Surgery for subaxial cervical spine injuries: which is better: anterior, posterior, or anterior–posterior combined approach?: a systematic review and meta-analysis

Abdul Hafid Bajamal; Eko Agus Subagio; Pandu Wicaksono; Asadullah; I Gusti Made Aswin Rahmadi Ranuh; Muhammad Faris; Budi Utomo

doi:10.31616/asj.2023.0266

Asian Spine J > Volume 18(4); 2024 > Article

Bajamal, Subagio, Wicaksono, Asadullah, Ranuh, Faris, and Utomo: Surgery for subaxial cervical spine injuries: which is better: anterior, posterior, or anterior–posterior combined approach?: a systematic review and meta-analysis

Review Article

Asian Spine Journal 2024;18(4):594-607.

Online first: August 21, 2024

DOI: https://doi.org/10.31616/asj.2023.0266

Surgery for subaxial cervical spine injuries: which is better: anterior, posterior, or anterior–posterior combined approach?: a systematic review and meta-analysis

Abdul Hafid Bajamal¹

, Eko Agus Subagio²

, Pandu Wicaksono²

, Asadullah²

, I Gusti Made Aswin Rahmadi Ranuh¹

, Muhammad Faris²

, Budi Utomo³

¹Department of Neurosurgery, Universitas Airlangga Hospital, Faculty of Medicine, Universitas Airlangga, Surabaya, Indonesia

²Department of Neurosurgery, Dr. Soetomo General Academic Hospital, Faculty of Medicine, Universitas Airlangga, Surabaya, Indonesia

³Department of Public Health and Preventive Medicine, Dr. Soetomo General Academic Hospital, Faculty of Medicine, Universitas Airlangga, Surabaya, Indonesia

Corresponding author: Muhammad Faris, Department of Neurosurgery, Dr. Soetomo General Academic Hospital, Faculty of Medicine, Universitas Airlangga, Surabaya, East Java, Indonesia,
Tel: +62-315501086, Fax: +62-315501088, E-mail: mfarisns@fk.unair.ac.id

Received August 18, 2023 Revised April 1, 2024 Accepted April 3, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Both anterior and posterior approaches have shown insignificant differences in good clinical outcomes with one over another advantages and disadvantages. This review aimed to provide evidence for the best management of subaxial cervical spine injuries and discuss the clinical outcomes and complications. Clinical studies of anterior versus posterior and anterior versus anterior–posterior (combined) approaches to subaxial cervical spine injury were searched electronically from PubMed, Medline, ScienceDirect, Cochrane Library, and other Internet databases. Clinical improvement, complication rates, and mortality rates showed no significant differences with an odds ratio of 1.09 (95% confidence interval [CI], 0.79–1.49; p=0.61) for the anterior versus posterior approach and an odds ratio of 1.05 (95% CI, 0.35–3.18; p=0.93) for the anterior versus the combined approach. Surgical duration and blood loss were significantly different between the anterior and posterior groups with a mean difference of −42.84 (95% CI, −64.39 to 21.29; p<0.0001); −212.91 (95% CI, −417.60 to 8.22; p=0.04), respectively, whereas the length of hospitalization did not (p=0.16). No difference was found between the groups when compared by clinical improvement and complication rate. Meanwhile, the anterior approach was superior to the posterior approach in terms of surgical duration, blood loss, and hospitalization length.

Keywords: Injury; Spinal cord trauma; Subaxial cervical injury

Introduction

Acute subaxial cervical spine injury remains the most devastating problem, in which >50% of cases are located between C5 and C7 after trauma [1,2]. Acute subaxial cervical spine injury consists of trauma to C3–C7 [3,4].

An individual with subaxial cervical spine injury often presents with a spectrum of damage to anatomical structures, including bone fractures, ligamentous injuries, and traumatic disk injuries, usually involving spinal cord and nerve roots. Irreducible fracture or dislocation, instability, and spinal cord compression usually are important considerations for selecting an appropriate treatment strategy, be it anterior, posterior, or combined approach [5–10].

The anterior approach is often used in cases of retropulsion of bone fragments or disc herniation that causes spinal cord injury and for its advantages, such as less iatrogenic injury and reduced incidence of secondary spinal cord injuries, postoperative axial pain risk in the posterior approach, need of concomitant anterior approach [11,12], and infection rate [1,13–20].

The posterior approach is often used in the case of posterior band injury and irreducible dislocation fracture and for its advantages, such as easy reduction and rapid postoperative intensive care [1,17–20].

For the last two decades, both anterior and posterior approaches have shown insignificant differences in good clinical outcomes with one over another advantages and disadvantages. This review aimed to provide evidence for the best management of subaxial cervical spine injuries and discuss the clinical outcomes and complications.

