REVIEW

Recent update on basic mechanisms of spinal cord injury

Syed A. Quadri1,2 & Mudassir Farooqui3 & Asad Ikram3& Atif Zafar3 & Muhammad Adnan Khan1,2

& Sajid S. Suriya1,2 &

Chad F. Claus4 & Brian Fiani5 & Mohammed Rahman6& Anirudh Ramachandran7

& Ian I. T. Armstrong1,2&

Muhammad A. Taqi1,2 & Martin M. Mortazavi1,2

Received: 8 May 2018 /Revised: 20 June 2018 /Accepted: 6 July 2018# Springer-Verlag GmbH Germany, part of Springer Nature 2018

AbstractSpinal cord injury (SCI) is a life-shattering neurological condition that affects between 250,000 and 500,000 individuals eachyear with an estimated two to three million people worldwide living with an SCI-related disability. The incidence in the USA andCanada is more than that in other countries with motor vehicle accidents being the most common cause, while violence beingmost common in the developing nations. Its incidence is two- to fivefold higher in males, with a peak in younger adults. Apartfrom the economic burden associated with medical care costs, SCI predominantly affects a younger adult population. Therefore,the psychological impact of adaptation of an average healthy individual as a paraplegic or quadriplegic with bladder, bowel, orsexual dysfunction in their early life can be devastating. People with SCI are two to five times more likely to die prematurely, withworse survival rates in low- and middle-income countries. This devastating disorder has a complex and multifaceted mechanism.Recently, a lot of research has been published on the restoration of locomotor activity and the therapeutic strategies. Therefore, itis imperative for the treating physicians to understand the complex underlying pathophysiological mechanisms of SCI.

Keywords Spinal cord injury .Mechanism of spinal cord injury . Primary phase . Secondary phase . Chronic phase

Introduction

Traumatic spinal cord injury (SCI) is a life-shattering neuro-logical disorder, which affects between 250,000 and 500,000individuals each year [1–4]. Trauma to the spinal cord second-ary to motor vehicle accidents (MVAs), falls, violence, and

sports injuries are the leading causes with road traffic acci-dents being the most common etiology [1, 2, 5–10]. The inci-dence is two- to fivefold higher in males as compared to fe-males, with a peak in younger adults. Studies have reportedannual economic burden of billions of dollars on the federalhealthcare systems [11–13]. Since it predominantly affects theyounger adult population, the psychological impact of adap-tation of a healthy individual to a paraplegic or quadriplegic intheir early life can be devastating [14–17]. Moreover, it isreported that people with an SCI are two to five times morelikely to die prematurely than people without an SCI, withworse survival rates in low- and middle-income countries [1,18–20].

Multiple medical specialties are involved in the manage-ment and treatment of patients with SCI. Such specialties in-clude neurosurgery, trauma surgery, neurology, psychiatry,pain management, and rehabilitation [16]. Recently, extensiveresearch and attention have been focused on the restoration oflocomotor activity and therapeutic strategies for SCI, includ-ing cell transplantation, surgical neurostimulation, and drugadministration. Therefore, it is essential for physicians inter-ested in SCI research to understand the underlying mecha-nisms of injury of this devastating condition [21–25].Though the subject is quite vast, in this article, the authors

* Syed A. Quadridr.saqader@gmail.com

1 California Institute of Neuroscience, 2100 Lynn Road, Suite 120,Thousand Oaks, CA 91360, USA

2 National Skull Base Center, Thousand Oaks, CA, USA3 Department of Neurology, University of New Mexico,

Albuquerque, NM, USA4 Department of Neurosurgery, St. John Providence Hospital and

Medical Centers, Michigan State University, Southfield, MI, USA5 Department of Neurosurgery, Desert Regional Medical Center, Palm

Springs, CA, USA6 Department of Neurology, Desert Regional Medical Center, Palm

Springs, CA, USA7 College of Osteopathic Medicine of the Pacific, Western University

of Health Sciences, Pomona, CA, USA

Neurosurgical Reviewhttps://doi.org/10.1007/s10143-018-1008-3

provide a glance into the complex and multifaceted mecha-nism of SCI.

Different levels and severity of SCI

SCI is characterized by a wide array of symptoms includingparalysis, paresthesia, spasticity, pain, and cardiovascular,bowel, bladder, or sexual dysfunction [26]. It has been shownto cause substantial autonomic dysfunction, with neurogenicshock being one of the leading causes of death followingtraumatic SCI [19, 27–29]. The amount of disability dependson the severity and the level of spinal cord injury.

The International Standards for Neurological Classificationof Spinal Cord Injury (ISNCSCI) developed by the AmericanSpinal Injury Association (ASIA) is used commonly to assessthe severity of injury [30, 31]. The severity of the injury isclassified into different grades in the ASIA Impairment Scale(Table 1).

Various epidemiological studies have demonstrated thatSCI occurs more at the cervical levels, being the most mobilesegment of the spine and limited by two stable segments at itstwo ends being the cranium superiorly and the thorax inferi-orly, although the distribution varies substantially based on thegeographic location of the studies [1, 2, 6].

Mechanism of SCI

It is well established that the pathophysiology of acute SCIoccurs in two stages: the initial, immediate mechanical injurycaused by a permanent or temporary compression [32, 33]

usually causing contusion of the spinal cord, followed by asecondary phase, characterized by destructive and self-propagating biochemical changes in neuronal and glial cellsthat lead to increased dysfunction and eventual cell death overhours to weeks after the initial insult [13, 34–38]. The spatialextent of secondary injury events spread rostrally and caudallyfrom the site of the impact causing structural and functionaldisturbance along the spinal cord [37, 38]. Cell death beginslocally and centrally, ultimately leading to the destruction ofthe central gray matter along with the partial or complete lossof adjacent white matter tracts [39–42]. The primary phase ofthe SCI is usually unexpected and could lead to significantdelay in the management. In contrast, the secondary phase ofSCI is subacute and can be manageable [39]. A proper under-standing of the mechanisms underlying the secondary phaseof SCI may help the physician manage it promptly with ameasure to decrease this secondary reaction and optimize thetreatment, limiting the extent of the spinal cord injury [43].The mechanism of SCI is summarized in Fig. 1.

Primary SCI phase

The spinal cord can be contused or transected. A contu-sion is the most frequent finding [44–46]. Persistent com-pression after the impact triggers a cascade of events lead-ing to disability after the SCI [47–50]. Primary injuryoccurs at the time of the initial impact. An example iscompression from a retropulsed component of a burstfracture [35, 51, 52]. Hallmarks of primary injury arelocal hemorrhage, edema, and ischemia that progressand initiate the secondary phase [53]. The mechanisms

Table 1 The American SpinalInjury Association (ASIA)Impairment Scale

Grade Degree of spinal cord dysfunction

A Full dysfunction

• Complete loss of motor and sensory function below the level of SCI

• Includes S4–S5 loss, leading to perianal numbness (anal area, see Fig. 1)

B Partial dysfunction

• Complete loss of motor function below the level of SCI

• Variable loss of sensory function below the level of SCI, sparing S4–S5 sacral innervation(anal sensations intact, see Fig. 1)

C Partial dysfunction

• Partial: more than 50% loss of motor function below the level of SCI

• Motor movements cannot be made against the gravity of affected muscles (grade 1/5 or 2/5)

D Partial dysfunction

• Partial: less than 50% loss of motor function below the level of SCI

• Motor movements can be made against the gravity in affected muscles (grade 3 or more out of 5)

E Normal neurologic examination

• Complete neuronal recovery

• No motor or sensory loss

Neurosurg Rev

of injury can be compression, distraction, or translation ofthe spinal column [54]. Disruption of local vascular struc-tures at the site of insult often leads to localized hemor-rhages within the spinal cord tissue [55–58]. Since thespinal cord tissues are located within a confined space,framed by vertebral bodies, clotting and edema shift theposition of the neuronal and glial tissues within its con-fined area, increasing local pressures on the tissues, thusenhancing ischemia. All of the blood flow to the injurysite decreases so that local neurons and glial cells aredeprived of oxygen and glucose. The ischemia and mem-brane damage in surviving cells initiate the secondaryphase of spinal cord injury. Edema is common amongspinal cord injuries and is a significant proponent of thedevelopment of secondary injury. The initial traumaticimpact increases the permeability of the blood-spinal cordbarrier and can induce both vasogenic and cytotoxic fac-tors, all of which can cause osmotically active substancesto enter and contribute to the injury edema [39, 40, 59,60]. The initial primary mechanical insult on the spinalcord instantly injures or destroys resident tissue, and thebony or disk fragments cause the laceration and transec-tion of the spinal cord and the surrounding structures. In

theory, the gray matter is irreversibly damaged during thefirst hour after the SCI, but the white matter may survivethe insult up to 72 h after the SCI [61].

Secondary SCI phase

The primary injury causes delayed damage and death to adja-cent cells that survive the original trauma. This biological andself-propagating response to SCI is known as the secondaryinjury phase and is characterized by multiple cascades of bio-chemical events that cause further tissue loss and dysfunction.These cascading events [33, 36, 57, 58, 62] can be dividedinto three distinct yet often continuous sequences: acute, sub-acute, and chronic phases.

Acute phase

In the acute phase, the damage is a direct result of the primarymechanical trauma that, within seconds, disrupts the structuralintegrity of the tissue and immediately leads to physical andbiochemical alterations. These local and systemic events in-clude spinal shock, vascular dysfunction, ischemia, membrane

Fig. 1 The mechanism of spinalcord injury (SCI). Demonstrationof various phases of SCI aftertrauma to the spinal cord

Neurosurg Rev

compromise, ionic dysregulation, and neurotransmitter accu-mulation. Many of these events overlap into the subacutephase.

