This study was designed to systematically review and meta-analyze the functional and radiological outcomes between lateral and posterior approaches in adult degenerative scoliosis (ADS). Both lateral (lumbar, extreme, and oblique) and posterior interbody fusion (posterior lumbar and transforaminal) are used for deformity correction in patients with ADS with unclear comparison in this cohort of patients in the existing literature. A literature search using three electronic databases was performed to identify studies that reported outcomes of lateral (group L) and posterior interbody fusion (group P) in patients with ADS with curves of 10°–40°. Group P was further subdivided into minimally invasive surgery (MIS-P) and open posterior (Op-P) subgroups. Data on functional, radiological, and operative outcomes, length of hospital stay (LOHS), fusion rates, and complications were extracted and meta-analyzed using the random-effects model. A total of 18 studies (732 patients) met the inclusion criteria. No significant difference was found in functional and radiological outcomes between the two groups on data pooling. Total operative time in the MIS-P subgroup was less than that of group L (233.86 minutes vs. 401 minutes,
Adult degenerative scoliosis (ADS) or
Lateral lumbar interbody fusion (LLIF), extreme lateral interbody fusion (XLIF), and oblique lateral interbody fusion (OLIF) are new techniques performed from the lateral approach, which works on the principle of indirect decompression by increasing the foraminal height. The lateral approach is gaining popularity because of minimal trauma to muscles, large surface area for fusion, large size cages, and fusion segment saving [
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used to conduct this systematic review [
The following are the inclusion criteria for the studies: (1) full-text articles published in English; (2) randomized controlled trials (RCT) or observational or cross-sectional study design; (3) adult cohort of >40 years of age, either sex, with primary or
The following are the exclusion criteria: (1) non-English articles, without full-text; (2) case reports, review articles, systematic reviews, experimental studies, biomechanical studies, and conference abstracts; (3) patients with age of <40 years; (4) idiopathic, secondary adult, and syndromic or iatrogenic scoliosis; (5) heterogeneous cohort of patients operated for degenerative disc disease or spondylolisthesis alone without any deformity; (6) severe curves requiring osteotomy or circumferential fusion; (7) surgical procedure involving decompression alone or interbody fusion with anterior approach (anterior lumbar interbody fusion, ALIF) as a primary procedure; and (8) clinical outcomes reported other than VAS and ODI.
Critical appraisal of the selected studies was independently performed by two reviewers (S.M. and P.V.S.) using the National Institutes of Health (NIH) quality assessment tool that uses 12 objective criteria regarding study objectives, selection criteria, eligibility, representativeness of study participants, sample size, intervention, outcome measures, blinding, statistical analysis, outcome assessment, lost to follow-up, and group level assessment [
All data were extracted using text, tables, figures, and any supplementary material. Two authors (S.M. and P.V.S.) independently extracted the data from the selected articles, and any discrepancies were resolved after discussion with the senior authors (P.K. and B.S.). Data of each surgical procedure was separately extracted in the case of studies comparing the difference between two surgical procedures. Studies were divided into two groups as follows: group L (lateral approach, including LLIF/XLIF/OLIF) and group P (posterior approach, including PLIF and TLIF). Group P was further divided into two subgroups, the open posterior (Op-P) and MIS-posterior subgroups (MIS-P). The baseline characteristics included the first author’s name, country, year of publication, study design, sample size, gender, inclusion criteria, type of surgery, and duration of follow-up. The primary outcome included functional outcome reported as a change in VAS or ODI or both. Secondary outcome measures included the following: (1) radiographic outcomes, such as changes in Cobb’s angle, lumbar lordosis (LL), and sagittal vertical axis (SVA); (2) operative time and blood loss; (3) fusion rate; (4) length of hospital stay (LOHS); and (5) complications.
