A comprehensive comparison of posterior lumbar interbody fusion versus posterolateral fusion for the treatment of isthmic and degenerative spondylolisthesis: A meta-analysis of prospective studies

Clinical Neurology and Neurosurgery 188 (2020) 105594

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro

A comprehensive comparison of posterior lumbar interbody fusion versus posterolateral fusion for the treatment of isthmic and degenerative spondylolisthesis: A meta-analysis of prospective studies

T

Yucai Lia, Zhifeng Wub, Dongze Guob, Hao Youb, Xing Fanb,* a b

People's Hospital of Rizhao, Rizhao 276800, China Beijing Neurosurgical Institute, Capital Medical University, Beijing 100050, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Spondylolisthesis Posterolateral fusion Posterior lumbar interbody fusion Meta-analysis

Posterolateral fusion (PLF) and posterior lumbar interbody fusion (PLIF) are both frequently-used procedures for the surgical treatment of isthmic and degenerative spondylolisthesis. The current meta-analysis aimed to perform a comprehensive comparison between PLF and PLIF in terms of the therapeutic effects on spondylolisthesis. PubMed, Embase, Web of Science and the Cochrane Central Register of Controlled Trials were searched for relevant prospective studies. Measures of clinical outcome, postoperative complication rate, fusion rate, and blood loss are presented as odds ratio (OR), mean difference and corresponding 95 % confidence interval (CI) as appropriate. Eight prospective studies comprising 723 patients were eventually enrolled in the meta-analysis. Patients who underwent PLIF had a better clinical outcome (pooled OR, 1.63 [95 % CI, 1.02–2.61]; p = 0.04) and a higher fusion rate (pooled OR, 3.33 [95 % CI, 1.88–5.90]; p < 0.01) than those who underwent PLF. No significant difference between the two procedures was identified for postoperative complication rate and blood loss. The results showed that PLIF was superior to PLF in clinical outcome and fusion rate, and equal to PLF in terms of blood loss and the rate of postoperative complications. Here we provide the most effective evidence currently available for the comparison between PLF and PLIF, which has guiding significance for clinical practice.

1. Introduction Spondylolisthesis is defined as the forward displacement of one lumbar vertebra over the subjacent vertebra. The most common causes of spondylolisthesis are pars interarticularis abnormalities and degenerative changes [1,2], and the prevalence of isthmic spondylolisthesis (IS) and degenerative spondylolisthesis (DS) in general population is approximately 10 % [3,4]. When symptomatic, spondylolisthesis can result in a variable clinical syndrome of low back and/or lower extremity pain, which includes varying degrees of neurological deficits. In recent decades, surgical intervention has played a major role in the treatment of spondylolisthesis. Previous studies have identified that surgical intervention may lead to greater improvements in pain and function than nonsurgical treatment [5,6]. Arthrodesis has proven to be an effective, reliable, and safe surgical treatment strategy for spondylolisthesis. Various surgical procedures for arthrodesis including posterolateral fusion (PLF), anterior lumbar interbody fusion, posterior lumbar interbody fusion (PLIF), transforaminal lumbar interbody fusion

⁎

and so on, have been applied in clinic. However, it is difficult for spine surgeons to arrive at a consensus on the optimal procedure, and relevant comparisons have never ceased [7–9]. PLF and PLIF are both widely used surgical procedures for the treatment of spondylolisthesis. To date, numerous studies have compared the two procedures in patients with IS and DS, and corresponding meta-analyses have been conducted. However, after a cursory review, we found that it was still not easy to summarize which procedure is better for the treatment of spondylolisthesis. Some previous studies showed that PLIF was superior to PLF in terms of clinical outcome and fusion rate [10,11], while others suggested that there was no significant difference between the two procedures [12,13]. Furthermore, even existing meta-analyses arrived at inconsistent conclusions [14–19]. Evidently, it is necessary to obtain a clear conclusion on this issue to guide clinical practice. Through the literature review, we found that all previous meta-analyses had included retrospective studies. Therefore, a synthesis of more rigorous evidence (i.e., analysis using prospective studies only), would be helpful to get a more exact result. In the current

Corresponding author at: South West fourth ring road 119, Fengtai District, Beijing, China. E-mail address: [email protected] (X. Fan).

https://doi.org/10.1016/j.clineuro.2019.105594 Received 24 August 2019; Received in revised form 2 November 2019; Accepted 10 November 2019 Available online 11 November 2019 0303-8467/ © 2019 Elsevier B.V. All rights reserved.

