Posterior stabilisation without formal debridement for the treatment of non-tuberculous pyogenic spinal infection in frail and debilitated population – A systematic review and meta-analysis

Abstract

Non-tuberculous pyogenic spinal infection (PSI) incorporates a variety of different clinical conditions. Surgical interventions may be necessary for severe cases where there is evidence of spinal instability or neurological compromise. The primary surgical procedure, for late-stage PSI, focuses on the anterior approach with aggressive debridement of the infected tissue regions. An alternative treatment method that employs a posterior approach without any formal debridement, is seen as controversial. To the best of our knowledge, few case series and no systematic reviews are assessing the value of this posterior technique. We aim to evaluate the effectiveness of the posterior approach without formal debridement and the associated clinical outcomes, for PSI cases requiring surgical intervention.

Several databases including MEDLINE, NHS Evidence, and the Cochrane database were searched from the date of creation of each database to December 16, 2019. A selection of the keywords used includes: “posterior approach”, “debridement” and “discitis”. Studies were excluded if they involved the anterior approach, carried out formal debridement, or were tuberculous spinal infection cases. We accepted any study type which included adult patients, with spinal infection at any level of the vertebral column. The Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines were used to follow standard systematic review structure. The main clinical outcomes evaluated include pain, neurological recovery (Frankel Grading System, FGS) post-operative complications, and functional outcomes (Kirkaldy-Willis Criteria and Spine Tango Combined Outcome Measure Index, COMI).

Post-surgical neurological improvement was demonstrated with a mean FGS improvement of 1.12 in 102 patients over the included four articles. Post-operative neurological function was found to be improved at a statistically significant level when a random-effects model was applied, with the effect size found to be at 0.68 (p < 0.001). Pain level was improved significantly postoperatively. There were also enhanced functional outcomes post-intervention when the Kirkaldy-Willis criteria and COMI scores were assessed in certain studies.

Within the limit of the available literature, our results showed that the posterior approach with posterior stabilisation without formal debridement can result in successful infection resolution, improved pain scores and neurological outcomes. However, Larger series with longer follow-up duration is strongly recommended.

1. Introduction

Pyogenic spinal infection (PSI) is a bacterial infection of the spine that incorporates a variety of clinical conditions, including osteomyelitis, spinal epidural abscess (SEA), spondylitis, and spondylodiscitis, amongst several others.^1,2 Patients with these conditions tend to present with neck or back pain and may also have neurological signs, weight loss, and/or fever.² Due to the rich capillary network of the vertebral endplate, localised bacterial colonization can lead to destruction of the intervertebral disc and adjacent vertebrae.³ The sequelae of this process are potential spinal instability and neurological compromise that can be both life-altering and eventually, life-threatening. The incidence of PSI stands at between 0.2 and 2 cases per 100,000 per annum.¹ At present, the major cause of bacterial spinal infections tends to be Staphylococcus aureus,⁴ with incidences varying considerably between 30 and 80%.⁵ It is critical to identify the causative organism to ensure that both surgical intervention (if needed) and targeted antibiotic therapy can occur synergistically, to ensure the optimum patient outcome.⁶

One of the key issues clinically is the poor specificity of signs and symptoms associated with PSI, which can closely mimic other metabolic conditions, malignancy, and inflammatory problems. This may partially explain its delayed diagnosis, at which point the patient may be suffering from an advanced stage condition, that can lead to a substantial impact on the quality of life.^1,2,7

1.1. Description of the intervention

The mainstay of treatment for PSI tends to be conservative; involving antibiotics, bed rest, and extrinsic spinal stabilisation.⁷ Surgery is only indicated in those patients who have a poor response to conservative methods, vertebral body collapse, deformity or have a neurological deficit, possibly associated with SEA compression on the spinal cord.^2,3,7, 8, 9, 10

