Surgical management protocol for anterior-only reduction and fixation for acute, delayed and old subaxial cervical facet dislocation: a retrospective study of 87 consecutive cases in China

Article information

Asian Spine J. 2026;.asj.2025.0324

Publication date (electronic) : 2026 January 6

doi : https://doi.org/10.31616/asj.2025.0324

Zhengfeng Zhang

Department of Orthopedics, Xinqiao Hospital, Army Medical University, Chongqing, China

Corresponding author: Zhengfeng Zhang, Department of Orthopaedics, Fuling Hospital, Chongqing University, 2 Gaosuntang Road, Fuling District, Chongqing 408000, China, Tel: +86-23-68774628, Fax: +86-23-68755608, E-mail: zhangz3@126.com

*Current affiliation: Department of Orthopaedics, Fuling Hospital, Chongqing University, Chongqing, China

Received 2025 June 14; Revised 2025 July 26; Accepted 2025 September 9.

Abstract

Introduction

The surgical management of traumatic cervical facet dislocations requires attention to four key elements: decompression, reduction, fixation, and fusion. Variations in treatment strategies arise from differences in reduction success between closed reduction, anterior open reduction, and posterior open reduction techniques, the requirement for three-column spinal fixation, the presence of traumatic intervertebral disc herniation, and the timing of injury (acute, delayed, or old dislocation). This complexity has led to the adoption of multiple surgical approaches, including anterior-alone, posterior-alone, combined anterior-posterior, posterior-anterior, and even staged anterior-posterior-anterior strategies [1–3].

Among these options, the anterior-only approach (discectomy, open reduction, and instrumental fusion) is widely recommended. It has two main advantages: lower risk of secondary spinal cord injury in the presence of traumatic intervertebral disc herniation [4–6], and reduced surgical trauma and shorter operative time compared with posterior or combined approach [3,7,8]. However, the anterior-only approach carries two principal limitations: anterior reduction failure rates ranging from 4% to 40% [3,9]; and approximately 30% failure rate of anterior vertebral screw fixation at C6/7 and C7/T1 levels or in cases of severe vertebral and articular process fractures [10,11]. Both scenarios may necessitate additional posterior surgery. The posterior-alone approach facilitates reduction and three-column pedicle screw fixation [12], but is associated with greater surgical morbidity and a higher risk of neurological deterioration when retropulsed disc fragments are present [13]. Therefore, the posterior approach is typically reserved for cases involving posterior cord compression due to fractured posterior elements or irreducible fracture dislocations refractory to open or closed anterior reduction. To establish itself as a standalone treatment, an anterior-only procedure must achieve both a 100% reduction success rate and provide sufficiently rigid anterior fixation to obviate the need for supplemental posterior surgery.

To address these challenges, the authors developed two novel anterior reduction techniques: (1) modified kyphotic paramedian distraction using Caspar pins and (2) anterior facetectomy [14]. Additionally, a specialized anterior cervical pedicle screw plate system was designed to optimize fixation [15]. The initial clinical outcomes were reported in a 2019 case series of 63 patients [16], which subsequently led to the development of a comprehensive surgical protocol incorporating these techniques. The present retrospective study evaluated the efficacy of this protocol, which integrates both the reduction techniques and the instrumentation, in 87 consecutive patients with acute, delayed, or old subaxial cervical facet dislocations, with a minimum follow-up of 12 months.

Materials and Methods

Patients

Between January 2014 and December 2023, a consecutive series of 89 patients with subaxial cervical facet dislocations were admitted or transferred to and treated surgically by the investigator’s group at Xinqiao Hospital. Two patients with contraindications to anterior surgery were excluded: one with tracheostomy and one with prior adjacent-level anterior cervical discectomy and fixation. Ultimately, 87 patients underwent anterior-only reduction and fixation. This study extends our initial 2019 cohort of 63 patients with additional cases and longer follow-up [14]. Written informed consent was obtained from all patients.

