Introduction
Acute spinal cord injury (SCI) is considered one of the most life-threatening traumatic conditions, frequently resulting in permanent neurological deficits, severe functional disability, and increased mortality [1–3]. The neurological outcomes after SCI are determined not only by the magnitude of the initial mechanical insult but also by a complex cascade of secondary injury mechanisms, including ischemia, inflammation, excitotoxicity, oxidative stress, and apoptotic cell death. These secondary processes evolve over hours to days following injuries and represent critical targets for early therapeutic intervention [4].
Thus, the contemporary management of SCI prioritizes rapid neurological assessment, the prevention of secondary insults, hemodynamic optimization, and timely spinal cord decompression and stabilization. Growing clinical evidence supports early decompressive surgery, particularly within 24 hours after injury, as a potentially modifiable factor associated with improved neurological recovery, although the optimal timing in ultra-early windows and patient selection should be further investigated [5,6].
Although pharmacologic neuroprotection has long been investigated in SCI, it remains controversial. High-dose methylprednisolone was historically recommended based on early randomized trials, but its routine use has declined due to concerns about limited neurological benefits and increased risk of complications [7,8]. Novel neuroprotective agents, such as riluzole and minocycline, have shown promising signals in clinical trials, but definitive evidence supporting their widespread adoption is still lacking [9,10].
Beyond acute stabilization and decompression, novel therapeutic approaches, including neuroregeneration and neuromodulation, have gained increasing attention, emphasizing the evolving and multidisciplinary nature of SCI management [4,5,11]. Given the rapidly changing landscape of SCI treatment and the absence of universally accepted standards of care, a comprehensive synthesis of current evidence is needed. This review aimed to summarize the contemporary management strategies for SCI, critically evaluate ongoing controversies, and highlight novel therapies that may shape future clinical practice (Table 1).
Pathophysiology and Secondary Injury Cascade
The pathological mechanisms of SCI are classified into primary and secondary injury processes. Primary injury is the immediate mechanical disruption of neural and vascular structures at the time of trauma and is largely irreversible. On the other hand, secondary injury comprises a complex and time-dependent cascade of molecular and cellular events that evolve over hours to days after the initial insult and critically influence neurological outcomes [8,12].
Secondary injury is initiated by microvascular disruption, resulting in hemorrhage, edema, and impaired spinal cord perfusion. Subsequent ischemia and hypoxia exacerbate tissue damage through mitochondrial dysfunction and energy failure, while the breakdown of the blood–spinal cord barrier facilitates the infiltration of inflammatory mediators into the injured cord [8,12]. Moreover, excitotoxicity represents a key mechanism of secondary neuronal injury. Excessive glutamate release and the sustained activation of ionotropic receptors result in pathological calcium influx, oxidative stress, and the activation of apoptotic signaling pathways, contributing to neuronal and oligodendroglial cell loss and impaired axonal conduction [12]. Moreover, inflammatory responses further amplify secondary injuries through the activation of resident microglia and the recruitment of peripheral immune cells. Although certain inflammatory processes may contribute to debris clearance and tissue remodeling, excessive or prolonged inflammation promotes tissue destruction and compromises neurological recovery [13].
Since secondary injury mechanisms remain active beyond the initial traumatic event, they represent major therapeutic targets in SCI. Contemporary management strategies, such as hemodynamic optimization and early surgical decompression, are primarily aimed at reducing secondary damage rather than reversing the primary injury. Furthermore, understanding these mechanisms provides the biological rationale for time-sensitive and multimodal interventions in SCI [8,13].
Early Clinical Evaluation and Prognostic Markers
An early and standardized neurological assessment is fundamental in the management of SCI, as it guides initial treatment decisions, establishes baseline injury severity, and provides the framework for prognostication and longitudinal outcome assessments. The International Standards for Neurological Classification of SCI (ISNCSCI), which was developed by the American Spinal Injury Association (ASIA) and the International Spinal Cord Society, remains the global reference standard for the clinical assessment and classification of SCI [14–16]. The ISNCSCI framework integrates the ASIA Impairment Scale (AIS), detailed motor and sensory examinations, and the determination of the neurological level of injury. Since its initial introduction in 1982, the ISNCSCI has undergone multiple revisions to improve clarity, inter-rater reliability, and clinical applicability. The most recent major revision, released in 2019, replaced the previous 2015 edition and addressed several limitations in acute clinical settings [14,15].
