The occurrence of adjacent segment pathology (ASP) after spinal fusion has been well described. Rates of reoperation after index fusion have ranged from 25% to 35% within a 10-year follow-up period.1,2 Because of these relatively high reoperation rates, ASP has been the driving force for the development of motion preservation technology. While it is clear that ASP does occur after fusion, what is less clear is the etiology. Many authors have suggested that ASP is a progression of natural degenerative disease. Others have suggested that ASP is largely iatrogenic and secondary to the higher biomechanical stresses at the adjacent level because of the fusion.
To assess whether ASP is a progression of natural arthritic disease in the spine, it is important that we improve our understanding of the natural history of spinal degenerative disease in the absence of a fusion. By establishing normative reference baseline rates of spinal degenerative disease, one can compare the rates of ASP to assess whether they are comparable. Presumably, if the rates of ASP after fusion are comparable with the rate of baseline spinal spondylosis, this would give credence to the hypothesis that ASP is a continuation of natural degenerative disease. Conversely, if the rate of ASP is significantly greater than the rate of natural spinal degenerative disease, this may suggest that there is another contributing factor.
The purpose of this systematic review was to address the following key questions: (1) What is the population risk of radiographical degeneration in the lumbar and cervical spines and does it vary by age? (2) Among patients who receive fusion for lumbar or cervical degeneration, what is the risk of radiographical adjacent segment pathology (RASP)? (3) Among patients who do not receive fusion for lumbar or cervical pathology, but who were eligible, what is the risk of developing additional pathology?
MATERIALS AND METHODS
We conducted a systematic search in PubMed and the Cochrane Collaboration Library for literature published between January 1970 and March 30, 2012. The search results were limited to human studies published in the English language. Reference lists of key articles were also systematically checked to identify additional eligible articles. For key question 1, we included longitudinal population-based studies with at least 2 time periods assessing radiographical degeneration of the lumbar or cervical spine. We excluded studies that selected samples on the basis of spine symptoms and cross-sectional studies (Table 1). However, using this criteria, we found that no studies provided information on age-specific risks of radiographical degeneration; thus, for the purpose of addressing this question only, we expanded our search criteria to include population-based cross-sectional studies that specifically addressed the relationship between age and radiographical degeneration in the lumbar or cervical spine. All other inclusion/exclusion criteria remained the same. For key questions 2 and 3, we included patients who either received fusion or were eligible for fusion but received nonoperative care while enrolled in a comparative study (randomized controlled trial [RCT] or cohort study) for a symptomatic degenerative lumbar or cervical spine. We excluded studies of fusion for tumor, trauma, infection, or inflammatory disease. We also excluded studies reporting on clinical ASP without evidence of radiographical degeneration. Full text of potential articles meeting the inclusion criteria were reviewed by 2 independent investigators (J.R.D., E.B.) to obtain the final collection of included studies (Figure 1).
From the included articles, the following data were extracted: study design, patient demographics, follow-up duration and the rate of follow-up, definition of radiographical spinal degeneration or RASP, and prevalence and/or incidence (see Appendix, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A693).
We report the cumulative incidence or the prevalence of radiographical degeneration. The cumulative incidence of radiographical degeneration was defined as the proportion of people who had been degeneration-free at the time of the initial evaluation (key question 1), or of patients RASP-free at the time of enrollment in an RCT involving spinal fusion (key questions 2 and 3), that had subsequent development of new pathology at final follow-up. Prevalence was defined for key question 1 as the proportion of people at any follow-up. For key question 2, prevalence was defined as the proportion of patients with RASP at follow-up given that some patients had RASP upon enrollment in an RCT involving spinal fusion. Data were summarized in tables and figures but were not pooled between the studies because of the limited number of studies available and the heterogeneity of patient populations, follow-up periods, and outcomes.
