Imaging of interstitial cryotherapy - An in vitro comparison of ultrasound, computed tomography, and magnetic resonance imaging

Rationale and objectives. To evaluate the imaging capabilities of ultrasoun d (US), computed tomography (CT), and magnetic resonance imaging (MRI) in m onitoring interstitial cryotherapy and to compare them with visual control. Methods. An experimental MR-compatible, vacuum-insulated and liquid nitroge n-cooled cryoprobe was inserted under ill vitro conditions into a porcine l iver, which was kept at a temperature of 37 +/- 1 degrees C, in a water bat h with continuous stirring. The freezing procedure was controlled macroscop ically, by US (Toshiba Sonolayer, 7.5-MHz linear array transducer), by CT ( Siemens Somatom Plus, slice thickness 2-8 mm, 165-210 mA at 120 kV), and by MRI (Philips Gyroscan ACS-NT, FFE TR/TE/FA = 15/5.4/25 degrees, T1-SE 550/ 20, T2-TSE 1800/100) after the iceball reached its maximum size. Results. The maximum iceball diameter around the probe tip was 12.0 mm by v isual control, 12.4 mm by US, 12.7 mm by CT, and within 12.8 mm by spin ech o sequences and 11 mm by gradient echo sequence. Due to the nearly signal-f ree appearance of the frozen tissue on MR images, the ice/tissue contrast o n T1-weighted and gradient echo images was superior to T2-weighted images a nd CT images. Sonographically, the ice formation appeared as a hyperechoic sickle with nearly complete acoustic shadowing. Conclusion. Due to the better ice/tissue contrast, T1-weighted or gradient echo MR images were superior to CT and US in monitoring interstitial cryoth erapy. Gradient echo sequences generally underestimated the ice diameter by 15%. (C) 1999 Academic Press.