Spinal shock

Spinal shock is a transient physiological disruption in functionand reflexes as a result of the initial insulting injury as ex-plained in a study on a rat model [63]. Although specificaspects of shock differ according to the site of cord injury,the most shock is characterized by sensory deficits, flaccidparalysis, the absence of deep tendon reflexes, absent reflexsomatic activity, and thermoregulatory dysfunction below thelevel of injury. Testing for spinal reflexes in a rectal exam isone the most reliable methods for lower spinal cord injuries[27, 59, 64]. The bulbocavernosus reflex, a reflex arc betweenthe bulbocavernosus muscle of the penis and the external analsphincter, is the best clinical indicator of whether a patient is inspinal shock [64–67]. Various human studies confirmed thatthe absence of this reflex indicates the spinal shock in thepatient, whereas the presence of this reflex rules out spinalshock and validates the accuracy of the clinical exam findings[65–67]. The neurogenic shock might follow. It is character-ized by severe arterial hypotension, bradycardia, and hypo-thermia [63, 68]. It is a direct consequence of autonomic mal-function caused by the sympathetic disruption, as indicated ina study on the rat model of vasospasm-related SCI [69].Descending sympathetic tracts are lost, leading to unopposedintact parasympathetic tone [28, 39, 40, 59].

Vascular dysfunction

Vascular injury during the acute phase of spinal cord injurycontributes immensely to the progression of secondary injuryas described in various human and animal studies [51, 55]. Asstated prior, traumatic mechanical injury to the spinal cordelicits severe hemorrhages throughout the gray matter, leadingto hemorrhagic necrosis and subsequent central myelomalacia[56, 70]. Holtz et al. explain the reversible spinal cord com-pression injury in a rat model; significant changes and disrup-tion occur in the local microcirculation, mainly at the capillaryand venule level [71]. Coupled with systemic hypotension andlocal hemorrhage, microcirculatory loss causes a major reduc-tion in blood flow at the lesion. This substantial decrease inblood flow leads to tissue ischemia that can become progres-sively worse over the first few hours of insult [71]. The quickprogression of ischemia is somewhat unclear but may be dueto vasospasm or vasoactive amines released during the injury,as studied in a rat model of SCI [69]. In addition to the dis-ruption of the microcirculation, the ischemia and subsequentreperfusion induce pronounced endothelial damage of the in-jured vessels. This endothelial damage is the consequence ofoxygen-derived free radicals and other toxic by-products

released during the early ischemic and reperfusion period.Highly reactive oxygen and nitrogen species contribute tothe oxidative stress involved in the endothelial damage, vas-cular permeability, and edema that are intimately related toprogression and cascading events responsible for secondaryspinal cord injury [39–41, 59, 60].

Membrane and ionic dysregulation

The integrity of the neuronal cell membrane is critical to thefunction and protection of the highly controlled intracellularenvironment. Traumatic insult to the membrane results in dis-ruption of the integrity, leading to the hyperpermeability for theions. Most important is the calcium influx into the intracellularspace, activating proteases, disrupting mitochondrial function,and activating apoptotic pathways. Also, surviving cells sur-rounding the injury may also experience deleterious conse-quences due to adjacent ionic disruption from plasma mem-brane compromise [39, 40]. Ouyang et al. studied a guinea pigmodel of 80% spinal cord compression, either briefly or con-tinuously for 30 min; it is interesting to note the role of myelindisruption and potassium channel dysregulation compression-mediated conduction block, given the apparent elementaryconcept of membrane integrity and the cytoskeletal and struc-tural distinctness between somal membranes and axonal mem-branes [72]. One of the most substantial discrepancies in mem-brane sealing between neurons and other cell types is the timecourse. Neuronal membranes have been shown to require mi-nutes to hours while fibroblast requires seconds to minutes toproperly seal [73]. Perhaps, this longer membrane disruptionexposure has a larger contribution to the secondary spinal cordinjury which is yet to be defined in future studies.

In the event of direct mechanical injury, membrane disrup-tion permits ions to move down their electrochemical gradientacross the damaged membrane. An essential element in sec-ondary injury is the rapid influx and excessive intracellularaccumulation of Ca2+. As an essential secondary messenger,calcium influx results in numerous enzyme activations thatlead to mitochondrial dysfunction, cytoskeleton destruction,free radical production, axonal degeneration, glutamate dis-charge, and eventually, apoptotic or necrotic pathways. Also,some evidence suggests that sodium and potassium also play arole in the secondary injury process. Although axons in thewhite matter lack NMDA receptor-mediated and voltage-sensitive calcium channels, they do express α-amino-3-hy-droxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor-mediated sodium channels. This accumulation of sodium cancause intracellular edema, acidosis, and potentiate calciuminflux through the Na/Ca exchanger [39, 40, 57, 62].

The role of a node of Ranvier in neuronal signal conductionis studied meticulously during the last decade [74]. The actionpotential propagation along myelinated nerve axons requiresintact node of Ranvier, but it has scarcely been mentioned in

Neurosurg Rev

the articles published on the disorders of the central nervoussystem (CNS) [74, 75]. Saltatory conduction through myelin-ated neurons involves Nav1.1, Nav1.6, and KCNQ channels[75, 76]. The node of Ranvier, paranodal region, andjuxtaparanodal regions have variable densities of these chan-nels (Fig. 2), which usually get disrupted due to the stretch orelongation of axons during SCI [72, 77]. This leads to theionic dysequilibrium and contributes to the pathophysiologyof the SCI by altering the function of these myelinated neu-rons by exposing K+ channels [77, 78]. Human and animaltrials are now being performed to remyelinate the nerves dam-aged during the SCI, aiming at functional restoration of themyelinated nerves [79].

Neurotransmitter toxicity

Glutamate is a well-studied, abundant excitatory neurotrans-mitter found throughout the central nervous system. Usually,glutamate is stored in vesicles inside neurons near the axonterminals where they are released into the synapse upon cal-cium influx. Synaptic glutamate is subsequently collected bysurrounding astrocytes, is converted to glutamine by gluta-mine synthetase, and is transferred back to the neuron foradditional glutamate synthesis. Upon traumatic spinal cordinjury, extracellular levels of glutamate accumulate to neuro-toxic levels around the injury site due to excessive release andimpaired uptake [80–82]. These toxic levels are well known to

Fig. 2 Diagram demonstratingthe molecular architecture of themyelinated and demyelinatedspinal cord axons. Myelinatedfibers display a highly organizedmolecular structure, in which theNa+ channels are localized at thenodes of Ranvier, and the Kv1.1and Kv1.2 K+ channel subunitsare located under the compactmyelin sheets in thejuxtaparanodal zones. The nodaland juxtaparanodal regions aredisjointed by the paranode, whichis identified by the presence ofcontactin-associated protein.Myelin loss after injury ordysmyelination results in markedchanges and disruption in themolecular organization of axons

Neurosurg Rev

produce direct and indirect damage to the spinal cord by amassive influx of Ca2+ into the neurons. Intracellular Ca2+

causes glutamate release and activation of a variety of recep-tors including Ca2+ ion channels which play a role in celldeath after SCI [83–86]. Voltage-gated Ca2+ activation or cal-cium leakage after an injury to the cell membrane causes fur-ther elevation of Ca2+ inside the cell and, in turn, increasesrelease of glutamate [86]. This process is known as excitotoxiccell death (Fig. 3). The increase in intracellular Ca2+ is causednot only by the influx of extracellular Ca2+ into the intracel-lular space but also by the release of intracellular Ca2+ that isstored within intracellular storages (mitochondria, endoplas-mic reticulum (ER) [87]), into the cytosol of injured spinalcord neuron and the entire length of an axon [88, 89].Axonal degeneration is mediated by the opening of ER-derived Ca2+ from inositol triphosphate (IP3) receptors andryanodine receptors (RyRs), leading to the activation of themitochondrial permeability transition pore (mPTP) that

contributes to axonal degeneration, as described in a studyby Villegas et al. in a rat model of SCI [90].

Glutamate, however, must first bind to receptor proteins toelicit this influx of ions into the neuron. Two classes of gluta-mate receptors reside on the post-synaptic neuron: metabotro-pic and inotropic. Metabotropic receptors are coupled to Gproteins and act through secondary messengers. They inducecalcium release from endoplasmic reticulum stores throughthe phospholipase C (PLC) pathway and alone are not partic-ularly neurotoxic. Instead, they appear to potentiate AMPAcurrent and exacerbate the inotropic receptor-mediatedexcitotoxicity. The inotropic receptors include NMDA,AMPA, and kainate that control channels for the cations,Na+, Ca2+, and K+ [36, 58, 62, 91].

These glutamate receptors have been shown to be presenton neurons and glial cells, especially on oligodendrocytes andastrocytes, leaving them particularly vulnerable to thisexcitotoxicity. This toxic accumulation of glutamate disrupts

Fig. 3 Diagram illustratingseveral mechanisms of axonaldegeneration. The influx ofextracellular Ca2+ into thecytoplasm leads to the activationof neuronal cell death processes.Intracellular calcium storesrelease Ca2+ from endoplasmicreticulum (ER) through inositoltriphosphate (IP3) receptors andryanodine receptors (RyRs),leading to the activation of themitochondrial permeabilitytransition pore (mPTP) thatfurther adds to the axonaldegeneration of reactive oxygenspecies (ROS) and calpainproteases. Glutamate release alsocontributes to the cell injury

Neurosurg Rev

ionic homeostasis and impairs normal mitochondrial function-ing, which leads to free radical production and subsequentlipid peroxidation. This noxious cascade results in demyelin-ation of axons and loss of neurons around the injury site. Withthe loss of myelin, axons are directly exposed to the lethaleffects of free radicals and inflammatory cytokines.Conduction delays or blockage become clinically evident asthe demyelination leads to significant motor and sensory def-icits [39, 40, 75].

Subacute phase

The secondary phase of spinal cord injury can last up to weeksor even months. During the subacute phase of injury, the areaof trauma distinctly enlarges and features a continuation of theacute stages with the addition of more novel mechanisms. Thesubacute phase is highlighted by the introduction of free rad-ical production, lipid peroxidation, and immune-mediatedneurotoxicity [33, 36, 62].

