Hemoglobin and hemoglobin-derived products are the most important DAMPs released by ruptured red blood cells (RBCs) in the subarachnoid space and are involved in the inflammatory activation of SAH (Bozza and Jeney, 2020). Hemoglobin and its products, such as methemoglobin, heme, heme chloride, and oxygenated hemoglobin, bind to the TLR-4 receptor as DAMPs. Methemoglobin is water soluble, which can lead to extensive TLR-4 activation with cerebrospinal fluid (CSF) circulation. In addition, heme stimulates the formation of more neutrophil extracellular traps (NETs) while activating TLR-4.

Hemoglobin consists of four globin chains tightly bound to the heme group. Oxyhemoglobin and its metabolites are considered to be the main sources of reactive oxygen species (ROS) in the pathophysiological process of SAH (Hugelshofer et al., 2018). After hemolysis, tetrameric hemoglobin is released from RBCs and degrades gradually, producing toxic intermediates (Stokum et al., 2021). Among them, heme is considered more toxic than hemoglobin (Bulters et al., 2018). In the ferrous (Fe²⁺) and trivalent (Fe³⁺) states, heme can react with hydrogen peroxide to generate hydroxyl radicals through the Fenton reaction, which can damage lipid membranes and lead to the production of lipid ROS, cell dysfunction and even ferroptosis (Marnett et al., 2003; Yang and Stockwell, 2016).

These hemoglobin-derived products induce toxicity by producing ROS that can cause oxidation of cell lipids, proteins and DNA, leading to programmed cell death (Sinha et al., 2013). ROS can further activate TLR/NF-κB/MAPK, KEAP1–NRF2–ARE, eicosanoid and other signaling pathways as well as NLRP3 inflammasomes to mediate inflammatory responses (Reczek and Chandel, 2015). Oxyhemoglobin has been shown to induce cerebral artery contraction after SAH by inhibiting voltage-dependent K⁺ channels in cerebral arteries and inducing R-type Ca²⁺ channel expression in cerebral arteries (Ishiguro et al., 2006; Link et al., 2008).

As an important endogenous vasodilator, NO can be produced by endothelial cells, neurons and microglia (Bulters et al., 2018). After SAH, peroxynitrite can be produced by the reaction of NO with superoxide radicals, a highly oxidative species. These peroxides damage the neurons that produce NO and reduce NO production. More importantly, hemoglobin released by subarachnoid blood inhibits the activity of endothelial NO synthase (eNOS), exacerbating the decrease in NO production (Sabri et al., 2011; Lenz et al., 2021). As a result, the availability of NO to vascular smooth muscle cells is reduced, leading to vasoconstriction (Li Q. et al., 2016, Li J. et al., 2016). In addition to regulating cerebrovascular tension, NO inhibits the formation of thrombocytopenic microthrombosis (Voetsch et al., 2004). As the result of the combination of these factors, cerebral perfusion is significantly reduced leading to neuronal dysfunction (Takeuchi et al., 2006). Reduced bioavailability of NO reduces the cortical diffusion depolarization threshold (Petzold et al., 2008). This leads to diffuse ischemia and neuronal death (Terpolilli et al., 2016).

Many studies have shown that cerebral edema in SAH is a biphasic phenomenon (Serrone et al., 2015). The formation of early cerebral edema is a direct result of early ischemic injury during initial bleeding, while subsequent delayed edema appears to be caused by BBB dysfunction (Hayman et al., 2017). There is evidence to show that hemoglobin and its metabolites can cause brain edema. In a model of intracranial hemorrhage (ICH), the iron deposition around the hematoma gradually increases after injection of autologous blood into the brain parenchyma of rats. Meanwhile, the water content in the brain tissue around the hematoma also gradually increases (Huang et al., 2002). However, the degree of cerebral edema in rats is significantly reduced by chelating agents. Immunohistochemical analysis shows that aquaporin (AQP)4 is highly expressed in astrocytes. Therefore, iron overload and AQP4 may play a key role in the formation of hemoglobin-mediated brain edema after ICH (Qing et al., 2009). Hemoglobin-induced oxidative stress can increase expression of matrix metalloproteinase (MMP)-9 and lead to BBB dysfunction (Katsu et al., 2010; Ding et al., 2014). In view of this, delayed edema after SAH is thought to be caused at least in part by hemoglobin and its breakdown products (Urday et al., 2015).

