The use of neuraxial (intrathecal and epidural) analgesia has been suggested in treatment guidelines put forth for the treatment of refractory cancer pain. We review the literature and present our algorithm for using neuraxial analgesia. We also present our outcomes using this algorithm over a 28-month period. We used neuraxial analgesia in 87 of 4,107 patients, approximately 2% of those seen for pain consultation. Evaluation of those patients at an 8-week follow-up revealed improved pain control. After institution of neuraxial analgesia, there was a significant reduction in the proportion of patients with severe pain (defined as a "pain worst" score in the severe range of 7-10), from 86% to 17%, noted to be highly statistically significant. At follow-up, numerical pain scores decreased significantly from 7.9 +/- 1.6 to 4.1 +/- 2.3. No difference was noted between the intrathecal and epidural groups. Oral opioid intake after instituting neuraxial analgesia revealed a significant decrease from 588 mg/day oral morphine equivalents to 294 mg/day. At follow-up, self-reported drowsiness and mental clouding (0-10) also significantly decreased from 6.2 +/- 3.0 and 5.4 +/- 3.4 to 3.2 +/- 3.0 and 3.1 +/- 3.0, respectively. This retrospective review shows promising efficacy of neuraxial analgesia in the context of failing medical management.