Clinical Neurology and Neurosurgery 188 (2020) 105594

Y. Li, et al.

study, we conducted a meta-analysis of prospective studies comparing PLF and PLIF in patients with IS and DS. We investigated betweengroup differences in clinical outcome, postoperative complication rate, fusion rate, operation time and blood loss. 2. Methods 2.1. Search strategy and selection criteria The current study was performed according to the Preferred Reporting Items for Systematic Review and Meta-Analyses of individual participant data (PRISMA-IPD) guidelines [20]. We systematically searched PubMed, Embase, Web of Science and the Cochrane Central Register of Controlled Trials (CENTRAL) for all relevant articles from database inception to August 18, 2019. The search strategy applied was a combination of the following key words: “PLF” or “posterolateral lumbar fusion” or “posterolateral fusion”, “PLIF” or “posterior lumbar interbody fusion” or “posterior interbody fusion” and “spondylolisthesis” or “spondylolistheses” or “spondylolysis” or “spondylolyses”. Retrieved results were transferred to Endnote X7 software (Thomson Reuters, New York, NY, USA), where a check for duplicates was performed. The titles and abstracts of the remaining articles were then screened with the following exclusion criteria: 1) articles published in languages other than English; 2) reviews, guidelines or classifications; 3) letters to the editor; 4) case reports or small case series (< 5 cases); 5) meeting abstracts or abstract-only studies; 6) in vitro and animal experiment studies; 7) other irrelevant studies. Subsequently, full-text articles of potentially relevant studies were obtained and assessed for eligibility. Studies that met the following inclusion criteria were included: 1) studies on patients with spondylolisthesis, or studies from which data on spondylolisthesis patients could be extracted independently; 2) studies comparing PLF and PLIF; 3) prospective studies; 4) duration of follow-up ≥1 yr; 5) the ability to extract data related to the indicators used for comparison (including clinical outcome, postoperative complication rate, fusion rate, operating time, and blood loss).

Fig. 1. Flow diagram for the study.

Galbraith plot was created. Otherwise, a fixed-effect model was used. Sensitivity analysis and/or subgroup analysis was performed if necessary.

2.2. Data extraction and quality assessment

3. Results

The literature search was performed by one author. Two authors screened the retrieved records and completed data extraction from all eligible studies using a standardized data extraction form. In cases of conflict, a consensus meeting was held with another author. The overall quality of the evidence was evaluated according to the grades of recommendation assessment, development and evaluation (GRADE) system with GRADE profiler software (version 3.6.1, McMaster University, Hamilton, Ontario, Canada) [21]. After considering risk of bias, inconsistency, indirectness, imprecision, and publication bias, evidence quality was defined as “high”, “moderate”, “low”, or “very low”.