Methods

Search strategy

Article selection was performed using the Preferred Items for Systematic Review and Meta-Analysis (PRISMA) Protocol reporting guidelines and the Cochrane Collaboration guidelines [21,22]. The literature was searched electronically on July 26, 2023, from PubMed, Medline, ScienceDirect, Cochrane Library, and other Internet databases. The titles and abstracts were screened by combining the following terms: “(subaxial cervical spine injury*) AND ([anterior approach] OR [anterior-only approach]) AND ([posterior approach] OR [posterior-only approach]) AND ([anterior–posterior approach] OR [combined approach]) AND (‘cervical vertebrae’[Mesh] OR ‘cervical spine’[Mesh]) AND (‘spinal fractures’[Mesh]).”

Articles published in English between 2000 and 2023 were extracted. The reference lists of published articles were also comprehensively screened using Mendeley Reference Manager Software (Elsevier, Amsterdam, Netherlands) to ensure the inclusion of all possible studies. Unpublished data were not included in the review.

Selection criteria

Articles included were clinical comparative studies comparing anterior versus posterior and anterior versus anterior–posterior (combined)surgical approach for subaxial cervical spine injuries; randomized or controlled clinical trials; patients with traumatic subaxial cervical spinal injury based on computed tomography and plain radiographs; patients aged ≥18 years; follow-up >12 months; and full-text articles. The articles must present primary outcomes, including clinical outcomes, complication rates, blood loss, operative duration, and mortality rate.

Proceeding articles, editorials, commentaries, publications before a peer-review process, systematic reviews, and meta-analyses were excluded. Articles that were duplicate reports and articles whose full texts could not be acquired were also excluded.

Data extraction

The following information was collected: basic characteristics, including study design, age, sex, and number of enrolled patients; injury information, including classification and level of cervical injury, mechanism of injury, and neurological status; surgical information, consisting of surgical strategy; primary outcomes, including neurological improvement, complication rates, surgical duration, blood loss, length of hospitalization, and mortality rates.

Statistical analysis

This meta-analysis utilized RevMan ver. 5.4 (The Cochrane Review; The Cochrane Collaboration, London, UK). Quantitative data are presented as odds ratio for dichotomous outcomes and mean difference (MD) for continuous outcomes. Data were given based on a 95% confidence interval (CI). The level of significance was set at p=0.05. Heterogeneity was calculated using the χ² test and I² statistics (p-value for the χ² test of 0.10 was considered significant or I² >50%). Fixed-effects model (FEM) was applied unless the heterogeneity was significant, and the random-effects model was used. In the subgroup analysis, the outcomes between surgical strategies were compared in terms of surgical duration, blood loss, and hospitalization length using sensitivity analyses because of data heterogeneity.

Data collection and study quality assessment

Two reviewers extracted data from the included studies, whereas another two reviewers checked the extracted data independently. Disagreements between reviewers were resolved by all review members. For each missing data, our senior review members contacted the author for additional data. The PRISMA flow diagram represents the literature searches (Fig. 1) [23], and the included literature is summarized in Table 1 [1,16–20,24–30]. The review included 13 studies, where 12 were nonrandomized cohort studies and one was a randomized prospective study.

Risk of bias analysis

Each author assessed the risk of bias in the included articles independently using risk of bias in nonrandomized intervention studies (ROBINS-I) for nonrandomized studies (Fig. 2A, B) [1,17–20,24–30] and risk of bias in randomized trials (RoB 2) for randomized studies (Fig. 3A, B) [16,22,31]. The results of each assessment were discussed by all authors. The review included one randomized study. In the meta-analysis, the overall risk of bias assessment of twelve nonrandomized studies and one randomized study was low; meanwhile, five studies had a high risk of bias for selective reporting, and one study had a high risk of bias for outcomes.

Results

Anterior versus posterior approach

A total of 10 studies compared the anterior and posterior approaches in terms of clinical improvement, complication rates, mortality rates, surgical duration, blood loss, and length of hospitalization. Pooled data on clinical improvements, complication rates, and mortality rates showed no significant differences with an odds ratio of 1.09 (95% CI, 0.79–1.49; p=0.61). The FEM was applied to the pooled data (overall heterogeneity: I²=35%) (Fig. 4) [1,16–20,24,25,27,30].

Clinical improvement

Six studies with a total of 305 patients reported data regarding clinical improvements between the anterior and posterior groups, revealing no significant difference between the two groups. The odds ratio was 1.14 (95% CI, 0.72–1.81; p=0.59).