Typically, intrusive anterior debridement of the infected area, with concomitant structural stabilisation, is the accepted method of surgical intervention. In a debilitated and elderly population, however, this can lead to a significant risk of mortality.¹¹ The main infected structure, in conditions such as pyogenic spondylodiscitis, is the anterior aspect of the vertebrae⁸ In addition to this, a growing number of published articles believe a degree of posterior stability is required to maintain spinal structure, thus the implementation of laminectomies, via a posterior approach, is thought to further worsen this integrity.^11,12 For these reasons, along with the more direct nature of access to the infected loci, the anterior approach has generally been the preferred method of choice for many surgeons.11, 12, 13, 14, 15 Nonetheless, several complications have been identified with this anterior approach, including respiratory problems and graft or vertebral body collapse leading to spinal kyphosis.^2,8,13 Interventions using a posterior approach have therefore been conducted and the results have been encouraging. Earlier mobilization, rapid rehabilitation and potentially higher fusion rates are just some of the benefits that have been seen in comparison to an anterior approach.^6,16, 17, 18 Positive findings have been found in patients with end-stage PSI who have received anterior fusion with the use of autografts and full debridement.13, 14, 15 Multi-comorbid states in patients, however, can often lead to other complications of major spinal surgery, including wound dehiscence and implant failure.⁸ With all this in mind, the intervention we aim to critically analyze is the posterior approach without formal debridement and the associated clinical outcomes, for end-stage PSI cases.

1.2. Importance of conducting this review

Several studies have investigated the posterior approach for PSI patients but these small-scale studies, usually taken from a single centre, have little statistical power.^2,6, 7, 8^,10 Additionally, many of these studies tend to be retrospective case series, the weakest form of evidence.¹⁹ This systematic review aims to collate the current primary studies in this area and provide a foundation for larger-scale primary studies, with long-term follow-up, and eventually lead to further systematic reviews. To the best of the authors’ knowledge, we are unaware of any existing systematic review that specifically analyses this surgical technique.

We aim to clarify whether there is any tangible benefit of using the posterior approach, without formal debridement, for the treatment of PSI patients that have indications for spinal surgery. Outcomes will specifically focus on the resolution of infection, improvements in pain, and neurological status after using the posterior approach.

2. Methods

2.1. Search strategy

As aforementioned, no existing systematic reviews are assembling primary studies in this research area, at least to our knowledge. Two independent authors searched different search engines between November and December 2019, which included MEDLINE, NHS evidence, and Cochrane (for any existing systematic reviews) but also Google Scholar for completeness. We literature searched from the date of creation of the databases to the last day of search (December 16, 2019). To ensure we collated all the relevant studies needed for the review, we studied the references from eligible studies to see whether there were any more studies we could include. Further searches were performed alongside clinical librarians at two different National Health System (NHS) trusts; this equated to 4 different people reviewing the current evidence in this area. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA)²⁰ guidelines were used throughout the study to ensure we conformed to standard systematic review structure. Different combinations of keywords used to search the relevant literature (Table 1).

Table 1.

Search strategies utilised the following search terms to identify relevant research papers.

1	((Stabilisation OR fixation OR ‘pedicle screw’ OR hook) NOT (‘minimally invasive’ OR ‘percutaneous’))
2	‘′Orthopaedic Fixation Devices’’/
3	1 OR 2
4	((Debridement AND infect∗) OR spondylodiscitis OR discitis)
5	Debridement
6	Discitis
7	4 OR 5 OR 6
8	3 AND 7
9	anterior OR posterior
10	8 AND 9

The search was limited to just the English language and no other limits were placed. Inclusion and exclusion criteria are summarised as in Table 2. We included retrospective case series studies only, as these were the only available primary study design on this topic area. We only included case reports that reviewed greater than 3 patients, this was to ensure that we increased the statistical power of the review. Included studies involved interventions that were posterior, that did not include any formal debridement – whereby the surgeon intentionally seeks out infectious material and removes it – and those studies that involved fixation. This search yielded 39 of the 45 total results.

Table 2.

Inclusion and exclusion criteria for the study.

Inclusion	Exclusion
No restrictions on online database search	Clinical diagnosis of PSI as Tuberculosis-related
No restriction on follow-up period	Interventions were non-posterior in nature
All study types	Invasive anterior approach to treatment
Adults >18 years old
All spinal regions
No regard to type of spinal infection, age, gender or severity of disease
All types of fixations/type of equipment used

3. Results

3.1. PRISMA flow diagram and study selection

A total of 45 studies were retrieved from the electronic database, 2 studies removed as they were duplicates (Fig. 1). The 43 studies then underwent title and abstract screening. Two authors read the titles and abstracts of the papers and determined whether these fit the eligibility criteria. For completeness, both authors read the full-text article and individually determined whether that specific paper would be included within our review. Any disagreements were resolved through collaboration. After this screening phase, 7 papers were identified as eligible.

Fig. 1.

PRISMA flow diagram and study selection.

