We have submitted the records of 115 patients operated upon for intervertebral-disc lesion to a rather searching analysis. The results in these patients have been classified according to arbitrarily established criteria and as many variables as possible were eliminated from the study. In addition, the figures were all submitted to statistical analysis. Significant findings are:

1. Patients with good historical, physical, and myelographic evidence of nerve-root compression obtain better results from disc surgery.

2. Patients who, at operation, are found to have completely extruded discs obtain better results.

3. In this series, the results in women are not significantly worse than in men.

4. In women, the majority of disc lesions are located at the fourth lumbar interspace. Many of these were also associated with degenerative discs at the fifth lumbar interspace.

5. In men, there appears to be no such predilection for any given interspace.

6. Fourteen (12 per cent) of the patients had recurrence; ten of these were on the same side as the original lesion, and four were on the opposite side.

7. Twenty-eight patients (24.3 per cent) probably should not have had an operation performed.

It seems clear that if we remember the past history of low-back pain and sciatica and if we understand the pathological changes in the disc with the resulting altered function of the disc unit, we will limit surgery to those who continue to be disabled as a result of pain, or have progressive neurological alteration in spite of adequate conservative therapy. At the present time, it is our opinion that where conservative treatment fails and after careful study exploration reveals only a degenerated disc, spine fusion should be done. In those cases in which the operative finding is one of definite nerve-root compression, simple disc excision seems adequate. When the original operation fails or the patient has recurrence, fusion should be a part of the secondary operative procedure.