Evaluation of cementless femoral stem level on mediolateral projection radiographs.

OBJECTIVE: To evaluate the accuracy of measuring cementless femoral stem level on mediolateral projection radiographs. STUDY DESIGN: Benchtop cadaveric. SAMPLE POPULATION: Twelve canine cadaver femurs. METHODS: Cementless femoral stems were inserted into 12 canine cadaver femurs at three levels of subsidence. Mediolateral radiographs were obtained for each femur at 0 degrees , 10 degrees , and - 10 degrees frontal plane angulation and at 0 degrees , 15 degrees , 30 degrees , -15 degrees , and - 30 degrees axial plane rotation. Stem level was measured physically on specimens and on radiographs, and a proportion was used to calculate corrected stem level. Stem level was assessed relative to the greater trochanter and relative to the intertrochanteric fossa. Analysis of variance tests were used to compare actual, radiographically measured, and corrected stem level. RESULTS: No differences were detected between radiographically measured and actual stem level relative to the greater trochanter at 0 degrees , 15 degrees , 30 degrees , and - 30 degrees axial rotation with 0 degrees frontal plane angulation; introduction of 10 degrees or - 10 degrees frontal plane angulation resulted in differences between radiographically measured and actual stem level. Errors >0.5 mm were observed in 82% of radiographic measurements on the basis of the intertrochanteric fossa. The use of a corrective proportion did not improve the accuracy of radiographic measurements. CONCLUSION: Femoral stem level was accurately quantitated on mediolateral projection radiographs in this cadaver model. Frontal plane angulation distorted this measurement. CLINICAL SIGNIFICANCE: Femoral stem subsidence may be assessed on well-positioned mediolateral projection radiographs if the landmarks are visible.