Abstract

Introduction: Accurate templating prior to hip replacement requires that the magnification of the radiograph is known. This magnification is usually measured using a scale marker ball or disc of known diameter, but this method is inaccurate when the marker is not precisely positioned in the coronal plane of the hips.

Our aims were to design a novel scale marker which does not require such precise positioning, and to compare the accuracy of this new marker with a standard single ball marker.

Methods: The new marker consists of two separate markers: one behind the patient’s pelvis, the other at the front. It can be shown that the radiographic magnification of such markers is consistently related to the magnification of the hips.

The posterior marker consists of a 75x75cm square foam mat, incorporating multiple 25.4mm metal rods arranged in series down the centre. The anterior marker is made from five 25.4mm steel balls, linked in series at 20mm intervals. The mat is positioned just underneath the patient’s pelvis as they lie supine for their radiograph. The five balls are placed in the midline over the patient’s suprapubic region, and the x-ray is then taken. The radiographic dimensions of the ball and rod which are located between the hips are then measured. The magnification of the hips may then be calculated from these dimensions using a simple equation.

To validate the new “double” marker, it was compared with a conventional single marker ball. 74 hip arthroplasty patients undergoing routine radiographic follow up were recruited. Both the new double marker and the single marker were applied at the time of x-ray, the magnification according to each was calculated, and these were compared to the true radiographic magnification as determined from the known dimensions of the prosthesis. All markers were positioned by independent radiographers trained in their use.

Results: The correlation between true and predicted magnification was excellent using the double marker (r=0.90), but only moderate for the single marker (r=0.50). The median error of the single marker was 4.8%, but only 1.1% for the double marker (p< 0.001). The reliability of the double marker as a predictor of true magnification was very good (intraclass correlation coefficient, ICC=0.89), but was poor for the single marker (ICC=0.32). The accuracy of the double marker was unaffected by the patient’s body mass index. The inter and intraobserver variability of the new method were both excellent (ICC> 0.94).

Discussion: The double marker method is significantly more accurate and reliable than the single marker method when used in a clinical setting, as it does not rely on precise positioning of the marker by the user. We believe that this technique may become the gold standard method of calculating radiographic hip magnification in clinical practice.