The aim of the study was to evaluate the long-term outcome of various surgical procedures for lumbar spinal stenosis. Operations were performed on 117 consecutive patients for lumbar spinal stenosis between 1987 and 1992. Pre- and intraoperative data were recorded in a standardized manner. Three treatment groups were distinguished: group I consisting of 39 patients submitted to undercutting decompression; group II, 51 patients, submitted to laminectomy and foraminal decompression alone; and group III, 27 patients, who underwent foraminal decompression and laminectomy with instrumented fusion. Eight years (5-10 years) after surgery a questionnaire was mailed to the patients containing the outcome scales according to Greenough and Fraser [6] and Turner et al. [22] together with questions about residual pain, necessity of treatment and satisfaction with the operative outcome. A total of 72 questionnaires (61.6%) gave enough information for analysis. After a mean follow-up of 8 years, walking capacity had increased significantly in all groups (P < 0.001). Compared to preoperative values, pain had decreased significantly in all groups (P < 0.01). In group I 36% had good-to-excellent outcomes, and 30.8% and 23.8% in groups II and III (P > 0.05). Forty percent of group I patients were unsatisfied with the result, compared to 38.4% and 33.3% in the other groups (P > 0.05). Overall, 25 of 72 patients (34.7%) had severe constant back and/or leg pain requiring daily administration of analgesics. We conclude that the long-term outcome of decompressive surgery of the lumbar spinal canal, without and with instrumented fusion, is less favourable than was previously reported.