Lumbar fusion is not necessarily the best option in chronic low back pain patients


Anne Mannion (Spine Center, Schulthess Klinik, Zürich, Switzerland) reported the results of an 11-year follow-up study that showed that chronic low back pain patients treated with lumbar fusion surgery did not have significantly better outcomes compared with patients treated with non-operative management.

Speaking at the British Association of Spine Surgeons meeting (BASS; 13–15 March, Norwich, UK), Mannion said that chronic low back pain was a “major challenge” to healthcare systems and that management of the condition remained controversial. She added that four randomised controlled trials that compared lumbar fusion surgery with conservative treatment did not provide conclusive results, with one indicating that surgery was superior to conservative management and three others not finding a significant difference between the two treatment options. Furthermore, the question of how the long-term outcomes of surgery compare with the long-term outcomes of non-surgical treatment has never been addressed. Mannion commented that finding an answer was important in terms of ensuring that the patient was able to give their informed consent when agreeing to a treatment option (either surgery or non-operative treatment).

The aim of the present study, Mannion reported, was to compare “clinical outcomes at long-term follow-up of patients randomised to either surgery or multidisciplinary cognitive behaviour therapy and exercise rehabilitation.” All of the patients in the study had chronic low back pain and were potential candidates for fusion. Mannion et al created a new booklet of questionnaires and sent it to patients from all of the previous randomised trials that compared fusion with non-surgical treatment. Patients were followed-up at an average of 11 years after treatment. Mannion stated: “The primary outcome was the Oswestry Disability Index score at follow-up and we had various secondary outcomes.” The secondary outcomes included pain (back and leg), quality of life, global treatment outcome, pain frequency, pain medication use, work status, and satisfaction with care.

Of the 473 patients enrolled in the study, only 261 (140 in the surgery group and 121 in the non-operative care group) were available at the 11-year follow-up point. Mannion commented that this was “not the best follow-up in the world” but added that it was “acceptable” given the length of the follow-up.  

In both the intention-to-treat analysis and the as-treated analysis, after adjusting for confounding variables (ie. baseline Oswestry Disability Index, age, gender, smoking, previous surgery, and duration of low back pain), Mannion et al did not observe any significant differences in the primary outcome (Oswestry Disability Index score at follow-up) between the groups. Additionally, there were no significant differences between the groups in terms of the secondary outcomes. Mannion commented, “These data are fairly convincing because no differences were observed between the groups regardless of the outcome examined or the method of analysis.”

Summarising the results, Mannion stated that the study was: “The largest randomised trial on surgery for chronic low back pain with the longest follow-up, so there are no issues regarding insufficient statistical power.” She added: “The study provides essential information for future patients regarding the likely long-term benefits of spinal surgery and non-operative treatments. Given the increased risks of surgery and the lack of deterioration in the non-operative patients over time, the use of lumbar fusion in chronic low back pain patients should not be favoured when programmes of multidisciplinary rehabilitation are available.”

According to Mannion, further work is needed to “hone indications for patient selection for both treatments.” She told Spinal News International these studies could examine various individual baseline factors (demographic variables, clinical history, imaging findings, clinical status, etc.) to try to identify whether differences in any of these were able to statistically predict better results after fusion compared with non-surgical treatment. Mannion added: “In this way, one would identify who benefits most from surgery and who would do just as well (or better) with non-operative treatment.