During the last two decades, new microinvasive techniques have been introduced into the treatment of lumbar disc herniation. The potential benefits of microinvasive disc surgery are the reduced surgical trauma to the tissue, increased safety due to good visualization of the operation field under the microscope and, consequently, reduced postoperative morbidity and shorter hospitalization in comparison to conventional surgery. In this study, we evaluated the use of microdiscectomy and percutaneous nucleotomy in the treatment of patients with lumbar disc herniation. Of the 237 patients who underwent microdiscectomy for virgin single-level lumbar disc herniation, 92% informed that their sciatic pain had completely recovered or markedly diminished during a median postoperative follow-up of 2 years and 79% of these patients had returned to work. The outcome of the patients operated on for a disc protrusion was to some extent less satisfactory than the outcome of the patients operated on for a prolapse or a sequestrum. Further, of the 45 patients who underwent percutaneous nucleotomy, the sciatic pain had completely recovered or markedly diminished in 38 (84%) patients during a mean postoperative follow-up of 2 years and 78% of them had returned to work. Also in these patients, a protrusion-type of disc herniation was associated with an inferior outcome. After a mean postoperative follow-up of 3 years, the patients treated with microdiscectomy underwent repeated clinical examination. Segmental instability of the lumbar spine was observed in 22% of 190 patients studied. Lumbar instability correlated significantly (P < 0.0001) with an unsatisfactory long-term outcome in these patients. Clinical instability was preoperatively detected in 24% of the 45 patients treated with percutaneous nucleotomy. Also in these patients, instability predicted significantly (P < 0.05) an inferior outcome. On the first postoperative day, there was an extradural hematoma in 86% of the 44 patients studied with MRI. The incidence of hematomas was associated with the surgical method used: all 28 patients treated with microdiscectomy but only 10 (63%) of the 16 patients treated with percutaneous nucleotomy had a hematoma (P = 0.001). In addition, 25 (61%) of the 41 patients studied had an edematous epidural mass effect mimicking preoperative disc herniation. During follow-up for 6 months, the mass effect resolved completely in 10 of these patients but in 15 patients, MRI subsequently revealed either a disc prolapse or a protrusion. In addition, a protrusion was detected in 7 patients with no postoperative mass effect. There was, however, no association between these MRI findings and the clinical outcome of the patients.(ABSTRACT TRUNCATED AT 400 WORDS)