▷ Thoracolumbar trauma: integrating imaging, stability, and classification into surgical decision-making - Ocronos

X (Twitter) Facebook Pinterest LinkedIn Email Reddit WhatsApp

Síguenos en Pinterest

Incluido en la revista Ocronos. Vol. IX. N.º 11–Noviembre 2026.

Pág. Inicial: Vol. IX; N.º 11: 27

Autor principal (primer firmante): Alberto Maldonado Olvera

Fecha recepción: 05/10/2026

Fecha aceptación: 01/11/2026

Ref.: Ocronos. 2026;9(11): 27

https://doi.org/10.58842/HPUO3216

Article type: Narrative review / literature review

Autores:

Maldonado Olvera, Alberto1; Affiliations: 1 Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, Universidad Autónoma de Nuevo León (UANL), Monterrey, N.L., Mexico; Professional category: Alberto Maldonado Olvera, MD, Orthopedics and Traumatology; ORCID: Alberto Maldonado Olvera: 0009-0003-4883-2869; Corresponding author: Alberto Maldonado Olvera, MD; Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, UANL, Av. Francisco I. Madero y Av. Gonzalitos s/n, Col. Mitras Centro, C.P. 64460, Monterrey, N.L., Mexico.
Valeriano Michel, Yazmin2; Affiliations: 2 Department of Radiology and Imaging, Hospital Zambrano-Hellion, Tecnológico de Monterrey (ITESM), San Pedro Garza García, N.L., Mexico; Professional category: Yazmin Valeriano Michel, MD, Radiology and Imaging; ORCID: Yazmin Valeriano Michel: 0009-0006-1015-0151
Reyes Fernández, Pedro Martín3; Affiliations: 3 Professor at Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, Universidad Autónoma de Nuevo León (UANL), Monterrey, N.L., Mexico; Professional category: Pedro Martín Reyes Fernández, MD, PhD, Orthopedics and Traumatology; ORCID: Pedro Martín Reyes Fernández: 0009-0006-8576-9151
González Pérez, Aldo Domingo1; Affiliations: 1 Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, Universidad Autónoma de Nuevo León (UANL), Monterrey, N.L., Mexico; Professional category: Aldo Domingo González Pérez, MD, Orthopedics and Traumatology; ORCID: Aldo Domingo González Pérez: 0009-0004-2840-504X
Acosta Olivo, Carlos Alberto3; Affiliations: 3 Professor at Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, Universidad Autónoma de Nuevo León (UANL), Monterrey, N.L., Mexico; Professional category: Carlos Alberto Acosta Olivo, MD, PhD, Orthopedics and Traumatology; ORCID: Carlos Alberto Acosta Olivo: 0000-0002-2025-1865
Peña Martínez, Víctor Manuel4; Affiliations: 4 Chief of Department of Orthopedics and Traumatology, Hospital Universitario “Dr. José Eleuterio González”, Universidad Autónoma de Nuevo León (UANL), Monterrey, N.L., Mexico; Professional category: Víctor Manuel Peña Martínez, MD, PhD, Orthopedics and Traumatology; ORCID: Víctor Manuel Peña Martínez: 0000-0003-2924-1329.

Number of tables: 5. Number of figures: 3.

Author contributions: AMO: conceptualization, literature search and selection, writing of the original draft, and preparation of tables. YVM: conceptualization, analysis and interpretation of imaging findings, preparation of figures, and critical revision of the manuscript. PMRF: supervision, clinical interpretation, and critical revision for important intellectual content. ADGP: literature search and writing of the initial draft. CAAO: supervision, clinical interpretation, and critical revision for important intellectual content. VMPM: supervision, clinical interpretation, and critical revision for important intellectual content. All authors approved the final version of the manuscript.

