Introduction

Aneurysmal subarachnoid hemorrhage (SAH) accounts for 5% of all stroke cases and affects up to 30,000 North Americans yearly [1]. Early brain injury that occurs at the time of bleed is the leading cause of mortality (30–70%) after SAH [1, 2]. SAH survivors are at risk of developing delayed cerebral vasospasm, delayed cerebral ischemia, or delayed ischemic neurological deficits during the hospital course [2]. Delayed vasospasm develops in approximately 70% of patients between 3 and 14 days after SAH [1, 2]. For decades, delayed vasospasm has been considered the single and the most important cause of delayed cerebral ischemia and poor outcome [3]. The basic and clinical research has been focused on finding strategies to prevent and/or treat delayed vasospasm. However, lack of prevention of delayed cerebral ischemia and improved outcome in a recent clinical trial (CONSCIOUS-1) that successfully prevented the development of delayed vasospasm has raised doubts on the importance of vasospasm in delayed ischemic injury and the outcome after SAH [4]. Recent reviews of the experimental and clinical literature indicate that the presence of delayed vasospasm is not a pre-requisite for delayed ischemic injury and poor outcome after SAH [2, 5]. In fact, 21% of SAH survivors, who do not develop vasospasm, develop delayed ischemic injury, and only 20–30% of those, who do develop delayed vasospasm actually, suffer from delayed ischemic injury [2]. It maybe that the pathological mechanisms that activate within minutes after SAH and lead to early brain injury play an important role in the pathogenesis of delayed ischemic injury and poor outcome [6]. This manuscript summarizes the animal and human literature addressing the mechanisms of early brain injury after SAH and the importance of its early treatment.

Early Brain Injury by SAH

Early brain injury is the product of pathological mechanisms triggered in the brain during the first 72 h after SAH (Fig. 1). These mechanisms are activated at aneurysm rupture and evolve with time affecting the course and the outcome of SAH (Table 1) [7–9]. Below, we discuss the pathological mechanisms most pertinent to early brain injury after SAH.

Fig. 1
figure 1

Mechanisms of early brain injury after SAH: A number of changes in cerebral environment and function occur during the first 72 h after SAH. Some of the major changes are listed. See text for explanation. ICP intracranial pressure, CPP cerebral perfusion pressure, CBF cerebral blood flow, NO nitric oxide, NOS nitric oxide synthase, ET-1 endothelin-1

Full size image
Table 1 The timeline of pathological alterations leading to early brain injury after SAH
Full size table

Mechanical Trauma

The first injury to the brain after the aneurysm rupture is mechanical. This trauma evokes constriction of the artery harboring the ruptured aneurysm and its compression by blood filling in the subarachnoid cisterns [10, 11]. Sudden rise in intracranial pressure that may reach as high as 2,000 mm H2O (161.8 mmHg) [12] stops further bleed and compresses cerebral arteries and tissue. Depending upon the amount released, blood not only stretches the subarachnoid space, but also flows into the branching channels and envelops branches of the conducting artery [13]. The stretching of the subarachnoid space by blood is mechanically transferred to the vessels near the aneurysm leading to spasm of surrounding arteries [14]. Over the course of its presence, the subarachnoid blood clot evokes the early brain injury [15] and the delayed spasm [16].

Altered Cerebral Physiology

Intracranial Pressure

Intracranial pressure (ICP) rises when blood released at the time of aneurysm rupture fills up the subarachnoid cisterns displacing the cerebrospinal fluid (CSF). Most awake patients describe this moment as the onset of “the worst headache of my life” [12]. Two patterns of ICP rise are recognized and compress brain tissue by different mechanisms. In the first pattern, observed in most patients, ICP peaks to a value near diastolic blood pressure and then falls and settles near but slightly above the baseline [17]. In this pattern, the volume of bleed is small but cerebral edema is present [12]. In the second pattern, observed in some patients, ICP remains elevated, due to the mass effect from enlarging hematoma or due to the development of acute hydrocephalus [12, 18]. The second pattern of ICP rise is associated with high mortality [12, 19]. The terms “ischemic-edematous lesion” and “hemorrhagic-compressive lesion” have been used to differentiate the brain compressing forces associated with the two ICP patterns [12, 19]. The intensity of ICP rise is linked to hemorrhage volume, obstruction of the CSF outflow, diffuse vasoparalysis, and distal cerebral arteriolar vasodilation. The severity of ICP is associated with changes in cerebral metabolism, inflammation, fall in cerebral blood flow (CBF), and development of early and delayed cerebral ischemia [20]. Hence, the extent of ICP rise is often used to predict outcome of SAH [15, 21] (Fig. 2).

