PURPOSE: We investigated whether cell-kill modelling could be used as a mean for predicting the outcome of microwave thermotherapy for benign prostate hyperplasia (BPH). METHODS: The two models--Henriques' damage integral and Jung's compartment model--were implemented in a computer program. Real treatment data for 22 patients with BPH who were in chronic retention were used as input, including measured intraprostatic temperatures and microwave power. To test if modelling gives results that are consistent with actual observations, comparison with transrectal ultrasound (TRUS) measurements of the prostate volume before and after treatment was made. The sensitivity of the computer model for variations in the heat cytotoxicity and the temperature probe location in the adenoma was also tested. RESULTS: The average TRUS volume reduction 3 months after treatment was 26 cc, whereas the corresponding cell kill calculation was 27 cc. The computer model appears to be rather insensitive to minor uncertainties in heat sensitivity and location of the intraprostatic reference temperature sensors. CONCLUSION: Cell-kill modelling appears to give results that are consistent with actual observations. The coagulated tissue volume is calculated in real time during the treatment, thereby providing an immediate prediction of the treatment outcome. By using cell-kill modelling, the endpoint of a treatment can be set individually; e.g., when a certain volume reduction has been achieved.
PURPOSE: We investigated whether cell-kill modelling could be used as a mean for predicting the outcome of microwave thermotherapy for benign prostate hyperplasia (BPH). METHODS: The two models--Henriques' damage integral and Jung's compartment model--were implemented in a computer program. Real treatment data for 22 patients with BPH who were in chronic retention were used as input, including measured intraprostatic temperatures and microwave power. To test if modelling gives results that are consistent with actual observations, comparison with transrectal ultrasound (TRUS) measurements of the prostate volume before and after treatment was made. The sensitivity of the computer model for variations in the heat cytotoxicity and the temperature probe location in the adenoma was also tested. RESULTS: The average TRUS volume reduction 3 months after treatment was 26 cc, whereas the corresponding cell kill calculation was 27 cc. The computer model appears to be rather insensitive to minor uncertainties in heat sensitivity and location of the intraprostatic reference temperature sensors. CONCLUSION: Cell-kill modelling appears to give results that are consistent with actual observations. The coagulated tissue volume is calculated in real time during the treatment, thereby providing an immediate prediction of the treatment outcome. By using cell-kill modelling, the endpoint of a treatment can be set individually; e.g., when a certain volume reduction has been achieved.