Malnourished adolescents and young adults with anorexia nervosa (AN) are at high risk for skeletal deficits. To examine whether low-magnitude mechanical signals (LMMS) could preserve bone mineral density (BMD) throughout 6 months in adolescents and young adults with AN. This double-blind, sham-controlled randomized clinical trial, conducted in a hospital-based specialty clinic, assessed female adolescents and young women without medical comorbidity or medication use that would compromise bone health. A total of 837 female adolescents were screened from January 1, 2012, to December 31, 2019, of whom 317 met the study criteria. Data analysis was performed from 2020 to 2024. Platform delivering low-magnitude mechanical signals (LMMS) (0.3 g at 32-37 Hz) or sham (ie, placebo) signals for 10 minutes daily for 6 months. The primary outcome was trabecular volumetric BMD (vBMD) as measured by peripheral quantitative computed tomography of the tibia at baseline and 6 months. Secondary outcomes included cortical vBMD, cross-sectional area (CSA), areal BMD and body composition measured by dual-energy x-ray absorptiometry, and serum bone turnover markers. Forty female adolescents and young women (median [IQR] age, 16.3 [15.1-17.6] years; median [IQR] percentage median BMI for age, 87.2% [81.0%-91.6%]) completed the trial. Total bone vBMD changes were nonsignificant in both groups (95% CI for difference in median change between groups, -57.11 to 2.49): in the LMMS group, vBMD decreased from a median (IQR) of 313.4 (292.9-344.6) to 309.4 (290.4-334.0) mg/cm3, and in the placebo group, it increased from a median (IQR) of 308.5 (276.7-348.0) to 319.2 (309.9-338.4) mg/cm3. Total CSA at the 4% tibia site increased from a median (IQR) of 795.8 (695.0-844.8) mm2 to 827.5 (803.0-839.4) mm2 in the LMMS group, whereas in the placebo group, it decreased from 847.3 (770.5-915.3) mm2 to 843.3 (828.9-857.7) mm2 (95% CI for difference in median change between groups, 2.94-162.53). Median (IQR) trabecular CSA at the 4% tibia site increased from 616.3 (534.8-672.3) mm2 to 649.2 (638.0-661.4) mm2 in the LMMS group but decreased in the placebo group from 686.4 (589.0-740.0) mm2 to 647.9 (637.3-661.9) mm2 (95% CI for difference in median change between groups, 2.80-139.68 mm2). Changes in cortical vBMD, cortical section modulus, and muscle CSA were not significant between groups. The 6-month changes in trabecular and total bone CSA at the tibia 4% site (weight-bearing trabecular bone) were significantly different between groups (these measures increased in the LMMS group but decreased in the placebo group; total bone CSA: 95% CI, 2.94-162.53; P = .01; trabecular CSA: 95% CI, 2.80-139.68; P = .02). Greater increases in body mass index were seen in the placebo group (median [IQR] gain, 0.5 [-0.3 to +2.1]) than in the LMMS group (median [IQR] gain, +0.4 [-0.3 to +2.1]), perhaps due to differences in fat mass accrual. No adverse events occurred related to the LMMS intervention. In this randomized clinical trial of female adolescents and young women with AN, a 6-month LMMS intervention did not yield improvement in tibial trabecular vBMD. However, LMMS led to increases in total and trabecular CSA at the tibia. These results suggest an early positive response of increased bone turnover and trabecular bone quantity due to the LMMS intervention. Future studies should use a longer duration of intervention, consider strategies to optimize adherence, and potentially focus on a more profoundly malnourished patient population. ClinicalTrials.gov Identifier: NCT01100567.