorrhagic Stroke Administered at Earliest Time

GTN

glyceryl trinitrate

ICH

Intracerebral hemorrhage

IVH

Intraventricular hemorrhage

MISTIE

Minimally Invasive Surgery plus Alteplase in Intracerebral Hemorrhage Evacuation

mRS

Modified Rankin Scale

PATCH

Platelet Transfusion in Cerebral Hemorrhage