Abstract

Objective: To synthesize the evidence on how mechanical stability, the Thoracolumbar Injury Classification and Severity Score (TLICS), the AO (Arbeitsgemeinschaft für Osteosynthesefragen) Spine classification, and imaging findings come together in decision-making for thoracolumbar trauma.
Methods: We conducted a narrative review based on a targeted search of PubMed / MEDLINE, Google Scholar, AO Spine / AO Foundation sources, and the reference lists of key publications, updated to September 28, 2026.
Results: Forty-three potentially relevant publications were identified; after screening of titles, abstracts, and full texts when available, 29 were included in the final narrative synthesis. Computed tomography is the reference modality for the initial assessment of morphology and alignment. Posterior wall involvement or canal narrowing alone does not equate to instability or to an indication for surgery. Magnetic resonance imaging should be used selectively when ligamentous or neural injury is suspected, when a neurological deficit is present, or when clinical and CT findings are discordant. Frank disruption of the posterior ligamentous complex raises concern for instability, whereas isolated edema on fluid-sensitive sequences should not be automatically equated with complete rupture. In burst fractures without neurological deficit, current evidence supports an individualized decision.
Conclusions: Treatment decisions should integrate morphology, neurological status, stability, ligamentous integrity, and patient-specific factors.

Keywords: thoracolumbar trauma; spinal stability; surgical decision-making; AO Spine classification; posterior ligamentous complex

Introduction

Thoracolumbar spine trauma is a clinically relevant subset of spinal injuries because it involves a biomechanical transition zone between the relatively rigid thoracic spine and the more mobile lumbar region. Injuries range from generally stable compression fractures to tension band injuries and highly unstable translational patterns. Although ligamentous tension band injuries and translational injuries usually favor surgical stabilization, some purely transosseous flexion-distraction patterns may be suitable for nonoperative management in selected patients. ^1–3 In clinical practice, the challenge is not only to identify a fracture but also to characterize its morphology, mechanical stability, and neurological impact.

The concept of stability has historically been central to the evaluation of thoracolumbar trauma. Denis’ three-column model provided a useful framework for understanding how disruption of supporting and tension-bearing structures may predispose to progressive deformity, neurological deterioration, or failure of conservative treatment. ⁴ Although modern systems have moved beyond this model, its underlying principle remains clinically relevant: surgical decision-making depends on recognizing which injuries can remain stable under physiological loads and which require surgical stabilization.

Current management relies on a multidimensional assessment that integrates injury morphology, neurological status, integrity of the posterior ligamentous complex (PLC), sagittal alignment, and patient-specific factors. The Thoracolumbar Injury Classification and Severity Score (TLICS) was developed to standardize this assessment by incorporating morphology, neurological status, and PLC integrity. ^5,6 Similarly, the AO (Arbeitsgemeinschaft für Osteosynthesefragen) Spine thoracolumbar classification of the AO Foundation provides an internationally recognized framework based on morphology, neurological status, and clinical modifiers, supported by a scoring system and a treatment algorithm that guide decision-making. ^1,7–9

Imaging is essential because classification systems and treatment algorithms depend on an accurate characterization of osseous, ligamentous, and neurological injury. Computed tomography (CT) is the main initial modality and defines morphology, vertebral height loss, posterior wall involvement, retropulsion, spinal canal narrowing, posterior element fractures, facet alignment, and translational or rotational displacement. ^2,10–12 Magnetic resonance imaging (MRI) provides complementary information on the PLC, spinal cord, conus medullaris, cauda equina, epidural hematoma, traumatic disc injury, and occult soft-tissue damage; it is most useful in selected cases in which these findings may change the classification or the treatment strategy. ^12–18

Among imaging findings, PLC disruption is particularly relevant because it may change the injury classification and raise concern for mechanical instability. ^13–16,18 However, isolated ligamentous edema on MRI should not be automatically equated with complete mechanical rupture. Likewise, posterior wall involvement or spinal canal narrowing alone does not constitute an indication for surgery. Scenarios such as burst fractures without neurological deficit and suspected ligamentous injury with equivocal CT findings require the integration of clinical examination, classification, and targeted imaging. ^{9,15–17,19–22}

Objectives

General objective

To synthesize the current evidence on how mechanical stability, classification systems, and imaging findings are integrated into surgical decision-making in thoracolumbar trauma.

Specific objectives

To describe the concept of spinal stability in thoracolumbar trauma and the imaging markers that suggest mechanical instability.
To compare the TLICS and AO Spine thoracolumbar classification systems and their score-based treatment recommendations.
To define the roles of CT and MRI in the initial assessment, with emphasis on evaluation of the PLC.
To propose a practical, management-oriented decision algorithm (Figure 1) and imaging checklist for spine surgeons and radiologists.

Methods

We designed a narrative review to synthesize the relevant evidence on thoracolumbar trauma, spinal stability, classification systems, imaging assessment, and surgical decision-making. The search was updated to September 28, 2026 (Table 1).

