Skip to main content
Download PDF
  • Systematic review
  • Open access
  • Published:

Risk factors for adjacent segment degeneration after posterior lumbar fusion surgery in treatment for degenerative lumbar disorders: a meta-analysis

Journal of Orthopaedic Surgery and Research volume 15, Article number: 582 (2020) Cite this article

Abstract

Study design

A meta-analysis.

Objective

We performed a meta-analysis to explore the incidence and risk factors of adjacent segment degeneration (ASD) after posterior lumbar fusion surgery.

Methods

An extensive search of the literature was performed in English database of PubMed, Embase, and Cochrane Library, and Chinese database of CNKI and WANFANG (up to May 2020). We collected factors including demographic data, surgical factor, and sagittal parameters. Data analysis was conducted with RevMan 5.3 and STATA 12.0.

Results

Finally, 19 studies were included in the final analysis. In our study, the rate of ASD after posterior lumbar fusion surgery was 18.6% (540 of 2896). Our data also showed that mean age, body mass index (BMI), the history of smoking and hypertension, preoperative adjacent disc degeneration, long-segment fusion, preoperative superior facet violation, high lumbosacral joint angle, pre- and post-operative L1-S1 sagittal vertical axis (SVA), post-operative lumbar lordosis (LL), and preoperative pelvic incidence (PI) were associated with the development of ASD. However, gender, history of diabetes, bone mineral density (BMD), preoperative Oswestry Disability Index (ODI) and Japanese Orthopedic Association (JOA), the type of fusion (PLIF vs TLIF), type of bone graft (auto- vs allograft), fusion to S1(vs non-fusion to S1), diagnose (lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis), preoperative pelvic tilt (PT), LL and sacral slope (SS), post-operative SS, PT and PI were not associated with the development of ASD.

Conclusions

In our study, many factors were correlated with the risk of ASD after posterior lumbar fusion surgery. We hope this article can provide a reference for spinal surgeons in treatment for lumbar degenerative diseases.

Introduction

Due to good clinical results, posterior lumbar fusion surgeries have been widely used in treatment for various lumbar degenerative diseases. Though the initially good clinical results after fusion, biomechanical change of the spine caused by fusion may accelerate the degeneration of the adjacent segment [1]. So, adjacent segment disease or adjacent segment degeneration (ASD) is considered to be a potential long-term complication of spinal fusion. The rate of ASD, considered radiographic changes without symptom, ranges from 5.2 to 49% in various studies after posterior lumbar fusion surgery [2].

Some researchers found that ASD may be caused by lumbar fusion, which can induce abnormal intradiscal pressure and too much movement at the adjacent spinal levels, resulting in abnormal discal stress distribution [3, 4]. However, Battie et al. [5] found that ASD after fusion was a natural process that was not related to fusion surgery. Recent articles have reported the risk factors for ASD including older age, female, expression of the estrogen receptor, the number of instrumented level, pre-existing degenerative condition at an adjacent motion segment, sagittal alignment change [6,7,8,9,10]. As far as we know, the risk factors for ASD remain controversial. Therefore, this study aims to explore the incidence and risk factors of ASD following posterior lumbar fusion surgery for degenerative lumbar disorders.

Methods

Search strategy

We searched for the English and Chinese language studies with the keywords: “adjacent segment degeneration” or “ASD”, and “lumbar surgery” in English database of PubMed, Embase, and Cochrane Library and Chinese database of CNKI and WANFANG. There was no limitation on the date of publication, which covered all previously published studies up to May 2020.

Eligibility criteria

Included articles must satisfy: (1) study population must be adult patients; (2) measured outcomes of the incidence and risk factors of ASD; (3) comparison: ASD group and non-ASD group; (4) the study must be meet the definition of ASD (defined as a radiological change in which narrowing of disc height was ≥3 mm, the progressive slipping of adjacent segments was ≥3 mm (in comparison with preoperative flexion and extension lateral radiographs), and the posterior opening of adjacent segments was 5°; (5) follow-up of more than 2 years; and (6) patients with lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis. Studies were excluded if they (1) were abstracts, letters, reviews, or case reports; (2) had repeated data; (3) did not report outcomes of interest;(4) patients treated for lumbar trauma, tumor, infection, inflammation, scoliosis; and (5) patients underwent any other lumbar surgery.

Data extraction and outcome measures

The data included the general characteristics of each study and the outcomes measured. General characteristics included first author, year of publication, country, the number of ASD patients and total patients, follow-up time, type of article, shown in Table 1. The outcomes include the rate of every risk factors. When the same population was reported in several publications, we retained only the most informative article or complete work to avoid duplication of information. Data were extracted independently by two authors. Any disagreements concerning paper eligibility were resolved by discussion and consensus. Test for risk of publication bias. We performed a visual inspection of the funnel plot for publication bias. The funnel plot should be asymmetric when there is publication bias and symmetric in the case of no publication bias. We performed Egger and Begg tests to measure the funnel plot asymmetry using a significance level of p<0.10. The trim and fill computation was used to estimate the effect of publication bias. Sensitive analysis overall because of the low heterogeneity of every factor, so we do not calculate sensitive analysis.

