The aim of this study was to investigate if the distribution of subchondral volumetric bone mineral density (vBMD) from peripheral quantitative computed tomography (pQCT) is related to estimates of knee joint loads calculated during walking gait in healthy young people. We recruited 19 young (age 18–40 years) healthy people with no self-reported knee pain or pathology. For all participants we collected two forms of data:​ (1) pQCT data at 2% of tibia length (from the proximal joint line) using a Stratec XCT3000 scanner at 0.2 × 0.2 mm in plane resolution;​ and (2) indices of joint loading, specifically external joint moment, at the indexed knee during walking gait. Joint moments were calculated from motion capture and ground reaction force data. pQCT scans were performed immediately prior to gait analysis. A sub-group of 9 participants attended a second scanning session to establish the reproducibility of the pQCT workflow. vBMD was extracted for four sub-regions (anteromedial, anterolateral, posteromedial and posterolateral). Reproducibility of the pQCT workflow was good to excellent (ICCs 0.832–0.985) with minimal detectable differences ranging from 2.3–39.5 mg HA/cm³. Significant independent correlations were identified between the external rotation moment and the medial-to-lateral (r = 0.517), posteromedial-to-posterolateral (r = 0.627) and posteromedial-to-anterolateral (r = 0.518) vBMD ratios, and between the knee adduction moment and the medial-to-lateral (r = −0.476) and posteromedial-to-posterolateral (r = −0.497) vBMD ratios. There appear to be significant relationships between measures of vBMD from pQCT and indices of joint loading in healthy people. These data are the first to combine imaging at the resolution available with pQCT and indices of joint loading in the same cohort.