Approach to junctional fractures in spine: A systematic review and meta-analysis

Abstract

Background

The management of thoracolumbar burst fractures presents a challenge due to the absence of consensus on the most effective surgical approach. Surgeons commonly employ anterior, posterior, or combined anterior-posterior instrumentation methods to achieve fracture reduction, stabilization, and neural canal decompression. Despite the availability of these techniques, there is a lack of comparative studies evaluating their efficacy. This gap in the literature underscores the necessity for research to thoroughly assess and compare the various surgical approaches for thoracolumbar spine fractures. The present study aims to compare and critically evaluate the different approaches to surgical management of thoracolumbar spine fractures in order to identify the best treatment.

Material and method

A literature search was performed in 13 electronic databases for articles published between 2000 and August 31, 2023. Two separate reviews were conducted to ensure thoroughness and reliability in article selection. Ultimately, thirteen articles meeting the inclusion criteria were chosen, comprising nine non-randomized control trials (NRCTs) and four randomized control trials (RCTs). To assess the risk of bias in the NRCTs, the Newcastle-Ottawa scale was utilized, while the ROB-b tool was applied to evaluate the RCTs.

Results

Thirteen studies meeting the predefined inclusion and exclusion criteria were selected, comprising nine non-randomized control trials and four randomized control trials. A meta-analysis was conducted, revealing that patients undergoing the posterior approach tended to experience longer surgery durations (anterior vs posterior = 5.38) and higher blood loss (anterior vs posterior = 26.21) compared to those undergoing the anterior or combined anterior-posterior approaches. Additionally, immediate postoperative improvement in kyphotic correction (anterior vs posterior = 2.15 and anterior – posterior vs posterior-2.97) was observed in the posterior approach group. However, follow-up assessments indicated a subsequent loss of kyphotic angle in this group (anterior vs posterior = 1.41 and anterior – posterior vs posterior-2.00).The systematic review comprises only seven studies comparing anterior and posterior approaches and three studies evaluating posterior and combined approaches. Furthermore, the utilization of different clinical scales across the studies complicates drawing definitive conclusions. Also, heterogeneity of the population, including variations in comorbidities, types of fractures, follow-up duration, timing of intervention, surgeon expertise, and surgical techniques were noted across the included studies. Additionally, there is variability in sample sizes, ages, genders, and follow-up periods. Most of the included studies in the present study were performed on ventral cord compression and kyphosis correction cases rather than simple junctional fracture cases.

Conclusion

Each surgical approach possesses its own set of advantages and disadvantages. However, the lack of consensus and standardized protocols underscores the need for further research to establish definitive guidelines for surgical decision-making in thoracolumbar spine fractures.

1. Introduction

The thoracolumbar junction (T10-L2) is particularly prone to traumatic injury due to its distinct biomechanical characteristics. Positioned between the rigid and less mobile thoracic spine, which includes the ribs and sternum, and the more flexible lumbar spine below, it represents a transitional zone with unique vulnerabilities.¹

Injuries to the thoracolumbar spine are prevalent, often stemming from high-energy incidents, accounting for a substantial portion (40 %–80 %) of cases. These incidents include motor vehicle accidents, falls from significant heights, recreational mishaps, and occupational injuries.^1,2 Furthermore, approximately 25 % of patients with such fractures sustain additional spine fractures elsewhere or experience injuries to two or more other organ systems due to the force of impact. Commonly associated injuries encompass pneumothorax, haemothorax, rib fractures, bronchial disruption, myocardial or pulmonary contusion, injuries to major blood vessels, hemopericardium, cardiac tamponade, and diaphragmatic rupture.³

Magerl et al. (1994)⁴ conducted a study analysing 1445 consecutive thoracolumbar injuries, revealing that the upper and lower ends of the thoracolumbar spine, as well as the T10 level, were the least frequently injured sites. They introduced a comprehensive AO classification system for thoracolumbar fractures, which categorizes them based on the mechanism of injury and the morphological pattern of the fracture. Type A fractures (vertebral body compression) accounted for 66.1 % of cases, while type B fractures (involving anterior and posterior element injuries with distraction) comprised 14.5 %, and type C fractures (involving anterior and posterior element injuries with rotation) constituted 19.4 %. Additionally, regarding neurological deficits, the overall incidence was 22 %, with the highest occurrence observed in type C fractures at 55 %.

