February 01, 2014
15 min read

Understanding of adjacent-segment pathology increasing, but questions still remain

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Questions still exist about adjacent-segment pathology and the role that natural degeneration and any effects previous surgery play in the condition. Publications that defined adjacent-segment pathology and data registries currently in development are increasing the understanding that neurosurgeons and orthopedic surgeons have of the phenomenon.

“The controversy isn’t so much whether this happens or not, but how much of it is just the natural history and how much of it is the surgery itself?” Alan S. Hilibrand, MD, Joseph and Marie Field Professor of Spinal Surgery, Director of Orthopaedic Medical Education and Professor of Neurological Surgery at Jefferson Medical College/The Rothman Institute, in Philadelphia, said.

“Degenerative changes are going to occur as we get older, no matter what we do to our bodies. Some people say there is no such thing as adjacent-segment pathology because it is just the natural history of aging. It is possible that this does entirely explain this phenomenon. However, it is certainly easier to identify adjacent-segment pathology in people who have had previous surgery, just because we know who those people are,” Hilibrand told Spine Surgery Today.

It is rare for clinical adjacent-segment pathology to occur, according to Alan S. Hilibrand, MD, but he said it typically requires surgical treatment.

It is rare for clinical adjacent-segment pathology
to occur, according to Alan S. Hilibrand, MD,
but he said it typically requires surgical treatment.

Image: Daniel Burke Photography

Until recently, a universally accepted classification system that addresses clinical and radiological issues for adjacent-segment disease and adjacent-segment degeneration did not exist. In 2012, Michael G. Fehlings, MD, PhD, FRCSC, FACS, and colleagues conducted a systematic review to determine how these terms are classified and defined in the peer-reviewed literature. They stated that adjacent-segment disease and adjacent-segment degeneration refer to degenerative changes that occur after reconstructive spine surgery, typically at an immediately adjacent functional spinal unit. The changes can include disc degeneration, instability, spinal stenosis, facet degeneration and deformity. However, based on their review, no formal classification system existed for cervical or thoracolumbar adjacent-segment disorders.

Because of these findings, in a 2012 special issue of Spine, Anderson and colleagues proposed “adjacent-segment pathology” as the general term to describe the changes that occur adjacent to a previously operated level. Under this umbrella term, radiographic adjacent-segment pathology (ASP) includes degenerative arthritic changes on imaging. When the radiographic changes then correspond to clinical symptoms, such as pain, difficulty walking or nerve pain and radiculopathy, it is known as clinical ASP.

Incidence rates

The landmark 1999 study by Hilibrand and colleagues of 374 consecutive patients followed for 21 years after anterior cervical arthrodesis for treatment of cervical spondylosis revealed a 2.9% yearly incidence rate of ASP within the first 10 years after surgery. They predicted 25.6% of patients would have ASP within 10 years after the initial surgery.

Lawrence and colleagues conducted a systematic review on ASP following cervical fusion and found a prevalence of symptomatic ASP of 11% to 12% at 5 years, 16% to 38% at 10 years and 33% at 15 years. They estimated the mean rate of symptomatic degeneration development in the cervical spine after arthrodesis at 1.6% to 4.2% per year. The mean rate of reoperation was 0.8%. These authors also conducted a systematic review of ASP after lumbar fusion and found an annual incidence of 0.6% to 3.9%.


The reoperation rates vary based on which part of the spine is involved in the ASP, the prior operation and the surgeon, according to Hilibrand. Data from the Spine Patient Outcomes Research Trial (SPORT), which is not surgeon- or institute-specific, revealed that at 4 years, the overall reoperation rate—not just for ASP—was 13% for patients who underwent surgery for spinal stenosis and 15% for patients who underwent surgery to treat degenerative spondylolisthesis, he said.

“If you think about it on a yearly basis, it comes out to be about 3% of patients per year,” Hilibrand said.


The debate in ASP is whether it is caused by degenerative changes and natural history or if previous surgery plays a role.

James D. Kang

James D. Kang

“The truth is there are biomechanical alterations that occur due to fusion and patients already have some predisposition to degeneration or spondylosis, so it is probably somewhere in the middle,” James D. Kang, MD, professor of orthopaedic and neurological surgery, UPMC endowed chair in spine surgery, vice chairman of the Department of Orthopaedic Surgery and director of Ferguson Laboratory for Spine Research at the University of Pittsburgh School of Medicine, said.

In fact, patients with asymptomatic changes on imaging who do not undergo surgery often show ASP changes on imaging years later.

