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 ASP, and information about postsurgical malalignment as a risk factor for cervical ASP (see Supplemental Digital Material, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A700).
Study Quality and Overall Strength of Body of Literature
Level of evidence ratings were assigned to each article independently by 2 reviewers (E.M.V., A.C.S.) using criteria set by The Journal of Bone & Joint Surgery, American Volume,5 for prognostic studies and modified to delineate criteria associated with methodological quality as described elsewhere6 (see Supplemental Digital Material, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A700 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 group7 and recommendations made by the Agency for Healthcare Research and Quality.8 Risk of bias was evaluated during the individual study evaluation described earlier in the section “Study Quality.” 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.6 The supplemental digital material contains the details of how we arrived at the strength of evidence for each key question (see Supplemental Digital Material, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A700).
Where the data were available, the risk (cumulative incidence) of ASP in studies of malalignment was reported. Data were summarized but were not pooled between studies due to the limited number of studies available and the heterogeneity of malalignment definitions, patient populations, and outcomes.
To evaluate the effect of malalignment as a risk factor for ASP, we calculated the unadjusted risk ratios (RRs) with corresponding confidence intervals and P values where data were available. RRs more than 1.0 indicated an increased risk of ASP associated with the factor, whereas those that were less than 1.0 indicated a decreased risk. If the 95% confidence interval included the value of 1.0, the association was deemed not statistically significant. All calculations were performed using Stata 9.0 (StataCorp LP, College Station, TX).9
Clinical Recommendations and Consensus Statements
Clinical recommendations or 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 impart a deliberate separation of the strength of the evidence (i.e., high, moderate, low, or insufficient) from the strength of the recommendation. Where 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.6
The search strategy yielded a total of 338 citations, of which 311 could be eliminated on the basis of the title or abstract; full text was retrieved for 27 studies; and of these there were 5 that provided data for risk of ASP in the context of malalignment after cervical spine surgery.10–14 All reported on RASP. No studies reported on CASP or additional intervention to address ASP. Each study used different definitions for malalignment, precluding any pooling of data. No study examined the effect of malalignment on ASP after posterior surgery. No study examined the effect of malalignment in the coronal plane on ASP after any surgery.
Does the Presence or Magnitude of Postsurgical Malalignment in the Coronal (Scoliosis) or Sagittal Plane (Kyphosis/Lordosis) Affect the Risk of Cervical ASP?
Table 2 summarizes characteristics of included studies. Study populations tended to be primarily male and most had undergone surgery for degenerative cervical disc disease. The reported follow-up times ranged from 1.5 to 23 years and the follow-up in these studies was either low or not calculable. All were retrospective and considered level III evidence. Additional information on included studies and their quality assessment is contained in the supplemental digital material (see Supplemental Digital Material, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A700).
The overall risk (i.e., cumulative incidence) of RASP ranged from 17% to 75% in 4 studies.10–13 The definition of ASP was primarily based on radiographical or MRI findings. Only 1 study attempted to include symptomatology as part of the ASP definition.11 Overall risk could not be calculated from the information provided regarding patients in the fifth study.14
Three studies provide data from which unadjusted RRs and 95% confidence intervals could be calculated.10,12,13 In 2 of the 3 studies, the risk of RASP associated with malalignment was approximately 2-fold with the third study showing a slightly lower risk, which failed to reach statistical significance (Figure 2). Details of these studies are described below.
Faldini et al10 defined lordotic (normal alignment) patients as those with postoperative sagittal segmental alignment (SSA) of more than 0°, and kyphotic as SSA 0° or less; in this study, 27% (n = 18/66) of persons with lordosis and 61% (n = 25/41) of those with kyphosis developed RASP. The SSA was measured by using the angle between a line parallel to the upper border of the vertebral body proximal to the disc space involved and a line parallel to the lower border of the underlying vertebral body for the fused vertebrae. A tendency to kyphosis in postoperative SSA was determined to correlate with the development of RASP (r = −0.428; P < 0.001), as did a tendency to kyphosis of SSA at last follow-up (r = −0.543; P < 0.001). It is not clear at what point postoperatively measurements were taken. The sagittal alignment of the cervical spine was calculated from the line parallel to the upper border of the C2 vertebral body and the line parallel to the lower border of the C7 vertebral body; it is also positive in case of lordotic alignment and negative when kyphotic. No correlation was demonstrated between the postoperative sagittal alignment of the cervical spine and the development of RASP, or between the sagittal alignment of the cervical spine at last follow-up and RASP. All data were obtained from radiographs and patients' files, without recalling any patients at the time of this study. There were 271 patients who underwent surgery during the study, but 128 were excluded due to other pathologies present, age more than 65 years at last follow-up, or incomplete radiographs with all vertebral bodies from C2–C7 clearly visible; a further 36 patients did not have a minimum follow-up of 10 years and thus were not available for analysis.
