The aim of this study was: to assess the long-term efficacy and safety of percutaneous vertebroplasty (PVP) for treating painful vertebral osteoporotic fractures, and to estimate the risk of vertebral fracture in the vicinity of a cemented vertebra. A prospective open study was conducted. PVP were carried out between July 1995 and September 2000 for 16 patients with symptomatic osteoporotic vertebral fracture that had not responded to extensive conservative medical therapy. All the patients were followed-up for more than 1 year. The efficacy of the PVP was assessed by the changes over time in pain on Huskisson's visual analog scale (VAS) and on the McGill-Melzack scoring system (MGM). The efficacy of the procedure was also assessed by measuring the changes over time in quality of life assessed by the Nottingham Health Profile (NHP instrument): twenty-one vertebrae treated by PVP in 16 patients were evaluated. The mean duration of follow-up was 35 months. Pain assessed by the VAS significantly decreased from a mean of 71.4 mm+/-13 before PVP to 36 mm+/-30 after 6 months, and to 39 mm+/-33 at the time of maximal follow-up ( p<0.05 for both comparisons). The results were also significant for the MGM: 3.00+/-0.57 before PVP to 1.6+/-1.4 at the long-term follow-up ( p<0.05). The solely statistically significant decrease for quality of life was noted for pain. A slight but not significant improvement was noted for 3/6 dimensions of the NHP scores. A slight but significant increase in social isolation was also found. No severe complication occurred immediately after PVP. At the long term follow-up (35 months) there was a slight but not significantly increased risk of vertebral fracture in the vicinity of a cemented vertebra: odds ratio 3.18 (95% confidence interval (CI) 0.51-19.64). The odds ratio of a vertebral fracture in the vicinity of an uncemented fractured vertebra was 2.14 (95% CI: 0.17-26.31). In conclusion, PVP appears to be safe and effective for treating persistent painful osteoporotic fractures. Controlled studies with long-term follow-up are needed to evaluate the risk of vertebral fractures in the vicinity of a cemented vertebra.