“There are people who have a predisposition to degenerative disc disease,” Pierce D. Nunley, MD, director of the Spine Institute of Louisiana, said. “I have had many patients over the years who have had disc disease, and although they did not have surgery, they have come back 5 or 10 years later with an adjacent-segment problem.”

Joseph S. Cheng, MD, MS, associate professor of neurological surgery and director of the Neurosurgery Spine Program at Vanderbilt University Medical Center, in Nashville, Tenn., said that he has had patients with lumbar spondylosis who do not have surgery, yet they develop adjacent-segment degeneration at the levels just from progressive changes.

“Because we haven’t taken them to surgery, you can’t blame the surgical fixation, and so one of the problems with this concept of ASP is there is an underlying natural history of arthritic progression of spondylosis around the area. From a biomechanical standpoint on top of that, we know that how we limit the motion of one joint will affect the areas around it with increasing stress and strain,” he said.

Patients born with congenital fusion, or Klippel-Feil syndrome, are also at risk for ASP, according to Fehlings, professor of neurosurgery, Gerald and Tootsie Halbert chair in neural repair and regeneration, vice chair of the Department of Surgery, and co-chairman of the Spinal Program at University of Toronto. “The adjacent-segment pathology also can occur as part of the natural body response to spondylosis in the spine, which has nothing to do with the fact that there is increased mechanical stress on that segment,” he said. “It may just reflect another natural history of the patient.”

Changes in biomechanics

Changes in spinal biomechanics caused by surgery are believed to be one cause of ASP.

“Surgery could be a decompression that causes instability at the next level. It could be a fusion, which is the most common cause and what we see most frequently; and it can even be a disc replacement, because even with a disc replacement, the biomechanics of the operated motion segment are changed,” Hilibrand said.

Anything that affects motion across a joint and therefore causes additional stress and strain or increased load bearing of the surrounding joints can lead to ASP, Cheng said.

Joseph S. Cheng

Joseph S. Cheng

“It can be anything from fusing that area to even disc collapse from natural degeneration,” he said.

In the case of total disc arthroplasty (TDA), if the arthroplasty does not fit the motion properly or reflect the necessary load bearing across the segment, it can lead to ASP.

“I think that is the argument between the fixed center rotation devices vs. the variable translation devices for arthroplasty,” Cheng said.

In vertebroplasty, if surgeons significantly strengthen one bone segment, it no longer absorbs the bone even though it is considered ceramic material. “If you are solidifying that bone with cement, then any forces that go through that are immediately translated through without any absorption of the energy and so, therefore, the ceramic segments of bone carry a higher risk for fracture,” Cheng said.


Lawrence and colleagues identified several factors that increased the risk for ASP in the cervical spine: age younger than 60 years, fusing adjacent to C5-6 and/or C6-7, preexisting disc herniation, and/or dural compression secondary to spinal stenosis with a mean anteroposterior diameter spinal canal of 13 mm or less.

Fehlings noted the presence of kyphosis may also increase the risk for ASP in the cervical spine.

“This would be due to the anterior cervical plate as part of the cervical reconstruction procedure,” he said. “If this plate is too close to the disc level above, it could predispose the disc above to degeneration.”

In the lumbar spine, Lawrence’s group found the following factors increased the risk for developing ASP: age older than 60 years, preexisting facet degeneration, degenerative disc disease, performing a multilevel fusion, stopping a construct at L5, performing a laminectomy adjacent to a fusion, and excessive disc height distraction during posterior interbody fusion.

Treatment options

Radiographic ASP is common and rarely requires surgery.

“Clinical ASP is pretty uncommon,” Hilibrand said. “I believe it is less than 3% per year in my lumbar spine patients and definitely less than 3% per year in my cervical spine patients. But when it occurs in the lumbar spine, it usually ends up requiring surgery.”

Nunley suggested surgeons first try to treat ASP conservatively. “If you have done a fusion and the patient has broken down the first level, you will just start knocking the dominoes down,” he said.

“About half of patients with adjacent-segment pathology won’t need any treatment. They will respond to nonoperative treatment,” said Fehlings, who is a member of the Spine Surgery Today Editorial Board.

If a patient requires surgery, the most standard way to treat symptomatic ASP is to extend the fusion a level above. “The results generally are excellent,” he said.

“If it comes to surgery, I am a fan of motion-sparing technology, when it is indicated,” Nunley said. “If the facets are problematic or the patient has spondylolisthesis or instability, for example, you certainly don’t want to put a disc replacement there. Unfortunately, I think the majority of adjacent-segment pathology, because of the kind of pathology that is present and the severity at the time it comes to surgery, requires a fusion.”

When surgically treating ASP, Cheng said it is necessary to understand the spine biomechanics.