Study Quality and Overall Strength of Body of Literature
Level of evidence ratings were assigned to each article independently by 2 reviewers (J.R.D., E.B.) using criteria set by The Journal of Bone & Joint Surgery, American Volume3 for prognostic studies and modified to delineate criteria associated with methodological quality and described elsewhere4 (see Appendix, Supplemental Digital Content available at http://links.lww.com/BRS/A693 for individual study ratings). The overall body of evidence with respect to each clinical question was determined on the basis of precepts outlined by the Grades of Recommendation Assessment, Development and Evaluation working group5 and recommendations made by the Agency for Healthcare Research and Quality.6 Risk of bias was evaluated during the individual study evaluation described earlier. This system that derives a strength of evidence grade for each outcome or clinical question of “high,” “moderate,” “low,” or “insufficient” is described in further detail in the methods article for this focus issue.7
Consensus statements were made through a modified Delphi approach by applying the Grades of Recommendation Assessment, Development and Evaluation/Agency for Healthcare Research and Quality criteria that imparts a deliberate separation between the strength of the evidence (i.e., high, moderate, low, or insufficient) from the strength of the recommendation. When appropriate, recommendations of statements “for” or “against” were given. “Strong” or “weak” designations were based on the quality of evidence, the balance of benefits/harms, and values and patient preferences. In some instances, costs may have been considered. A more thorough description of this process can be found in the focus issue methods article.7
We identified a total of 15 studies from our search strategy that met the inclusion criteria, 8 addressing key question 1 and 7 addressing key questions 2 and/or 3 (Figure 1). For key question 1, our initial search produced 111 possible studies. We excluded 82 after abstract review, the majority of which were review articles, studies on the heritability/genetics of disc degeneration/back pain, or studies in specific populations (elite athletes, manual laborers, etc.). Among the 29 articles retrieved for full-text review, 23 were excluded, primarily because they were not population-based or they were cross-sectional studies. Following the decision to include cross-sectional studies reporting age-specific data, we re-examined our excluded articles from the first search and found 2 that met the new criteria and included them. We also conducted a new search that yielded 64 potential articles, none of which fulfilled the inclusion criteria. For key question 2, our initial search produced 124 total citations. We excluded 88 articles after abstract review. Among the 36 articles that underwent full-text review, 29 were excluded because there was no radiographical evaluation of ASP. A list of all excluded articles can be obtained in the supplemental information (see Appendix, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A693).
Overview of Studies Included for Key Question 1
Six prospective, longitudinal, population-based studies were identified, which provided information regarding the population rate of radiographical degeneration in the lumbar (4 studies)8–11 and cervical (2 studies)12,13 spines (Table 2). The included populations were all subsets of larger longitudinal studies investigating various diseases such as osteoarthritis (OA) and heart disease. Two of the lumbar studies were conducted in identical patient populations taken from the Framingham Heart Study but different criteria were used for defining radiographical disc degeneration. For the purposes of this study, these articles were treated as 1 study population.10,11 Two prospective, cross-sectional, population-based studies were identified, which provided information regarding the age-specific rate of radiographical degeneration in the lumbar spine.14,15 One study was conducted as an ancillary project to the Framingham Heart Study.15
Across the 3 longitudinal lumbar studies, a total of 1555 participants were analyzed, with females comprising 64.8% to 100% of the populations. Mean ages were similar across the studies, ranging from 53.8 to 54.7 years. Mean follow-up periods varied from 9 to 25 years. In 2 studies, the percentage of patients followed was 21.8% and 79.4% (reflecting the subset of patients included in the analysis from the entire available study population); 1 study did not report the percentage of patients followed.8 Radiographical lumbar spine degeneration was defined in various ways across the studies (see Table 2 for details).