3.1. Study selection Our systematic search of the literature yielded 127 publications from PubMed, 219 from Embase, 297 from Web of Science and 31 from CENTRAL. After removing 283 duplicates, 391 references remained for screening of the titles and abstracts. Ultimately, 28 articles were read in their entirety and 8 studies met the eligibility criteria for the current study. Fig. 1 shows the flow diagram. 3.2. Study characteristics Details for the included studies are shown in Table 1. Among 8 eligible studies, 4 were randomized controlled trials (RCTs) [10,22–24]; 2 were clinical controlled trials (CCTs) [25,26]; 2 were prospective nonrandomized studies [12,27]. 6 studies included patients with IS only, while 2 studies included patients with both IS and DS. A total of 723 patients with IS or DS were enrolled in the 8 selected studies (349 patients in 4 RCTs, 159 patients in 2 CCTs and 215 patients in 2 prospective non-randomized studies), with sample sizes ranging from 50 to 163. Only 2 studies enrolled over 100 patients [12,22]. Overall, 595 patients (82.3 %) with IS and 128 patients (17.7 %) with DS were analyzed, including 340 patients (47.0 %) who underwent PLF and 383 patients (53.0 %) who underwent PLIF. The overall strength of evidence for each comparable parameter was evaluated with GRADE criteria. The overall strength of the evidence was rated as “moderate” for fusion rate, “low” for clinical satisfaction and postoperative complication rate, and “very low” for blood loss.

2.3. Statistical analysis Statistical analysis was performed using Review Manager software (RevMan, version 5.3, Nordic Cochrane Centre, Cochrane Collaboration, Copenhagen, Denmark) and Stata software (version 15.0, Stata corporation, College station, Texas, USA) as appropriate. Forest plots were used to illustrate the results graphically. The odds ratio (OR) and corresponding 95 % confidence interval (CI) were used as measures of qualitative variables including clinical outcome, postoperative complication rate, and fusion rate, while the mean difference (MD) and corresponding 95 % CI were used to estimate quantitative variables such as operating time and blood loss. Heterogeneity was assessed with Cochran's Q test, and the I2 statistic. In the presence of significant heterogeneity (I2 > 50 %; Cochran's Q test p-value < 0.10 and I2 > 25 %), data were pooled with a random-effects model, and a 2

Clinical Neurology and Neurosurgery 188 (2020) 105594

Y. Li, et al.

Table 1 Baseline characteristics of included studies. Study

Country

Cheng et al. [22] Dantas et al. [25] Dehoux et al. [27]

China Brazil France

Ekman et al. [12]

Sweden

Farrokhi et al. [23] Feng et al. [26] Lee et al. [24] Müslüman et al. [10]

Iran China Korea Turkey

Study design

RCT CCT Prospective non-randomized study Prospective non-randomized study RCT CCT RCT RCT

Follow-up time (months)

Sample size

Age (mean, range/SD)

Sex (PLF)

Sex (PLIF)

Total

IS

DS

PLF

PLIF

PLF (yrs)

PLIF (yrs)

M

F

M

F

48 12 72

138 60 52

56 14 52

82 46 /

68 30 25

70 30 27

48 (38–63) 52.4 42.4 (14–63)

49 (36–62) 47.6 39.5 (14–56)

36 17 14

32 13 11

39 10 /

31 20 /

> 24

163

163

/

77

86

39

40

38

39

33

53

12 24 > 36 39.6

80 99 81 50

80 99 81 50

/ / / /

40 36 39 25

40 63 42 25

49.66 ± 9.01 60.2 ± 5.2 53.4 ± 2.3 47.3

50.35 ± 11.30 57.1 ± 6.9 53.7 ± 2.1 50.6

/ / 21 9

/ / 18 16

/ / 23 8

/ / 19 17

RCT, randomized controlled trial; CCT, clinical controlled trial; IS, isthmic spondylolisthesis; DS, degenerative spondylolisthesis; PLF, posterolateral fusion; PLIF, posterior lumbar interbody fusion; SD, standard deviation.

between subgroups was identified (I2 = 0 %, p = 0.71, Cochran's Q test, Fig. 4). These results suggested that the difference in study design did not contribute to overall heterogeneity. Subsequently, a Galbraith plot was created to detect potential sources of between-study heterogeneity. As shown in Fig. 5, the two studies conducted by Dehoux et al. [27] and Ekman et al. [12] are the two outliers under the dominant model. After removing these two studies, the heterogeneity was significantly reduced (I2 = 33 %; p = 0.20, Cochran's Q test, Fig. 6). The result remained insignificant (pooled OR, 2.06 [95 % CI, 0.84–5.06]; Z = 1.58, p = 0.11, Fig. 6).