“I don’t believe in motion preservation,” he told Spine Surgery Today. “I believe in load preservation, and so I treat it by going back to the biomechanics of the spine. If all I do is extend the fusion up without taking into consideration the alignment and center of gravity and how that segment is going to be loaded, then there is no wonder the next level will be the next one to break down because I haven’t fixed the problem.”

Outcomes after surgery for ASP are generally not as good as after the initial surgery.

Michael G. Fehlings

Michael G. Fehlings

“The outcomes are probably 10% to 20% less in terms of global total patient satisfaction and outcomes,” Kang said. “But our goal is, of course, to treat the neurologic symptoms and prevent further deterioration or damage. In that regard, the surgery is often successful.”

Preventing ASP

ASP can often not be prevented because it is part of the natural history of degeneration. However, to reduce the occurrence after initial surgery, “the surgeon should try to reconstruct the spine to maintain balance,” Fehlings said. “That means in the cervical spine that you need to have a degree of cervical lordosis and you also want to avoid injuring the adjacent-level discs in the cervical spine by avoiding placing it too close to the disc above or below.”

Surgeons should provide the most anatomical type of stabilization in the initial surgery, according to Hilibrand.

“If you are going to fuse the spine, try to fuse it in the way that it was built,” he said. “Don’t take away or add to the lordosis; don’t change the overall spinal balance.”


In the lumbar spine, Nunley cautioned that when using percutaneous screws, surgeons should try to reduce the trauma to the next level.

“We’re looking at minimally invasive techniques and how they can reduce trauma to an adjacent level,” he said. “Often, I’ll use smaller screws at the top — what is called a ‘soft landing’ — where you have a little flexibility right at the top of the curve to maybe make it not so stiff. All sorts of techniques are coming out and people are collecting data, but unfortunately, none of these techniques have been proven to decrease the chances of symptomatic ASP. But we are working on it.”

In addition, surgeons should address all pathology, Hilibrand said. “If a patient definitely requires surgery, then all compressive pathology should be addressed at the time of surgery,” he said.

There is also some debate in the literature as to whether or not arthroplasty reduces the risk for ASP.

“There is weak evidence that there is a slightly reduced rate of adjacent-segment pathology with arthroplasty, but it still remains unclear,” Fehlings said.

Pierce D. Nunley

Pierce D. Nunley

Researchers have argued that because TDA allows the disc to keep moving, then the adjacent segment should not deteriorate. “However, now the 5- or 6-year data coming out show that the adjacent-segment deterioration next to total disc replacements is not different than that of cervical fusion,” said Kang, who is a member of the Spine Surgery Today Editorial Board. “That might lead directly to the conclusion that this is the natural history of the patient, but we don’t know that for certain because TDA may not provide good motion. Just because the disc moves doesn’t mean it is normal, natural motion. It can still be altered by biomechanics. We don’t know the answer to that.”

Nunley also called for the creation of algorithms to determine how a patient should best be treated to avoid ASP.

“We are going to have to find ways specifically to say, for example, if you’re between the ages of 30 [years] and 50 [years], if you don’t have any lumbar disc disease, you don’t have osteoporosis, it shows that you are going to do better with disc replacement,” he said. “Or, if you are 60 years old, you have the beginnings of osteoporosis and concurrent lower degenerative disc disease, you need a fusion. Those types of algorithms created by evidence-based data need to get created. As they do, I think we will see the cost-effectiveness.”

Registries to collect data

No nationwide registries are currently available to track these patients and ASP cases. However, Hilibrand told Spine Surgery Today that AOSpine and the Cervical Spine Research Society (CSRS) are working to develop registries. Nunley is also involved in the development of two registries, one by the Scoliosis Research Society and the other by the International Spine Study Group.

“The International Spine Study Group has a very large registry looking at adjacent-segment pathology and we have a couple of ongoing projects,” Nunley said. “Our data are still being analyzed, but it appears the number is somewhere from 2.5% to 4% per year for the number of symptomatic or clinical adjacent-segment pathology cases.”

In addition, Nunley’s research foundation received a substantial CSRS grant to study all patients in four FDA cervical TDA studies. They performed MRIs in all the patients at 5 years postoperatively.


“We are now analyzing that data, but unlike any of the previous studies, we’ll probably have at least a 90% follow-up evaluating radiographic ASP and we will have symptomatic and asymptomatic patients equally,” Nunley said.

He and his colleagues plan to present the data at the CSRS Annual Meeting in December.

Hilibrand noted that studies like the SPORT trial, which is almost 10 years out, should provide helpful data for tracking ASP.