Katsuura et al12 divided patients into 4 groups based on the alignment of the cervical spine, those with lordotic (normal) alignment, and those with malalignment being described as straight, kyphotic, or sigmoid. In this study, 33% (n = 18/33) of those with lordosis developed RASP, compared with patients with straight alignment (100%; n = 3/3), kyphotic (88%; n = 7/8), or sigmoid alignment (50%; n = 2/4). These alignments for the cervical spine were determined from the overall shape and angles formed from tangent lines to the posterior edges of C2–C7. The SSA was determined from the angle formed by the upper plane and the lower plane of the fused segment. On the basis of multiple regression analysis, the preoperative alignment of the cervical spine was associated with a higher incidence of RASP (P < 0.0001), and the postoperative SSA could be a possible determinant (P = 0.0671). Data were available for 42 of 93 patients, who agreed to return to participate in the study, and who had a follow-up of more than 5 years from surgery. To calculate the RR, those with any alignment of the cervical spine other than lordosis were compared with those who had lordosis.
Kulkarni et al13 described their patients as those for whom alignment of the cervical spine was maintained, of which 67% developed RASP, versus those with kyphotic alteration, 88%. Kyphotic alteration was defined as a lordotic spine that became straight or kyphotic, or a straight spine that became kyphotic, as measured on lateral cervical radiographs obtained before surgery and at follow-up. The classification was determined by a line drawn from the posterior of the inferior endplate of C2 to C7; the spine was classified as lordotic if all vertebrae between these points were ventral to the line, as straight if any touched the line, and as kyphotic if 1 or more intersected the line. No statistical association between the risk of RASP and this classification of malalignment was seen.
The remaining 2 studies provided limited information regarding the risk of RASP related to postsurgical malalignment.
Ishihara et al11 (n = 112) compared the mean fusion alignment between those who were classified as having RASP and those who did not, with lordosis represented by a positive number of degrees and kyphosis as a negative number of degrees. Alignment was measured radiographically for the angle from C2 to C7 on a lateral radiograph taken in the standing position. Among those with ASP, the mean degree of fusion alignment (n = 19) was 1.27° ± 2.82°, and among those without ASP (n = 93), −1.00° ± 6.78°, and there was no statistically significant association (P = 0.217). The large SDs call the stability of these measurements into question.
Matsumoto et al14 reported MRI results related to degenerative disease for 64 individuals who had undergone anterior cervical discectomy and fusion (ACDF). They describe different categories of degenerative changes seen on MRI relative to “fusion in kyphosis” of 5° or more or less than 5°. There were smaller percentages of patients with decreased signal intensity, disc space narrowing, or foraminal stenosis among those with less kyphosis (<5°), but more patients in this group had posterior disc protrusion. However, it seems that individuals could have findings in more than 1 category. It was not possible to determine the proportion of individuals in each group that was considered to have RASP. The authors state that only the number of fused levels was significantly associated with development of RASP.
The overall strength of evidence on postsurgical sagittal malalignment as a risk factor for RASP after anterior procedures in the cervical spine was low, indicating that we have low confidence that the evidence reflects the true effect. Further research is likely to change our confidence in the effect estimate and it is likely to change the estimate of effect given that the evidence either is unavailable or does not permit a conclusion. No studies on CASP or the influences of coronal malalignment or posterior procedures on the development of ASP were identified, thus the strength of evidence was insufficient (Table 3).
Case Study 1
PW was 41 year old when she presented with significant right shoulder and lateral arm pain. On examination, she had 4+/5 power on shoulder abduction, hyper-reflexic, and a positive Hoffman sign on the right. The imaging demonstrated a disc extrusion at C4–C5 with slight right eccentricity (Figure 3). She underwent a C4–C5 ACDF to treat her pathology with good result and complete resolution of her symptoms when reviewed at follow-up by 6 weeks. In our hands, the ACDF procedure requires placement of the graft in the interbody space ventrally, to distract the space open and restore ventral height, and therefore the trend toward lordosis. Preoperatively her C2–C7 Cobb angle was 20° and at postoperative follow-up 3 years later, 18°. There was a significant trend to lordosis at the segmental level. The angle at the fusion level at C4–C5 changed from −12° to 3°. This in turn led to changes at C3–C4 from −3° to 16° and at C5–C6 from −2° to 3°. There was no evidence of RASP or CASP in follow-up to 3 years postsurgery.