Fig. 2
figure 2

The timeline of physiological alterations during the first 72 h after SAH. Within seconds after SAH: ICP rises, CBF and CPP fall, and BP increases [17]. Five minutes after SAH: ICP declines towards basal value and CPP and BP recover to the basal value, CBF remains decreased, and CBF autoregulation is impaired [17, 37]. Sixty minutes after SAH: ICP stabilizes to a new plateau that is above the basal value, CPP and BP have recovered, CBF is still decreased, and CBF autoregulation still impaired [17, 37]. Twenty-four hours after SAH: ICP is still at the 60-min value, CPP and BP are recovered, CBF is decreased, and CBF autoregulation is still impaired [140]. Seventy-two hours after SAH: ICP, CPP, and BP are at baseline [140], CBF is recovered or decreased [140], and CBF autoregulation is still impaired [141]. Insert: In animals, the higher the ICP rise at SAH and the lower the 60-min CBF recovery, the smaller the changes of 24 h survival [142]

Full size image

Cerebral Perfusion Pressure

CPP falls profoundly during and immediately after SAH. Animal studies indicate that CPP fall is not sufficient to cause perfusion arrest [17]. Similarly, in SAH patients, CPP reductions are not clearly associated with poor neurological outcome. Consequently, decreased CPP may contribute to early ischemic brain injury but is not solely responsible for it [17, 22].

Cerebral Blood Flow

CBF falls after SAH and may or may not recover depending upon the severity of the bleed [17]. In animals, CBF reduction after SAH is accompanied by constriction of large cerebral blood vessels [23]. By contrast, in humans, little arteriographic evidence of acute arterial spasm is found [24, 25]. Therefore, initial fall in CBF in humans is often attributed to the brief period of no-reflow, due to elevated ICP and decreased CPP [26].

CBF Autoregulation

Autoregulatory mechanisms of CBF are frequently impaired after SAH leading to inadequate CBF response to a change in systemic blood pressure (pressure autoregulation) or to a change in partial pressure of carbon dioxide (chemoregulation) [22, 27]. In animals, a severe disturbance in autoregulation occurs within 2–3 h and continues for months after SAH [22]. In patients, this impairment is most pronounced during the first 72 h and correlates well with the SAH severity [27].

Altered Ionic Homeostasis

A rapid alteration in ionic homeostasis occurs after SAH and affects especially sodium, potassium, calcium, and magnesium ions. The alteration evokes immediate effects such as vasoconstriction, an electrical activity disturbance, and slowly developing effects, such as activation and expression of proteins, that develop in a delayed fashion but last for a long time. Both of these effects can be detrimental to the injured brain. The timeline of SAH-derived alteration in ionic homeostasis is presented in Table 2.

Table 2 The timeline of ionic alterations during the first 72 h after SAH
Full size table

Cortical Spreading Depolarization

Cortical Spreading Depolarization (CSD) describes a wave of mass neuronal depolarization associated with net influx of cations and water [28] and is an effect of breakdown of ion homeostasis in the cerebral cortex. CSD is associated with massive neuronal influx of sodium and calcium. Elevated intracellular calcium is possibly the predominant mediator of neuronal death from ischemia [29]. Animal and human studies indicate that CSDs occur early and late after SAH. In human SAH, CSDs can occur as clusters or as isolated events [28]. The cooperative study on brain injury and depolarizations (COSBID group) found that clustered CSDs occurred in spatial and temporal correlation to the development of early and delayed brain damage. Electrocortical and regional cerebral blood flow recordings provided evidence of three different neurovascular responses to CSD in SAH patients, similar to the findings in animals: (1) spreading hyperemia, (2) spreading ischemia, and (3) neurovascular uncoupling [28]. Experimental evidence suggests that subarachnoid oxyhemoglobin, elevated extracellular potassium, decline in NO availability, glutamate, and endothelin-1 are involved in the development of CSD and spreading ischemia after SAH [30, 31].