The targeted search strategy included PubMed / MEDLINE, Google Scholar, AO Spine / AO Foundation sources, journal-specific databases, and manual screening of the reference lists of key publications. The following terms were combined: thoracolumbar trauma, thoracolumbar fracture, spinal stability, AO Spine, TLICS, posterior ligamentous complex, computed tomography, magnetic resonance imaging, burst fracture, operative, nonoperative, conservative treatment, and surgical decision-making.

Enviar artículo para publicar

Enviar artículo para publicar

We included peer-reviewed, English-language publications relevant to the classification, stability, imaging, or management of thoracic, thoracolumbar, or lumbar trauma. Priority was given to foundational studies, guidelines and recommendations, prospective studies, diagnostic accuracy studies, systematic reviews, and meta-analyses. We excluded isolated case reports, small case series without broader clinical implications, non-spinal trauma, studies limited to the cervical spine, and publications whose content was adequately covered by more recent reviews or guidelines with a higher level of evidence.

The targeted search and manual screening identified 43 potentially relevant publications. After review of titles, abstracts, and, when available, full texts, 14 publications were excluded because of thematic overlap with more recent reviews or guidelines, an overly narrow or peripheral focus relative to the aim of the review, or limited applicability to the current classification and management framework. The final narrative synthesis included 29 publications. Given the narrative nature of this work, neither a PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram nor a formal risk-of-bias assessment was applied.

This review did not include individual patient data, identifiable images, or original research involving human participants; therefore, ethics committee approval and informed consent were not required. The illustrative images in Figure 2 were obtained from the institutional archive and were anonymized.

Results

Forty-three potentially relevant publications were preselected; 14 were excluded and 29 were included in the final narrative synthesis. The included publications comprised 6 foundational studies on classification or treatment algorithms, 4 clinical guidelines or recommendations, 8 systematic reviews or meta-analyses, 2 imaging-focused narrative reviews, and 9 primary diagnostic, observational, or therapeutic studies. The evidence was organized into the following themes: biomechanics and stability, classification systems, CT and MRI assessment, PLC injury, injury patterns, conservative versus surgical management, and controversial scenarios.

The thoracolumbar junction as a transition zone

The thoracolumbar junction is vulnerable because it marks the transition between the thoracic spine, stabilized by the rib cage and physiological kyphosis, and the lumbar spine, which is more mobile and bears greater axial loads. This transition creates a zone of mechanical stress, particularly between T11 and L2, where axial compression, flexion-distraction, and translational mechanisms can produce traumatic fractures. ^1,4,10,11

From a surgical perspective, small morphological differences can have important consequences. An isolated anterior wedge fracture may behave as a stable injury, whereas a morphologically similar injury with disruption of the posterior tension band may be unstable. The mechanism of injury should therefore be interpreted together with the classification, the neurological examination, and imaging.

Denis’ three-column model remains a useful conceptual framework. The anterior column comprises the anterior longitudinal ligament, the anterior annulus, and the anterior vertebral body; the middle column comprises the posterior vertebral body, the posterior annulus, and the posterior longitudinal ligament; and the posterior column comprises the neural arch, facet joints, interspinous and supraspinous ligaments, ligamentum flavum, and posterior musculoligamentous structures. ⁴

Compression injuries result from axial loading with or without flexion and range from wedge fractures to burst fractures. Burst fractures involve failure of the posterior vertebral wall and may be associated with retropulsion. Distraction injuries reflect failure of the posterior or anterior tension band, whereas translation/rotation injuries indicate severe disruption of both osseous and ligamentous stabilizers. ^1,7,8

Injury mechanism and morphology help estimate stability, the risk of neurological injury, and the likelihood of failure of conservative treatment, but they should not be interpreted in isolation. Imaging findings must be integrated with alignment, ligamentous integrity, neurological status, and patient-specific factors. ^1,2,7,8

The concept of stability in thoracolumbar trauma

Stability as a central principle

Spinal stability is central because it directly influences the choice between conservative and surgical management. A stable injury is one that is expected to maintain alignment and protect neural structures under physiological loads. Conversely, an unstable injury may progress to deformity, mechanical pain, neurological deterioration, or failure of nonoperative treatment. ^1,4–8

Mechanical and neurological stability

Mechanical stability refers to the ability of the injured segment to maintain alignment without collapse or progressive deformity. Neurological stability implies the absence of current or potential compromise of the spinal cord, conus medullaris, cauda equina, or nerve roots. A patient may be neurologically intact and still have a mechanically unstable injury, particularly in distraction or translation injuries.

