Percutaneous kyphoplasty: Risk Factors for Recollapse of Cemented Vertebrae

Percutaneous kyphoplasty can offer pain relief and restoration of vertebral height immediately after the procedure; however, little is known about how many vertebrae recollapse during follow-up or why recollapse occurs. In the present study, we define recollapse of a treated vertebra, assess how common it is following percutaneous kyphoplasty, and investigate risk factors for the condition. In total, 203 consecutive patients who underwent percutaneous kyphoplasty were reviewed after an average 12.7 months to assess what proportion of cement-augmented vertebrae had recollapsed. Potential risk factors for recollapse included age, gender, body weight, body height, body mass index, treated level, duration of symptoms, follow-up duration, preoperative T-scores, surgical approach, the intravertebral cleft, contact of polymethyl methacrylate (PMMA) with endplates, cement volume, cement leakage, and midline vertebral body height. Stepwise multivariate linear regression was conducted to predict recollapse as quantified by midline vertebral height loss. Overall, 38.9% of the augmented vertebrae recollapsed. In the recollapse group, the average midline vertebral height ratio and kyphotic angles statistically significantly changed during follow-up (P < 0.05). Pain scores decreased immediately after percutaneous kyphoplasty and generally remained low at follow-up. Significant predictors of midline vertebral height loss at follow-up included presence of an intravertebral cleft, postoperative vertebral height, and non-PMMA-endplate-contact. Together, these factors accounted for 28% of the variability in midline height loss. Benefits of percutaneous kyphoplasty are partly offset by subsequent recollapse. Recollapse is greater if there is an intravertebral cleft, non-PMMA-endplate-contact and an increase in the post vertebral height.