Evaluation of a novel method to calculate cementless femoral stem level on craniocaudal projection radiographs.

OBJECTIVE To assess the accuracy of a proportion-corrected radiographic cementless femoral stem level calculation in a cadaver model. STUDY DESIGN Cadaveric study. SAMPLE POPULATION Twelve cadaveric canine femurs. METHODS Cementless femoral stems were implanted in 12 cadaver femurs at three stem levels. Craniocaudal projection radiographs of the implanted femurs were obtained at 0°, 10°, 20°, 30°, and 45° angulations in the sagittal plane (relative to the radiographic plate) and at 0°, 15°, and 30° internal and external axial rotation. A proportional calculation including actual stem length, radiographically measured stem length, and radiographically measured stem level relative to the greater trochanter (uncorrected stem level) was solved to yield a corrected stem level. The errors of the stem level measurements were compared. RESULTS With 0° or 15° of internal rotation, the true stem level differed by more than 0.5 mm from corrected stem level in 2 of 30 tested femoral positions. It differed from the uncorrected stem level in 10 of 30 positions. With 15° and 30° external rotation or 30° internal rotation, the corrected stem level was not more accurate than the uncorrected stem level for any femoral position. CONCLUSION Use of a proportion correction minimized variability in the stem-level measurements due to sagittal angulation (up to 30°) of the femur at 0° or 15° of internal rotation. CLINICAL SIGNIFICANCE A proportion correction can improve the accuracy of femoral stem-level assessment on craniocaudal projection radiographs, which may aid in the detection of subsidence after cementless total hip replacement.