Across the 2 longitudinal cervical studies, a total of 3212 participants were analyzed, all from the Clearwater Osteoarthritis Study. Only 1 study reported the sex and mean age of its population, with 66.8% women and a mean age of 66.8 years.12 Mean follow-up periods were similar between the studies, 5.6 and 5.8 years. The percentage of patients followed was reported by 1 study only and was 71.8% (reflecting the subset of patients included in the analysis from the entire available study population).13 Radiographical cervical disc degeneration was defined in the same way in both studies, as OA grades 2, 3, or 4 according the OA criteria of Kellgren.16
Across the 2 cross-sectional lumbar studies, a total of 1230 participants were analyzed. Only 1 study reported that the sex and mean age of its population, with 44% female and a mean age of 52.6 years.15 The mean follow-up periods and the percent of patients followed were not reported in either study. The age-specific prevalence of lumbar disc degeneration was reported by decade in both studies. Radiographical lumbar spine degeneration was defined in a variety of ways across the 2 studies.
Overview of Studies Included for Key Questions 2 and/or 3
Seven studies were identified, which provided information regarding the risk of RASP among patients who received fusion for lumbar (3 studies)17–19 and cervical (4 studies)20–23 degeneration (Table 3). Of the lumbar studies, there was 1 RCT and 2 retrospective cohorts with 278 patients who underwent fusion. Males comprised between 49.4% and 68.0% of the populations and mean ages ranged from 39 to 58 years. Mean follow-up periods ranged from 5.0 to 12.6 years. Only 1 study reported a follow-up rate that was 81.8%.17 RASP was defined in a variety of ways across the 3 studies (see Table 3 for details). Of the 4 cervical studies, there was 1 RCT and 3 prospective cohorts with a total of 310 patients who underwent fusion. Males comprised between 42.9% and 63.0% of the populations and mean ages ranged from 43.9 to 53.0 years. Mean follow-up periods ranged from 1.6 to 3.0 years. Only 1 study reported a follow-up rate which was 86.5%.20 RASP was defined in a variety of ways across the 4 studies (see Table 3 for details).
Only 1 RCT was found, which provided information regarding the rate of lumbar RASP in patients who were eligible but did not receive lumbar fusion.17 There were 34 patients who received nonoperative care; 55.9% were male and the mean age was 37 years. The mean follow-up period was 12.6 years and the follow-up rate was only 50.0%. No studies were found that looked at this risk in the cervical spine.
What Is the Population Risk of Radiographical Degeneration in the Lumbar and Cervical Spine and Does it Vary by Age?
Regardless of how degeneration was defined, the cumulative incidence of radiographical spine degeneration ranged from 12.7% to 51.5% across 3 studies8,10,11,13 during a 5- to 25-year period (Figure 2). In the lumbar spine, the highest incidence was seen for disc space narrowing (45.2%), followed by endplate sclerosis (23.3%), and spondylolisthesis (19.7% and 12.7%). One of the studies also reported the sex-specific incidence of lumbar spine degeneration during a 25-year period with males and females having similar incidences of endplate sclerosis but females showing a slightly greater incidence of disc space narrowing following 25-years of follow-up (Figure 3).11 Two studies also reported the prevalence of radiographical degeneration in the lumbar spine. In 1 study, the prevalence of any lumbar disc degeneration ranged from 0% to 14.4% at baseline and 20.1% to 59.6% at 25-year follow-up (Figure 4).10,11 Disc space narrowing was the most prevalent finding at both baseline and follow-up examinations, followed by endplate sclerosis and spondylolisthesis, respectively. The second study reported the prevalence of lumbar degeneration at the onset of the study only and was 67.6% for disc space narrowing and 91.3% for anterior vertebral osteophytes.9 However, this same study also reported the proportion of patients with progressive lumbar disc degeneration (an increase in grade in an affected vertebra at baseline) during the mean 9-year follow-up period with rates of progression of 34.7% for anterior vertebral osteophytes and 28.8% for disc space narrowing among their population (Figure 5).