The literature indicates that around 27 % of patients with thoracolumbar junction injuries experience significant disability, deformity, and neurological deficits.^{1, 2, 3} Consequently, ongoing efforts have been directed towards developing effective treatment strategies. The management of these fractures is geared towards achieving long-term correction of vertebral deformity and either restoring or preserving neurological function.^5,6,7

In cases of stable thoracolumbar spine fractures without neurological deficits, conservative management typically involves the use of a well-molded brace or hyperextension cast. However, if a neurological deficit is present or if the injury pattern is deemed unstable, surgical management is recommended. Surgical intervention is also favored for patients who are unable to tolerate extended periods in a cast, as well as those with multiple extremity injuries or skin lesions.^{5, 6, 8, 9}

The surgical management of spinal conditions has indeed experienced significant advancements in recent decades.⁶ Similarly, various approaches are utilized to address thoracolumbar fractures, including the anterior approach, posterior approach, and combined anterior-posterior approach. However, each of these approaches and surgical interventions comes with its own distinct advantages and disadvantages.^{7, 10, 11}

Various studies have explored the management of thoracolumbar fractures, yet few have comprehensively evaluated and compared the outcomes of different surgical approaches. A recent systematic literature review by Verlaan et al..¹⁰ highlighted the inadequacy and scarcity of studies addressing surgical approaches for thoracolumbar injuries. Furthermore, controversies persist regarding the optimal approach and type of surgery. Oprel et al.,¹¹ in their systematic review, exclusively compared posterior approaches with combined anterior-posterior approaches, neglecting to include comparisons with the anterior approach. Consequently, there remains a significant gap in the literature regarding the comparison of all three approaches. Given the absence of a consensus on managing these fractures, there is a pressing need to address this research gap. Therefore, the present study aims to compare and critically evaluate the different surgical approaches in the management of thoracolumbar spine fractures.

2. Material and methods

2.1. Protocol

This systematic review adhered to the guidelines set forth in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, ensuring transparent reporting of the review process. Furthermore, the systematic review protocol was registered on the PROSPERO platform, enhancing transparency and accountability in the research process.

2.2. Focused question

Which is the effective approach in the surgical management of thoracolumbar spine fractures?

P	Patients with thoracolumbar spine fractures
I	Posterior approach
C	Anterior approach, combined approach
O	Duration of surgery, kyphotic angle correction, blood loss

2.3. Information sources

The data search primarily involved conducting electronic searches across multiple well-known databases, including PubMed, Google Scholar, Scopus, Embase, EBSCO, Web of Science, Clinical Trials Registry India, The Cochrane Oral Health Group's Trials Register, The Cochrane Central Register of Controlled Trials (CENTRAL), The PROSPERO International prospective register of systematic reviews, and the National Institutes of Health Trials Register. The search focused on articles published between January 1, 2000, and August 31, 2023, ensuring comprehensive coverage of literature comparing different approaches in the surgical management of thoracolumbar spine fractures. Only articles published in English or those with a detailed English summary were considered. Additionally, references to both included and excluded studies were reviewed to identify any additional relevant reports.

The literature search utilized a specific set of keyword combinations to ensure comprehensive coverage across all 12 databases. These combinations of MeSH term included "THORACOLUMBAR JUNCTION FRACTURE" AND "SURGICAL MANAGEMENT", as well as "JUNCTIONAL FRACTURES" AND "SURGICAL APPROACHES" OR "SURGICAL MANAGEMENT".

2.4. Eligibility criteria

The inclusion criteria were as follows.

• 1.Original studies
• 2.Full-text, English-language studies
• 3.Case control trials; Randomized and non-randomized, interventional studies, or cohort studies were included.
• 4.Patients with thoracolumbar spine fractures undergoing surgical management
• 5.Articles reporting information about the outcomes in terms of clinical, radiological and surgical outcome of thoracolumbar spine fractures

The exclusion criteria were as follows.

• 1.Systematic reviews with or without meta-analysis, case reports, case series, and expert opinions will be excluded.
• 2.Animal research or in vitro studies.
• 3.Studies focusing on surgical management of thoracolumbar fractures lack information on approaches or outcomes.
• 4.Only abstracts or if full text unavailable

2.5. Study selection and data extraction

The study selection process will consist of two stages. Initially, two reviewers will independently evaluate the chosen articles based on their titles and abstracts. Using predefined selection criteria, the studies will be categorized as either relevant or irrelevant. Subsequently, the full texts of the relevant studies will undergo a second and final analysis, again applying the same criteria for labeling. And irrelevant articles were excluded. If needed, the third author was called in.

From the finalized papers, the following information was retrieved: 1) Study characteristics, such as the authors, country, and publication year; 2) study type and sample; 3) total sample size and sample groups; 4) follow-up period; 5) findings of the study (primary and additional outcomes); and 6) conclusion.

2.6. Risk of bias and quality of assessment

The randomized controlled studies were assessed using the Cochrane risk-of-bias tool, RoB 2. The non-randomized controlled studies were evaluated using the Newcastle-Ottawa quality assessment scale.

2.7. Statistical analysis

Meta-analyses were conducted using a random effects model (REM) in RevMan 5.4 (developed by The Nordic Cochrane Centre, Copenhagen). The primary outcomes considered were kyphotic angle correction, length of hospital stay, and blood loss in subjects treated with three different surgical approaches for managing thoracolumbar spine fractures.