Microglia act as the resident macrophages of the CNS and are an important component of innate and adaptive immune responses. TLR-mediated microglial activation leads to production of several inflammatory mediators that rapidly respond to different stimuli, such as DAMPs (Atangana et al., 2017). TLR4 is most abundantly expressed on microglial membranes (Lehnardt, 2010; Li et al., 2021). TLR4 and other PRRs lead to the activation of downstream inflammatory signaling cascades, including the NF-κB, MyD88/TRIF, and MAPK pathways (Geraghty et al., 2019). However, at different stages in SAH, microglia exhibit different phenotypes: the proinflammatory M1 phenotype predominates in the early stages and this is gradually replaced by the anti-inflammatory M2 phenotype as the disease progresses (Zheng et al., 2020). Between them, M1 has the ability to release proinflammatory cytokines, such as TNF-α and IL-6 (Hu et al., 2015). In animal models, increased expression of proinflammatory cytokines is associated with poor prognosis of SAH (Kooijman et al., 2014; Wang et al., 2021). Neuronal cell death and microglial cell accumulation follow similar time courses. Thus, microglial accumulation can cause secondary brain damage after SAH (Schneider et al., 2015). Therefore, early post-SAH promotion of activation of these microglia toward an anti-inflammatory phenotype may have a neuroprotective effect (Schneider et al., 2018).

Astrocytes are the most abundant glial cells in the CNS. They play an important role in maintaining the integrity of the BBB and supporting the activity of neurons (Cunningham et al., 2019). Astrocytes can differentiate into different phenotypes under different stimuli, namely proinflammatory type A1 and anti-inflammatory type A2 (Shinozaki et al., 2017; Hyvärinen et al., 2019). DAMP-mediated activation of TLR contributes to the acquisition of A1 phenotype in astrocytes (Xu et al., 2018). TLR expression is low in the astrocytes of healthy individuals. However, TLRs, especially TLR-4, are abundant on the surface of the astrocyte membrane in the event of injury or inflammation (Azam et al., 2019). After SAH, various DAMPs are released into the subarachnoid space, promoting the activation of proinflammatory phenotype through TLRs (Ghaemi et al., 2018). One study found that changes in astrocyte Ca²⁺ signaling after SAH led to a neurovascular coupling response that shifts blood vessels from a diastolic to a constrictive state, and ultimately exacerbates brain tissue damage (MacVicar and Newman, 2015; Pappas et al., 2015). At the same time, the dysfunction of glutamate uptake mediated by astrocytes may be the possible mechanism of DCI after SAH (Tao C. et al., 2020, Tao K. et al., 2020). Additionally, A1 astrocytes cause depression-like behavior and cognitive dysfunction in mice (Zhang et al., 2020). Current interventions targeting type A1 astrocytes in animal models reduce neuronal death as well as neurogenesis decline and cognitive impairment (Lu et al., 2020; Zhang et al., 2021). Therefore, therapies targeting astrocytes may help improve outcomes in patients with SAH.

Neutrophils are the most abundant type of white blood cells in peripheral blood, and they are significantly increased after SAH (Gris et al., 2019; Morga et al., 2020). Activation of astrocytes challenges the integrity of the BBB and vascular permeability is increased. Many chemokines are produced in the local inflammatory response after SAH. Under the combined action of the two, numerous neutrophils enter the subarachnoid space from the peripheral circulation (Coulibaly et al., 2020; Osuka et al., 2021). Related observational studies have found that increased neutrophil-to-lymphocyte ratio is inversely associated with prognosis in patients with SAH (Giede-Jeppe et al., 2019). Neutrophils have also been found to mediate early cortical hypoperfusion in animal models (Neulen et al., 2019). On the one hand, neutrophils can release IL-6, transforming growth factor-β1 and other inflammatory cytokines to produce a cascade reaction, which plays an important role in post-SAH inflammation (Takizawa et al., 2001). On the other hand, these neutrophils can produce peroxide by NADPH oxidase (NOX) and myeloperoxidase (MPO), causing damage to neurons and other support cells in the CNS and even apoptosis and cerebral cortex insufficiency (Chu et al., 2015; Neulen et al., 2019). Recent studies have shown that the presence of NETs released by neutrophils can transform microglia into a more proinflammatory phenotype, thereby aggravating neuroinflammation after SAH and leading to poor prognosis (Hanhai et al., 2021).