Decreased Serum Magnesium

Approximately 38% of the patients admitted within 48 h after SAH have abnormally low serum magnesium [32]. In animals, decrease in serum and CSF magnesium occurs within minutes and in humans within hours after SAH [33]. Because magnesium is a physiological antagonist of calcium and controls the NMDA receptor-derived calcium influx, its decrease after SAH contributes to the rise in cellular calcium. In addition, magnesium dilates blood vessels, inhibits aggregation of platelet, inhibits release of excitatory amino-acids, and inhibits synthesis of endothelin-1 (ET-1) [34]. Therefore, decrease in magnesium after SAH exacerbates early brain injury and promotes mechanisms of delayed brain damage.

Mechanical and Biochemical Alterations

The timeline of SAH-derived mechanical and biochemical alterations is presented in Fig. 3. These alterations are described in the next three paragraphs.

Fig. 3
figure 3

The timeline of biochemical alterations during the first 72 h after SAH. Sixty minutes after SAH: Glutamate concentration in cerebral interstitial fluid is increased [23], ventricles swell and brain water content increases [35], and the status of plasma electrolytes at this time is not established. Twenty-four hours after SAH: Glutamate concentration in cerebral interstitial fluid is still increased [143], hydrocephalus [35, 96] and hyponatremia [40, 96] have set in. Seventy-two hours after SAH: Glutamate concentration cerebral interstitial fluid is still increased [143], hydrocephalus [35] and hyponatremia [41] are still present. Insert: SAH patients with large bleeds and poor clinical status at admission are more likely to develop acute hydrocephalus and have poor outcomes [38]

Full size image

Hydrocephalus

Approximately 20% to 30% patients develop acute hydrocephalus within the first 3 days after SAH [35]. In most cases, these patients have larger hemorrhages, poor cerebral perfusion, and reduced CBF at admission [36]. In animals, symptoms of hydrocephalus are present within an hour after initial bleed and are associated with intensity of CBF reduction and cerebral ischemia [37]. The mechanisms of acute hydrocephalus include sudden obstruction of cerebrospinal fluid circulation [35], presence of blood in the ventricles, hemorrhage from a posterior circulation aneurysm, diffused spread of subarachnoid blood, rebleeding, hypertension, and increased sympathetic activity [38].

Increase in Extracellular Glutamate

Cerebral glutamate level increases within minutes after SAH and peaks at approximately 40 min [23]. This biochemical change is associated with the intensity of initial insult [39]. An increased interstitial glutamate concentration after SAH is linked to cellular leakage, altered synaptic transmission, blood–brain barrier opening, and inhibited glutamate uptake [39].

Hyponatremia

Hyponatremia is either present in 10% to 30% of SAH patients at admission or develops within 1–2 days from the initial bleed [40]. Cerebral salt-wasting syndrome and inappropriate secretion of anti-diuretic hormone are implicated in its development. SAH-related hyponatremia is difficult to treat and is associated with the risk of developing cerebral ischemia and infarctions [41].

Vascular Pathology

Cerebral vasculature constricts in response to SAH [42–45]. In animals, constriction of large and small (≤100 μm) parenchymal vessels is visible within minutes after the initial bleed (Table 3) [23, 46]. In humans, vascular imaging is mostly restricted to large vessels; these imaging studies reveal that the large vessels constrict with a delay of 3–7 days after the SAH. However, more recently, Uhl et al. confirmed constriction of small vessels in patients undergoing surgery within the first 72 h after SAH [45]. Thus, it appears that the response of small vessels to SAH in humans (constriction) is the same as in animal models.