3. Result

A low kappa value (0.85) indicates acceptable agreement among the three investigators. Initially, electronic searches yielded 1605 articles, which were then analyzed by the two investigators. The entire process of study selection is outlined in the PRISMA flow chart depicted in Fig. 1. Subsequently, 13 articles were selected to be included in the systematic review.

Fig. 1.

Flowchart of study selection.

Among the 13 included studies, 4 were randomized controlled trials (RCTs) and 9 were non-randomized controlled trials (non-RCTs). The summary of all the included studies were tabulated (Table 1).

Table 1.

characteristics of included studies.

S. No	Author, year, & Study design	Sample	Follow up period months	CLINICAL OUTCOME	RADIOLOGICAL OUTCOME	SURGICAL OUTCOME	CONCLUSION
1.	Hitchon et al.,12 2006 Retrospective cohort	Total-63 Male-45 (71.4 %) Female −18 (28.6 %) Mean age-42 (22–64) Group I (Posterior approach)- 25 patient Group II (Anterolateral approach) - 38	1.8 years	Rand SF-36 questionnaire Frankel scores Preoperative Anterolateral −3.7 ± 1.1 posterior- 3.5 ± 1.4 Postoperative Anterolateral −4.2 ± 0.8 posterior- 9.8 ± 9.4	Angular Deformity Preoperative Anterolateral −11.9 ± 9.7° posterior- 4.1 ± 7.1° Postoperative Anterolateral −4.5 ± 9.3° posterior- 9.8 ± 9.4°	Operative Time (Min) Anterolateral-285 min posterior- 203 min Hospital stay 14.6 days	When burst fractures are accompanied by neurological deficits and angular deformities, opting for the anterior approach with anterior-column reconstruction seems to present advantages.
2.	Wood et al.13 2005 RCT	Total-38 Mean age-42 Group I (Posterior approach)- 18 patient Group II(Anterior approach)-20	44 months	Visual Analog Pain Scale Preoperative Anterior −7.1 posterior- 5.8 Postoperative Anterior −3.1 posterior- 3.6 Roland and Morris preinjury functional disability scores Anterior −1.4 posterior- 1.2 Oswestry disability score Anterior −21 posterior- 20.1	Angular Deformity Preoperative Anterior −10.475° posterior- 10.75° Postoperative Anterior −11.05° posterior- 12.5°	Operative Time (Min) Anterior-233 min Posterior −205 min Hospital stay Posterior −10.1days Anterior −7.2 days Blood loss Anterior- 780 ml Posterior-460 ml Complications For posterior approaches, complications may include posterior wound dehiscence, instrumentation or bone failure, urinary tract infection, instrumentation breakage, deep wound infection, pseudarthrosis, and seroma formation. In contrast, anterior approaches may lead to complications such as prolonged neuropraxia, postoperative diabetic ketoacidosis, and pseudarthrosis.	The complication rate after anterior treatment of burst fractures may be notably lower than that following posterior instrumentation and fusion.
3.	Been et al.,14 1999 RCT	Total-46 Mean age-44 (24–64) Group I (Anterior decompression/stabilization with iliac bone graft + posterior stabilization)- 27 patient Group II (Posterior distraction instrumentation and stabilization wit AO internal fixator) - 19	6 years	Pain scores group I −23/27 no pain (85 %) Group II- 15/19 no pain (68 %)	Kyphotic Deformity Postoperative Group I-3.30 (SD = 7.70) Group II- 4.10 (SD = 12.40).		No significant correlation between
4.	Briem et al.15 2004 Prospective RCT	Total-20 Mean age-49.6 years Group I (primary dorsal stabilization with fixateur interne + ventral fusion with autologous bone graft)- 10 Group II (posterior stabilization with bi-segmental fixateur interne, no bone graft.) - 10	47.5 months	SF-36 quality of life questionnaire	Cobb angle Postoperative Group I-2.80° Group II-2.1°	Loss of vertebral body height (LOVBH) (at 4yrs) Group I-0.77 Group II-0.88	SF-36 and age, loss of correction, Cobb angle and sagittal index.
5.	Wu et al.,16 2013 Retrospective cohort	Total-94 Male-56 Female −38 Mean age-44.7 years (range 18–77 years) Group I (Posterior approach)- 35 Group II (anterior approach) - 24 Group III (para spinal approach) - 32	18 months (10–36 months)	Postoperative Oswestry disability score Anterior −13 ± 2.4 posterior- 12 ± 0.7 Paraspinal–6±0.5	Cobb angle Preoperative Anterior −21.2 ± 5.7 posterior- 20.6 ± 5.3 Paraspinal-20.4 ± 6.2 Postoperative (1yr) Anterior −4.7 ± 1.1 posterior- 4.6 ± 1.3 Paraspinal-4.6 ± 1.4	Operative Time (Min) Anterior-176.3 min ± 20.7 min, Posterior −94.1 min ± 13.7 min Para spinal −94.1min ± 13.7min length of incision Anterior-18.6 ± 2.4 cm Posterior −15.4 ± 2.1 cm Paraspinal- 9.3 mm ± 0.7 mm Blood loss Anterior- 255.1 ± 38.4 ml Posterior-143.3 ml ± 28.3 ml Paraspinal-86.7 ml ± 20.0 ml	Paraspinal approach provides us a new alternative in managing thoracolumbar burst fractures
6.	Boswell et al.17 2012 Retrospective	Total-41 Burst fracture −12 Male-19 Female −22 Mean age-44.7 years (range 18–77 years) Group I (Posterior approach)- 41	27.7 (0.5–60 months)	Postoperative Oswestry disability score Posterior- 13.9 Post-operative Back Pain Index Posterior-16	–	–	Open vertebroplasty can be a safe and effective treatment for selected patients who are not candidates for conventional spinal fusion techniques
