Basivertebral foramen could be connected with intravertebral cleft: a potential risk factor of cement leakage in percutaneous kyphoplasty

BACKGROUND CONTEXT: Among different types of cement leakage in percutaneous kyphoplasty (PKP) for osteoporotic vertebral body compression fractures, leaks into the spinal canal are considered to be the most common complication. One potential structure causing this type of cement leakage is the potential connection between the basivertebral foramen and the intravertebral cleft, which is revealed clearly on magnetic resonance (MR) images, but is often ignored in the literature. PURPOSE: The purpose of this study is to assess the incidence rate of different types of cement leakage in PKP with or without intravertebral clefts and to determine whether the basivertebral foramen could be connected to the intravertebral cleft. STUDY DESIGN: This study is a retrospective assessment of the presence of an intravertebral cleft in osteoporotic vertebral bodies and the different types of cement leakage after PKP on radiographs, computed tomographic (CT) scans, and MR images. PATIENT SAMPLE: A total of 164 consecutive patients underwent PKP to treat 204 osteoporotic vertebral compression fractures. OUTCOME MEASURES: Outcome measures include the occurrence of different types of cement leakage in the groups with an intravertebral cleft and without intravertebral clefts. METHODS: A total of 204 vertebrae in 164 consecutive patients who underwent PKP to treat osteoporotic vertebral compression fractures were classified into two patterns based on preoperative radiographs, CT scans, and/or MR images of the treated levels: cleft pattern (with an intravertebral cleft in the vertebral body) and trabecular pattern (without intravertebral clefts). When an intravertebral cleft was identified, the investigators examined the basivertebral foramen and looked for a communication between the two structures on three-dimensional CT scans and MR images. On direct postoperative images, the patterns of cement leakage were classified as five types: type A, through a cortical defect into the paraspinal soft tissues; type B, through the basivertebral foramen; type C, via the needle channel; type D, through a cortical defect into the disc space; and type E, via the paravertebral vein. The association of the distribution of the cement leakage and the presence of an intravertebral cleft was analyzed retrospectively. Moreover, the association of type B leakage with the communication between the basivertebral foramen and the intravertebral cleft was also assessed. RESULTS: The average interobserver kappa values for determining the type of cement leakage and the presence of intravertebral cleft were 0.916 (range, 0.792-1) and 0.935, respectively. In 41 of 204 vertebrae (19.9%), an intravertebral cleft was confirmed on preoperative images. A communication between the intravertebral cleft and the basivertebral foramen was seen in 10 vertebrae (24.4%). Cement leakage was 36.2% in the group with a trabecular pattern and 41.5% in the group with a cleft pattern (p>.05). Leaks through the basivertebral foramen (type B; N=30, 14.7%) and through cortical defects into the disc space (type D; N=14, 6.9%) were more common than other types. Twenty of 163 vertebrae with the trabecular pattern (12.3%) and 10 of 41 vertebrae with the cleft pattern (24.4%) were identified as type B leaks, which reached statistical significance (p<.05). There was no statistical difference between the trabecular pattern and the cleft pattern on other types of leaks. CONCLUSIONS: Type B leaks are more common in vertebrae with an intravertebral cleft, which supports the presence of a connection between an intravertebral cleft and the basivertebral foramen. Thus, care must be taken when PKP is performed in these patients to avoid direct cement leakage into the spinal canal through the basivertebral foramen. (C) 2014 Elsevier Inc. All rights reserved.