Complication rates

Ten studies with a total of 655 patients reported complication rates in the comparison between the anterior and posterior groups and showed no significant difference between the two groups. The odds ratio was 1.02 (95% CI, 0.63–1.65; p=0.95).

Mortality rates

Three studies with a total of 82 patients presented the mortality rates between the anterior and posterior groups. This subgroup analysis revealed no significant difference between the two groups. The odds ratio was 1.18 (95% CI, 0.42–3.31; p=0.76).

Surgical duration

Four studies with a total of 260 patients reported the surgical duration between the anterior and posterior groups. The random-effects model revealed significant differences between the two groups, with a MD of −42.84 (95% CI, −64.39 to −21.29; p<0.0001), and subgroup analysis was performed because of the dominance of the anterior approach (Fig. 5) [1,17–19].

Blood loss

Four studies with a total of 260 patients reported blood loss between the anterior and posterior groups. The random-effects model revealed significant differences between the two groups, with a MD of −212.91 (95% CI, −417.60 to −8.22; p=0.04), and a subgroup analysis was performed for patients aged <50 years (Fig. 6) [1,17–19].

Length of hospitalization

Five studies with a total of 523 patients reported the length of hospitalization between the anterior and posterior groups. The random-effects model revealed no significant differences between the two groups, with a MD of −2.11 (95% CI, −5.06 to 0.84; p=0.16). A subgroup analysis was performed for patients aged <50 years (Fig. 7) [1,16,17,19,30].

Anterior versus combined groups

Seven studies in total compared the anterior and combined groups in terms of clinical improvement and complication rates. The pooled clinical improvement and complication rate data showed no significant differences, with an odds ratio of 1.05 (95% CI, 0.35–3.18; p=0.93). The random-effects model was applied to the pooled data (overall heterogeneity: I²=75%) (Fig. 8) [24–30].

Clinical improvement

Four studies with a total of 177 patients reported clinical improvement between the anterior and combined groups. The random-effects model revealed significant differences between the two groups. The odds ratio was 2.45 (95% CI, 1.12–5.40; p=0.03).

The significant heterogeneity for the clinical improvement outcomes between the groups might have a strong correlation with a wide range of clinical improvements between the articles, which are attributed to the mechanisms of injury, number of levels affected, and type of injury, and the random-effects model was used for the quantitative analysis.

Complication rates

Seven studies with a total of 440 patients reported complication rates between the anterior and combined groups and showed no significance between the two groups. The odds ratio was 0.81 (95% CI, 0.15–4.41; p=0.81).

Along with the clinical improvements in the anterior and combined subgroups, significant heterogeneity in the complication rates was found. This might be strongly correlated with a major gap in the complication rates between the articles, which are attributed to the mechanisms of injury, number of levels affected, type of injury, and surgeon’s preferred technique, and the random-effects model was used for the quantitative analysis.

Discussion

Clinical improvement

Regarding clinical or neurological improvement, overall, no difference was found between the anterior and posterior approaches. Belirgen et al. [1] reported that the anterior approach had a slightly better American Spinal Injury Association (ASIA) score improvement than the posterior group in 33 patients. Lenga et al. [17] reported that the posterior approach achieved slightly better outcomes than the anterior approach in 28 octogenarian (aged ≥80 years) patients, and both the anterior and posterior approaches led to significant improvements in motor weakness.

Moawad and El-sawy [18] reported no difference between the groups in 40 patients with ligamentous subaxial cervical dislocation with no bone disruption and no cervical disk herniation, whereas postoperative pain was more intense in the posterior group than in the anterior group.

Ren et al. [19] did not find significant differences between the groups in their analysis of 159 patients. Thus, six patients in the anterior group and one in the posterior group had Frankel grade A preoperatively [19]. Similarly, Al Samouly and Taha [20] reported 60 patients with distractive flexion injury (DFI) without any other cervical spine injuries or multilevel injuries, and 24 of 60 patients had Frankel grade A preoperatively (anterior group, n=18; posterior group, n=6). Mizuno et al. [24] also did not find differences between the groups in terms of clinical (Frankel grade) improvement in 11 patients with facet dislocation. This finding might be affected by six of 11 patients (anterior group, n=5) who had Frankel grade A or B injuries with no clinical improvement.

In terms of clinical improvement between the anterior and combined approaches, overall, no difference was found between the groups, except in one study conducted by Rezaee et al. [25] who described 72 patients who underwent early cervical stabilizing surgeries and reported that the two-stage combined approach was superior to the anterior, posterior, and one-stage combined approaches; thus, the finding was not significant.