From the 7 papers making it to the eligibility stage, 1 paper was removed because it only included a case report of a single patient. Another paper was removed because the outcomes it reported were distinctly different from the other 5 papers. The authors, therefore, narrowed the papers down to 5 for qualitative analysis and 4 papers for quantitative analysis. This was because one of the 5 papers did not have suitable data for extraction and analysis. The four papers analyzed quantitatively had low heterogeneity, thus were amenable to statistical combination.

From four studies, mean and standard deviation were computed for Pre-operative/follow-up neurological status scores. Meta-analysis model using random effects was performed to estimate pooled effect size. The reason for using random effects was that the variation among studies was very likely due to different designs of studies and characteristics of populations.^21,22 The heterogeneity between-studies was examined using Q and I² statistics.²³ The R 4.0.0 package using metafor library was employed to perform meta-analysis.

3.2. Background characteristics of included patients

From the data available, the mean age of all 108 patients in the 5 studies included for qualitative analysis was 61.6 years. Only Chin Lin et al. did not provide information regarding the sex of patients. Of the 60 patients in the other 4 studies, however, only 14 (23%) were females.

Perhaps most importantly, all the papers provided some information on co-morbidities. The patient cohorts were vulnerable and high-risk for surgery. Dobran et al. showed that 18 co-morbidities, including diabetes mellitus, end-stage renal disease (ESRD), and Rheumatoid Arthritis (RA) were present in its’ patient group. Mohamed et al. disclosed that the mean American Society of Anaesthesiologists (ASA) score of its’ 15 patients was 2.8, which would classify the overall patient group as being between mild systemic disease (ASA grade 2) and severe systemic disease that is not a constant threat to life (ASA grade 3). Similarly, the ASA for the 21 patients included in Aljawadi et al. series was 2.76. Chin Lin et al. showed that 22 co-morbidities (with DM and RA being most prevalent) were present within its’ 48-patient group. Only 3 patients (14%) had no record of a co-morbidity in the work by Aljawadi et al. with all other patients living with at least one co-morbidity and 12 patients (57%) having multiple. Some of these co-morbidities included lifestyle factors such as smoking and alcoholism, although the most common co-morbidities in this cohort were DM and hypertension. Finally, each of the patients within the study by Fushimi et al. had at least one co-morbidity.

The level of spinal infection was reported for 61 patients only. The thoracic spine was most commonly affected (64%), followed by the lumbar spine (33%), while cervical spine infection was reported in only 3% of patients.

3.3. Quality of the evidence (NIH quality Appraisal tool)

Once the 5 selected studies had been obtained, the National Institute of Health critical appraisal tool (NIH)²⁴ was used to determine the quality of the studies collected. There are 9 criteria set out in this appraisal tool and two authors decided whether each of the studies either met the criteria, did not meet the criteria, or other (could not determine the criteria or the criteria was not reported). The greater the percentage of the criteria that are met, the more reliable the study was likely to be. Using these percentage values, the author ratings for the percentage of criteria met was used to determine whether the paper should get a ‘good’, ‘fair’, or ‘poor’ rating. The results of this appraisal can be seen in Table 3, with the overall quality of four of the articles being rated as ‘Good’, whilst one study rated as ‘Fair’. The reasons for determining the overall quality is also given in Table 3.

Table 3.

NIH quality appraisal tool for case series.

Paper	Percentage (%) of criteria met	Percentage (%) of criteria not met	Percentage (%) of criteria ranked as other (CD or ND∗)	Overall Quality Rating (Good, Fair or Poor)	Reason for Overall Quality Rating
Aljawadi et al, 20192	77.8	11.1	11.1	Good	A 21-patient study with clinically relevant outcomes such as key bloods and the Frankel grading system. EQ-5D follow-up data was missing for 3 patients and 11 patients did not have pre-intervention data for the COMI questionnaire, however.
Mohamed S et al, 201410	66.7	22.2	11.1	Good	A 15-patient study that clearly defined its’ purpose with at least 10 patients having a 2-year follow-up. Lack of detailed baseline characteristics made comparison between patients difficult, however. Furthermore, lack of a detailed description on the statistical analysis was also evident.
Lin L et al, 20128	77.8	0.0	22.2	Good	A 48-patient study that precisely stated its’ methodology. It included clinically relevant outcomes as well as functional. There was also a lengthy average follow-up of 64 months. There was uncertainty in whether patients were consecutively enrolled and there was no explicit inclusion or exclusion criteria, however.
Dobran M et al, 20167	66.7	11.1	22.2	Good	An 18 patient study that recruited patients over a 5-year period (2008–2013) but it was unclear whether patients were enrolled consecutively and there was no explicit inclusion or exclusion criteria. There was also a lack of a quality of life measure used in determining outcomes, despite the study having a robust methodology. The objective and study protocols were clear with a high mean follow-up period (30.16 months).
Fushimi K et al, 20126	55.6	22.2	22.2	Fair	A 6-patient study with a clear objective and study protocol but lacked inclusion and exclusion criteria. Furthermore, there was no clear indication as to over which time period patients were enrolled or whether they were consecutive inclusions. The follow-up period was more than adequate (average of 33.6 months) and they did report clinical and objective quality of life outcomes. The outcome data was not suitable for extraction and use in the meta-analysis, however.