A central feature of the 2019 ISNCSCI update is the explicit documentation of clinical judgment when the neurological examination findings are influenced by factors unrelated to SCI. In acute trauma care, motor or sensory testing may be confounded by pain-limited effort, immobilization due to associated fractures, reduced patient cooperation, sedation, or concomitant peripheral nerve injury. Moreover, the revised standards encourage clinicians to clearly document these non-SCI–related influences rather than attributing all deficits to spinal cord pathology. This approach reduces the risk of erroneous AIS grading and improves the interpretability of baseline and follow-up evaluations. Importantly, the inappropriate use of the “not testable” (NT) designation may result in the misclassification of injury completeness. When the NT scores are assigned without a clear explanation, the baseline AIS grade may be poorly determined. In such cases, apparent neurological “recovery” on subsequent assessments may reflect the correction of the initial classification rather than true neurological improvement. This issue is particularly relevant in the early postinjury period, when spinal shock, intoxication, or critical illness frequently confound the initial evaluation. Accordingly, repeat ISNCSCI assessments after physiological stabilization are strongly recommended in both clinical practice and research settings [14–16].
Magnetic resonance imaging (MRI) provides complementary prognostic information by characterizing intramedullary injury patterns that are not captured by clinical examination alone. MRI features that are consistently associated with poorer neurological outcomes include intramedullary hemorrhage, extensive cord edema, cord swelling, and severe spinal cord compression. Of these, intramedullary hemorrhage is strongly associated with unfavorable neurological recovery and widely regarded as a high-risk imaging biomarker in SCI [17,18].
The Brain and Spinal Injury Center (BASIC) score is a practical MRI-based grading system derived from axial T2-weighted images that was designed to stratify injury severity based on the transverse extent of spinal cord signal abnormality. The grading system of the BASIC score is as follows: grade 0, no abnormal intramedullary signal; grade 1, focal T2 hyperintensity involving the gray matter; grade 2, T2 hyperintensity extending into the white matter but not involving the entire cord cross-section; grade 3, diffuse T2 hyperintensity involving the entire cord cross-section; and grade 4, T2 hyperintensity with associated intramedullary hemorrhage (Fig. 1).
Clinically, increasing BASIC grades are associated with progressively worse neurological outcomes. Patients with BASIC grades of 0–1 generally present with a higher likelihood of significant neurological recovery, whereas those with BASIC grades of 3–4 are strongly associated with severe initial deficits and limited recovery potential. As such, the BASIC score provides a pragmatic adjunct to ISNCSCI classification, supporting early prognostic stratification, the intensity of monitoring, and informed patient counseling [19].
Acute Management Strategies
Hemodynamic optimization and pharmacologic support
The optimization of spinal cord perfusion remains a fundamental component of SCI management. Practice guidelines and consensus recommendations derived from neurosurgical societies suggest maintaining a mean arterial pressure (MAP) between 85 mm Hg and 90 mm Hg for up to 7 days following acute cervical SCI to minimize secondary ischemic injuries and support spinal cord perfusion [20]. This reflects the impaired autoregulatory capacity of the injured spinal cord, which renders neural tissue particularly vulnerable to systemic hypotension.
Observational clinical studies have shown that higher MAP values, as well as reduced exposure to MAP levels below 85 mm Hg, are associated with improved neurological recovery, although a causal relationship has not been established [21,22]. Accordingly, MAP augmentation is widely adopted in contemporary neurocritical care practice, with continuous invasive arterial blood pressure monitoring recommended during the acute phase.
In many patients, MAP targets cannot be achieved with volume resuscitation alone and commonly require pharmacologic vasopressor support. Norepinephrine is commonly considered a first-line agent due to its balanced α-adrenergic vasoconstrictive effect with modest β-adrenergic activity, allowing for an effective MAP augmentation while preserving cardiac output and spinal cord blood flow. Phenylephrine, a pure α-agonist, may be used as an alternative but can be associated with reflex bradycardia and reduced cardiac output, particularly in patients with high cervical injuries or pre-existing cardiac dysfunction. Dopamine has also been historically used but is increasingly avoided because of the higher rates of arrhythmias and cardiogenic complications associated with its use, especially in older patients and those with cervical SCI. Thus, the selection of vasopressor agents should be individualized, taking into account the baseline heart rate, cardiac reserve, and autonomic instability [23].