Mizuno et al. [24] did not find significant differences between the anterior and combined approaches in terms of clinical (Frankel grade) improvement in 11 patients with facet dislocation who were treated surgically. This finding might be affected by the preoperative condition. Because the preoperative condition might affect the outcome, Song and Lee [26] reported no difference between the groups (n=50) because both groups had extensive ASIA score improvements, except for six patients with anterior cord syndrome or complete spinal cord injury. According to Lee et al. [27], the surgeons’ skill or preferred technique might also affect the neurological or clinical outcomes.

Moreover, Kim et al. [28] reported no difference between the groups in 63 patients with bilateral facet dislocation; no patients had Frankel grade A. Similarly, Liu and Zhang [29] reported no difference in 93 patients with subaxial cervical facet dislocation.

Complication rates

Overall, no significant difference was found between the anterior and posterior groups in terms of complication rates; some of the complications (e.g., implant dislocation, transient dysphagia, transient odynophagia, transient hoarseness, cerebrospinal fluid leakage, carotid artery injury, and vertebral artery injury) might have a strong relationship with the surgeons’ skill or the preferred technique.

Regarding infections, Kwon et al. [16] reported 42 patients with unilateral facet injuries at the subaxial level, and five patients in the posterior group had surgical site infections, and one of those five had a deep wound infection, while no infections occurred in the anterior group. These infections might affect the length of stay of the posterior group [16]. Kim et al. [28] reported wound infections in the dorsal side in two patients in the combined approach.

Mizuno et al. [24] reported cerebrospinal fluid leak complications in one patient with a complex vertebral fracture in the combined group and one patient with neurological deterioration due to delayed disk herniation in the posterior group.

In implant-related complications, Belirgen et al. [1] reported one patient who underwent anterior surgery and then required additional posterior fusion after failing within the first week. Moreover, Ren et al. [19] reported two cases of bilateral dislocations in patients in the anterior group, who had screw loosening that required posterior stabilization, and one patient in the posterior group who had wound infections. Al Samouly and Taha [20] reported screw loosening in five patients in the posterior group, which was significant. In addition, Song and Lee [26] reported complications in seven patients in the anterior group that required fusion, and no significant difference was found between the groups.

In terms of fusion rate-related complications, Moawad and El-sawy [18] reported that the posterior approach was associated with a higher number of pseudoarthrosis (no definite solid bony union) than the anterior group with a 2:1 ratio. Moreover, one patient in the posterior group had a delayed subluxation due to screw loosening [18]. Moreover, Song and Lee [26] described that the anterior approach was disadvantageous compared with the posterior approach in regard to an increased fusion time, whereas the posterior approach was disadvantageous in regard to morbidity because of longer surgical duration; however, this approach was superior in terms of stability. Thus, anterior fusion should be performed first, followed by delayed posterior fusion for patients with DFI with unilateral or bilateral dislocation [26].

Liu and Zhang [29] reported three patients in their anterior group whose neurological state deteriorated by 1–2 points (converting ASIA scores A–E into 1–5) and one patient in the combined group by 1 point according to the Japanese Orthopaedic Association score. Lee et al. [27] also reported two cases in the anterior group who had aggravated neurological deficits postoperatively.

Lenga et al. [17] and Rezaee et al. [25] reported no significant difference between the groups in regard to postoperative complications. Lenga et al. [17] also stated that surgical duration and blood loss were significantly related to the increased risk of morbidity and mortality.

El-Hajj et al. [30] reported higher perioperative complications in the posterior group (n=292), although this was not statistically significant. This finding might be affected by the posterior group being the oldest, followed by those in the combined group and then the anterior group [30].

Transient dysphagia, transient odynophagia, and transient hoarseness are the most common transient complications of the anterior approach; as Belirgen et al. [1] reported, two patients in the anterior group developed transient hoarseness and dysphagia that resolved on follow-up. Kwon et al. [16] described 11 patients in the anterior group who had transient odynophagia or a change in their voice at the time of discharge. Ten described a resolution at the 6th week of follow-up, and one described a resolution at the 3rd month of follow-up [16].

Surgical duration

In terms of surgical duration, seven studies described this primary outcome, and the anterior approach was far superior to other groups (posterior and combined approaches) in six studies [1,17–19,26,29]. Liu and Zhang [29] described that intraoperative positioning may play a role in terms of skin-to-skin duration.

One article was excluded because of the dominance of the anterior approach, which was attributed to the surgeon’s preference for the anterior approach [19]. Kwon et al. [16] described that the posterior approach had a shorter mean surgical duration than the anterior approach (103 minutes versus 134 minutes, respectively), although Kwon et al. [16] mentioned that this time did not reflect skin-to-skin time because this time includes patient positioning and setups.