For these 87 patients (67 men and 10 women), ages ranged from 20 to 73 years (mean 52.6±18.5 years). Injury levels spanned C3/C4 to C7/T1, distributed as follows: C3/C4 (n=6), C4/C5 (n=16), C5/C6 (n=39), C6/C7 (n=23), and C7/T1 (n=3). The cohort included 23 unilateral and 64 bilateral facet dislocations. Traumatic disc herniation was present in 24 patients. Neurological status, assessed using the American Spinal Injury Association (ASIA) Scale [17], ranged from A to E (A: n=28, B: n=7, C: n=6, D: n=25, E: n=21). Vertebral or articular process fractures were identified in 48 patients. Nine patients had old dislocations, defined as >3 weeks after injury, with durations ranging from 21 to 64 days. Surgical procedures were performed between 0 and 64 days postinjury. Follow-up ranged from 12 to 24 months (mean 16.3±4.2 months) (Table 1).

Table 1

Demographic data of patients with cervical facet dislocation

Preoperative and postoperative evaluation included standard anterior-posterior, lateral, and dual oblique radiographs to assess spinal injury, reduction, and internal fixation. Magnetic resonance imaging (MRI) at 1.5 or 3.0-Tesla, with T1-weighted and T2-weighted sequences, was used to evaluate traumatic disc herniation, cord compression, and intramedullary signal changes on sagittal and axial planes. Computed tomography (CT) was performed to characterize the injury in detail, assess spinal alignment, and confirm fusion.

Surgical management protocol

The surgical protocol was individualized according to patient-specific factors (clinical condition and neurological status) and technical requirements (reduction and fixation) (Fig. 1). Patients were stratified into three temporal categories based on the interval from injury to surgery: acute (0–3 days), delayed (4–21 days), and old (>3 weeks). The largest subgroup was the delayed group, which mainly included patients who were initially asymptomatic and overlooked, or those transferred from external hospitals or emergency departments.

Fig. 1

The surgical management protocol and patients’ information.

Regarding reduction strategies, patients were classified into three groups: (1) acute and delayed dislocation cases requiring urgent surgery within 24 hours of admission (n=66); (2) acute and delayed dislocation cases with contraindications to urgent surgery (n=12); and (3) old dislocation cases who underwent selective operation (n=9). Patients requiring urgent surgery and old dislocation (77 cases in total) underwent the anterior-only procedure, which incorporated two sequential reduction techniques: modified Caspar pin kyphotic paramedian distraction, followed by anterior facetectomy, if the former technique failed. Contraindications to urgent surgery at admission included fever, brain trauma, respiratory compromise, shock, and other unstable conditions that necessitated surgical delay. These 12 patients initially underwent closed reduction. If successful, anterior discectomy and fixation were performed after an appropriate delay. if closed reduction failed, anterior open reduction was undertaken when surgery became feasible.

For anterior fixation, the methods were selected according to injury level and severity. Anterior vertebral screw plates were used for C3/4, C4/5, and selected C5/6 cases, whereas pedicle screw plates were applied for C6/7, C7/T1, and selected C5/6 cases with severe vertebral and articular process fractures. All procedures were performed by the investigator.

Reduction techniques

Closed reduction

In most cases, the patients were awake and treated with Gardner-Wells traction, starting with 5 kg plus 2.5 kg for each level of injury below C1. After gradual traction application, anterior rotation and lateral flexion were carefully performed away from the side of the dislocated facets. Once the locked facets disengaged, rotation was applied in the opposite direction. When a click was heard or felt, the neck was extended [3,18] (Fig. 2A).

Fig. 2

Illustrations of three reduction techniques. (A) Closed reduction. (Left) Lateral view of facet dislocation; (Middle) Gradual increase of in-line traction until complete articular process disengagement; (Right) Maintenance of traction with manual caudal pressure on the upper vertebra to complete reduction. (B) Principle of modified kyphotic paramedian distraction with Caspar pins. (Left) Direction of the upper pin placement at the dislocation side in the axial plane. (Middle) Positioning Caspar pins at approximately 10° to 20° angle with respect to each other in the sagittal plane. Gradual distraction under fluoroscopy until disengagement of locked facets was observed on the lateral view. (Right) Application of dorsal and rotational force to the rostral vertebra to complete reduction. (C) Principle of anterior facetectomy. (Left) Facet locking persists after kyphotic paramedian distraction. (Middle) Partial resection of the dislocated superior facet edge following anteromedial foraminotomy. (Right) Application of dorsally-directed and rotational force to the rostral vertebra to complete reduction.