Table 3 The timeline of vascular alterations and cell death during the first 72 h after SAH
Full size table

The morphology and function of small vessels is assessed in animals and in autopsy samples from patients who died within the first 72 h after SAH. These studies show corrugation, disruption, and detachment of the endothelium from the basement membrane [7, 43]. Therefore, it is not surprising that therapeutic agents that require a functional endothelium to elicit response are ineffective during the early hours after SAH [47, 48]. Another morphological alteration found in small vessels after SAH is destruction of the basement membrane [49–51]. Although at present destruction of basement membrane is established in experimental studies only, its pathological consequence—increased vessel permeability—is established in animals and in humans [50, 52, 53]. In most cases, an increase in vascular permeability precedes and correlates well with the development of delayed cerebral ischemia and poor clinical outcome [50, 52].

Cell Death

Necrosis, apoptosis, and autophagy cell death pathways activate early in the brain after SAH (Table 3). Cerebral targets of these cell death pathways include brain cells (neurons and glia) and cerebral vasculature (smooth muscle and endothelium) [6, 51, 54]. It appears that more than one cell death pathway is active at any given time after SAH. For example; Dreier and colleagues found necrotic and apoptotic cell death and cerebral infarction in animals 24 h after SAH [55]. Similarly, Lee et al. report neuronal apoptosis in the superficial layers of the fronto-basal cortex and autophagy in deep cortical structures of animals 24 h after SAH [56]. For how long, after SAH, do these cell death pathways remain active is not clear at present. However, animal and human autopsy studies indicate that apoptotic cell death of neurons increases during the first 7 days and then decreases by 11 days after SAH [9, 57]. In SAH animals, the early cell death is associated with neurological deficits [54, 58].

Molecular Alterations

The timeline of SAH-derived molecular alterations is presented in Fig. 4. These molecular alterations are described below.

Fig. 4
figure 4

The timeline of molecular alterations during the first 72h after SAH. Ten minutes after SAH: Cerebral NO level is decreased [59] and platelet aggregates are present in parenchymal vessels [62]. Sixty minutes after SAH: cerebral NO level remains decreased [59], platelet aggregates persist in parenchymal vessels [62], plasma ET-1 level increases [74], oxidative stress is in progress [77, 78], and inflammatory cytokines are expressed [144]. Three hours after SAH: Cerebral NO level is increasing towards recovery [59], platelet aggregates are still present in the cerebral vessels [62], oxidative stress persists [77, 78], and inflammatory cytokines are expressed. Twenty-four hours after SAH: Cerebral NO level increases above basal value [60, 61], platelet aggregation in parenchymal vessels continues [62], plasma ET-1 level remains increased [73], oxidative stress persists [145], expression of inflammatory cytokine persists [144], and their markers appear in serum and CSF [88]. Seventy-four hours after SAH: CSF level of NO [61] and of ET-1 is increased [73], oxidative stress persists [145], CSF inflammatory cytokine level remains increased [88], and blood platelet count remains decreased indicating activation, sequestration/aggregation in the brain [146]. Insert: Antioxidant system activity is decreased and lipid peroxidation products accumulate within 72 h after SAH and correlate well with poor clinical conditions and outcome [79, 145]

Full size image

Nitric Oxide/Nitric Oxide Synthase Pathway

A time-dependent alteration in nitric oxide (NO)/nitric oxide synthase (NOS) pathway occurs during the first 24 h after SAH. In animals, three phases of alteration in cerebral NO are recognized: a decrease within 10 min, return to basal value at 3 h [59], and an increase above basal value at 24 h after SAH [60]. In humans, due to early timing, the first two phases of cerebral NO alteration are not illustrated; the third phase, presenting an increase in NO 24 h after SAH, however, is established [61].

The NO/NOS pathway plays a major role in regulating the cerebral hemodynamic. Therefore, any alteration in this pathway can have pathological consequences. For example, NO regulates CBF and blood pressure by dilating blood vessels and by inhibiting platelet aggregation and leukocyte adherence to the vascular endothelium. After SAH, as cerebral NO level falls, CBF falls, cerebral vessels constrict, platelets aggregate, and neutrophils adhere to the vascular endothelium [59, 62, 63]. Similarly, the pathological rise in cerebral NO, 24 h after SAH, can exacerbate the brain injury. For example, NO as a free radical itself and in the form of peroxynitrite (powerful oxidant) can attack cell membrane; cause damage to the mitochondria, vascular endothelium, and smooth muscle cells [64]l; and activate cell death [65]. The initial fall and later rise in cerebral NO are linked to pathogenesis of delayed vasospasm and poor clinical outcome after SAH [61, 66, 67]. Consequently, alterations in NO occurring during the first 24 h after SAH carry acute, delayed, and prolonged consequences.

