Neurorepair strategies
In addition to neurological damage, acute brain injury induces a series of neurorestorative events [
58]. In some cases, the central nervous system is able to remodel itself following insults that impair tissue homeostasis. Neurorestorative events include neurogenesis, gliogenesis, angiogenesis, synaptic plasticity and axonal sprouting. These processes are stimulated by endogenous growth-related factors and may continue for weeks to months, facilitating functional and structural recovery. Unfortunately, these restorative processes are largely ineffective for the severity of damage usually encountered in TBI or stroke. Accordingly, providing such injured tissue with a milieu that enhances neuroregenerative processes has become an important therapeutic target.
Mesenchymal stromal cells
Infusion of mesenchymal stromal cells (MSCs) can improve structural and functional outcomes in different brain injury models [
59]. MSCs secrete growth and neurotrophic factors or induce their production by resident brain cells, including microglia. The interaction between MSCs and inflammatory microenvironments is crucial. MSCs can reprogram the local microenvironment from a detrimental function to a beneficial role, reducing toxic events and promoting endogenous restorative processes [
60,
61].
The protection induced by MSCs is extremely variable across studies. In addition to methodological differences between injury models and laboratories, heterogeneity in MSC populations also contributes to disparate outcomes. MSCs obtained from different sources (for example, bone marrow versus umbilical cord blood versus amniotic fluid) are characterized by different protective potency. Furthermore, MSCs obtained from similar sources but from different donors can display different effects [
62]. To predict
in vivo potency, additional experimental work is needed to identify the specific properties of each MSC subpopulation, and to understand determinants of intrinsic heterogeneity. Before translating MSC-based therapies to the clinical environment, safety and consistency also need to be confirmed.
Remote ischemic conditioning
Brief repeated cycles of peripheral vascular occlusion and de-occlusion in dogs prior to induction of coronary ischemia reduce myocardial infarct size [
63]. This remote ischemic preconditioning is presumed to induce humoral factors that prevent reperfusion injury in several organs, including the brain. Protection occurs through modification of intracellular kinase activity, mitochondrial permeability and the inflammatory response to reperfusion [
64]. One potential means by which ischemic conditioning can be achieved is by application of a standard blood pressure cuff to the arm and alternating 5-minute cycles of inflation and release [
65].
Ischemic conditioning may be applied before (preconditioning), during (perconditioning) or after (postconditioning) a cerebral ischemic event. In patients, perconditioning as an adjunct to treatment with intravenous alteplase was associated with a reduction in tissue risk of infarction after acute thrombotic stroke [
66]. Preconditioning has also been associated with prevention of recurrent stroke in patients with intracranial arterial stenosis [
67]. Nevertheless, several questions remain unanswered: would remote ischemic conditioning reduce infarct size if administered before thrombolytic reperfusion of acute thrombotic stroke; can repeated (daily for weeks to months) ischemic conditioning improve long-term outcomes after cerebral ischemia; and in what other settings could ischemic conditioning reduce reperfusion injury and produce better clinical outcomes?
Volatile anesthetic agents for neuroprotection
Volatile anesthetic agents may have neuroprotective properties. Pretreatment with isoflurane improved long-term neurological outcomes after experimental hypoxic/ischemic bran injury or focal brain ischemia [
68] and post-treatment provided neuroprotection in rats [
69]. The inducible form of nitric oxide synthase may mediate the tolerance to ischemia. Other factors that could be involved are the inhibition of excitatory neurotransmission and regulation of intracellular calcium responses during ischemia. Although attractive for their potential benefit on ischemic damage, these experiments have not been translated into clinical studies because use of a volatile agent may also induce vasodilatation, increasing CBF and, consequently, elevations in ICP.
In SAH patients, however, in whom an increase in CBF may be beneficial, the availability of a pragmatic bedside dispensing device and continuous ICP and regional CBF monitoring has made the inhalation of isoflurane for neuroprotection in the ICU practical, and this approach has been assessed in a small pilot study [
70]. Nevertheless, at this stage, the use of volatile agents remains unproven and not ready for clinical implementation [
71].
The brain can use alternative substrates beyond glucose, including lactate, pyruvate and ketone bodies, particularly in conditions of increased energy demand and limited glucose availability (for example, exercise, starvation, hypoglycemia or hypoxia/ischemia). Preferential use of lactate over glucose has been demonstrated in healthy human subjects [
72] and diabetic patients [
73]. Sodium lactate infusion has been shown to be neuroprotective in several brain injury models, both
in vitro and
in vivo [
74-
77]. Hypertonic solutions (containing sodium chloride or sodium lactate) may prevent brain edema and elevations of ICP following TBI [
78,
79]. Solutions containing lactate have favorable cerebral metabolic effects in patients with TBI [
78]. Acetyl-L-carnitine infusion may also exert protective effects on the injured cells [
80].
