OBJECTIVES: The purpose of this study is to compare the reliability and accuracy of the screw protuberance method (SPM) and overlay method (OM) for measuring femoral neck shortening on anterior-posterior (AP) radiographs. The secondary aim is to investigate the changes in reliability and accuracy with varying femoral rotation. METHODS: Radio-opaque femur sawbone models were fitted with either 3 cancellous screws or a sliding hip screw implant. Anterior-posterior radiographs were obtained using C-arm fluoroscopy with femoral neck shortening up to 15 mm and with the femoral shaft in 30 degrees of internal rotation to 30 degrees of external rotation (ER). Four observers measured femoral neck shortening at 2 time points. Intraobserver and interobserver reliability were calculated using the intraclass coefficient. Accuracy was analyzed through a Bland-Altman agreement statistic stratified by femoral rotation. RESULTS: Both measurement techniques displayed excellent reliability, regardless of femoral rotation or implant. There was a significant difference in femoral neck shortening measurements with rotation for both the OM (P < 0.001) and SPM (P < 0.001). Both methods are accurate within 1 mm of the actual magnitude of shortening from 30-degree internal rotation to 15-degree ER. At 30-degree ER, shortening was underestimated by -2.10 mm using the OM (95% confidence interval, -2.43 to -1.76; P < 0.01) and by -1.64 mm using the SPM (95% confidence interval, -1.83 to -1.45; P < 0.01). CONCLUSION: This study demonstrates that both the OM and SPM are accurate and reliable assessments for femoral neck shortening; however, both methods are sensitive to extreme ER. Given the simplicity of the SPM technique, it may have increased utility for pragmatic research studies.
OBJECTIVES: The purpose of this study is to compare the reliability and accuracy of the screw protuberance method (SPM) and overlay method (OM) for measuring femoral neck shortening on anterior-posterior (AP) radiographs. The secondary aim is to investigate the changes in reliability and accuracy with varying femoral rotation. METHODS: Radio-opaque femur sawbone models were fitted with either 3 cancellous screws or a sliding hip screw implant. Anterior-posterior radiographs were obtained using C-arm fluoroscopy with femoral neck shortening up to 15 mm and with the femoral shaft in 30 degrees of internal rotation to 30 degrees of external rotation (ER). Four observers measured femoral neck shortening at 2 time points. Intraobserver and interobserver reliability were calculated using the intraclass coefficient. Accuracy was analyzed through a Bland-Altman agreement statistic stratified by femoral rotation. RESULTS: Both measurement techniques displayed excellent reliability, regardless of femoral rotation or implant. There was a significant difference in femoral neck shortening measurements with rotation for both the OM (P < 0.001) and SPM (P < 0.001). Both methods are accurate within 1 mm of the actual magnitude of shortening from 30-degree internal rotation to 15-degree ER. At 30-degree ER, shortening was underestimated by -2.10 mm using the OM (95% confidence interval, -2.43 to -1.76; P < 0.01) and by -1.64 mm using the SPM (95% confidence interval, -1.83 to -1.45; P < 0.01). CONCLUSION: This study demonstrates that both the OM and SPM are accurate and reliable assessments for femoral neck shortening; however, both methods are sensitive to extreme ER. Given the simplicity of the SPM technique, it may have increased utility for pragmatic research studies.