Imaging markers of instability

Findings that raise suspicion of instability include vertebral translation, facet malalignment or dislocation, posterior distraction, interspinous widening, and other signs of posterior tension band failure. ^{2,10,11,16,18} Marked loss of vertebral height and segmental kyphosis indicate greater morphological severity and risk of deformity, but they should not, on their own, be equated with mechanical instability. Similarly, posterior wall involvement defines burst fracture morphology but does not by itself establish instability or constitute an indication for surgery.

Stability and classification

TLICS translates the injury assessment into a score that integrates morphology, neurological status, and PLC integrity. ^5,6 AO Spine classifies compression injuries as type A, tension band injuries as type B, and translation injuries as type C, and also incorporates neurological status and clinical modifiers. ^1,7,8 B2, B3, and C injuries raise greater concern for instability, whereas purely transosseous B1 injuries require individualized assessment. ^2,3

Classification systems for surgical decision-making

Why classification systems matter

Classification systems provide a standardized language to describe morphology, estimate stability, communicate findings across specialties, and guide treatment. Their value lies not only in naming a fracture pattern but also in estimating whether the patient might benefit from conservative management, external immobilization, stabilization, decompression, or a combined approach. ^1,5–8

TLICS

TLICS includes three domains: injury morphology, neurological status, and PLC integrity. Morphology is scored 1 point for simple compression, 2 for burst fracture, 3 for translation/rotation, and 4 for distraction. The PLC is scored 0 points if intact, 2 if its integrity is indeterminate or suspected to be injured, and 3 if injured. The neurological component assigns 0 points to an intact patient, 2 to nerve root injury or complete spinal cord/conus injury, and 3 to incomplete spinal cord/conus injury or cauda equina syndrome. ^5,6 The total score provides a guiding treatment recommendation: scores of 3 or less favor nonoperative management, a score of 4 represents an intermediate zone, and scores of 5 or more favor operative management. These thresholds do not replace clinical judgment and should be interpreted in light of comorbidities, associated injuries, deformity, bone quality, ability to mobilize, and patient preferences. ^2,5,6

AO Spine thoracolumbar classification

The AO Spine thoracolumbar classification is a hierarchical system based primarily on morphology. Type A injuries are compression injuries, subdivided from A0 to A4; type B injuries represent tension band failure, with B1 being purely transosseous, B2 predominantly ligamentous or osteoligamentous posterior, and B3 hyperextension injuries with anterior tension band failure; and type C injuries show translation or displacement in any plane. The system also incorporates neurological status and clinical modifiers. ¹

Within type A injuries, A3 is an incomplete burst fracture involving the posterior wall and a single endplate, whereas A4 is a complete burst fracture involving both endplates. Type B injuries involve tension band failure, and type C injuries represent loss of alignment due to translation; the latter are considered highly unstable. ^1,7,8 This structure allows the injury to be described anatomically and placed on a severity gradient, although the final decision still depends on the clinical context.

AO Spine Injury Score and treatment algorithm

The Thoracolumbar AO Spine Injury Score (TL AOSIS) assigns severity values to morphology, neurological status, and modifiers. For morphology, A0 scores 0 points; A1, 1; A2, 2; A3, 3; A4 and B1, 5; B2, 6; B3, 7; and C, 8. Neurological status is scored as N0 = 0, N1 = 1, N2 = 2, N3 = 4, N4 = 4, and NX = 3; the M1 modifier adds 1 point, whereas M2 has no fixed score. ⁷ In the original treatment algorithm, a score of 3 or less favors a trial of nonoperative treatment, a score of 4–5 represents an intermediate zone in which either strategy is acceptable, and a score greater than 5 favors operative management. ⁸ Contemporary studies show that the likelihood of a surgical recommendation increases with higher morphological category and score, particularly for A4, type B, and type C injuries. ⁹

Limitations

Classification systems have limitations. Interobserver agreement may vary, especially for complex patterns and PLC assessment. ^13–16,23 In addition, these systems do not fully capture frailty, bone quality, polytrauma, pain, functional demands, or surgical risk. They should therefore be used as tools for structured reasoning rather than as rigid mandates.