3.4. Meta-Analysis – Assessing preoperative and postoperative neurological function using frankel scores

Frankel grading system employs five classifications from A (complete motor and sensory deficit) to E (a patient free of neurological symptoms). A meta-analysis was performed for Frankel scores across the four studies (total of 102 patients) using a random-effects model, which was used to estimate the overall effect. This was based on a standardized mean-change measure from pre-test (Pre-operative Frankel scores) and post-test (follow-up Frankel scores) designs.²⁵ For the four studies included in the meta-analysis, the test of heterogeneity was insignificant (I² = 5.52%). The overall effect size was significantly large (effect size = 0.68, 95% CI: 0.51 to 0.85), (Fig. 2). Hence, the pre-operative and follow-up neurological status differed significantly in degree of change. From the meta-analysis, seen in Fig. 2, we determined that the post-operative Frankel scores were reduced compared to the pre-operative Frankel scores at a statistically significant level.

Fig. 2.

Forest plot for Pre-operative and follow-up neurological function.

7 patients had preoperative FGS grade A (complete motor and sensory loss). This had improved after surgery to grade E (normal motor and sensory function) in 1 patient and to grade D (incomplete motor loss, able to mobilise with or without aids) in 2 patients. The remaining 4 patients had no improvement in their neurological status. In the same context, FGS grade B (complete motor loss, incomplete sensory loss) was reported in 6 patients preoperatively. This had improved to grade D in 4 patients, grade E in one patient and one patient had deterioration of his neurological state to grade A postoperatively.

3.5. Outcomes not evaluated in the meta-analysis

Other results were not included in the meta-analysis due to the high levels of inconsistency, between studies, in the outcomes that did not involve pain before and after surgery. Furthermore, missing data in certain papers also made statistical analysis inappropriate. These other outcomes are important to consider, however.

3.5.1. Pre- and post-operative pain levels

Lin et al. had reported that posterior spinal stabilisation, without debridement of infected tissue, resulted in an improvement of pain as measured by the Visual Analogue Scale (VAS), which was declined from 7.2 before surgery to 2.2 after surgery.⁸ Similarly, Dobran et al. have reported that the VAS scores had improved to 1.38 ± 2.03 postoperatively from 9.16 ± 1.29 preoperatively, which was statistically significant (p < 0.05).⁷ Aljawadi et al. reported that 94% of patients assessed for postoperative pain level had reported having no pain, or only mild pain.² This was an improvement compared to preoperative data which showed that 90% of patients were suffering from pain.

3.5.2. Functional outcomes

Functional outcomes were also assessed in many of the papers. Dobran et al. showed that the average number of days that patients mobilised after surgery stood at 4.5 ⁷. Mohamed et al. also showed that 13 of 15 patients were ambulatory with or without an assistive device after the posterior spinal surgery.¹⁰ Lin et al. assessed functional outcome using the Kirkaldy-Willis criteria,^8,26 which showed that 32 of 48 patients (66.7%) felt their function outcome was excellent or good, whereas only 3 patients (6.3%) said their outcome was poor. Additionally, Aljawadi et al. carried out medium term-follow up of patients using the Spine Tango Combined Outcome Measure Index (COMI) questionnaire. Pre-operative and post-operative data were available for ten of the twenty-one patients. The COMI score improved significantly from 6.39 preoperatively to 4.05 postoperatively (p = 0.045). As well as Aljawadi et al. had assessed patients’ Quality of Life (QoL) using the Euro Qol EQ-5D questionnaire. The results were available for 18 patients out of twenty-one. 15 of the 21 patients (71.4%) reported as being ‘very satisfied’ with no patients stating that they were dissatisfied with the treatment. With regards to the overall outcome of the treatment, 14 patients (66.7%) stated that the treatment had ‘helped a lot’ with 20 patients overall stating that the treatment had helped to some extent (95%). Only 1 patient (4.8%), stated that the treatment did not help.

