The Clinical Follow-up Of Sandwich Level Fracture After Vertebral Cement Augmentation

[Objective]To explore the incidence and possible causative mechanism of SVF after vertebral cement infiltration, as well as the clinical indication for prophylactic cement reinforcement. A sandwich vertebra was referred to as an intact vertebral body that was located between two previously cemented levels.[Methods]From February 1, 2006 to December 31, 2010, four hundred and eighty-six patients with intractable painful skeletal diseases were treated with cement augmentation at our institution. After reviewing the postprocedural images of these 486 cases, patients with two adjoining septal operation levels (the middle intact level is so-called sandwich vertebra) were selected for further study. Sandwich vertebral fracture was identified by a hybrid method combining physical examination with image studies. A database was created containing patient gender, age, weight, height, body-mass index, follow-up duration, cumulative number of treated levels, TL junction or non TL junction (T11～L2), surgical approach, cement amount, cement disk leakage and pre- and postoperative kyphotic angulation of sandwich region. The rate of sandwich vertebral fracture was calculated and multiple logistic regression analysis was executed as well.[Results]The study was conducted on 42 patients. Three out of the 42 patients with underlying diseases other than benign osteoporotic vertebral compression fracture (OVCF) were excluded from the study (2 multiple myelomas, 1 metastatic pancreatic carcinoma), and one osteoporotic patient died of cardiovascular accident four months after the VP procedure, leaving 38 patients with 41 sandwich vertebral levels as the cohort for a long-term follow-up. The mean duration of follow-up after the procedure was 22.3 months (rang 9～47 months). The incidence of sandwich vertebral fracture per patients was 23.7% (9/38), while the incidence per sandwich levels was 22.0% (9/41). The gender, age, weight, height, body-mass index, follow-up duration, cumulative number of cemented levels, TL junction or non TL junction (T11～L2), surgical approach, cement amount and cement disk leakage of the patients did not correlate statistically with the risk of sandwich vertebral fracture. The local kyphotic angulation at the sandwich region (pre- and postoperative) with statistical significance was rendered for a multiple logistic regression. The result revealed that only preoperative kyphotic angulation at the sandwich region was positively associated with subsequent fracture of sandwich vertebra (OR=0.82, 95% CI=0.70～0.99, ward value=6.3, P=0.01).[Conclusions]The risk of sandwich vertebral fracture increases with the severity of local kyphotic angulation at the sandwich region. The sandwich vertebra, even though subjected to double load shifts, was not prone to structural failure compared with the osteoporosis natural history data.