Table 1 Characteristics of included studies
Full size table

Statistical analysis

Only dichotomous outcomes were mentioned in our study, so odd ratios (OR) and 95% confidence intervals (CI) were calculated for outcomes. A p value<0.05 was judged as statistically significant. Random-effects or fixed-effects models were used depending on the heterogeneity of the studies included. Heterogeneity was analyzed with both the chi-squared test I2 test, where p value of<0.10 for the chi-squared and I2>50% implied heterogeneity [11]. All statistical analyses were performed using Review Manager version 5.3 (The Cochrane Collaboration, Oxford, UK) and STATA 12.0 (Stata Corporation, College Station, TX, USA).

Results

Study identification and selection

Initially, we collected a total of 488 records by the database search. A total of 201 records were excluded due to repetition and 230 records were removed for review based on the titles and abstracts. The remaining 57 records were retrieved for inclusion criteria and 28 of them were excluded, 10 did not report outcomes of interest. Finally, 19 articles that met our inclusion criteria were included in the present meta-analysis. The selection process included in this meta-analysis is shown in Fig. 1.

Fig. 1
figure 1

Flow diagram of the study selection

Full size image

Baseline characteristics and quality assessment

The main characteristics of the 19 articles (from 23 to 630 patients) that were published before May 2020 included in the meta-analysis are presented in Table 1. A total of 540 patients were suffering from ASD after posterior lumbar fusion surgery in a total of 2896 patients. According to the 19 included studies, the rate of ASD was 18.6% (ranged from 8.5 to 69.4%).

Because all studies included were retrospective studies, we used the Newcastle Ottawa Quality Assessment Scale (NOQAS) to assess the quality of each study. This scale for non-randomized case-controlled studies and cohort studies was used to allocate a maximum of nine points for the quality of selection, comparability, exposure, and outcomes for study participants. Of these studies, 13 studies scored eight points and 6 studies scored seven points. Hence, the quality of each study was relatively high (Table 2).

Table 2 The quality assessment according to the Newcastle Ottawa Quality Assessment Scale (NOQAS) of each study
Full size table

Assessment of risk factors of ASD

Age

Ten studies (1560 of 2896 patients) [12,13,14,15,16,17,18,19,20,21] reported the age of patients at an operational time between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.43; I2 = 1%, Fig. 2). The meta-analysis showed that age was associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.02, SMD = 1.66, 95% CI [0.28, 3.04], Fig. 2).

Fig. 2
figure 2

a The standardized mean difference (SMD) estimate for preoperative age in 2 groups. b The standardized mean difference (SMD) estimate for preoperative body mass index in 2 groups. c The odds ratio (OR) estimate for the history of smoking. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Body mass index (BMI)

Four studies (571 of 2896 patients) [12, 15, 16, 21] reported BMI of patients at an operational time between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.62; I2 = 0%, Fig. 2). The meta-analysis showed that BMI was associated with a significant increase in the incidence of ASD (fixed-effects model; p<0.0001, SMD = 3.17, 95% CI [2.48, 3.87], Fig. 2).

History of smoking

Four studies (1250 of 2896 patients) [13, 16, 19, 22] reported a history of smoking between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.24; I2 = 29%, Fig. 2). The meta-analysis showed that the history of smoking was associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.0002, OR = 1.77, 95% CI [1.24, 2.52], Fig. 2).

Gender

Fifteen studies (2592 of 2896 patients) [12, 14,15,16,17,18,19,20,21,22,23,24,25,26,27] reported gender between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.89; I2 = 0%, Fig. 3). The meta-analysis showed that gender was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.83, OR = 0.97, 95% CI [0.76, 1.25], Fig. 3).

Fig. 3
figure 3

a The odds ratio (OR) estimate for gender. b The odds ratio (OR) estimate for the history of diabetes. CI = confidence interval, df = degrees of freedom, M-H=Mantel–Haenszel

Full size image

History of diabetes

Four studies (1167 of 2896 patients) [12, 19, 21, 22] reported a history of diabetes between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.49; I2 = 0%, Fig. 3). The meta-analysis showed that the history of diabetes was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.37, OR = 0.81, 95% CI [0.51, 1.29], Fig. 3).

Bone mineral density (BMD)

Two studies (158 of 2896 patients) [14, 28] reported BMD at the operational time between ASD group and non-ASD group. There was not significant in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.31; I2 = 5%, Fig. 4). The meta-analysis showed that BMD was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.24, SMD = −0.07, 95% CI [−0.19, 0.05], Fig. 4).