Blood loss

In terms of blood loss, five studies reported this primary outcome, and the anterior approach had superiority over other approaches (posterior and combined approaches) [1,17–19,29]. According to our subgroup analysis, the patient’s age might play a role in blood loss, as described by Lenga et al. [17]. Older populations have a potential for multiple comorbidities, and a higher number of vertebral levels affected causes more blood loss during the procedure [17].

Length of hospitalization

Seven studies described the length of hospitalization. The anterior approach is advantageous compared to the other groups. Thus, this finding was not significant statistically [1,16–19,25,30].

Belirgen et al. [1] reported that the posterior group had more patients treated with preoperative traction than the anterior group (11 versus eight, respectively), which prolonged the patient’s hospitalization stay and, thus, did not differ significantly. Kwon et al. [16] described that the posterior group had more infection complications, whereas no infections occurred in the anterior group, and this finding might affect the length of hospitalization.

According to the subgroup analysis, Lenga et al. [17] and Ren et al. [19] described that older patients had multiple comorbidities, more blood loss, and longer surgical duration. These factors are linear with the length of hospitalization. Ren et al. [19] also stated that the sutures in the posterior cervical incision cannot be removed as quickly as the anterior approach.

Mortality rates

Overall, no relationship was found between the mortality rate and approach type. Lenga et al. [17] reported no difference between the groups in terms of mortality rates. Moreover, Mizuno et al. [24] reported two patients (anterior approach) with Frankel grade A who died because of respiratory failure, and these patients had progressive cord swelling and ascending intramedullary high-signal intensity, which should be followed by posterior decompression.

On the contrary, Rezaee et al. [25] reported that the approach type was strongly related to the mortality rate. The one-stage combined approach had the highest early mortality compared with the other groups, and the most frequent cause of early mortality was pneumonia [25].

Limitations

Authors find it challenging to identify homogenous studies for some of the comparisons, whereas the heterogeneity between the studies and non-standardized outcomes reporting in the included studies might limit the interpretation of the pooled data. Only one study evaluated the efficacy and adverse events of the two stages combined approach; thus, this finding might be significant for the treatment of subaxial cervical spine injury. This review did not involve network meta-analysis, which did not permit a three-way comparison between the groups.

Conclusions

In this study, no difference was found in the comparisons between the anterior, posterior, and combined approaches in terms of clinical improvements and complication rates. Meanwhile, the anterior approach is superior to the posterior approach in terms of surgical duration, blood loss, and hospitalization length. However, the technique of choice is often based on surgeons’ preferences or clinical judgments, and this review offers another point of view.

Notes

Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Author Contributions

Conceptualization: MF, AR, HF, EAS; methodology: MF, AR, BU; data curation: MF, AR, AA, PW; formal analysis: AR, BU; visualization: AR; project administration: MF, AR; writing–original draft preparation: MF; writing–review and editing; MF, AR, AA; supervision: MF; and final approval of the manuscript: all authors.

Fig. 1

The Preferred Items for Systematic Review and Meta-Analysis flow diagram.

Fig. 2

(A) Risk of bias tool for non-randomized trial (ROBINS-I) graph. (B) ROBINS-I summary.

Fig. 3

(A) Risk of bias tool for randomized trial (Rob-2) graph. (B) Rob-2 summary.

Fig. 4

Anterior-only versus posterior-only subgroups of clinical improvement, complication rates, and mortality rates. M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom.

Fig. 5

Anterior-only versus posterior-only subgroups of surgical duration. SD, standard deviation; IV, inverse variance; CI, confidence interval; df, degrees of freedom.

Fig. 6

Anterior-only versus posterior-only subgroups of blood loss. SD, standard deviation; IV, inverse variance; CI, confidence interval; df, degrees of freedom.

Fig. 7

Anterior-only versus posterior-only subgroups of length of hospitalization. SD, standard deviation; IV, inverse variance; CI, confidence interval; df, degrees of freedom.

Fig. 8

Anterior-only versus combined subgroups of clinical improvement and complication rates. M-H, Mantel-Haenszel; CI, confidence interval; df, degrees of freedom.

Table 1.