Case Study 2
ZG was 44 year old when he presented with cervical myelopathy after a fall off a mountain bike months earlier. His imaging demonstrated spondylosis at C5–C6 and C6–C7 and evidence of degenerative disc disease at all levels in the cervical spine (Figure 4). The C2–C7 Cobb angle was 29°, at C2–C3 6°, at C3–C4 −5°, at C4–C5 −3°, and at C5–C7 18°. He underwent a C5–C6 and C6–C7 ACDF. Postoperatively his C2–C7 Cobb angle was 12°, at C2–C3 −3°, C3–C4 −10°, C4–C5 −9°, and the angle of the fusion, C5–C7, 14°. There was thus a reduction of 4° with the fusion at C5–C7. The spine was made more kyphotic at adjacent segments and overall alignment less lordotic. Four years later, he presented with neck pain and myelopathy shortly after a fall. Imaging demonstrated RASP to the previous C5–C7 ACDF at C4–C5. His C2–C7 Cobb angle was 6°, at C2–C3 −5°, at C3–C4 −13°, at C4–C5 −6°, and the angle of the fusion, as expected, remained at 14°. This case demonstrates the complexity of the management of cervical spondylosis. Although segmental alignment was relatively maintained, global alignment was not preserved. The adjacent segments thus became more kyphotic and it would seem that this contributed to the development of CASP. A C4–C5 ACDF was undertaken to address the ASP. The surgery paid particular attention to increasing the ventral height of the C4–C5 disc. After the revision surgery, the adjacent segments trended back toward lordosis. The Cobb angle for C2–C7 became 14°, C2–C3 −2°, C3–C4 −1°, and the angle of the fusion of C4–C7 13°. There was no evidence of ASP at 18 months postoperatively.
This article systematically reviews the available published evidence on the possible influence of alignment on the development of ASP. The overall strength of evidence to address this question with regard to sagittal malalignment was rated as low, with only 5 poor-quality retrospective studies (level of evidence III) articles found. Although the follow-up times are long in most studies, the poorly reported loss to follow-up in some studies and the low rate of follow-up reported in others leaves open the opportunity for substantial bias. Failure to evaluate and control for potentially confounding factors is another possible source of bias in these studies. The overall strength of evidence regarding the extent to which coronal malalignment may contribute to development of ASP is insufficient because no studies were identified.
Although the evidence is low, our results suggest that sagittal malalignment may contribute to development of RASP. No evidence was determined, however, for a role of coronal alignment in either RASP or CASP.
Cervical spondylosis is the most common progressive disease in the aging cervical spine.15 Adjacent-segment disease is a complication that has been described after surgical intervention for cervical spondylosis. Hilibrand et al4 published a series of 374 patients who demonstrated CASP with an annual incidence of 2.9% per year and 25.6% at 10 years. Recently, however, Wu et al16 reviewed 19,385 patients in a nationwide database that had ACDF and determined a yearly incidence of repeat ACDF surgery for CASP of 0.8%. At 10 years, 5.6% of the patients had undergone a second surgery for CASP. These annual and cumulative incidences point to the importance of reducing the risk of CASP development. The cumulative incidence of RASP from 17% to 75% across articles summarized in this systematic review underscores the importance of considering alignment and methods for decreasing the opportunity for malalignment. Unfortunately, no article addressed any role of malalignment in the development of ASP in patients with cervical disorders that have been treated with posterior surgery.
Spinal segmental alignment and overall cervical alignment were measured in different ways between the articles and this may affect results (see Demographic and Results Table in Supplemental Digital Material, Supplemental Digital Content 1, available at http://links.lww.com/BRS/A700). Faldini et al10 determined spinal segmental alignment by measuring the angle between a line drawn at the upper border of the vertebral body proximal to the disc space involved and the line parallel to the lower border of the body below. Overall alignment was measured by the same method of Katsuura et al,12 that is, by a line drawn parallel to the upper border of the body of C2 and the line parallel to the lower border of C7, the Cobb angle.17 To measure SSA, Katsuura et al12 used the posterior tangential method,18,19 by lines drawn parallel to the posterior vertebral bodies. SSA was determined by the angle formed by the upper plane and the lower plane of the fused segment. Kulkarni et al13 determined overall sagittal alignment by drawing a line between the posterior inferior edge of C2 and the similar point on C7. If the intervening bodies were anterior to this, then they were determined to be in lordosis, on the line straight and behind the line kyphotic. The validity of overall sagittal alignment measures has been investigated before in radiographs. Both Harrison et al18 and Ohara et al20 agreed that either the Cobb method or the posterior tangential method are robust valid measures of overall sagittal alignment. The standard error of measurement, however, is slightly less in the posterior tangential method. Thus, the measurements in the articles reviewed are reliable and not a source of significant bias. In future studies, it would seem prudent to use the tangential method, because it is slightly more accurate. The caveat to this is that these methods have not been compared in MRI and computed tomographic scans, where measurements may be more accurate due to the resolution of these imaging modalities.
