Introduction Ablative procedures have become an integral part of tumor therapy for hepatocellular carcinomas (HCC) or liver metastases of colorectal carcinoma (LMCRC) within the last years. Complete abla-tion of the tumor is particularly important in preventing local recurrence. For this purpose, a real-time display of the tissue temperature is useful. The aim of this phantom trial is to investigate dual energy computer tomography (DECT) for non-invasive temperature measurement during the abla-tion process, especially the dependence of radiation exposure and the influence of reconstruction as virtual monoenergetic images (VMI) and blinded images.
Methods This experiment was performed as an ex vivo biophantom on pig livers in a setting similar to a possible clinical procedure. One microwave probe and four micro thermometers were placed at a distance of 1-4 cm from the probe. Microwave ablation (MWA) was performed for 5 minutes, followed by a 5-minute cooling phase. DECT images in 80 kVp/ 135 kVp were acquired every 30 seconds. This test was repeated five times with varying tube current (10 mA, 20 mA, 40 mA, 80 mA, 160 mA). The raw-data, the blended images in 100 kVe and 120 kVe and the VMI be-tween 40 keV- 135 keV were analyzed, comparing the attenuation with the invasively measured temperatures. Furthermore, the spatial resolution was analyzed.
Results All experiments showed a negative correlation of attenuation and temperature with a Pearson R of -0,89 to -0,71. The diagnostic accuracy for temperatures above or below 70 °C was calculated in all experiments with over 67,5 %. The temperature sensitivity (TempSS) showed a change in attenua-tion (HU) from -0,88 HU/ °C to -3,09 HU/ °C. There were no significant differences in correla-tion, diagnostic accuracy and TempSS with increasing radiation exposure or in between the recon-structions. The spatial resolution showed that a measurement accurate to 1 °C is possible in a layer thickness of 32 mm at 160 mA. With decreasing exposure, the required thickness increases up to 92 mm.
Summary The detection of temperature differences in low radiation doses is just as possible as in high radia-tion doses, but the spatial resolution decreases with lower exposure. The spatial resolution of DECT, even at high radiation dose, is not sufficient to monitor an ablation process in the range of a few centimeters. An optimization of the spatial resolution of DECT is necessary to establish it for non-invasive thermometry in the clinical setting.
Der beschriebene Biophantomversuch untersucht die Dual Energy Computertomography (DECT) im Hinblick auf ihre Fähigkeit Temperaturen während einer Mikrowellenablation zu messen. Insbesondere wurde die Abhängigkeit zur Strahlendosis und zu den unterschiedlichen Rekonstruktionsmethoden der DECT (virtuell monoenergetische Bilder, geblendete Bilder) untersucht. Die reine Detektion von Temperaturunterschieden in niedrigen Strahlendosen ist genauso möglich wie bei hohen Strahlendosen, jedoch sinkt die räumliche Auflösung mit zunehmender Reduktion der Exposition. Die räumliche Auflösung der DECT ist, selbst bei hoher Strahlenexposition, nicht ausreichend, um einen Ablationsvorgang im Bereich von wenigen Zentimetern zu überwachen. Eine Optimierung der räumlichen Auflösung der DECT ist notwendig, um diese für die nicht inva-sive Thermometrie im klinischen Setting etablieren zu können.
