Background Lumbar canal stenosis (LCS) is a clinicopathological and radiological disease first described in 1949, causing a neurovascular compression leading to back pain, radicular, sensory or motor deficit which treatment is difficult.
Objectives The objectives of our study were appraisal of a protocolized treatment for 3 months and identification of predictive factors for failure of medical treatment. This is a prospective nonrandomized study with no control group. All patients satisfying inclusion criteria were included after clinical and radiological evaluation.
Methods Clinical inclusion criteria were neurogenic claudication, radiculopathy at rest with or without back pain, mono or multirooted sensory, motor or reflex abnormalities of the lower limbs. Radiological inclusion criteria were: anteroposterior dural sac diameter <10 mm, surface of the dural sac <100 mm2, reduction in the anteroposterior diameter of the spinal canal >25% on myelography in supine extended position.
Results Included patients had at least one clinical and one radiological criteria. Treatment was 2 epidural injections of 125 mg hydrocortancyl 15 days apart, combined with NSAIDs or analgesic, associated with physiotherapy sessions over 2 months.
147 patients were included. Mean age was 62±13 years. F/M sex ratio is 1.3. Mean BMI was 27.5±4.3, overweight was present in 72.1%. At baseline, mean duration for back pain and radicular pain were 9.3±10.5 and 4.4±5.7years. Primary objective was evolution of the Oswestry  score (a low back pain disability questionnaire) at 3 months as compared to baseline. Secondary objective was the evolution of pain.
At the end of treatment, the mean value for the Oswestry score decreased from 41.2 to 39.7 (p=0.17). Mean VAS for lumbar pain decreased from 53.6 to 42.1 (p<0.001) and root mean VAS from 57.7 to 49.1 (p=0.0025). At 3 months 38 patients improved (25.9%), 73 had stabilized (49.7%) and 30 had worsened (20.4%).
BMI, duration of back and radicular pain, mean lumbar radicular VAS and mean Oswestry score at inclusion were not significantly different in improved and non improved patients. Predictors of good response to treatment were the initial presence of impulsivity cough and nocturnal pain.
Conclusions In conclusion with this approach there was no significant decrease in Oswestry score, however there was a significant decrease in mean VAS lumbar and radicular pain. No predictor of poor response has been highlighted. Predictors of good response were the initial presence of impulsivity at cough and nocturnal pain. These results show the limits of medical treatment in the management of SCL.
Niskanen RO, The Oswestry Low Back Pain Disability Questionnaire. a two-year follow-up of spine surgery patients. Scand J Surg, 2002 91 (2):. P. 208-11.
Disclosure of Interest None declared