Advances in intracerebral haemorrhage management
Intracerebral haemorrhage (ICH) contributes 10-15% of strokes, is associated with high mortality (35-50% at 30 days) and only 20% of survivors are independent at 6 months [1,2]. Several features at ICH onset are useful indicators of prognosis: volume of the haematoma, level of consciousness and presence of hydrocephalus on initial evaluation are reliable predictors of 30-day mortality [3,4]. In addition, hyperglycaemia at presentation correlates with increased mortality . Recent studies have suggested that some markers can be added to this list of prognostic indicators: hyperthermia, leukocytosis, elevated fibrinogen levels , as well as increased levels of matrix metalloproteinases (MMPs), especially MMP-3, fibronectin-c, interleukin (IL)-6, glutamate and tumour necrosis factor-α (TNF-α) correlate with various detrimental features in acute ICH, including early neurological deterioration, perihaematomal oedema and mortality [7-9].
The traditional management of ICH has included the treatment of hypertension, possible associated coagulopathy and increased intracranial pressure (ICP), along with supportive measures such as maintaining the permeability of the airway, preventing/managing seizures and preventing systemic complications such as fever, electrolytic disturbances, deep-vein thrombosis and pulmonary embolism. More specific measures aimed at reducing the local effects of the ICH have included either surgical removal or limitation of growth of the haematoma.
The permeability of the airway is an essential need for the obtunded, lethargic or comatose patient with acute ICH, who is prone to hypoventilation, vomiting, poor handling of secretions and aspiration. This is best achieved by early endotracheal intubation, generally indicated for patients with a Glasgow Coma Scale (GCS) score of 8 or less, and is best performed with the administration of intravenous (i.v.) thiopental (0.5-1 mg kg−1) or lidocaine (1 mg kg−1). These short-acting agents block increases in ICP caused by tracheal and oro-pharyngeal stimulation, without prolonged effects on the level of consciousness or neurological function .
Hyperglycaemia, a factor that is associated with poor outcomes in ICH , needs to be corrected. Although there are no specific guidelines for patients with ICH, it seems prudent to follow those applied to patients with acute ischaemic stroke, in whom values above 300 mg dL−1 should be corrected .
The management of hypertension in the acute phase of ICH has been controversial. There is, on the one hand, the issue of possible (but still unproven) contribution of uncontrolled hypertension to the early growth of the haematoma [12-14], while at the same time there is a theoretical concern about creating perihaematomal ischaemia by lowering the blood pressure (BP) excessively . While these issues remain largely unresolved, the current recommendations regarding BP management in patients with acute ICH are not based on data from controlled clinical trials, but rather on observations derived from clinical series. These have suggested that BP management in this setting cannot aim for a specific level of BP (such as mean BP < 130 mmHg or systolic BP < 180 mmHg ), but rather needs to be individualized, taking into consideration issues such as age, baseline BP, mechanism of ICH and elevated ICP. It is expected that more specific guidelines will become available after the completion of the trials ATACH (Antihypertensive Treatment in Acute Cerebral Hemorrhage) and INTERACT (Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage), which are exploring the effects of targeted BP reduction on mortality and long-term disability in survivors.
ICH related to anticoagulant treatment (warfarin) is associated with high mortality, which relates to generally large and rapidly enlarging haematomas . Among its risk factors, excessive prolongation of the International Normalized Ratio (INR) frequently plays a primary role . Thus, once the diagnosis of ICH has been established by a computed tomography (CT) scan, there is an emergent need to correct the warfarin-induced coagulopathy, with the aim of bringing the INR to normal. This is generally achieved by a combination of measures, which include i.v. injection of 10 mg vitamin K1, and infusions of coagulation factors, the most commonly used being fresh frozen plasma (FFP) or prothrombin complex concentrate (PCC), which replenish the reduced levels of vitamin K-dependent coagulation factors (II, VII, IX, X). FFP has the disadvantages of delayed administration by requiring thawing and the infusion of large volumes of fluid, the latter exposing patients with heart disease to the risk of congestive heart failure . PCCs require smaller volumes of infusion and correct the coagulopathy faster than FFP , but have the disadvantages of limited availability and promoting thromboembolic complications. These limitations of FFP and PCCs, along with substantial disagreement among experts about the preferred ways of achieving rapid anticoagulation reversal , have stimulated the search for better alternatives. A promising approach has been the use of recombinant activated factor VII (rFVIIa), which can normalize the INR very rapidly, without the need for infusing large volumes of i.v. fluids . A small clinical series suggested the safety of this approach in patients with warfarin-related ICH , but this needs to be validated in a prospective trial, since rFVIIa use carries a potentially substantial risk of coronary and cerebral thromboembolic complications [24-26].