3.3. Clinical outcome Data on clinical outcome were extracted from 5 studies, although the assessment methods used were different. Global outcome (as assessed by patients) was used to evaluate clinical outcome in 2 studies [12,22], while 3 other studies evaluated the improvement in scores on the Oswestry Disability Index (ODI) [10], scores on the Prolo Economic and Functional Scale [25] and Beaujon scores [27], respectively. In the current meta-analysis, poor clinical outcome was defined as: a global outcome assessed by patients as“unchanged” or “worse”; an ODI score classified as“unchanged” or “worse”; a Prolo scale rating of grade 1 or 2; a Beaujon score of “poor result” or “failure”. Fig. 2 shows the forest plot. Patients who underwent PLIF seemed to have a significantly better clinical outcome than those who underwent PLF (pooled OR, 1.63 [95 % CI, 1.02–2.61]; Z = 2.04, p = 0.04; Fig. 2). No significant heterogeneity was identified (I2 = 0 %; p = 0.41, Cochran's Q test; Fig. 2).

3.5. Fusion rate Six studies compared the PLF group and PLIF group in terms of postoperative fusion rate. The result of meta-analysis is shown in Fig. 7. The combined result suggested that the PLIF group had a significantly higher fusion rate than did the PLF group (pooled OR, 3.33 [95 % CI, 1.88–5.90]; Z = 4.13, p < 0.01, Fig. 7), and no significant heterogeneity was found (I2 = 0 %; p = 0.42, Cochran's Q test; Fig. 7).

3.4. Postoperative complication rate Seven studies that included data on postoperative complications were analyzed. A random-effects model was used because high heterogeneity was observed (I2 = 66 %; p < 0.01, Cochran's Q test). The result showed that there were no significant statistical differences in complication rate between patients who received PLF and patients who received PLIF (pooled OR, 1.89 [95 % CI, 0.69–5.23]; Z = 1.23, p = 0.22, Fig. 3). To identify the potential sources of heterogeneity, a subgroup analysis based on study design was performed. No significant statistical difference in complication rate was observed among RCTs (pooled OR, 2.38 [95 % CI, 0.68–8.33]; Z = 1.35, p = 0.18, Fig. 4) or among non-RCTs (pooled OR, 1.57 [95 % CI, 0.27–9.29]; Z = 0.50, p = 0.62, Fig. 4). Moderate heterogeneity persisted in the RCTs group (I2 = 46 %, p = 0.13, Cochran's Q test, Fig. 4) while a significant high heterogeneity was observed in the non-RCTs group (I2 = 81 %, p < 0.01, Cochran's Q test, Fig. 4). Moreover, no heterogeneity

3.6. Operation time and blood loss Operation time was reported in 3 studies, the data are shown in Table 2. Overall, it seemed that the PLIF group had a longer operation time than did the PLF group. However, the data were not suitable for meta-analysis. As for blood loss, there were 3 studies that evaluated blood loss in two groups (Table 2). No difference in blood loss was identified (pooled MD, 46.83 ml [95 % CI, -142.33–236.00]; Z = 0.49, p = 0.63, Fig. 8), while great heterogeneity was found between studies (I2 = 88 %; p < 0.01, Cochran's Q test; Fig. 8). Subgroup analysis and Galbraith plot were not performed due to the limited number of studies.

Fig. 2. Forest plot for clinical outcome, with events stand for poor clinical outcome. 3

Clinical Neurology and Neurosurgery 188 (2020) 105594

Y. Li, et al.

Fig. 3. Forest plot for complication rate, with events stand for complications.