7	Stancic et al.,18 2001 Prospective, non-randomized trial	Total-25 Male-15 Female −10 Mean age-36 years (range 18–53 years) Group I (Posterior approach)- 12 Group II (anterior approach) - 13	12 months	ASIA/IMSOP grading scale Anterior-7 Posterior- 5 Prolo economic and functional rating scale Anterior-8/10 Posterior-9/10	–	Operative Time (Min) Anterior-250 (200–295) min Posterior −174 (130–215) min Hospital stay Anterior −17.3 (12–25) days Posterior −10.9 (7–15) days length of incision Anterior-17.8 (16–20) cm Posterior −13.2 (12–15) cm Blood loss Anterior- 1362 (1150–1500) ml Posterior-750 (500-1100) ml Cosmetic results Anterior - 3.4 (2–5) Posterior −4.1 (3–5)	Posterior surgery may be advised for emergency neuro decompression and stabilization of unstable thoracolumbar fractures due to its shorter operation time and reduced blood loss.
8	Knop et al.19 1996 Prospective Multicenter Study,	Total - 682 Male - 436 Female −246 Mean age - 39.5 years (range 7–83).	27 months	Hannover spine score: Posterior group Pre −93.3, Postoperative - 67.4, Follow up - 69.9 Anterior-posterior group Pre- 94.0, Post op- 63.1, Follow up- 65.1	Kyphotic deformity: Preoperative Anterior - 13.50 Posterior −14.90 Anterior-posterior - 16.10 Kyphotic at follow-up: Anterior - 4.90 Posterior - 1.90 Anterior-posterior- 5.80	Operation time Anterior- 218 min (range 108–520), Posterior - 134 min (range 30–390), anterior-posterior - 254 min (range 80–562). Blood loss: Anterior - 876 cc (range 200–5500), Posterior - 828 cc (range 0–8000), Anterior-posterior - 1387 cc (range 200–8800),
9	Schnee et al.20 1995 Retrospective Observational Study	Total- 25 Male −18 Female- 7 Mean age 34.4 years (range 14–59 years),	16 months	Minimal pain reported	Kyphotic deformities Anterior group Pre operative −16.80 Immediate Postoperative −2.90 Posterior group Pre operative - 7.50 Postoperative- 100 Anterior-posterior Pre operative −18.30 Postoperative- 1.00	-
10	Xu et al.21 2011 Retrospective	Total-48 Male-30 Female −18 Mean age-39.2 years (range, 23–52 years) Group I (anterior approach)- 48	32.4 months	minimal or no pain reported	Cobbs angle Preoperative Anterior −21.4 Postoperative Anterior −5.6°	Operative Time (Min) Anterior-224 min Hospital stay Anterior −7.5 days Blood loss Anterior- 950 ml	Performing anterior corpectomy and stability reconstruction using a titanic mesh autograft and Z-plate fixation in a single incision is deemed an effective technique for treating unstable thoracolumbar burst fractures, both with and without neurological deficits.
11	Cengiz et al.22 2008 RCT	Total-27 Male-18 Female −9 Mean age-41.4 ± 14.71 years Group I (Posterior approach with immediate stabilization within 8 h)- 12 Group II (Posterior approach with late stabilization within 3–15 days)- 15	14.5 (12–20) months	Preoperative Glascow coma score Group I-14.50 Group II -15.0 ASIA grading scale Pre-op Group I-2.00 ± 1.04 Group II -1.93 ± 0.96 Postoperative Group I- 3.67 ± 1.56 Group II -2.20 ± 1.15	–	Hospital stay Group I-12.50 days Group II -26.00 days Complications In group II -lung failure Group I-nil	Early surgical intervention could enhance neurological recovery, shorten hospitalization duration, and potentially reduce systemic complications in patients with thoracolumbar spinal cord injuries.
12	Shen et al. 201523 Obser national cohort	Total-129 Male-90 Female −39 Mean age-38.5 ± 9.1 years Group I (nonoperative management)- 104 Group II (anterior or Posterior approach)- 25	38.8 ± 13.9 months	Visual Analog Pain Scale Preoperative Group I-3.6 ± 1.3 Group II- 6.5 ± 1.9 Postoperative Anterior −3.1 posterior- 3.6 Roland and Morris preinjury functional disability scores Anterior −1.4 posterior- 1.2 Oswestry disability score Anterior −21 posterior- 20.1	IPD (Interpedicular distancea (IPD) Group I-11.2 ± 6.6 %, Group II-29.7 ± 5.5 % Kyphotic angle (KA) Group I-11.3 ± 4.2° Group II-22.7 ± 8.3° Loss of vertebral body height (LOVBH) Group I-29.4 ± 7.7 % Group II-31.5 ± 7.3 %	–	Conservative treatment should be decided with caution in patients with greater VAS scores or IPD
13	Danisa et al.24 1995 Retrospective cohort	Total-130 Mean age-35.4 years (range 19–62 years) Group I (combined anterior-posterior)- 6 Male-4 Female −2 Group II (anterior)- 16 Male-11 Female −5 Group III (posterior)- 27 Male-19 Female −8	27 months (range 6–54 months)	Dennis pain Scale Anterior −67 % Posterior −35 % Anterior posterior −40 %	Angular Deformity Preoperative Anterior −16.1° posterior- 15.2° Anterior-posterior- 26.0° Postoperative Anterior −19.8° posterior- 9.5° Anterior posterior −18.5°	Operative Time (Min) Anterior-438 min (±60 min) Posterior −219 min (±61 min) Combined anterior-posterior −569 min (±121 min) Hospital stay Anterior −13 days (±4.5 days) Posterior −10 days (±6.1 days) Combined anterior-posterior −22 days (±7.0 days) Blood loss Anterior- 1878 cc (±777 cc) Posterior-1103 cc (±793 cc) Anterior posterior −2541 cc (±1439 cc) Postoperative Complications Anterior -painful iliac crest bone graft scar, left apical pneumothoraces, fractured inferior Kaneda screw Posterior -deep back wound infec-tions, deep venous thrombosis Anterior posterior -iatrogenic laceration of the thoracic duct, apical pneumothorax after chest tube removal, osteoporotic patient had an L-1 burst fracture	Posterior instrumentation and fusion represent a safe, effective, and comparatively cost-effective treatment option for acute unstable thoracolumbar burst fractures.