Free radical injury

A well-characterized pathological process is occurring earlyafter SCI is the formation of reactive oxygen species (ROS)and reactive nitrogen species (RNS). This is a sequel to in-creased intracellular calcium levels, mitochondrial dysfunc-tion, arachidonic acid breakdown, and activation of induciblenitric oxide synthase. ROS and RNS cause lipid peroxidationas well as oxidative and nitrative damage to proteins andnucleic acids. Apart from cell membrane lysis leading to neu-ronal loss, free radicals invoke other types of damage, partic-ularly on the cytoskeleton and organelles. Also, oxidativedamage exacerbates mitochondrial dysfunction and contrib-utes to intracellular calcium overload which activates prote-ases resulting in the breakdown of cytoskeletal proteins [39,40].

Oxidative stress or free radical production is a well-characterized pathological process involved in secondaryspinal cord injury. It is simply defined as an imbalancebetween high levels of ROS and reactive nitrogen speciesand suboptimal levels of the antioxidative defense. Thesetoxic molecules are the consequence of oxidation-reduction reactions that are highly reactive and unstabledue to the odd number of electrons. Under physiologicalconditions, these ROS are essential for life and are in-volved in many roles throughout the cell. However, theCNS is uniquely vulnerable to free radical injury. TheCNS has only moderate levels of endogenous antioxidantsand antioxidative enzymes. Following acute spinal cordtrauma, these levels decrease dramatically. Additionally,the CNS also has a very high metabolic rate producingconstant levels of ROS, especially at times during energymetabolism compromise such as traumatic injury. Lastly,

the CNS contains significant amounts of transition metalssuch as copper and manganese that actively participate inthe free radical production. As a consequence, increasedintracellular calcium levels, mitochondrial dysfunction,and aberrant enzyme activation all lead to a massive surgeof free radical production that damage a wide range ofmolecular species including lipids, proteins, and nucleicacids. Aside from cell lysis and neuronal loss, free radicalsinvoke tremendous damage on cytoskeleton and organellesthat are essential for cellular function. Specifically, oxida-tive damage exacerbates the mitochondrial dysfunctionand contributes precipitously to the intracellular calciumoverload. This overload of calcium can activate a familyof calcium-dependent cysteine proteases known ascalpains. Calpains are a large family of proteases involvedin remodeling, signal transduction, differentiation, devel-opment, as well as apoptosis and necrosis [39, 40, 42, 92,93]. Interestingly, known substrates for these proteases arethe remarkable central and axonal cytoskeleton compo-nents responsible for both anterograde and retrogradetransports, electrical conduction, as well as neurotransmit-ter release in human and mammalian models of SCI [94,95].

Lipid peroxidation

ROS have very short half-lives while nitric oxide survivesrelatively longer. The reactive nitrogen species such asperoxynitrite (PN), a metabolite of NO and superoxide, isinvolved in one of the most investigated molecular mecha-nisms responsible for secondary spinal cord injury. It is theprincipal initiator of free radical-induced, iron-catalyzed lipidperoxidation and protein oxidative/nitrative damage in the in-jured spinal cord. Lipid peroxidation is a highly destructivereaction that is initiated when ROS attacks membrane poly-unsaturated fatty acids (PUFAs). These polyunsaturated fattyacids are quickly converted into lipid alkyl radicals that reactwith molecular oxygen to form a highly reactive lipid peroxylradical. These lipid peroxyl radicals react with other PUFAswithin the cell or neighboring cells, converting them intomorelipid alkyls and creating a progressive and continuous chainreaction. Therefore, this non-enzymatic-mediated lipid perox-idation pathway is a self-propagating process that can spreadmembrane damage to even nearby healthy cells. Of more sig-nificant concern, because CNS membranes contain compara-tively large amounts of fatty acids like linoleic acid and ara-chidonic acid, the CNS is exceptionally vulnerable to gener-ating large quantities of lipid peroxidation product. This prop-agation of membrane damage only further deteriorates theCa2+ homeostasis and additively contributes to the previouslymentioned neuronal degeneration and massive protease acti-vation in various mice models of spinal cord contusion andcompression [39, 40, 95].

Neurosurg Rev

Immune-associated neurotoxicity

The role of the immune system following a spinal cord injuryhas proven to be quite controversial throughout the literature,as described in human models of SCI [96]. Although inflam-mation has been primarily shown to be neurotoxic, there isplenty of evidence to suggest a neuroprotective role as well[97]. Therefore, the inflammatory role in spinal cord injuryappears to be two-sided, with no obvious approachable wayto delineate between the two. Mechanical injury to the spinalcord elicits a robust and highly coordinated inflammatory-immune response that directly and indirectly contributes tothe progression of secondary injury [96]. This inflammatoryimmune response can be further divided into a more spatialand chronological manner, briefly summarized as immediateneutrophil invasion, activation of resident microglia, recruit-ment of blood monocytes, and scar formation [39–41,98–101].

Immediate release of proinflammatory cytokines andchemokines from spinal cord cells like microglia near theprimary injury begins the inflammatory responses. Upon in-jury, neutrophils are the first blood-derived cell to arrive at thelesion and are primarily restricted to vessels and the hemor-rhagic area within the first 3–6 h. Upon 24 h, the neutrophillevel peaks and can persist for several months following theinjury. Although imperative to preventing further damage byphagocytosis and removing harmful cellular debris, they mayalso release potentially toxic factors such as free radicals, cy-tokines, hypochlorous acid, and proteinases. These toxins arethought to propagate secondary adjacent cellular injury andlead to degeneration of vital and mostly non-regenerative neu-ronal and glial elements. Inhibition of neutrophil adhesion tothe endothelial cell surface has been shown to reduce the se-verity of secondary injury following spinal cord trauma[39–41, 98–101].

Upon initial mechanical injury, resident macrophages andrecruited monocyte-derived macrophages quickly migrate tothe site of insult. Within 5–10 days, macrophages dominatethe site of injury and can reside for several months followinginjury. This rapid activation of microglia leads to a surge ofproinflammatory mediators including IL-1, IL-6, and TNFwhich also mediates the glial scar formation, as explained inmice models of SCI, and is detailed later [102]. Also, theyincrease the endothelial expression of chemoattractant andadhesion molecules, increasing extravasation of circulatingmyeloid cells and lymphocytes into the site of injury[39–41, 98–101]. This massive accumulation of macrophagescan have some implication in the progression of secondaryinjury [40]. A large number of macrophages that can accumu-late have been documented to be as high as 6000 cells/mm2

[103]. This mass effect can be loosely compared to the im-mune effect caused by a benign tumor in the brain and itsimpact on the surrounding uninjured parenchyma [40].

A topic of relatively new research interest and a hallmark ofimmune-inflammatory injury in spinal cord trauma is the pop-ulation of two types of macrophages: resident microglia andmonocyte-derived. They play a critical role in the degenera-tion and regeneration of tissues following spinal cord injury.Blood-borne monocytes are directed towards the lesion of SCIby chemotaxis due to the release of specific cytokines from thelesion [36, 104]. These inflammatory markers includingmonocyte chemoattractant protein 1 (MCP-1), matrix metal-loproteinase 9 (MMP-9), and stromal cell-derived factor 1(SCDF-1) synergistically facilitate migration of blood-bornemonocytes into the site of SCI [105]. These different macro-phages may also have various functions as bone marrow-derived macrophages migrate to the epicenter of injury whileresident macrophages localize to the edges of the lesion [101].Additionally, newly classified subsets of macrophages havebeen described to appear at the injured site: M1 and M2.Depending on the microenvironment in which they reside,their phenotype and activation can change to either initiatesecondary injury or initiate repair [33, 36, 62]. The M1 phe-notype is considered the proinflammatory subset that secretescytokines and chemokines immediately after the injury thatboth lead to further progression of secondary injury.Although M1 and M2 macrophages coexist at the lesion epi-center during the initial week after injury, it appears that onlythe M1 macrophage persists and can reside up to a month postinjury. The M2 phenotype is considered anti-inflammatoryand secretes cytokines and chemokines that appear immunesuppressive and lead to regeneration of injured and surround-ing spinal tissues [40, 41, 101]. For unknown reasons, expres-sion of M2 genes is only temporary following SCI and rapidlyreturns to pre-injury levels within a week [41]. Although it isunclear what determines the expression of these macrophagesubsets, it appears the activation is driven by the lesion-relatedfactors present at the time of injury. Myelin debris, as well aslarge TNF-α expression, are just some of the factors describedas preventing M1 to an M2 conversion, resulting in a state ofchronic inflammatory expression [41, 101].

General principles tell us that proper repair and tissuehealing require a coordinated inflammatory response, yet aprolonged inflammatory state is associated with poor tissuehealing. It is this feature of chronicity that is poorly under-stood following spinal cord injury. It is well known that clear-ance of cellular and myelin debris is significantly delayed inthe injured CNS compared to PNS. Oligodendrocytes havelittle to no capacity to clear myelin debris and therefore pro-vide incredibly potent inflammatory signals. It is suggestedthat perhaps the immune-privileged sites like the spinal cordexceed their ability to resolve inflammation following trauma.However, it is clear that specific cells like B cells, microglia,and macrophages persist longer and at higher levels well afterthe initial insult [41, 98]. This chronic state of spinal inflam-mation extends the possibility of exposed self-epitopes which

Neurosurg Rev

are usually shielded by the blood-spinal cord barrier. Also,uncleared apoptotic bodies can also release immunologicself-antigens, all presenting a risk of developing autoimmu-nity. As these self-antigens leak into circulation, they eventu-ally drain into secondary lymphoid organs where specificmature T and B lymphocytes form and eventually infiltratethe spinal cord barrier. These activated lymphocytes secretenumerous inflammatory and stimulatory cytokines.Additionally, the antibody/antigen complexes formed fromthis autoimmunity recruit and activate complement and Fcreceptor-bearing cells, leading to progressive axonal lossand demyelination [106–108]. Reduced inflammatory accu-mulation and functional recovery were all seen in mice lack-ing B cells and C3 complement components, providing adynamic role of the innate immune system following traumat-ic spinal injury [98]. As it appears, the dichotomized view ofinflammation as either good or bad is not only inaccurate butalso likely a reflection of our poor understanding of its spatialand temporal interactions following SCI.