Monocytes are innate immune cells produced mainly in the bone marrow. When they are released into the circulation, they quickly differentiate into macrophages and perform different functions. Within 24 h after SAH, there is a large number of monocytes infiltrating into the site of hemorrhage and gradually increasing (Jedrzejowska-Szypułka et al., 2010; Gris et al., 2019). Monocytes mediate cerebral vasospasm after SAH in animal models, which may be the mechanism related to DCI and DIND (Jackson et al., 2021). Once monocytes infiltrate brain tissue, they mature into macrophages and take on different morphological and biochemical characteristics (Coulibaly et al., 2020). Meanwhile, peripheral circulating macrophages are recruited by monocyte chemotactic protein-1 to enter the site of injury (Lu et al., 2009; Niwa et al., 2016). Similar to microglia, monocytes and macrophages are likely to be involved in the inflammatory response or oxidative damage after SAH and lead to poor prognosis (Kwan et al., 2019; Unda et al., 2020). Therefore, further research on its mechanism will help to develop more effective treatment regimens.

The toxic effect of immune cells on the CNS may be the result of combined action of cytokine-mediated inflammatory response and ROS-mediated oxidative stress.

Surgery is the only way to completely remove the blood clot. Especially for SAH patients with large hematomas or ruptures in the cerebroventricular system, surgery can effectively remove the hematoma and reduce the occurrence of vasospasm (Zhang et al., 2001). However, due to the anatomical structure of SAH, blood can be widely distributed in various parts of the subarachnoid space. Therefore, for patients with small hematomas and mild symptoms, surgical treatment is no longer appropriate.

For patients in whom craniotomy for removal of hematoma is not recommended, continuous drainage of CSF by external ventricular drainage and lumbar cistern drainage can be performed to reduce the hemoglobin concentration in CSF (Wolf, 2015). Continuous CSF drainage can significantly reduce the incidence of vasospasm (Klimo et al., 2004; Alcalá-Cerra et al., 2016). However, CSF drainage is of limited use and it does not clear the blood clot that has formed. Blood clots can still release hemoglobin, which can damage brain tissue. Therefore, CSF drainage results in a small improvement in long-term outcomes after SAH (Al-Tamimi et al., 2012).

In recent years, CSF drainage combined with intrathecal drug therapy has received attention. A randomized controlled trial in Japan used magnesium sulfate solution for intrathecal irrigation. Although vasospasm was significantly improved, the incidence of DCI and functional outcomes in patients with SAH were not significantly improved (Yamamoto et al., 2016). The combined intrathecal use of thrombolytic agents has not yielded satisfactory results (Etminan et al., 2013). It has also been found that intrathecal use of thrombolytic agents increases inflammation. However, this may be due to the release of hematoma breakdown products rather than thrombolytic drugs (Kramer et al., 2015).

Hemoglobin can be engulfed by cells in three ways: erythrophagocytosis, endocytosis of erythrocytes mediated by haptoglobin, and endocytosis of heme mediated by hemopexin (Figure 7; Bulters et al., 2018; Pan et al., 2020). These mechanisms of endogenous hemoglobin clearance provide an entry point for therapeutic regimens that target hemoglobin.

Steroid hormones are effective in combating inflammation and inhibiting the production of proinflammatory cytokines (Whitehouse, 2011). A retrospective study found that dexamethasone treatment appeared to reduce the risk of adverse outcomes after SAH (Czorlich et al., 2017). However, another study showed that dexamethasone use reduced poor outcomes, but not for DCI (Mohney et al., 2018). This may be because steroid hormones help maintain the water and electrolyte balance and reduce brain edema (Mistry et al., 2016).

