Ultrasound and magnetic resonance imaging for the detection of blood: An ex-vivo study.

Abstract

Early detection of bleeding into a joint is crucial in patients with haemophilia. This study was designed to evaluate the sensitivity of ultrasonography (USG) and magnetic resonance imaging (MRI) to detect the presence of blood in small concentrations in a simulated model to mimic joint bleeding. Different concentrations of blood in plasma, varying from 0.1% to 45%, were collected in 10-ml plastic syringes and imaged using 12 and 18MHz USG transducers and with 1.5T and 3T MRI scanners, at different intervals of time following dilution. The images were scored for the presence of blood by four experienced radiologists who were blinded to the concentration of blood. Within the first 2h, the 18MHz transducer was able to detect blood consistently up to 0.5%, whereas the 12MHz transducer could consistently identify blood up to 1.4%. After the first 12h, both transducers were able to detect blood up to 0.5% concentration. However, at concentrations below 0.5%, there was discordance in the ability to detect blood, with both transducers. There was no correlation between the signal intensities of MRI images and concentration of blood, at different time intervals, irrespective of the magnetic field strength. Detection of blood using the USG is dependent on variables such as the concentration of blood, frequency of the transducer used and timing of the imaging. As the concentration of blood decreases below 0.5%, the discordance between the observers increases, implying that the detection limit of USG affects its reliability at lower concentrations of blood. Caution is urged while interpreting USG imaging studies for the detection of blood in symptomatic joints.