4. Discussion Arthrodesis is currently considered to be the optimal surgical strategy for patients with IS and DS, but controversies still exist regarding the best procedure for it. PLF was developed by Watkins in the 1950s, while PLIF was first reported by Cloward in the 1960s. Both procedures were developed through long-term clinical practice and technical modification. To date, they have become the two most common choices for arthrodesis among various available procedures. A lot of studies have been carried out to determine which of the two is the optimal surgical procedure for spondylolisthesis, including six relevant meta-analyses [14–19]. Surprisingly, so much meta-analyses have failed to arrive at a definite conclusion, making this issue more confusing. Ye et al. first conducted a meta-analysis on this issue in 2013 and proposed that PLIF provided a higher fusion rate than PLF [14]. Two meta-analyses published subsequently also found that PLIF could offer not only an increased fusion rate, but also improved patient satisfaction [15] and a decreased complication rate [16]. However, later studies showed different results. Luo et al. demonstrated that although PLIF had a higher fusion rate, it was also associated with longer operation time, with no difference in clinical outcome compared to those achieved with PLF [17]. Moreover, Campbell et al. reported no significant difference in functional or operative outcomes between the two procedures [18]. In the latest relevant study published by Chen et al., no statistically significant difference in clinical outcome, complication

Fig. 5. Galbraith plot for complication rate.

rate, fusion rate or blood loss was identified [19]. Predictably, if we performed a meta-analysis without ensuring a study design more rigorous than that used in the above-mentioned studies, it would do little for clarifying this issue. Accordingly, only prospective studies were enrolled and analyzed in the current study, which made the results more accurate and reliable. Life quality is the main concern for patients with IS and DS postoperatively. Clinical practice has proven that both PLF and PLIF can provide excellent outcome for patients; however, is there a difference in

Fig. 4. Forest plot for complication rate with subgroup analysis (RCTs group and Non-RCTs group). 4

Clinical Neurology and Neurosurgery 188 (2020) 105594

Y. Li, et al.

Fig. 6. Forest plot for complication rate after removal of the study by Dehoux et al. and Ekman et al., which decreased heterogenicity.

Fig. 7. Forest plot for fusion rate, with events stand for nonunion. Table 2 The operation time and blood loss of relevant studies. Study

Cheng et al. [22] Dantas et al. [25] Dehoux et al. [27] Ekman et al. [12] Farrokhi et al. [23] Feng et al. [26] Lee et al. [24] Müslüman et al. [10]

Operation time (min)

Blood loss (ml)

FLF

PLIF

FLF

PLIF

192 / / / / / 126 ± 12 146(105–300)

210 / / / / / 156 ± 18 168(120–310)

/ / / / 747.37 ± 439 / 350 ± 25 1100 ± 280

/ / / / 837.07 ± 370.24 / 360 ± 30 830 ± 215

PLF, posterolateral fusion; PLIF, posterior lumbar interbody fusion.

Fig. 8. Forest plot for blood loss.

It is worth noting that this result was limited by the inconsistent use of various measurements used to report clinical outcomes, including the Oswestry Disability Index (ODI) [12,22–24], the Visual Analog Scale (VAS) [12,22–24], the median economic Prolo scale [25], the Beaujon score [27], the Disability Rating Index [12], and the Short Form of the Medical Outcomes Study (SF-36) [10,26]. Such inconsistency was not only a limitation for our study, or other meta-analyses, but also a restriction for the development of this field. It is difficult for readers to learn from studies with inconsistent outcome assessments. In the latest guidelines, the North American Spine Society recommends the ODI, VAS and SF-36 as some of the most clinically useful outcome measurements for spinal conditions [1]. With the development of evidence-

clinical outcome between the two procedures? Our results suggested that compared with patients who underwent PLF, patients who underwent PLIF had a significantly better clinical outcome. Understandably, PLIF provides superior mechanical strength owing to many biomechanical advantages, including immediate stabilization, maintenance of intervertebral disc height, support to the anterior column, restoration of load bearing capacity of ventral column and maybe, better sagittal balance [28,29]. By contrast, PLF usually requires more extensive exposure of the paravertebral muscles, which may lead to more severe low-back pain postoperatively. Each of the factors mentioned above may affect patients’ clinical satisfaction with the procedure. 5