All statistical analyses were performed using Metafor package in R statistical software ver. 4.0.0 (The R Foundation for Statistical Computing, Vienna, Austria; 2020) [
The literature search used PubMed, Google Scholar, and Scopus and identified 288 articles. From selected articles, 158 additional records were identified [
A total of 18 studies (732 patients) were included in the quantitative analysis. Most of the studies were retrospective (n=15) while only three were prospective [
A total of 17 studies reported the pre- and post-intervention VAS scores for back pain. Eight and nine studies showed an SMD of −3.88 (95% CI, −4.67 to −3.09;
A total of 12 studies reported the pre- and post-intervention VAS scores for leg pain. Eight and four studies showed an SMD of −3.26 (95% CI, −4.25 to −2.26;
A total of 12 studies reported the ODI score pre- and post-intervention. Four and nine studies showed an SMD of −21.28 (95% CI, −26.4 to −16.16;
A total of 18 studies reported the pre- and post-intervention Cobb’s angle. Eight and 10 studies showed an SMD of −13.66 (95% CI, −16.64 to −10.68;
A total of 17 studies reported the pre- and post-intervention LL angle. Seven and 10 studies showed an SMD of 15.6 (95% CI, 6.57 to 24.63;
A total of nine studies reported the pre and post-intervention SVA. Four and five studies showed an SMD of −5.59 (95% CI, −10.32 to −0.87;
Operative parameters were compared between group L, MIS-P subgroup, and Op-P subgroup. Only one study in group L [
Only one study in group L reported total operative time with a mean of 401 minutes (95% CI, 351 to 450.9) [
Only one study in group L reported a mean total blood loss of 477 mL (95% CI, 201.7 to 752.2) [
A total of 10 studies reported the fusion rate. Four studies in group L had pooled fusion rate of 97.8% (95% CI, 94.8 to 100;
Four studies in group L had pooled LOHS of 4.15 days (95% CI, 2.79 to 5.5;
Pooled dural tear rate across two studies [
Wound related complications including Wound infections, dehiscence, delayed healing, and seroma were reported in four studies in group L with a pooled rate of 7.5% (95% CI, 1 to 13.9) [
Pooled rate of permanent neurological deficit across two studies [
In group L, pooled rate of transient neurological deficit across three studies was 24.3 (95% CI, 13.1 to 35.4), which mainly included transient hip flexor weakness and thigh numbness [
Complications included screw loosening, pull-out, and rod breakage. Pooled rate in group L across threes studies [
Pooled re-surgery rate across seven studies [
Pooled pseudoarthrosis rate across one study [
Complications, such as postoperative thigh pain, retrograde ejaculation, and visceral injuries, were reported only in group L. Pooled rate of postoperative thigh pain across four studies was 7.7% (95% CI, 2.2 to 13.2) [
Funnels plot for clinical (VAS back, VAS leg, and ODI) and radiographic outcomes (Cobb’s angle, LL, and SVA) showed asymmetry on visual inspection (
ADS should be differentiated from adult idiopathic scoliosis as it is characterized by progressive spinal element degeneration due to aging, which is usually asymmetric, leading to coronal and sagittal malalignment. It is also known as “
The systematic reviews in existing literature mostly compared different lumbar interbody fusion procedures in degenerative disc disease [
Most of the studies included VAS (back and leg) and ODI as patient-reported outcome measures (PROM) while other HRQOL measures, such as EuroQol-5 dimension (EQ5D), 36-item Short Form Health Survey (SF-36), and Scoliosis Research Society-22 (SRS-22), were missing; thus, they could not be analyzed. All studies showed improvement in VAS back (lateral approach: −55% reduction, posterior approach: −58% reduction) and VAS leg (lateral group: −60% reduction, posterior group: −70% reduction) postoperatively. No statistically significant difference was found between the two groups. The maximum change in VAS was seen in studies with the highest preoperative VAS score for VAS back [
All the studies showed a reduced coronal Cobb’s postoperatively in both the groups (lateral group: −56.5%, posterior group: −62.3%). Studies with maximum preoperative Cobb’s showed maximum treatment effect, consistent with the study results by Ledonio et al. [
