Forty-five patients who had residual symptoms after lumbar spine surgery were re-evaluated and underwent additional surgery. After a minimum of 2 years' follow-up, these patients' results were reviewed to determine what variables might predict long-term outcome. Age, number of previous operations, and psychological diagnosis were not statistically significant, but a non-compensable injury, ability to return to work after surgery, a negative history of litigation, and achieving a solid fusion were statistically significant in predicting a good outcome. In this study of 45 patients who had repeat surgery, 82% were improved at an average follow-up of 28.2 months. Intrathecally enhanced computed tomography, magnetic resonance imaging, discography, and computed tomography-discography are required to thoroughly evaluate the previously operated lumbar spine, because a single imaging study showed surgical abnormalities in only 61% of the patients in this study.