Astrocytic glial scar formation

After SCI, astrocytes also play a critical role during thesubacute phase of the injury. Astrocytes surround neuronsin the entire CNS and are five times more in number thanneurons [109]. They are specialized supporting cells,which exert multiple crucial, complex functions in theCNS [109]. Reactive gliosis occurs after an injury toCNS, which is a pathological hallmark of CNS structurallesion [110–112]. Mature astrocytic scar is formed as aresult of reactive gliosis, that is composed of a narrowzone of astrocytes with elongated processes interminglingwith each other to surround the lesion core [109, 110].These scar-forming astrocytes are the consequence ofCNS insult, and hence, they are significantly higher innumber and densely packed around the lesion as com-pared to the regions of healthy tissues [110]. As theysecure the borders of SCI lesion, they play a neuroprotec-tive role by limiting the spread of inflammatory cells be-yond the margins of the astrocytic gliotic scar. In theexperimental studies, when astrocytic scar formation wasattenuated, the lesion size increased and demyelinationand neuronal loss were exacerbated which significantlyreduced the functional recovery of neurons after CNS in-jury [113–115]. These scar-forming astrocytes are alsothought to be a significant impediment to the axonal re-generation in in vitro studies as they secrete some proteo-glycans to inhibit axonal elongation [116]. Although somein vivo studies showed the crucial role of astrocytic brid-ges in axonal regeneration [111, 117, 118], further studieson this mechanism may clarify the exact role of this as-trocytic gliotic scar tissue.

Chronic phase

Glial scar formation

Oligodendrocyte apoptosis plays an important role in glialscar formation after SCI [42, 93, 119]. Following CNS injury,the immune-associated microglial response promotes and ac-tivates astrocytes, glial precursors, microglia, and fibroblaststo increase GFAP expression, become hypertrophic, prolifer-ate, and migrate to the epicenter of the lesion to begin theprocess of reactive astrogliosis, leading to the formation ofglial scar (Fig. 4) [41, 101, 102, 120, 121]. The role ofastrogliosis as either neuroprotective or neurotoxic remainsto be unclear [42, 97, 102]. Herrmann et al. used a rat modelof SCI and explained that the suppression of this scar forma-tion after SCI does elicit a decrease in proinflammatory cyto-kines and functional restoration, suggesting a more prominentneuroprotective role [102]. However, there are numerous ev-idences that suggest reactive gliosis plays an essential role ininjury repair [97]. Scar tissues produce some growth-promoting substances, such as fibronectin and laminin.Therefore, astrogliosis appears to be a defense response aimedto limit and repair local damage by isolating the lesion area,restricting the infiltration of peripheral leukocytes, providinggrowth factors, and restoring the blood-spinal cord barrier.Also, activated microglia promote revascularization and up-take of excess glutamate, maintaining and promoting a morefavorable environment for surviving neurons [36, 62, 119].Many genetic studies support this protective role, demonstrat-ing a lack of scar formation that leads to widespread lesioninduction with increasing neuronal cell death and sensory-motor deficits [93, 120, 121].

This seemingly favorable role of glial scar formation isquickly surpassed by overwhelming physical and chemicalbarriers inhibiting any possibility of neuronal regenerationor repair (Fig. 4). The sheer number of astrocytes alonemigrating into and around the lesion site is enough to con-stitute a physical barrier to axonal regeneration. However, itis the upregulation of numerous growth inhibitory mole-cules that contribute to the non-permissive nature of scarformation. The CNS extracellular matrix (ECM) plays asignificant role in facilitating cellular migration, guidance,and synaptogenesis during the development of the CNS[61]. Likewise, the ECM composition and integrity becomedetrimental to the regeneration and repair following spinalcord injury. The ECM composition becomes dramaticallyaltered as the glial scar upregulates and produces a largenumber of molecules that prevent repair and repel axonalgrowth. Among them are tenascin, semaphoring 3A, keratinsulfate proteoglycan (KSPG), myelin-associated glycopro-tein, Nogo, oligodendrocyte-myelin glycoprotein, and mostimportantly, chondroitin sulfate proteoglycan (CSPG).CSPGs are an increasingly studied family of highly sulfated

Neurosurg Rev

glycosaminoglycan-protein chain molecules that are pre-dominantly responsible for inhibiting regeneration of sev-ered axons. CSPGs form numerous connections with otherECM molecules and form unique structures calledperineuronal nets (PNNs). These structures wrap aroundneurons, creating a strong chemical barrier and preventingelongation. These non-permissive PNNs cause a steric hin-drance of growth-promoting adhesion molecules likelaminins and integrins, attenuating their activation and sup-pressing neurite growth. CSPGs also contribute to the func-tion of chemorepulsive proteins by altering the structure ofaxon guidance cues. Upon physical interaction with theglycosaminoglycans (GAGs) of CSPGs, guidance cues likeSema5A can convert from an attractive cue to an inhibitorycue. Additionally, CSPGs can also errantly bind to extra-cellular calcium or its channels and alter calcium influxinto cells, further affecting neuronal growth [121].

The inhibition of axonal growth by CSPGs appears likelymultifactorial, thus making it a challenging therapeutic target.However, recent studies indicate a much more potent targetthat mediates CSPG inhibitory effects. CSPGs have beenshown to induce growth inhibition by binding and activatingseveral essential receptor proteins, including common leuko-cyte antigen-related (LAR) phosphatase, PTPσ, and Nogo re-ceptor 1 (NgR1) and NgR3. PTPσ and common LAR phos-phatase are transmembrane protein receptors that modulatetyrosine phosphorylation within the cell and have been shownto bind CSPGs with high affinity and mediate CSPG inhibi-tory effects.

Nogo receptors

Nogo receptors (NgRs) are a family of membrane proteins thatshare very similar structures. NgR1 has been known to bind

Fig. 4 Glial scar formation after the spinal cord injury at the site of impact, leading to recruitment of immune cells

Neurosurg Rev

myelin inhibitors like Nogo. Specifically, Nogo-A is a myelintransmembrane protein responsible for stabilizing axonal con-nectivity and has been shown to collapse growth cones andstop neurite elongation [121]. NgR1 and NgR3 bind CSPGGAGs and mediate inhibition of neuronal growth [121, 122].All of these receptors act intracellularly and activate the GTP-binding signaling protein RhoA. RhoA interacts with numer-ous molecules throughout the cell and regulates neuronal mor-phogenesis. More importantly, it can bind and activate theRho kinase signaling pathway that leads to phosphorylationof target proteins like Akt and Erk, subsequently inactivatingthem, as explained in various human and animal models ofSCI [121]. Nogo-A has two domains, Nogo-66 and amino-Nogo, each having a distinct mechanism of action. The Nogo-66 receptor is present on the axonal membrane, and Nogo-66itself is a neuron-specific inhibitor. In contrast, amino-Nogo isa non-specific inhibitor. The second messenger activity ofNogo-66 destabilizes the cellular cytoskeleton and hence im-pediment to the initial growth [123–126]. Nogo and variousmolecules also act non-specifically and simultaneously as areceptor and ligand by a complex mechanism on neuronal,oligodendrocytic, and glial surface cells [127]. Nogo antibodyis also a subject of interest recently and shows that progressiveaxonal sprouting and reorganization occur, but these antibod-ies also act non-specifically of other receptors, henceexplaining the diverse role of these molecules in the recoveryfrom SCI [128, 129]. Recently, a phase I/IIa clinical trial of aRhoA inhibitor successfully showed beneficial and safe treat-ment [36, 58, 130].

Current treatment approaches

Keeping in view the basic underlying mechanism of SCIdiscussed above, many treatment options are currently being

pursued in SCI patients. Blood pressure control to optimizethe perfusion pressure is vital for the neurologic recovery, andincreased ISP can worsen the ischemia in the neurons after anacute SCI. Squair et al. presented an interesting finding thatthe spinal cord perfusion pressure (SCPP) is a better predictorof neurologic recovery after an SCI than the traditionally mea-sured mean arterial pressure (MAP) [131]. Authors recom-mended that keeping SCPP > 50 mmHg is a good indicatorof improved neurologic outcomes following SCI [131]. Somecritiques debate that as 2013 guidelines recommend the MAPgoal of 85–90 mmHg for up to 7 days, there could be otherfactors affecting SCPP so it might not be accurate [132].However, as the SCPP is derived from MAP, it directly re-flects the changes in SCPP and can be considered an excellenttool to predict neurologic recovery. Management focused onmonitoring and controlling the ISP may improve outcomesand recovery from the acute SCI. Tykocki et al. presented asystemic review discussing ISPmonitoring and recommendedbony decompression with durotomy or duroplasty overlaminectomy alone to optimize the ISP for the improved re-covery of injured neurons after an SCI [133].

In general, there are three main strategies for treatment ofchronic SCI: (1) cellular transplantation, in which transplan-tation takes advantage of plasticity to harness residual circuit-ry; (2) pharmacological treatment, in which treatment aims atencouraging active regeneration of injured neurons; and (3)neurostimulation which aims at using technology to restorefunction without restoring neural architecture. The choice oftreatment approach depends on the severity of SCI whether itis a complete or incomplete spinal cord injury outlined inFig. 5 [131].