Total number of patients −1368.

In terms of comparisons made within the studies, three of them compared all three surgical approaches (anterior, posterior, and combined anterior-posterior), while five studies specifically compared the anterior and posterior approaches. Additionally, two studies compared two different surgical techniques with the posterior approach. Notably, one study was an observational investigation of the anterior approach alone. Furthermore, there was a single study that compared the anterior approach with the anterolateral approach.

The considered studies consisted of 1368 patients. The number of participants in each study varied between 20 and 65 individuals. No gender distribution was reported in two studies, whereas the remaining eleven studies had male predominance. Regarding follow-up periods, the duration varied across studies, ranging from 12 months to 6 years.

A Rand SF-36 questionnaire to evaluate preoperative and postoperative subjective symptoms like pain, neurological deficit, and quality of life was done in two studies. Other scales or grading systems used in the included studies are the VAS pain scale, Oswestry disability score, Roland and Morris preinjury functional disability scores, Frankel scores, ASIA/IMSOP grading scale, Prolo economic and functional rating scale, Hannover spine score, and Dennis pain scale. Pain scores and disability scores are lower in the posterior approach group compared to other approaches. The angular deformity outcome was assessed in eleven studies. In surgical outcomes, operative time, hospital stay, blood loss, and complications in the follow-up period were mentioned in ten articles.

3.1. Risk of bias and quality assessment

Table 2, Fig. 2, Fig. 3 provide a summary of the overall bias assessment using the Newcastle-Ottawa risk of bias tool for the non-randomized controlled trials (non-RCTs). Notably, the risk of bias was low.

Table 2.

Risk of bias assessment for cohort studies: Newcastle–Ottawa scale cohort studies.

References	Selection			Comparability		Outcome
	Case definition adequate	Representation of cohort	Selection of controls	Control of confounders cases and controls (smoking)	Control of confounders cases and controls (alcohol)	Assessment of outcome	Length of follow-up	Loss to follow-up	Total Score
Hitchon et al. (2006	Y		Y	Y	Y	Y			5
Xu et al. (2011)	Y		Y			Y			3
Knop et al. (1996)	Y	Y	Y	Y		Y			5
Wu, (2013)	Y	Y	y		Y	Y			5
Boswell et al. (2012)	Y		Y	Y		Y			4
Stancic, (2001)	Y	Y	Y		Y	Y			5
Schnee et al.2 (1995)	y		Y	y		Y			4
Danisa et al. (1995)	Y	Y	Y			Y			4
Shen et al. (2015)	Y	Y	Y			Y			4

Fig. 2.