Fig. 4
figure 4

a The standardized mean difference (SMD) estimate for bone mineral density in 2 groups. b The standardized mean difference (SMD) estimate for preoperative ODI score in 2 groups. c The standardized mean difference (SMD) estimate for preoperative JOA score in 2 groups. df = degrees of freedom, ODI = Oswestry disability index. JOA = Japanese Orthopedic Association, M-H = Mantel–Haenszel

Full size image

Preoperative Oswestry Disability Index (ODI)

Two studies (120 of 2896 patients )[20, 28] reported preoperative ODI between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.41; I2 = 0%, Fig. 4). The meta-analysis showed that preoperative ODI was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.77, SMD = −0.59, 95% CI [−4.66, 3.47], Fig. 4).

Preoperative Japanese Orthopedic Association (JOA)

Three studies (161 of 2896 patients) [17, 20, 28] reported preoperative JOA between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.42; I2 = 0%, Fig. 4). The meta-analysis showed that preoperative JOA was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.75, SMD = −0.21, 95% CI [−1.54, 1.11], Fig. 4).

History of hypertension

Three studies (650 of 2896 patients) [12, 13, 19] reported a history of hypertension between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.31; I2 = 15%, Fig. 5). The meta-analysis showed that the history of hypertension was associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.001, OR = 2.29, 95% CI [1.37, 3.82], Fig. 5).

Fig. 5
figure 5

a The odds ratio (OR) estimate for the history of hypertension. b The odds ratio (OR) estimate for preoperative Pfirrmann’s classification. c The odds ratio (OR) estimate for the length of fusion (short vs long fusion). CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Preoperative Pfirrmann’s classification at the adjacent segment

Eight studies (1750 of 2896 patients) [15, 16, 18, 19, 21, 22, 25, 26] reported Pfirrmann’s classification between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.49; I2 = 0%, Fig. 5). The meta-analysis showed that preoperative Pfirrmann’s classification of more than three levels was associated with a significant increase in the incidence of ASD (fixed-effects model; p<0.0001, OR = 0.36, 95% CI [0.26, 0.50], Fig. 5).

Short versus long fusion

Two studies (172 of 2896 patients) [24, 26] reported the length of fusion between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.20; I2 = 14%, Fig. 5). The meta-analysis showed that long fusion ≥3 was associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.02, OR = 0.29, 95% CI [0.10, 0.85], Fig. 5).

Preoperative superior facet violation

Two studies (958 of 2896 patients) [15, 22] reported a preoperative superior facet violation between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.53; I2 = 0%, Fig. 6). The meta-analysis showed that preoperative superior facet violation was associated with a significant increase in the incidence of ASD (fixed-effects model; p<0.00001, OR = 29.74, 95% CI [17.20, 51.43], Fig. 6).

Fig. 6
figure 6

a The odds ratio (OR) estimate for preoperative superior facet violation. b The standardized mean difference (SMD) estimate for preoperative lumbar-sacral joint angle in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Preoperative lumbosacral joint angle

Two studies (99 of 2896 patients) [21, 28] reported a preoperative lumbosacral joint angle between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.22; I2 = 34%, Fig. 6). The meta-analysis showed that the preoperative lumbosacral joint angle was associated with a significant increase in the incidence of ASD (fixed-effects model; p<0.00001, SMD = −1.87, 95% CI [−2.50, −1.25], Fig. 6).

Type of fusion (PLIF versus TLIF)

Six studies (650 of 2896 patients) [15, 16, 21, 24,25,26] reported the type of fusion (PLIF versus TLIF) between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.80; I2 = 0%, Fig. 7). The meta-analysis showed that type of fusion (PLIF versus TLIF) was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.93, OR = 1.03, 95% CI [0.61, 1.73], Fig. 7).

Fig.7
figure 7

a The odds ratio (OR) estimate for the type of fusion (PLIF vs TLIP). b The odds ratio (OR) estimate for the type of graft (auto- vs allograft). c The odds ratio (OR) estimate for fusion to S1 (vs non-fusion to S1). CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Type of graft (Auto- versus allograft)

Three studies (232 of 2896 patients) [16, 21, 26] reported the type of graft (auto- versus allograft) between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.34; I2 = 4%, Fig. 7). The meta-analysis showed that type of graft (auto- versus allograft) was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.73, OR = 0.89, 95% CI [0.45, 1.75], Fig. 7).

Fusion to S1 (versus non-fusion to S1)

Six studies (1071 of 2896 patients) [15, 18, 23, 24, 26, 29] reported fusion to S1 between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (p for heterogeneity = 0.49; I2 = 0%, Fig. 7). The meta-analysis showed that fusion to S1 was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.89, OR = 0.97, 95% CI [0.60, 1.55], Fig. 7).

