We performed a prospective pilot study of subarachnoid anesthesia for kyphoplasty in 11 nonrandomized patients. Subarachnoid anesthesia was administered at the level of the best palpable intervertebral space below L3. Patients received intrathecally either hyperbaric or plain bupivacaine with or without fentanyl. Five patients experienced pain during the surgical procedure and received supplemental IV analgesia. One patient felt pain from the pressure on the ribs while in the prone position. The remaining patients were comfortable. In no patient was respiratory compromise or deep sedation observed. We conclude that subarachnoid anesthesia may be an adequate technique for kyphoplasty.