Statins, or 3-hydroxy-3-methylglutaryl-CoA inhibitors, can reduce total cholesterol and low-density lipoprotein (LDL), as well as triglyceride. They also increase high-density lipoprotein. Simvastatin has been shown to reduce the occurrence of DCI in animal models (Sugawara et al., 2008). This may be because statins exert a neuroprotective effect through a cholesterol-dependent mechanism (Tseng et al., 2007). However, a retrospective trial found that atorvastatin reduced cerebral vasospasm and infarction after SAH, but had little effect on long-term prognosis (Chen S. et al., 2020). Therefore, more evidence is needed to support the need for statins in patients with SAH (Bohara et al., 2021).

As mentioned above, COX is important in the pathophysiological process of SAH. Aspirin, as a nonselective COX inhibitor, inhibits the synthesis of TXA2 and prostacyclin simultaneously. Thus, aspirin inhibits the formation of microthrombosis and reduces the inflammatory response in SAH (Muroi et al., 2014; Darkwah Oppong et al., 2019). However, there are considerable differences of opinion regarding the use of NSAIDs. Although feasible in theory, they seem to have little effect in clinical practice (Young et al., 2012).

Minocycline, a tetracycline antibiotic, has recently been found to prevent inflammation and p53-related apoptosis induced by NLRP3 inflammasomes (Li Q. et al., 2016, Li J. et al., 2016). In addition, fluoxetine, which has been shown to have anti-inflammatory effects in many diseases, attenuates NLRP3 inflammasome and caspase-1 activation through autophagy activation in animal models (Li et al., 2017). Melatonin has also been found to reduce the inflammatory response and be beneficial for early post-SAH brain injury (Dong et al., 2016).

Nuclear factor-erythroid 2 p45-related factor 2 is a transcription factor that recognizes antioxidant response elements (AREs) to regulate the expression of a variety of genes (Figure 8; Shao et al., 2020). Nrf2 binds to the Kelch-like ECH-associated protein 1 (Keap1) in the cytoplasm through its binding domain. Therefore, the function of Nrf2 is regulated by Keap1 (Itoh et al., 1999). The Nrf2 system is widely expressed in CNS and is usually upregulated in response to inflammation and brain injury (Sandberg et al., 2014).

In animal experiments, Nrf2 expression is upregulated in cerebral arteries of rats after experimental SAH (Wang et al., 2010). It attenuates early brain injury such as cerebral edema, BBB injury and cortical cell apoptosis through the Nrf2–ARE pathway (Chen et al., 2011). Nrf2 can also reduce the occurrence of cerebral vasospasm (Zhao et al., 2016). After the Nrf2 gene is knocked out, brain injury of SAH rats is aggravated, including increased cerebral edema, BBB destruction, apoptosis of nerve cells, and severe neurological impairment (Li et al., 2014). Thus, existing studies have demonstrated that Nrf2 plays an important role in alleviating secondary complications induced by SAH.

Currently, sulforaphane, curcumin, astaxanthin, lycopene, melatonin, erythropoietin, and other Nrf2 system activators are available. All of this works by binding to Keap1 to release Nrf2. Released Nrf2 translocation into the nucleus leads to increased transcription (Zolnourian et al., 2019). Currently, there is growing clinical interest in the use of Nrf2 activators in the treatment of SAH (Guo et al., 2018; Zolnourian et al., 2020). We may use this mechanism to treat patients with SAH in the future.

Superoxide dismutase, glutathione peroxidase (GPX), and catalase are important peroxidase scavengers in the CNS (Figures 9, 10; Lewén et al., 2000). However, the antioxidant capacity of these enzyme systems is reduced after SAH (Marzatico et al., 1993). This leads to a negative effect on the antioxidant stress after SAH. For example, decreased concentrations of SOD in plasma and CSF have been shown to be associated with long-term poor outcomes after SAH (Krenzlin et al., 2021). In addition, gene transfer of SOD was found to reduce cerebral vasospasm after experimental SAH (Watanabe et al., 2003). So far, no clinical treatment associated with exogenous antioxidant enzymes has been reported. In the future, exogenous supplementation of these enzymes will possibly reduce the level of oxidative stress in SAH patients, reduce the damage of peroxides to the CNS, and improve prognosis.