Diagnose (lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis)

Four studies (392 of 2896 patients) [15, 16, 24, 26] reported diagnose (lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis) between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (three p for heterogeneity = 0.41, 0.42, 0.24; I2 = 0, 0, 29%, Fig. 8). The meta-analysis showed that diagnose (lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis) was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.33, 0.39, 0.83; OR = 0.03, 95% CI [−0.03, 0.10]; OR = 1.47, 95% CI [0.61, 3.51]; OR = 1.13, 95% CI [0.37, 3.42], respectively, Fig. 8).

Fig. 8
figure 8

a The odds ratio (OR) estimate for diagnosis (lumbar spinal stenosis vs lumbar spondylolisthesis). b The odds ratio (OR) estimate for diagnosis (lumbar disc herniation vs lumbar spinal stenosis). c The odds ratio (OR) estimate for diagnosis (lumbar disc herniation vs lumbar spondylolisthesis). CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Pre- and post-operative L1-S1 sagittal vertical axis (SVA)

Two studies (714of 2896 patients) [16, 22] reported pre- and post-operative L1-S1 SVA between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (two p for heterogeneity = 0.65, 0.84; I2 = 0, 0%, Fig. 9). The meta-analysis showed that both pre- and post-operative L1-S1 SVA were associated with a significant increase in the incidence of ASD (fixed-effects model; p<0.00001, SMD = 6.94, 95% CI [4.85, 9.03]; SMD = 3.87, 95% CI [2.33, 5.40], respectively, Fig. 9).

Fig. 9
figure 9

a The standardized mean difference (SMD) estimate for preoperative L1-S1 sagittal vertical axis (SVA)in 2 groups. b The standardized mean difference (SMD) estimate for post-operative L1-S1SVA in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Pre- and post-operative pelvic tilt (PT)

Five studies (669 of 2896 patients) [19, 20, 22, 29, 30] reported pre- and post-operative PT between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (two p for heterogeneity = 0.34, 0.55; I2 = 11, 0%, Fig. 10). The meta-analysis showed that both pre- and post-operative PT were not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.53, 0.14, SMD = −0.50, 95% CI [−2.06, 1.06]; SMD = 1.89, 95% CI [−0.59,4.36], respectively, Fig. 10).

Fig. 10
figure 10

a The standardized mean difference (SMD) estimate for preoperative pelvic tilt (PT)in 2 groups. b The standardized mean difference (SMD) estimate for post-operative PT in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Pre- and post-operative sacral slope (SS)

Eight studies (1492 of 2896 patients) [14, 16, 19, 20, 22, 23, 29, 30] reported pre- and post-operative SS between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (two p for heterogeneity = 0.92, 0.66; I2 = 0, 0%, Fig. 11). The meta-analysis showed that both pre- and post-operative SS was not associated with a significant increase in the incidence of ASD (fixed-effects model; p = 0.07, 0.21, SMD = −1.27, 95% CI [−2.63, 0.09]; SMD = −1.73, 95% CI [−4.42, 0.96], respectively, Fig. 11).

Fig. 11
figure 11

a The standardized mean difference (SMD) estimate for preoperative sacral slope (SS) in 2 groups. b The standardized mean difference (SMD) estimate for post-operative SS in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Pre- and post-operative pelvic incidence (PI)

Five studies (997 of 2896 patients) [18,19,20, 23, 30] reported pre- and post-operative PI between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (two p for heterogeneity = 0.47, 0.33; I2 = 0, 0%, Fig. 12). The meta-analysis showed that preoperative PI was associated with a significant increase in the incidence of ASD; however, post-operative PI was not associated with it (fixed-effects model; p = 0.02, 0.67, SMD = −2.13, 95% CI [−3.95, −0.30]; SMD = 0.90, 95% CI [−3.19, 4.99], respectively, Fig. 12).

Fig. 12
figure 12

a The standardized mean difference (SMD) estimate for preoperative pelvic incidence (PI) in 2 groups. b The standardized mean difference (SMD) estimate for post-operative PI in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Pre- and post-operative lumbar lordosis (LL)

Twelve studies (1423 of 2896 patients) [12, 14, 15, 17, 19,20,21, 23, 26, 28,29,30] reported pre- and post-operative LL between ASD group and non-ASD group. There was no significance in the test for heterogeneity and the studies had low heterogeneity (two p for heterogeneity = 0.04, 0.41; I2 = 45, 0%, Fig. 13). The meta-analysis showed that preoperative LL was not associated with a significant increase in the incidence of ASD; however, post-operative LL was associated with it (fixed-effects model; p = 0.10, 0.002, SMD = −0.75, 95% CI [−1.65, 0.14]; SMD = −3.70, 95% CI[−5.99, −1.42], respectively, Fig. 13).

Fig. 13
figure 13

a The standardized mean difference (SMD) estimate for preoperative lumbar lordosis (LL) in 2 groups. b The standardized mean difference (SMD) estimate for post-operative LL in 2 groups. CI = confidence interval, df = degrees of freedom, M-H = Mantel–Haenszel

Full size image

Publication bias

After detection of publication bias by STATA 12.0, there was no publication bias found for all included studies (all p > 0.05).

