P32. Can CT-based bone mineral density predict proximal junctional fracture?

Abstract

BACKGROUND CONTEXT Proximal junctional fractures (PJFr) can be a catastrophic complication associated with adult spinal deformity surgery. Presence of osteoporosis has been shown to be one of the risk factors for PJFr. Dural-Energy X-Ray Absorptiometry (DEXA) scans are standard screening tools for diagnosis of osteoporosis. However, spinal DEXA scores have been reported to be flawed with regional sclerosis associated with spinal degeneration representing a possible confounder to the study. Recently, investigators have reported on the use of a computed tomography (CT) scan surrogate in place of spinal DEXA scores. PURPOSE The primary focus of this study is to investigate whether examining CT based assessments of bone mineral density (BMD) preoperatively can predict the possibility of PJFr. STUDY DESIGN/SETTING Retrospective two-center study with review of adult spinal deformity patients (consecutive surgeries between 2012 and 2015) who underwent instrumented fusions to the pelvis encompassing at least 5 fusion levels. PATIENT SAMPLE n=92 patients. OUTCOME MEASURES Using CT scans we calculated the mean sagittal and axial BMD values in order to predict the risk possibility of PJFr. METHODS Reviewed and documented patients´ demographics, and measured preoperative, postoperative, and final follow up radiographic parameters. In CT scans, mean sagittal and axial BMD values were calculated by defining a region of interest within vertebral. Vertebra at most cranial end of final construct defined as upper instrumented level (UIV). Preoperative BMD values were measured at UIV, UIV+1(proximal), and UIV -1(distal). The measured BMD values were reported in Hounsfield units (HU).Patients were categorized into three groups: no proximal junctional deformity (No PJD, control group), proximal junctional kyphosis (PJK), and PJFr groups. BMD values, demographics and radiographic variables of PJK and PJFr were compared with control group, separately. ANOVA tests used for continuous variable and Chi square tests for categorical variables. RESULTS Study sample n= 92, mean age 64 yrs (42 to 81), mean follow up 1.5 yrs (3 months to 4 years), No PJD group (n=59), PJK group (n=22), PJFr group (n=11). Significantly low preoperative BMD values were noticed for fracture group in all sagittal views at UIV+1 (0.007), UIV (0.04) and UIV-1 (0.05) and in all axial views at UIV+1 (0.02), UIV (0.03), except UIV -1 (0.19) UIV+1, compared to control group. No significant differences in BMD values were noticed for PJK group compared to control group. Osteoporosis history was not statistically different between fracture and control groups (9% vs 20%, p=0.41). CONCLUSIONS Adult spinal deformity patients, who undergo fusions to the pelvis with at least 5 fusion levels, without osteoporosis history appear to be at risk for fractures at proximal junctional vertebras with lower BMD measurements. BMD measurements can be helpful for counseling of patients regarding the risk of fractures, they may also be used for planning of the proximal extent of constructs as well as potential preventative surgical methods. FDA DEVICE/DRUG STATUS computed tomography (CT) scans (Approved for this indication), Dural-Energy X-Ray Absorptiometry (DEXA) scans (Approved for this indication)