Anterior cervical plating, total disc arthroplasty, and noninstrumented fusion (NIF) have been used in the treatment of cervical disc disease. With these procedures, adjacent-segment pathology (ASP) in the cervical spine has been described. Distinct from degenerative forms of ASP, adjacent-level ossification development (ALOD) is a type of ASP that results in adjacent-level marginal osteophyte formation. This results in heterotopic bone formation along the anterior longitudinal ligament at the adjacent levels. Few studies have compared these 3 surgical procedures to evaluate the risk, timing, and outcomes related to ALOD.
With anterior cervical plating, factors such as the length of the plate and the plate-to-disc distance (PDD) have been proposed to affect the risk of ALOD. Some have advocated using the shortest plate possible, to minimize intrusion on the adjacent segment and to minimize the risk of ALOD.1,2 This is distinct from other types of ASP such as adjacent-segment degeneration where osteophyte formation, disc space degeneration, and endplate sclerosis are the components to the pathology.
The purpose of this systematic review was to examine the following key questions regarding the development of ALOD by comparing 3 different techniques used to treat cervical degenerative disc diseases:
- What is the risk of ALOD in patients receiving noninstrumented cervical fusion, instrumented cervical fusion with a plate, or cervical total disc arthroplasty?
- What are the risk factors for ALOD?
- What is the time course for the development of ALOD?
- Does ALOD affect outcomes and rates of reoperation?
MATERIALS AND METHODS
Electronic Literature Search
A systematic search was conducted for articles published between January 1990 and December 2011. We used 2 search strategies to identify articles from PubMed, Embase, and the Cochrane Library. First, we used key words to detect articles that used the term “adjacent level ossification” combined with cervical spine fusion (instrumented or noninstrumented) or total disc arthroplasty. Next, we searched for all articles that reported heterotopic ossification (HO) in the same 3 treatments. Among the articles describing HO, we evaluated the full text to determine if there were any cases of HO specifically identified as occurring at an adjacent segment. We limited our results to humans and to articles published in the English language. We included articles with adult cervical patients who had spondylosis, radiculopathy, or disc herniation; were treated with anterior cervical plates, total disc replacement (TDR), or NIF; and had the risk of or risk factors for ALOD reported. Articles were excluded if patients were (<18 yr of age), or treated for tumor or trauma. Other exclusions included posterior fusions, or case reports or studies with fewer than 10 subjects, as described in Table 1. Full text of potential articles meeting the inclusion criteria by both methods were reviewed by 2 independent investigators (J.R.D, K.D.R.) to obtain the final collection of included studies (Figure 1).
From the included articles, the following data were extracted: patient demographics, inclusion and exclusion criteria, follow-up duration and the rate of follow-up, outcomes assessed, risks (%) of ALOD, and potential risk factors (see Web Appendix, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A692).
Study Quality and Overall Strength of Body of Literature
Level-of-evidence ratings were assigned to each article independently by 2 reviewers (C.E., J.R.D.) using criteria set by the Journal of Bone & Joint Surgery, American Volume,3 for prognostic studies and modified to delineate criteria associated with methodological quality and described elsewhere4 (see supplemental digital material for individual study ratings, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A692). 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 as described earlier. This system, which 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 The supplemental digital material (Supplemental Digital Content 1, available at http://links.lww.com/BRS/A692) contains the details of how we arrived at the strength of evidence for each key question.
Where the data were available, we reported the cumulative incidence or the prevalence of ALOD using the number of events (adjacent levels with ossification) in the numerator and the number of adjacent levels at risk in the denominator. For studies only providing the patient as the denominator, the proportion of patients with 1 or more levels of ossification was considered a count of 1 in the numerator, and recorded and noted in the text and appropriate tables. The cumulative incidence (%) of ALOD was defined as the proportion of patients who had been ALOD-free at the time of the index fusion and had subsequent development of ALOD at final follow-up. Prevalence was defined as the proportion of patients with ALOD at follow-up from among all patients in the study, some of whom had ALOD at the time of the index procedure. Data were summarized in tables but were not pooled between studies because of the limited number of studies available and the heterogeneity of patient populations and outcomes.