Discussion

Degenerative lumbar diseases are common diseases in the clinic, especially in the elderly population. Posterior lumbar fusion surgery is a popular surgical procedure in treatment for patients with degenerative spinal disorders. ASD, as a complication of posterior lumbar spinal fusion surgery, always attracts the attention of spine surgeons [3, 6, 8, 9]. In 2004, Paul Park [31] reviewed articles involved ASD after lumbar fusion surgery and concluded that age, posterior lumbar interbody fusion, injury to the facet joint of the adjacent segment, long-segment fusion, sagittal alignment, pre-existing degenerated disc at the adjacent level, LL, osteoporosis, female gender, post-menopausal state were potential risk factors for ASD. Furthermore, in 2012, Brandon D [32] performed a mate-analysis on ASD indicating that age more than 60 years, male sex, facet degeneration, degenerative disc disease, adjacent to the fused segment, multilevel fusion, fusion to L5, and excessive disc height distraction. However, this article was limited to higher-quality studies because of only 5 included studies. Although many scholars pay more attention to ASD after spine surgery, the risk factors associated with ASD are controversial.

Thus, we perform a meta-analysis to evaluate the risk factors associated with ASD [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]. The rate of ASD after posterior lumbar fusion surgery was 18.6% (ranged from 8.5 to 69.4%) in this study. The pooled results from this meta-analysis suggested that gender of patients, history of diabetes, BMD, preoperative ODI and JOA, the type of fusion (PLIF vs TLIF), type of bone graft (auto- vs allograft), fusion to S1 (vs non-fusion to S1), diagnose (lumbar disc herniation, lumbar spinal stenosis, lumbar spondylolisthesis), preoperative PT, LL and SS, post-operative SS, PT, and PI were not were associated with a significant increase in the incidence of ASD. However, older age, BMI, the history of smoking and hypertension, preoperative adjacent disc degeneration, long-segment fusion, superior facet violation, high lumbosacral joint angle, pre- and post-operative L1-S1 SVA, post-operative LL, and preoperative PI were associated with a significant increase in the incidence of ASD.

Aota [33] observed that patients older than 55 years of age were at risk of ASD. However, some articles [12, 16, 19] indicated that age was not a significant factor of ASD. In our study, the older age demonstrated an increased risk of developing ASD. The reason that the older spine is less flexible and more difficult to adapt to the biomechanical alterations after fusion might partially explain this difference [34].

Bagheri [35] demonstrated that patients who had higher preoperative BMI showed a statistically increase in the risk of developing ASD. Wang [15] reported that BMI more than 25 kg/m2 was found to be a risk factor for ASD. Our results were consistent with the previous studies [12, 15, 35]. In the present study, the history of smoking and hypertension was considered a risk of ASD, but the reason was unexplained.

Anandjiwala [26] demonstrated that preoperative disc degeneration at an adjacent level was a significant indicator of developing ASD. Our finding confirmed that patients with preoperative Pfirrmann’s classification of more than 3 in the radiographic adjacent segment were found to be a statistic risk factor of ASD. Compared with patients whose Pfirrmann’s classification was less than 3, biomechanical alterations caused by fusion make it more vulnerable to experience degeneration at the adjacent level that preoperative Pfirrmann’s classification was more than 3. Additionally, we also found that preoperative superior facet violation was related to the increasing rate of ASD. Actually, preoperative superior facet violation is a form of degeneration at the adjacent segment, causing deduced adaptability to biomechanical change.

Long fusion that was more than 3 levels demonstrated a significant relationship with the incidence of ASD. Ghiselli et al. [36] reported that multiple-level fusion had a three times higher risk for developing ASD than single-level fusion. Decreased elasticity and increased stiffness of the lumbar segment caused by long fusion are difficult to accommodate biomechanical changes at the adjacent motion segment including stress concentration and intradiscal pressures, which make it easier to experience degeneration at the adjacent segment.

It remains controversial as to whether an association exists between sagittal malalignment and ASD. Anandjiwala [26] showed that sagittal alignment parameters were not associated with the development of ASD. Zhong [19] demonstrated the same results. While other articles showed that patients with post-operative sagittal imbalance have a statistically significant increased chance of developing ASD [37, 38], our finding showed that partial sagittal parameters had a close relationship with the development of ASD. Djurasovic [39] found the patients developing ASD with a significantly lower level of LL. Wu [40] reported that the post-operative angle of LL was 7.9° higher than the preoperative angle in patients after PLIF. Nakashima [41] concluded that appropriate post-operative LL after surgery could play a crucial role in the prevention of ASD. In the present study, post-operative LL was found to be a risk of ASD, while preoperative LL was not. Nakashima [42] believed that the patients with a high preoperative PI value have a significantly higher risk of ASD after spinal fusion, but he did not explore whether post-operative PI is the risk of ASD. In our study, regarding the role of preoperative PI to ASD, we were consistent with Nakashima. Nevertheless, post-operative PI was not associated with a significant increase in the incidence of ASD.

