Evaluating atypical features of femur fractures: How change in radiological criteria influenced incidence and demography of atypical femur fractures in a community setting†

Identification of atypical femoral fractures (AFFs) can be challenging. To assist in the radiological assessment, an ASBMR Task Force developed a case definition for AFFs in 2010, revising it in 2013. How the revised definition performs in a community setting compared to the 2010 definition is unknown. We applied the 2013 criteria to 372 femoral fractures that occurred between January 1, 1996, and June 30, 2009, employing two independent expert physician reviewers. We used radiographs that had been categorized in a previous study on the incidence of atypical fractures using 2010 ASMBR criteria (BEAK1). In this follow-up study (BEAK2), the same reviewers reviewed all previously identified femoral shaft fractures (FSFs) (N = 197) and distal femur fractures (N = 131) plus a 15% random sample of intertrochanteric fractures (N = 49). After initial review, agreement between the two reviewers ranged from 63-100% for specific features, and 84% of radiographs received the same overall classification. Fewer fractures met the 2013 compared to 2010 ASMBR case definition of AFFs (37 per 2013 criteria vs. 74 per 2010 criteria). Forty-three radiographs (58%) categorized as AFFs according to 2010 criteria were no longer AFFs when 2013 criteria were applied, and an additional 12 non-atypical FSFs according to 2010 criteria were reclassified as AFFs according to 2013 criteria. The major cause of AFF reclassification was the change in the definition of transverse configuration. The modification of the comminution, non-traumatic, and periosteal/endosteal thickness criteria resulted in the reclassification of non-atypical FSFs to AFFs. Incidence rate of AFFs according to 2013 ASBMR criteria was lower overall during the 13 years of observation than when the 2010 ASBMR criteria were applied, although we saw a slight increase starting in 2000. As in BEAK1, we found that those with AFFs were younger, more often female, and had a higher exposure rate to bisphosphonates than those with non-atypical FSFs. As we continue to unravel the demographics of those who suffer from AFFs, our study adds information about how the change in criteria influences epidemiological work. This article is protected by copyright. All rights reserved