Although continuous infusion of local anesthetic through nerve sheath catheters provides excellent pain control after amputation, the influence of this technique on the incidence of phantom limb pain is controversial. The current retrospective study examined the influence of continuous nerve sheath catheter analgesia and primary anesthetic technique (general or regional anesthesia) on the incidence of phantom limb pain. After institutional review board approval, data were gathered on patients who had amputation from 1990 to 1999. Medical records were reviewed to determine the level of and indication for amputation, age at time of amputation, current disease status, date of diagnosis and surgery, and primary anesthetic technique. Six months after amputation, preoperative pain, phantom limb sensations, and phantom limb pain were assessed using a verbal scale ranging from 0 (no pain) to 10 (worst pain) and methods of pain control also were evaluated. Thirty-nine patients completed the study. The incidence of phantom limb pain (visual analog scale score > or = 3) was 67% and was lower than the historic incidence (80%) before the use of this analgesia technique. Incidence of phantom limb pain was higher for patients requiring proximal versus distal amputations. The primary anesthetic technique (general versus regional) did not affect the incidence of phantom limb pain. Long-term followup showed that the incidence of phantom limb pain in patients receiving continuous nerve sheath catheter infusion is lower than previously reported for patients who had amputation for oncologic indications.