“In the cervical spine, individual surgeons are now tracking their data over a long period of time,” he said. “I think we will see more information about individual surgeon experience.”

In addition to registries, basic science research is needed to evaluate the effects of a surgical fusion on transferring range of range of motion to the upper cervical spine, Hilibrand said. “Right now most biomechanical studies only look at the lower cervical spine and the effects of fusion on other levels of the lower (C2 to T1) cervical spine,” he said. – by Tina DiMarcantonio

Anderson PA. Spine. 2012. doi:10.1097/BRS.0b013e31826d62ed.
Hilibrand AS. J Bone Joint Surg Am. 1999;81:519-528.
Kraemer P. Spine. 2012. doi:10.1097/BRS.0b013e31826d7dd6.
Lawrence BD. Spine. 2012. doi:10.1097/BRS.0b013e31826d60fb.
Lawrence BD. Spine. 2012. doi:10.1097/BRS.0b013e31826d60d8.
Weinstein JN. J Bone Joint Surg Am. 2009. doi:10.2106/JBJS.H.00913.
Weinstein JN. Spine. 2010. doi:10.1097/BRS.0b013e3181e0f04d.
For more information:
Joseph S. Cheng, MD, MS, can be reached at T-4224 Medical Center North, Nashville, TN 37232-2380; email: joseph.cheng@vanderbilt.edu.
Michael G. Fehlings, MD, PhD, FRCSC, FACS, can be reached at Toronto Western Hospital, West Wing, 4th Floor, Room 4WW449, 399 Bathurst St., Toronto, ON M5T 2S8, Canada; email: michael.fehlings@uhn.ca.
Alan S. Hilibrand, MD, can be reached at 925 Chestnut St., 5th Floor, Philadelphia, PA 19107; email: deborah.bauer@rothmaninstitute.com.
James D. Kang, MD, can be reached at Department of Orthopaedics, University of Pittsburgh, Kaufman Medical Building, 3471 Fifth Ave., Suite 1010, Pittsburgh, PA 15213; email: kangjd@upmc.edu.
Pierce D. Nunley, MD, can be reached at 1500 Line Ave., Suite 200, Shreveport, LA 71101; email: pnunley@louisianaspine.org.
Disclosures: Cheng has no relevant financial disclosures. Fehlings receives royalties for the Discover Cervical Disc. Hilibrand receives royalties from Aesculap/B. Braun, Alphatec Spine, Amedica, Biomet, and Zimmer, and has stock or stock options in Amedica, Benvenue Medical, Lifespine, Nexgen, Paradigm Spine, Pioneer Surgical, PSD, Spinal Ventures, Syndicom and Vertiflex. Kang’s institution has a research grant with Johnson & Johnson (Synthes). Nunley’s institution, the Spine Institute of Louisiana Research Foundation, has received routine research support.


What procedures have you performed most often to treat or manage adjacent-segment pathology?


Depends on several factors

What operation we use to treat a patient with symptomatic adjacent-segment pathology depends on numerous factors. These include how many levels are involved, how many levels the original operation involved, the level of the adjacent-segment pathology (ASP), the state of degeneration of that level, the specific pathology at the level causing symptoms, and age of the patient.

K. Daniel Riew

K. Daniel Riew

For example, a young patient who had a single-level anterior cervical discectomy and fusion (ACDF) or arthroplasty with an adjacent-segment centrally herniated disc would be an excellent candidate for another ACDF or arthroplasty. If the symptoms are caused by a foraminal disc herniation or bony stenosis, they might also be a candidate for a posterior foraminotomy with or without discectomy. On the other hand, an older person with an anterior decompression and fusion from C4-7 who breaks down at C3-4 and C7-T1 would probably do best from a posterior C3-T2 decompression and fusion. Likewise, a patient who has a segment that is difficult to approach from the anterior is more likely to have a posterior operation and vice versa. Examples include a T1-2 herniated disc, which might be more easily done posteriorly. Alternatively, someone with a history of posterior infection, laminectomy and wound issues with a single-level problem is more likely to have an anterior operation.


In rare cases, a circumferential procedure is necessary. Most commonly, this involves a case where there is severe central stenosis below a long anterior fusion in a patient at risk for pseudarthrosis (eg, C3-7 with a large central disc herniation at C7-T1). An ACDF at this level is already at risk for pseudarthrosis due to the long lever arm of the fused segment above. If the patient has additional risk factors, such as tobacco abuse or diabetes, it might be prudent to operate circumferentially to ensure a high rate of fusion.