Endothelin-1

ET-1 is a potent vasoconstrictor released by astrocytes and leukocytes in response to inflammation and early ischemia after SAH [68, 69]. ET-1 is implicated in early brain injury and in the pathogenesis of delayed vasospasm and delayed ischemic deficits after SAH [70–72]. A number of observations support this concept: (1) ET-1 level increases in serum and plasma within minutes after SAH and expression of its receptors increases 24 to 48 h later [73–75]; (2) ET-1 has the capacity to produce long-lasting constriction [73]; (3) ET-1 level is increased at the time that cerebral NO is reduced (see above) and thus has a perfect unopposed opportunity to elicit a sustained contraction in cerebral vessels after SAH; and (4) ET-1 creates the degenerative morphological changes in the vascular wall similar to those that occur after SAH [76]. Therefore, ET-1 provides yet another mechanism that activates minutes after SAH and has early, delayed, and prolonged consequences.

Oxidative and Nitrosative Stress

Animal and human studies indicate that oxygen-free radicals (ROS) are generated early after SAH and consume enzymatic and non-enzymatic antioxidant defense systems [77–79]. ROS are mostly generated during lipid peroxidation and hemoglobin auto oxidation and induce oxidative stress that contributes to rapidly developing early and more slowly developing delayed ischemic injury after SAH [77, 78]. The mechanisms of ROS-induced brain injury after SAH include: (1) damage of vascular smooth muscle and endothelium, (2) disruption of blood–brain barrier, (3) production of strong spasmogens, and (4) induction of pro-apoptosis enzymes [6, 77]. Accordingly, mice that overexpress superoxide dismutase exhibit ameliorated delayed vasospasm and significantly reduced 24-h mortality [80, 81].

Inflammation

Substantial amount of data supports early activation of inflammatory cascade after SAH. Components especially important in post-SAH inflammation and injury include adhesion molecules, cytokines, leukocytes, and complement. Adhesion molecules (such as vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), and E-selectin) are required for leukocyte migration and represent inflammation. In animals, endothelial expression of adhesion molecules and their serum levels increase within 24 h after SAH and their selective inhibition improves outcome [82, 83]. Similarly, in SAH patients, serum ICAM-1 and VCAM-1 concentration increases at the day of hemorrhage, remains high for 6–8 days, and associates with delayed ischemic injury [84–86].

An early increase in pro-inflammatory cytokines (such as interlukin-6, interleukin-1 receptor antagonist, and tumor necrosis factor-alpha) is also noted in the serum and CSF of SAH patients and relates to early and delayed ischemia and poor outcome [87, 88]. Similarly, systemic complement, another promoter of inflammation, activates within the first 48 h in SAH patients and associates with delayed neurological complications [89]. In animals, early inhibition of complement prevents pathogenesis of delayed vasospasm [90].

Platelets

Platelets activate within minutes after SAH. A reduction in venous jugular platelet count and shape change indicating sequestration and activation is observed 5 min after SAH in animals and 48 h after ictus in SAH patients [91, 92]. Furthermore, platelet aggregates in the lumen of small cerebral vessels are found within 10 min after SAH in animals [62] and within 2 days after SAH in human autopsy studies [8].

The presence of platelets in the small arteries leads to “no-reflow” phenomenon: the absence of vascular filling after a period of global cerebral ischemia [93]. In addition, luminal platelet aggregates activate and promote mechanisms that cause structural injury and functional deficits in small vessels and devastate the already compromised brain. For example, they (1) mechanically obstruct and biochemically constrict (via releasing platelet-derived serotonin, ADP and PDGF) the vessel lumen to promote hypoperfusion [53, 94], (2) injure the vascular endothelium to promote further aggregation [7, 95], and (3) digest the major protein, collagen IV of the vascular basement membrane (via releasing collagenases such as matrix metalloproteinases-2 and 9) to increase vascular permeability and gain access to the brain parenchyma [7]. In brain parenchyma, platelets may activate additional inflammatory mechanisms to further aggravate brain injury after SAH.

