Effect of head size, head material, and radiation dose on the repeatability of CT-RSA measurements of femoral head penetration.

Abstract

The risk of early revision of total hip arthroplasty (THA) for polyethylene wear is now low, but there remains a need to perform wear measurements in patients for clinical surveillance. The gold standard of wear measurements has been radiostereometric analysis (RSA), which has limited availability. The use of computed tomography (CT) to perform THA wear measurement was described a decade ago and found to have acceptable accuracy and precision, but high radiation dose was a concern. Additionally, the use of larger femoral head sizes and ceramic femoral heads has risen in the past decade. The objectives of the study were to determine the effect of femoral head size, femoral head material, and lowered radiation dose on femoral head penetration measurement repeatability. A cadaveric hip was implanted with a cementless THA implant system. CT scans were acquired at a conventional radiation dose and at a reduced dose and repeated for a 32 mm and 36 mm cobalt-chromium femoral heads and a 32 mm ceramic femoral head. Apparent translation of the femoral head versus the acetabular cup was measured between the repeated scans using a CT-RSA software, where deviations from zero indicated measurement precision. The mean and standard deviation of translations in all planes was <0.200 mm. There was no effect for 3D translation of increasing cobalt-chromium head size (p = 0.2252). Cobalt-chromium heads had superior repeatability compared to ceramic heads at reduced dose (p = 0.022), but not at conventional dose (p = 0.1265). Further, superior repeatability was achieved with the reduced dose scan for the cobalt-chromium head (p = 0.0058), however there was no difference between doses for the ceramic head (p = 0.8148). CT-based wear measurement repeatability is excellent and consistent with prior literature even when implementing a larger femoral head, a ceramic femoral head, or reducing radiation dose to 25% of a conventional clinical scan.