Interpretation of hip fracture patterns using areal bone mineral density in the proximal femur

Bone mineral density scans are currently interpreted based on an average score of the entire proximal femur. Improvements in technology now allow us to measure bone density in specific regions of the proximal femur. The study attempts to explain the pathophysiology of neck of femur (NOF) and intertrochanteric/basi-cervical (IT) fractures by correlating areal BMD (aBMD) scores with fracture patterns, and explore possible predictors for these fracture patterns.

This is a single institution retrospective study on all patients who underwent hip surgeries from June 2010 to August 2012. A total of 106 patients (44 IT/basi-cervical, 62 NOF fractures) were studied. The data retrieved include patient characteristics and aBMD scores measured at different regions of the contralateral hip within 1 month of the injury. Demographic and clinical characteristic differences between IT and NOF fractures were analyzed using Fisher's Exact test and two-sample t test. Relationship between aBMD scores and fracture patterns was assessed using multivariable regression modeling.

After adjusted multivariable analysis, T-Troc and T-inter scores were significantly lower in intertrochanteric/basi-cervical fractures compared to neck of femur fractures (P = 0.022 and P = 0.026, respectively). Both intertrochanteric/basi-cervical fractures (mean T.Tot -1.99) and neck of femur fractures (mean T.Tot -1.64) were not found to be associated with a mean T.tot less than -2.5. However, the mean aBMD scores were consistently less than -2.5 for both intertrochanteric/basi-cervical fractures and neck of femur fractures. Gender and calcium intake at the time of injury were associated with specific hip fracture patterns (P = 0.002 and P = 0.011, respectively).

Hip fracture patterns following low energy trauma may be influenced by the pattern of reduced bone density in different areas of the hip. Intertrochanteric/basi-cervical fractures were associated with significantly lower T-Troc and T-Inter scores compared to neck of femur fractures, suggesting that the fracture traversed through the areas with the lowest bone density in the proximal femur. In the absence of reduced T.Troc and T.Inter, neck of femur fractures occurred more commonly. T-Total scores may underestimate the severity of osteoporosis/osteopenia and measuring T-score at the neck of femur may better reflect the severity of osteoporosis and likelihood of a fragility fracture.