Published 1/1/2009
Annie Hayashi

Award-winning study examines outcomes based on novel diagnostic indications

Study is move toward evidence-based decisions in spinal surgery

“Chronic low back pain is a symptom, not a diagnostic entity from which treatment outcomes can be readily extrapolated,” said Steven D. Glassman, MD, at the recent North American Spine Society Meeting.

“Asking whether fusion is a good treatment for low back pain is like asking whether antibiotics are a good treatment for shortness of breath.”

To effectively translate outcome data into clinical decision-making or evidence-based guidelines, according to Dr. Glassman, greater diagnostic specificity is needed. The report of his attempt to measure outcomes based on diagnostic subgroups was honored with a “2008 Outstanding Paper Award” from The Spine Journal.

Measuring outcomes based on diagnosis
Dr. Glassman’s single-center study involved 327 patients with lumber degenerative disease who underwent lumbar spine fusion and completed 2 years of follow-up. Each patient was assigned to a diagnostic subgroup prior to undergoing primary or revision lumbar fusion surgery. The diagnostic subgroups for primary and revision surgery and the numbers of patients in each are shown in
Table 1.

Patient-reported outcome measures—Oswestry Disability Index (ODI), Medical Outcome Study Short Form-36 (MOS SF-36), and numeric rating scales (NRSs) for back pain and leg pain—were collected at baseline, and postoperatively at 6 months, 1 year, and 2 years. The SF-36 physical component score (PCS) was used in the latter portion of the study.

At 2 years postsurgery, the mean SF-36 PCS and the mean ODI scores for the entire study population had improved. But the diagnostic subgroups had varying levels of improvement in their ODI scores (Table 2).

“Assessment of outcomes based on a specific diagnosis revealed the most substantial and most consistent improvement in the spondylolisthesis group,” Dr. Glassman reported.

The spondylolisthesis subgroup scored at or near the maximum for all health-related quality of life (HRQOL) measures. In addition, 71 percent of that subgroup reached the minimum clinically important difference on the ODI scales—a higher percentage than in any other subgroup—despite a greater incidence of complications. The spondylolisthesis and scoliosis subgroups demonstrated the most substantial improvement in ODI scores at 1 and 2 years postsurgery.

Among stenosis patients, however, a statistically significant deterioration in ODI scores was seen between the first and second year postsurgery.

Although stenosis patients were older (average age: 64 years) and had a high incidence of subsequent adjacent level deterioration, Dr. Glassman reported that the reason for the ODI deterioration was not readily obvious.

SF-36PCS outcomes
When measuring SF-36 PCS, researchers found very different results. The greatest improvement was seen in the disk pathology and spondylolisthesis subgroups; patients in the nonunion and adjacent level degeneration subgroups had the least improvement.

Scoliosis subgroup
Dr. Glassman expressed surprise that the scoliosis subgroup scored as well as it did because the patients in that group tended to be older (average age: 63 years) and had to undergo a more extensive procedure (mean 2.7 levels) when compared with other groups.

“Although the group was small,” said Dr. Glassman, “the incidence of complications and the need for adjacent level revision trended high in this group, consistent with the complexity of the pathology.”

Revision outcomes
Among patients in the revision surgery groups, results for those in the postdiskectomy group were “somewhat more optimistic,” according to Dr. Glassman. “These patients have been typically reported to obtain inferior HRQOL benefits compared with primary surgery cases.

Mean improvement in ODI (14.0 points) and SF-36 PCS (5.3 points) was competitive with some of the primary fusion categories.”

Within the 2-year study timeframe, 34 patients required revision surgery. Revision rates were highest in the nonunion subgroup (26.1 percent) and in the adjacent level degeneration category (17 percent).

Call for diagnostic specificity
Although Dr. Glassman concedes that the diagnostic subgroups outlined in this study may not be sufficient, he believes “it is a step in the right direction.”

“Diagnostic specificity is a critical component in building an improved evidence base for lumbar fusion surgery. Without diagnostic specificity for entities beyond spondylolisthesis, the absence of well-defined study populations will continue to limit our ability to move toward evidence-based decision making,” he concluded.

The authors have the following disclosures:

Dr. Glassman: Medtronic Sofamor Danek; Leah Yacat Carreon, MD: Medtronic Sofamor Danek Norton Healthcare; Mladen Djurasovic, MD: Medtronic Sofamor Danek; John R. Dimar II, MD: Medtronic Sofamor Danek; John R. Johnson, MD: Medtronic Sofamor Danek; Rolando M. Puno, MD: Medtronic Sofamor Danek; Mitchell J. Campbell, MD: Medtronic Sofamor Danek

Annie Hayashi is the senior science writer for AAOS Now. She can be reached at hayashi@aaos.org