Clinical Neurology and Neurosurgery 188 (2020) 105594

Y. Li, et al.

based medicine, those recommendations need to be further clarified in the future. The current study also found that PLIF presented a higher fusion rate than PLF. For successful fusion, an appropriate fusion site and prepared tissue bed are crucial. For PLIF, the cancellous bone of the vertebral body may offer a superior fusion bed and provide a larger surface area to support the fusion [22]. Additionally, performing fusion under compression is also a possible reason for the increased fusion rate of PLIF [23]. As for operation time, blood loss, and complication rate, the results of previous studies have been controversial. Some authors considered that PLIF was more invasive, involving extensive dissection that could lead to prolonged operation time and more blood loss, thus resulting in a higher complication rate [12,28]. Others reported that PLIF was associated with a shorter operating time, less blood loss, and a lower complication rate [22,27]. Here we also tried to compare PLF and PLIF in terms of these three variables. Operation time was not analyzed due to insufficient data. No significant difference in blood loss was identified. As for postoperative complication rate, both procedures could lead to common complications including pain, neurological deficits, spinal stenosis, infection, thrombosis, and hematoma. PLF was usually related to complications associated with hardware biomechanics, such as screw fracture and loosening of the construct, while PLIF was often correlated with complications resulting from its invasive nature, such as incidental dural injury. But overall, both procedures were similar in terms of postoperative complication rate. Notably, the results need to be interpreted with caution due to the limited number of studies included (3 for blood loss) and the existence of between-study heterogenicity (for blood loss and for complication rate). We speculated that the source of the heterogenicity might be the amount of collective experience among members of each surgical team. The current study had its limitations. Except for the variety in outcome measures, the number of included studies was limited; sample sizes were relatively small in some studies. More high-quality studies are needed.

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

5. Conclusion

[18]

In the current study, we performed a comprehensive comparison between PLF and PLIF in terms of therapeutic effects on spondylolisthesis. The results showed that PLIF was superior to PLF in terms of clinical outcome and fusion rate, while no significant difference was identified between procedures in terms of in postoperative complication rate and blood loss. The current study provides the most effective evidence for the comparison between PLF and PLIF, and the results may be used for guiding clinical practice.

[19]

[20]

[21]

Ethical approval [22]

This article does not contain any studies with human participants or animals performed by any of the authors.

[23]

Declaration of Competing Interest

[24]

The authors declare that they have no conflict of interest concerning this publication.

[25]

[26]

References

[27]

[1] D.S. Kreiner, J. Baisden, D.J. Mazanec, R.D. Patel, R.S. Bess, D. Burton, N.B. Chutkan, B.A. Cohen, C.H. Crawford 3rd, G. Ghiselli, A.S. Hanna, S.W. Hwang, C. Kilincer, M.E. Myers, P. Park, K.A. Rosolowski, A.K. Sharma, C.K. Taleghani, T.R. Trammell, A.N. Vo, K.D. Williams, Guideline summary review: an evidencebased clinical guideline for the diagnosis and treatment of adult isthmic spondylolisthesis, Spine J. 16 (12) (2016) 1478–1485. [2] P.G. Matz, R.J. Meagher, T. Lamer, W.L. Tontz Jr., T.M. Annaswamy, R.C. Cassidy, C.H. Cho, P. Dougherty, J.E. Easa, D.E. Enix, B.A. Gunnoe, J. Jallo, T.D. Julien, M.B. Maserati, R.C. Nucci, J.E. O’Toole, K. Rosolowski, J.N. Sembrano,

[28]

[29]