We separately performed a subgroup analysis for operative parameters, such as blood loss, operative time, and LOHS, for open and MIS, due to the heterogeneity in the surgical procedures. The pooled total operative time of group L (401 minutes) was significantly more than the MIS-P subgroup (238.8 minutes), while it was comparable to the Op-P subgroup (380.2 minutes). Positioning of patients from lateral to prone for posterior instrumentation after interbody fusion (LLIF/XLIF/OLIF) explains the increased operative time in the lateral approach for ADS deformity correction. Additionally, the average number of interbody fusion levels was significantly more in group L compared with group P, this could explain the shorter operative time in group P. The pooled total blood loss of group L (477 mL) was comparable with the MIS-P subgroup (385 mL), while significantly different from the Op-P subgroup (1,325 mL). Li et al. [
No significant difference was found in the fusion rates and pseudoarthrosis between the two groups in this study. All included studies had a minimum follow-up period of 1 year; however, the range of follow-up was very wide from 1 to 13 years, which could have led to a bias in the comparison of fusion and pseudoarthrosis rates. In a meta-analysis conducted by Teng et al. [
Complications, such as transient sensorimotor weakness, were found significantly more in group L (24.3%) than in group P (5.6%). Transient thigh numbness was most commonly reported in group L, followed by transient hip flexor weakness. Hijji et al. [
This study has a few limitations. Due to the lack of comparative studies or RCT, high heterogeneity was noted due to the difference in inclusion criteria of patients, surgeon’s expertise, and choice of surgical procedure across the selected studies. Only one study [
Prospective comparative studies and RCTs with SRS-22, EQ5D, and SF-36 as PROM are required in the future for high-quality evidence. Additionally, the measurement of radiographic parameters should be uniform, preferably using an erect 91.44-cm anteroposterior and lateral X-ray of the whole spine with reporting of coronal, sagittal, and pelvic parameters.
Both posterior (PLIF/TLIF) and lateral (LLIF/XLIF/OLIF) lumbar interbody fusion procedures achieve similar functional and radiographic outcomes in patients with mild to moderate curves of ADS. Deformity correction from the lateral approach has an advantage in blood loss and LOHS over the Op-P subgroup. The MIS-P subgroup has less operative time than group L but with comparable blood loss and LOHS. No significant difference was found in the fusion rates, pseudoarthrosis, and complications between the two groups except for transient sensorimotor deficits. Few complications were found as approach-specific in each group.
No potential conflict of interest relevant to this article was reported.
Conception: SM, KA, BS, PK; data acquisition: SM, PVS, KA, SI, NG; data analysis: SM, PVS, KA, SI, GY, NG, VV, BS, PK; draft preparation: SM, SI, SS; interpretation of data: PVS, GY, VV; draft revision: GY, SS, NG; critical reviewing draft: BS, PK; and final approval of the manuscript: all authors.
Supplementary materials can be available from
Literature search strings.
National Institutes of Health quality assessment of included studies.
Flow diagram of the study as per PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocol. ALIF, anterior lumbar interbody fusion.
Forest plot showing Visual Analog Scale (VAS) back (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing Visual Analog Scale (VAS) leg (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing Oswestry Disability Index (ODI) (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing coronal Cobbs (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing lumbar lordosis (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing sagittal vertical axis (SVA) (subgroup 1: group L, subgroup 2: group P). CI, confidence interval.
Forest plot showing total operative time (subgroup 1: group L, subgroup 2: minimally invasive surgery [MIS] posterior subgroup, subgroup 3: open posterior subgroup). CI, confidence interval; NA, not available.
Forest plot showing total blood loss (subgroup 1: group L, subgroup 2: minimally invasive surgery [MIS] posterior subgroup, subgroup 3: open posterior subgroup). CI, confidence interval; NA, not available.
Forest plot showing fusion rates (subgroup 1: group L, subgroup 2: minimally invasive surgery [MIS] posterior subgroup, subgroup 3: open posterior subgroup). CI, confidence interval; Ev/Trt, number of events/number of treatments.
Forest plot showing length of hospital stay (subgroup 1: group L, subgroup 2: minimally invasive surgery posterior subgroup, subgroup 3: open posterior subgroup). CI, confidence interval.
Funnel plots.
