MRI Only Occult Geriatric Hip Fractures: Is Displacement Common with Nonoperative Treatment?

Purpose: Geriatric fractures of the proximal femur are common and typically warrant surgical stabilization. Surgery is also typically advocated for patients who have a hip fracture not evident on orthogonal radiographs or computed tomography (CT), but display increased bone edema within the femoral neck or intertrochanteric regions with magnetic resonance imaging (MRI). However, it is not known how often these fractures displace when treated nonoperatively. The purpose of our study was to report the rate of displacement requiring secondary surgery in nonoperatively treated occult geriatric hip fractures.

Methods: All nonoperatively treated femoral neck or intertrochanteric femur fractures (AO/OTA 31A and 31B) at our institution from 2003 to 2016 were identified using an institutional geriatric hip fracture database and closed treatment CPT code search. Patients older than 65 with no evident fracture on radiographs or CT but increased osseous edema on MRI consistent with an occult fracture were included (Figure 1). Patients who died prior to evidence of fracture displacement or radiographic and clinical evidence of healing were excluded. The primary outcome measure was fracture displacement necessitating surgery.

Results: 13 patients met the inclusion and exclusion criteria. There were six femoral neck fractures and seven intertrochanteric fractures. The mean age of the cohort was 84.4 years (range, 70 - 91 years) and 69% (9/13) were female. Treatment typically consisted of restricted weight bearing which varied in duration based on the treating surgeon.

Of the entire cohort, two fractures displaced necessitating surgery (2/13; 15.4%). Both of the displaced fractures occurred in the femoral neck cohort (2/6; 33%) compared to none in the intertrochanteric group (0/7; 0%). The difference in displacement between fracture locations was not significantly different (p=0.19).

Conclusions: In our study of nonoperative geriatric femoral neck and intertrochanteric femur fractures detected by MRI only, 15% of fractures displaced necessitating surgery. Both displacements occurred in the femoral neck group. Both patients had subsequent surgical treatment (one compression hip screw, one total hip arthroplasty) and went on to recover without further complication. Our study represents the largest investigation detailing the natural history of nonoperatively treated occult geriatric hip fractures. However, no definitive treatment recommendations can be made due to our small cohort size. We recommend that surgeons continue to manage occult hip fracture patients utilizing their preferred treatment algorithm.