A practical comparison of TLICS and AO Spine is presented in Table 2.

Imaging in the initial assessment

Imaging as a decision-making tool

Imaging should be viewed as a decision-support tool rather than merely a diagnostic step. It should characterize morphology, identify markers of instability, assess neurological risk, and provide the information needed to apply TLICS and AO Spine. ^{1,5–8,10–12}

Computed tomography

CT is the primary modality in the initial assessment because it provides a rapid, high-resolution evaluation of osseous structures and alignment. It should define the injured level, whether the injury is single or multiple, morphology, posterior wall involvement, retropulsion, spinal canal narrowing, vertebral height loss, segmental kyphosis, posterior element fractures, facet alignment, and translational or rotational displacement. ^2,10–12

CT findings relevant to classification and planning include posterior wall involvement, retropulsion, degree of spinal canal narrowing, vertebral height loss, kyphotic deformity, posterior element fractures, and facet abnormalities. A distinction should be made between findings that define morphology and those that suggest instability. In particular, posterior wall involvement distinguishes a burst fracture from a simple compression fracture, but it is not in itself evidence of PLC failure or an indication for surgical stabilization. Translation, distraction, and facet malalignment or dislocation are more closely associated with mechanically unstable injuries. ^{1,2,10,11,16,18}

Magnetic resonance imaging

MRI is not routinely required for every thoracolumbar fracture. Its main value lies in the assessment of the PLC, spinal cord, conus medullaris, cauda equina, epidural hematoma, traumatic disc injury, and soft tissues when there is a specific clinical or CT-based question. ^12–17 In neurologically intact patients with a diagnostic CT and no signs of ligamentous or neural injury, routine MRI is of limited value.

According to the American College of Radiology (ACR) Appropriateness Criteria, MRI of the region of interest without intravenous contrast is usually the appropriate next study when ligamentous, spinal cord, or nerve root injury is suspected or confirmed. ¹² It should be considered in the presence of neurological deficit, suspected PLC injury, equivocal CT findings, suspected disc or epidural injury, an unreliable neurological examination, or clinical–radiological discordance. A meta-analysis of four studies including 554 patients found that MRI changed the classification in approximately 17 % of cases and the treatment decision in 22 %, mainly by reclassifying type A injuries as type B; these results support selective use when PLC status remains indeterminate after CT. ¹⁷

Communication with the surgeon

A management-oriented report should communicate the findings that modify classification, stability estimation, and treatment planning (Tables 4 and 5; Figure 3). The report should summarize the relevant radiological findings without replacing the clinical and neurological integration required to make treatment decisions. ^{2,10,12,16,24}

Posterior ligamentous complex injury

The PLC is a key stabilizing structure in thoracolumbar trauma (Figure 2). Its integrity maintains the function of the posterior tension band and resists flexion forces. Frank PLC disruption raises concern for instability, progressive kyphosis, and failure of conservative treatment, and may change the classification and the treatment score. ^{5,6,13–16,18}

The PLC generally comprises the supraspinous ligament, interspinous ligament, ligamentum flavum, and facet joint capsules. In flexion-distraction or translation injuries, one or more of these components may be injured, ranging from edema or partial tears to complete disruption with posterior hematoma and interspinous widening.

CT may show indirect signs of PLC injury, including vertebral translation, facet malalignment or subluxation/dislocation, interspinous widening, spinous process avulsion fracture, and horizontal fractures of the lamina or other posterior elements. Not every facet widening or laminar fracture should be considered equivalent to PLC injury: a vertical laminar fracture may accompany burst fracture morphology without implying disruption of the posterior tension band. ^16,18 The combination of multiple CT signs increases the likelihood of ligamentous injury. A meta-analysis of eight studies including 1,440 patients estimated a pooled sensitivity of 75 % and a specificity of 87 % for CT in detecting PLC injury, using MRI as the reference standard. ¹⁸

MRI allows direct assessment of ligamentous continuity, as well as edema and hematoma. Frank discontinuity of the normal hypointense band of the supraspinous ligament or ligamentum flavum is more specific for rupture than isolated hyperintensity on short tau inversion recovery (STIR) sequences; therefore, edema or fluid without discontinuity should be interpreted with caution and may warrant an indeterminate category rather than an assumption of complete PLC injury. ^13–16 For management purposes, the PLC can be reported as intact, indeterminate/suspected, or frankly disrupted. In TLICS, PLC status modifies the score; in AO Spine, posterior tension band injury may correspond to a type B pattern, whereas M1 is reserved for indeterminate tension band injury. ^1,5–8

Management-oriented injury patterns

Injury patterns should be interpreted not only as morphological descriptions but also as indicators of stability, neurological risk, and the potential need for operative treatment. ^1,5–8 A management-oriented summary is shown in Table 3.