Study characteristics
Author | Year | Country | Study period | Study design | No. of patients | Mean age (yr) | Sex (M/F) | Inclusion criteria | Open/MIS | Interbody fusion | Adjunct | No. of levels of interbodytotal (mean±SD) | FU (mo) |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Wang et al. [ |
2010 | USA | NS | Retrospective | 23 | 64.4 (range, 42–84) | 6/17 | ADS (coronal Cobbs >20° and/or loss of global sagittal balance) | MIS | LLIF | BMP-2 | 86 (3.7±1.25) | 13.4 |
Philips et al. [ |
2013 | USA | NS | Prospective | 107 | 68 (range, 45–87) | 29/78 | ADS (between T8 and S1; age >45 yr; scoliosis >10; ODI >30) | MIS | XLIF | 344 (3.2±2.5) | Minimum 24 | |
Khajavi et al. [ |
2014 | USA | 2005–2009 | Retrospective | 21 | 70.1 (range, 48–79) | 7/14 | ADS | MIS | XLIF | BMP-2 | 70 (3.3±2) | 24 (range,12–36) |
Baghdadi et al. |
2014 | USA | 1997–2011 | Retrospective matched cohort | 33 | 66 | 7/26 | DLS | MIS | LLIF | 181 (5±2) | 21.6 | |
Kim et al. [ |
2018 | South Korea | 2013–2016 | Retrospective | 32 | 68 (range, 58–78) | 4/28 | ADS (degenerative scoliosis; degenerative kyphoscoliosis) | MIS | OLIF | 122 (3.8±1) | 26.1±6.2 | |
Takatori et al. [ |
2018 | Japan | NS | Retrospective | 12 | 72 (range, 65–76) | 0/12 | ADS (type 1 adult scoliosis) | NS | XLIF | 34 (2.8±1.2) | 18.24 | |
Katz et al. [ |
2019 | USA | 2012–2016 | Retrospective | 27 | 66 | 10/17 | ADS (LLIF ≥3 levels; preoperative coronal Cobb angle ≥10°) | MIS | LLIF | Allograft DBM or synthetic BGS mixed with bone marrow aspirate | 100 (3.7±1) | 23±12 |
Yang et al. [ |
2021 | China | Jul 2017–Dec 2018 | Retrospective | 12 | 67.2 | 2/10 | DLS (coronal Cobb angle of >10° or SVA >5 cm; Silva and Lenke types 2–4) | MIS | OLIF | Allogenic bone graft | 34 (2.8±0.5) | 26.8±1.8 |
Wu et al. [ |
2008 | Taiwan | Apr 2000–Apr 2004 | Retrospective | 26 | 64.2 | 11/15 | DLS (failed conservative management) | Open | PLIF | 83 (2.4±0.7) | 36 (range, 24–72) | |
Tsai et al. [ |
2011 | Taiwan | May 2004–Jan 2007 | Retrospective | 58 | 68.9 | 11/47 | DLS (coronal Cobbs >10°; age >50 yr; refractory to medical treatment, minimal 2-yr FU) | Open | PLIF | (2.8±1.0) | 38.7±11.0 | |
Sabou et al. [ |
2019 | UK | 2009–2015 | Retrospective | 64 | 70.26 (range, 49–90) | 13/51 | DLS (coronal Cobbs >20°; PI–LL >11°; PT >25°) | Open | PLIF | 209 (3.2±0.75) | 64 (range, 36–108) | |
Scheufler et al. [ |
2010 | Germany | Jul 2006–Oct 2008 | Prospective | 30 | 73.2 (range, 64–88) | 12/18 | DLS (Cobbs >25°; back pain; progressive lumbar radiculopathy) | MIS | TLIF | BMP-2, propyhlactic cement augmentation in 10 patients | 134 (4.46±0) | 19.6 (range, 10–32) |
Wang et al. [ |