Kumar [43] showed that patients with a normal C7 plumb line alignment had a lower incidence of adjacent-level change following lumbar spinal fusion. Liang [16] found that patients with normal post-operative lumbar sagittal alignment had a lower incidence of ASD. In addition, he concluded that the preoperative L1-S1 SVA was found to be a potential risk factor for predicting ASD after lumbar spine fusion. Our findings showed that both preoperative and post-operative L1-S1 SVA were associated with a significant increase in the incidence of ASD. Correction of sagittal alignment by spinal fusion plays an important role in the development of ASD. It ensures the proper conditions for fusion and facilitates the preservation of the adjacent segment. Abnormal sagittal alignment can have a deleterious effect on the adjacent segment.

There were several limitations to this study. First, we evaluated only radiological and asymptomatic ASD, but symptomatic one was not considered. Additionally, some factors had two included studies. The mentioned above might impact the accuracy of the results. Second, some factors, like PI-LL, might be risk factors for ASD. Because related studies were few and could not get pooled result, we excluded them. Third, follow-up time varied between the studies and thus may influence our results. Fourth, all the included studies come from Asian countries, which may affect the bias of results.

In conclusion, older age, BMI, the history of smoking and hypertension, preoperative adjacent disc degeneration, long-segment fusion, superior facet violation, high lumbosacral joint angle, pre- and post-operative L1-S1 SVA, post-operative LL, and preoperative PI were associated with a significant increase in the incidence of ASD. In this meta-analysis, we can clearly see which kind of people more likely had ASD after surgery. We hope this article can provide a reference for spinal surgeons in the treatment of lumbar degeneration diseases. Meanwhile, it is helpful for future study on ASD. Further large-scale, well-designed studies are urgently needed.

Availability of data and materials

Yes.

Abbreviations

ASD:

Adjacent segment degeneration

BMI:

Body mass index

BMD:

Bone mineral density

LL:

Lumbar lordosis

PI:

Pelvic incidence

PT:

Pelvic tilt

SS:

Sacral slope

SVA:

Sagittal vertical axis

References

  1. Cheh G, Bridwell KH, Lenke LG, et al. Adjacent segment disease following lumbar/thoracolumbar fusion with pedicle screw instrumentation: a minimum 5-year follow-up. Spine. 2007;32(20):2253–7.

    Article  Google Scholar 

  2. Park P, Garton HJ, Gala VC et al. Adjacent segment disease after lumbar or lumbosacral fusion: review of the literature. Spine (Phila Pa 1976). 2004 29:1938–1944. doi: 00007632-200409010-00019[pii].

  3. Lee CS, Hwang CJ, Lee SW, et al. Risk factors for adjacent segment disease after lumbar fusion. Eur Spine J. 2009;11:1637–43.

    Article  Google Scholar 

  4. Quinell RC, Stockdale HR. Some experimental observations on the influence of a single floating fusion on the remaining lumbar spine. Spine. 1981;6:263–7.

    Article  Google Scholar 

  5. Battie MG, Videman T, Parent E. Lumbar disc degeneration: epidemiology and genetics influences. Spine. 2004;29:2679–90.

    Article  Google Scholar 

  6. Puvanesarajah V, Cancienne JM, Werner BC, et al. Perioperative complications associated with posterolateral spine fusions: a study of elderly medicare beneficiaries. Spine J. 2016;16:S266–7 [Epub ahead of print].

    Article  Google Scholar 

  7. Park P, Garton HJ, Gala VC, et al. Adjacent segment disease after lumbar or lumbosacral fusion: review of the literature. Spine (Phila Pa 1976). 2004;29:1938–44.

    Article  Google Scholar 

  8. Kumar MN, Baklanov A, Chopin D. Correlation between sagittal plane changes and adjacent segment degeneration following lumbar spine fusion. Eur Spine J. 2001;10:314–9.

    Article  CAS  Google Scholar 

  9. Etebar S, Cahill DW. Risk factors for adjacent-segment failure following lumbar fixation with rigid instrumentation for degenerative instability. J Neurosurg. 1999;90:163–9.

    CAS  PubMed  Google Scholar 

  10. Ha KY, Chang CH, Kim KW, Kim YS, Na KH, Lee JS. Expression of estrogen receptor of the facet joints in degenerative spondylolisthesis. Spine (Phila Pa 1976). 2005;30:562–6 00007632-200503010-00015[pii].

    Article  Google Scholar 

  11. Xiao SW, Jiang H, Yang LJ, Xiao ZM. Anterior cervical discectomy versus corpectomy for multilevel cervical spondylotic myelopathy: a meta-analysis. Eur Spine J. 2015;24:31–9. https://doi.org/10.1007/s00586-014-3607-1.