Overall Risk of bias summary for non-RCT studies.

Fig. 3.

Risk of bias for individual non-RCT studies.

In contrast, the randomized controlled trials (RCTs) were evaluated using the ROB-b tool, revealing a low risk of bias in each individual study. However, the overall risk of bias was high, as illustrated in Fig. 4, Fig. 5.

Fig. 4.

Risk of bias summary for RCT studies.

Fig. 5.

Risk of bias for individual RCT studies.

3.2. Meta-analysis

Meta-analyses were performed comparing anterior approaches vs. posterior approaches and combined approaches vs. posterior approaches.

• I.
  Anterior vs posterior approach

  • A. Based on Duration of surgery

Seven studies (Danisa et al., Xu et al., Knop et al., Hitchon et al., Wu et al., Stancic et al., and Wood et al.) assessed duration of surgery were included. The results indicated that patients undergoing the posterior approach had significantly longer surgery durations compared to those undergoing the anterior approach (95 % CI = 68.35–146.63; Z value = 5.38; p < 0.0001) (I² = 100 %; random effects model (REM), as illustrated in Fig. 6. Funnel plot revealed an asymmetrical distribution of studies, suggesting a higher likelihood of publication bias, as depicted in Fig. 7.

• B.Based on kyphotic correction at immediate post-op:

Fig. 6.

Forest plot depicting comparison of anterior versus posterior approaches for the Duration of surgery parameters.

Fig. 7.

funnel plot assessing the publication bias in studies comparing the anterior versus posterior approaches for the Duration of surgery parameter.

Six studies (Danisa et al., Hitchon et al., Knop et al., Schnee et al., Wood et al., and Wu et al.) evaluated kyphotic correction at immediate post-operative assessment. The analysis revealed that patients treated with the posterior approach demonstrated significantly higher improvement in kyphotic correction at immediate post-operative assessment compared to those treated with the anterior approach, (95 % CI = 0.33–7.33; Z value = 2.15; p = 0.03) (I² = 96 %; REM), as shown in Fig. 8. Funnel plot reveals an asymmetrical distribution indicating a higher likelihood of publication bias, as depicted in Fig. 9.

• C.Based on kyphotic correction at follow-up post-op period:

Fig. 8.

Forest plot depicting comparison of anterior versus posterior approaches for the kyphotic correction at immediate post-op parameters.

Fig. 9.

funnel plot assessing the publication bias in studies comparing the anterior versus posterior approaches Based on kyphotic correction at immediate post-op parameter.

Eight studies (Been et al., Briem et al., Danisa et al., Hitchon et al., Knop et al., Schnee et al., Wood et al., and Wu et al.) evaluated kyphotic correction at the follow-up post-operative period. Interestingly, the analysis revealed that patients treated with the anterior approach demonstrated greater kyphotic correction at the follow-up postoperative period compared to those treated with the posterior approach (95 % CI = −49.10-8.06; Z value = 1.41; p = 0.16), (I² = 100 %; REM), as illustrated in Fig. 10. Funnel plot indicates an asymmetrical distribution and suggests a higher possibility of publication bias, as shown in Fig. 11.

• D.Based on blood loss during surgery:

Fig. 10.

Forest plot depicting comparison of anterior versus posterior approaches for the kyphotic correction at follow up post-op period parameter.

Fig. 11.

funnel plot assessing the publication bias in studies comparing the anterior versus posterior approaches Based on kyphotic correction at follow up post-op period parameter.

Four studies (Danisa et al., Stancic et al., Wood et al., and Wu et al.) assessed estimated blood loss during surgery. The analysis revealed that patients treated with the posterior approach exhibited greater blood loss during surgery compared to those treated with the anterior approach, with a mean difference of 26.21 (95 % CI = −11.79–64.21; Z value = 1.35; p = 0.18), (I² = 99 %; REM), as shown in Fig. 12. Funnel plot demonstrated an asymmetrical distribution indicating a higher likelihood of publication bias, as depicted in Fig. 13.

Fig. 12.

Forest plot depicting comparison of anterior versus posterior approaches based on blood loss during surgery parameter.

Fig. 13.

funnel plot assessing the publication bias in studies comparing the anterior versus posterior approaches based on blood loss during surgery parameter.