The principle aim of surgery is to remove the pathology causing neural compression, although the method of intervention may differ. The amount of neural compression will be least in the anatomical neutral position of the cervical spine. The anatomical neutral position in the sagittal plane is one of slight lordosis, although in cervical spondylosis, where ventral disc height is reduced, the spine may trend toward kyphosis.15 It would seem reasonable then that any surgical intervention should attempt to address both the symptomatic process and return the spine toward the normal. The logical question is then: Does malalignment contribute to the development of ASP?
The strength of evidence to evaluate this question is low based primarily on methodological considerations. From a clinical perspective, there are some additional aspects to consider. Studies that provided sufficient data for some level of analysis evaluated RASP and none evaluated CASP or reported on the number of persons treated for symptomatic ASP. Matsumoto et al14 quantified the amount of adjacent level pathology on a point scale system that used MRI to evaluate the level of degeneration by looking for the amount of disc protrusion, disc space narrowing, decrease in signal intensity of the disc, and foraminal stenosis. Although the grading system determined that more degeneration was seen in those fused in a kyphotic angle greater than or equal to 5°, 1 parameter in their grading system (i.e., posterior disc protrusion) showed a lower prevalence in those with kyphosis (65% vs. 78%). In other words, those with a more lordotic alignment demonstrated a higher prevalence of posterior disc protrusions. This demonstrates the importance of addressing clinically relevant ASP because radiological findings do not necessarily manifest with clinically significant pathology. The extent to which radiological degeneration at adjacent segments might represent an intermediate (indirect) outcome and is or is not linked to patient symptoms or function might be a consideration in determining a potential downgrade for Grades of Recommendation Assessment, Development, and Evaluation.
Conceptually, it would be logical to attempt to restore normal alignment in surgical procedures. As outlined in case study 1, it is our practice to attempt to restore sagittal balance to the extent possible with anterior surgery. By distracting the ventral disc space, some lordosis can be restored due to the anatomy of the cervical spine.15 Case study 2 presents a case that demonstrates the complexity of the management of cervical spondylosis. Although segmental alignment was relatively maintained at the fusion site, the global alignment was not corrected. There was evidence of some degenerative changes in the adjacent disc. The patient subsequently developed ASP. This case shows that where global alignment cannot be addressed, the patient should be carefully observed with regard to the development of ASP where attention to adjacent alignment was not optimum. The fusion actually reduced the lordosis in the fused segment, thus apparently contributing to the development of adjacent pathology. The revision surgery attended to this and the patient has no evidence of adjacent pathology in follow-up.
Future studies providing high-quality evidence on this particular topic may be challenging to design. However, high-quality cohort analyses using multivariate statistical methods are required to verify any association between ASP and postoperative malalignment. This is essential for minimizing confounding variables in a condition with multifactorial etiology. These confounding variables may include age, sex, smoking history, preoperative alignment, or level of disease. Such studies should clearly define criteria for determining RASP and clear definitions for CASP. Evaluation of functional and patient-reported outcomes and the extent to which they are or are not associated with radiographical findings is also needed. Studies to determine the most accurate and reliable measures for cervical segmental and overall alignment are obviously necessary, because there seem to be no studies in the literature with regard to computed tomographic or MRI measurements. Implications of global spine balance would also need to be addressed; does the patient with lumbar sagittal imbalance have a higher likelihood of developing CASP in the cervical spine after an ACDF, for example?
Although the quality of available evidence is low, these studies do suggest that sagittal deformity is a risk factor in the development of ASP after reconstructive surgery for symptomatic cervical spondylosis.
It is recommended that an attempt be made to maintain or restore sagittal lordosis in surgical intervention for cervical disorders.
