Does the Thermal Damage Estimate Correlate With the Magnetic Resonance Imaging Predicted Ablation Size After Laser Interstitial Thermal Therapy?
Abstract BACKGROUND Magnetic resonance guided laser induced thermal therapy (LITT) is a minimally invasive method to treat a wide range of intracranial pathologies. The Arrhenius model is used to generate a thermal damage estimate (TDE) predicting ablation extent. OBJECTIVE Evaluation and correlation of the TDE to magnetic resonance imaging (MRI)-estimated ablation extent in human cases. METHODS The Medtronic Visualase system (Medtronic Inc, Dublin, Ireland) was utilized. Postablation axial T1-contrast enhanced images were acquired and intraoperative TDE image was obtained from the Visualase console. OsiriX DICOM Viewer (Pixmeo Inc, Bernex, Switzerland) was utilized to calculate cross-sectional area on MRI. ImageJ (National Institutes of Health, Bethesda, Maryland) was utilized for TDE area. Two blinded raters performed all measures. Statistical testing included Pearson correlation and the Student's t-test. RESULTS Twenty-two cases including tumor and epilepsy were evaluated. Average MRI predicted tumor ablation area was 4.72 ± 2.22 cm2 and average predicted epilepsy ablation area was 4.12 ± 1.89 cm2. Average tumor TDE was 4.02 ± 1.95 cm2 and average epilepsy TDE was 4.36 ± 2.21 cm2. Rater 1’s ablation areas and TDEs correlated with r = 0.89 (P < .0001) and no significant difference (P > .5). Rater 2’s ablation areas and TDEs correlated with r = 0.91 (P < .0001) and no significant difference (P > .7). Rater 1 vs Rater 2 showed a strong correlation for TDE (r = 0.98, P < .000001) and ablation area (r = 0.96, P < .0001) and no significant difference (P > .5). CONCLUSION The TDE is an accurate and reliable measure of ablated area in LITT in human cases as assessed on postoperative MRI. Future studies should be larger and assess reliability of the TDE when multiple lasers and planes are used.