Percutaneous kyphoplasty (PKP) has been used to treat osteoporotic vertebral compression fractures for over 10 years; however, clinically speaking it is still controversial as to whether the use of unipedicular PKP or bipedicular PKP is best. Our study aimed to compare the different effects of unipedicular PKP and bipedicular PKP on the stiffness of compression fractured vertebral bodies (VBs), as well as to assess how cement distribution affect the bilateral biomechanical balance of the VBs. During this study, 30 thoracic VBs were compressed, creating vertebral compression fracture models; then they were augmented by unipedicular (group A and B) PKP and bipedicular (group C) PKP. In group A (unipedicular PKP), the cement was injected into one side and the augmentation was limited to the same side of the VB. In group B (unipedicular PKP), the cement was injected at only one side but the augmentation extended across the midline and filled both sides of the VB. In group C (bipedicular PKP), the cement was injected into both sides and thus achieved the bilateral augmentation. For the unipedicular PKP, the amount of cement injected was 15% of the original VB volume; while in bipedicular PKP, the amount of cement injected was a total of 20% of the original VB volume (10% was injected into each side). Using a MTS-858, we examined three phases of the VBs (intact, pre-augmented, post-augmented), by applying loads axially to the total vertebra and bilateral sides of the vertebra for each of three cycles, respectively. The changes of force and displacement were then recorded and the stiffness of the total vertebra and bilateral sides of the vertebra were calculated. For the pre-augmentation stage, the total VB stiffness of groups A, B and C significantly decreased when the compression fracture models were established (P < 0.05). After the cement augmentation (the post-augmentation stage), both groups A and B, showed that the stiffness could be restored to the initial, intact state; however, in group C, the stiffness was significantly higher than the initial, intact state (P < 0.01). The stiffness of the augmented side of group A was significantly higher than the non-augmented side (P < 0.001). In groups B and C, no significant differences were observed in the stiffness between total VB and each individual side. Thus, we can conclude that both unipedicular PKP and bipedicular PKP significantly increase the total VB stiffness. Bipedicular PKP creates stiffness uniformly across both sides of the vertebrae, while unipedicular PKP, creates a biomechanical balance depending on the distribution of cement. If bone cement is augmented only on one side, the stiffness of non-augmented side will be significantly lower than the augmented side, which might lead to an imbalance of stress on the VB. However, when cement augmentation crosses the midline, stiffness of both sides increase comparatively and biomechanical balance is thus achieved.