6

A.T. Villavicencio, J.P. Witt, Guideline summary review: an evidence-based clinical guideline for the diagnosis and treatment of degenerative lumbar spondylolisthesis, Spine J. 16 (3) (2016) 439–448. S. Jacobsen, S. Sonne-Holm, H. Rovsing, H. Monrad, P. Gebuhr, Degenerative lumbar spondylolisthesis: an epidemiological perspective: the Copenhagen Osteoarthritis Study, Spine 32 (1) (2007) 120–125. L. Kalichman, D.H. Kim, L. Li, A. Guermazi, V. Berkin, D.J. Hunter, Spondylolysis and spondylolisthesis: prevalence and association with low back pain in the adult community-based population, Spine 34 (2) (2009) 199–205. H. Moller, R. Hedlund, Surgery versus conservative management in adult isthmic spondylolisthesis–a prospective randomized study: part 1, Spine 25 (13) (2000) 1711–1715. J.N. Weinstein, J.D. Lurie, T.D. Tosteson, B. Hanscom, A.N. Tosteson, E.A. Blood, N.J. Birkmeyer, A.S. Hilibrand, H. Herkowitz, F.P. Cammisa, T.J. Albert, S.E. Emery, L.G. Lenke, W.A. Abdu, M. Longley, T.J. Errico, S.S. Hu, Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis, N. Engl. J. Med. 356 (22) (2007) 2257–2270. X.Y. Liu, Y.P. Wang, G.X. Qiu, X.S. Weng, B. Yu, Meta-analysis of circumferential fusion versus posterolateral fusion in lumbar spondylolisthesis, J. Spinal Disord. Tech. 27 (8) (2014) E282–93. J.M. Levin, J.E. Tanenbaum, M.P. Steinmetz, T.E. Mroz, S.C. Overley, Posterolateral fusion (PLF) versus transforaminal lumbar interbody fusion (TLIF) for spondylolisthesis: a systematic review and meta-analysis, Spine J. 18 (6) (2018) 1088–1098. S.J. McAnany, E.O. Baird, S.A. Qureshi, A.C. Hecht, J.G. Heller, P.A. Anderson, Posterolateral fusion versus interbody fusion for degenerative spondylolisthesis: a systematic review and meta-analysis, Spine 41 (23) (2016) E1408–E1414. A.M. Musluman, A. Yilmaz, T. Cansever, H. Cavusoglu, I. Colak, H.A. Genc, Y. Aydin, Posterior lumbar interbody fusion versus posterolateral fusion with instrumentation in the treatment of low-grade isthmic spondylolisthesis: midterm clinical outcomes, Journal of neurosurgery, Spine 14 (4) (2011) 488–496. Z. Yijian, L. Hao, Y. Huilin, P. Bin, Comparison of posterolateral fusion and posterior lumbar interbody fusion for treatment of degenerative spondylolisthesis: analysis of spino-pelvic sagittal balance and postoperative chronic low back pain, Clin. Neurol. Neurosurg. 171 (2018) 1–5. P. Ekman, H. Moller, T. Tullberg, P. Neumann, R. Hedlund, Posterior lumbar interbody fusion versus posterolateral fusion in adult isthmic spondylolisthesis, Spine 32 (20) (2007) 2178–2183. J.C. Urquhart, N. Alnaghmoosh, K.R. Gurr, S.I. Bailey, C. Tallon, S. Dehens, M.P. Rosas Arellano, C.S. Bailey, Posterolateral versus posterior interbody fusion in lumbar degenerative spondylolisthesis, Clin. Spine Surg. 31 (9) (2018) E446–E452. Y.P. Ye, H. Xu, D. Chen, Comparison between posterior lumbar interbody fusion and posterolateral fusion with transpedicular screw fixation for isthmic spondylolithesis: a meta-analysis, Arch. Orthop. Trauma Surg. 133 (12) (2013) 1649–1655. X. Liu, Y. Wang, G. Qiu, X. Weng, B. Yu, A systematic review with meta-analysis of posterior interbody fusion versus posterolateral fusion in lumbar spondylolisthesis, Eur. Spine J. 23 (1) (2014) 