Neurostimulation strategies such as deep brain stimulation(DBS), spinal cord stimulation (SCS), motor cortex stimula-tion (MCS), transcutaneous direct current stimulation (tDCS),and repetitive transcranial magnetic stimulation (rTMS) in

Fig. 5 Current treatment approaches after complete and incomplete spinal cord injury (SCI), simplified in as a flowchart

Neurosurg Rev

affecting pain, sensorimotor symptoms, and autonomic dys-regulation, all of which are important sequelae in SCI, areused in incomplete severely affected chronic SCI (ASIA gradeB/C) [23, 134]. Electrical stimulation to restore the function ofspinal cord injury has been used for quite some time now [64,135, 136]. Neuroprosthesis that can stimulate the spinal cordand the muscles is seen to revive the spinal cord circuits con-trolling the motor functions after the SCI, thus aiding the re-habilitation of such patients [22, 137]. It has been demonstrat-ed that epidural spinal cord stimulation can produce walk-likemotion of the lower limbs [136, 138], where the electrodes areplaced on the dorsal surface of the spinal cord at the lumbo-sacral region [137, 139].

In patients with cervical SCI who are unable to sustainindependent ventilation because of a disruption of diaphragminnervation, percutaneous stimulation and pacing of the dia-phragm is done to achieve respiratory functioning [140–143].This technique relies on intact phrenic nerve function, butlately, phrenic nerve reconstruction with intercostal nervegrafting has expanded its indications and helps reduce or elim-inate ventilator support in high-cervical (C3–C5) SCI patients[141, 144, 145]. The Brindley procedure consists of implan-tation of a sacral anterior root stimulator and a rhizotomy ofthe dorsal sacral roots to control detrusor function [146–149].Intermittent stimulation of the device on the skin by an exter-nal transmitter enabled emptying of bladder, defecation, anderection [146, 150, 151]. The device is implanted either byintrathecal or extradural administration in patients with com-plete SCI to eliminate neurogenic detrusor overactivity [146,150, 152–154]. However, the current percutaneous implanta-tion techniques are limited to superficial extrapelvic nervesthat expose patients to various complications like infectionsand can lead to migration and dislocation of skin.Laparoscopic implantation of neuroprosthesis (LION) proce-dures permit easy access and visibility to all pelvic nerves andplexuses as well as the diaphragm for the implantation ofneurostimulators [155–157]. This technique allows protectionagainst abovementioned complications and avoids previouslyused arduous and risky open surgical approaches.

Recently, a lot of attention has been directed at therapeuticstrategies for SCI, including cell transplantation and drug ad-ministration [21, 57, 58, 91, 158–160]. Riluzole, abenzothiazole anticonvulsant approved for treatment of amyo-trophic lateral sclerosis (ALS) patients, has neuroprotectiveproperties based on the inhibition of pathologic glutamatergictransmission in synapses of neurons via sodium channelblockade [159]. It has also shown to support the motor andrespiratory recovery by promoting neuronal survival and func-tion of the neural network below the SCI level following in-jury in a cervical (C2) spinal cord hemisection model [161].Clinical trials for the treatment of acute traumatic SCI withRiluzole are underway [24, 162]. The results of phase IIplacebo-controlled randomized trial of minocycline in acute

SCI also showed some encouraging results warranting furtherinvestigations [25].

Stem cell therapy can be considered in the complete (ASIAgrade A) SCI [57, 58, 91]. There is abundant preclinical evi-dence supporting the regeneration of neurons using stem celltherapy, while in some cases, conflicting evidence is also pres-ent [163–165]. The therapeutic potential of olfactoryensheathing cells andmesenchymal stem cells in SCI has beenestablished, and clinical trials have been carried out [158, 166,167]. From the present review of literature, there seem to be avery strong need for more phase II clinical trials and evenfurther studies to explore the efficacy and clinical translationof the existent preclinical data [163, 168].

Conclusion

SCI is a complex and multifaceted mechanism. Increased cy-tosolic intracellular calcium plays a crucial role. Apart fromthe basic neuroscientists, it is imperative for the treating phy-sicians to understand the complex underlying mechanism ofSCI. The authors have cataloged these fundamental injurymechanisms that intensify SCI.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict ofinterest.

Ethical approval All procedures were done in accordance with the eth-ical standards.

Informed consent No human subject was involved, and no informedconsent was required for this review.

References

1. WHO (2017) Spinal cord injury. Available from: http://www.who.int/mediacentre/factsheets/fs384/en/

2. Wyndaele M, Wyndaele JJ (2006) Incidence, prevalence and ep-idemiology of spinal cord injury: what learns a worldwide litera-ture survey? Spinal Cord 44(9):523–529

3. Selassie A, Cao Y, Saunders LL (2015) Epidemiology of traumaticspinal cord injury among persons older than 21 years: apopulation-based study in South Carolina, 1998-2012. TopSpinal Cord Inj Rehabil 21(4):333–344

4. National Spinal Cord Injury Statistical Center (2015) Facts andfigures at a glance. University of Alabama at Birmingham.Available at: http://www.msktc.org/lib/docs/Data_Sheets_/MSKTC_SCIMS_Fact_Fig_2015.pdf, Birmingham

5. Jazayeri SB, Beygi S, Shokraneh F, Hagen EM, Rahimi-Movaghar V (2015) Incidence of traumatic spinal cord injuryworldwide: a systematic review. Eur Spine J 24(5):905–918

6. Singh A et al (2014) Global prevalence and incidence of traumaticspinal cord injury. Clin Epidemiol 6:309–331

7. Rahimi-Movaghar V, Sayyah MK, Akbari H, Khorramirouz R,Rasouli MR, Moradi-Lakeh M, Shokraneh F, Vaccaro AR

Neurosurg Rev

(2013) Epidemiology of traumatic spinal cord injury in develop-ing countries: a systematic review. Neuroepidemiology 41(2):65–85

8. Klebine P (2015) Understanding spinal cord injury: part 1—thebody before and after injury in collaboration with the model sys-tems knowledge translation center. The National Spinal CordInjury Statistical Center

9. Mortazavi M, Gore PA, Chang S, Tubbs RS, Theodore N (2011)Pediatric cervical spine injuries: a comprehensive review. ChildsNerv Syst 27(5):705–717

10. Mortazavi MM, Dogan S, Civelek E, Tubbs RS, Theodore N,Rekate HL, Sonntag VKH (2011) Pediatric multilevel spine inju-ries: an institutional experience. Childs Nerv Syst 27(7):1095–1100

11. Krueger H et al (2013) The economic burden of traumatic spinalcord injury in Canada. Chronic Dis Inj Can 33(3):113–122

12. Collie A, Keating C, Pezzullo L, Gabbe B, Cooper J, Brown D,Olver J, McCartin F, Trethewey P (2010) Brain and spinal cordinjury in Australia—economic cost and burden of disease. Inj Prev16(Suppl 1):A25–A26

13. Sekhon LH, Fehlings MG (2001) Epidemiology, demographics,and pathophysiology of acute spinal cord injury. Spine (Phila Pa1976) 26(24 Suppl):S2–S12

14. Hohmann GW (1975) Psychological aspects of treatment and re-habilitation of the spinal cord injured person. Clin Orthop RelatRes 112:81–88

15. Orbaan IJ (1986) Psychological adjustment problems in peoplewith traumatic spinal cord lesions. Acta Neurochir 79(1):58–61

16. Nas K et al (2015) Rehabilitation of spinal cord injuries. World JOrthop 6(1):8–16

17. Mortazavi MM,Mariwalla NR, Horn EM, Tubbs RS, Theodore N(2011) Absence of MRI soft tissue abnormalities in severe spinalcord injury in children: case-based update. Childs Nerv Syst 27(9):1369–1373

18. Yang XX, Huang ZQ, Li ZH, Ren DF, Tang JG (2017) Riskfactors and the surgery affection of respiratory complication andits mortality after acute traumatic cervical spinal cord injury.Medicine (Baltimore) 96(36):e7887

19. Majdan M, Plancikova D, Nemcovska E, Krajcovicova L,Brazinova A, Rusnak M (2017) Mortality due to traumatic spinalcord injuries in Europe: a cross-sectional and pooled analysis ofpopulation-wide data from 22 countries. Scand J Trauma ResuscEmerg Med 25(1):64

20. Biering-Sorensen F et al (2017) International Spinal Cord InjuryCore Data Set (version 2.0)—including standardization ofreporting. Spinal Cord 55(8):759–764

21. Sobani ZA, Quadri SA, Enam SA (2010) Stem cells for spinalcord regeneration: current status. Surg Neurol Int 1:93

22. Jackson A, Zimmermann JB (2012) Neural interfaces for the brainand spinal cord—restoring motor function. Nat Rev Neurol 8(12):690–699

23. Chari A, Hentall I, Papadopoulos M, Pereira E (2017) Surgicalneurostimulation for spinal cord injury. Brain Sci 7(2)

24. Fehlings MG, Wilson JR, Frankowski RF, Toups EG, Aarabi B,Harrop JS, Shaffrey CI, Harkema SJ, Guest JD, Tator CH, BurauKD, Johnson MW, Grossman RG (2012) Riluzole for the treat-ment of acute traumatic spinal cord injury: rationale for and designof the NACTN phase I clinical trial. J Neurosurg Spine 17(1Suppl):151–156

25. Casha S, Zygun D, McGowan MD, Bains I, Yong VW, JohnHurlbert R (2012) Results of a phase II placebo-controlled ran-domized trial of minocycline in acute spinal cord injury. Brain135(Pt 4):1224–1236

26. Hayta E, Elden H (2017) Acute spinal cord injury: a review ofpathophysiology and potential of non-steroidal anti-inflammatory

drugs for pharmacological intervention. J ChemNeuroanat 87:25–31

27. Phillips AA, Krassioukov AV (2015) Contemporary cardiovascu-lar concerns after spinal cord injury: mechanisms, maladaptations,and management. J Neurotrauma 32(24):1927–1942

