Percutaneous vertebroplasty (PVP), whereby polymethylmethacrylate cement is injected into the vertebral body (VB), has been used to successfully treat various spinal lesions. The mechanism responsible for the palliative effect of PVP is unknown, but it may be the result of neural damage caused by heat liberated during polymerization of the polymethylmethacrylate. The purpose of the current study was to measure in vitro temperature histories at three key locations (anterior cortex, center, spinal canal) in VBs injected with one of two different bone cements (Simplex P and Orthocomp) to determine the role temperature plays in PVP. Twelve VBs (T11-L2) from three elderly female spines were instrumented with thermocouples and injected with 10 cc of one of the two cements. Temperatures were measured with the VBs in a bath (37 degrees C) for 15 min after injection. A Student's paired t-test was used to determine differences in peak temperature and time above 50 degrees C between the two cement groups. Peak temperatures and temperatures above 50 degrees C were significantly higher and longer, respectively, at the center of VBs injected with Simplex P (61.8 +/- 12.7 degrees C; 3.6 +/- 2.1 min) than those injected with Orthocomp (51.2 +/- 6.2 degrees C; 1.3 +/- 1.4 min). There was no significant difference in peak temperature between cements at the spinal canal location; temperature there did not rise above 41 degrees C. Although thermal damage to intraosseous neural tissue caused by cement polymerization cannot be ruled out as a potential mechanism for pain relief experienced by patients subsequent to PVP, it seems unlikely based on the worst-case conditions tested in the current study.