SUN-395 Pelvic Bone Density Is Lower Than Bone Density of Hip and Femoral Neck in Postmenopausal Women

Abstract

Pelvic fractures represent 7% of all fragility fractures; they account for 5% of the cost of osteoporotic fracture care, and are commonly (> 50%) associated with loss of independence in the elderly. The incidence of pelvic fractures has increased significantly over the past 3 decades. However, little is known about the relationship between bone mineral density (BMD) and pelvic fractures. We have conducted a pilot cross-sectional study to establish a method of measuring pelvic BMD and to correlate BMD of the pelvis with BMD at other skeletal sites. Postmenopausal women without a history of pelvis and hip fragility fractures were enrolled. Hip, spine, and pelvis DXA scans were obtained using a Hologic DXA machine. Pelvic BMD was calculated using Hologic Research Software from 3 areas of the pelvis (R1: public symphysis, R2: inferior pubic ramus, and R3: superior pubic ramus), corresponding to common fracture locations. Pelvis BMD was the average of the 3 pelvis sites. Pelvic BMD measurement precision error was calculated using the root mean square method (Recommended by International Society of Clinical Densitometry (ISCD)). Statistical analysis was used to compare BMD at different sites. Alpha error was set at 0.05. Of 73 postmenopausal women who were enrolled in the study (average age 64 years, average 15 years postmenopausal), 3% had chronic kidney disease, 7% had type 2 DM, 3% were on corticosteroids and none were smokers. BMD of femoral neck assessed on pelvic DXA was not significantly different from femoral neck BMD measured on standard DXA (P=0.09). To assess pelvis BMD measurement precision, 15 patients underwent 3 separate pelvic DXA images after repositioning. BMD precision error was 0.011g/cm2 which is slightly lower than the precision total hip BMD at our center (0.007 g/cm2). BMD of R1, R2, and R3 pelvic areas were measured as 0.44±0.15, 0.41± 0.15, and 0.62 ±0.19 g/cm2, respectively. Notably, BMD of R3 was significantly higher than the other 2 areas (P<0.001, ANOVA). Average BMD (0.49±0.14 g/cm2) of pelvis was significantly lower than BMD of femoral neck (0.72± 0.16 g/cm2), total hip (0.86±17 g/cm2) and spine (0.97± 19 g/cm2)(P <0.001). Average BMD of pelvis was significantly lower in participants with osteopenia and osteoporosis of the hip and femoral neck compared to participants with normal BMD in those locations. In summary, we report a precise method of measuring BMD of commonly fractured areas of the pelvis. Pelvis BMD is lower than hip, femoral neck, and spine. Bone density of the pelvis correlates with hip and femoral neck bone density. The results of this pilot study can be used for future studies looking at pelvic low bone density in patients with pelvic fragility fractures which could help identify patients at risk for pelvic fragility fractures and change how osteoporosis is defined based on DXA images.