3.3. Combined anterior-posterior vs posterior approach

• A.Based on kyphotic correction at immediate post-op:

Three studies (Danisa et al., Knop et al., and Schnee et al.) comparing kyphotic correction at immediate post-operative assessment. The analysis revealed that patients treated with the posterior approach demonstrated significantly greater kyphotic correction at immediate post-operative assessment compared to those treated with the combined approach (95 % CI = 2.75–13.39; Z value = 2.97; p = 0.003), (I² = 79 %; REM) as illustrated in Fig. 14. The difference between the two groups was statistically significant. Funnel plot indicates an asymmetrical distribution and suggests a higher likelihood of publication bias, as depicted in Fig. 15.

• B.Based on kyphotic correction at follow-up post-op period:

Fig. 14.

Forest plot depicting comparison of combined anterior-posterior versus posterior approaches for the kyphotic correction at immediate post-op parameters.

Fig. 15.

funnel plot assessing the publication bias in studies comparing the combined versus posterior approaches Based on kyphotic correction at immediate post-op parameter.

Three studies (Danisa et al., Knop et al., and Schnee et al.) assessed kyphotic correction at the follow-up post-operative period. The analysis revealed that patients treated with the posterior approach exhibited statistically significant higher kyphotic correction at the follow-up post-operative period compared to those treated with the combined anterior-posterior approach, (95 % CI, 0.09–7.91; Z value = 2.00; p = 0.05), (I² = 88 %; REM) as depicted in Fig. 16. An asymmetrical distribution suggesting a higher likelihood of publication bias was noted in the funnel plot, as shown in Fig. 17.

Fig. 16.

Forest plot depicting comparison of combined anterior-posterior versus posterior approaches for the kyphotic correction at follow up post-op period parameters.

Fig. 17.

funnel plot assessing the publication bias in studies comparing the combined versus posterior approaches Based on kyphotic correction at follow up post-op period parameter.

4. Discussion

The medical community continues to debate the optimal surgical approach for treating thoracolumbar fractures. Surgeons frequently rely on their personal experience rather than carefully weighing the specific indications, advantages, and drawbacks of each approach when making their choice. There is a pressing need for an evidence-based approach to systematically enumerate and critically evaluate the various surgical approaches available. Establishing a protocol for the selection of surgical approaches for thoracolumbar spine fractures is essential to ensuring optimal patient outcomes and standardizing treatment practices. By conducting thorough research and synthesizing existing evidence, we can develop guidelines that aid surgeons in making informed decisions tailored to individual patient needs and fracture characteristics.

In our current study, it's important to acknowledge the methodological differences among the included studies, which could significantly impact the results. Out of the total, seven studies were observational cohort studies, while only four were randomized controlled trials (RCTs). Additionally, one of the studies stood out as a multicenter study with a substantial sample size, potentially enhancing the robustness and generalizability of the findings.

The results of our analysis revealed notable variations in outcomes across different surgical approaches. Specifically, mean volume blood loss, total hospitalization, and physician charges were found to be significantly higher in cases where the anterior or combined anterior-posterior approach was utilized. Conversely, operative time and postoperative hospital stay were observed to be significantly longer when employing the posterior approach or combined anterior-posterior approach.

The anterior approach to spinal surgery indeed relieves compression of the spinal canal and restores anterior column stability, which can be beneficial in certain cases.⁶ However, its complexity arises from the involvement of organs in the chest and abdomen, major blood vessels, and other vital structures, necessitating a skilled surgeon. Additionally, the resection of vertebrae through this approach can be more traumatic, leading to increased complications.^{6, 7, 10}

Studies, such as the one conducted by Wu et al..¹⁶, have shown that the anterior approach tends to have longer surgery durations, increased incision lengths, and greater blood loss compared to the posterior approach. These differences are statistically significant and are important factors to consider when selecting the appropriate surgical approach.

While the anterior approach may be indicated for severe Denis type B fractures, it may not be suitable for Denis type A fractures due to its invasiveness and associated risks.¹⁶ Overall, the decision to utilize the anterior approach should be carefully weighed based on the specific needs and circumstances of each patient.

The posterior approach to spinal surgery offers several advantages, including its safety profile, avoidance of vital visceral and vascular structures, and generally satisfactory clinical outcomes. This approach is suitable for various indications, including distraction injuries without neural compression, isolated deficits of nerve root with intact posterior ligaments, and different scenarios involving intact or disrupted posterior ligaments, along with neurological status.

The results of the present Meta-analysis indicate that the posterior approach typically results in longer surgery durations and high blood loss. In contrast, several studies reveal that application of the posterior approach in simple junctional fractures has shown less blood loss and lesser operating time. This could be attributed to avoidance of vital viceral and vascular structures.^{19, 20, 21, 22, 23, 25, 24}

Most of the included studies in the present study were performed on ventral cord compression and kyphosis correction cases rather than simple junctional fracture cases. These reasons could have caused contradictory results in the present Meta-analysis.

Signifacnt improvement in the kypotic angle were noted in the posterior approach group compared to anterior and combined anterior-posterior approach.