Compression fractures

Compression fractures usually correspond to AO type A injuries. An A1 wedge fracture or an A2 split/pincer injury without posterior wall involvement, tension band injury, or neurological deficit is generally considered mechanically stable and can be treated conservatively. If the posterior wall is involved, the injury is no longer a simple compression fracture and should be classified as an A3 or A4 burst fracture, without assuming that this alone implies mechanical instability. ^1,2,7,8

Burst fractures

Burst fractures are characterized by failure of the vertebral body with posterior wall involvement. In AO Spine, A3 corresponds to an incomplete burst fracture and A4 to a complete burst fracture. ¹ Management depends on neurological status, tension band integrity, alignment, deformity, ability to mobilize, pain, and patient factors; the percentage of canal compromise should not be used as a stand-alone criterion for surgery. ^{2,8,9,19–22}

Burst fractures without neurological deficit remain a controversial scenario (see Discussion). In patients with an intact PLC, no translation or distraction, and acceptable alignment, nonoperative management can achieve satisfactory clinical outcomes. ^{2,19,20,22,25} Earlier evidence, including a prospective randomized trial, its long-term follow-up, and a Cochrane review, had already shown no consistent functional advantage of surgery in selected patients, albeit with limited certainty of evidence. ^26–28

Flexion-distraction and posterior tension band injuries

Flexion-distraction injuries require a distinction between B1 and B2 patterns. AO B1 injuries are purely transosseous and monosegmental and therefore retain the potential for bone healing. Contemporary data suggest that some neurologically intact patients with isolated, minimally displaced B1 injuries can be managed without surgery under close clinical and radiographic follow-up. ³ In contrast, B2 injuries involve predominantly ligamentous or osteoligamentous disruption of the posterior tension band, raise greater concern for instability, and therefore usually favor surgical stabilization. ^1,2,7,8

Extension-distraction, translation, and fracture-dislocation injuries

B3 extension-distraction injuries involve failure of the anterior tension band and are particularly relevant in the rigid or ankylosed spine, where they can be highly unstable even after low-energy trauma. Translation/rotation injuries correspond to type C and represent the most unstable patterns, frequently requiring stabilization. Fracture-dislocations involve loss of alignment and require urgent evaluation by the spine team. ^1,2,7,8

Conservative versus surgical management

The goals of treatment are to preserve or restore stability, protect neurological structures, prevent progressive deformity, relieve pain, and allow early mobilization and functional recovery. The choice between conservative and surgical management depends on morphology, stability, neurological status, tension band integrity, alignment, comorbidities, ability to mobilize, functional demands, and patient preferences. ^{1,2,5–8,22,25}

Conservative management may be appropriate for mechanically stable injuries in neurologically intact patients with an intact PLC, no unstable translation or distraction, acceptable alignment, manageable pain, and feasible follow-up. It includes analgesia, early mobilization, rehabilitation, and clinical and radiographic monitoring; bracing may be used selectively but should not be considered mandatory for every stable, neurologically intact burst fracture. ^2,25

Surgical management should be considered in cases of significant mechanical instability, type C translational injuries, B2/B3 injuries, fracture-dislocation, progressive deformity, neurological deficit associated with relevant neural compression, failure of conservative treatment, or patient factors favoring early stabilization. ^1,2,7,8 Confirmed PLC disruption substantially increases concern for instability, although its therapeutic implications must be weighed together with morphology and neurological status. Purely transosseous B1 injuries are a potential exception and can be treated nonoperatively in carefully selected cases. ³ The goals of surgery include restoring stability, correcting or preventing deformity, protecting neural structures, decompressing when indicated, and facilitating mobilization.