Overall, these data support the hypothesis that supplementation of the injured human brain with alternate energy fuels may be beneficial after acute brain injury [
81]. Although a large trial of prehospital hypertonic saline infusions did not show beneficial effects on outcome [
26], administration of isotonic or hypertonic solutions containing lactate, pyruvate or ketones may be a better option for the treatment of brain edema and cerebral ischemia following acute brain injury by enhancing brain energetics and, in turn, neurological recovery.
Sex hormones
In numerous models, early parenteral administration of sex hormones has anti-apoptotic, anti-inflammatory and anti-oxidant properties and can accelerate reparative processes that prevent long-term sequelae [
82]. One potential long-term reparative action of estrogens relates to their effects on sonic hedgehog, a signaling protein that controls and directs differentiation of neural stem cells, thus influencing brain repair by generating new neurons whenever necessary. Estrogen-induced acceleration of sonic hedgehog production may represent a potential pathway for neuroregeneration and neuroprotection [
83]. Laboratory data have shown that the relevant effects of sex hormones in neurological emergencies, such as TBI, stroke and spinal cord injury, are neither cell-type specific nor insult specific [
84,
85].
Several clinical trials of sex hormones in TBI have recently been completed. The phase II RESCUE–TBI study (Clinical Trials NCT00973674) evaluated the safety and feasibility of administering a single dose of intravenous conjugated estrogens (Premarin; Pfizer Inc., New York, NY, USA) in patients with severe TBI, but no results are yet available. The National Institutes of Health-sponsored ProTECT III study evaluated the effects of intravenous progesterone (started within 4 hours of injury and given for a total of 96 hours) versus placebo in patients with moderate to severe TBI, but was stopped for futility [
86]. The SyNAPSe Trial, comparing intravenous progesterone with placebo within 8 hours of severe TBI for a total of 120 hours, completed enrollment but failed to demonstrate a benefit [
87]. There is therefore currently no clinical evidence to support sex hormone usage in TBI.
Hyperoxia in neuroprotection
Oxygen is an essential substrate for the brain; however, the safety margin between effective and toxic oxygen doses is relatively narrow. Oxygen may be toxic to the lungs (for example, tracheobronchitis, absorption atelectasis, hypoxic pulmonary vasoconstriction and hyperoxia-induced lung injury), the circulation and the brain tissue itself (seizures or lipid peroxidation) [
88], and hyperoxia has been associated with increased mortality in patients with various acute neurological disease processes [
24,
89-
91]. However, hyperoxia can increase PbtO
2, restore mitochondrial redox potential, decrease ICP, restore aerobic metabolism and improve pressure autoregulation [
92-
94]. Administering 100% oxygen at normal atmospheric pressure (normobaric hyperoxia) is inexpensive, widely available and can be started promptly after TBI or stroke (for example, by paramedics). Results in humans, however, have been mixed [
24,
93,
95-
97]. There is probably a narrow effective dose, and benefit may be limited to at-risk tissue. Moreover, treatment may only be effective in specific subgroups of patients or may depend on the metabolic state [
97]. Furthermore, whether improvements in brain metabolism translate into better outcome is unclear [
96].
Hyperbaric oxygen has been shown to reduce infarct volume, blood–brain barrier disruption, edema and neurologic deficits in animal models of ischemic brain injury [
98]. In experimental TBI, hyperbaric oxygen decreased neuron injury and edema [
99]. In a small group of severe TBI patients, hyperbaric oxygen improved brain metabolism and decreased ICP [
100]. However, administering hyperbaric oxygen can be clinically challenging, requiring that patients are moved out of the ICU to the hyperbaric oxygen chamber, an expensive facility available in only a few centers. In a small study of 42 TBI patients, the combination of hyperbaric and normobaric hyperoxia was associated with significant outcome benefits compared with standard therapy [
101]. These results need confirmation in larger studies.
PbtO
2 monitoring is often used to optimize oxygenation targets and evaluate the utility of therapeutic interventions. Decreases in PbtO
2 are associated with chemical markers of brain injury and with both mortality and unfavorable outcome after TBI [
102], with similar (but less robust) associations in SAH. Addition of PbtO
2-based care to conventional ICP-based and CPP-based care has been associated with improved outcomes after severe TBI [
103,
104]. This issue has been evaluated in a multicenter phase II clinical trial (Clinical Trials NCT00974259). Preliminary results demonstrate the feasibility and safety of PbtO
2-based care and suggest a benefit to outcome, but phase III trials are necessary to confirm these findings.