Development and validation of a novel overhead method for anteroposterior radiographs of fractured rat femurs.

We aimed to establish a new method of obtaining femur anteroposterior radiographs from live rats. We used five adult male Sprague-Dawley rats and created a femoral fracture model with an 8 mm segmental fragment. After the surgery, we obtained two femoral anteroposterior radiographs, a novel overhead method, and a traditional craniocaudal view. We obtained the overhead method three times, craniocaudal view once, and anteroposterior radiograph of the isolated femoral bone after euthanasia. We compared the overhead method and craniocaudal view with an isolated femoral anteroposterior view. We used a two-sample t-test and intraclass correlation coefficient (ICC) to estimate the intra-observer reliability. The overhead method had significantly smaller differences than the craniocaudal view for nail length (1.53 ± 1.26 vs. 11.4 ± 3.45, p < 0.001, ICC 0.96) and neck shaft angle (5.82 ± 3.8 vs. 37.8 ± 5.7, p < 0.001, ICC 0.96). No significant differences existed for intertrochanteric length/femoral head diameter (0.23 ± 0.13 vs. 0.23 ± 0.13, p = 0.96, ICC 0.98) or lateral condyle/medial condyle width (0.15 ± 0.16 vs. 0.13 ± 0.08, p = 0.82, ICC 0.99). A fragment displacement was within 0.11 mm (2.4%). The overhead method was closer to the isolated femoral anteroposterior view and had higher reliability.