Summary of the studies

Author	Study design	Age (yr)	Male gender	Trauma mechanism	Injury mechanism	Level of injury	Neurological improvement	Complication rate	Surgical duration (min)	Blood loss (mL)	Length of stay (day)	Mortality rate	Summary
El-Hajj et al. [30] (2023)	Prospective cohort	62±25	212 (292)	NA	Distraction (84), compression (36), burst (172)	NA	NA	AO (3), PO (8), AP (12)	NA	NA	AO (3±2), PO (3±3), AP (4±5)	NA	The complication rate did not differ between the groups. The longest length of hospitalization was the posterior approach group, while the shortest was the anterior one.
Lee et al. [27] (2023)	Retrospective cohort	52.1	12 (12)	MVA (7), fall (3), other (2)	Distraction (12)	C3–4 (1), C4–5 (5), C5–6 (3), C6–7 (3)	AO (2), PO (1), AP (2)	AO (0), PO (0), AP (0)	NA	NA	NA	NA	A gentle reduction via posterior approach would be safer in terms of avoiding iatrogenic cord injury in distraction-flexion injury patients.
Lenga et al. [17] (2023)	Retrospective cohort	AO (85.2±1.5), PO (82.2±1.1)	AO: 9 (13), PO: 8 (15)	NA	NA	NA	AO (82.4±10.1), PO (89.5±12.1)	AO (5), PO (6)	AO (182.5±32.1), PO (235±58.4)	AO (87.5±18.1), PO (661.5±100.1)	AO (10±6.2), PO (11.3±5.2)	AO (3), PO (3)	In the neurological outc om e , t he re wa s no difference between the groups. Surgical duration and blood loss are the significant risk factors for complications, while the patient’s age, sex, and number of segments did not relate to the complication rate.
Moawad et al. [18] (2022)	Prospective cohort	37.4	27 (40)	MVA (22), fall (18)	Distraction (40)	C3–4 (2), C4–5 (4), C5–6 (14), C6–7 (20)	AO (17), PO (10)	AO (11), PO (18)	AO (93±17.71), PO (140±20.18)	AO (100±29.15), PO (230±91.36)	NA	AO (0), PO (0)	In terms of blood loss, the anterior approach was better. While the surgical duration and neurological outcome did not differ between the groups. In terms of postoperative pain, the anterior approach was better.
Rezaee et al. [25] (2021)	Retrospective cohort	40.7±16.5	51 (72)	MVA (48), fall (11), other (13)	Distraction (52), burst (31)	C3–4 (14), C4–5 (29), C5–6 (33), C6–7 (23), C7–T1 (9)	AO (11.2±13), PO (3.1±4.9), AP (5.6±9.5)	AO (22), PO (6), AP (15)	NA	NA	NA	AO (9), PO (4), AP (6)	The neurological outcome, length of hospitalization, and complication were not related to the approach. While the early mortality rate was related to the combined approach at one stage. The two stages combined approach had the least loss of correction and the posterior approach had the most.
Ren et al. [19] (2020)	Prospective cohort	AO (53.1±14.2), PO (54.7±15.6)	AO: 63 (92), PO: 44 (67)	NA	Distraction (159)	C3–4 (20), C4–5 (49), C5–6 (55), C6–7 (31), C7–T1 (4)	AO (53), PO (36)	AO (2), PO (1)	AO (72.1±9.2), PO (93±11.3)	AO (71.5±14.6), PO (102.4±18.5)	AO (8.6±1.5), PO (13.4±2.3)	NA	Neurological outcomes did not differ between the groups. The posterior approach had greater loss of correction, blood loss, and longer surgical duration.
Liu et al. [29] (2019)	Retrospective cohort	AO (48.8±12.2), AP (47.1±11.6)	AO: 55 (63), AP: 25 (30)	MVA (24), fall (28), other (41)	Distraction (93)	C3–4 (6), C4–5 (13), C5–6 (33), C6–7 (33), C7–T1 (8)	AO (40), AP (11)	AO (0), AP (0)	AO (88.6±35), AP (274±114.7)	AO (92.5±84.0), AP (275.0±183.2)	NA	NA	The neurological outcome did not differ between the groups. The surgical duration and blood loss were better in the anterior approach group.
Al Samouly et al. [20] (2018)	Retrospective cohort	38.2±14.4	42 (60)	MVA (51), fall (9)	Distraction (60)	C3–4 (6), C4–5 (21), C5–6 (25), C6–7 (8)	AO (13), PO (14)	AO (7), PO (6)	NA	NA	NA	NA	An anterior approach was better in realignment, fusion, and risk of complications. The anterior approach was better in DFI stage 3 and 4, while the posterior approach was better for DFI stage 1 and 2.
Belirgen et al. [1] (2013)	Retrospective cohort	AO: 36 (14–62), PO: 37 (18–67)	AO: 13 (18), PO: 12 (15)	MVA (28), fall (3), other (2)	Distraction (29), compression (4)	C3–4 (2), C4–5 (5), C5–6 (11), C6–7 (12), C7–T1 (3)	AO (17), PO (11)	AO (2), PO (1)	AO (147.2±20.07), PO (210±53.71)	AO (79.7±23.54), PO (200±91.31)	AO (13.4±20.39), PO (15.9±8.74)	NA	The neurological outcomes and complications rate did not differ between the groups. Blood loss and surgical duration were better in the anterior approach group.
Song et al. [26] (2008)	Retrospective cohort	AO (47.3±11.9), AP (46.1±16.6)	40 (50)	MVA (37), fall (9), other (4)	Distraction (50)	NA	AO (6.49±18.67), AP (15.28±20.39)	AO (8), AP (0)	AO (96.68±17.70), AP(238.75±30.17)	NA	NA	NA	The fusion time was longer in the DFI stage 3 which was treated with the anterior approach than in the DFI stages 3 and 4 which was treated with the posterior approach. The surgical duration was longer in the posterior approach group, while the neurological outcomes did not differ between the groups.
Kwon et al. [16] (2007)	RCT	AO (37.5±3.6), PO (33.0±3.1)	AO: 14 (20), PO: 17 (22)	NA	Distraction (42)	C4–5 (6), C5–6 (10), C6–7 (25), C7–T1 (1)	NA	AO (2), PO (5)	AO (134), PO (103)	NA	AO (3.41±0.9), PO (5.36±3.29)	NA	The posterior approach group had a greater infection rate, although was not significant statistically. The surgical duration was longer in the posterior approach group. In terms of postoperative pain, the anterior group was better.
Kim et al. [28] (2007)	Retrospective cohort	AO (57.4), AP (43.3)	AO: 36 (47), AP: 12 (18)	MVA (36), fall (19), other (10)	Distraction (65)	C3–4 (6), C4–5 (14), C5–6 (18), C6–7 (23), C7–T1 (3)	AO (25), PO (5)	AO (0), AP (2)	NA	NA	NA	NA	The neurological and the radiological outcomes did not differ between the groups. Two patients of the combined approach group had wound infection. Anterior approach is an effective and safe compared to the combined approach for the bilateral facet dislocation.
Mizuno et al. [24] (2007)	Retrospective cohort	44.3±16.35	8 (11)	NA	Distraction (11)	NA	AO (1), PO (2), AP (2)	AO (2), PO (1), AP (1)	NA	NA	NA	NA	In the anterior approach group, 2 patients with Frankel A died due to respiratory failure. One patient had neurological deterioration due to disk herniation after posterior fixation. Postoperative kyphosis or recurrent facet dislocation did not occur in this study.