K. Daniel Riew, MD, is the Mildred B. Simon Distinguished Professor of Orthopedic Surgery, professor of neurological surgery, chief of cervical spine surgery, and director of the Orthopedic & Rehab Institute for Cervical Spine Surgery at Washington University Orthopedics in St. Louis. He is McDonnell Academy Ambassador at Barnes-Jewish Hospital & Washington University School of Medicine and a member of the Spine Surgery Today Editorial Board.
Disclosure: Riew currently receives royalties from Medtronic and Biomet and has received royalties in the past from Osprey. He also owns stocks in Amedica, Benvenue, Expanding Orthopedics, Nexgen Spine, Osprey, Paradigm Spine, Spinal Kinetics, Spineology and Vertiflex. He has received research support from AO Spine, Cerapedics, CSRS, Medtronic and Spinal Dynamics.


Studies show new insights in treatment

The clinical course of adjacent-segment pathology (ASP) has been a point of contention in the literature. In 1990, Herkowitz and colleagues randomly assigned patients to anterior cervical discectomy and fusion (ACDF) vs. posterior cervical foraminotomy (PCF) without fusion for the surgical management of cervical radiculopathy. While 41% of patients in the ACDF group developed ASP, 50% of patients in the PCF group developed ASP. These early studies suggest ASP reflects the natural history of cervical degeneration. Almost 10 years later, with the advent of advanced diagnostic modalities, Hilibrand and colleagues found the cumulative incidence of ASP was 25.6% after 10 years.

Ali Bydon

Ali Bydon

While some authors contend ASP is part of the natural history of cervical spondylosis, others argue it is a consequence of instrumented fixation of the spinal segments. We recently published a series of three ACDF manuscripts in which the medical records of 888 patients who underwent ACDF for cervical spondylosis during a 20-year period were reviewed. We found among those ACDF cases, symptomatic ASP required reoperation in 12% of patients (n=108). Thus, whether ASP is a natural progression of spondylosis or a sequela of ACDF, degeneration of adjacent spinal segments may still be amenable to operative management.

Revision surgery for ASP may follow an anterior or posterior operation. In our 2013 clinical series, we found anterior revision surgery for ASP after ACDF led to higher rates of postoperative radiculopathy and redevelopment of ASP when compared with posterior revision surgery. Are laminectomy and fusion the answer to the ASP conundrum? Not necessarily. While posterior operations conferred a lower rate of redevelopment of ASP necessitating secondary revision surgery, the posterior approach also had higher rates of postoperative complications, such as wound infection and discharge to rehab. Moreover, in comparison with anterior revisions, more spinal levels were fusions in the posterior operations, which increases the limitation of patients’ cervical range of motion.

The outcomes of these patients requiring second-time and even third-time reoperations have provided new insights into the progression and pathogenesis of ASP. In our series, the incidence of ASP requiring reoperation statistically increased from 12% after first-time ACDF to 25% after a second cervical fusion (n=108; Xu and colleagues). Patients with a second cervical fusion via an anterior approach had a higher chance for developing recurrent ASP vs. patients with a posterior approach. Furthermore, the time to ASP development decreased from 47 months after the first ACDF to 30 months after a second cervical fusion. Because patients who undergo a second cervical fusion develop ASP at both higher and faster rates, immobilization of the cervical spine with instrumented fusion may indeed contribute, at least in part, to the progression of ASP.


All other factors taken equally, however, the location of ASP may influence the risk for repeat operation. Commensurate with Hilibrand and colleagues, we found the C5-6 segment saw the highest rate of cervical spinal degeneration followed by C6-7. Rostral ASP was also more common than caudal ASP. On the other hand, neither the location of the index ACDF nor the number of spinal levels fused correlated with the propensity to develop ASP.

ASP is a complex phenomenon in spine surgery with several prognostic factors, including the number of reoperations, time to reoperation, surgical approach and location of the spinal instability. Although conservative therapy may be appropriate in selected patients, 12% of patients required reoperation after an ACDF.

Ali Bydon, MD, a member of the Spine Surgery Today Editorial Board, is associate professor in the department of neurosurgery, codirector of neurosurgical medical student education, director of neurosurgical undergraduate education, director of the Spinal Column Biomechanics and Surgical Outcomes Lab, and clinical director of the JHBMC Spine Program at Johns Hopkins Neurosurgical Spine Center.
Disclosure: Bydon received a research grant from DePuy Spine. He serves on the clinical advisory board of MedImmune LLC.

Bydon M. Neurosurgery. 2013. doi:10.1227/NEU.0000000000000204.
Herkowitz HN. Spine. 1990;15:1026-1030.
Hilibrand AS. J Bone Joint Surg Am. 1999;81:519-528.
Xu R. Spine. 2014. doi:10.1097/BRS.0000000000000074.