Therapeutic Options

Animal studies demonstrate that treatment of early brain injury improves outcome after SAH. Human data supporting these finding, however, are lacking as delayed vasospasm and delayed ischemic injury remain the focus of treatment in clinics and clinical trials. The success of these clinical trials unfortunately has been limited. It is time that a new strategy for treating SAH, aimed at reducing the progression of early activated injurious mechanisms, identified in the section “Early Brain Injury by SAH”, is considered. As explained earlier, these mechanisms activate within minutes after aneurysm rupture and may evolve with time and contribute to poor outcome.

Treatment under this new strategy will begin soon after SAH patient is stabilized and will use pharmacological agents that decrease the progression of the mechanisms identified in the section “Early Brain Injury by SAH”. Pharmacological agents may include a vasodilator (such as NO donors, calcium channel blockers, magnesium or ET-1 antagonists) to prevent further constriction, improve CBF, and reduce the intensity of spreading cerebral ischemia and progressing brain injury. Similarly, an antioxidant, anti-inflammatory, or antiplatelet agent may also be used to reduce ongoing inflammation and oxidation stress. In some cases, inhibition of a single mechanism may not provide substantial protection, and an agent or a combination of agents that inhibit multiple injurious mechanisms may be needed. In such a scenario, nitric oxide donors may be of special interest; they dilate cerebral arteries, recover CBF, and inhibit platelet aggregation. Care, off course, will need to be taken not to exacerbate the brain injury. Consequently, stabilization of the patient and continuous monitoring of vital signs, including ICP, CBF, BP, and heart rate, will be of crucial importance for an early treatment to begin and continue. Nevertheless, given the failure of current therapeutic focus in improving outcome, it is clear that the new strategy aimed at prevention of early brain injury to improve SAH outcome needs be considered. The following section discusses the therapies that are found successful in preventing early brain injury in animals and the clinical trials that have used similar treatments against delayed developing complications (Fig. 5).

Fig. 5
figure 5

Therapeutic strategies against early brain injury after SAH: A battery of compounds working via different pathways has been examined against early brain injury after experimental SAH. Many of them have also been tested against delayed vasospasm and DIC. See text for explanation. ECE endothelin-converting enzyme, PKC protein kinase C, ERK1/2 extracellular signal-regulated kinase, ET-1 endothelin-1, NO nitric oxide, NOS nitric oxide synthase, GSNO S-nitrosoglutathione, SNP sodium nitroprusside, GTN nitroglycerin, eNOS endothelial nitric oxide synthase

Full size image

Calcium Channel Blockers

Blockade of dihydropyridine-type calcium channel is found beneficial against SAH. Nimodipine is the most common agent used for this purpose. Animal studies show that nimodipine used 30 min to 6 h after SAH attenuates constriction and improves cerebral blood supply [96]. In SAH patients, nimodipine reduces the incidence of ischemic complications and the risk of poor outcome. Nimodipine is approved for use in SAH patients in the USA [1]. Current clinical practices call for oral administration within 4 days after SAH ictus for 21 days [97, 98].

The mechanisms underlying the beneficial effects of nimodipine in SAH patients are not clear. However, it is clear that reversal of delayed vasospasm is not one of them, as little reduction in angiographic vasospasm in patients on nimodipine is found [1]. Recovery of CBF and vasodilation, leading to cerebral protection, observed in animals, may explain nimodipine’s benefits, but remains to be established in SAH patients.