2013 | USA | NS | Retrospective | 25 | 72 (range, 62–84) | 8/17 | ADS (Cobbs >20°; failed conservative) | MIS | TLIF | BMP-2 | 80 (3.2±0.76) | Minimum 12 |
Du et al. [ |
2017 | China | Jan 2012–Nov 2014 | Retrospective | 96 | 63.9 | 43/53 | ADS (DLS with neurological symptoms; without central LCS) | Robot-assisted MIS | TLIF | Gelatin sponge impregnated with a mixture of 3 drugs to intraoperative nerve root block | 130 (1.4±0.5) | 16.4 (range, 13–24) |
Zhao et al. [ |
2018 | China | Jan 2008–Jan 2014 | Retrospective | 22 | 63.7 | 8/14 | DLS (Cobbs >10°; one level lumbar stenosis; refractory to medical management) | MIS | TLIF | 22 (1±0) | 24 | |
Author | Year | Country | Study period | Study design | No. of patients | Mean age (yr) | Sex (M/F) | Inclusion criteria | Open/MIS | Interbody fusion | Adjunct | No. of levels of interbodytotal (mean±SD) | FU (mo) |
Crandall et al. |
2009 | USA | NS | Prospective non-randomized | 20 | 67 (range, 49–81) | 2/18 | ADS (progressive deformity/spinal imbalance/failed conservative management 12 months) | Open | TLIF | BMP-2 | 54 (2.7±0.75) | 31 (range, 24–42) |
Burniekiene et al. [ |
2012 | USA | Feb 2003–Oct 2009 | Retrospective | 29 | 65.9 (range, 49–83) | 7/22 | DLS (DLS with neurogenic claudication and painful lumbar degenerative disc disease; refractory to conservative management) | Open | TLIF | 37 (2.2±0.75) | 30 (range, 15–47) | |
Zhu et al. [ |
2014 | China | Jan 1999–Dec 2007 | Retrospective | 95 | 58.5 | 37/58 | DLS (age >50 yr; coronal Cobbs >25°; minimal 5-yr FU) | Open | TLIF | 236 (2.8±1.0) | 93.6 (range, 60–156) |
M, male; F, female; MIS, minimally invasive surgery; SD, standard deviation; FU, follow-up; NS, not specified; ADS, adult degenerative scoliosis; LLIF, lateral lumbar interbody fusion; BMP, bone morphogenic protein; ODI, Oswestry Disability Index; XLIF, extreme lateral interbody fusion; DLS, degenerative lumbar scoliosis; OLIF, oblique lumbar interbody fusion; DBM, demineralized bone matrix; BGS, bone graft substitutes; SVA, sagittal vertical axis; PLIF, posterior lumbar interbody fusion; PI, pelvic incidence; LL, lumbar lordosis; PT, pelvic tilt; TLIF, transforaminal lumbar interbody fusion; LCS, lumbar canal stenosis.
Only LLIF group patients included.
Only TLIF group patients included.
Clinical and radiographic outcomes
Author | VAS back | VAS leg | ODI | Cobbs (°) | LL (°) | SVA (cm) | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
| |||||||||||||
Preop | Postop | Change | Preop | Postop | Change | Preop | Postop | Change | Preop | Postop | Change | Preop | Postop | Change | Preop | Postop | Change | |
Wang et al. [ |