Values are presented as mean±standard deviation or number (%) unless otherwise stated.

NA, not available data; AO, anterior-only approach; PO, posterior-only approach; AP, anterior-posterior approach; MVA, motor vehicle accident; DFI, distractive flexion injury.

References

1. Belirgen M, Dlouhy BJ, Grossbach AJ, Torner JC, Hitchon PW. Surgical options in the treatment of subaxial cervical fractures: a retrospective cohort study. Clin Neurol Neurosurg 2013;115:1420–8.

2. Aebi M. Surgical treatment of upper, middle and lower cervical injuries and non-unions by anterior procedures. Eur Spine J 2010;19(Suppl 1): S33–9.

3. Joaquim AF, Patel AA. Subaxial cervical spine trauma: evaluation and surgical decision-making. Global Spine J 2014;4:63–70.

4. Torretti JA, Sengupta DK. Cervical spine trauma. Indian J Orthop 2007;41:255–67.

5. Aebi M, Zuber K, Marchesi D. Treatment of cervical spine injuries with anterior plating: indications, techniques, and results. Spine (Phila Pa 1976) 1991;16(3 Suppl): S38–45.

6. Koh YD, Lim TH, You JW, Eck J, An HS. A biomechanical comparison of modern anterior and posterior plate fixation of the cervical spine. Spine (Phila Pa 1976) 2001;26:15–21.

7. Dvorak MF, Fisher CG, Fehlings MG, et al. The surgical approach to subaxial cervical spine injuries: an evidence-based algorithm based on the SLIC classification system. Spine (Phila Pa 1976) 2007;32:2620–9.

8. Henriques T, Olerud C, Bergman A, Jonsson H Jr. Distractive flexion injuries of the subaxial cervical spine treated with anterior plate alone. J Spinal Disord Tech 2004;17:1–7.

9. Lambiris E, Kasimatis GB, Tyllianakis M, Zouboulis P, Panagiotopoulos E. Treatment of unstable lower cervical spine injuries by anterior instrumented fusion alone. J Spinal Disord Tech 2008;21:500–7.

