Aims: We undertook an analysis to determine the prognostic indicators of successful outcome following decompression for radiculopathy from lumbar spinal stenosis. Methods: 203 patients underwent spinal decompression in a þve year period till June 2001 and were subsequently followed up. Age, sex, number of levels, the speciþc levels involved, type of stenosis, symptoms, duration, bilaterality were preoperative factors looked at. The type of decompression, number of levels decompressed, the speciþc levels and intra-operative complications were noted. Postoperative resolution of pain, duration to alleviation of pain were assessed. Patient satisfaction and discharge from clinic indicated successful outcome. Referral to the pain clinic reßected a failure of treatment. Results: 65% of patients who had primary decompression experienced satisfactory improvement in symptoms. Logistic Regression analysis showed that the presence of radicular pain at þrst review signiþcantly increased the likelihood of failure of surgery and referral to the pain clinic for (p=0.02) for leg symptoms. 57% of patients who had decompression following previous surgery at the same level were relieved of leg pain. The duration of leg pain alone adversely affected þnal outcome (p=0.01) amongst all the factors assessed including complications from surgery. The complication rate from revision surgery (7/30) was signiþcantly greater (p=0.01) than primary decompression (16/173). Conclusions: Persistance of radicular pain early after decompression increases the likelihood of eventual failure to improve symptoms Patients with long periods of pain prior to decompression following previous back surgery should be cautioned about the decreased likelihood of success and an increased risk of complications.