A key decision is to distinguish stabilization from decompression. Stabilization addresses mechanical instability, whereas decompression addresses neural compression. Spinal canal narrowing caused by retropulsion does not necessarily require decompression in a neurologically intact patient. In contrast, neurological deficit with concordant compression, cauda equina syndrome, progressive deterioration, epidural hematoma with mass effect, or traumatic disc herniation may justify decompression. ^1,2,5–8,12

Discussion

Controversial scenarios

Despite current classification systems, several scenarios remain controversial because they do not fit neatly into binary stable-or-unstable categories. These include A3/A4 burst fractures without neurological deficit, spinal canal compromise in neurologically intact patients, equivocal PLC injury, CT–MRI discordance, polytrauma, and the ankylosed spine. ^{2,9,15–17,19–22,24,29}

Burst fractures without neurological deficit

The central question is whether surgery improves long-term functional outcomes compared with nonoperative management in neurologically intact patients. A 2024 meta-analysis of 19 studies and 1,056 patients found little or no difference in pain or disability at 6 months or later, although surgery achieved better radiographic correction of kyphosis; the certainty of evidence was very low. ¹⁹ In a 2026 prospective international cohort of 198 patients, the time to reach a clinically important improvement in the Oswestry Disability Index (ODI) was similar with surgical and nonsurgical treatment. ²⁰ Because this cohort was not randomized, its results may be affected by confounding by indication. A 2026 meta-analysis suggested an earlier return to work after surgery, but no advantage in the overall rate of return to work, pain, disability, or quality of life; heterogeneity for time to return to work was high. ²¹ Taken together, the current evidence favors an individualized decision rather than an indication based solely on burst fracture morphology. ²²

Spinal canal compromise without neurological deficit

Retropulsion can produce striking spinal canal narrowing, but there is no universal percentage threshold of canal compromise that, on its own, constitutes an indication for surgery in a neurologically intact patient. This finding should be correlated with neurological status, stability, actual neural compression, epidural hematoma, disc injury, and clinical course. ^2,9,19,20,22

Equivocal PLC injury and CT–MRI discordance

CT may suggest PLC injury through indirect signs, whereas MRI may show ligamentous or soft-tissue edema that does not always imply complete mechanical failure. Frank ligamentous discontinuity is more specific than isolated hyperintensity. In these cases, the practical question is whether the additional finding changes the classification, the estimation of stability, or the treatment strategy; a truly indeterminate PLC status may justify MRI if that information could change management. ^13–18,29

Polytrauma, advanced age, and the ankylosed spine

In polytrauma, the decision also depends on hemodynamic stability, associated injuries, and surgical timing. In patients with ankylosing spondylitis or diffuse idiopathic skeletal hyperostosis (DISH), the rigid spine behaves biomechanically like a long bone, and apparently minor trauma can cause highly unstable injuries. The threshold for CT should be low, and multilevel or noncontiguous injuries should be actively sought; when a fracture is found in an ankylosed spine, imaging of the entire spine may be necessary. MRI is particularly important in the presence of neurological deficit, an unreliable examination, suspected epidural hematoma or discoligamentous injury, or persistent high suspicion despite inconclusive CT. In AO Spine, these conditions are captured by the M2 modifier. ^1,2,7,12 In older patients or those with poor bone quality, frailty, surgical risk, baseline mobility, and tolerance of immobilization should also be taken into account.

Surgeon-oriented imaging checklist

In thoracolumbar trauma, the imaging report should go beyond fracture detection and provide information that supports classification, stability estimation, and treatment planning. The most useful report for the surgeon describes the injury pattern, identifies signs of tension band failure, assesses the PLC when possible, evaluates neural structures, and highlights the findings that may change management. ^{2,10,12,16,18,24}

The checklist should address where the injury is located, whether it is single or multilevel, the dominant morphology, the presence of posterior wall involvement and retropulsion, the degree of spinal canal narrowing, height loss and kyphosis, posterior element injuries, facet alignment, translation or distraction, and signs of PLC injury. When MRI is available, neural compression, spinal cord/conus/cauda equina injury, epidural hematoma, and traumatic disc injury should also be described. Finally, the report should summarize which findings favor stability or suggest instability and indicate whether there is a reasonable indication for MRI or urgent correlation with the spine team.