Surgical timing in SCI
Surgical decompression is among the few potentially modifiable interventions that may influence neurological recovery after SCI. The Surgical Timing in Acute Spinal Cord Injury Study reported that decompression performed within 24 hours can be safely achieved and was associated with a higher likelihood of significant neurological improvement (e.g., ≥2 AIS grade improvement at 6 months) compared with later surgery [24].
More recently, ultra-early decompression, most commonly defined as surgery within ≤12 hours after injury, has gained interest in clinical practice. Observational cohorts have reported that ultra-early surgery (≤12 hours) tends to be associated with higher rates of AIS conversion or neurological improvements compared with later time windows [25]. However, other cohorts—particularly those using MRI-confirmed decompression and stratifying by injury severity—have emphasized that the effect of timing on long-term neurological recovery remains uncertain and may be confounded by injury severity, imaging biomarkers, and selection effects [26]. Systematic reviews/meta-analyses have suggested the potential advantages of ultra-early (≤12 hours) decompression in specific subgroups (e.g., complete cervical SCI); however, the evidence base is still dominated by nonrandomized studies [26]. Similar questions are being assessed in thoracolumbar SCI, wherein meta-analytic data are used to evaluate even earlier windows (e.g., ≤8 hours), although heterogeneity remains substantial [27].
Damage control spine surgery and percutaneous fixation (Fig. 2)
In patients with polytrauma or physiologically fragile conditions, some centers extrapolate principles from damage control orthopedics to spine trauma, favoring abbreviated stabilization and/or staged procedures to reduce operative time and physiologic stress. Although this approach has been covered in dedicated reviews focused on damage control concepts applied to spinal trauma, high-level SCI-specific outcome validation remains limited [27].
Within this framework, percutaneous pedicle screw fixation is increasingly used as an “abbreviated stabilization” option. In a retrospective polytrauma cohort, percutaneous fixation was associated with better clinical outcomes than open fixation, despite worse baseline clinical conditions, supporting its potential role when physiologic reserve is limited [28]. Consensus-style work on damage control orthopedics also identifies percutaneous interventions as beneficial in spinal trauma contexts, reinforcing its contemporary role as a less invasive stabilization strategy within staged management [29].
Anterior surgical approaches versus posterior surgical approaches and neurological recovery
Approach selection should be individualized based on the predominant site of compression, fracture–dislocation morphology, alignment goals, and overall patient physiology. With regard to cervical facet injuries and unstable cervical injuries, randomized trials comparing anterior versus posterior stabilization have reported that both approaches are valid options, without consistent superiority of one approach for neurological recovery as a primary outcome; the secondary outcomes may differ by injury type and surgical goals [30,31]. A recent meta-analysis comparing anterior and posterior approaches in cervical SCI has similarly reported that clear evidence favoring one approach over the other for neurological outcomes remains limited, supporting the interpretation that neurological recovery is more strongly influenced by the adequacy/timing of decompression and stability restoration rather than the approach itself [32].
The long-term observational data in specific entities (e.g., traumatic cervical dislocation with SCI) suggest that anterior approaches may better restore or preserve alignment in some contexts, but such findings should not be generalized as showing superior neurological recovery [33].
Controversial Medical Therapies
Steroids: NASCIS trials versus contemporary guidelines
The high-dose administration of methylprednisolone remains one of the most debated pharmacologic interventions in SCI. The National Acute SCI Studies (NASCIS) II and III reported modest neurological improvements when methylprednisolone was administered within a narrow therapeutic window after injury, leading to its widespread clinical adoption during the 1990s [7,34]. However, these reported advantages were largely derived from post hoc subgroup analyses, and the absolute magnitude of neurological improvement was small.
Subsequent re-analyses and guideline updates have raised significant concerns regarding the NASCIS trials, including methodological limitations, statistical interpretation, and an increased risk of serious adverse events such as infection, gastrointestinal bleeding, and metabolic complications [35]. As a result, contemporary clinical practice guidelines no longer recommend the routine administration of high-dose steroids in the treatment of SCI. Current recommendations emphasize that methylprednisolone should not be considered a standard treatment and, if used at all, should be offered only after a careful discussion of its potential risks and uncertain benefits [36–38].