3.5.3. Complications rate

Out of the 54 patients over three papers^2,8,10 that discuss post-operative complications, only 11 exhibited any form of complication. This puts the complication rate at 20.4% following the posterior stabilisation procedure.

4. Discussion

The first-line treatment of PSI is by conservative measures including antibiotics, spinal support, and bed rest. Surgical intervention should only be considered for patients who failed to respond satisfactorily to the conservative treatment, or if there is any indication for surgery (such as vertebral body collapse with significant deformity, or severe neurological deficit due to compression by an epidural collection).^8,27 All patients included in this systematic review have had a trial of conservative management (unless surgery was indicated at presentation), and they were only considered for surgery after failure of conservative measures.

Posterior stabilisation without debridement for pyogenic spinal infection is currently a surgical treatment option, to potentiate the effect of antibiotics and improve healing of the infection, particularly in elderly patients with comorbidities, and patients who will not tolerate extensive surgical approaches.²⁸ Spine infection generally affects the anterior vertebral elements, thereby making the anterior surgical procedure a more direct intervention to debride the infected material. However, anterior spinal surgery is a more extensive surgery, associated with more blood loss, longer operative time, and more prolonged postoperative recovery and rehabilitation. The posterior approach is more attractive as it is associated with earlier mobilization, more rapid rehabilitation, and faster fusion achieved with posterior instrumentation.^16,29 Furthermore, it is reported that posterior surgery could improve neurological outcome in half of the patients.²⁷ the operative time of posterior surgery, blood loss, and complications are less compared to the anterior approach.⁶ Traditionally, debridement of infected tissue, avoiding hardware placement, if possible, to minimise the risk of bacterial colonization was the mainstay when surgery indicated for musculoskeletal infection. However, recently multiple studies from both Neurosurgery and Orthopaedics literature for patients with osteomyelitis, discitis, and abscess underwent spinal stabilisation while acutely infected, had reported successful infection resolution. Acute instrumentation for pyogenic spinal infections patients reduced the likelihood to require additional surgery compared to patients without spinal instrumentation.¹⁰ The position of the PSI in the spinal column did not affect the surgical option and a posterior approach without debridement was performed for all patients.

In a retrospective study conducted by Mohamed et al. posterior long-segment fixation, without formal debridement, resulted in resolution of spinal infection in a series of 15 patients and significant neurological recovery in almost all cases.¹⁰ This surgical technique, when combined with aggressive antibiotic therapy and a multidisciplinary team approach, is effective in managing severe spinal infections in a challenging patient population.¹⁰ The main aim of this systemic review is to evaluate the effectiveness of posterior spinal surgery, without debridement, for complex and progressed PSI. The emergence of this technique has been due to the increased mortality seen in elderly patients who undergo spinal surgery via an anterior approach. With earlier primary studies seeming to indicate a potential benefit of the posterior approach, the authors felt that a review and meta-analysis of the literature was needed.

From the meta-analysis of four papers results, which included 102 patients, we were able to identify a statistically significant difference between the pre-operative and post-operative neurological status as measured by the Frankel Grading system. The heterogeneity between the neurological assessment in four papers analyzed was insignificant, suggesting that conducting a combined meta-analysis was both appropriate and valid. In patients that presented with at least neurological deficit, there was a mean improvement of 0.68 on the FGS for the 102 patients assessed in the four papers. It is important to highlight that, from the reviewed articles, posterior approach without debridement had been performed even for patients who had complete motor dysfunction before surgery. The severity of pre-operative neurological status didn’t seem to affect the surgical approach for these patients. The results from these 4 articles showed that 15 patients presented either with neurological dysfunction of Grade A, or B FGS (complete motor loss), nonetheless, 8 of these patients had improvement in their neurological status postoperatively.

Moreover, functional outcomes that were assessed indicate that this particular intervention can provide significant benefits. Aljawadi et al.² portrayed a statistically significant post-surgical improvement in the COMI score in a group of 10 patients, whilst Lin et al.⁸ showed that 66.7% of patients felt their functional outcome was excellent or good post-intervention. Mohammed et al. had reported that residual symptoms of pain had no or minor effect on the work or usual activities in 52% of subjects, with 88% reported having either no or mild problems with mobility.¹⁰ These results all favour the use of a posterior approach with no formal debridement.

