OBJECTIVE To determine the frequency of phantom limb pain in pediatric patients with cancer-related amputations and the relationship between phantom limb pain and chemotherapy (CHRx). METHODS We retrospectively reviewed the medical records of all pediatric patients who underwent major limb amputations between 5 and 17 years of age during the period from Oct. 1, 1983, to Oct. 1, 1993, and were treated at the Mayo Clinic. METHODS One investigator reviewed 293 medical records and selected cases on the basis of specific inclusion and exclusion criteria. The study group consisted of 75 patients (67 with cancer-related and 8 with trauma-associated amputations). All reports of phantom limb pain and its temporal relationship to administration of CHRx were recorded. Data were analyzed by use of descriptive statistics. RESULTS Phantom limb pain was reported in 48% of patients with cancer-related amputations in comparison with 12% of patients with trauma-associated amputations. Among patients with cancer, phantom limb pain was experienced by 74% who were exposed to CHRx before or at the time of amputation, 44% who received CHRx after amputation, and 12% who never received CHRx. Patients with trauma-related amputations and those who did not receive CHRx reported phantom limb pain at similar intervals after amputation (a mean of 6 days). In comparison, 76% of patients with cancer and a history of exposure to CHRx reported phantom limb pain within 72 hours after amputation. Three of four patients who first received CHRx after amputation reported phantom limb pain on the day CHRx was begun (a mean of 12.3 days after amputation). CONCLUSIONS Our observations suggest that (1) phantom limb pain occurs more often in pediatric amputees with cancer than in those with trauma and (2) the hypothesis that CHRx may be a risk factor in the development of phantom limb pain in pediatric amputees should be investigated further.