The Lifting Intervention For Training Muscle and Osteoporosis Rehabilitation for Men (LIFTMOR-M) trial examined efficacy and safety of an eight-month, supervised, high-intensity progressive resistance and impact training (HiRIT) program compared with machine-based isometric axial compression (IAC) training in middle-aged and older men with low areal bone mineral density (aBMD). The primary purpose of the current work was to explore change in thoracic kyphosis and incident fracture from vertebral morphology following eight-months of HiRIT or IAC training. The secondary purpose was to explore change in clinical kyphosis measures for HiRIT, IAC and a non-randomized, matched control group. Men (≥ 45yrs), with low aBMD, were recruited and randomized to HiRIT or IAC, or designated control. Clinical measures of thoracic kyphosis with inclinometry were determined. Cobb angle of kyphosis and vertebral fracture assessment using the Genant semi-quantitative method were determined from lateral thoracolumbar DXA (Medix DR, Medilink, France). Per-protocol (n = 40) and intention-to-treat (n = 93) analyses were conducted. Forty participants (HiRIT n = 20, IAC n = 20; 66.1 ± 7.8yrs.; lumbar spine T-score - 0.1 ± 0.8; femoral neck T-score - 1.5 ± 0.5) underwent clinical kyphosis measures and thoracolumbar DXA at baseline and follow-up. No between-group differences were detected in kyphosis change, however, within-group improvements in neutral (HiRIT - 2.3 ± 0.8°; IAC - 2.5 ± 0.8°) and 'standing tall' (HiRIT - 2.4 ± 0.8°; IAC - 2.0 ± 0.8°) postures were observed (p < 0.05). HiRIT improved Cobb angle (- 3.5 ± 1.5°, p = 0.027) from baseline. Over the 8months, no incident vertebral fractures nor progression of prevalent vertebral fractures occurred for HiRIT participants. Five incident fractures of thoracic vertebrae occurred for IAC and one wedge fracture progressed. Ninety-three participants underwent clinical kyphosis measures at both time-points (HiRIT n = 34, IAC n = 33, control n = 26). HiRIT exhibited a reduction in 'standing tall' kyphosis compared to control (- 2.3 ± 0.6° versus 1.4 ± 0.7°, p < 0.05), but no other between-group differences were detected. Although there was no difference in change between intervention groups, thoracic kyphosis appeared to improve in both HiRIT and IAC with exercise exposure. HiRIT improved 'standing tall' posture in comparison to usual activities. HiRIT was not associated with vertebral fracture progression or incident vertebral fracture, but for some IAC participants there was evidence of progression of vertebral fracture severity and incident vertebral fractures, in our small sample. Larger trials are required to confirm the observations of the current work, which was exploratory in nature.