The administration of postoperative antibiotics should be guided by accurate microbiological diagnosis obtained from multiple deep tissue samples taken at the time of surgery.²⁷ Aljawadi et al. had reported that 16 out of the 21 patients included in their retrospective study have had a positive growth (which was Staphylococcus Aureus in 50%).² Mohammed et al. have mentioned that their choice of antibiotics was guided by the infectious disease specialist based on the Culture and sensitivity results.¹⁰ None of the five included studies advised any specific antibiotics protocol to be followed for these patients. Dobran et al. reported that the mean duration of antibiotics therapy was 2.11 months of IV antibiotics followed by 5.33 months of oral antibiotics.⁷

4.1. Rationale of posterior stabilisation without debridement

This approach had been described for the management of PSI in elderly debilitated patients with multiple comorbidities. It was reported that performing formal debridement of the infected tissue followed by posterior stabilisation for the management of PSI in elderly patients who have more than one comorbidity can result in high rates of morbidity (50%) and mortality (17%).²⁸ Hence, it was suggested that performing formal debridement would increase total surgical time and the amount of blood loss, which can explain the high rate of postoperative morbidity and mortality.² Most of the patients included in this review had multiple comorbidities, and the mean ASA score was available for 36 patients only and was around 2.8. The mean surgical time for performing posterior stabilisation without formal debridement was reported by Fushimi et al. to be around 227 min (however, no data was provided about the mean surgical time from the other 4 papers included in this study).

Precise physiological mechanisms by which the posterior approach, without debridement, leads to infection recovery, and spinal fusion is still yet to be fully elucidated. One theory that may explain the biological plausibility could be that secure stabilisation across the region of infection can lead to better antibiotic access and enhanced tissue recovery.^2,10 This could partially be down to the enhanced blood flow to the region, an aftereffect of the rigid fixation.^30,31 Furthermore, as long as appropriate antibiotic therapy is initiated, then infection presence does not affect the fusion of the bone.⁸ Hence, complete debridement does not seem fundamental, especially when previous studies have shown that internal fixation and bone fusion can both occur harmlessly, even in the presence of co-existing spinal infection.^32,33 Despite this, the use of metallic surgical equipment to treat spinal infection remains distinctly controversial amongst many surgeons.²⁹ One of the primary reasons for this is the fact that the instrumentation can act as an adherence site for bacteria and lead to the formation of a biofilm that can significantly reduce antibiotic efficacy.⁷ This should not discourage the use of the posterior approach, however, as instrumentation may only span healthy tissue, avoiding the infectious field. Crucially, the fixation of the vertebral column with metallic implants can also secure the region to aid control of inflammation.⁷ The physiological reserve of the patient ultimately determines the choice of approach. A posterior approach can lead to less comprehensive tissue dissection, decreased operating times, and intraoperative blood loss.^2,6,8 If a patient is healthy, young, and has very few comorbidities, then the anterior approach would still generally be the first line surgical intervention of choice, due to the more radical debridement and fusion.³ This can lead to a faster recovery.² Despite this, many studies have concluded that the posterior approach, without debridement, will most likely be effective in those patients who are most debilitated and would not tolerate aggressive anterior debridement procedures.^2,7,10

4.2. Limitations of this review

There are some key limitations to consider with this review. Firstly, we were only able to find case series as a study type in this highly specialised research area. There is a need for future reviews to utilise other study types with greater quality, like randomised control trials or case-control studies. Crucially, as a consequence of the case series study type, there was no control group used as a comparator. The lack of a well-defined control group, with similar baseline characteristics to the intervention group, reduces the validity of our work and makes it difficult to draw meaningful conclusions.

Furthermore, the GRADE criteria could have also been applied to ascertain with confidence whether the true effect is close to the estimate given by each study. This would have also allowed us to determine the optimal information size, which is the required sample size in each study that would enable us to obtain reliable conclusions about the intervention.

Additionally, it is worth noting that a meta-analysis was only conducted for the change in neurological status between pre-operative and post-operative patients. This was mainly due to the lack of homogenous outcomes for quality of life and functional results reported between the four papers. Future systematic reviews should look at the outcomes more holistically and include meta-analyses of results for function and patient experience post-operatively.

5. Conclusion

Taking into consideration the limitation of this review, the results showed that the posterior approach with posterior stabilisation without formal debridement of the infected tissue can result in resolution of infection, with improved neurological status. As well as this approach was helpful to decrease postoperative pain scores level and functional scores. However, a larger series with a longer follow-up duration is required to enrich the evidence regarding this surgical approach.

Disclaimer

The authors declare that no part of this study has been taken from existing published or unpublished materials without due acknowledgement and that all secondary materials used herein has been fully referenced.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.