Management of increased intracranial pressure
In the management of increased ICP, the aims are to achieve normal levels of ICP while maintaining a cerebral perfusion pressure of 70 mmHg or higher . Early attempts at reducing morbidity and mortality of ICH with the use of anti-brain oedema agents, such as glycerol  and steroids , yielded negative results, which in the case of steroids (dexamethasone) resulted in worse outcomes in the treated group as a result of infectious complications. The currently available options are limited, since the commonly used approaches of hyperventilation, osmotic diuresis with mannitol, cerebrospinal fluid (CSF) drainage, barbiturate coma and induced hypothermia are all associated with major side-effects and complications, while their value remains for the most part unproven.
A number of measures are important in reducing ICP or preventing its transient or persistent elevation. These include: (1) Elevation of the head. An angle of head flexion of 30°, while avoiding head rotation, favours intracranial venous drainage and lowers ICP; (2) Fever control. Fever is an independent predictor of poor outcomes in patients with ICH [6,30], at least partially by contributing to increased ICP . Aggressive management of fever is thus mandatory in patients with ICH. Whether induced hypothermia has a place in this setting is less clearly established. Although cooling to 32-34°C lowers ICP, the risks of pulmonary infections, coagulopathy and rebound intracranial hypertension upon reversal may not result in improved outcome ; (3) Sedation and analgesia. For the intubated patient with increased ICP, agitation results in transient elevations of ICP that can further compromise neurological function and promote tissue displacements. Sedation with propofol or midazolam, coupled with analgesia with fentanyl, generally accomplishes the goal of avoiding agitation-induced elevations of ICP; (4) Seizure control and prophylaxis. Seizures in ICH patients are particularly common in those with lobar haemorrhages, and they tend to occur from the time of onset onwards . In addition, non-convulsive seizures are often documented by continuous electroencephalogram (EEG) monitoring in the ICU . For these reasons, vigilance for seizures has to be close, and treatment with i.v. agents should be started promptly, as their presence is associated with poor outcome . The use of prophylactic anticonvulsants in ICH patients without clinical or electrographic evidence of seizures is controversial, since only limited data  justify it in the subgroup with lobar ICH, and no prospective controlled trials have examined the issue; (5) Hyperventilation. Although hyperventilation aiming for PCO2 levels of 30-35 mmHg is a proven and rapid method of lowering ICP, it has the disadvantages of lowering cerebral blood flow and having a brief duration effect . For these reasons, it is mostly recommended as a measure to be used for short periods of time, and is especially indicated in emergency situations for the management of impending tissue shifts and herniation; (6) Osmotic diuresis. The use of mannitol is associated with decreases in ICP by a combination of mechanisms that include shifting of water from the brain parenchyma into the intravascular space, decreased CSF production, reduction of blood viscosity and vasoconstriction, all resulting in a decrease in brain volume . Despite these effects, its use is also associated with serious negative consequences, the most important ones being hypovolemia, hyperosmolality and electrolytic disturbances. Possibly as a result of these detrimental side-effects, the efficacy of mannitol in this setting has not been documented in placebo-controlled studies ; (7) Barbiturate coma. High-dose i.v. barbiturates reduce cerebral metabolism and cerebral blood flow, resulting in reduced ICP. I.v. pentobarbital, given in an initial loading dose of 3-10 mg kg−1 followed by 1-2 mg kg−1 h−1, produces rapid reductions in ICP . Severe hypotension is its main serious side-effect, and inotropic support is often necessary to maintain adequate BP levels ; and (8) Ventricular drainage. Ventriculostomy with CSF removal effectively reduces ICP, and in addition allows for ICP monitoring. It is appropriate for patients with raised ICP resulting from hydrocephalus, a situation most commonly seen in those with thalamic, cerebellar or caudate haemorrhages . Unfortunately, this technique has a risk of bacterial meningitis [39,40], and is associated with high mortality .
