Number of segments fused a risk factor for adjacent segment disease

Issue: March 2011

The number of segments fused in a posterior lumbar interbody fusion procedure — as well as previous lumbar decompressive surgery — can be considered risk factors for adjacent segment disease, according to a study presented at the 2010 North American Spine Society Annual Meeting in Orlando, Fla.

During his presentation, William R. Sears, MBBS, FRACS, also noted that patients who underwent single-level posterior lumbar interbody fusion (PLIF) for lytic spondylolisthesis appear to be at lower risk for developing adjacent segment disease (ASD) than those operated on for degenerative spondylolisthesis.

“The existence of [ASD] as an entity related to spinal fusion or the fusion adjacent is controversial,” Sears said. “Whether it is perhaps genetic factors or somewhere in between is still unclear. What is also unclear from the literature are the rates of development of [ASD], but this is what is driving the whole new motion preservation technology industry, so I think it deserves some attention.

“I think we have always understood that single-level fusions have a relatively low risk of ASD, but patients often question the ‘domino’ effect of fusing the spine,” Sears told Orthopedics Today. “This study has helped me to inform patients about the real level of their risk of developing ASD, both low and high.”

Flexion plain X-ray of the lumbar spine of an 81 year-old woman, 2 years following L3-S1 posterior lumbar interbody fusion.

Extension plain X-ray of the lumbar spine of an 81 year-old woman, 2 years following L3-S1 posterior lumbar interbody fusion.

Flexion/extension plain X-rays of the lumbar spine of an 81 year-old woman, 2 years following L3-S1 posterior lumbar interbody fusion. The patient complained of increasing back pain and leg weakness.

Images: Sears WR

Survey follow-up

Sears conducted postal and telephone surveys of 912 consecutive patients who underwent a total of 1,000 PLIF procedures for degenerative spine disease between October 1993 and November 2009 from a prospectively acquired lumbar fusion database. Median age of this group was 63 years, with a range of 14 to 92 years. Single-level fusion was performed in 61% of the PLIF procedures, two levels were fused in 20%, three levels in 5% and four levels in 5%. Five or more levels were fused in 9% of the procedures.

The surveys inquired as to whether patients had undergone further surgery since their index procedure. In the case of deceased patients, Sears said a record was made of whether they had undergone further surgery prior to death.

Sears reported a follow-up rate of 91%, including the 10% of patients who had died. Mean follow-up duration was 59 months, and a minimum of 5-year follow-up data was available on 413 fusions.

Risk factors

A Kaplan-Meier analysis revealed a mean time of 166 months to ASD surgery following all 1-level fusions. Mean time to ASD surgery following 2-level fusions was 146 months, and the mean time to ASD surgery following 3- and 4-level fusions was 126 months.

Furthermore, with patients who had a minimum 5-year follow-up, 8.2% of 1-level patients had undergone further surgery, compared with 15.1% for 2-level and 10.8% for 3- and 4-level patients.

CT myelogram of a 75 year-old woman suffering claudicant back and leg pain.
CT myelogram of a 75 year-old woman suffering claudicant back and leg pain. Severe degenerative change and canal stenosis is evident, adjacent to a previous instrumented and relatively kyphotic, postero-lateral fusion at L4/5.

Sears noted that 1-level fusions for degenerative spondylolisthesis had a significantly higher incidence of subsequent ASD surgery: annual incidence of 1.1% and 5-year prevalence of 11%, compared with 2.4% and 11% for lytic spondylolisthesis, respectively. The degenerative spondylolisthesis group reportedly showed no difference in development of ASD between men and women.

Sears said the average annual incidence was approximately 2.5%, but noted that this was not uniform due to relative risk factors.

“Statistically, the greatest significance — and perhaps the most clinical significance — is that if you fuse one level, you have a 1.7% risk,” he said. “You double that if you do two levels, you triple it if you do three or four levels.”

The take-home message

“The risk of developing ASD in young patients following single-level procedures is relatively low when compared with the high risk following multilevel fusions, especially in older patients,” Sears told Orthopedics Today, noting a prevalence of 29% at 5 years and 40% at 10 years following three- or four-level fusions. “Perhaps then, we should be directing more of our clinical research efforts on evaluating dynamic stabilization devices in this patient group rather than the usual, relatively young patient populations with single-level devices that have tended to be the subject of large, prospective, randomized trials.” – by Robert Press

Reference:

Sears WR. Adjacent segment disease following 1,049 posterior lumbar interbody fusions: A retrospective review. Paper #154. Presented at the 2010 North American Spine Society Annual Meeting. October 5-9, 2010. Orlando, Fla.

William R. Sears, MBBS, FRACS, can be reached at sears.public@mac.com.

Disclosure: Sears is a consultant to and receives royalties for an interbody fusion device from Medtronic. He is also a consultant to and holds stock options with Paradigm Spine GmbH.

Revision surgery for ASD is an important driver of the cost of spine surgery, and a clinically significant sequela of lumbar fusion surgery. Accurate information on the prevalence of revision surgery for symptomatic adjacent segment degeneration is useful for the patient and the physician in guiding informed decision making. Identification of risk factors for ASD may guide surgical strategies and alternative approaches to the management of lumbar degenerative disorders.

The investigators reviewed the largest series of consecutive patients treated with PLIF surgery to identify rates of ASD. The data demonstrate that diagnosis and the number of levels fused may be independent predictors of ASD rates. The strength of the study is the large cohort under study, and the high rate of follow-up. However, the study has significant limitations including lack of a definition of the cause of revision surgery, and incomplete identification of risk factors for revision surgery. The study would be strengthened by a multivariate analysis to control for the possibility that number of levels fused may be dependent on diagnosis or other confounding variables.

Overall, ASD is an important cause of revision lumbar surgery. This study offers important information on the prevalence and timing of revision surgery after PLIF, and may be useful for informing patients on expected outcomes of care after PLIF surgery. Surgeons and patients are most interested in the question: What is the rate of ASD compared with the alternative? Further study may include a matched cohort analysis of ASD rates in patients treated with circumferential fusion compared with alternatives including posterolateral-only fusion, anterior-only fusion and dynamic stabilization or arthroplasty. Further identification of specific ASD risk factors including segmental and regional alignment, success of arthrodesis, decompression at the upper instrumented vertebra, and patient-specific factors may require further investigation of the cohort studied.

— Sigurd H. Berven, MD
Director of Clinical Fellowship in Spine Surgery and Resident Education Committee
Department of Orthopaedic Surgery
UC San Francisco
San Francisco, Calif.
Disclosure: He is a consultant for Medtronic and DePuy Spine.