Osteocytes are recognized as having a pivotal role in bone tissue homeostasis, and stimuli that increase osteocyte death result in decreased bone tissue quality. Previous in vitro studies have shown that mechanical stimulation prevents osteocyte death; however, in vivo evidence of this protective effect is limited. The aim of this study was to investigate if mechanical stimulation provided by voluntary exercise reduces osteocyte death caused by estrogen deficiency. Thirty-two female Wistar rats (5 months old) were either sacrificed as baseline controls (BSL, n = 7), ovariectomized or sham-operated and housed in cages with a voluntary running wheel (OVXEX, n = 7; SHAMEX, n = 6), or ovariectomized or sham-operated and housed in standard cages of equivalent size (OVXSED, n = 6; SHAMSED, n = 6) and sacrificed at age 14 months. Histomorphometric analysis of femur mid-diaphysis cortical bone revealed a significantly higher osteocyte number (N.Ot) and lower empty lacunae number (N.Lc) in both the OVXEX and SHAMEX groups compared to their SED counterparts. Intracortical porosity (Po.Ar) was also lower in both EX groups compared to their SED counterparts and significantly correlated with N.Lc (r = 0.616; P < 0.001). Three-point bending testing showed a significantly higher Young's modulus and ultimate stress in OVXEX compared to OVXSED and significant correlations between N.Lc and both yield stress (r = -0.376, P < 0.05) and ultimate stress (r = -0.369, P < 0.05) and between intracortical porosity and bone ultimate stress (r = -0.451, P < 0.05). Our results show that voluntary exercise prevented osteocyte death and that this protective effect was associated with increases in femur ultimate stress, which could be partially explained by decreases in Po.Ar.