Modified kyphotic paramedian distraction with Caspar pins

Patients remained under Gardner-Wells traction. A standard Smith-Robinson anterior cervical approach was employed in all cases. The kyphotic distraction technique with Caspar pins was first described in 2000 [7]. Briefly, two Caspar pins were inserted at 10°–20° sagittal angulation [7]. The upper pin entry point and trajectory were adjusted to produce an ipsilateral offset toward the dislocated facet [16]. This configuration generated kyphotic distraction (reproducing segmental flexion) and paramedian force application, concentrating distraction at the dislocated joint. The combined forces synergistically facilitated facet unlocking. Following discectomy, gradual distraction was maintained until facet disengagement was confirmed either by tactile feedback or by fluoroscopic visualization on the lateral view. A dorsally directed force with concurrent rotation was then applied to the rostral vertebra to achieve reduction (Fig. 2B).

Anterior facetectomy

The anterior facetectomy procedure for reduction followed our previously described protocol [14,16]. When kyphotic paramedian distraction with Caspar pins failed to achieve reduction, an anteromedial foraminotomy was performed by resecting the posterior foraminal portion of the uncovertebral joint. After cephalad retraction of the nerve root within the neuroforamen, the edge of the dislocated superior facet was fractured to achieve reduction. The Caspar retractor was then directed posteriorly to achieve posterior translation of the upper segment and the fractured lower segment (part of the superior facet). Once the ipsilateral facet was reduced, the retractor was further advanced and rotated cephalad to facilitate posterior translation of the contralateral superior segment. If rotational reduction of the contralateral facet was unsuccessful, a facetectomy was performed on that side using the same technique (Fig. 2C).

Fixation

Following reduction, a matched polyetheretherketone cage with autologous bone was inserted. Two types of anterior cervical locking plates were used, selected according to the injury level and fracture severity. A specially designed anterior pedicle screw plate (Z3; Wego, Weihai, China) was applied for C7/T1, C6/7, selected C5/6 cases with severe vertebral and articular process fractures, and in patients who underwent the anterior facetectomy procedure [15]. Anterior pedicle screws were inserted along the pedicle axis under fluoroscopic guidance or O-arm navigation, following our established protocol [15,16,19]. For C3/4, C4/5, and selected C5/6 levels, anterior vertebral screw palates were used. Postoperatively, all patients received closed suction drainage for 48–72 hours. Cervical orthoses were applied for 6–8 weeks, except in cases stabilized by anterior pedicle screw fixation. Fusion was assessed by CT at 6 months postoperatively and defined as trabecular bone bridging between adjacent vertebrae. All scans were independently evaluated by two radiologists.

Results

All 87 patients successfully underwent anterior surgical procedures. No intraoperative vertebral artery or nerve root injuries occurred, and none of the patients required supplemental posterior surgery. All patients achieved successful reduction (Table 2). Success rates were 67% for closed reduction, 81% for modified Caspar pin kyphotic paramedian distraction, and 100% for anterior facetectomy. No supplemental posterior procedures were required. Most failed closed reductions occurred at C6/7 and C5/6. The modified kyphotic paramedian distraction was successful in all C3/4 and C4/5 cases, as well as most C6/7 cases. Anterior facetectomy achieved reduction at all C5/6, C6/7, and C7/T1 levels, including old dislocations. Four patients required bilateral anterior facetectomy due to old dislocations with severe articular process fractures. No vertebral artery or nerve root injuries occurred during facetectomy. Among old dislocations, the success rate was 11% with the modified Caspar pin kyphotic-paramedian distraction versus 100% with anterior facetectomy. Compared with modified kyphotic paramedian distraction, anterior facetectomy achieved a higher success rate of reduction but required longer operative time (p<0.01 for both) (Table 2).