To evaluate the effect of a treatment type (plate vs. total disc arthroplasty) on the risk of ALOD in a randomized trial, we calculated the risk difference, 95% confidence interval (CI), and the number needed to harm. The number needed to harm represents the number of patients one would need to treat with the intervention (anterior plate) to cause 1 case of ALOD compared with the control (total disc arthroplasty). To evaluate the effect of specific risk factors on ALOD, we calculated the unadjusted risk ratios with corresponding CIs and P values. All calculations were performed using Stata 9.0 (Agency for Healthcare Research and Quality, Rockville, MD).8
Recommendations 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 impart 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 or statements “for” or “against” were given “strong” or “weak” designations based on the quality of the 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 442 total citations from our literature search. Of these, 424 were excluded by the title/abstract and 18 full-text articles were evaluated to determine if they met the inclusion criteria. From these 18 studies, 13 were excluded for the following reasons: HO was reported but the location of ossification was not specified to the adjacent segment (n = 5); HO was reported at the same (treated) level (n = 7) but not at the adjacent level; and no ossification was reported (n = 1).
Details of the excluded articles can be found in the supplemental digital material (Supplemental Digital Content 1, available at http://links.lww.com/BRS/A692). The remaining 5 studies9–13 met our inclusion criteria and are summarized in this report (Figure 1). We did not identify any studies comparing plates with NIF nor TDR with NIF. However, we identified 1 randomized controlled trial (RCT) comparing anterior plate fusion versus TDR,9 1 single-arm study that looked for ALOD in NIF,13 and 2 single-arm studies that looked for ALOD after fusion using anterior plates.10–12 The RCT used a single center's data from the pivotal Food and Drug Administration investigational device exemption multicenter trial comparing the Bryan artificial disc (Medtronic, Memphis, TN) with anterior cervical discectomy and fusion (ACDF). Study characteristics can be found in Table 2.
Four studies reported risks of ALOD after 2 years: 1 RCT using the patient as the unit of measure evaluating ACDF with anterior plates compared with TDR9; 2 single-arm studies evaluating only ACDF with anterior plating using adjacent levels as the unit of measure10,12; and 1 single-arm study assessed NIF, also using adjacent levels as the unit of measure.13 Among the 3 studies evaluating ACDF using anterior plates, the risk of ALOD ranged from 64% (the proportion in patients) to 41% (proportion of levels) (Figure 2). The risk of ALOD after TDR in 1 study was 24% (proportion in patients)9 and after NIF in 1 study, 6% (proportion of levels).13
One RCT used a single site's data from a multicenter Investigational Device Exemption Food and Drug Administration trial to determine whether ALOD occurs more frequently in patients undergoing fusion with plate fixation than it does in patients undergoing total disc arthroplasty.9 Pre- and postoperative radiographs were read by 2 independent spine fellows blinded to the treatment. The surgical level was blacked-out through photo editing, making blind assessment possible. ALOD was rated using the following classification system10: Grade 1, ossification occupying less than 50% of the adjacent disc space height; Grade 2, ossification extending beyond 50% of the adjacent disc space height; and Grade 3, bridging of 100% of the adjacent disc space with bone (Figure 3). The fusion group had significantly higher ossification grades than the arthroplasty group at both the 2-year (P = 0.003) and the 4-year follow-up interval (P = 0.004). The proportion of patients who developed any ALOD was greater after anterior plate fusion, 64% versus 24%, respectively (risk difference, 40%; 95% CI, 14%–66%; P = 0.006; number needed to harm, 2.5).
