Femur head height error on AP pelvic x-rays with alterations of source-to-image distance, central ray position, and object-to-film distance

Objective: To identify those factors that produce leg-length insufficiency (LLI) projection errors on AP pelvic x-rays: altered source-to-image distance (SID), altered central ray (CR) position, or altered object-to-film distance (OFD). Method: Femur head height discrepancy as apparent LLI was measured using a simulated femur head model during altered source-to-image distance (SID) of 100 cm and 180 cm, and altered central ray (CR) projections at the level of the xiphisternum (X), umbilicus (U), at the femur head height (FH), and at the level of the femur heads with a tube tilt to the umbilicus level (FHT) on AP pelvic x-rays. Altered OFD of +3 cm, 0 cm, and -3 cm from neutral were created by manipulating the model. Results: The greatest LLI error was at the CR position of X and U with altered OFD of +3 cm and -3 cm, which produced LLI errors of up to 9 and 8.5 mm respectively. The SID of 100 cm resulted in greater LLI error across conditions than the SID of 180 cm. The condition with the lowest LLI error was the CR at FH with a SID of 180 cm. Conclusion: The use of standard AP full-spine x-ray centred at X, and the standard AP lumbo- pelvic x-ray centred at U, to determine LLI is invalid due to the possibility of high LLI error at these CR positions created by pelvic rotation. This work calls into question the validity of measuring LLI with x-ray unless a CR at FH is used with a SID of 180 cm.