Measures intended to decrease haematoma size
Surgical haematoma removal
The generally disappointing effects of the measures discussed above, coupled with the obvious intuitive advantage of removing or decreasing the size of the ICH, have led to a number of attempts at documenting the value of surgical removal of the haematoma. Many of the early studies were small, non-randomized clinical series that produced conflicting results based on the differences in patient populations studied, timing and type of surgical procedure and use of different parameters for measuring outcome (Table 1) [42-49]. Despite these inconclusive data, certain types of ICH appeared to benefit from surgical therapy, thus justifying a large-scale randomized clinical trial comparing the surgical and conservative management of ICH.
The international Surgical Trial in Intracerebral Haemorrhage (STICH)  addressed the issue by including 1033 patients from 83 hospitals in 27 countries, randomly assigned to either surgery or conservative management within 72 h of symptom onset, with haematomas larger than 2 cm in diameter to be operated within 24 h from randomization. Patients in poor neurological condition (GCS < 5) were excluded. The study population included those patients in whom the participating neurosurgeon determined that there was uncertainty about the preferred approach to management (the clinical ‘equipoise'), while those considered to be good candidates for surgical treatment were not included in the randomization process. The specific operative technique used in those randomized to surgery was the decision of the participating neurosurgeon, and included craniotomy in 75%, while the rest had less invasive procedures. The primary outcomes were death and disability at 6 months, the latter measured with the extended Glasgow Outcome Scale (eGOS), while secondary outcomes were the Barthel Index (BI), modified Rankin Scale (mRS) and mortality at 6 months. The 506 surgical and 530 medical group patients were well balanced with regard to risk factors, haematoma location (39% lobar and 42% basal ganglionic in the surgical group vs. 40% and 42% in the medical group, respectively) and volume (median of 40 cm3 vs. 37 cm3, respectively), the latter placing the majority of patients in the category of moderate-size ICH. There was a high (26%) rate of cross-over from the medical to the surgical arm, mostly on account of neurological deterioration or recurrent bleeding, causes that accounted for 85% of the cross-overs. The cross-over subjects underwent craniotomy as the surgical technique in 85% of the cases.
In the intention-to-treat analysis, there were no significant differences between the two groups, as a similar percentage (26% in the surgical, 24% in the medical group) had a favourable outcome, measured as an mRS of 2 or less and a BI higher than 95 at 6 months. Similarly, mortality rates at 6 months were virtually identical (36% in the surgical group, 37% in the medical group). Subgroup analyses identified patients with lobar haematomas located at a depth of 1 cm or less from the surface, as well as those with GCS of 9-12, and patients treated with craniotomy (as opposed to other surgical techniques), who benefited from surgery, but the differences did not reach statistical significance. Another group that showed a potential benefit from surgery was the small group with anticoagulant-related ICH . On the other hand, comatose patients (GCS of 5-8) did less well with surgery. Overall, the trial results failed to document superiority of one treatment modality over the other in this population, but potentially selected subgroups may benefit from craniotomy, particularly those with superficially located lobar haemorrhages who present with a GCS of 9 or higher. This approach will be tested in STICH II, which will randomize patients with lobar ICH of 2-100 cm3 in volume located within 1 cm of the cortical surface, with GCS of 5 or higher, to be randomized within 48 h of ICH onset, with surgery to follow within 12 h from randomization, using the same clinical ‘equipoise' principle used in STICH .
Reduction of haematoma expansion by a non-surgical approach - the factor VII experience
In view of the lack of evidence in favour of surgical treatment of ICH, there has been a renewed interest in assessing medical measures that may have an impact in the prognosis of ICH. The non-surgical approach of ICH has been based on the notion, currently well established, that the intracerebral haematoma has a high tendency to increase its volume in the first hours after onset [12-14]. It has been shown that almost 40% of the intracerebral haematomas grow substantially in size during the first few hours after the onset of symptoms . In addition, this volume increase is associated consistently with deterioration in neurological function. This has stimulated an interest in evaluating measures able to stop or decrease early haematoma growth. The haemostatic agent rFVIIa that has been approved by the Food and Drug Administration (FDA) since 1999 for the treatment of bleeding complications in patients with haemophilia and circulating inhibitors (antibodies) against factor VIII or IX. Due to its ability to generate thrombin and thus stop the haemorrhagic process, the use of this agent in the first hours after the onset of ICH was tested recently in a phase IIb, dose-finding controlled clinical trial by Mayer and colleagues . A group of 399 patients were randomly assigned to receive, within 4 h of ICH onset, one of three doses of rFVIIa (40, 80 or 160 μg kg−1) or placebo. The ICH was documented by CT scan and its volume was measured at baseline and then within 24 h of receiving the assigned treatment, with the primary aim of comparing the magnitude of haematoma growth in the three groups treated with rFVIIa against placebo. The primary outcome measure was the percentage of change in volume of the haematoma in CT scan after 24 h from treatment. Secondary end-points included the mRS, BI, eGOS and NIHSS, as well as mortality, all assessed at 90 days.