The management-oriented radiological impression should avoid vague descriptions. For example, instead of “L1 burst fracture with canal narrowing,” it is more useful to report an acute L1 burst fracture, specifying A3 or A4 when feasible, the approximate percentage of height loss, retropulsion and degree of spinal canal narrowing, the presence or absence of translation or facet malalignment, and CT signs of tension band injury. When MRI is available, PLC integrity should be reported as intact, indeterminate, or frankly disrupted. The report may suggest neurological correlation or MRI if the result could change the classification or management, but the final choice between conservative and surgical treatment requires clinical integration.

Proposed practical algorithm

A practical algorithm should begin with clinical and neurological assessment, followed by CT with multiplanar reconstructions. CT is then used to define morphology, identify signs of tension band failure, and evaluate the spinal canal and alignment. Non-contrast MRI of the region of interest is considered in the presence of neurological deficit, suspected ligamentous or neural injury, equivocal CT findings, suspected disc or epidural injury, an unreliable neurological examination, or clinical–radiological discordance. ^12,16–18 TLICS and / or AO Spine can subsequently be used to structure the treatment decision without replacing clinical judgment. ^1,2,5–9

Conservative management is favored when the patient is neurologically intact, morphology suggests stability, the PLC is intact, there is no unstable translation or distraction, alignment is acceptable, pain allows mobilization, and follow-up is feasible. Surgery is favored for B2/B3 or C injuries, translation, fracture-dislocation, frank tension band disruption with instability, neurological deficit with relevant compression, progressive deformity, or failure of conservative management. A TLICS score of 4 and a TL AOSIS score of 4–5 represent intermediate zones in which the decision should be individualized. ^2,3,5,6,8

Strengths, limitations, and future directions

A management-oriented approach improves interdisciplinary communication by aligning radiological interpretation with surgical decision-making. This framework is useful for residents, orthopedic trauma surgeons, radiologists, and spine surgeons because it offers a shared language based on TLICS and AO Spine. ^1,5–8

The available evidence has limitations. Interobserver agreement may vary, particularly for complex patterns and PLC assessment. ^13–16,23 Algorithms do not fully capture frailty, bone quality, polytrauma, functional demands, surgical risk, or patient preferences. In addition, controversial scenarios such as burst fractures without neurological deficit continue to show heterogeneity in both practice and comparative evidence, even in contemporary studies. ^9,19–22

This is a narrative rather than a systematic review; therefore, it does not provide pooled estimates or a formal risk-of-bias assessment. In addition, the restriction to English-language publications and a non-exhaustive search may have introduced selection and language bias. Future research should improve the reproducibility of PLC assessment, validate treatment algorithms across different trauma systems, and identify imaging findings that predict failure of conservative management.

Conclusions

Thoracolumbar trauma requires a structured, multidisciplinary approach that integrates morphology, neurological status, mechanical stability, tension band integrity, patient factors, and imaging findings. TLICS and AO Spine provide useful frameworks to standardize assessment and support treatment planning; however, their scores should be interpreted as decision-support tools rather than substitutes for clinical judgment. ^1,2,5–9

CT remains the cornerstone of the initial assessment because it defines morphology, alignment, posterior wall involvement, retropulsion, the spinal canal, posterior elements, and translational or rotational displacement. ^2,10–12 MRI is complementary and potentially decisive when ligamentous or neural injury is suspected, but its use should be selective and directed at a specific clinical question. ^12,16–18

Surgical decisions should not be based on isolated descriptors such as “burst fracture,” “posterior wall involvement,” or “spinal canal narrowing.” Stable compression fractures and a proportion of A3/A4 burst fractures without neurological deficit can be managed nonoperatively, whereas B2/B3 injuries, type C translational patterns, fracture-dislocations, progressive deformity, and neurological compromise with concordant compression favor surgical consideration. ^{2,3,8,9,19–22} An imaging checklist focused on morphology, stability, and neural structures can improve communication between radiology and spine surgery without turning the radiology report into a stand-alone treatment recommendation.

Declarations

Funding: none.
Conflicts of interest: the authors declare no conflicts of interest.
Ethics statement: this narrative review did not include individual patient data, research involving human participants, or animal research. The CT and MRI images in Figure 2 were obtained from the institutional archive, were anonymized, and contain no identifiable data; in accordance with institutional regulations, their use for illustrative purposes did not require individual consent. Therefore, ethics committee approval was not applicable.