Although nutritional status may adversely affect various health outcomes, the relationship between anthropometry and outcomes after hematopoietic cell transplantation (HCT) has not been fully studied in children. We analyzed the impact of pre-HCT body mass index (BMI), arm muscle area, and arm fat area on outcomes in 733 patients age 2-18 years who underwent allogeneic HCT for a hematologic malignancy between 1985 and 2009. We evaluated these 3 variables according to patient group based on age- and sex-adjusted percentiles for BMI, arm muscle area (<5th, 5th-24th, 25th-94th, and ≥95th), and arm fat area (<25th, 25th-94th, and ≥95th). Cox proportional hazards regression models for event-free survival (EFS), relapse, and nonrelapse mortality (NRM) at 100 days and 3 years after HCT, as well as grade II-IV acute graft-versus-host disease (GVHD) and chronic GVHD, were performed using the 3 major variables and adjusted for covariates. BMI was <5th percentile in only 3% of patients and ≥95th percentile in 15% of patients, but outcomes for both groups were similar to those for the BMI 25th-94th percentile group. The BMI 5th-24th percentile group had lower EFS (P = .01) and higher relapse (P = .003) at day +100 post-HCT, but these associations did not hold at 3 years post-HCT. Arm muscle area was <5th percentile in 8% of patients, and arm fat area was <25th percentile in 10%. Analysis of arm muscle area showed that the <5th percentile group had lower EFS and higher NRM and relapse rate at day +100 (P = .002, .04, and .01, respectively) and 3 years (P = .0004, .008, and .01, respectively) post-HCT. Arm fat area <25th percentile was associated with lower EFS at day +100 (hazard ratio, 1.5; P = .05), but not at 3 years post-HCT. Anthropometry variables were not associated with acute or chronic GVHD. In conclusion, arm muscle area <5th percentile appears to be a stronger predictor than BMI of poor outcomes after HCT in children with hematologic malignancies.