Riluzole and other neuroprotective pharmacologic agents
In an effort to identify safer and more effective neuroprotective strategies, the effectiveness of several pharmacologic agents in the treatment of SCI has been investigated. Riluzole, a sodium channel blocker with antiglutamatergic properties, showed acceptable safety and encouraging signals of neurological improvement in early-phase clinical trials [39]. These findings led to the multicenter Riluzole in SCI Study (RISCIS) randomized controlled trial, which investigated the treatment efficacy of riluzole in patients with acute cervical SCI. Although the subgroup analyses suggested potential neurological benefits in selected patients (acute incomplete cervical SCI AIS B–C, particularly AIS C, treated within 12 hours of injury), the primary endpoint of the RISCIS trial was not met, and riluzole has not been adopted as a standard therapy for acute SCI [40]. These results emphasize the challenges of translating promising preclinical neuroprotective mechanisms into consistent clinical efficacy.
The effectiveness of minocycline, a tetracycline antibiotic with anti-inflammatory and antiapoptotic properties, in the treatment of acute SCI has also been assessed. A phase II randomized trial showed that minocycline was safe and suggested its possible neurological benefits, particularly in the treatment of cervical SCI. However, the study was not powered to establish definitive efficacy [9]. To date, larger confirmatory trials are limited, and the use of minocycline remains an investigational therapy. Overall, despite the extensive preclinical rationale, no pharmacologic agent beyond supportive care has shown sufficient efficacy to warrant routine clinical use in acute SCI.
Therapeutic hypothermia
Therapeutic hypothermia has been investigated as a neuroprotective strategy based on experimental evidence suggesting the attenuation of secondary injury mechanisms. Early clinical studies and small prospective series have confirmed the feasibility of inducing modest systemic hypothermia in selected patients with SCI [41]. However, subsequent clinical experience has revealed significant logistical challenges and potential complications, including coagulopathy, infection, and cardiopulmonary instability.
To date, clinical trials assessing hypothermia in SCI have been compromised by small sample sizes, heterogeneous protocols, and inconsistent outcome measures. No high-level evidence has established a clear or reproducible neurological benefit of hypothermia in this setting [42]. Consequently, therapeutic hypothermia remains experimental and is not recommended outside controlled clinical trials.
Future Directions in SCI Therapy
Cellular therapy
Cell-based therapies represent one of the most actively investigated strategies for promoting neurological recovery after SCI. The proposed mechanisms include trophic support, immunomodulation, remyelination, and the facilitation of axonal sprouting rather than direct neuronal replacement, and most clinical experience remains limited to early-phase studies focused primarily on feasibility and safety [43,44].
In Japan, the clinical translation of intravenous autologous bone marrow-derived mesenchymal stromal/stem cells (MSCs) expanded in autologous serum has been pioneered by Honmou et al. [45] at Sapporo Medical University. In a prospective case series involving patients with traumatic SCI, the intravenous infusion of auto-serum–expanded autologous MSCs was reported to be feasible and generally well tolerated, with exploratory neurological and functional changes observed during follow-up. However, the uncontrolled design and small sample size preclude definitive conclusions regarding efficacy [45].
Based on the accumulated safety data and exploratory effectiveness signals, Japan’s regulatory authority granted conditional and time-limited approval to STEMIRAC (Stemirac for Injection) under the national regenerative medicine framework, requiring strict post-marketing surveillance and confirmatory data collection. This regulatory pathway highlights both the potential for accelerated clinical access and the continuing need for high-quality randomized trials with harmonized endpoints.
Beyond MSCs, other cellular approaches—including neural stem/progenitor cells, Schwann cells, and olfactory ensheathing cells—have entered early clinical testing. While feasibility and relative safety have been proven, reproducible and clinically significant neurological recovery has not yet been consistently shown.
