Investigation of the thermal and tissue injury behaviour in microwave thermal therapy using a porcine kidney model

Abstract

Minimally invasive microwave thermal therapies are being developed for the treatment of small renal cell carcinomas (RCC, d< 3cm). This study assessed the thermal history and corresponding tissue injury patterns resulting from microwave treatment of the porcine renal cortex. Three groups of kidneys were evaluated: (1) in vitro treated, (2 in vivo with 2-h post-treatment perfusion (acute) and (3) in vivo with 7-day post-treatment perfusion (chronic). The kidneys were treated with an interstitial water-cooled microwave probe (Urologix, Plymouth, MN) that created a lesion centered in the renal cortex (50W for 10min). The thermal histories were recorded at 0.5cm radial intervals from the probe axis for correlation with the histologic cellular and vascular injury. The kidneys showed a reproducible 2cm chronic lesion with distinct histologic injury zones identified. The thermal histories at the edge of these zones were found using Lagrangian interpolation. The threshold thermal histories for microvascular injury and stasis appeared to be lower than that for renal epithelial cell injury. The Arrhenius kinetic injury models were fit to the thermal histories and injury data to determine the kinetic parameters (i.e. activation energy and frequency factor) for the thermal injury processes. The resultant activation energies are consistent in magnitude with those for thermally induced protein denaturation. A 3-D finite element thermal model based on the Pennes bioheat equation was developed and solved using ANSYS (V7.0). The real geometry of the kidneys studied and temperature dependent thermal properties were used in this model. The specific absorption rate (SAR) of the microwave probe required for the thermal modelling was experimentally determined. The results from the thermal modelling suggest that the complicated change of local renal blood perfusion with temperature and time during microwave thermal therapy can be predicted, although a first order kinetic model may be insufficient to capture blood flow changes. The local blood perfusion was found to be a complicated function of temperature and time. A non-linear model based on the degree of vascular stasis was introduced to predict the blood perfusion. In conclusion, interstitial microwave thermal therapy in the normal porcine kidney results in predictable thermal and tissue injury behaviour. Future work in human kidney tissue will be necessary to confirm the clinical significance of these results. © 2004 Taylor & Francis Ltd.