Table 2

Clinical data of three reduction techniques

All patients achieved stable fixation with an anterior vertebral screw plate (n=59) or anterior pedicle screw plate (n=28). Compared with anterior vertebral screw palate fixation, anterior pedicle screw fixation required longer fixation time (8.6±4.3 minutes vs. 46.3±15.3 minutes), longer operation time (69.2±23.5 minutes vs. 110.6±26.5 minutes), and greater blood loss (68.6±24.1 mL vs. 96.3±75.2 mL) (p<0.05 for all) (Table 3). Postoperative radiographs, MRI, and CT confirmed appropriate cervical alignment and accurate pedicle screw placement within the pedicle cortex. At ≥12 months’ follow-up, fusion was achieved in all patients, as verified by CT (Fig. 3). No instrumentation failures occurred. Neurological improvement of ≥1 ASIA grade was observed in 29 patients (43.9%), with no cases of neurological deterioration (Table 4).

Table 3

Clinical data of two fixation techniques

Fig. 3

Preoperative and postoperative imagines of an illustrative case of C6/7 facet dislocation. (A) Preoperative sagittal computed tomography (CT) demonstrating C6/C7 right facet joint subluxation (arrow). (B) Preoperative sagittal CT revealing the C6/C7 dislocation with C7 vertebral fracture (arrow). (C) Preoperative sagittal CT depicting C6/C7 left facet joint dislocation with fracture (arrow). (D) Preoperative three-dimensional-CT image illustrating facet joints dislocation (arrows). (E) Preoperative T2 sagittal magnetic resonance imaging (MRI) showing C6/C7 dislocation with C7 vertebral fracture (arrow). (F) Postoperative T2 sagittal MRI demonstrating C6/C7 reduction and decompression. (G) Postoperative anteroposterior and oblique radiograph images confirming restoration of sagittal alignment and fixation. (H) Postoperative sagittal CT image showing the C6/7 right facet joint reduction and correct pedicle screw fixation (arrow). (I) Six months postoperative sagittal CT images confirming C6/7 reduction and solid fusion. (J) Postoperative sagittal CT image showing C6/7 left facet joint reduction and appropriate pedicle screw fixation (arrow). (K) Postoperative axial CT images verifying optimal anterior pedicle screws and vertebra screws at C6 and C7.

Table 4

Neurologic outcome (ASIA grade)

Discussion

In this study, the anterior-only procedure achieved successful reduction, complete decompression, rigid internal fixation, and fusion in patients with cervical facet dislocation. The advantages of this approach are threefold: (1) reduced risk of secondary spinal cord injury and minimized surgical trauma; (2) a 100% reduction success rate with the combined use of modified kyphotic paramedian distraction and anterior facetectomy, applicable in acute, delayed, as well as old dislocations of <8 weeks; and (3) provision of a three-column fixation with the anterior cervical pedicle screw plate. To the best of our knowledge, this represents the most effective algorithm for anterior-only reduction and fixation of cervical facet dislocations.

There is no clear consensus on the optimal surgical approach (anterior, posterior, or combined) for cervical facet dislocation. Contributing factors include traumatic intervertebral disc herniation, variable failure rates of different reduction techniques, level-specific fixation requirements, extent of posterior element injury, and chronicity of dislocations. Conventional anterior techniques are limited by their inability to apply focal force directly to locked facets, resulting in failure rates of 4%–40% for anterior open reduction [3,7,8]. In such cases, supplemental posterior reduction or a posterior-only approach is often required. The posterior approach achieves reliable reduction and provides three-column stabilization with pedicle screw fixation, but is associated with greater iatrogenic soft-tissue injury and an increased risk of secondary spinal cord injury in the presence of traumatic intervertebral disc herniation [4–6]. The anterior reduction techniques used in the present study achieved a 100% success rate, supporting the effectiveness of this anterior-only algorithm.