Biologics and neuroregeneration
Biologic approaches target molecular pathways that drive the secondary injury cascade and inhibit axonal regrowth (e.g., inflammation, myelin-associated inhibitors, and intracellular growth-inhibitory signaling). A well-known example is Rho/ROCK pathway inhibition. However, a randomized controlled trial of the local delivery of the Rho inhibitor VX-210 (Cethrin) in acute cervical SCI was discontinued early for futility and did not meet its primary efficacy endpoint, highlighting the translational gap between strong preclinical rationale and clinical efficacy [46].
These negative or neutral late-stage results have shifted attention toward combination paradigms—associating biologics with optimized decompression, rehabilitation, neuromodulation, and improved patient stratification—rather than expecting a single agent to produce large neurological gains across heterogeneous SCI phenotypes [3,47].
Neuromodulation and spinal cord stimulation
Neuromodulation has been considered one of the most promising avenues for functional restoration after SCI. Epidural electrical stimulation has shown that spinal motor circuits retain latent capacity for activation even in individuals with chronic and clinically complete SCI when stimulation is paired with task-specific training. Subsequent work has shown that neuromodulating spinal circuitry could allow voluntary movement even in individuals diagnosed with clinically complete paralysis, reframing recovery potential as partially “state-dependent” and modifiable through stimulation and training [48,49].
More recently, a pivotal multicenter trial of non-invasive cervical spinal cord stimulation (ARC-EX; Up-LIFT) reported clinically significant improvements in upper limb strength and function for a substantial proportion of participants with chronic cervical SCI when stimulation was paired with structured rehabilitation [50].
In parallel, high-profile “digital bridge” approaches associating cortical signals with spinal stimulation have shown feasibility for restoring overground walking in selected individuals with chronic tetraplegia, suggesting that neurotechnology platforms may be individualized and adaptive in the future [51].
Artificial intelligence prognostication models
As trial designs increasingly emphasize patient stratification and individualized counseling, artificial intelligence/machine learning (ML) approaches are being developed to predict neurological and functional outcomes using registry-scale clinical data and multimodal features. An unsupervised machine learning model derived from the Rick Hansen SCI Registry (RHSCIR) was developed to prognosticate walking-related outcomes, suggesting how large registries can support clinically useful prediction tools beyond traditional linear models [52].
Additional ML work has investigated the prediction of AIS grades and recovery trajectories in newly injured patients. However, broad external validation and transparent reporting (including calibration and clinical utility) remain essential before routine deployment [53].
Conclusions
SCI remains a life-threatening condition that is associated with significant and often lifelong consequences. Despite the advances in trauma systems and surgical techniques, neurological recovery after SCI is still largely determined by early pathophysiological events and the effectiveness of timely, coordinated interventions. Thus, contemporary management emphasizes rapid neurological assessment, the prevention of secondary injury, hemodynamic optimization, and early spinal cord decompression and stabilization as the core pillars of acute care.
Growing clinical evidence has supported early surgical decompression, particularly within 24 hours after injury, as a potentially modifiable factor associated with improved neurological outcomes. However, the optimal role of ultra-early decompression, approach selection, and staged strategies in physiologically unstable patients continues to improve, emphasizing the importance of individualized, physiology-driven decision-making rather than rigid time-based mandates.
Pharmacologic neuroprotection remains an area of active investigation and ongoing controversy. Historical reliance on the high-dose administration of methylprednisolone has declined due to limited efficacy and safety concerns, while newer agents, such as riluzole and minocycline, have shown only exploratory or subgroup-specific signals without definitive evidence to support their routine use. Similarly, therapeutic hypothermia has not yet shown reproducible clinical advantages and remains experimental.
Novel therapies—including cellular transplantation, biologic modulation, neuromodulation, and data-driven prognostication—provide promising avenues to complement established acute management strategies. Experiences such as the Honmou–STEMIRAC program show both the potential and the challenges of translating regenerative therapies into clinical practice, highlighting the need for a rigorous trial design, harmonized outcome measures, and long-term safety surveillance. Advances in neuromodulation and artificial intelligence further suggest that functional recovery and prognostication may increasingly rely on personalized, multimodal approaches.
In summary, the effective management of SCI requires an integrated, multidisciplinary framework that combines evidence-based acute care with a careful evaluation of novel therapies. Its continued progress will depend on collaborative multicenter research, refined patient stratification, and the balanced interpretation of evolving evidence to improve outcomes for individuals with SCI.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