28. Meister R et al (2014) Neurogenic shock. Rev Med Suisse10(438):1506–1510

29. Marx JA, Rosen P (2014) Rosen’s emergency medicine: conceptsand clinical practice, 8th edn. Elsevier/Saunders, Philadelphia

30. American Spinal Injury Association (ASIA) (1998) Abstracts. JSpinal Cord Med 21(2):151–194

31. Roberts TT, Leonard GR, Cepela DJ (2017) Classifications inbrief: American Spinal Injury Association (ASIA) ImpairmentScale. Clin Orthop Relat Res 475(5):1499–1504

32. Anderson DK, Means ED, Waters TR, Green ES (1982)Microvascular perfusion and metabolism in injured spinalcord after methylprednisolone treatment. J Neurosurg 56(1):106–113

33. Tubbs RS, Blouir MC, Romeo AK,Mortazavi MM, Cohen-GadolAA (2011) Spinal cord ischemia and atherosclerosis: a review ofthe literature. Br J Neurosurg 25(6):666–670

34. Kobayashi T (1968) Experimental study on pathologicalphases of whiplash injury. Nihon Seikeigeka Gakkai Zasshi42(1):1–12

35. Allen A (1911) Surgery of experimental lesion of spinal cordequivalent to crush injury of fracture dislocation of spinal column:a preliminary report. J Am Med Assoc LVII(11):878–880

36. Mortazavi MM, Verma K, Deep A, Esfahani FB, Pritchard PR,Tubbs RS, Theodore N (2011) Chemical priming for spinal cordinjury: a review of the literature. Part I—factors involved. ChildsNerv Syst 27(8):1297–1306

37. Ray SK, Dixon CE, Banik NL (2002) Molecular mechanisms inthe pathogenesis of traumatic brain injury. Histol Histopathol17(4):1137–1152

38. Rossignol S, SchwabM, Schwartz M, FehlingsMG (2007) Spinalcord injury: time to move? J Neurosci 27(44):11782–11792

39. Oyinbo CA (2011) Secondary injury mechanisms in traumaticspinal cord injury: a nugget of this multiply cascade. ActaNeurobiol Exp (Wars) 71(2):281–299

40. Borgens RB, Liu-Snyder P (2012) Understanding secondary inju-ry. Q Rev Biol 87(2):89–127

41. Bastien D, Lacroix S (2014) Cytokine pathways regulating glialand leukocyte function after spinal cord and peripheral nerve in-jury. Exp Neurol 258:62–77

42. Beattie MS, Farooqui AA, Bresnahan JC (2000) Review of cur-rent evidence for apoptosis after spinal cord injury. J Neurotrauma17(10):915–925

43. Tator CH et al (1984) Management of acute spinal cord injuries.Can J Surg 27(3):289–293 296

44. Dumont RJ, Verma S, Okonkwo DO, Hurlbert RJ, Boulos PT,Ellegala DB, Dumont AS (2001) Acute spinal cord injury, partII: contemporary pharmacotherapy. Clin Neuropharmacol 24(5):265–279

45. Benzel EC, Larson SJ (1986) Functional recovery after decom-pressive operation for thoracic and lumbar spine fractures.Neurosurgery 19(5):772–778

46. Suwanwela C, Alexander E Jr, Davis CH Jr (1962) Prognosis inspinal cord injury, with special reference to patients with motorparalysis and sensory preservation. J Neurosurg 19:220–227

47. Kojima Y et al (1979) Evoked spinal potentials as a monitor ofspinal cord viability. Spine (Phila Pa 1976) 4(6):471–477

48. Saul TG, Carol M, Ducker TB (1982) Immediate mini-myelography in acute cervical cord injuries. Am Surg 48(9):463–468

49. Oiwa T (1983) Experimental study on the post-laminectomy de-terioration in cervical spondylotic myelopathy—influences of the

Neurosurg Rev

meningeal treatment and persistent spinal cord block. NihonSeikeigeka Gakkai Zasshi 57(6):577–592

50. Elia C, Hariri OR, Duong J, Dong F, Sweiss R, Miulli D (2018)Use of a pediatric craniotome drill for cervical and thoracic spinedecompression: safety and efficacy. World Neurosurg 113:e486–e489

51. Tator CH (1995) Update on the pathophysiology and pathology ofacute spinal cord injury. Brain Pathol 5(4):407–413

52. Hariri OR (2018) Posterior-only stabilization for traumaticthoracolumbar burst fractures. Cureus 10(3):e2296

53. Young W (1993) Secondary injury mechanisms in acute spinalcord injury. J Emerg Med 11(Suppl 1):13–22

54. Pang D, Wilberger JE Jr (1982) Spinal cord injury without radio-graphic abnormalities in children. J Neurosurg 57(1):114–129

55. Tator CH, Fehlings MG (1991) Review of the secondary injurytheory of acute spinal cord trauma with emphasis on vascularmechanisms. J Neurosurg 75(1):15–26

56. Wolman L (1965) The disturbance of circulation in traumatic para-plegia in acute and late stages: a pathological study. Paraplegia 2:213–226

57. Mortazavi MM, Verma K, Tubbs RS, Theodore N (2011) Cellularand paracellular transplants for spinal cord injury: a review of theliterature. Childs Nerv Syst 27(2):237–243

58. Mortazavi MM, Verma K, Deep A, Esfahani FB, Pritchard PR,Tubbs RS, Theodore N (2011) Chemical priming for spinal cordinjury: a review of the literature part II—potential therapeutics.Childs Nerv Syst 27(8):1307–1316

59. Popa C et al (2010) Vascular dysfunctions following spinal cordinjury. J Med Life 3(3):275–285

60. Losey P, AnthonyDC (2014) Impact of vasculature damage on theoutcome of spinal cord injury: a novel collagenase-induced modelmay give new insights into the mechanisms involved. NeuralRegen Res 9(20):1783–1786

61. Blight AR, Young W (1989) Central axons in injured cat spinalcord recover electrophysiological function followingremyelination by Schwann cells. J Neurol Sci 91(1–2):15–34

62. Mortazavi MM, Verma K, Harmon OA, Griessenauer CJ, AdeebN, Theodore N, Tubbs RS (2015) The microanatomy of spinalcord injury: a review. Clin Anat 28(1):27–36

63. Guha A, Tator CH, Rochon J (1989) Spinal cord blood flow andsystemic blood pressure after experimental spinal cord injury inrats. Stroke 20(3):372–377

64. Minassian K et al. (2002) Effective spinal cord stimulation (SCS)for evoking stepping movement of paralyzed human lower limbs:study of posterior root muscle reflex responses. na

65. Vogel C, Rukwied R, Stockinger L, Schley M, Schmelz M,Schleinzer W, Konrad C (2017) Functional characterization ofat-level hypersensitivity in patients with spinal cord injury. JPain 18(1):66–78

66. Taylor SW, Laughlin RS, Kumar N, Goodman B, Klein CJ, DyckPJ, Dyck PJB (2017) Clinical, physiological and pathologicalcharacterisation of the sensory predominant peripheral neuropathyin copper deficiency. J Neurol Neurosurg Psychiatry 88(10):839–845

67. Berger MJ, Kimpinski K, Currie KD, Nouraei H, Sadeghi M,Krassioukov AV (2017) Multi-domain assessment of autonomicfunction in spinal cord injury using a modified autonomic reflexscreen. J Neurotrauma 34(18):2624–2633

68. Lee JS, Fang SY, Roan JN, Jou IM, Lam CF (2016) Spinal cordinjury enhances arterial expression and reactivity of alpha1-adrenergic receptors-mechanistic investigation into autonomicdysreflexia. Spine J 16(1):65–71

69. Anthes DL, Theriault E, Tator CH (1996) Ultrastructural evidencefor arteriolar vasospasm after spinal cord trauma. Neurosurgery39(4):804–814

70. Tator CH, Koyanagi I (1997) Vascular mechanisms in the patho-physiology of human spinal cord injury. J Neurosurg 86(3):483–492

71. Holtz A, Nystrom B, Gerdin B (1990) Relation between spinalcord blood flow and functional recovery after blocking weight-induced spinal cord injury in rats. Neurosurgery 26(6):952–957

72. Ouyang H, SunW, Fu Y, Li J, Cheng JX, Nauman E, Shi R (2010)Compression induces acute demyelination and potassium channelexposure in spinal cord. J Neurotrauma 27(6):1109–1120

73. Hendricks BK, Shi R (2014) Mechanisms of neuronal membranesealing following mechanical trauma. Neurosci Bull 30(4):627–644

74. Arancibia-Carcamo IL, Attwell D (2014) The node of Ranvier inCNS pathology. Acta Neuropathol 128(2):161–175

75. Vacher H, Mohapatra DP, Trimmer JS (2008) Localization andtargeting of voltage-dependent ion channels in mammalian centralneurons. Physiol Rev 88(4):1407–1447

76. Lorincz A, Nusser Z (2010) Molecular identity of dendriticvoltage-gated sodium channels. Science 328(5980):906–909

77. Sun W, Fu Y, Shi Y, Cheng JX, Cao P, Shi R (2012) Paranodalmyelin damage after acute stretch in guinea pig spinal cord. JNeurotrauma 29(3):611–619

78. Fu Y, Sun W, Shi Y, Shi R, Cheng JX (2009) Glutamateexcitotoxicity inflicts paranodal myelin splitting and retraction.PLoS One 4(8):e6705

79. Khalaj AJ, Hasselmann J, Augello C, Moore S, Tiwari-WoodruffSK (2016) Nudging oligodendrocyte intrinsic signaling toremyelinate and repair: estrogen receptor ligand effects. J SteroidBiochem Mol Biol 160:43–52

80. Faden AI et al (1989) The role of excitatory amino acids andNMDA receptors in traumatic brain injury. Science 244(4906):798–800

81. Katayama Y, Becker DP, Tamura T, Hovda DA (1990) Massiveincreases in extracellular potassium and the indiscriminate releaseof glutamate following concussive brain injury. J Neurosurg 73(6):889–900