The results showed a significantly smaller increase in haematoma volume after 24 h from treatment in the groups which received rFVIIa, corresponding to increases in volume of 16%, 14% and 11% for the doses of 40, 80 and 160 μg kg−1, respectively, in comparison with 29% of median growth of the haematoma in the placebo group (P = 0.01 for the three rFVIIa groups compared with placebo). This observed difference in haematoma growth was only significant in patients who were treated within 3 h of onset of symptoms, with a median haematoma growth of 13% for the combined rFVIIa groups vs. 39% for placebo (P = 0.004), while those treated between 3 and 4 h did not show a difference in haematoma growth in comparison with placebo (16% vs. 14%, respectively). With regard to the secondary end-points, mortality was significantly lower in the rFVIIa group (18%, in comparison with 29% for the placebo group, a relative risk reduction of 38%). The functional scales showed a trend towards a higher percentage of patients with minimal functional deficits at 90 days in the rFVIIa groups, with a suggestion for a dose relationship, as the largest dose (160 μg kg−1) showed generally higher efficacy in comparison with the two smaller doses, but these differences did not reach statistical significance.
These positive effects of rFVIIa were unfortunately accompanied by an increase in the thromboembolic complications in the rFVIIa group in comparison with placebo: they occurred in 7% of the rFVIIa group, and in only 2% in the placebo group; furthermore, when only the arterial thromboembolic complications (myocardial infarction, ischaemic stroke) were considered, these occurred in 5% of the rFVIIa group and in none in the placebo group (P = 0.01). This tendency for an increase in arterial thromboembolic complications in the study of Mayer and colleagues  has also been observed with the use of rFVIIa for other indications . In view of these observations, a phase III trial (FAST, for rFVIIa in Acute Haemorrhagic Stroke Treatment), using lower doses of rFVIIa (20 and 80 μg kg−1) in comparison with placebo, and with inclusion and exclusion criteria essentially identical to those of the Mayer and colleagues  study, has been recently completed. The use of rFVIIa at a lower dose was aimed at reducing its thromboembolic complications, ideally with replication of the therapeutic value of this agent in limiting the growth of the ICH. In the event of confirmatory phase III trial results, it is likely that this treatment will be incorporated into the acute management of ICH; despite the high medication cost, a recent cost-effectiveness analysis of rFVIIa use in ICH suggested that this agent is both cost-effective and cost-saving . Furthermore, the success of this therapy in spontaneous ICH could eventually lead to its evaluation in cases of ICH provoked by therapeutic measures, such as haemorrhages associated with anticoagulant and thrombolytic treatment. In the setting of warfarin-related ICH, the issue of the risk of thromboembolic complications with rFVIIa will be particularly relevant, since the use of this agent, coupled with the discontinuation of warfarin, will be expected to be additive with regard to the potential for thromboembolism (i.e. the prothrombotic effect of rFVIIa plus the thromboembolic risk of the disorder that was the original indication for chronic anticoagulant treatment). Finally, the eventual documentation of the value of this agent in ICH could stimulate future interest in associating it with the surgical evacuation of ICH, with the hope of using this haemostatic agent in combination with minimally invasive surgery , which may result in outcomes that are superior to the purely conservative management of ICH.