    Article  PubMed  Google Scholar 

  12. Zheng G, Wang C, Wang T, et al. Relationship between postoperative lordosis distribution index and adjacent segment disease following L4-S1 posterior lumbar interbody fusion. J Orthop Surg Res. 2020;15(1):129. https://doi.org/10.1186/s13018-020-01630-9.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Lin H, Zhaopeng C, Keng C, et al. Retrospective analysis of preoperative risk factors of adjacent segment degeneration after posterior lumbar interbody fusion. J Clin Orthop Res. 2017;2(1):1–5.

    CAS  Google Scholar 

  14. Mi H, Yu M, Xiaoguang L, et al. Correlative factors for adjacent segment degeneration after lumbar spinal fusion. Chin J Spine Spinal Cord. 2014;24(3):199–203.

    Google Scholar 

  15. Wang H, Ma L, Yang D, et al. Incidence and risk factors of adjacent segment disease following posterior decompression and instrumented fusion for degenerative lumbar disorders. Medicine (Baltimore). 2017;96(5):e6032. https://doi.org/10.1097/MD.0000000000006032.

    Article  Google Scholar 

  16. Liang J, Dong Y, Zhao H. Risk factors for predicting symptomatic adjacent segment degeneration requiring surgery in patients after posterior lumbar fusion. J Orthop Surg Res. 2014;9:97. https://doi.org/10.1186/s13018-014-0097-0.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Makino T, Honda H, Fujiwara H, Yoshikawa H, Yonenobu K, Kaito T. Low incidence of adjacent segment disease after posterior lumbar interbody fusion with minimum disc distraction: A preliminary report. Medicine (Baltimore). 2018;97(2):e9631. https://doi.org/10.1097/MD.0000000000009631.

    Article  Google Scholar 

  18. Heo Y, Park JH, Seong HY, et al. Symptomatic adjacent segment degeneration at the L3–4 level after fusion surgery at the L4–5 level: evaluation of the risk factors and 10-year incidence. Eur Spine J. 2015:24(11). https://doi.org/10.1007/s00586-015-4188-3.

  19. Zhong ZM, Deviren V, Tay B, et al. Adjacent segment disease after instrumented fusion for adult lumbar spondylolisthesis: incidence and risk factors. Clin Neurol Neurosurg. 2017;2017(156):29–34. https://doi.org/10.1016/j.clineuro.2017.02.020.

    Article  Google Scholar 

  20. Ma Z, Huang S, Sun J, Li F, Sun J, Pi G. Risk factors for upper adjacent segment degeneration after multi-level posterior lumbar spinal fusion surgery. J Orthop Surg Res. 2019;14(1):89. https://doi.org/10.1186/s13018-019-1126-9.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Ushio S, Hirai T, Yoshii T, et al. Preoperative risk factors for adjacent segment degeneration after two-level floating posterior fusion at L3-L5. Spine Surg Relat Res. 2019;4(1):43–9. https://doi.org/10.22603/ssrr.2019-0003.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Bagheri SR, Alimohammadi E, Froushani AZ, et al. Adjacent segment disease after posterior lumbar instrumentation surgery for degenerative disease: Incidence and risk factors. J Orthop Surg. 2019;27(2):1–6.

    Article  Google Scholar 

  23. Bae JS, Lee S-H, Kim J-S, et al. Adjacent segment degeneration after lumbar interbody fusion with percutaneous pedicle screw fixation for adult low-grade isthmic spondylolisthesis: minimum 3 years of follow-up. Neurosurgery. 2010;67:1600–8. https://doi.org/10.1227/NEU.0b013e3181f91697.

    Article  PubMed  Google Scholar 

  24. Cho K-S, Kang S-G, Yoo D-S, et al. Risk factors and surgical treatment for symptomatic adjacent segment degeneration after lumbar spine fusion. J Korean Neurosurg Soc. 2009;46:425–30.

    Article  Google Scholar 

  25. Soh J, Lee JC, Shin BJ. Analysis of risk factors for adjacent segment degeneration occurring more than 5 years after fusion with pedicle screw fixation for degenerative lumbar spine. Asian Spine J. 2013;7(4):273–81.

    Article  Google Scholar 

  26. Anandjiwala J, Seo J-Y, Ha K-Y, et al. Adjacent segment degeneration after instrumented posterolateral lumbar fusion: a prospective cohort study with a minimum five-year follow-up. Eur Spine J. 2011;20:1951–60. https://doi.org/10.1007/s00586-011-1917-0.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Miyagi M, Ikeda O, Ohtori S, et al. Additional decompression at adjacent segments leads to adjacent segment degeneration after PLIF. Eur Spine J. 2013;22:1877–83. https://doi.org/10.1007/s00586-013-2694-8.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Chen B-L, Wei F-X, Ueyama K, et al. Adjacent segment degeneration after single-segment PLIF: the risk factor for degeneration and its impact on clinical outcomes. Eur Spine J. 2011;20:1946–50. https://doi.org/10.1007/s00586-011-1888-1.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Weishi L, Zhouran S, Guo Y, et al. Effect of spinopelvic sagittal alignment on the development of adjacent segment degeneration after posterior lumbar fusion: investigation on cases with a minimum of 6 years of follow up. Chin J Spine Spinal Cord. 2018;28(10):865–72.