In the cervical spine, Wilder et al13 reported a baseline prevalence of 21.7% among their population. In their follow-up study, the authors reported the progression of cervical disc degeneration during a mean follow-up period of 5.8 years.12 The proportion of individuals who had evidence of progression was 47.9% and was similar between males (49.4%) and females (47.2%). The overall rate of progression of cervical disc degeneration per 100 person-years was also similar for males and females (8.9% and 8.0%, respectively); however, the age and sex-specific rates of progression per 100 person-years reveal that between ages 40 and 60 years, cervical disc degeneration progresses at a greater rate in females than in males. However, between ages 60 and 79, males had a faster rate of cervical degeneration progression than females (Figure 6). Similar progression rates were seen between males and females during the eighth (and greater) decade of life.
Regarding age-specific risk, the prevalence of lumbar disc degeneration increased with age across both cross-sectional studies (Figure 7). In one study, the proportion of participants aged 18 to 29, 30 to 39, 40 to 49, and 50 years or older with any degeneration was 42%, 48%, 70%, and 88%, respectively, and those with moderate to severe degeneration was 28%, 31%, 35%, and 60%, respectively.14 The second study looked at the prevalence of 4 different features of lumbar degeneration across 4 age groups in their population.15 Evidence of disc space narrowing was seen in only 21% of participants who were less than 40 years of age but by the fourth decade the prevalence had more than doubled to 52% and continued to increase to 68% in the fifth decade and to 81% in those aged 60 years and older. The prevalence of facet joint OA followed a similar pattern among the age groups: 24%, 45%, 74%, and 84%, respectively. Degenerative spondylolisthesis was rarely seen in patients younger than 50 years (range: 0%–2%); between the ages of 50 and 59, the prevalence was 11%, increasing to 34% in those aged 60 and older. Spinal stenosis was most prevalent in those aged 60 years and older (12%) with younger participants at a lower risk (range: 1%–5%).
Among Patients Who Receive Fusion for Lumbar or Cervical Degeneration, What Is the Risk of RASP Over Time?
The definition of RASP varied across 7 studies17–23 (Table 3 and Figure 8). Regardless of how degeneration was defined, the cumulative incidence of RASP following fusion ranged from 6.3% to 44.4% during 1.6 to 12.6 years of mean follow-up. The risk of cervical RASP was higher than that of lumbar RASP despite the shorter mean follow-up periods.
Among Patients Who Do Not Receive Fusion for Lumbar or Cervical Degeneration, but Who Were Eligible, What Is the Risk of Further Degeneration Over Time?
The cumulative incidence of additional degeneration in the lumbar spine in patients who did not receive (but were eligible for) lumbar fusion was low during a mean follow-up of 12.6 years in the only comparative study of nonoperative care that reported this outcome (Figure 9).17 Both posterior disc height reduction (>2 SD over the mean of the posterior disc height reduction in the entire nonoperative group) and remaining mean disc height (<20% of anterior vertebral height) had cumulative incidences of only 5.9%. No incidences of worsening of the University of California Los Angeles score or totally reduced posterior disc height (0 mm) were seen in these patients during 12.6 years. However, it must be noted that only 50% of the nonoperative patients from this study were available for the 12-year follow-up. Although it is unknown how many of those unavailable for follow-up ended up having surgery because of additional degeneration, any cases requiring surgery would not be reflected in these data. As a result, the risk of additional degeneration in this population is likely underestimated.
The overall strength of evidence evaluating the population risk of radiographical degeneration in the lumbar and cervical spine is “low,” that is, we have low confidence in the estimate of the absolute risk and further research may change the estimate. Regarding age-specific risk, the overall strength of evidence that the prevalence increases with age is moderate, that is, we have moderate confidence that evidence reflects the true effect and further research may change the estimate. The overall strength of evidence evaluating the risk of RASP among patients who receive fusion for lumbar or cervical degeneration and among patients who do not receive fusion for lumbar or cervical degeneration, but who were eligible, is “low.”