Endothelin-1 Antagonism

At least four approaches that block ET-1-mediated constriction of cerebral arteries are studied after SAH. These include: (1) blocking ET-1 biosynthesis [99, 100], (2) reducing extracellular ET-1 levels [101], (3) blocking ET-1 receptors [102, 103], and (4) inhibiting upregulation of endothelin receptors [104]. ET-1 receptor blockade has provided the most promising results. In animal studies, ET-1 receptor antagonists recover CBF when used 60 and 120 min after SAH [102]. In clinical trials, ET antagonist, Clazosentan, prevents vasospasm but does not improve the quality of life, supporting dissociation between the two measures [103]. As ET-1-mediated constriction contributes to brain injury beginning minutes after SAH, perhaps, a treatment strategy that prevents this contribution is warranted to maximize the benefits, improved quality of life, of ET-1 antagonism.

Magnesium Sulfate

A number of investigators have studied the effect of increasing cerebral magnesium against brain injury after SAH. Animal studies find that magnesium pretreatment decreases the duration of ischemic depolarization and reduces ischemic brain lesions upon acute SAH [105]. Clinical studies have so far examined the safety of magnesium treatment within the first 72 h after SAH. These small pilot studies report that continuous intravenous infusion of magnesium to obtain serum magnesium levels of 1.6–2.3 mmol/L or a rise of CSF magnesium level to 11% to 21% for 10 or 14 days is well tolerated [106]. Encouraged by the results of pilot studies, a large randomized, placebo-controlled, double-blind, multicenter phase III clinical trial (IMASH) was conducted [107]. The results could not confirm clinical benefits of intravenous magnesium infusion over placebo in SAH patients [108]. This failure may have resulted from the low CSF penetration of peripherally infused magnesium or a requirement of an even earlier administration to protect brain against injury.

Antioxidants

Antioxidants successfully prevent oxidative stress and decrease early brain injury in animals after SAH [109, 110]. However, clinical studies with the focus on delayed brain injury have not found these compounds effective [111, 112].

Methylprednisolone (a synthetic glucocorticoid) and tirilazad mesylate (a 21-aminosteroid) are the most studied antioxidants. In animals, methylprednisolone used early (immediately or 30 min) after SAH attenuates CBF reduction and a rise in cerebral resistance [113]. In addition, it prevents vasoactive prostanoid and ecosinoid release [109], reduces lipid peroxidation, and preserves an antioxidant enzyme system [114]. In a recent clinical study, methylprednisolone used within 24 to 48 h after SAH for 3 days improved 1-year functional outcome [111]. This study supports the idea that treating early brain injury after SAH improves outcome.

Similarly, tirilazad, when used in animals within 3 h after SAH prevents CBF and CPP changes [110], protects microvascular endothelium and blood–brain barrier [115]. In clinical trials, tirilazad therapy that began within 34–48 h after SAH and continued for 10 days showed improved outcome and decreased mortality in poor grade (grades IV and V) male patients only [116]. However, these results could not be reproduced and a meta-analysis that included five randomized placebo-controlled trials found no evidence that tirilazad reduces the risk of death or disability after SAH [112].

Free radical scavengers—ebselen and edaravone—are also examined against vasospasm in SAH patients. Little preclinical data exist on the efficacy of ebselen during the early phase of SAH. Experimental data on edaravone (MCI-186) show that its use within 24 h after SAH decreases lipid peroxidation (decreases malondialdehyde and increases SOD activity), decreases caspase-3 activation, increases 48 h survival, and improves neurological outcome [117]. In humans, ebselen and edaravone treatment beginning 4 days after SAH is associated with a trend towards lower incidence of vasospasm, cerebral infarction, and delayed ischemic injury [118].

Antiplatelet Agents

Surprisingly, only little animal and more clinical data are available on the use of antiplatelet agents after SAH. Acetyl salicylic acid (aspirin) and ticlopidine are the most common antiplatelet agents studied after SAH. In vitro and in vivo experimental studies demonstrate that Aspirin prevents prolonged vasoconstriction produced by blood [119]. Meta-analysis of the published data shows a trend towards better outcome in patients treated with antiplatelet agents compared with patients who received no antiplatelet agent [120]. In a small study, ticlopidine, used after cisternal drainage, reduced platelet aggregation and improved functional outcome [121]. At least two studies have examined the outcome in patients who used aspirin prior to or soon after SAH. The results are contradictory. One study found increased hemorrhage size and poor outcome in women who consumed more than 15 adult aspirin tablets per month prior to SAH [122]; the other study found a significantly reduced risk of cerebral infarction in patients with significant urinary salicylate level 48 h after SAH [123]. Clearly, a larger study is needed to ascertain the effect of dose and timing of aspirin intake on the outcome after SAH.