1. Broderick JP, Brott T, Tomsick T et al.
. Intracerebral hemorrhage more than twice as common as subarachnoid hemorrhage. J Neurosurg
2. Counsell C, Boonyakarnkul S, Dennis M et al.
. Primary intracerebral haemorrhage in the Oxfordshire community stroke project. 2: prognosis. Cerebrovasc Dis
3. Broderick JP, Brott TG, Duldner JE, Tomsick T, Huster G. Volume of intracerebral hemorrhage: a powerful and easy-to-use predictor of 30-day mortality. Stroke
4. Diringer MN, Edwards DF, Zazulia AR. Hydrocephalus: a previously unrecognized predictor of poor outcome from supratentorial intracerebral hemorrhage. Stroke
5. Fogelholm R, Murros K, Rissanen A, Avikainen S. Admission blood glucose and short term survival in primary intracerebral haemorrhage: a population based study. J Neurol Neurosurg Psychiat
6. Leira R, Davalos A, Silva Y et al.
. Early neurologic deterioration in intracerebral hemorrhage: predictors and associated factors. Neurology
7. Alvarez-Sabin J, Delgado P, Abilleira S et al.
. Temporal profile of matrix metalloproteinases and their inhibitors after spontaneous intracerebral hemorrhage: relationship to clinical and radiological outcome. Stroke
8. Silva Y, Leira R, Tejada J et al.
. Molecular signatures of vascular injury are associated with early growth of intracerebral hemorrhage. Stroke
9. Castillo J, Davalos A, Alvarez-Sabin J et al.
. Molecular signatures of brain injury after intracerebral hemorrhage. Neurology
10. Diringer MN. Intracerebral hemorrhage: pathophysiology and management. Crit Care Med
11. Adams HP, Adams RJ, Brott T et al.
. Guidelines for the early management of patients with ischemic stroke: a scientific statement from the Stroke Council of the American Stroke Association. Stroke
12. Kazui S, Naritomi H, Yamamoto H, Sawada T, Yamaguchi T. Enlargement of spontaneous intracerebral hemorrhage: incidence and time course. Stroke
13. Brott T, Broderick J, Kothari R et al.
. Early hemorrhage growth in patients with intracerebral hemorrhage. Stroke
14. Fujii Y, Takeuchi S, Sasaki O, Minakawa T, Tanaka R. Multivariate analysis of predictors of hematoma enlargement in spontaneous intracerebral hemorrhage. Stroke
15. Powers WJ, Zazulia AR, Videen TO et al.
. Autoregulation of cerebral blood flow surrounding acute (6 to 22 hours) intracerebral hemorrhage. Neurology
16. Broderick JP, Adams HP, Barsan W et al.
. Guidelines for the management of spontaneous intracerebral hemorrhage: a statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke
17. Hart RG, Boop BS, Anderson DC. Oral anticoagulants and intracranial hemorrhage: facts and hypotheses. Stroke
18. Fang MC, Chang Y, Hylek EM et al.
. Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med
19. Schulman S. Care of patients receiving long-term anticoagulant therapy. N Engl J Med
20. Lankiewicz MW, Hays J, Friedman KD, Tinkoff G, Blatt PM. Urgent reversal of warfarin with prothrombin complex concentrate. J Throm Haemost
21. Aguilar MI, Hart RG, Kase CS et al.
. Treatment of warfarin-associated intracerebral hemorrhage: literature review and expert opinion. Mayo Clin Proc
22. Deveras RA, Kessler CM. Reversal of warfarin-induced excessive anticoagulation with recombinant human factor VIIa concentrate. Ann Intern Med
23. Freeman WD, Brott TG, Barrett KM et al.
. Recombinant factor VIIa for rapid reversal of warfarin anticoagulation in acute intracranial hemorrhage. Mayo Clin Proc
24. Mayer SA, Brun NC, Begtrup K et al.
. Recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med
25. O'Connell KA, Wood JJ, Wise RP, Lozier JN, Braun MM. Thromboembolic adverse events after use of recombinant human coagulation factor VIIa. JAMA
26. Sugg RM, Gonzalez NR, Matherne DE et al.
. Myocardial injury in patients with intracerebral hemorrhage treated with recombinant factor VIIa. Neurology
27. Chambers IR, Banister K, Mendelow AD. Intracranial pressure
within a developing intracerebral hemorrhage. Br J Neurosurg
28. Yu YL, Kumana CR, Lauder IJ et al.
. Treatment of acute cerebral hemorrhage with intravenous glycerol: a double-blind, placebo-controlled, randomized trial. Stroke