Two studies assessing fusion with anterior plates reported on the risk of ALOD based on whether the ossification developed cranially or caudally to the fused segment. Both reported a 2-fold higher risk of ALOD in the cranial adjacent segment compared with the caudal (59% vs. 29%11 and 58% vs. 28%12) (Table 3). The single study on NIF reported 4.8% cranial ossification compared with 1.2% caudal.13
Risk Factors for ALOD
Two studies reported on the relationship between PDD and risk of ALOD,9,10 and 1 study evaluated the progression of ALOD stratified by PDD.11 Park et al10 in 2005, reviewed 118 patients who had solid fusion after anterior fusion with a plate for a degenerative cervical condition. Follow-up was for a minimum of 1 year. Eighteen caudal adjacent-level segments were not included in the analysis because of inadequate visualization on radiographs. They determined a higher risk of any ALOD when the plate was less than 5 mm away from the disc compared with greater than 5 mm: 58.6% of the levels versus 11.5% (risk ratio, 5.1; 95% CI, 2.5–10.4). When the investigators looked at the moderate to severe cases of ossification (grade 2 or 3), the risk in the group with a PDD less than 5 mm was 16.6% of the levels compared with 5.0% in the group with a PDD of 5 mm or more (risk ratio 3.3; 95% CI, 0.8–13.4; P = 0.076). Similarly, the post hoc analysis by Garrido et al9 described earlier reported a higher risk of any ALOD when the PDD was less than 5 mm versus 5 mm or more: 95% and 50% of patients, respectively. Park et al,11 in 2007, studied 62 patients, most of whom had developed ALOD after anterior fusion and plating, and retrospectively evaluated the progression of the ossification over time. They determined that progression from no or mild ossification (grade 0 or 1) in adjacent levels by 3, 6, or 12 months to moderate or severe ossification (grade 2 or 3) was 2 to 2.7 times more likely in patients with a PDD less than 5 mm compared with patients with a PDD of 5 mm or more (Figure 4).
Single-Level Anterior Cervical Corpectomy and Fusion Versus 2-Level ACDF
Park et al,12 in 2010, reviewed the lateral cervical radiographs of 97 patients who had a solid fusion after a single-level anterior cervical corpectomy and fusion (ACCF) or 2-level ACDF for the treatment of cervical spondylosis. All patients underwent anterior fusion using allograft and screw-plate fixation. The choice of operation was based on the presence of retrovertebral compression. A corpectomy was used if the decompression could be performed more easily using that procedure. The risk of ALOD was higher in the single-level ACCF group compared with the 2-level ACDF group at final follow-up (24 mo), 50% of the levels versus 33% (risk ratio, 1.5; 95% CI, 0.99–2.27; P = 0.051).
Time Course for the Development of ALOD
Three studies reported risks of ALOD over time, and 1 study evaluated the progression of ALOD over time (Figure 5). One study observed an increased risk in the prevalence for ALOD from 64% of patients at 2-year follow-up to 84% at 4-year follow-up when performing fusion with plates.9 The same study also observed an increased prevalence for ALOD using a TDR, with the prevalence increasing from 25% of patients at 2 years to 52% at 4 years. Park et al,12 in 2010, noted that 5% of the levels developed adjacent-level ossification as early as 6 weeks after anterior plate fusion that rose to 41% of the levels at the 24-month follow-up. The third study in patients undergoing NIF reported a small risk of ossification that remained relatively constant at the 6-, 12-, and 24-month follow-ups.13
In the Park et al11 2007 study described earlier, the authors noted that the absence of ossification in the early postoperative period was no guarantee of avoiding advanced ossification later. They reported that 23.5% and 14.9% of adjacent levels in patients without ossification 3 and 6 months after anterior fusion and plating progressed to moderate or severe (grade 2 or 3) ALOD by the 24-month follow-up. Among patients without ossification at the 12-month follow-up, only 1.8% of the adjacent levels went on to develop moderate or severe ALOD. Among patients with grade 0 or 1 ossification 24 months after surgery, none of the 80 levels progressed to grade 2 or 3 ossification.
ALOD and Outcomes
We determined no clinical studies assessing outcomes or rates of reoperation comparing patients with ALOD with patients without ALOD.
The overall strength of evidence evaluating the absolute risk of ALOD after anterior fusion with a plate or total disc arthroplasty is “low,” that is, we have low confidence in the absolute estimate and further research may change the estimate. The overall strength of evidence evaluating the absolute risk of ALOD after NIF is “insufficient,” that is, the evidence does not permit a conclusion. The overall strength of evidence evaluating the PDD as a risk factor for ALOD is “moderate,” meaning that we have moderate confidence that the evidence reflects the true effect, although further research may change the estimate. The overall strength of evidence for key question 3 evaluating the timing of ALOD and key question 4 evaluating the outcomes and reoperation rates of ALOD are “low” (low confidence in the evidence reflecting the true effect) and “insufficient” (the evidence is not available), respectively (Table 4).