43–56. X.Y. Liu, G.X. Qiu, X.S. Weng, B. Yu, Y.P. Wang, What is the optimum fusion technique for adult spondylolisthesis-PLIF or PLF or PLIF plus PLF? A meta-analysis from 17 comparative studies, Spine 39 (22) (2014) 1887–1898. J. Luo, K. Cao, T. Yu, L. Li, S. Huang, M. Gong, C. Cao, X. Zou, Comparison of posterior lumbar interbody fusion versus posterolateral fusion for the treatment of isthmic spondylolisthesis, Clin. Spine Surg. 30 (7) (2017) E915–E922. R.C. Campbell, R.J. Mobbs, V.M. Lu, J. Xu, P.J. Rao, K. Phan, Posterolateral fusion versus interbody fusion for degenerative spondylolisthesis: systematic review and meta-analysis, Glob. Spine J. 7 (5) (2017) 482–490. Y.C. Chen, L. Zhang, E.N. Li, L.X. Ding, G.A. Zhang, Y. Hou, W. Yuan, Comparison of posterolateral fusion and posterior lumbar interbody fusion in the treatment of lumbar spondylolithesis: a meta-analysis, J. Invest. Surg. (2018) 1–8. L.A. Stewart, M. Clarke, M. Rovers, R.D. Riley, M. Simmonds, G. Stewart, J.F. Tierney, P.-I.D. Group, Preferred reporting items for systematic review and meta-analyses of individual participant data: the PRISMA-IPD statement, JAMA 313 (16) (2015) 1657–1665. G.H. Guyatt, A.D. Oxman, S. Sultan, P. Glasziou, E.A. Akl, P. Alonso-Coello, D. Atkins, R. Kunz, J. Brozek, V. Montori, R. Jaeschke, D. Rind, P. Dahm, J. Meerpohl, G. Vist, E. Berliner, S. Norris, Y. Falck-Ytter, M.H. Murad, H.J. Schunemann, G.W. Group, GRADE guidelines: 9. Rating up the quality of evidence, J. Clin. Epidemiol. 64 (12) (2011) 1311–1316. L. Cheng, L. Nie, L. Zhang, Posterior lumbar interbody fusion versus posterolateral fusion in spondylolisthesis: a prospective controlled study in the Han nationality, Int. Orthop. 33 (4) (2009) 1043–1047. M.R. Farrokhi, A. Rahmanian, M.S. Masoudi, Posterolateral versus posterior interbody fusion in isthmic spondylolisthesis, J. Neurotrauma 29 (8) (2012) 1567–1573. G.W. Lee, S.M. Lee, M.W. Ahn, H.J. Kim, J.S. Yeom, Comparison of posterolateral lumbar fusion and posterior lumbar interbody fusion for patients younger than 60 years with isthmic spondylolisthesis, Spine 39 (24) (2014) E1475–80. F.L. Dantas, M.N. Prandini, M.A. Ferreira, Comparison between posterior lumbar fusion with pedicle screws and posterior lumbar interbody fusion with pedicle screws in adult spondylolisthesis, Arq. Neuropsiquiatr. 65 (3B) (2007) 764–770. Y. Feng, L. Chen, Y. Gu, Z.M. Zhang, H.L. Yang, T.S. Tang, Restoration of the spinopelvic sagittal balance in isthmic spondylolisthesis: posterior lumbar interbody fusion may be better than posterolateral fusion, Spine J. 15 (7) (2015) 1527–1535. E. Dehoux, E. Fourati, K. Madi, B. Reddy, P. Segal, Posterolateral versus interbody fusion in isthmic spondylolisthesis: functional results in 52 cases with a minimum follow-up of 6 years, Acta Orthop. Belg. 70 (6) (2004) 578–582. M. van Dijk, T.H. Smit, S. Sugihara, E.H. Burger, P.I. Wuisman, The effect of cage stiffness on the rate of lumbar interbody fusion: an in vivo model using poly(l-lactic Acid) and titanium cages, Spine 27 (7) (2002) 682–688. K.T. Kim, S.H. Lee, Y.H. Lee, S.C. Bae, K.S. Suk, Clinical outcomes of 3 fusion methods through the posterior approach in the lumbar spine, Spine 31 (12) (2006) 1351–1357 discussion 1358.