10. Woodworth RS, Molinari WJ, Brandenstein D, Gruhn W, Molinari RW. Anterior cervical discectomy and fusion with structural allograft and plates for the treatment of unstable posterior cervical spine injuries. J Neurosurg Spine 2009;10:93–101.

11. Robertson PA, Ryan MD. Neurological deterioration after reduction of cervical subluxation: mechanical compression by disc tissue. J Bone Joint Surg Br 1992;74:224–7.

12. Olerud C, Jonsson H Jr. Compression of the cervical spine cord after reduction of fracture dislocations: report of 2 cases. Acta Orthop Scand 1991;62:599–601.

13. Reindl R, Ouellet J, Harvey EJ, Berry G, Arlet V. Anterior reduction for cervical spine dislocation. Spine (Phila Pa 1976) 2006;31(6): 648–52.

14. Ordonez BJ, Benzel EC, Naderi S, Weller SJ. Cervical facet dislocation: techniques for ventral reduction and stabilization. J Neurosurg 2000;92(1 Suppl): 18–23.

15. Du W, Wang C, Tan J, Shen B, Ni S, Zheng Y. Management of subaxial cervical facet dislocation through anterior approach monitored by spinal cord evoked potential. Spine (Phila Pa 1976) 2014;39:48–52.

16. Kwon BK, Fisher CG, Boyd MC, et al. A prospective randomized controlled trial of anterior compared with posterior stabilization for unilateral facet injuries of the cervical spine. J Neurosurg Spine 2007;7:1–12.

17. Lenga P, Gulec G, Kiening K, Unterberg AW, Ishak B. Anterior cervical discectomy and fusion with plate versus posterior screw fixation after traumatic subaxial fractures in octogenarians: complications and outcomes with a 2-year follow-up. Acta Neurochir (Wien) 2023;165:1145–54.

18. Moawad AM, El-Sawy MM. Subaxial cervical spine ligamentous instability anterior versus posterior fixation: prospective nonrandomized study. Egypt J Neurol Psychiatr Neurosurg 2022;58:120.

19. Ren C, Qin R, Wang P, Wang P. Comparison of anterior and posterior approaches for treatment of traumatic cervical dislocation combined with spinal cord injury: minimum 10-year follow-up. Sci Rep 2020;10:10346.

20. Al Samouly HM, Taha AM. Comparative study of surgical approaches for distractive flexion injuries of sub-axial cervical spine. Open J Mod Neurosurg 2018;8:342–51.

21. Moher D, Shamseer L, Clarke M, et al. Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) 2015 statement. Syst Rev 2015;4:1.

22. Higgins JP, Li T, Deeks JJ. Choosing effect measures and computing estimates of effect. Higgins JP, Thomas J, Chandler J, . In: Cochrane Handbook for Systematic Reviews of Interventions version 6.1 (updated September 2020) [Internet] London: Cochrane. 2020 [cited 2023 Aug 10]. Available from: https://www.training.cochrane.org/handbook

23. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.

24. Mizuno J, Nakagawa H, Inoue T, Nonaka Y, Song J, Romli TM. Spinal instrumentation for interfacet locking injuries of the subaxial cervical spine. J Clin Neurosci 2007;14:49–52.

25. Rezaee H, Keykhosravi E, Mashhadinejad M, Pishjoo M. Comparison of anterior, posterior, and combined surgical approaches on the outcomes of patients suffering from subaxial cervical spine injuries. Bull Emerg Trauma 2021;9:133–7.

26. Song KJ, Lee KB. Anterior versus combined anterior and posterior fixation/fusion in the treatment of distraction-flexion injury in the lower cervical spine. J Clin Neurosci 2008;15:36–42.

27. Lee KS, Park EJ, Min WK. Surgical outcome of locked facet in distractive flexion injury of the subaxial cervical spine: single institution retrospective study. Medicine (Baltimore) 2023;102:e33028.

28. Kim KH, Cho DC, Sung JK. The management of bilateral interfacetal dislocation with anterior fixation in cervical spine: comparison with combined antero-posterior fixation. J Korean Neurosurg Soc 2007;42:305–10.

29. Liu K, Zhang Z. Comparison of a novel anterior-only approach and the conventional posterior-anterior approach for cervical facet dislocation: a retrospective study. Eur Spine J 2019;28:2380–9.

30. El-Hajj VG, Singh A, Blixt S, Edstrom E, Elmi-Terander A, Gerdhem P. Evolution of patient-reported outcome measures, 1, 2, and 5 years after surgery for subaxial cervical spine fractures, a nation-wide registry study. Spine J 2023;23:1182–8.

31. Sterne JA, Savovic J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898.