Background Premature bone loss after childhood chemotherapy may be underestimated in patients with bone sarcoma. Methotrexate (MTX), a standard agent in osteosarcoma protocols, reportedly reduces bone mineral density (BMD). The literature, however, has reported cases of BMD reduction in patients with Ewing's sarcoma treated without MTX. Thus, it is unclear whether osteoporosis after chemotherapy relates to MTX or to other factors. Questions/purposes We therefore asked whether (1) young patients with a bone sarcoma had BMD reduction, (2) patients treated with MTX had lower BMD, and (3) other factors (eg, lactose intolerance or vitamin D deficiency) posed additional risks for low BMD. Methods We retrospectively reviewed 43 patients with malignancies who had dual-energy x-ray absorptiometry (DEXA) (lumbar, femoral); 18 with Ewing's sarcoma (mean age, 26 ± 8 years), and 25 with an osteosarcoma (mean age, 27 ± 10 years). The mean time since diagnosis was 8 ± 4 years in the group with Ewing's sarcoma and 7 ± 5 years in the group with osteosarcoma. At last followup we determined BMD (computing z-scores), fracture rate, and lifestyle, and performed serum analysis. Results BMD reduction was present in 58% of patients (37% had a z-score between À1 and À2 SD, 21% had a z-score less than À2 SD) in at least one measured site. Seven of the 43 patients (16%) had nontrauma or tumor-associated fractures after chemotherapy. Findings were similar in the Ewing and osteosarcoma subgroups. We found vitamin D deficiency in 38 patients (88%) and borderline elevated bone metabolism; lactose intolerance was present in 16 patients (37%). Conclusion Doctors should be aware of the possibility of major bone loss after chemotherapy with a risk of pathologic fracture. Vitamin D deficiency, calcium malnutrition, and lactose intolerance may potentiate the negative effects of chemotherapy, and should be considered in long-term patient management. Level of Evidence Level II, prognostic study. See Guidelines for Authors for a complete description of levels of evidence.