    Google Scholar 

  30. Guo Y, Zhouran S, Siyu Z, et al. The effect of pre-existing degeneration at adjacent segment on postoperative adjacent segment degeneration and surgical clinical outcome. Spine Spinal Cord. 2020;30(2):103–10.

    Google Scholar 

  31. Park P, Hugh J. Garton, et al. Adjacent segment disease after lumbar or lumbosacral fusion: review of the literature. SPINE. 2004;29(17):1938–44.

    Article  Google Scholar 

  32. Lawrence BD, Wang J, et al. Predicting the risk of adjacent segment pathology after lumbar fusion. SPINE. 2012;37(22S):S123–32..

    Article  Google Scholar 

  33. Aota Y, Kumano K, Hirabayashi S. Postfusion instability at the adjacent segments after rigid pedicle screw fixation for degenerative lumbar spinal disorders. J Spinal Disord. 1995;8:464–73.

    Article  CAS  Google Scholar 

  34. Kim YJ, Bridwell KH, Lenke LG, Glattes CR, Rhim S, Cheh G. Proximal junctional kyphosis in adult spinal deformity after segmental posterior spinal instrumentation and fusion: minimum five-year follow-up. Spine. 2008;33(20):2179–84. https://doi.org/10.1097/BRS.0b013e31817c0428.

    Article  PubMed  Google Scholar 

  35. Bagheri SR, Alimohammadi E, Zamani Froushani A, et al. Adjacent segment disease after posterior lumbar instrumentation surgery for degenerative disease: incidence and risk factors. J Orthop Surg (Hong Kong). 2019;27:2309499019842378.

    Article  Google Scholar 

  36. Ghiselli G, Wang JC, Bhatia NN, Hsu WK, Dawson EG. Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg Am. 2004;86-A:1497–503.

    Article  Google Scholar 

  37. Zencica P, Chaloupka R, Hladikova J, et al. Adjacent segment degeneration after lumbosacral fusion in spondylolisthesis: a retrospective radiological and clinical analysis. Acta Chir Orthop Traumatol Cech. 2010;77:124–30.

    CAS  PubMed  Google Scholar 

  38. Chung KJ, Suh SW, Swapnil K, et al. Facet joint violation during pedicle screw insertion: a cadaveric study of the adult lumbosacral spine comparing the two pedicle screw insertion techniques. Int Orthop. 2007;31:653–6.

    Article  Google Scholar 

  39. Nakashima H, Kawakami N, Tsuji T, et al. Adjacent segment disease afterposterior lumbar interbody fusion: based on cases with a minimum of 10 years of follow-up. Spine (Phila Pa 1976). 2015;40:E831–41.

    Article  Google Scholar 

  40. Wu CH, Wong CB, Chen LH, et al. Instrumented posterior lumbar interbodyfusion for patients with degenerative lumbar scoliosis. J Spinal Disord Tech. 2008;21:310–5.

    Article  Google Scholar 

  41. Djurasovic MO, Carreon LY, Glassman SD, et al. Sagittal alignment as a riskfactor for adjacent level degeneration: a case-control study. Orthopedics. 2008;31:546.

    Article  Google Scholar 

  42. Nakashima H, Kawakami N, Tsuji T, et al. Adjacent segment disease afterposterior lumbar interbody fusion: based on cases with a minimum of 10 years of follow-up. Spine. 2015;40(14):E831–41.

    Article  Google Scholar 

  43. Kumar MN, Baklanov A, Chopin D. Correlation between sagittal planechanges and adjacent segment degeneration following lumbar spinefusion. Eur Spine J. 2001;10(4):314–9.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

Not applicable.

Author information

Authors and Affiliations

  1. Wuxi No.9 People Hospital, Wuxi, China

    Tao Wang

  2. Department of Spinal Surgery, The Third Hospital of Hebei Medical University, No. 139 Ziqiang Road, Shijiazhuang, 050051, China

    Wenyuan Ding

Authors
  1. Tao Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wenyuan Ding
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Wang Tao collected data and write the essay; Ding Wenyuan is responsible for the design. The author (s) read and approved the final manuscript.

Corresponding author

Correspondence to Wenyuan Ding.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Yes.

Competing interests

There are no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, T., Ding, W. Risk factors for adjacent segment degeneration after posterior lumbar fusion surgery in treatment for degenerative lumbar disorders: a meta-analysis. J Orthop Surg Res 15, 582 (2020). https://doi.org/10.1186/s13018-020-02032-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13018-020-02032-7

Keywords

Journal of Orthopaedic Surgery and Research

ISSN: 1749-799X