Three principal reduction techniques are available for facet dislocations: closed reduction, anterior open reduction, and posterior open reduction [20]. Closed reduction provides the earliest opportunity for reduction to relieve spinal cord compression, but its success rate ranges from 30% to 100% [21], and it carries a risk of secondary spinal cord injury in the presence of traumatic intervertebral disc herniation [5,6,8]. In the present study, the surgical management protocol stratified patients into three groups: (1) candidates for urgent surgery at admission, (2) those with contraindications to urgent surgery, and (3) those with old dislocations. Emergency surgeries were performed promptly after admission to decompress the spinal cord, mirroring the timing of early closed reduction procedures. Patients unsuitable for urgent surgery underwent early closed reduction. As a tertiary referral center, our cohort consisted mainly of delayed-presentation cases, most of whom were transfers from other hospitals or emergency departments with initially missed injuries. Particularly, our modified anterior techniques were effective not only for severe dislocations but also for delayed (3–21 days) and old (>21 days) cases. The maximum interval from injury to open reduction in this series was 64 days (Table 1). When the two techniques were compared, anterior facetectomy showed superior success at C6/7 and C7/T1 levels as well as in old dislocations compared with kyphotic paramedian distraction (Table 2). Anterior facetectomy achieved a 100% reduction rate, comparable to posterior open reduction outcomes for delayed and old cases. However, due to the limited number of old cases, it is unclear whether this technique can be applied to bilateral dislocations of over 8 weeks duration with severe facet joint fractures.

Cervical facet dislocation represents a three-column injury in the AO classification [22]. Consequently, some authors argue that anterior-only fixation may provide insufficient stability in cases with articular process injury [10,11,23], recommending supplemental posterior fixation for severe posterior column damage [24]. Others have reported failure rates of approximately 30% for anterior vertebral screw fixation at C6/7 and C7/T1, necessitating additional posterior surgery [10,11]. These concerns underscore the need to tailor fixation methods to the specific injury pattern at each level. In the present algorithm, anterior vertebral screw plates were applied for C3/4, C4/5, and selected C5/6 cases, whereas anterior pedicle screw plates were used for C7/T1, C6/7, selected C5/6, and in patients undergoing facetectomy, to achieve three-column stabilization. At ≥12 months of follow-up, no implant failures were observed, indicating stability comparable to posterior fixation.

The present study achieved satisfactory clinical outcomes while minimizing operative time and blood loss. Consistent with our prior investigation [25], the anterior-only approach maintained a 100% reduction success rate while substantially reducing surgical morbidity compared with conventional posterior-anterior techniques. Neurological improvement in this series was also comparable to outcomes reported with posterior-only approaches [12,13]. Previous studies have documented neurologic improvement rates of 20%–76.2% with posterior or combined approaches [9,13,26], whereas our anterior-only technique resulted in improvement of at least one ASIA grade in 43.9% of patients, with no cases of neurologic deterioration (Table 4). Notably, despite the predominance of delayed (48 patients) and old (nine patients) dislocations (Table 2), neurological outcomes remain favorable (Table 4).

While the clinical outcomes are encouraging, several limitations should be acknowledged. First, this study represents a single-surgeon experience at a specialized spine center and involves a modest sample size. Second, patients with systemic comorbidities (e.g., osteoporosis, ankylosing spondylitis) or posterior pathologies requiring intervention (e.g., epidural hematomas, ligamentum flavum rupture, laminar fractures) were excluded. Third, the old dislocation subgroup was small (n=9), with a maximum delay of 64 days; alternative approaches may be preferable in more complex cases involving prolonged delay, severe comminution, or multilevel injuries. Fourth, although anterior facetectomy and cervical pedicle screw fixation proved safe in this series [15,27], both techniques require advanced technical expertise due to their anatomical complexity.

Conclusions

In summary, the anterior-only reduction approach, incorporating modified kyphotic paramedian distraction with Caspar pins and anterior facetectomy, achieved a 100% reduction rate in acute, delayed, and old cervical facet dislocations. Anterior cervical pedicle screw plate fixation provided reliable three-column stabilization with high fusion rates at follow-up. This protocol eliminated the need for posterior or combined approaches in all cases, demonstrating both safety and efficacy in the management of cervical facet dislocations.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Article information Continued