7.3±1.9 | 3.35±2.01 | −3.95±2.76 | 4.35±2.9 | 1.57±1.7 | −2.78±3.36 | - | - | - | 31.4±13.37 | 11.4±9.2 | −20±16.23 | 37.7±8.5 | 45.5±6.9 | 7.8±10.9 | - | - | - |
| ||||||||||||||||||
Philips et al. [ |
- | - | −3.4±3.1 | - | - | −3.5±3.9 | - | - | - | 20.9±10.4 | 15.2±10.6 | −5.7±14.85 | - | - | - | - | - | - |
| ||||||||||||||||||
Khajavi et al. [ |
7±1.5 | 2.9±2.17 | −4.1±2.63 | 5.6±2.8 | 3.3±2.17 | −2.3±3.54 | 48.4±14.3 | 24.4±15.6 | −24±21.16 | 27.6±7.9 | 16.6±10.68 | −11±13.28 | 31.8±13.5 | 44±12.07 | 12.2±18.1 | - | - | - |
| ||||||||||||||||||
Baghdadi et al. [ |
8±2 | 2±3 | −6±3.6 | 4±4 | 1.5±3 | −2.5±5 | - | - | - | 39±18 | 16±11 | −23±21.09 | 38±14 | 44±14 | 6±19.79 | 4.9±4.7 | 3.8±3.2 | −1.1±5.68 |
| ||||||||||||||||||
Kim et al. [ |
5.6±1 | 2±0.7 | −3.6±1.2 | 5.1±2.6 | 1.5±1.1 | −3.6±1.94 | 55.4±16 | 22.6±5.8 | −32.8±17.01 | 21.55±6.97 | 9.6±4.9 | −11.95±8.52 | 5.79±16.04 | 46.54±6.11 | 40.75±17.12 | 13.66±6.32 | 2.94±2.63 | −10.72±6.84 |
| ||||||||||||||||||
Takatori et al. [ |
7.1±2.2 | 4.7±2.2 | −2.4±3.1 | 5.7±3.7 | 4.2±1.6 | −1.5±4.03 | - | - | - | 30.0±12.2 | 9.8±4.5 | −20.2±13 | 8.1±27.0 | 39.8±15.8 | 31.7±31.28 | 12.3±4.7 | 4.3±3.8 | −8±6.04 |
| ||||||||||||||||||
Katz et al. [ |
6±0.7 | 3.57±0.86 | −2.44±1.13 | 5.86±0.49 | 3.12±0.8 | −2.74±0.93 | 41.5±2.29 | 25.68±4.86 | −15.82±5.37 | 22.1±1.99 | 8.7±1.08 | −13.3±2.26 | 47.3±2.06 | 52.7±2.14 | 5.4±2.97 | - | - | - |
| ||||||||||||||||||
Yang et al. [ |
7.3±0.9 | 2.1±0.8 | −5.2±1.2 | 7.8±0.9 | 1.8±0.8 | −6±1.2 | 38.8±2.3 | 21.9±2 | −16.9±3.04 | 19.6±4.8 | 6.9±3.8 | −12.7±6.12 | 29.4±8.6 | 40.8±5.8 | 11.4±10.37 | 4.3±4.3 | 1.5±1 | −2.8±4.4 |
| ||||||||||||||||||
Wu et al. [ |
- | - | - | - | - | - | 58±11.5 | 25.8±19 | −32.2±22.2 | 16.5±5.7 | 7.4±3.4 | −9.1±6.63 | 22.2±15.7 | 30.1±13.3 | 7.9±20.57 | - | - | - |
| ||||||||||||||||||
Tsai et al. [ |
7.4±2 | 2.4±2 | −5±2.8 | - | - | - | 28.1±8 | 12.2±8.8 | −15.9±11.89 | 19.3±6.8 | 7.7±5.4 | −11.6±8.68 | 30.0±13.7 | 29.0±9.5 | −1±16.67 | −0.33±3.4 | −0.42±2.72 | −0.09±4.35 |
| ||||||||||||||||||
Sabou et al. [ |
7.5±2.25 | 3.7±2.8 | −3.7±3.59 | 6.5±3.75 | 3.75±3.2 | −2.75±4.93 | - | - | - | 30.34±13.88 | 7.98±7.08 | −22.36±15.58 | 28.05±18.21 | 43.8±10.46 | 15.75±21 | 6.08±5.68 | 4.96±3.82 | −1.12±6.84 |
| ||||||||||||||||||
Scheufler et al. [ |
7.5±0.8 | 2.68±0.76 | −4.82±1.1 | - | - | - | 57.2±6.9 | 25.5±5.3 | −31.7±8.7 | 42±13.32 | 10.3±7.8 | −31.7±15..43 | 8.8±8.9 | 36±6.9 | 27.2±11.26 | 3.16±1.52 | 0.8±0.84 | −2.36±1.73 |
| ||||||||||||||||||
Wang et al. [ |
7.6±1.7 | 3.4±1.7 | −4.2±2.4 | 5.1±2.02 | 1.8±0.99 | −3.3±2.25 | 44.9±11.8 | 24.1±11.6 | −20.8±16.54 | 29.2±9.3 | 9.0±5.0 | −20.2±10.55 | 27.8±12.9 | 42.6±12.1 | 14.8±17.68 | 7.4±4.9 | 4.3±5.7 | −3.1±7.51 |
