Trabecular structure analysis using C-arm CT: comparison with MDCT and flat-panel volume CT

Purpose This paper assesses interscan, interreader, and intrareader variability of C-arm CT and compares it to that of flat-panel volume-CT (fpVCT) and high-definition multi-detector-CT (HD-MDCT). Methods Five cadaver knee specimens were imaged using C-arm-CT, fpVCT, and HD-MDCT. Apparent (app.) trabecular bone volume fraction (BV/TV), app. trabecular number (TbN), app. trabecular spacing (TbSp), and app. trabecular thickness (TbTh) of the proximal tibia were measured by three readers. Interreader, intrareader, and interscan variability for C-arm CT was expressed as coefficient of variation (CV), standard deviation (SD), and intraclass correlation coefficient (ICC). Results With the exception of app.TbSp (CV: 7.05–9.35%, SD: 0.06–0.09, ICC: 0.89–0.94), the variability of C-arm CT was low (CV: 2.41–6.43%, SD: 0.01–0.048, ICC: 0.65–0.98). Its interreader reliability (CV: 2.66–4.55%, SD: 0.01–0.03, ICC: 0.81–0.95) was comparable to that of HD-MDCT (CV: 2.41–4.08%, SD: 0.014–0.016, ICC: 0.95–0.96), and fpVCT (CV: 3.13–5.63%, SD: 0.009–0.036, ICC: 0.64–0.98) for all parameters except app.TbSp. Conclusions C-arm CT is a reliable method for assessing trabecular bone architectural parameters with the exception of app.TbSp due to spatial resolution limitation.