PURPOSE: To compare different magnetic resonance imaging sequences for correct representation and discrimination between non-ablated, partially ablated, and completely ablated tissue after radiofrequency ablation (RFA) in the ex vivo, non-perfused bovine liver model. MRI is the method of choice for imaging of the induced coagulation zones ex vivo. However, the optimal imaging sequence has not been determined yet. MATERIALS AND METHODS: RFA, which is a method to locally destroy malignant cells with heat and which has been established for the treatment of non-resectable liver tumors, was applied to bovine livers. After the RF ablation procedure, the livers were sliced for visual inspection of the coagulation zones. As a result, non-ablated, partially ablated, and completely ablated tissues were distinguished. For the subsequent MR imaging, markers were used to indicate the borders of the different zones. For every imaging sequence the tissue representation at the site of these markers was classified as "exact", "overestimating", "underestimating" and "not determinable". Additionally, T 1, T 2, and proton density (PD) maps were measured. RESULTS: It was found that both T 1 and T 2 values gradually decrease from non-ablated to ablated tissue, with T 1 showing a stronger relative change compared to T 2. For PD no change was observed between non-ablated and partially ablated tissue and a clear decrease was detected for the completely ablated zone. CONCLUSION: Sequences with a strong PD influence resulted in the highest accuracy for discrimination between completely ablated and partially ablated tissue. The differences in image quality and contrast could be explained by the observed T 1, T 2 and PD changes. Georg Thieme Verlag KG Stuttgart, New York.
PURPOSE: To compare different magnetic resonance imaging sequences for correct representation and discrimination between non-ablated, partially ablated, and completely ablated tissue after radiofrequency ablation (RFA) in the ex vivo, non-perfused bovine liver model. MRI is the method of choice for imaging of the induced coagulation zones ex vivo. However, the optimal imaging sequence has not been determined yet. MATERIALS AND METHODS: RFA, which is a method to locally destroy malignant cells with heat and which has been established for the treatment of non-resectable liver tumors, was applied to bovine livers. After the RF ablation procedure, the livers were sliced for visual inspection of the coagulation zones. As a result, non-ablated, partially ablated, and completely ablated tissues were distinguished. For the subsequent MR imaging, markers were used to indicate the borders of the different zones. For every imaging sequence the tissue representation at the site of these markers was classified as "exact", "overestimating", "underestimating" and "not determinable". Additionally, T 1, T 2, and proton density (PD) maps were measured. RESULTS: It was found that both T 1 and T 2 values gradually decrease from non-ablated to ablated tissue, with T 1 showing a stronger relative change compared to T 2. For PD no change was observed between non-ablated and partially ablated tissue and a clear decrease was detected for the completely ablated zone. CONCLUSION: Sequences with a strong PD influence resulted in the highest accuracy for discrimination between completely ablated and partially ablated tissue. The differences in image quality and contrast could be explained by the observed T 1, T 2 and PD changes. Georg Thieme Verlag KG Stuttgart, New York.