The etiology for ASP has been suggested to be either a natural progression of arthritic disease or an iatrogenic biomechanical effect of the fusion. The purpose of this systematic review was to assess the rate of spinal degeneration in patients without fusion and compare it to the rate of ASP in patients with fusion. If found to be comparable, it may suggest that the etiology of ASP is a natural progression of cervical spondylosis. If the ASP rate was substantially greater than the rate of spinal degeneration in patients without fusion, this may suggest that another factor (such as biomechanical effects) may be contributing to ASP.
There are limitations to this systematic review. First, there is no consensus on definition or classification of spine degeneration. As a result, authors use different outcome measures for degeneration that make comparisons from study to study not possible. Second, the outcome we used was radiographical degeneration. Although clinically symptomatic degeneration is of more interest, we chose to use radiographical degeneration as our outcome because population studies examining the risk of symptomatic degenerative disease were largely lacking in the literature. In an effort to maintain similar outcomes of measure in each group, we opted to examine radiographical outcomes as these were more readily available within the literature. However, radiographical degeneration may not necessarily correlate with clinically symptomatic disease. One strength of this review includes the use of longitudinal population studies that estimate the incidence of degeneration.
We found that the population risk of radiographical spinal degeneration ranged from 12.7% to 51.5% during a 5- to 25-year follow-up period. Among those who had spine fusion for degenerative disease, the risk of RASP in the spine ranged from 6.3% to 44.4% during a 1.6- to 12.6-year follow-up period. Among individuals who did not receive fusion for lumbar or cervical degeneration, but who were eligible, the risk of degeneration in this population was 5.9% during 12 years. However, it must be noted that more than 50% of the study population were unavailable for follow-up. It is likely that some of those not returning for follow-up had subsequent surgical procedures for spine degeneration making these estimates suspect.
These data suggest that although the overall risk of RASP is comparable to the rate of spinal degeneration, the follow-up period ranges were different. Specifically the follow-up in the spinal degeneration group was substantially longer. This suggests that the rate of RASP may be greater than the rate of natural degeneration; however, this statement should be interpreted cautiously. As noted, there is substantial variation in the follow-up and the definition of what constituted ASP. As with many clinical questions, further investigation is needed to better answer our key questions. To that end, a clear, universally agreed upon definition of ASP needs to be established. Secondly, parity between radiographical measures between the 2 groups needs to be established. Assessing the rate of adjacent segment disc degeneration to the rate of de novo spondylolisthesis will likely not provide valuable comparative information. However, comparing the rate of de novo spondylolisthesis versus the rate of adjacent segment spondylolisthesis may provide improved clinical insight. Furthermore, controlling for other variables such as the region of the spine (cervical vs. lumbar) and other potential covariates (i.e., age, sex, comorbidity) is needed for an accurate assessment. Finally and most importantly, parity between clinical measures between the 2 groups needs to be established. It may be challenging to interpret a comparison between patients with ASP and radiculopathy versus patients with ASP and axial back pain.
ASP may occur at a higher rate than the rate of natural spinal pathology and suggests that another factor (such as biomechanical effect of fusion) may accelerate pathologic changes.
Strength of Statement: Weak
- Reported rates of ASP in patients with fusion seem to be higher than the rates of de novo degeneration in patients without fusion, suggesting that the fusion itself may have a contributory effect in the development of ASP.
- These studies have significant variation in follow-up, definitions of radiographical degeneration and other covariates.
- Clarification of ASP definitions and classification is needed to optimally compare different patient populations.
The authors are indebted to Ms. Nancy Holmes, RN, for her administrative assistance.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.spinejournal.org).
1. Ghiselli G, Wang JC, Bhatia NN, et al. Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg Am 2004;86-A:1497.
2. Hilibrand AS, Carlson GD, Palumbo MA, et al. Radiculopathy and myelopathy at segments adjacent to the site of a previous anterior cervical arthrodesis. J Bone Joint Surg Am 1999;81:519.
3. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1.
4. Norvell DC, Dettori JR, Fehlings MG, et al. Methodology for the systematic reviews on an evidence based approach for the management of chronic LBP. Spine (Phila Pa 1976) 2011;36:S10–8.
5. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.
6. West S, King V, Carey TS, et al. Systems to Rate the Strength of Scientific Evidence. Maryland: Agency for Healthcare Research and Quality; 2002.
7. Norvell DC, Dettori JR, Skelly AC, et al. Methodology for the systematic reviews on adjacent segment pathology. Spine 2012;37:S10–S17.
8. Aono K, Kobayashi T, Jimbo S, et al. Radiographic analysis of newly developed degenerative spondylolisthesis in a mean twelve-year prospective study. Spine (Phila Pa 1976) 2010;35:887.
9. Hassett G, Hart DJ, Manek NJ, et al. Risk factors for progression of lumbar spine disc degeneration: the Chingford Study. Arthritis Rheum 2003;48:3112.
10. Kauppila LI, Eustace S, Kiel DP, et al. Degenerative displacement of lumbar vertebrae. A 25-year follow-up study in Framingham. Spine (Phila Pa 1976) 1998;23:1868.
11. Kauppila LI, McAlindon T, Evans S, et al. Disc degeneration/back pain and calcification of the abdominal aorta. A 25-year follow-up study in Framingham. Spine (Phila Pa 1976) 1997;22:1642.
12. Wilder FV, Fahlman L, Donnelly R. Radiographic cervical spine osteoarthritis progression rates: a longitudinal assessment. Rheumatol Int 2011;31:45.
13. Wilder FV, Hall BJ, Barrett JP. Smoking and osteoarthritis: is there an association? The Clearwater Osteoarthritis Study. Osteoarthritis Cartilage 2003;11:29.
14. Cheung KM, Karppinen J, Chan D, et al. Prevalence and pattern of lumbar magnetic resonance imaging changes in a population study of one thousand forty-three individuals. Spine (Phila Pa 1976) 2009;34:934–40.
15. Kalichman L, Guermazi A, Li L, et al. Association between age, sex, BMI and CT-evaluated spinal degeneration
features. J Back Musculoskelet 2009;22:189.
16. Kellgren J H, Jeffrey M R, Ball J, et al. Atlas of Standard Radiographs. Vol II. The Epidemiology of Chronic Rheumatism. Oxford: Blackwell Scientific, 1963.
17. Ekman P, Moller H, Shalabi A, et al. A prospective randomised study on the long-term effect of lumbar fusion on adjacent disc degeneration. Eur Spine J 2009;18:1175.
18. Kanayama M, Hashimoto T, Shigenobu K, et al. Adjacent-segment morbidity after Graf ligamentoplasty compared with posterolateral lumbar fusion. J Neurosurg 2001;95:5–10.
19. Satoh I, Yonenobu K, Hosono N, et al. Indication of posterior lumbar interbody fusion for lumbar disc herniation. J Spinal Disord Tech 2006;19:104–8.
20. Coric D, Nunley PD, Guyer RD, et al. Prospective, randomized, multicenter study of cervical arthroplasty: 269 patients from the Kineflex C artificial disc investigational device exemption study with a minimum 2-year follow-up: clinical article. J Neurosurg Spine 2011;15:348–58.
21. Kim SW, Limson MA, Kim SB, et al. Comparison of radiographic changes after ACDF versus
Bryan disc arthroplasty in single and bi-level cases. Eur Spine J 2009;18:218–31.
22. Maldonado CV, Paz RD, Martin CB. Adjacent-level degeneration after cervical disc arthroplasty versus
fusion. Eur Spine J 2011;20(suppl 3):403.
23. Park SB, Jahng TA, Chung CK. Remodeling of adjacent spinal alignments following cervical arthroplasty and anterior discectomy and fusion. Eur Spine J 2012;21:322.