Nitric Oxide

The effect of increasing NO bioavailability or prolonging the duration of NO-mediated mechanisms against early brain injury is examined. Methods used to increase NO bioavailability include intracarotid infusion of NO-saturated saline [124], administration of an NO donor [23, 124–127], and increase eNOS expression and/or activity [128]. S-nitrosoglutathione and nitroglycerin, NONOate, glyceroltrinitrate, and diazeniumdiolate are some of the NO donors studied after SAH [23, 125, 126]. Studies with NO donor find that its early use recovers CBF, dilates large and small cerebral vessels, and prevents excitotoxic glutamate release after SAH [23, 46]. Approaches used to increase eNOS expression and activity such as pre-SAH statin use [129] have also produced beneficial effects—reduction in the intensity of arterial spasm 2 days after SAH [129]. Similarly, patients who were taking statin before SAH exhibit decrease risk of symptomatic vasospasm and significantly lower incidence of cerebral infarctions [130, 131].

Another approach that prolongs NO-mediated mechanisms is inhibition of cyclic guanosine 3′,5′-monophosphate (cGMP) degradation by phosphodiesterase [132]. cGMP is a mediator of many of NO-induced effects. In animals, inhibition of phosphodiesterase activity improves NO-mediated vasodilatation [132]. In SAH patients, papaverine, a non-specific phosphodiesterase inhibitor, is routinely used to dilate constricted arteries during aneurysm surgery and for the treatment of cerebral vasospasm [1]. The short half-life and risks associated with this agent, however, outweigh its benefits [133].

Overall, it appears that increasing NO level at the vascular bed is beneficial after SAH, and the sooner this is done the better the outcome is. These experimental findings remain to be duplicated in a clinical setting.

Statins—Acute Treatment

Statins are hydroxymethylglutaryl coenzyme A reductase inhibitors and potent inhibitors of cholesterol synthesis. In addition, statins inhibit platelet aggregation, reduce excitotoxic effects of glutamate, prevent endothelial and neuronal apoptosis, reduce inflammation, enhance angiogenesis, and upregulate and activate endothelial nitric oxidase synthase and nitric oxide production.

The effect of acute (within minutes from ictus for animals and within 3 days for humans) statin treatment against early and delayed brain injury after SAH has been examined. In animals, simvastatin injected at 30 min and 24 and 48 h after SAH reduced perivascular granulocyte migration and basilar artery vasospasm [134]. Animal studies also show that the protective effects of statin are rapidly lost upon its withdrawal [135]. In contrast to animals, some clinical studies do [136] and others do not [137] find acute statin treatment beneficial in SAH patients. In most of these studies, therapy began within 72 h after SAH and continued for 14 days, and delayed outcomes such as vasospasm and delayed ischemic injury were evaluated. Recently, two separate groups conducted a meta-analysis on the effect of statin on SAH outcome [138, 139]. The results are inconsistent. One study found that post-SAH statin use decreases the overall incidences of delayed vasospasm, delayed ischemic injury, and mortality [139] , while the other found that this treatment does not improve neurological outcomes [138]. Consequently, at present, the effect of post-SAH statin use on outcome remains unresolved. Perhaps a large randomized trial that studies the effect on early brain injury in addition to the delayed complications as an outcome is warranted before a decision on acute statin use after SAH can be made.

Summary

Therapies against SAH are designed to treat vasospasm with the ultimate goal of preventing delayed ischemic injury and improving outcome. The success of these therapies in reducing incidence of delayed vasospasm without reduction in delayed ischemic injury and improved quality of life indicates that treating vasospasm alone may not achieve this goal. The results of animal and human studies indicate that mechanisms leading to brain injury activate minutes after SAH and may contribute to the pathogenesis of delayed ischemic injury. Therefore, a therapy that is directed towards inhibiting early brain injury may prove more beneficial in preventing delayed ischemic injury and improving quality of life in this setting.