82. Palmer AM, Marion DW, Botscheller ML, Bowen DM, DeKoskyST (1994) Increased transmitter amino acid concentration in hu-man ventricular CSF after brain trauma. Neuroreport 6(1):153–156

83. Hayes RL, Jenkins LW, Lyeth BG (1992) Neurotransmitter-mediated mechanisms of traumatic brain injury: acetylcholineand excitatory amino acids. J Neurotrauma 9(Suppl 1):S173–S187

84. Olney JW (1969) Brain lesions, obesity, and other disturbances inmice treated with monosodium glutamate. Science 164(3880):719–721

85. Weber JT (2004) Calcium homeostasis following traumatic neu-ronal injury. Curr Neurovasc Res 1(2):151–171

86. Young W (1992) Role of calcium in central nervous system inju-ries. J Neurotrauma 9(Suppl 1):S9–S25

87. Stirling DP, Cummins K, Wayne Chen SR, Stys P (2014)Axoplasmic reticulum Ca(2+) release causes secondary degener-ation of spinal axons. Ann Neurol 75(2):220–229

88. Young W (1985) The role of calcium in spinal cord injury. CentNerv Syst Trauma 2(2):109–114

89. Pelisch N, Gomes C, Nally JM, Petruska JC, Stirling DP (2017)Differential expression of ryanodine receptor isoforms after spinalcord injury. Neurosci Lett 660:51–56

90. Villegas R,Martinez NW, Lillo J, Pihan P, HernandezD, Twiss JL,Court FA (2014) Calcium release from intra-axonal endoplasmicreticulum leads to axon degeneration through mitochondrial dys-function. J Neurosci 34(21):7179–7189

91. MortazaviMM,HarmonOA, Adeeb N,DeepA, Tubbs RS (2015)Treatment of spinal cord injury: a review of engineering usingneural and mesenchymal stem cells. Clin Anat 28(1):37–44

Neurosurg Rev

92. Fatima G, Sharma VP, Das SK, Mahdi AA (2015) Oxidativestress and antioxidative parameters in patients with spinalcord injury: implications in the pathogenesis of disease.Spinal Cord 53(1):3–6

93. Liu XZ, Xu XM, Hu R, du C, Zhang SX, McDonald JW, DongHX, Wu YJ, Fan GS, Jacquin MF, Hsu CY, Choi DW (1997)Neuronal and glial apoptosis after traumatic spinal cord injury. JNeurosci 17(14):5395–5406

94. Ma M (2013) Role of calpains in the injury-induced dysfunctionand degeneration of the mammalian axon. Neurobiol Dis 60:61–79

95. Mustafa AG, Wang JA, Carrico KM, Hall ED (2011)Pharmacological inhibition of lipid peroxidation attenuatescalpain-mediated cytoskeletal degradation after traumatic braininjury. J Neurochem 117(3):579–588

96. Kaplan AE (1977) Spinal cord influences on immune responsesdemonstrated following interruption of the spinal cord as a resultof injury. Zh Vopr Neirokhir Im N N Burdenko 6:40–43

97. Allan SM, Rothwell NJ (2003) Inflammation in central nervoussystem injury. Philos Trans R Soc Lond Ser B Biol Sci 358(1438):1669–1677

98. Schwab JM, Zhang Y, Kopp MA, Brommer B, Popovich PG(2014) The paradox of chronic neuroinflammation, systemic im-mune suppression, autoimmunity after traumatic chronic spinalcord injury. Exp Neurol 258:121–129

99. Plemel JR, Wee Yong V, Stirling DP (2014) Immune modulatorytherapies for spinal cord injury—past, present and future. ExpNeurol 258:91–104

100. Anthony DC, Couch Y (2014) The systemic response to CNSinjury. Exp Neurol 258:105–111

101. Zhou X, He X, Ren Y (2014) Function of microglia and macro-phages in secondary damage after spinal cord injury. NeuralRegen Res 9(20):1787–1795

102. Herrmann JE, Imura T, Song B, Qi J, Ao Y, Nguyen TK, KorsakRA, Takeda K, Akira S, SofroniewMV (2008) STAT3 is a criticalregulator of astrogliosis and scar formation after spinal cord injury.J Neurosci 28(28):7231–7243

103. Moriarty LJ, Borgens RB (1999) The effect of an applied electricfield on macrophage accumulation within the subacute spinal in-jury. Restor Neurol Neurosci 14(1):53–64

104. Ma M, Wei T, Boring L, Charo IF, Ransohoff RM, Jakeman LB(2002) Monocyte recruitment and myelin removal are delayedfollowing spinal cord injury in mice with CCR2 chemokine recep-tor deletion. J Neurosci Res 68(6):691–702

105. Zhang H, Trivedi A, Lee JU, Lohela M, Lee SM, Fandel TM,Werb Z, Noble-Haeusslein LJ (2011) Matrix metalloproteinase-9and stromal cell-derived factor-1 act synergistically to supportmigration of blood borne monocytes into the injured spinal cord.J Neurosci 31(44):15894–15903

106. Zu J, Wang Y, Xu G, Zhuang J, Gong H, Yan J (2014) Curcuminimproves the recovery of motor function and reduces spinal cordedema in a rat acute spinal cord injury model by inhibiting theJAK/STAT signaling pathway. Acta Histochem 116(8):1331–1336

107. Li X, Chen W, Sheng J, Cao D, Wang W (2014) Interleukin-6inhibits voltage-gated sodium channel activity of cultured rat spi-nal cord neurons. Acta Neuropsychiatr 26(3):170–177

108. Oropallo MA, Goenka R, Cancro MP (2014) Spinal cord injuryimpacts B cell production, homeostasis, and activation. SeminImmunol 26(5):421–427

109. Sofroniew MV, Vinters HV (2010) Astrocytes: biology and pa-thology. Acta Neuropathol 119(1):7–35

110. Wanner IB, Anderson MA, Song B, Levine J, Fernandez A,Gray-Thompson Z, Ao Y, Sofroniew MV (2013) Glial scarborders are formed by newly proliferated, elongated astro-cytes that interact to corral inflammatory and fibrotic cells

via STAT3-dependent mechanisms after spinal cord injury. JNeurosci 33(31):12870–12886

111. Kawaja MD, Gage FH (1991) Reactive astrocytes are substratesfor the growth of adult CNS axons in the presence of elevatedlevels of nerve growth factor. Neuron 7(6):1019–1030

112. Burda JE, Sofroniew MV (2014) Reactive gliosis and the multi-cellular response to CNS damage and disease. Neuron 81(2):229–248

113. Bush TG, Puvanachandra N, Horner CH, Polito A, Ostenfeld T,Svendsen CN, Mucke L, Johnson MH, Sofroniew MV (1999)Leukocyte infiltration, neuronal degeneration, and neurite out-growth after ablation of scar-forming, reactive astrocytes in adulttransgenic mice. Neuron 23(2):297–308

114. Drogemuller K, Helmuth U, Brunn A, Sakowicz-Burkiewicz M,Gutmann DH, Mueller W, Deckert M, Schluter D (2008)Astrocyte gp130 expression is critical for the control of toxoplas-ma encephalitis. J Immunol 181(4):2683–2693

115. Faulkner JR et al (2004) Reactive astrocytes protect tissue andpreserve function after spinal cord injury. J Neurosci 24(9):2143–2155

116. Silver J, Edwards MA, Levitt P (1993) Immunocytochemicaldemonstration of early appearing astroglial structures that formboundaries and pathways along axon tracts in the fetal brain. JComp Neurol 328(3):415–436

117. Williams RR, Henao M, Pearse DD, Bunge MB (2015)Permissive Schwann cell graft/spinal cord interfaces for axon re-generation. Cell Transplant 24(1):115–131

118. Zukor K, Belin S,Wang C, Keelan N,Wang X, He Z (2013) Shorthairpin RNA against PTEN enhances regenerative growth ofcorticospinal tract axons after spinal cord injury. J Neurosci33(39):15350–15361

119. Casha S, YuWR, FehlingsMG (2001) Oligodendroglial apoptosisoccurs along degenerating axons and is associated with FAS andp75 expression following spinal cord injury in the rat.Neuroscience 103(1):203–218

120. Spejo AB, Oliveira AL (2015) Synaptic rearrangement followingaxonal injury: old and new players. Neuropharmacology 96(Pt A):113–123

121. Ohtake Y, Li S (2015) Molecular mechanisms of scar-sourcedaxon growth inhibitors. Brain Res 1619:22–35

122. Dickendesher TL, Baldwin KT, Mironova YA, Koriyama Y,Raiker SJ, Askew KL, Wood A, Geoffroy CG, Zheng B,Liepmann CD, Katagiri Y, Benowitz LI, Geller HM, Giger RJ(2012) NgR1 and NgR3 are receptors for chondroitin sulfate pro-teoglycans. Nat Neurosci 15(5):703–712

123. Dergham P, Ellezam B, Essagian C, Avedissian H, Lubell WD,McKerracher L (2002) Rho signaling pathway targeted to promotespinal cord repair. J Neurosci 22(15):6570–6577

124. Fournier AE, Strittmatter SM (2001) Repulsive factors and axonregeneration in the CNS. Curr Opin Neurobiol 11(1):89–94

125. Fournier AE, Takizawa BT, Strittmatter SM (2003) Rho kinaseinhibition enhances axonal regeneration in the injured CNS. JNeurosci 23(4):1416–1423

126. Madura T, Yamashita T, Kubo T, Fujitani M, Hosokawa K,Tohyama M (2004) Activation of Rho in the injured axons fol-lowing spinal cord injury. EMBO Rep 5(4):412–417

127. Laurén J, Hu F, Chin J, Liao J, Airaksinen MS, Strittmatter SM(2007) Characterization of myelin ligand complexes with the neu-ronal Nogo-66 receptor family. J Biol Chem 282(8):5715–5725