| ||||||||||||||||||
Du et al. [ |
6.54±1.17 | 2.9±0.4 | −3.64±1.23 | 7.13±1.35 | 1.48±0.27 | −5.65±1.37 | 58.2±11.3 | 12.9±4.4 | −45.3±12.12 | 33.7±10.2 | 16.6±4.4 | −17.1±11.1 | 44.6±9.2 | 61.4±9.8 | 16.8±13.44 | 4.80±0.61 | 1.73±0.51 | −3.07±0.78 |
| ||||||||||||||||||
Zhao et al. [ |
6.2±1.8 | 2.2±0.7 | −4±1.93 | 8.2±0.7 | 1.4±1.4 | −6.8±1.56 | 62.4±16.1 | 24.2±9.3 | −38.2±18.59 | 20.7±7 | 12.7±7.1 | −8±9.97 | 39.5±13.6 | 43.6±10.6 | 4.1±17.24 | - | - | - |
| ||||||||||||||||||
Crandall et al. [ |
6.75±1.75 | 3.59±2 | −3.16±2.65 | - | - | - | 46.5±12.5 | 27.9±12.2 | −18.6±17.46 | 23.95±13.5 | 7.5±3.75 | −16.4±14 | 44.8±13.2 | 48.4±4 | 3.6±13.79 | - | - | - |
| ||||||||||||||||||
Burniekiene et al. [ |
7.6±1.5 | 3.6±2 | −4±2.5 | - | - | - | - | - | - | 32.3±10 | 15.4±12 | −16.8±15.62 | 37.6±9.75 | 40.5±8.3 | 2.9±12.8 | - | - | - |
| ||||||||||||||||||
Zhu et al. [ |
8.9±2 | 2±1.2 | −6.9±2.3 | - | - | - | 32.2±8.6 | 11.1±6.8 | −21.1±10.96 | 31.1±10.3 | 8.3±3.6 | −22.8±10.9 | 9.3±7.6 | 30.1±9.5 | 20.8±12.16 | - | - | - |
Values are presented as mean±standard deviation.
VAS, Visual Analog Scale; ODI, Oswestry Disability Index; LL, lumbar lordosis; SVA, sagittal vertical axis; Preop, preoperative; Postop, postoperative.
Complications
Complications | Subgroup | No. of studies | No. of patients | No. of complications | Pooled rate (95% CI) | |
---|---|---|---|---|---|---|
Dural tear | Lateral | 2 | 50 | 2 | 4 (1.4 to 9.4) | 0 |
Posterior | 4 | 107 | 10 | 7 (2 to 12) | 10.9 | |
Wound related | Lateral | 4 | 179 | 12 | 7.5 (1 to 13.9) | 44.32 |
Posterior | 6 | 368 | 14 | 2.7 (1.1 to 4.3) | 0 | |
Neurological deficit | Lateral | 2 | 140 | 11 | 7.8 (3.4 to 12.3) | 0 |
Posterior | 4 | 284 | 10 | 2.4 (0.1 to 4.8) | 41.21 | |
Transient sensori-motor defcit | Lateral | 3 | 56 | 14 | 24.3 (13.1 to 35.4) | 0 |
Posterior | 3 | 76 | 5 | 5.6 (5 to 10.7) | 0 | |
Implant or construct related | Lateral | 3 | 68 | 6 | 7.3 (1.2 to 13.5) | 0 |
Posterior | 5 | 439 | 71 | 14.9 (8.7 to 21.1) | 74.4 | |
Resurgery | Lateral | 7 | 235 | 23 | 7.9 (4.3 to 11.4) | 5.9 |
Posterior | 10 | 468 | 34 | 5.9 (2.5 to 9.3) | 71.38 | |
Pseudoarthrosis | Lateral | 1 | 23 | 1 | 4.3 (−4 to 12.7) | NA |
Posterior | 3 | 155 | 9 | 6.3 (−1 to 13.7) | 61.27 | |
Others | ||||||
Postoperative thigh pain | Lateral | 4 | 88 | 8 | 7.7 (2.2 to 13.2) | 0 |
Visceral injuries | Lateral | 2 | 139 | 3 | 2 (−2 to 6) | NA |
Retrograde ejaculation | Lateral | 1 | 27 | 1 | 3.7 (3.4 to 10.8) | NA |
Adult degenerative scoliosis | Posterior | 8 | 414 | 43 | 8.6 (5.2 to 12.2) | 43.18 |
CI, confidence interval; NA, not available.