Meta-analyses have shown that the posterior approach typically results in shorter surgery durations, less blood loss, and significant improvements in the kyphotic angle. However, there may be some challenges with maintaining kyphotic correction over time. Studies by Wu et al.¹⁶ and Hitchen et al.¹² have reported similar findings regarding the long-term progression of kyphotic angulation despite initial improvements.

The anterior approach seems to offer advantages in decompression and vertebral column reconstruction, particularly when combined with a posterior approach for reconstructing the tension band. However, the meta-analysis results showed no significant differences in terms of all outcomes with other approaches. The clinical indications of Anterior approach are, 1. In cases with osteoporosis and requiring internal fixation anteriorly and posteriorly, 2. In locations where anterior instrumentation is not safe i.e. in low lumbar or high thoracic because of the presence of anatomical structures.

The results of the present study highlights that each approach has its own pros and cons. The surgeon should consider the medical and trauma status of the patient, type of fracture, and economy before deciding on the approaches. The present study emphasizes the lacking and inadequate criteria or no definite protocol for decision-making on the thoracolumbar approaches. Also, there are insufficient randomized control studies on the focused question, so the results of the present study must be considered with caution, taking this into account.

The drawbacks of this study is the heterogeneity of the population, including variations in comorbidities, types of fractures, follow-up duration, timing of intervention, surgeon expertise, and surgical techniques across the included studies. Additionally, there is variability in sample sizes, ages, genders, and follow-up periods. Also, most of the included studies in the present study were performed on ventral cord compression and kyphosis correction cases rather than simple junctional fracture cases which could have caused longer duration of surgery and higher blood loss.

The systematic review comprises only seven studies comparing anterior and posterior approaches and three studies evaluating posterior and combined approaches. Furthermore, the utilization of different clinical scales across the studies complicates drawing definitive conclusions.

Despite these challenges, this study represents the first comprehensive appraisal comparing various approaches to managing thoracolumbar spine fractures. Although individual studies generally exhibit a low risk of bias, issues such as randomization and confounding biases are noted. Despite its limitations, this review presents the most recent and pertinent evidence available, which can serve as a valuable resource for clinicians and researchers alike.

5. Conclusions

The anterior approach to surgery offers advantages in vertebral resection and reconstruction of vertebral height. However, the anterior approach is more complex and involves vital organs compared to the posterior approach. Conversely, the posterior approach is commonly utilized for most thoracolumbar fractures with indirect reduction and generally entails fewer complications than the anterior approach, though it has its own limitations. Nonetheless, the posterior approach is familiar to the spine surgeons.

In the present systematic review, various studies examining different surgical approaches for thoracolumbar fractures were analyzed. The results of the meta-analysis reveals that patients undergoing the posterior approach tended to experience longer surgery durations and higher blood loss compared to those undergoing the anterior or combined anterior-posterior approaches. Additionally, immediate postoperative improvement in kyphotic correction was observed in the posterior approach group. However, follow-up assessments indicated a subsequent loss of kyphotic angle in this group. The drawbacks of the present systematic review study is the heterogeneity of the population, including variations in comorbidities, types of fractures, follow-up duration, timing of intervention, surgeon expertise, and surgical techniques across the included studies. Additionally, there is variability in sample sizes, ages, genders, and follow-up periods.

The reults of the present study must be dealt with caution as most of the included studies in the present study were performed on ventral cord compression and kyphosis correction cases rather than simple junctional fracture cases. These reasons could have caused longer surgery durations and higher blood loss in metaanalysis results.

A notable finding of this review was that each approach demonstrated promising outcomes in terms of clinical, surgical, and radiological measures. Hence, the present study emphasizes the inadequate criteria or no definite protocol for decision-making on the thoracolumbar approaches. These observations underscore the importance of increasing awareness among surgeons and clinicians regarding the varied outcomes and potential complications associated with different surgical approaches. We recommend conducting further research with rigorous study designs to establish definitive protocols for selecting the most appropriate approach for thoracolumbar fracture management.

Authors’ statements

All authors have made significant contributions to the manuscript and have reviewed and approved the final version of the manuscript for submission.

Framing the concept and design- Dr. Bikram Kesari Kar.

Corresponding author- Dr. Punit Gaurav.

Analyzing the statistics- Dr. Roshan Lal Goyal.

Reviewing the literature- Dr. Priyanshu Pandey.

Writing of Manuscript- Dr. Punit Gaurav.

Supervision- Dr. Dushyant Chouhan.

Software Validation- Dr. Aakash Mishra.

Visualization and Investigation- Dr. Punit Gaurav.

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this manuscript.

Ethical clearance

Ethical clearance for the study was obtained from the institutional ethics committee, and all procedures were conducted in accordance with the Declaration of Helsinki.

Patient consent

Informed written consent was obtained from all patients involved in this study.

Source of funding

The authors did not receive any financial support or funding for this study.