128. Bareyre FM, Haudenschild B, Schwab ME (2002) Long-lastingsprouting and gene expression changes induced by the monoclo-nal antibody IN-1 in the adult spinal cord. J Neurosci 22(16):7097–7110

129. Bregman BS, Kunkel-Bagden E, Schnell L, Dai HN, Gao D,Schwab ME (1995) Recovery from spinal cord injury mediated

Neurosurg Rev

by antibodies to neurite growth inhibitors. Nature 378(6556):498–501

130. Fehlings MG, Theodore N, Harrop J, Maurais G, Kuntz C,Shaffrey CI, Kwon BK, Chapman J, Yee A, Tighe A,McKerracher L (2011) A phase I/IIa clinical trial of a recombinantRho protein antagonist in acute spinal cord injury. J Neurotrauma28(5):787–796

131. Squair JW, Bélanger LM, Tsang A, Ritchie L, Mac-Thiong JM,Parent S, Christie S, Bailey C, Dhall S, Street J, Ailon T, PaquetteS, Dea N, Fisher CG, Dvorak MF, West CR, Kwon BK (2017)Spinal cord perfusion pressure predicts neurologic recovery inacute spinal cord injury. Neurology 89(16):1660–1667

132. Satyarthee GD (2018) Reader response: spinal cord perfusionpressure predicts neurologic recovery in acute spinal cord injury.Neurology 90(19):904

133. Tykocki T, Poniatowski Ł, Czyż M, Koziara M, Wynne-Jones G(2017) Intraspinal pressure monitoring and extensive duroplasty inthe acute phase of traumatic spinal cord injury: a systematic re-view. World Neurosurg 105:145–152

134. Donati ARC, Shokur S, Morya E, Campos DSF, Moioli RC, GittiCM, Augusto PB, Tripodi S, Pires CG, Pereira GA, Brasil FL,Gallo S, Lin AA, Takigami AK, Aratanha MA, Joshi S, BleulerH, Cheng G, Rudolph A, Nicolelis MAL (2016) Long-term train-ing with a brain-machine interface-based gait protocol inducespartial neurological recovery in paraplegic patients. Sci Rep 6:30383

135. Guru K, Mailis A, Ashby P, Vanderlinden G (1987) Postsynapticpotentials in motoneurons caused by spinal cord stimulation inhumans. Electroencephalogr Clin Neurophysiol 66(3):275–280

136. Dimitrijevic MR, Gerasimenko Y, PinterMM (1998) Evidence fora spinal central pattern generator in humans. Ann N YAcad Sci860(1):360–376

137. Harkema S, Gerasimenko Y, Hodes J, Burdick J, Angeli C, ChenY, Ferreira C, Willhite A, Rejc E, Grossman RG, Edgerton VR(2011) Effect of epidural stimulation of the lumbosacral spinalcord on voluntary movement, standing, and assisted stepping aftermotor complete paraplegia: a case study. Lancet 377(9781):1938–1947

138. Gerasimenko YP, Makarovskii A, Nikitin O (2002) Control oflocomotor activity in humans and animals in the absence ofsupraspinal influences. Neurosci Behav Physiol 32(4):417–423

139. Jilge B, Minassian K, Rattay F, Pinter MM, Gerstenbrand F,Binder H, Dimitrijevic MR (2004) Initiating extension of the low-er limbs in subjects with complete spinal cord injury by epidurallumbar cord stimulation. Exp Brain Res 154(3):308–326

140. Garara B, Wood A, Marcus HJ, Tsang K, Wilson MH, Khan M(2016) Intramuscular diaphragmatic stimulation for patients withtraumatic high cervical injuries and ventilator dependent respira-tory failure: a systematic review of safety and effectiveness. Injury47(3):539–544

141. Kaufman MR et al (2015) Diaphragmatic reinnervation inventilator-dependent patients with cervical spinal cord injuryand concomitant phrenic nerve lesions using simultaneous nervetransfers and implantable neurostimulators. J Reconstr Microsurg31(5):391–395

142. Posluszny JA Jr et al (2014) Multicenter review of diaphragmpacing in spinal cord injury: successful not only in weaning fromventilators but also in bridging to independent respiration. JTrauma Acute Care Surg 76(2):303–309 discussion 309–10

143. Dalal K, DiMarco AF (2014) Diaphragmatic pacing in spinal cordinjury. Phys Med Rehabil Clin N Am 25(3):619–629 viii

144. Nandra KS, Harari M, Price TP, Greaney PJ,WeinsteinMS (2017)Successful reinnervation of the diaphragm after intercostal tophrenic nerve neurotization in patients with high spinal cord inju-ry. Ann Plast Surg 79(2):180–182

145. Krieger LM, Krieger AJ (2000) The intercostal to phrenic nervetransfer: an effective means of reanimating the diaphragm in pa-tients with high cervical spine injury. Plast Reconstr Surg 105(4):1255–1261

146. Martens FM, Heesakkers JP (2011) Clinical results of abrindley procedure: sacral anterior root stimulation in combi-nation with a rhizotomy of the dorsal roots. Ther Adv Urol2011:709708

147. Van Kerrebroeck PE et al (1996) Results of the treatment of neu-rogenic bladder dysfunction in spinal cord injury by sacral poste-rior root rhizotomy and anterior sacral root stimulation. J Urol155(4):1378–1381

148. Brindley GS, Polkey CE, Rushton DN, Cardozo L (1986) Sacralanterior root stimulators for bladder control in paraplegia: the first50 cases. J Neurol Neurosurg Psychiatry 49(10):1104–1114

149. Brindley GS, Polkey CE, Rushton DN (1982) Sacral anterior rootstimulators for bladder control in paraplegia. Paraplegia 20(6):365–381

150. Brindley GS (1994) The first 500 patients with sacral anterior rootstimulator implants: general description. Paraplegia 32(12):795–805

151. MacDonagh RP, Sun WM, Smallwood R, Forster D, Read NW(1990) Control of defecation in patients with spinal injuries bystimulation of sacral anterior nerve roots. BMJ 300(6738):1494–1497

152. Hansen J et al (2005) Treatment of neurogenic detrusor overactiv-ity in spinal cord injured patients by conditional electrical stimu-lation. J Urol 173(6):2035–2039

153. Hamel O, Perrouin-Verbe B, Robert R (2004) Brindley techniquewith intradural deafferentation and extradural implantation by asingle sacral laminectomy. Neurochirurgie 50(6):661–666

154. Sauerwein D, Ingunza W, Fischer J, Madersbacher H, Polkey CE,Brindley GS, Colombel P, Teddy P (1990) Extradural implantationof sacral anterior root stimulators. J Neurol Neurosurg Psychiatry53(8):681–684

155. Possover M (2014) The LION procedure to the pelvic nerves fortreatment of urinary and faecal disorders. Surg Technol Int 24:225–230

156. Possover M (2009) The sacral LION procedure for recovery ofbladder/rectum/sexual functions in paraplegic patients after ex-plantation of a previous Finetech-Brindley controller. J MinimInvasive Gynecol 16(1):98–101

157. Onders RP, Elmo MJ, Khansarinia S, Bowman B, Yee J, Road J,Bass B, Dunkin B, Ingvarsson PE, Oddsdóttir M (2009) Completeworldwide operative experience in laparoscopic diaphragm pac-ing: results and differences in spinal cord injured patients andamyotrophic lateral sclerosis patients. Surg Endosc 23(7):1433–1440

158. Tetzlaff W, Okon EB, Karimi-Abdolrezaee S, Hill CE, SparlingJS, Plemel JR, Plunet WT, Tsai EC, Baptiste D, Smithson LJ,KawajaMD, FehlingsMG,KwonBK (2011) A systematic reviewof cellular transplantation therapies for spinal cord injury. JNeurotrauma 28(8):1611–1682

159. Nagoshi N, Nakashima H, Fehlings MG (2015) Riluzole as aneuroprotective drug for spinal cord injury: from bench to bedside.Molecules 20(5):7775–7789

160. Adeeb N, Mortazavi MM (2014) The role of FGF2 in spinal cordtrauma and regeneration research. Brain Behav 4(2):105–107

161. Satkunendrarajah K, Nassiri F, Karadimas SK, Lip A, Yao G,Fehlings MG (2016) Riluzole promotes motor and respiratoryrecovery associated with enhanced neuronal survival and functionfollowing high cervical spinal hemisection. Exp Neurol 276:59–71

162. Fehlings MG, Nakashima H, Nagoshi N, Chow DSL, GrossmanRG, Kopjar B (2016) Rationale, design and critical end points forthe Riluzole in Acute Spinal Cord Injury Study (RISCIS): a

Neurosurg Rev

randomized, double-blinded, placebo-controlled parallel multi-center trial. Spinal Cord 54(1):8–15

163. Fan X et al (2017) Stem cell transplantation for spinal cord injury:a meta-analysis of treatment effectiveness and safety. NeuralRegen Res 12(5):815–825

164. Kothari M, Goel A (2013) The stem cell promise: the future ofstemocytology. J Craniovertebr Junction Spine 4(2):47–48

165. Svendsen C (2002) Stem cells: hype or hope? Drug Discov Today7(8):455–456

166. Anna Z, Katarzyna JW (2017) Therapeutic potential of olfactoryensheathing cells and mesenchymal stem cells in spinal cord inju-ries. Stem Cells Int 2017:3978595

167. Park JH, Kim DY, Sung IY, Choi GH, Jeon MH, Kim KK, JeonSR (2012) Long-term results of spinal cord injury therapy usingmesenchymal stem cells derived from bone marrow in humans.Neurosurgery 70(5):1238–1247 discussion 1247

168. Goel A (2016) Stem cell therapy in spinal cord injury: hollowpromise or promising science? J Craniovertebr Junction Spine7(2):121–126

Neurosurg Rev