29. Poungvarin N, Bhoopat W, Viriyavejakul A et al.
. Effects of dexamethasone in primary supratentorial intracerebral hemorrhage. N Engl J Med
30. Schwarz S, Hafner K, Aschoff A, Schwab S. Incidence and prognostic significance of fever following intracerebral hemorrhage. Neurology
31. Rossi S, Zanier ER, Mauri I et al.
. Brain temperature, body core temperature and intracranial pressure
in acute cerebral damage. J Neurol Neurosurg Psychiat
32. Schwab S, Georgiadis D, Berrouschot J, Schellinger PD, Graffagnino C, Mayer SA. Feasibility and safety of moderate hypothermia after massive hemispheric infarction. Stroke
33. Passero S, Rocchi R, Rossi R et al.
. Seizures after spontaneous supratentorial intracerebral hemorrhage. Epilepsia
34. Vespa PM, O'Phelan K, Shah M et al.
. Acute seizures after intracerebral hemorrhage: a factor in progressive midline shift and outcome. Neurology
35. Stocchetti N, Maas AIR, Chieregato A, van der Plas AA. Hyperventilation in head injury: a review. Chest
36. Misra UK, Kalita J, Ranjan P, Mandal SK. Mannitol in intracerebral hemorrhage: a randomized controlled study. J Neurol Sci
37. Wijman CAC, Kase CS. Intracerebral hemorrhage: medical considerations. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM, eds. Stroke: Pathophysiology, Diagnosis, and Management
, 3rd edn. New York, USA: Churchill Livingstone Inc., 1998: 1359-1372.
38. Schwab S, Spranger M, Schwarz S, Hacke W. Barbiturate coma in severe hemispheric stroke: useful or obsolete? Neurology
39. Lozier AP, Sciacca RR, Romagnoli MF et al.
. Ventriculostomy-related infections: a critical review of the literature. Neurosurgery
40. Halloway KL, Barnes T, Choi S et al.
. Ventriculostomy infections: the effect of monitoring duration and catheter exchange in 584 patients. J Neurosurg
41. Schade RP, Schinkel J, Visser LG, Van Dijk JM, Voormolen JH, Kuijper EJ. Bacterial meningitis caused by the use of ventricular or lumbar cerebrospinal fluid catheters. J Neurosurg
42. McKissock W, Richardson A, Taylor J. Primary intracerebral haemorrhage: a controlled trial of surgical and conservative treatment in 180 unselected cases. Lancet
43. Juvela S, Heiskanen O, Poranen A et al.
. The treatment of spontaneous intracerebral hemorrhage: a prospective randomized trial of surgical and conservative treatment. J Neurosurg
44. Batjer HH, Reisch JS, Allen BC et al.
. Failure of surgery to improve outcome in hypertensive putaminal hemorrhage: a prospective randomized trial. Arch Neurol
45. Morgenstern LB, Frankowski RF, Shedden P et al.
. Surgical treatment for intracerebral hemorrhage (STICH): a single-center, randomized clinical trial. Neurology
46. Zuccarello M, Brott T, Derex L et al.
. Early surgical treatment for supratentorial intracerebral hemorrhage: a randomized feasibility study. Stroke
47. Auer LM, Deinsberger W, Niederkorn K et al.
. Endoscopic surgery versus medical treatment for spontaneous intracerebral hematoma: a randomized study. J Neurosurg
48. Teernstra OPM, Evers SMAA, Lodder J, Leffers P, Franke CL, Blaauw G. Stereotactic Treatment of Intracerebral Hematoma by Means of a Plasminogen Activator: a Multicenter Randomized Controlled Trial (SICHPA). Stroke
49. Hattori N, Katayama Y, Maya Y, Gatherer A. Impact of stereotactic hematoma evacuation on activities of daily living during the chronic period following spontaneous putaminal hemorrhage: a randomized study. J Neurosurg
50. Mendelow AD, Gregson BA, Fernandes HM et al.
. Early surgery versus initial conservative treatment in patients with spontaneous supratentorial intracerebral haematomas in the international Surgical Trial in Intracerebral Haemorrhage (STICH): a randomised trial. Lancet
51. Rabinstein AA, Wijdicks EFM. Surgery for intracerebral hematoma: the search for the elusive right candidate. Rev Neurol Dis
52. Earnshaw SR, Joshi AV, Wilson MR, Rosand J. Cost-effectiveness of recombinant activated factor VII in the treatment of intracerebral hemorrhage. Stroke
53. Broderick JP. The STICH trial: what does it tell us and where do we go from here? Stroke