Level of Evidence: Low
Strength of Statement: Weak
- There is low-level evidence that postsurgical sagittal alignment increases the incidence of cervical RASP.
- There are no published studies investigating the role of postsurgical coronal alignment in the development of cervical ASP.
- The posterior tangential method seems to convey the most accurate measure of sagittal alignment in the cervical spine in radiographs, although other methods such as the Cobb angle are accurate when the spine is in lordosis.
- There are no published studies determining the best method of cervical alignment measurement in computed tomography or MRI.
The authors thank Ms. Nancy Holmes, RN, and Chi Lam, MS, for their administrative assistance. The author M.G.F. contributed to study concept, interpretation, manuscript preparation, and manuscript revision; M.A.H. and H.J.K.: interpretation, manuscript preparation, and manuscript revision; A.C.S.: data analysis and interpretation, manuscript preparation, and manuscript revision; E.M.V.: data analysis, manuscript preparation, and manuscript revision.
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. Carette S, Fehlings MG. Clinical practice. Cervical radiculopathy. N Engl J Med 2005;353:392–9.
2. Fehlings MG, Arvin B. Surgical management of cervical degenerative disease: the evidence related to indications, impact, and outcome. J Neurosurg Spine 2009;11:97–100.
3. Kepler CK, Hilibrand AS. Management of adjacent segment disease after cervical spinal fusion. Orthop Clin North Am 2012;43:53–62, viii.
4. 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;81:519–28.
5. Wright JG, Swiontkowski MF, Heckman JD. Introducing levels of evidence to the journal. J Bone Joint Surg Am 2003;85-A:1–3.
6. Norvell DC, Dettori JR, Skelly AC, et al. Methodology for the systematic reviews on adjacent segment pathology. Spine 2012;37:S10–7.
7. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.
8. 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): Agency for Healthcare Research and Quality. Rockville, MD; 2002.
9. StataCorp. Stata Statistical Software: Release 9th ed. College Station, TX: StataCorp LP; 2005.
10. Faldini C, Pagkrati S, Leonetti D, et al. Sagittal segmental alignment as predictor of adjacent-level degeneration after a Cloward procedure. Clin Orthop Relat Res 2011;469:674–81.
11. Ishihara H, Kanamori M, Kawaguchi Y, et al. Adjacent segment disease after anterior cervical interbody fusion. Spine J 2004;4:624–8.
12. Katsuura A, Hukuda S, Saruhashi Y, et al. Kyphotic malalignment after anterior cervical fusion is one of the factors promoting the degenerative process in adjacent intervertebral levels. Eur Spine J 2001;10:320–4.
13. Kulkarni V, Rajshekhar V, Raghuram L. Accelerated spondylotic changes adjacent to the fused segment following central cervical corpectomy: magnetic resonance imaging study evidence. J Neurosurg 2004;100(suppl Spine):2–6.
14. Matsumoto M, Okada E, Ichihara D, et al. Anterior cervical decompression and fusion accelerates adjacent segment degeneration: comparison with asymptomatic volunteers in a ten-year magnetic resonance imaging follow-up study. Spine (Phila Pa 1976) 2010;35:36–43.
15. Shedid D, Benzel EC. Cervical spondylosis
anatomy: pathophysiology and biomechanics. Neurosurgery 2007;60:S7–13.
16. Wu JC, Liu L, Wen-Cheng H, et al. The incidence of adjacent segment disease requiring surgery after anterior cervical diskectomy and fusion: estimation using an 11-year comprehensive nationwide database in Taiwan. Neurosurgery 2012;70:594–601.
17. Cobb JR. Outlines for the study of scoliosis. In: Edwards JW ed. Instructional Course Lecture. Ann Arbor, MI: American Academy of Orthopaedic Surgeons; 1948:261–75.
18. Harrison DE, Harrison DD, Cailliet R, et al. Cobb method or Harrison posterior tangent method: which to choose for lateral cervical radiographic analysis. Spine (Phila Pa 1976) 2000;25:2072–8.
19. Harrison DD. The Physics of Spinal Correction. Washington DC: National Library of Medicine; publication number WE 725 H318C; 1986:319.
20. Ohara A, Miyamoto K, Naganawa T, et al. Reliabilities of and correlations among five standard methods of assessing the sagittal alignment
of the cervical spine. Spine (Phila Pa 1976) 2006;31:2585–91; discussion 92.
cervical spondylosis; degenerative disc disease; spine surgery; sagittal alignment
Supplemental Digital Content
© 2012 Lippincott Williams & Wilkins, Inc.