Surgery reduced pain and disability in lumbar spinal stenosis better than nonoperative treatment. Commentary

Question: In patients with lumbar spinal stenosis, what is the effectiveness of decompressive surgery compared with nonoperative treatment with respect to pain and disability? Design: Randomized (allocation concealed)*, unblinded, controlled trial with 24-month follow-up. Setting: 4 university hospitals in Finland. Patients: 94 patients (mean age, 63 y; 67% women) with lumbar spinal stenosis. Inclusion criteria were back pain radiating to lower limbs or buttocks, fatigue or loss of sensation in the lower limbs aggravated by walking, persistent pain without progressive neurologic dysfunction, spinal canal narrowing on imaging, and symptoms and signs lasting for >6 months. Exclusion criteria were lumbar spinal stenosis severe enough to require surgery or mild enough to exclude surgery; spinal stenosis not caused by degeneration; spondylolysis and spondylolytic spondylolisthesis; a previous back operation because of stenosis or instability; a herniated lumbar disc in the previous 12 months; another spinal disorder; intermittent claudication due to atherosclerosis; severe osteoarthritis, arthritis, or neurologic disease affecting the lower limbs; psychiatric disorders; or alcoholism. 87 patients (93%) were available for 24-month follow-up. Intervention: 50 patients were allocated to surgery involving segmental decompression and an undercutting facetectomy of the affected area. Lumbar instability was treated at the surgeon's discretion with fusion of the lumbar spine augmented by transpedicular instrumentation, if necessary. 44 patients were allocated to nonoperative treatment involving follow-up by a physiatrist who assessed the need for individual treatment. Patients in the nonoperative group received nonsteroidal anti-inflammatory drugs and saw a physiotherapist 1 to 3 times. Patients in both groups received a brochure describing spinal stenosis. Main outcome measures: Change from baseline on the Oswestry Disability Index (ODI) (range, 0 [no disability] to 100 [worst possible disability]), leg and low-back pain during walking (range, 0 [no pain] to 10 [worst possible pain]), and self-reported walking ability. Main results: Analysis was by intention to treat. ODI scores improved more in patients who received surgery than in patients who received nonoperative treatment (Table). Patients in the surgery group also had less leg and back pain during walking. Walking ability improved in both groups with no difference between them. Conclusions: In patients with moderate lumbar spinal stenosis, whose severity of symptoms would justify either surgical or nonoperative treatment, decompressive surgery reduced pain and disability more than nonoperative treatment did. The effectiveness for pain and disability was sustained over 2 years, but the objective measure of walking ability improved in both groups, with no statistical difference between operative and nonoperative groups.