On the basis of our systematic review, anterior cervical fusion with a plate and decreased PDD (<5 mm) demonstrated higher rates of ALOD. Although the etiology of ALOD is not known, anterior cervical plates and specific plating technique seem to play a role in its development. This is evidenced by higher rates of ALOD seen in plated ACDF compared with noninstrumented ACDF and total disc arthroplasty.9,10,12,13 In addition, when plates are applied less than 5 mm from the adjacent levels, the risk for the development of ALOD was 2 to 5 times higher.9,10 Thus, in theory, ALOD may be caused by an inflammatory reaction between the plate and anterior longitudinal ligament leading to ossification.
Although both ALOD and ASP can result in bone formation, there are differences between the 2. It has been our experience that patients with ALOD typically have better preservation of disc height compared with patients with ASP. In addition, patients with ASP are more likely to have disc protrusions, facet arthrosis, spondylolisthesis, endplate sclerosis, and/or osteophyte formation. Because ALOD in its severest stages causes ankylosis and a decrease in motion, the result is a “relative protection” of the adjacent level from degenerative changes. Also, ALOD forms nonmarginal bone, caused by ossification of the anterior longitudinal ligament, as opposed to degeneration of the adjacent motion segment where you get marginal osteophyte formation. Finally, it is not clear whether ASP is due to surgery or the natural history of degenerative disease. ALOD, on the contrary, seems to be an iatrogenic problem that has a high likelihood of developing if the plate is placed too close to the adjacent segment (Figure 6).
There is insufficient evidence to suggest a role for surgical treatment of ALOD and no study has examined or verified inferior outcomes with its development. Although this may question the clinical significance of ALOD, current outcomes-based questionnaires are not designed specifically for assessing the consequences of ALOD. Range-of-motion analysis is necessary to define the effect of ALOD on adjacent-segment motion. Thus, meticulous surgical technique, with care to minimize adjacent-level encroachment by a plate, should be used.
On the basis of moderate evidence, we determined that plates increase the risk of ALOD and the risk is increased if the plate is placed less than 5 mm away from the adjacent disc. Therefore, we recommend that the surgeons make every effort to keep the plate as far away from the adjacent disc as possible.
Strength of Statement: Strong
- Instrumentation with an anterior plate increases the risk of ALOD compared with NIFs of the cervical spine.
- A PDD of more than 5 mm is recommended to minimize the risk for the development of ALOD.
- ALOD occurs more commonly in ACDF compared with TDR.
The authors thank 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. Goffin J, van Loon J, Van Calenbergh F, et al. Long-term results after anterior cervical fusion and osteosynthetic stabilization for fractures and/or dislocations of the cervical spine. J Spinal Disord 1995;8:500–8; discussion 499.
2. Mähring M. Segmental changes in the cervical spine following cervical spondylodeses of unstable injuries. Unfallchirurgie 1988;14:247–58.
3. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1–3.
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 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. Evidence Report/Technology Assessment No. 47. Prepared by the Research Triangle Institute-University of North Carolina Evidence-based Practice Center, Contract No. 290-97-0011. Rockville, MD: 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:S10–7.
8. StataCorp. Stata Statistical Software: Release 9. College Station, TX: StataCorp LP; 2005.
9. Garrido BJ, Wilhite J, Nakano M, et al. Adjacent-level cervical ossification after Bryan cervical disc arthroplasty compared with anterior cervical discectomy and fusion. J Bone Joint Surg Am 2011;93:1185–9.
10. Park JB, Cho YS, Riew KD. Development of adjacent-level ossification in patients with an anterior cervical plate
. J Bone Joint Surg Am 2005;87:558–63.
11. Park JB, Watthanaaphisit T, Riew KD. Timing of development of adjacent-level ossification after anterior cervical arthrodesis with plates. Spine J 2007;7:633–6.
12. Park Y, Maeda T, Cho W, et al. Comparison of anterior cervical fusion after two-level discectomy or single-level corpectomy: sagittal alignment, cervical lordosis, graft collapse, and adjacent-level ossification. Spine J 2010;10:193–9.
13. Yang JY, Song HS, Lee M, et al. Adjacent level ossification development after anterior cervical fusion without plate
fixation. Spine (Phila Pa 1976) 2009;34:30–3.