Lumbar total disc replacement (TDR) is a treatment option for selected patients with chronic low back pain (LBP) that is non-responsive to conservative treatment. The long-term results of disc replacement compared with multidisciplinary rehabilitation (MDR) have not been reported previously.

We aimed to assess the long-term relative efficacy of lumbar TDR compared with MDR.

We undertook a multicenter randomized controlled trial at five university hospitals in Norway.

The sample consisted of 173 patients aged 25-55 years with chronic LBP and localized degenerative changes in the lumbar intervertebral discs.

The primary outcome was self-reported physical function (Oswestry Disability Index [ODI]) at 8-year follow-up in the intention-to-treat population. Secondary outcomes included self-reported LBP (visual analogue scale [VAS]), quality of life (EuroQol [EQ-5D]), emotional distress (Hopkins Symptom Checklist [HSCL-25]), occupational status, patient satisfaction, drug use, complications, and additional back surgery.

Patients were randomly assigned to lumbar TDR or MDR. Self-reported outcome measures were collected 8 years after treatment. The study was powered to detect a difference of 10 ODI points between the groups. The study has not been funded by the industry.

A total of 605 patients were screened for eligibility, of whom 173 were randomly assigned treatment. Seventy-seven patients (90%) randomized to surgery and 74 patients (85%) randomized to rehabilitation responded at 8-year follow-up. Mean improvement in the ODI was 20.0 points (95% confidence interval [CI] 16.4-23.6, p≤.0001) in the surgery group and 14.4 points (95% CI 10.7-18.1, p≤.0001) in the rehabilitation group. Mean difference between the groups at 8-year follow-up was 6.1 points (95% CI 1.2-11.0, p=.02). Mean difference in favor of surgery on secondary outcomes were 9.9 points on VAS (95% CI 0.6-19.2, p=.04) and 0.16 points on HSCL-25 (95% CI 0.01-0.32, p=.04). There were 18 patients (24%) in the surgery group and 4 patients (6%) in the rehabilitation group who reported full recovery (p=.002). There were no significant differences between the groups in EQ-5D, occupational status, satisfaction with care, or drug use. In the per protocol analysis, the mean difference between groups was 8.1 ODI points (95% CI 2.3-13.9, p=.01) in favor of surgery. Forty-three of 61 patients (70%) in the surgery group and 26 of 52 patients (50%) in the rehabilitation group had a clinically important improvement (15 ODI points or more) from baseline (p=.03). The proportion of patients with a clinically important deterioration (six ODI points or more) was not significantly different between the groups. Twenty-one patients (24%) randomized to rehabilitation had crossed over and had undergone back surgery since inclusion, whereas 12 patients (14%) randomized to surgery had undergone additional back surgery. One serious adverse event after disc replacement is registered (< 1%).

Substantial long-term improvement can be expected after both disc replacement and MDR. The difference between groups is statistically significant in favor of surgery, but smaller than the prespecified clinically important difference of 10 ODI points that the study was designed to detect. Future research should aim to improve selection criteria for disc replacement and MDR.

Polls results

On a scale of 1 to 10, rate how much this article will change your clinical practice?

NO change
BIG change
84% Article relates to my practice (53/63)
6% Article does not relate to my practice (4/63)
9% Undecided (6/63)

Will this article lead to more cost-effective healthcare?

60% Yes (38/63)
19% No (12/63)
20% Undecided (13/63)

Was this article biased? (commercial or personal)

6% Yes (4/63)
80% No (51/63)
12% Undecided (8/63)

What level of evidence do you think this article is?

34% Level 1 (22/63)
31% Level 2 (20/63)
28% Level 3 (18/63)
4% Level 4 (3/63)
0% Level 5 (0/63)